text
stringlengths 4.16k
144k
|
|---|
been discussed by Wang et al., which confirmed the potential regulatory mechanism of FTH1P3 on breast cancer paclitaxel resistance through the miR-206/ABCB1 axis, thus providing a novel insight for the breast cancer chemoresistance [88]. Since pseudogenes have high sequence homology with parental genes, they may also share a pool of miRNAs that bind to common MREs, and hence serve as miRNA sponges to regulate parental RNAs. The lncRNAs of the pseudogenes are therefore endowed with a competitive ability against the parental gene, which is based on the sponging of common miRNAs. This competitive ability is theoretically more relevant for gene families made up of a large number of pseudogenes, and the ferritin gene family belongs to this group. Indeed, the FTH1 gene:pseudogenes network has been reported to be relevant in prostate cancer, possibly by affecting iron balance [56]. By miRNA pulldown experiments, FTH1 and some of its pseudogenes (FTH1P2, FTH1P8, FTH1P11, and FTH1P16) were identified as targets for miR-181a-5p, miR-19b-1-5p, miR-19b-3p, miR-210-3p, miR-362-5p, miR-616-3p, and miR-638, which also showed ceRNA activity [56]. In DU145 and PC3 models of prostate cancer FTH1 and FTH1P11 and FTH1P16 show near equimolar expression level, further underlining the fulfilment of an important criterion for an effective parental gene:pseudogene ceRNA network [56]. As described in the previous section, FTH1 expression is regulated by iron availability, whereby an iron excess induces FTH1 expression and, conversely, iron depletion provokes FTH1 downregulation. Thus, providing or removing iron to cells is a physiological approach to fine regulate FTH1 expression. Interestingly, in DU145 and PC3 models of prostate cancer, the addition of iron induced the expression not only of FTH1, but also of FTH1P11 and FTH1P16, two pseudogenes whose expression is not regulated by IREs, while iron depletion down-regulated FTH1, FTH1P11, and FTH1P16 [56]. This type of experimental demonstration, whereby physiologically modulating the expression level of a gene of the ceRNA pathway, a modification in the same direction of the corresponding gene of the same pathway is observed provides more robust and convincing evidence of the efficacy of the ceRNA pathway, especially when compared with the most common approaches of ectopic overexpression or downregulation of gene/pseudogenes. Further investigation with mutated MREs, confirmed that the reciprocal regulation of FTH1 and FTH1P11/16 expression was mediated by ceRNA interactions [56]. Conclusions Since the link between iron balance, oxidative stress, and tumorigenesis is widely recognized, the parental FTH1/pseudogenes ceRNAs could have a much broader role in oncology than that shown so far. Indeed, the ceRNAs identified in the prostate cancer model may also have an impact in other tumor settings; likewise, other lncRNA of ferritin pseudogenes may act in a similar way, which is why the measurement of lncRNA-FTH1-pseudogenes expression levels is desirable in those tumor models in which FTH1, iron balance, and oxidative stress appear to be involved. This approach would provide an additional level of understanding of the intricate connections between the various levels of gene expression and biological outcomes of cancer relevance, such as proliferation, survival, metastasis, and drug resistance, and would represent a paradigmatic example to extend to those oncogenes or tumor suppressor genes represented by a large family of pseudogenes. Another field that needs further investigation is the link between the FTH1/pseudogenes networks, the miRNAs involved in these networks, and other tumor genes. An example of such possible connections is depicted in Figure 4, which summarizes the possible connection between the FTH1/pseudogenes networks that have so far been experimentally demonstrated and oncogenes/tumor suppressor genes, based on common miRNA targets reported in the literature. Metabolic reconstruction and experimental verification of glucose utilization in Desulfurococcus amylolyticus DSM 16532 Desulfurococcus amylolyticus DSM 16532 is an anaerobic and hyperthermophilic crenarchaeon known to grow on a variety of different carbon sources, including monosaccharides and polysaccharides. Furthermore, D. amylolyticus is one of the few archaea that are known to be able to grow on cellulose. Here, we present the metabolic reconstruction of D. amylolyticus’ central carbon metabolism. Based on the published genome, the metabolic reconstruction was completed by integrating complementary information available from the KEGG, BRENDA, UniProt, NCBI, and PFAM databases, as well as from available literature. The genomic analysis of D. amylolyticus revealed genes for both the classical and the archaeal version of the Embden-Meyerhof pathway. The metabolic reconstruction highlighted gaps in carbon dioxide-fixation pathways. No complete carbon dioxide-fixation pathway such as the reductive citrate cycle or the dicarboxylate-4-hydroxybutyrate cycle could be identified. However, the metabolic reconstruction indicated that D. amylolyticus harbors all genes necessary for glucose metabolization. Closed batch experimental verification of glucose utilization by D. amylolyticus was performed in chemically defined medium. The findings from in silico analyses and from growth experiments are discussed with respect to physiological features of hyperthermophilic organisms. Electronic supplementary material The online version of this article (10.1007/s12223-018-0612-5) contains supplementary material, which is available to authorized users. Introduction Archaea were described as an independent phylogenetic group of microorganisms as early as 1977 (Woese and Fox 1977). The first two established archaeal phyla were the Euryarchaeota and the Crenarchaeota (Woese et al. 1990). Although archaea exhibit many prokaryotic characteristics, they also possess metabolic pathways that are markedly different from other organisms (Bräsen et al. 2014;Sato and Atomi 2011). The central carbohydrate metabolism of archaea comprises unique variants of enzymes within the Embden-Meyerhof-Parnas (EMP) pathway and the Entner-Doudoroff (ED) pathway (Bräsen et al. 2014;Verhees et al. 2003). The EMP pathway includes the phosphorylation of glucose or fructose to alpha-Dglucose-6-phosphate and fructose-6-phosphate, the cleavage of fructose-1,6-bisphosphate to glyceraldehyde-3p h o s p h a t e , a n d t h e f u r t h e r o x i d a t i o n t o glycerinaldehyde-3-phosphate by a glycerinaldehyde-3phosphate:ferredoxin oxidoreductase (GAPOR). This is followed by phosphoglycerate mutase (PGM), an enolase, pyruvate kinase resulting in pyruvate, and the oxidation from pyruvate to acetyl-CoA by a pyruvate-ferredoxin oxidoreductase (PFOR). Anaerobic archaea possess the enzymes to convert acetyl-CoA to acetate by ADPforming acetyl-CoA syntethase, while sulfur-, O 2 -, and nitrate-reducing archaea use the tricarboxylic acid cycle to oxidize it into two carbon dioxide (CO 2 ) molecules (Siebers and Schönheit 2005). Based on the results of automated archaeal genome annotations, complete metabolic routes for known pathways could initially not be confirmed due to missing information on the archaeal enzymatic machinery. This was also the case when the CO 2fixing pathway was identified in Thaumarchaeota (Sato and Atomi 2011). The ability to grow autotrophically is widespread among the Crenarchaeota and can be found among the orders Sulfolobales, Thermoproteales, and Desulfurococcales. Genome analysis The genome of D. amylolyticus DSM 16532 (Z-1312) was sequenced by Susanti et al. 2012. The genome contains 1,384,116 bp with a GC content of 44.8%. One thousand seventy-five out of 1475 protein-coding genes were predicted to have known functions (Susanti et al. 2012). Metabolic reconstruction of the D. amylolyticus genome was manually performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa and Goto 2000), the Comprehensive Enzyme Information System (BRENDA) (Schomburg et al. 2002), the Universal Protein Resource (UniProt) (Boeckmann et al. 2005) domain database by the European Bioinformatics Institute (EMBL-EBI), and the Protein Family database (PFAM) (Finn et al. 2016). Enzyme information for the carbon metabolism (red), carbon fixation pathways in prokaryotes (red), glycolysis and glyconeogenesis (green), pyruvate metabolism (green), pentose phosphate pathway (orange), citrate cycle (yellow), starch and sucrose metabolism (blue), fructose and mannose metabolism (purple), galactose metabolism (gray), glyoxylate and dicarboxylate metabolism (pink), ABC transporter systems, and the secretion system were superimposed onto KEGG pathway maps ( Supplementary Fig. 1). NCBI database entries of gene-ID or protein-ID for each enzyme obtained from KEGG and BRENDA were analyzed using Basic Local Alignment Search (BLAST) (Altschul et al. 1990). BLAST was performed with the following settings: tblastn, tblastx, and blastn within the database reference genomic sequences (refseq_genomic) and blastp within the database reference proteins (refseq_protein). For a shorter calculation time, searches were limited to the taxa Desulfurococcales (taxid: 114380). Blastn calculations were extended to the BLAST algorithm Bsomewhat similar sequences.R esults for Query coverage, E-value, and Identity can be found in the supplemental material (Supplementary Tables 1 and 2). Supplementary Table 1 shows the complete list of enzymes, previously reported in KEGG enzyme maps, while Supplementary Table 2 shows enzymes of KEGG maps, which have not yet been assigned to the D. amylolyticus genome. The PFAM domain database was used to verify known protein family homologs based on the protein information received from BLAST results. All gene-IDs and protein-IDs shown in the proposed genome of D. amylolyticus have been checked for homologs PFAM families found in Supplementary Table 3. The identity threshold to determine homology was always ≥ 95%, except for PF00389 and PF02826 where the identity threshold was 70%. Chemicals CO 2 , N 2 , 20 Vol.-% CO 2 in N 2 , and CO were of test gas quality (Air Liquide, Schwechat, Austria). All other chemicals were of highest grade available. Closed batch cultivation Cultures of D. amylolyticus were grown anaerobically at 5·10 4 Pa under either 20 Vol.-% CO 2 in N 2 , 100% N 2, or 100% CO 2 in a closed batch set-up (Rittmann and Herwig 2012). The following carbon sources were individually tested: arabinose, cellulose, fructose, glucose, lactose, maltose, starch, sucrose, CO, and CO 2 , with or without YE at 5 g/L, except for cellulose, which was applied at 2 g/L. Two different approaches for the closed batch experiments were performed. In the closed batch experiments shown in Table 1, the pH was adjusted in each serum flask separately. The serum bottles were agitated at 100 rpm in an air bath (Labwit-Zwyr-2102c, Lab Xperts Laboratory Solutions Austria, Klosterneuburg, Austria). The pre-culture for inoculation was obtained from a fructose-grown D. amylolyticus culture. All experiments were performed in duplicates together with a negative control and reproduced three times. In the closed batch experiments shown in Table 2, the pH of the medium was adjusted in a 1-L gas-tight flask (pressure plus+ GL 45, clear, Duran Group, Mainz, Germany) before being distributed to the individual serum bottles sealed by rubber stoppers (Ochs Glasgerätebau, Langerwehe, Germany). The serum bottles were agitated at 200 rpm in an air bath (Labwit-Zwyr-2102c, Lab Xperts Laboratory Solutions Austria, Klosterneuburg, Austria). In the closed batch experiments shown in Table 2, fructose pre-grown D. amylolyticus cells were harvested by centrifugation (Eppendorf Centrifuge 5415R, Eppendorf, Hamburg, Germany) for 20 min and 15,700 g. The supernatant was removed and the resulting pellet washed with the respective medium. All experiments were performed in quadruplicates together with a negative control and reproduced twice. Pressure was always determined before samples for microscope analysis were obtained. For inoculation, 5% (v/v) of pre-culture was added anaerobically in the anaerobic glove box (Coy Laboratory Products, Grass Lake, USA) using a gas-tight syringe (Soft-Ject, Henke Sass Wolf, Tuttlingen, Germany). After inoculation, the headspace of the bottles was filled with the respective gas and incubated at 80°C in an air bath (Labwit-Zwyr-2102c, Lab Xperts Laboratory Solutions, Klosterneuburg, Austria). Depending on the specific growth rate (μ) of D. amylolyticus on different carbon substrates, samples of 1 mL of suspension were taken for cell counts at regular intervals. After sampling, the serum bottle headspace was re-prezurized to 5·10 4 Pa using the respective gas. Cultivation in chemically defined medium was used to examine how μ and the final cell concentration of D. amlyloyticus are affected by the presence of the different carbon sources. Cell counting Biomass samples were withdrawn from the serum bottles by using syringes (Soft-Ject, Henke Sass Wolf, Tuttlingen, Germany) and hypodermic needles (Sterican size 14, B. Braun, Melsungen, Germany). D. amylolyticus cells were counted by applying 10 μL of sample onto a Neubauer improved cell-counting chamber (Superior Marienfeld, Lauda-Königshofen, Germany) with a grid depth of 0.1 mm. Cultures were counted using a Nikon microscope (Nikon Eclipse 50i, Nikon, Amsterdam, Netherlands). Measurement of the absorbance for estimation of cell concentration could not be performed due to particle interference inside the cultivation media. Data analysis The maximum specific growth rate (μ max [1/h]) and the mean specific growth rate (μ mean [1/h]) were calculated as follows: e μt with N, cell concentration [cells per mL]; N 0, initial cell concentration [cells per mL]; t, time [h]; and e, Euler number. μ was calculated from the delta cell counts (intervals) from the growth curves. |
μ max is the maximum of the slope of all μ. μ mean is the mean of all μ of the growth curve where an increase of cell concentration was visible. Substrate uptake According to the metabolic reconstruction of D. amylolyticus, the organism is able to import a variety of carbon compounds by using a variety of genome encoded sugar transporters ( Supplementary Fig. 1). These compounds include the monosaccharides glucose, fructose (hexoses) and arabinose (pentose), the disaccharides maltose, lactose and sucrose, and the polysaccharides starch and cellulose. This reconstruction differs from previous work in which glucose was shown not to be metabolized by D. amylolyticus (Perevalova et al. 2005). A complete list of all enzymes and PFAM domains can be found in Supplementary Tables 1, 2 and 3. D. amylolyticus DSM 16532 exhibited the highest gene similarity to D. amylolyticus 1221n (formerly known as Desulfuroccus kamchatkensis) and to D. amylolyticus Z-533 T (Supplementary Table 1). This finding is not very surprising as D. amylolyticus DSM 16532 and D. amylolyticus 1221n were reclassified as synonyms of D. amyloyticus ). According to our BLAST analyses, the carbon substrates could be channeled to the central catabolic pathway by 97 ABC transporter family genes, e.g., Desfe_0184, Desfe_0187, Desfe_0620, Desfe_0639, Desfe_0721, and Desfe_0754 as summarized in Supplementary Tables 1 and 2. However, the biochemistry of these ABC transporters is not known. Additionally, other enzymes in relation to sugar transport were identified in the genome, such as the multiple sugar transport system ATP-binding protein (Desfe_1188), which could channel the sugar via ABC transporter permease (Desfe_0355 and Desfe_0366) and the ABC transporter substrate-binding protein (Desfe_0354). Fermentative growth D. amylolyticus possesses all necessary genes for gluconeogenesis and glycolysis (Fig. 1). The glucose, fructose, and mannose degradation pathways lead to generation of beta-Dfructose-6-phosphate, where they enter glycolysis. This step could be followed by phosphofructokinase (PFK) (Desfe_0717 and Desfe_0968), fructose-bisphosphate aldolase, class I (Desfe_0718), and fructose 1,6-bisphosphate aldolase/phosphatase (FBPase, Desfe_1349). From results of the metabolic reconstruction, D. amylolyticus would be able to use the classical and the archaeal (modified) EMP to convert glyceraldehyde-3-phosphate to glycerate-3-phosphate. For the classical EMP, glyceraldehyde 3-phosphate dehydrogenase (GAPDH, Desfe_0262) and phosphoglycerate kinase (Desfe_0261) could be used, which would result in the production of NADPH and ATP. It must be noted that, except for halophilic archaea, GAPDH is involved in gluconeogenesis not in glyclolysis (Siebers and Schönheit 2005). However, in the archaeal EMP pathway, glyceraldehyde ferredoxin oxidoreductase (GAPOR, Desfe_0557) or a non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Desfe_0067) could be used, which would result in 2 mol of reduced ferredoxin (Fd 2− ) or NADH respectively. FBPase (Desfe_1349) as well as GAPDH (Desfe_0067) are enzymes used in gluconeogenesis and counteract the irreversible reactions of the modified EMP, like PFK, GAPOR, and pyruvate kinase (Siebers and Schönheit 2005). The archaeal EMP pathway was shown to have many variations, one of which has been shown to operate in Pyrococcus furiosus (Siebert and Schönheit 2005). Pyruvate, the end product of the EMP pathway (Kengen et al. 1996), was obtained via a 2,3-bisphosphoglycerate-independent phosphoglycerate mutase (Desfe_0416), enolase (Desfe_0063), pyruvate kinase (Desfe_1347), or pyruvate water dikinase (Desfe_0879). Pyruvate might be decarboxylated by PFOR to acetyl coenzyme A (acetyl-CoA). D. amylolyticus possesses 2 PFOR homologs: two alpha subunits (Desfe_0503 and Desfe_1298), two beta subunits (Desfe_0502 and Desfe_1299), two gamma subunits (Desfe_0505 and Desfe_1296), and two delta subunits (Desfe_0504 and Desfe_1297). This combined action would result in 2 Fd 2− . D. amylolyticus possesses a dihydrolipoamide dehydrogenase (Desfe_0667) which could result in the production of 1 NADH through pyruvate degradation. However, the function of dihydrolipoamide dehydrogenase is only known in bacteria, where it catalyzes the oxidation of dihydrolipoamide. This reaction has not been shown in Archaea (Jolley et al. 1996). D. amylolyticus utilizes ATPdependent sugar kinases during glycolysis (Kengen et al. 1994;Hansen and Schonheit 2000). Therefore, D. amylolyticus DSM 16532 must regenerate ATP for the anabolic reactions. Some of the ATP can be produced via substrate level phosphorylation through AMP-utilizing phosphoenolpyruvate synthase. Another possibility for ATP production could be chemiosmotic phosphorylation involving Fd 2− oxidation. Fd 2− can be produced through the action of GAPOR and/or PFOR. The oxidation of Fd 2− could be coupled to the generation of a proton motive force and subsequent production of H 2 (Bräsen et al. 2014). D. amylolyticus possesses several homologs of hydrogenases, including a membrane-bound hydrogenase, which could be responsible for the generation of a proton motive force and concomitant H 2 production. Despite very high cell-specific H 2 production rates (3.41-8.42 fmol/cell h) were achieved during batch experiments in bioreactors, the volumetric H 2 production rates of D. amylolyticus are still too low to be of biotechnological relevance (Reischl et al. 2018). The metabolic routes for substrates to enter glycolysis differ for the different carbon compounds. With respect to monosaccharides, glycolysis may commence with the degradation of the monosaccharide glucose starting with a ROK family protein (Desfe_0578) to alpha-D-glucose-6-phosphate. From that point, the bifunctional phosphoglucose/phosphomannose isomerase (Desfe_1128) is used to generate beta-D-fructose 6phosphate. These steps would consume 1 ATP. Fructose could enter via fructokinase (Defe_0717 and Desfe_0968) and may be converted to beta-D-fructose 6-phosphate while also consuming one ATP. Arabinose degradation in archaea is still unresolved and needs further investigation (Brouns et al. 2006). However, the gene annotations presented here indicate that D. amylolyticus possesses an alcohol dehydrogenase (Desfe_1240) that could form D-arabino-1,4-lactone and contains a D-arabino-1 dehydrogenase, which is a homolog to the dehydrogenase found in Sulfolobus solfataricus (SSO1300). Clusters of orthologous groups (COG) from various organisms for arabinose degradation such as COG3970, COG4948, COG0129, and COG0179 (Brouns et al. 2006) and genes for arabinose degradation found in S. solfataricus, e.g., SSO3124, SSO3117 and SSO3118 (Peng et al. 2011) or SSO3107, SSO1303 (Brouns et al. 2006), could not be detected in the genome of D. amylolyticus. Only a homolog of the 2keto-3-deoxy-D-arabinonate dehydratase (COG3970), which is responsible for arabinose degradation, was found. Like in several Burkholderia species and in Azospirillum brasiliense, this gene could be replaced in D. amylolyticus by a dihydrodipicinolate synthase family protein (COG0329) (Brouns et al. 2006). When homologous genes from Haloferax volcanii are used for identification of genes for arabinose degradation in D. amylolyticus, only a gene encoding for a NAD-dependent epimerase (Desfe_0989) homologous to the H. volcanii gene HVO_B0032, which forms L-arabinoate, could be detected. Furthermore, no H. volcanii homologs for arabinose degradation (e.g. HVO_B0038A, Supplementary Fig. 1 HVO_B0027, or HVO_B0039 (Johnsen et al. 2013) could be identified in the genome of D. amylolyticus. Concerning utilization of disaccharides by D. amylolyticus, sucrose could be split into D-fructose and alpha-D-glucose by a hypothetical protein, also annotated as sucrose alphaglucosidase (Desfe_0611) and further transformed to alpha-D-glucose-phosphate by a ROK family protein (Desfe_0578) making use of 1 ATP. Lactose could be partitioned into Dglucose and D-galactose by beta-glucosidase (Desfe_0624) followed by a ROK family protein (Desfe_0578) generating alpha-D-glucose-6-phosphate. Maltose is split into 2 mol of glucose (Schäfer and Schönheit 1992) and could be degraded by maltokinase (Desfe_0406) using one ATP followed by the use of starch synthase (maltosyl-transferring, Desfe_0644) producing amylose and further metabolized to ADP-glucose by a starch synthase (glycosyl-transferring, Desfe_0403). With a sugar-phosphate nucleotidyltransferase (Desfe_0962) or via glucose-1-phosphate adenylyltransferase (Desfe_0189) that finally enters the glycolysis as alpha-D-glucose-1-phosphate, homologs of maltokinases (EC 2.7.1.175) are present in almost all known bacterial phyla as well as in some Crenachaeota (Fraga et al. 2015). With respect to polysaccharide utilization, D. amylolyticus seems to possess several options. Starch could be broken down by starch phosphorylase (Desfe_0264) to amylose and alpha-D-glucose-1-phosphate. Amylose can be broken down by starch synthase (glycosyl-transferring) (Desfe_0403) to ADP-glucose. From ADP-glucose, glycolysis commences via nucleotidyltransferase (Desfe_0189) or sugar-phosphate nucleotidyltransferase (Desfe_ 0962) again forming alpha-Dglucose-1-phosphate. Cellulose could be degraded to cellobiose by an endoglucanase (Desfe_0691) and further to beta-Dglucose by a beta-glucosidase (Desfe_0624). Interestingly, no sequence homology to any known cellulose degrading enzyme could be found in the genome of D. amylolyticus. As our strain is unique among Desulfurococcus spp. to utilize cellulose (Graham et al. 2011), we compared the enzymes endoglucanase (Desfe_0691) and beta-galactosidase (Desfe_0624) to D. amylolyticus Z-533 T enzymes SPHMEL_RS03930 and SPHMEL_RS03240. There is a slight difference in the sequence: the endoglucanase gave an identity of 99% while the beta-galactosidase only showed 97% identity. Since it is known that the presence of cellulose genes does not assure the ability of an organism to be able to metabolize this substrate (Graham et al. 2011), the difference in gene identity was very interesting to note. Although the difference is not very high, the gene might be a new type of cellulase and the putative reason why the strain D. amylolyicus DSM 16532 is able to utilize cellulose. Hence, metabolomics and subsequent biochemical characterization seem to be necessary to elucidate the cellulose degradation pathway of this organism. Further degradation of cellobiose could be achieved by an ATP consuming step to beta-D-glucose-6-phosphate induced by a ROK family protein (Desfe_0578). Beta-Dglucose-6-phosphate then could enter glycolysis with the bifunctional phosphoglucose/phosphomannose isomerase (Desfe_1128) resulting in beta-D-fructose-6-phosphate. The metabolic end product acetate could in theory be generated from acetyl-CoA by acetyl-CoA synthetase (Desfe_0782 and Desfe_1050), resulting in formation of 1 ATP. For synthesis of lactate and ethanol, reducing equivalents such as NAD(P)H and Fd 2− would have to be used. Aldehyde dehydrogenase (Desfe_0067) could catalyze the reaction from acetate to acetaldehyde, a very thermodynamic unfavorable reaction (Thauer et al. 1977), followed by alcohol dehydrogenase (Desfe_0019 and Desfe_1240) catalysis, which would form ethanol. Two NAD(P)H would be needed for these two reactions. The formation of the metabolic end product lactate would only require 1 NADH and could be catalyzed by a lactate dehydrogenase (Desfe_1212). Incomplete CO 2 -fixation pathways Despite intensive examination and analysis of the D. amylolyticus genome, only an incomplete reductive citrate cycle could be identified, where certain essential genes are absent. To complete the reductive citric acid cycle in this organism, it would be necessary to find candidates for the missing enzymes in the genome and to perform biochemical characterization of the respective candidate enzymes. Unfortunately, this incomplete citric acid cycle is the case for many archaeal genomes (Vanwonterghem et al. 2016). D. amylolyticus could utilize the existing reactions of the reductive citric acid cycle for converting phosphoenolpyruvate (Desfe_0003) or pyruvate (Desfe_0591) to oxaloacetate. From there, oxaloacetate could be reduced to (S)-malate by malate dehydrogenase (Desfe_0284). The dehydration of malate to fumarate remains unresolved, as it is the case in P. furiosus and Thermococcus kodakarensis, although these two organisms are able to use a malic enzyme (EC 1.1.1.39) to dehydrate malate to pyruvate (Fukuda et al. 2005). However, no malic enzyme homologs could be identified in D. amylolyticus. Fumarate and succinate are interconverted by fumarate reductase/succinate dehydrogenase flavoprotein domain protein (Desfe_0481). Succinyl-CoA synthetase beta subunit (Desfe_1156) and succinyl-CoA synthetase alpha subunit (Desfe_1155) convert succinate to succinyl-CoA. The 2-oxoglutarate ferredoxin oxidoreductase complex including two alphas (Desfe_0475 and Desfe_0499), two betas (Desfe_0474 and Desfe_0498), one gamma (Desfe_0497), and one delta subunit (Desfe_0500) catalyze the reaction from succinyl-CoA to 2-oxoglutarate. Homologs for isocitrate dehydrogenase linking isocitrate to 2-oxoglutarate could not be identified in the genome. Citrate and isocitrate isomerization could be performed by aconitase (Desfe_0217). To be able to produce 2-oxoglutarate and aspartate, D. amylolyticus would b Growth of D. amylolyticus on disaccharides. c Growth of D. amylolyticus on monosaccharides be able to use aspartate aminotransferase (Desfe_0590) and further glutamate dehydrogenase (Desfe_0075) converting glutamate to 2-oxoglutarate while producing NADPH. Additionally, fumarate and aspartate utilization could be coupled via adenylosuccinate synthetase (Desfe_0482) and adenylosuccinate lyase (Desfe_0494). The presence of other CO 2 -fixation pathways of Archaea, such as the dicarboxylate-4-hydroxybutyrate cycle, the 3hydroxypropionate bicycle, and the 3-hydroxypropionate-4hydroxybutyrate cycle, in the genome of D. amylolyticus were also investigated (Supplementary Table 4). However, despite intensive in silico examinations, no complete CO 2 -fixation pathway (Bar-Even et al. 2011;Berg et al. 2007;Berg et al. 2010;Huber et al. 2008;Thauer 2007) could be identified. Within the dicarboxylate-4-hydroxybutyrate cycle (Berg et al. 2010), only the conversion from succinate to succinyl-CoA (Desfe_1155 and Desfe_1156), the oxidation to succinate semialdehyde (Desfe_0067) and the transformation to 4hydroxybutyrate by homologs of alcohol dehydrogenase (Desfe_1240 and Desfe_0019), and the subsequent transformation from acetoacetyl-CoA to two acetyl-CoA (Desfe_0849) could be proposed to take place in D. amylolyticus. The other steps necessary for performing the dicarboxylate-4-hydroxybutyrate cycle (Bar-Even et al. 2011;Huber et al. 2008) remain undetected. Similarly, only a few genes |
encoding reactions of the 3-hydroxypropionate bicycle (Bar-Even et al. 2011;Berg et al. 2010) were found in the genome, including the enzyme for formation of acrylyl-CoA from 3-hydroxypropionate (Desfe_0019 and Desfe_1240) and the enzyme conversion of acrylyl-CoA to propionyl-CoA (Desfe_0880). Our extensive genome re-annotation revealed that D. amylolyticus harbors several genes involved in various CO 2 fixation pathways, but from a metabolic reconstruction perspective, this organism seems to rely on the fermentative growth mode for maintaining its carbon and energy metabolism. Determination of physiological characteristics Based on the metabolic reconstruction, closed batch experiments were conducted to determine the carbon source resulting in the highest μ max , μ mean , and highest final cell densities. Therefore, D. amylolyticus was grown on selected carbon compounds (arabinose, cellulose, fructose, glucose, lactose, maltose, sucrose, and starch) and on CO 2 and CO in a closed batch system. D. amylolyticus was able to grow on all tested poly-, di-, and monosaccharides. Growth curves of D. amylolyticus grown in a medium containing one of these sugars and YE is shown in Fig. 2. Unambiguously, growth on starch (Fig. 2a, Table 1), provided the best growth conditions resulting in a μ max of 0.059 1/h and a cell concentration of 3.00·10 7 confirming earlier reports (Perevalova et al. 2005). When considering the metabolic burden of sugar transport into the cell to produce cellular energy, the constraints of the fermentative growth of D. amylolyticus on starch, accompanied by acetate formation, could result in the best growth conditions and highest cellular energy gain. However, during the cultivation of D. amylolyticus on starch solid particles, aggregation occurred inside the growth medium which made cell counting difficult to almost impossible (as indicated in the standard deviation shown in Fig. 2a). Slow growth and growth to low cell densities could be observed when D. amylolyticus was grown on disaccharides ( Fig. 2b and Table 1). This could be explained by putative unintended by-product formation and accumulation due to the Maillard reaction (Lerche et al. 2002), which was inhibiting growth of D. amlyolyticus. According to a study on anaerobic bacteria, melanoidins, the product of the Maillard reaction, have strong prebiotic potential and can be used as carbon source by particularly Bifidobacterium spp. (Borrelli and Fogliano 2005). However, the effect on archaea could be different. Another reason could be that trace elements (e.g., tungsten and/or selenium) were limiting growth. Possibly also the sulfur source for growth of D. amylolyticus could be reconsidered, as sulfide is known to precipitate metal ions. As an alternative, growth tests using cysteine instead of sulfide could be performed. However, the generally encountered characteristic of hyperthermophilic organisms to grow only to low cell densities could be circumvented by applying cell retention systems, or by using multivariate optimization procedures for improving final biomass concentration values, as they have already been employed for thermophilic microorganisms (Abdel Azim et al. 2017;Seifert et al. 2014). Yet, there are still many unknown parameters that limit growth of hyperthermophiles to high cell densities (Chou et al. 2008;Pawar and Niel 2013). The fastest growth of D. amylolyticus on monosaccharides and YE was obtained from fructose Fig. 2c). Growth of D. amylolyticus on fructose comprised a μ max of 0.052 1/h and reached a cell concentration of 2.98·10 7 . From our metabolic reconstruction, there was no convincing indication that D. amylolyticus would not be able to grow on glucose (Fig. 1). Therefore, growth of D. amylolyticus was tested in a chemically defined medium containing fructose, cellulose, or glucose (Fig. 3). In contrast to Perevalova et al. 2005(Perevalova et al. 2005, growth of D. amlylolyticus on glucose was achieved at a μ max of 0.059 1/h. Even though the metabolic reconstruction revealed enzymes that could be involved in CO 2 fixation, almost no growth could be detected when CO 2 was used as sole source of carbon for growth ( Table 2). This is not surprising and might be due to the fact that in addition to CO 2 , another energy source, e.g., H 2 would be needed for cultivation of D. amylolyticus. Also, almost no growth could be observed when D. amylolyticus was grown on CO (Table 2). In the latter experimental setting, D. amylolyticus was always grown in a chemically defined medium lacking YE (Fig. 3). However, the addition of YE stimulated growth of D. amylolyticus (compare individual growth curves of Figs. 2 and 3). YE is very expensive and a source of rich complex nutrients, proteins, and minerals and therefore aimed to be omitted if a biotechnological production processes is envisioned (Willquist and van Niel 2012). Furthermore, the omission of YE is a prerequisite for physiological studies which aim to achieve fine-resolution mass balancing analyses. According to the metabolic reconstruction of D. amylolyticus ( Supplementary Fig. 1), growth on all tested carbohydrates would result in standard glycolytic ATP gain or loss. Therefore, the future research question would be how and if D. amylolyticus is able to gain and maintain redox equivalent homeostasis, or if cellular energy can also be obtained from co-assimilation of YE. As in their natural environment archaea could encounter carbon oligotrophic conditions, the organism could be growing mixotrophically. A putative co-assimilation of certain components contained in YE might be advantageous for D. amylolyticus. In this respect, also a rudimentary reverse citric acid cycle or a rudimentary 3-hydroxypropionate/4-hydroxybutyrate cycle could assist the balance of anaplerotic and cataplerotic reactions (Berg et al. 2010). An option could be to use carbon isotope labelling studies for elucidation of the CO 2 fixation potential or gene expression analysis. Another discussion point concerns the affinity of ABC transporters towards certain sugars (Bräsen et al. 2014;Willquist et al. 2010). The determination of ABC transporter specificity could be beneficial in designing future experiments designed to understand under which growth conditions and from which carbohydrate the highest biomass concentrations could be obtained. Such a physiological understanding would be necessary in order to achieve high biomass concentrations for subsequent biochemical, physiological, and biotechnological studies. In Silico Study of Silybum Marianum Targeting PARP protein (4UND protein) Title: Abstract-Silybum Marianum, is a plant belonging to the family Asteraceae. For many centuries it has been used a natural remedy for many diseases like Liver and Biliary tract diseases. It is effective as an antioxidant and is used in a variety of diseases. This study was conducted to check the effects of Silybum Marianum on PARP protein (4UND protein). The Molecular Docking techniques was chosen to check the effects of different chemical constituents of Silybum marianum on DNA damaging protein. For this purpose, different PARP inhibitor drugs were taken as standard. The Molecular Docking of the chemical constituents of Silybum marianum was performed using 4UND protein with the help of PyRx software along with BIOVIA Drug Discovery studio software. The result of molecular docking showed that some of the chemical constituent have higher binding affinity than standard PARP inhibitor drugs. and it has white-veined leaves with toothed spiny edges and purple flowers on the pinnacle of the stem and the ramifications ripe into 6-7 mm long brownish fruits (achenes) with a white silky pappus at their apex and has big prickly edged leaves included with undulating white patches and stems containing a milky juice [5][6] [7]. [4] Fig. Silybum Marianum Milk Thistle`s active component is a lipophilic extract obtained from the seeds of the plant, and it contains three isomer flavonolignans (silybin, silydianin, and silychristin), together known as silymarin [8] [9]. Because of the existence of phytoconstituents such as antioxidants and complete phenolics, milk thistle is used as a medicine for the treatment of a variety of diseases [9]. For years, Silybum Marianum has been used as a natural treatment for liver and biliary tract diseases [10]. A study that was done was to see how effective Silybum Marianum is at scavenging free radicals in vitro and how powerful it is as an antioxidant [11]. Milk thistle extracts are currently being researched for chemoprevention, recovery, and amelioration of chemotherapy side effects in cancer experimental therapeutics, but however, nomenclature precision has fallen behind [12]. Preclinical evidence for silymarin`s hepatoprotective and anti-carcinogenic effects, includes cancer cell growth inhibition in humans, prostate, skin, breast and cervical cells [13]. Extracts from milk thistle seeds have shown efficacy in halting the progression of human prostate carcinoma in a variety of in vitro and in vivo preclinical models over the last few years [14]. It has also been proposed that Milk Thistle can be used in diabetes to diminish insulin resistance [15]. NAD+ protein (ADP-ribosyl) transferase (polymerizing) (PARP) is a widely distributed chromatinassociated protein that is highly conserved and found in all eukaryotes except yeast [16]. PARP attaches to DNA strand, breaks and catalyses the transfer of ADP ribose from its substrate NAD+ to a number of nuclear acceptors, mostly to itself, in response to DNA damage induced by external genotoxic chemicals and endogenous cellular processes [17]. PARP-1 uses its two zinc fingers to detect DNA nicks, which it catalytically activates by binding to the break and helps to repair. The only other PARP enzyme activated by and assisting in the repair of DNA breaks is PARP-2, which has the closest structural resemblance to PARP-1 [18] [19]. Of all the proven use of PARP inhibitors the most promising ones are ischemia and cancer [20]. So, we have decided to do the research on cancer. The main bioactive component of medicinal plant (milk thistle) is silymarin (4% to 6%). Silymarin is the mixture of different flavonolignans such as, silybinin (60%) A and B (SBN A&B), iso-silybinin A and B (ISBN A&B), silychristin (SCN), and silydianin (SDN), it is a well-known Chinese herb. Silymarin is also present in the seeds, fruits as well as in the leaves of the Milk Thistle but the seed part has the maximum concentration of silymarin. Besides this Milk Thistle also includes other flavonoids such as-Rutin, quercetin, dihydrokaempferol, kaempferol, apigenin, naringin, eriodyctiol, chrysoeriol [4][9] [22]. Molecular Docking- The field of molecular docking has grown in popularity over the last three decades as a result of the structural molecular biology and structure-based drug discovery requirements [24]. Molecular docking is a type of computational modelling that helps predict the preferred binding orientation of one molecule (e.g., ligand) to another (e.g., receptor) when they interact to form a stable complex [25]. Materials- Hardware & Software-HP 15-da0300TU with Intel core CPU of 2.57 Ghz, 8 gb ram, and intel integrated graphics card of 4gb, and 64-bit windows operating system. PyRx a software [26] written in python language, which today can be used on any PC (personal computer), is used along with BIOVIA Drug Discovery software [27] for molecular docking. The files used for molecular docking were ligands which were downloaded from PubChem [28] in "sdf" format and Proteins which were downloaded from Protein bank in "pdb" format [29]. Methods- The protein files that were downloaded were first converted to "pdbqt" format from "pdb" format. The 8-Select the ligands using cntrl button on your keyboard and and using cursor. 9-Then on the screen the structure will appear make sure the whole structure is in the box. 10-Then click on Forward button and the docking will start. 11-After completion of the docking process click on "save as comma-separated vale" on the right side of your screen, and the binding affinity results will be saved in excel sheet. Table1 shows the binding affinity of standard PARP inhibitor drugs with 4UND PARP protein and Table 2 shows the binding affinity of chemical constituents of Silybum Marianum with 4UND PARP protein. The binding affinity of the other chemical constituents of Silybum Marianum which are also found to be good but not better than the standard PARP inhibitor drugs has been shown in Table 3 in appendix section as well as the binding affinity of the chemical constituents of Silybum Marianum whose binding affinity was found to be less than -6 is shown in Table 4 in appendix. (Table-1). The binding affinity of Talazoparib with 4UND protein was found to be -8.9 Sr.No kcal/mol and the binding affinity of Rucaparib with 4UND protein was found to be -7.8 kcal/mol. The (Table-1). The binding affinity of Olaparib with 4UND protein was found to be -9.8 kcal/mol and the binding affinity of Niraparib with 4UND protein was found to be -8. (Table-2). The binding affinity of Silydianin with |
4UND protein was found to be -11.2 kcal/mol and the binding affinity of Naringin with 4UND protein was found to be -11.2 kcal/mol. The binding affinity of Silydianin and Naringin with 4UND protein was found to be better than Campesterol, and Sitosterol, shows better binding affinity than Rucaparib. Silymarin, Silybinin, Apigenin, Quercetin, Campesterol and Sitosterol, shows better binding affinity than Niraparib. Discussion- For centuries, Silybum Marianum has been used in the treatment of various diseases mainly liver and biliary tract related diseases. It has been used as a natural remedy for all these centuries. In this study carried out by using Molecular Docking the binding affinity of chemical constituents of Silybum Marianum with 4UND PARP protein was carried out. The binding affinity of Silydianin and Naringin which are the chemical constituents of Silybum Marianum showed good interaction with 4UND protein. The binding affinity was found out to be as followed: The binding affinity of Standard PARP inhibitor drugs was found to be-1-Talazoparib: -8.9 kcal/mol. As we can see the best binding affinity among Standard PARP inhibitor drugs was found to be of OLAPARIB which is -9.8 kcal/mol. The binding affinity of chemical constituents of Silybum Marianum was found to be as followed: 1-Silymarin: -9.4 kcal/mol. 14-Rutin: -8.9 kcal/mol. As we can see the binding affinity of Silydianin which is found to be -11.2 kcal/mol and the binding affinity of Naringin which was also found to be -11.2 kcal/mol which is the best amongst all the chemical constituents of Silybum Marianum. From this we can say that the binding affinity of Silydianin and Naringin was found to be better than the binding affinity of Olaparib. And also, the binding affinity of Silychristin which is -10.1 kcal/mol which is also better than Olaparib, Niraparib, Rucaparib and Talazoparib. The previous proven activities of Silydianin, Naringin and Silychristin are-Silydianin which is an active chemical constituent of Silybum Marianum is proven to have inhibitory effect [30], antioxidant and cytotoxic activities in Caco-2 cells [31], used in prostate cancer [32], PTP1B inhibitor [33]. Naringin which is also a chemical constituent of Silybum Marianum shows antioxidant, anti-inflammatory, antiapoptotic, anti-ulcer, anti-osteoporotic and anti-carcinogenic activity [34], Anti-bacterial [35], it regulates p21 and survivin in triple negative breast cancer cells [36]. Silychristin which is also a chemical constituent of Silybum Marianum shows anti-oxidant, anti-inflammatory and Multidrug Resistance Modulation Activity [37], used in cancer prevention [38]. So, from these proven activities we can say that Silydianin, Naringin and Silychristin are previously being used for their anti-cancer activity and from the results of molecular docking we can also say that they can be used as PARP inhibitors in the future. The binding affinity of Rutin -8.9 kcal/mol was found to be equal to Talazoparib. The binding affinity of Sitosterol -8.8 kcal/mol was to be better than Rucaparib which is -7.8 kcal/mol. Olaparib was the first PARP inhibitor drug to get approved by European Medicines Agency in 2014, for mutant ovarin cancer [40] it also shows effect in late chemotherapy-refractory BRCA-deficient breast cancer and also olaparib is well tolerated and very active [41]. Rucaparib a PARP inhibitor drug is known to be tolerable and shows activity in patients with platinum-sensitive germline BRCA ½ mutated-HGOC [42]. Talazoparib which is a PARP inhibitor drug inhibits PARP catalytic activity, trapping PARP1 on damaged DNA and causing cell death in BRCA ½ -mutated cells [43]. Niraparib a potent PARP ½ inhibitor drug with good cell activity shows promising activity in cancer patients [44]. So, on the basis of the result obtained we can expect that Silydianin and Naringin also possess the inhibition activity against Oxidative DNA damage, because Silybum marianum extract has the ability to show protection against photo-carcinogenesis [45] and also on the basis of the results of molecular docking we can expect that in the future Silybum Marianum can be used to create PARP inhibitor drugs. Conclusion- The analysis of the results of Molecular Docking of chemical constituents of Silybum Marianum with 4UND DNA damaging protein helped to examine and evaluate the binding affinity of chemical constituents of Silybum Marianum with 4UND DNA damaging protein. The results obtained showed that Silydianin and Naringin the chemical constituents of Silybum Marianum showed the binding affinity of -11.2 kcal/mol respectively, which is better than the best binding affinity amongst the Standard PARP inhibitor drugs which was shown by Olaparib which is -9.8 kcal/mol. From the above result we can conclude that Silydianin and Naringin, the chemical constituents of Silybum Marianum showed better binding affinity than Talazoparib, Rucaparib, Olaparib, and Niraparib which are the standard PARP inhibitor drugs. Also, Silychristin which is a chemical constituent of Silybum Marianum shows the binding affinity of -10.1 kcal/mol, which is also better than Talazoparib, Rucaparib, Olaparib, and Niraparib. So, we can say that in the future we can expect to use Silybum Marianum to create PARP inhibitor drugs. Evaluation of Histological Changes in Back Muscle Injuries in Rats over Time Study Design Animal model study. Purpose The purpose of this study was to evaluate the histological variation in the injured muscle and production of calcitonin gene-related peptide in rats over time. Overview of Literature Vertebral surgery has been reported to cause atrophy of the back muscles, which may result in pain. However, few reports have described the time series histological variation in the injured muscle and changes in the dominant nerve. Methods We used 30 male, 8-week-old Sprague-Dawley rats. The right and left sides of the paravertebral muscle were considered as the injured and uninjured sides, respectively. A 115 g weight was dropped from a height of 1 m on the right paravertebral muscle. Hematoxylin and eosin (H&E) staining of the muscle was performed 1–3 weeks after injury for histological evaluation. Fluoro-Gold (FG) was injected into the paravertebral muscle. The L2 dorsal root ganglia on both sides were resected 1, 2, and 3 weeks after injury, and immunohistochemical staining for calcitonin gene-related peptide was performed. Results H&E staining of the paravertebral muscle showed infiltration of inflammatory cells and the presence of granulation tissue in the injured part on the ipsilateral side 1 week after injury. Muscle atrophy occurred 3 weeks after injury, but was repaired via spontaneous replacement of muscle cells/fibers. In contrast, compared with the uninjured side, the percentage of cells double-labeled with FG and calcitonin gene-related peptide in FG-positive cells in the dorsal root ganglia of the injured side was significantly increased at each time point throughout the study period. Conclusions These results suggest that sensitization of the dominant nerve in the dorsal root ganglia, which may be caused by cicatrix formation, can protract injured muscle pain. This information may be helpful in elucidating the underlying mechanism of persistent pain after back muscle injury. Introduction The incidence of spine surgery is estimated to be 1,800-2,050 and 400-725 million person-years in the United States and Japan, respectively [1]. According to research based on Medicare in the United States, the average rate of lumbar fusion increased by three-fold between 1992 and 2003 [2]. The Japanese Society for Spine Surgery and Related Research reported that the number of registered patients who underwent spinal surgery by their certified surgeons in 2011 had almost doubled that during the previous 10 years [3]. It has been reported that the conventional midline posterior approach for lumbar spine surgery causes significant muscle injury, which is particularly severe if powerful self-retaining retractors are used [4,5]. Gejo et al. [6] reported that muscle retraction for an extended time during the operation causes more severe back musclegdamage and a higher incidence of postoperative low back pain in human patients. Regarding skin injury, Kajita et al. [7] reported that painful skin scar had induced hyperalgia in rats, which might cause plastic changes in the central nervous system. However, the relationship between muscle injury and postoperative pain has not been clarified. Therefore, the purpose of the current study was to determine the relationship between the histological variation of injured muscle and the production of calcitonin gene-related peptide (CGRP) in the dominant nerve during the first 3 weeks after experimental back muscle injury in rats. Materials and Methods All animal procedures and protocols were approved by the ethics committee of our university and followed the United States National Institute of Health Guidelines for the Care and Use of Laboratory Animals (1996 revision). We used 30 male, 8-week-old Sprague-Dawley rats. Each rat weighed approximately 250 g at the time of muscle injury. Before injury induction, the rats were anesthetized with ethyl ether. If a withdrawal reflex occurred, an additional anesthetic was administered until no response was noted. Muscle contusion was then induced without a skin incision by dropping a 115 g weight from a height of 1 m onto an impactor placed on the right medial paravertebral musculature (Fig. 1). Fluoro-Gold (FG; Fluorochrome, Denver, CO, USA), a retrograde neuronal tracer, was injected into both sides of the paravertebral muscle to label afferent sensory nerves. Histology The muscle injured site (i.e., the right side) and uninjured site (i.e., the left side) were dissected from the back muscle under anesthesia with sodium pentobarbital (40 mg/kg, intraperitoneally) at 1, 2, and 3 weeks (12 rats, four rats per time period) after the muscle injury, and the sites were transcardially perfused with 0.9% saline, and subsequently with 500 mL of 4% paraformaldehyde in a phosphate buffer solution (PBS; 0.1 M, pH 7.4). Each formalin fixed tissue specimen was embedded in paraffin after dehydration for 14 hours using an ascending series of ethanol concentrations in Tissue-Tek VIP (M1500, Sakura Finetek Japan Co., Ltd., Tokyo, Japan). Subsequently, 4-µm-thick sections were made from these paraffin blocks using a sliding microtome (LS113, Yamato Kohki Industrial Co., Ltd., Saitama, Japan), following which they were placed on glass slides (#5116, Muto Pure Chemicals Co., Ltd., Tokyo, Japan). After deparaffinization with xylene and ethanol, these sections were stained with hematoxylin and eosin (H&E). The sections were finally overlaid with the Entellan new mounting medium (Merck KGaA, Fig. 1. We defined the right-side back muscle as the injured site and the left-side back muscle as the uninjured site. Under anesthesia, 8-week-old male Sprague-Dawley rats were injured on their right-side back muscle using a 115 g weight. The weight was dropped from a height of 1 m and was controlled by a pipe-shaped device. Asian Spine J 2017;11(1):88-92 Darmstadt, Germany) after dehydration with a graded series of ethanol and xylene concentrations. A professional animal pathologist observed the slides under a microscope (BH20, Olympus Corp., Tokyo, Japan), and the degree of each finding was evaluated semiquantitatively. We evaluated the presence or absence of histological degeneration, bleeding, and neutrophil recruitment for each slice. We also compared the histological changes over time in the injured site against those in the uninjured site. Immunohistochemistry for CGRP The L2 dorsal root ganglia (DRG) of the injured and uninjured sides were removed at 1, 2, and 3 weeks after the contusion injury. According to a previous report, the paravertebral musculature is under a significant degree of L2 control [8]. The proportion of FG-labeled neurons also showing immunoreactivity for CGRP, a marker of inflammatory pain, was then determined. Endogenous tissue peroxidase activity was quenched by soaking sections in 0.3% hydrogen peroxide solution in 0.01 M PBS for 30 minutes. The specimens were then treated for 90 minutes at room temperature in a blocking solution consisting of 0.3% Triton X-100 and 3% skim milk in 0.01 M PBS. The sections were labeled using a primary rabbit antibody to CGRP (Chemicon, Temecula, CA, USA) diluted to 1:1,000 in a blocking solution, and then incubated for 20 hours at 4°C. To detect CGRP immunoreactivity in DRG, the sections were incubated with a goat anti-rabbit Alexa Fluor 488 fluorescent antibody conjugate (1:400, Molecular Probes, Eugene, OR, USA). The sections were examined under a fluorescence microscope, following which the number of FG-labeled, CGRP-immunoreactive (ir), and both FG-labeled and CGRP-ir neurons were determined in 10 randomly selected areas of each DRG section. Statistical analysis We used the t test to compare the proportion of CGRPir FG-positive neurons in DRG at 1, 2, and 3 weeks after muscle injury. p<0.05 was considered significant. Histology Changes in the histology of the muscle samples over time at the injured site are shown in Fig. 2. H&E staining of the paravertebral muscle showed infiltration of the inflammatory cells and the presence of granulation |
tissue in the injured part as well as neovascular hyperplasia on the ipsilateral side compared with the contralateral side 1 week after injury. Moreover, the absorption of degenerated cell tissues and muscle fiber repair were observed 2 weeks after injury. Muscle atrophy occurred 3 weeks after injury; however, the atrophy was repaired via spontaneous replacement of muscle cells or fibers. Discussion In the present study, inflammation at the injured site peaked at 1 week after injury, and scarring developed by the third week. Kawaguchi et al. [9] reported that regeneration began within 1 week and was accomplished within 6 weeks in a back muscle injury model in which a selfretractor was used for 3 hours in rats. However, in our drop mass model, the back muscle was damaged by only a single contusion; therefore, the injured tissue may have improved earlier. In contrast, the pain neuropeptide in the DRG neurons innervating the injured muscle remained at an increased level 3 weeks after injury. In addition, in our previous study using an injured intervertebral disk model in rats, the inflammatory mediator at the injured site decreased to normal levels within 2 weeks; however, the CGRP-ir neurons among the DRG neurons innervating the responsible lesion remained at an increased level for at least 8 weeks [10]. Previously, it has been reported that a decrease in the trunk muscle strength is the main factor for chronic muscle pain [6]. We cannot deny that atrophy of the muscle tissue observed at 3 weeks could initiate postoperative muscle pain. However, our hypothetical inference was that the prolonged sensitization of the neuropeptide in DRG may produce postoperative back muscle pain. Alternatively, scarring in the damaged muscle may be responsible for a certain type of nociceptive signal being sent to DRG at 3 weeks after injury, regardless of the histological completion of local inflammation. Determining the causes for postoperative back muscle pain would require long-term research of the neuropeptide in DRG innervating the injured sites. Regardless, our results indicate that the amount of muscle damage should be minimized during surgery. A study that used a self-retaining retractor in rats reported that 5 min of retraction release every 40 or 60 minutes could prevent severe muscle injury during posterior lumbar spine surgery [11]. In addition, minimally invasive spine surgery techniques have been reviewed as procedures to decrease muscle crush injuries [5]. The present study confirms the aforementioned points. A limitation of our model was that the injury was caused by direct contusion; however, the power of traction and compression causes muscle damage during actual surgery. An intraoperative back muscle injury model should be constructed using retracting devices, such as those used by Kawaguchi et al. [9] in their muscle injury animal model. In addition, we did not quantify inflammation of the injured tissue through a histological method or immunohistochemistry technique. Moreover, a behavioral study should be performed to evaluate muscle pain, as performed by Miyagi et al. [12] and Sakuma et al. [13] using the CatWalk system in which an animal's gait can be analyzed. We are planning to resolve these issues through future research. The Association of Metabolic Syndrome and Its Components with Electrocardiogram Parameters and Abnormalities Among an Iranian Rural Population: The Fasa PERSIAN Cohort Study Background Metabolic syndrome (MetS) as a set of cardiac risk factors and its growing prevalence is one of the major concerns in different societies. In this study, we aimed to investigate the relationship between Mets and electrocardiogram (ECG) parameters and abnormalities as indicators for subclinical cardiovascular diseases (CVD). Methods In this sub-analysis study, we used the data from Fasa PERSIAN Cohort Study which includes subjects age 35–70 years. Subjects with available ECG data included in the study (n=7002) and subjects with missing data on MetS components and non-sinus rhythm ECG were excluded (n=44). The MetS definition based on the Adult Treatment Panel (ATP) III guidelines and also a 12-lead ECG was obtained from all participants. Results Our study population (n=6958) showed a mean age of 48.60±9.34 years and also 1656 (24.2%) subjects had MetS. Except for P duration, PR interval and S amplitude in men and P amplitude, S amplitude, Sokolow-Lyon Index, and QT interval in women, other ECG parameters differ significantly between subjects with and without Mets (P<0.05). Also among ECG abnormalities, prolonged P duration (≥120ms), QRS duration (≥100ms), and QTc interval (>450ms in male, >470ms in female) had a significant association with MetS in the total population. Waist circumferences (WC) showed the most count of significant relationship with ECG parameters in both genders. In males, WC more than ATP cut-points had significant associations with prolonged P and QRS duration, and also blood pressure (BP) had significant associations with prolonged P and QRS durations and QTc interval. In females, the MetS component except triglyceride had at least a significant relationship with prolonged P and/or QRS duration. Conclusion MetS and its component especially WC and BP were associated with ECG parameters and abnormalities. These associations with ECG as a marker of subclinical CVD showed the importance of MetS and each component in our population to monitor in the further longitudinal studies. Introduction The metabolic syndrome (MetS) or X syndrome refers to a set of obesity, hypertension, hyperlipidemia, microalbuminuria, and impaired glucose metabolism risk factors 1 first defined by Reaven in 1988. 2 With a growing prevalence, the MetS affected 10-23% of the world population between 2002 and 2004. 3 From 2003 to 2012, the prevalence of MetS in the United States was approximately 33%. 4 Studies conducted in the Middle East between 2003 and 2014 showed that MetS had a prevalence of about 25% and rapid growth, even more than the American and European populations. 5 The prevalence rate of MetS was more than 30% in different studies in Iran. 6,7 In another definition, this syndrome was defined as a set of cardiac risk factors 8 showing its role in developing cardiovascular disease (CVD). For example, MetS account for 25% of CVDs 9 . On the other hand, it increases the risk of CVD by 61%. 10 Each component of the MetS has a distinct role in the development of CVD; however, the distinct role of each component is greater than in the general. 11 CVD is the leading cause of premature mortality in the world with 17.9 million deaths in 2012, which is predicted to increase to 23 million deaths in 2030. Moreover, the low-and medium-income countries, specifically the Eastern Mediterranean countries, account for 50% of deaths and 80% of the global burden from this disease. The death from CVD in Iran increased from 26.6% in 1981 to 43.3% in 1995. 12 According to the GBD's reports, CVD has been the leading cause of mortality between 2010 and 2015 in Iran, accounting for more than 20% of all diseases and 46% of all deaths. 13 Electrocardiogram (ECG) can be regarded as a simple comprehensive method to check the heart's activity history. As a test with a high prediction value, it can be used to predict the chance of CVD development. 14 It is worth noting that the majority of studies have investigated the effects of MetS only on one or few markers from the patients' electrocardiograms, such as aortic stiffness, 15 T-wave axis, 16,17 resting heart rate, 18 QT interval, ST segment, and Q-wave, 14 or assessed the effect of only one component of the MetS, such as blood glucose, 19,20 obesity, 21 and/or abdominal adiposity 22 on the majority of ECG markers. This suggests that there are few similar studies in the world and no similar study in Iran as comprehensive as the present study. The genetic, cultural, and geographical differences, among many others, justify differences between various studies conducted in different geographic regions. In many areas, the cohort studies on metabolic subjects have investigated the effects of MetS and its components on the ECGs. It is worth noting that some of these studies produced different results, indicating that although a few relevant studies exist in the world, this relationship continues to remain unknown. 14,23 Besides, knowing the effects of MetS and its components on ECG can be helpful for CVD risk prediction, and used by health authorities and decision-makers to change its epidemic spread in Iran. 24 Study Design and Population In this sub-analysis of Fasa Cohort Study, as a part of the PERSIAN cohort study, near 11,100 people aged 35 to 70 years were included 25 Each participant has signed informed consent at the beginning of the study. All subjects with available data of ECGs and without reported or known CVD were included in the study (n=7002). All included subjects were at the same socioeconomic level, with the same ethnic and same residential region. Also, subjects with detected non-sinus rhythm ECG and missing data on Mets components haves been excluded (n=44,) and finally 6958 subjects remained in the study. Characteristics of Subjects Basic data including age, sex, smoking status, alcohol consumption, and chronic diseases such as diabetes and hypertension were questioned and recorded by an internetbased questionnaire. All medication that was taken within 2 weeks before registration has been questioned and recorded. A list of drugs (selected medications) which has been provided by CredibleMeds, as an American organization, including drugs with known risk and drugs with a possible risk for Torsades de pointes, which counts 187 cardiac and non-cardiac drugs in total was used to recognize the drugs which have been known as effective factors in QT interval prolongation. 26 Physical activity was measured by a 20-item questionnaire that can measure routine physical activities of rural Iranians. The amount of each activity in hours and minutes was determined; the MET-value of each activity was multiplied by its duration, and MET-min of each activity was calculated. Finally, the sum of all activities was calculated as the total physical activity (MET/24h). Measurements For the anthropometry calculations, height was measured by a stadiometer with an accuracy of 0.1 cm and weight was measured by a digital scale with an accuracy of 0.1 kilograms. Body mass index (BMI) was calculated using weight divided by the square of height (kg/m2). Waist circumference (WC) was measured at the midpoint of the inferior border of the lowest ribs to the anterior superior iliac spine, using an inelastic tape. Hip circumference was considered at the widest part of the participants' hip. Wrist circumference was measured just below the wrist bone. For evaluating blood pressure (BP), participants first rested for 15 minutes, after which with an interval of five minutes, two consecutive BP's were measured from participants and reported by an average of systolic and diastolic pressure in mmHg. Triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL), highdensity lipoprotein cholesterol (HDL), and fasting blood sugar (FBS) all obtained from venous blood samples taken in a 10-14h fasting condition. TC and HDL measures were assayed by using the Mindray BS380 autoanalyzer (Mindray Medical International, Shenzhen, China) as the biochemical tests. Laboratory testing of FBS was done using glucose oxidize test. We used the Friedewald formula to calculate LDL. 27 Electrocardiogram A 12-Lead ECG was obtained from the participants using a computer-based paperless device (Cardiax ®28 ) with a 2000 Hz sampling rate and 0.04 µV/bit (24-bit resolution). Patients were in the supine position 15 minutes before recording and were told to relax, breathe normally, refrain from moving and talking, but remain awake during the procedure. As this device provide us a more detailed value of the duration, amplitude, and axis of the ECG waves, we used ECG parameters such as Heart rate (bpm), P duration (ms), P amplitude II (mV), PR interval (ms), QRS duration (ms), R amplitude V5 (mV), S amplitude V1 (mV), QT interval (ms), QTc interval (ms), P axis (°) and QRS axis (°). All ECGs analyzed and reported Automatically by Cardiax software (version 3.50.2, International Medical Equipment Developing Co. Ltd., Budapest, Hungary) and exported to central data software. Also, prolonged PR interval was considered as PR interval≥ 200 ms, prolonged P duration was considered as P duration≥ 120 ms, prolonged QRS duration was considered as QRS duration ≥ 100 ms. Bazett's formula 29 was used to calculate heart-rate corrected QT interval (QTc) also we used clinical standard cut-offs for prolonged QTc interval (Man>450ms, Woman>470). Moreover, Sokolow-Lyon index was calculated as the sum of S amplitude V1 and R amplitude |
V5/6 30 Metabolic Syndrome Definition The MetS were assumed to be present provided that three or more of the following parameters are met, based on the NCEP ATP III guidelines: 31 1. The WC more than 94 cm for men and more than 80 cm for women 2. HDL less than 40 mg/dL in men and less than 50 mg/dL in women or receiving drug therapy for reduced HDL 3. TG level ≥ 150 mg/dL or receiving drug therapy for hypertriglyceridemia 4. BP ≥ 130/85 mmHg or receiving blood pressurelowering drugs 5. FBS ≥ 101 mg/dL or receiving glucose-lowering drugs Data Analysis Descriptive statistics were reported as number (percentage) or mean ± SD. Independent T-test was used to compare quantitative variables between the two groups. Also, both linear and logistic regression analyses performed for multivariate analysis. Regression adjusting was done with variables such as age, heart rate, smoking history, educational years, diabetes history, regular consumption of alcohol, physical activity, and selected medications (just for QT and QTc interval analysis). All the statistical analyses performed in SPSS 22.0 (IBM Corp., Armonk, N.Y., USA), Microsoft Excel 2019 software and for graphs, we used Prism version 8.00 (GraphPad Software, La Jolla California USA). Also, P-value<0.05 considered as statistically significant P-value. Ethics Approval Our study was following relevant guidelines and regulations of our regional and national research ethics committees, also the protocol of this study has the approval of regional and national research ethics committees (the equivalent of institutional review boards) of FUMS (reference number: IR.FUMS.REC. 1399.019). Results A total of 6958 participants, including 3001 (43.13%) male and 3957 (56.87%) female with a mean of age 48.60±9.34 years in total population were analyzed. DovePress women (80%) and being smoker and regular consumption of alcohol was higher in subjects without MetS. In Table 2, As can be seen, male participants with MetS had a significantly higher heart rate, longer QRS duration, and QTc interval (p<0.001); whereas, their P amplitude II, R amplitude V 5 , Sokolow-Lyon Index, QT interval, P axis, and QRS axis were lower (p<0.001). Among the female participants, those with MetS had a significantly higher heart rate (p<0.001), longer P duration (p=0.002), longer QRS duration (p<0.001), longer QTc interval (p<0.001), lower P axis (p=0.017), and lower QRS axis (p<0.001). Among the total population, 82 (1.1%) of subjects had prolonged PR interval, 670 (9.6%) had prolonged P duration, 2504 (35.98%) had Prolonged QRS duration, 691 (9.93%) had prolonged QTc interval. ECG abnormalities were more frequent in males. According to MetS, significantly higher frequency observed in prolonged P duration and prolonged QTc interval in males and prolonged P duration, QRS duration, and QTc interval in females (p<0.001). Table S1 shows the association between EGC parameters and abnormalities with MetS in the males and females. The highest positive beta-coefficient and 95% CI was related to heart rate and the highest negative was related to the QRS axis in males and P axis in females. Among ECG abnormalities prolonged P and QRS durations odds ratios (95% CI) respectively were 1.561 (1.112-2.191) and 1.394 (1.095-1.774) with P-value<0.05 in male. In the female gender, the both mentioned abnormalities were still significant, also in addition to these abnormalities, prolonged QTc interval was in a significant association with MetS(P-value=0.003) in females. Table 3 represents the results of the linear regression model of the association of MetS components with ECG parameters in both genders. Among MetS components, WC was significantly related to all ECG parameters in males. In females, just the QT interval did not show any significant association with WC. In both genders, it seems that FBS had the lowest count of significant relationships but the highest positive relationship in females was related to FBS and heart rate and in males was related to SBP and Sokolow-Lyon Index with a standardized beta coefficient 0.189 and 0.182 (P-value<0.001) respectively. Also, the highest negative relationship was related to WC and P axis with −0.268 and WC and R amplitude V5 with −0.226 (P-value<0.001), respectively in males and females. Also, Table S2 represents the results of the linear regression model of MetS components as binary outcomes with ECG parameters in both genders. Figure 1 represents a comparison of ECG parameters' means between different MetS scores in the total population. Overall, a significant increase of means was seen in multiple ECG parameters including heart rate, P duration, QRS duration, QTc interval, and P amplitude II. Also, a significant decrease of means was seen in the QT interval, P axis, QRS axis, R amplitude V 5, and Sokolow-Lyon Index. In ECG abnormalities, in males, just three of the MetS components including WC, BP, and HDL were significantly associated with prolonged P duration with odds ratios (95% CI), 1.849 (1.352-2.528), 1.323 (1.006-1.739) and 1.286 (1.013-1.631) respectively. Also, WC was related to prolonged QRS duration with 1.552 (1.231-1.956) and P-value<0.001, and BP was related to prolonged QT interval with 1.421 (1.067-1.893) and prolonged QRS duration with 1.215 (1.011-1.459) P-value<0.05. But in females, just prolonged P duration and prolonged QRS duration were in significant associations with MetS components, prolonged P duration with WC, FBS and BP, and prolonged QRS duration with WC, BP, and HDL. Among these variables, the highest OR was related to BP and prolonged P duration with 1.758 (1.392-2.221) and P-value<0.001 (Table 4). According to Table 5, it seems that WC and BP are the two MetS components (as binary outcomes) which have significant associations with having single or multiple ECG abnormalities. In male, WC with odds ratios (95% CI) 1.381 (1.095-1.742) and 1.571 (1.149-2.147) was related to having single or multiple ECG abnormalities (P-value<0.05) respectively. In females, such relationships observed with smaller odds ratios. Also, BP with an odds ratio (95% CI), 1.628 (1.265-2.094) and P-value<0.001, was related to having multiple ECG abnormalities in males but there was not any significant association between BP and having single abnormality (P-value=0.468) while in females it was significant with 1.356 (1.169-1.573) and P-value<0.001. In total, 707 (35.8%) of subjects with MetS score of 1, 693 (35.5%) of subjects with MetS score of 2, 445 (41%) of subjects with MetS score of 3, 167 (37.0%) of subjects with MetS score of 4, 49 (40.8%) of subjects with MetS score of 4 had one ECG abnormality which was an insignificant difference (P=0.021). Also, 171 (8.6%) of subjects with MetS score of 1, 183 (9.4%) of subjects with MetS score of 2, 108 (10%) of subjects with MetS score of 3, 60 (13.3%) of subjects with MetS score of 4, 24 (20%) of subjects with MetS score of 4 had multiple ECG abnormalities which were insignificant difference (P<0.001). ECG abnormalities' frequency in different MetS scores has been Notes: Data are presented as standardized beta coefficient P-value . The analysis was adjusted for age, heart rate, smoking history, educational years, diabetes history, regular consumption of alcohol, and physical activity. *P duration was added to analysis plus previously mentioned adjusted variables **QRS duration and selected medications were added to analysis plus previously mentioned adjusted variables. Significant results are bolded. P-value reference: a <0. presented in Figure 2 according to gender. Significant differences among different MetS scores in men were related to the frequency of multiple ECG abnormalities while both single/multiple ECG abnormalities showed significant differences in women (P<0.001). Also, Table S3 shows the odds ratios of having different MetS scores with Figure 1 Comparison of ECG parameters' means in different metabolic syndrome scores in the total population. a: significant difference to subjects with score 1, b: significant difference to subjects with score 2, c: significant difference to subjects with score 3, d: significant difference to subjects with score 4, e: significant difference to subjects with score 5. Abbreviations: MetS, Metabolic syndrome. Discussion In this study with a large number of Iranian population main findings of the study were i) In overall, the mean of Heart rate, P duration, PR Interval, QRS duration S amplitude, and QTc interval was higher in subjects with MetS but the mean of P amplitude, R amplitude, Sokolow-Lyon Index, QT interval, P axis, and QRS axis was lower in the subject with MetS. ii) ECG abnormalities including prolonged PR interval, P duration, QRS duration, and QTc interval frequencies were higher in subjects with metabolic syndrome. However, just prolonged P duration and QTc interval in men and prolonged P duration, QRS duration, and QTc interval in women were at a significant difference. iii) The MetS were associated with prolonged P duration and QTc interval in males. This association in female subjects was seen with all ECG abnormalities except prolonged PR interval. iv) Among MetS components, WC had the highest count of significant relationship with ECG parameters in both genders, and the highest positive relationship in females was related to FBS and heart rate and in males was related to SBP and Sokolow-Lyon Index. Also, the highest negative relationship was related to WC and P axis and R amplitude V5 respectively in males and females. v) Among relationships between MetS components and ECG abnormalities, WC had significant associations with prolonged P and QRS duration and BP had significant associations with prolonged P and QRS duration and QTc interval in males. Also, prolonged P and HDL had a significant association. In females, the MetS component except TG had at least a significant relationship with prolonged P and/or QRS duration. vi) Considering ATP cut-points for MetS components, Subjects with WC and BP more than cut-points seem to be more prone to have single/multiple ECG abnormalities which were higher in males than females. Also, MetS may increase the frequency of having multiple ECG abnormalities near 80% and 30% respectively in men and women significantly. It should be mentioned that these findings were independent of other risk factors such as age, heart rate, smoking history, educational years, diabetes history, regular consumption of alcohol, and physical activity. Metabolic Syndrome and ECG Parameters Our results showed that it seems MetS is in a significant positive association with heart rate as first ranked and following by QRS duration in men and QTc interval in women. Queen et al, 14 reported that the odds ratio in having MetS in subjects with HR more than 72 bpm in comparison to those with HR lower than 65.6 bpm is 2.38. They have also reported that the odds ratio in having MetS in subjects with QTcB more than 422 ms in comparison to those with QTc lower than 396 ms is 1.69. However, they had a different study design but it supports our result in the case of association of HR, QTc, and MetS. In a recent study with a smaller sample size, the PR interval had the strongest association with metabolic syndrome. 21 Surprisingly PR interval in our study was not at a significant relationship with metabolic syndrome. A cross-sectional analysis in the Netherlands 23 reported the mean of P and QRS axis 44.2 and 23.0 respectively in non-obese subjects which were near our results. Also, the same as our results, they observed a negative significant relationship between P and QRS axis with metabolic syndrome. In our study P axis was the first and QRS was the second parameter in having the highest negative relationship with MetS in women while vice versa was seen in men. A few studies are focusing on different ECG wave amplitudes with metabolic syndrome. A study that has been performed just on men, reported that a lower mean of P, R, and S amplitudes in subjects with MetS. 32 Among these amplitudes, just R amplitude was in the significant differences between the two groups. Our findings also have shown almost the same results. Metabolic Syndrome and ECG Abnormalities Our findings suggested that prolonged P duration in men and prolonged QRS in women among ECG abnormalities had the strongest significant association with metabolic syndrome. There was just another significant association in men among ECG abnormalities which were related to prolonged QRS duration. Also, the three statistically significant associations in women from highest to lowest association respectively were prolonged QRS duration, prolonged QT interval, and prolonged P duration. Ebong et al, 33 in a study on 6765 subjects aged from 45-84, reported that among those with MetS, 65.1% of men and 50% of women had abnormalities in their ECG. Although we had very different criteria for ECG abnormalities, in our study 61.8% of men and |
49.9% of women which was near the previously reported results. Also, having multiple ECG abnormalities seems to have a higher significant association than having one ECG abnormality in both genders. In men, MetS seem to prone subjects to have multiple ECG abnormalities more than 78% as subjects without MetS. Also, female subject with Mets had a higher significant association (near more than 35%) with having one ECG abnormality which did not apply to men. Each Metabolic Syndrome Component and ECG Parameters Waist circumference in men was at a significant positive association with all ECG parameters except P and R amplitude, Sokolow-Lyon index, P, and Q axis. The same relations were seen in females except for QT and QTc intervals and waist circumference which was not at a significant association which may be due to heart rate variability. Elffers et al, 22 reported that waist circumference positively was associated with QRS duration, P duration, and negatively was associated with the QRS axis and without any associations with the P axis. Their results support our findings however in the case of the P axis there was a negative association in our study. FBS in men was only associated with heart rate, R amplitude, and QRS axis. In women, FBS had a significant relation with heart rate, P, and R amplitudes. A few studies are focusing on blood glucose and ECG parameters but Paudyal et al on 100 participate, indicated a possible association between Impaired and left axis deviation. 34 Our negative association between FBS and QRS axis can indicate the same somehow but, this relationship was not significant. Another study reported a slight increase in PR interval in subjects with higher FBS. 35 We observed this association in both genders however it was one of the smallest and again insignificant. Both Systolic and diastolic BP in men were associated with ECG parameters except P duration and PR interval. Systolic BP also did not show a significant relationship with the QT interval. These exceptions in women were P amplitude. Also, Systolic BP did not show a significant relationship with the PR interval in addition to the previous one. Alonso et al, on 3180 participate from the MESA study suggested that Systolic and diastolic BP were not strongly associated with PR interval or P duration which can support our results in men. 36 In other studies, with focusing on QT interval and QRS duration, the authors observed that central aortic blood pressure can be a risk factor for QTc prolongation and longer QRS durations. 37,38 It seems BP after WC had the highest count of significant relationships especially diastolic blood pressure in both genders. TG in our study was the most different variable in having a significant relationship with ECG parameters between men and women. In men, TG was related to ECG parameters except for QTc interval and S amplitude but in women, it had significant associations with heart rate, R amplitudes, Sokolow-Lyon index, QT interval, P and QRS axis. Although we did not find any correlation between TG and PR interval in women and a slight significant relationship in men but Adegoke et al, 39 reported that triglyceride had a positive correlation with the PR interval (r = 0.3). This difference may be due to their different population and age because they have studied children with sickle cell anemia. In another study on 69 men, 32 TG was related to the P axis and QRS axis negatively with r=−0.374 and −0.363 respectively which was higher than our result which may be due to their small sample size and unadjusted results. HDL as another lipid profile among MetS components was at a significant association with heart rate, P amplitudes, QRS duration, R and S amplitudes, Sokolow-Lyon index, P and QRS axis in men. In women, P, R and S amplitudes, Sokolow-Lyon index, P axis, and QTc interval indicated a significant relationship with HDL. A Previous populationbased study reported a J-shaped association between QTc interval and HDL 40 which can be as a support for our results. In another study, 32 HDL was related to the P axis and QRS axis positively with r=0.426 and 0.219 respectively which was much higher in comparison with our result. Also, it should be mentioned that in their study HDL and QRS axis were not statistically significant while our results showed a significant positive relationship in men. Each Metabolic Syndrome Component and ECG Abnormalities In subjects with WC and BP more than ATP cut-points, it seems the chance of having prolonged P and/or QRS duration may be significantly higher. Moreover, HDL and BP had significant relationships with prolonged P duration and prolonged QTc interval. In males, WC had a higher value in comparison with BP and HDL in having ECG abnormality. In females, BP increases the frequency of prolonged P duration near 75% and increases the frequency of prolonged QRS duration near 50%. Interestingly HDL in our study show just a significant association in subjects without ECG abnormalities which can represent its protective effect on ECG abnormalities. Overall, it can be interpreted that MetS can increase the chance of having single or multiple ECG abnormalities, especially multiple ECG abnormalities in men. Possible Mechanism There are several possible mechanisms of components of MetS and electrical status and heart's structure which may manifest as ECG abnormalities. These effects of MetS on ECG variables may be direct, indirect, or interrelated with each other. Waist circumference as a representative of central obesity could result in increased sympathetic activity, diaphragm elevation, and increased cardiac output. 41,42 Hypertension also can result in elevated cardiac output and this elevation may cause increased left ventricular (LV) wall stress, which stimulates myocardial hypertrophy. 43 Also, the interrelation between obesity with an increased predisposition to hypertension is another point. Other studies have suggested leftward shifts of P and QRS axis were related to obesity. 44,45 Obesity also may increase cardiac loading and lead to remodeling of the heart muscle and finally PR prolongation. These effects of obesity may be due to hormones produced by the adipose tissue which can result in electrophysiological changes. 46 Higher FBS levels are associated with hypertension, hyperlipidemia, and a prothrombotic state which can interact synergistically to increase cardiac and ECG changes. 47 Also, as it is known atherosclerosis can increase myocardial ischemia which can lead to scarring and heart dysfunction and ultimately ECG changes. 43 Strengths and Limitations A strength in our study was our large study population (n=6960). Data collection of digital ECGs were done under highly standardized conditions and enabled us to report more detailed ECG data. Another strength was the extensive measurements of confounding factors in the Fasa PERSIAN Cohort Study which made us examine the relationship of MetS and its component with ECG parameters more accurately and independent from effective factors. In this study, we used ECG as a predictor for future CVD risks and examined the role of MetS and its components separately for better screening. Also, to our knowledge, this is the first study that aimed to investigate the relationship between metabolic syndrome, its components, and ECG parameters in an Iranian population. Our study is not free from limitations. First, as this study has a cross-sectional design, it limits our ability to infer any causality. For sure longer follow-ups in our study and other populations can reveal more results in case of these relationships. Second, our study was based on a rural population. As the subject we assessed may differ from a rural to urban subjects and the modernized urban population increasing in both developed or developing countries, it should be investigated in further studies with an urban Iranian population. Third, our study subjects were aged from 35-70 and we may not apply our findings to a younger Iranian population. Conclusion In conclusion, our study indicated that MetS and its components may have several different relationships on ECG parameters or abnormalities, and also, we suggest WC and BP as the two MetS components which had more and stronger associations with ECG parameters or abnormalities. Our results support this idea that MetS as an effective role player in ECG which is representative of subclinical CVD should be considered and screened to maintain subjects in a healthier lifestyle. Data Sharing Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request to the corresponding author. Ethics Approval and Consent to Participate The study protocol was following the Helsinki Declaration and confirmed by the Ethics Committee of Fasa University of Medical Sciences (Approval Code: IR.FUMS.REC. 1399.019). The participants were informed about the research objectives and the written informed consent was obtained from the subjects before starting the survey. Co-Administration of Curcumin and Bromocriptine Nano-liposomes for Induction of Apoptosis in Lung Cancer Cells Background: In recent years, nanotechnology with modern advances in the macromolecular design of nano-carriers has proved to be helpful in the development of drugs delivery systems. This research represents a novel co-administration of nano-vehicles, known as liposomes. Liposomes efficiently encapsulate curcumin and BR in a polymer structure, which results in enhanced aqueous solubility of the mentioned hydrophobic agents and higher bioavailability of the drugs. Methods: Preparation of curcumin and BR liposomes were carried out by the thin film method, and the amounts of purified drug and its release were analyzed. After dose determination, the human lung cancer cells (QU-DB) were exposed to BR and curcumin liposomes for 12, 24, and 48 h. Then the viability and apoptosis assays were carried out by using MTT and flow cytometry technique, respectively. Results: In this research, in vitro anti-cancer effects of former nano-formulations on lung cancer cells was confirmed, and no cytotoxicity effects of these nano-preparations were observed in the normal cells (HFLF-PI5). Conclusion: Our findings suggest the nano-liposomal drugs as effective anti-cancer agents; however, additional clinical examinations are required. INTRODUCTION he most common cancer in the world is lung cancer [1,2] . Despite numerous developments in surgery, chemotherapy, and radiotherapy over the past decades, the fatality rate of lung cancer has still continued largely unaffected, which is mostly because of its metastasis [3] . New treatment approaches for lung cancer patients are highly demanded because of the overall poor prognosis. Meanwhile, due to the low bioavailability of curcumin, its biological activity is severely limited, though the curcumin therapeutic index is promising [4] . For handling drug delivery issues, the most important nanoparticle systems are liposomes, polymer conjugates, polymeric micelles, dendrimers, nanoshells, proteins, and nucleic acid-based nanoparticles. Among these tools, liposomes and polymer-based nano-formulations create a common nanoparticle therapeutic agent, available for the clinical use [5] . Liposomes are artificially constructed vesicles consisting of a phospholipid bilayer and have the T Iran. Biomed. J. 24 (1): 24-29 25 ability to adjust the bio-distribution of drugs [6] . The current trials for various liposome constructions and a large number of commercially presented therapeutics seem to be promising [7] . The hydrophobic properties of curcumin has made it as a good candidate for liposome integration and encapsulation in the lipid layer of the liposomes [8] . Studies have suggested that the integration of curcumin into liposomes significantly augmentes its bio-availability. Moreover, the activity of liposomal curcumin is more favorable than that of curcumin alone [9] . A primary therapeutic drug for adenomas is BR. BR binds to the dopamine D2 receptors on pituitary epithelial cells to prevent prolactin secretion. Nowadays, BR is used to treat various diseases such as Parkinson, acromegaly, addiction, hyperprolactinemia, high growth hormone-secreting disorders, cyclical mastalgia, and type 2 diabetes [6] . BR stimulates dopamine receptors and has anti-mitotic and anti-tumor properties. D2 receptor mRNA has been identified in all BR-sensitive tumors [5,10] . The results of a previous research showed that significant changes occur in the expression of D2-like dopamine gene receptor in NSCLC [11] . In the present research, in vitro apoptosis occurrence by BR was examined in a lung adenocarcinoma cell line. The primary goal of this study was to estimate the possible curcumin and BR partitioning into the liposomes and adjustment of curcumin and BR encapsulation in liposomes. Also, we tried to assess the effect of curcumin-loaded and BR-loaded liposomes on lung cancer cells. To achieve this goal, DLS examination, anti-proliferative effects study using a MTT-based assay, and flow cytometry were applied. Chemicals and cells Curcumin and |
BR were obtained as a gift from Genetics Department of Tarbiat Modares University (Tehran, Iran). Soybean phospholipids with 75% phosphatidyl-choline, 2-distearoyl-sn-glycero-3phosphoethanol-amine, and sodium salt (DSPE-mPEG-2000) were procured from Lipoid GmbH (Ludwigshafen, Germany). Cholesterol was purchased from Sigma-Aldrich (St. Louis, MO, USA). All other chemicals and solvents used in this investigation were of analytical grade. Human lung carcinoma cell line, QU_DB, and human lung cell line, HFLF-PI5, (as a control) were provided from the National Cell Bank of Pasteur Institute of Iran, Tehran. Preparation and characterization Curcumin and BR liposomes were prepared by the thin film hydration-sonication method. Hydration was performed with 1300 μL deionized water at 65 °C for 60 minutes, using a rotary instrument (Heidolph, Germany). The resulting vesicles were then sonicated to reduce the mean size. The size distribution and polydispersity index of the drug-loaded liposome were evaluated using the DLS technique. All the measurements were carried out in triplicates at room temperature, and their mean value was reported. Drug entrapment efficiency Free unentrapped drugs were separated from drugloaded liposomes using the dialysis membrane (cut off: 12-14 kDa). After digesting the liposomal vesicles with isopropanol (99% purity), the amounts of liposomal drug entrapped were analyzed by a UV spectrophotometer at 422 nm [12] . Release assay The release of drug from liposomes against PBS was monitored by dialysis at pH 7 at 37 °C for 96 hours. The calculation of drug release was performed at different times in the PBS solution. Further, drug release calculation was conducted using a calibration curve by a UV spectrophotometer. Flow cytometry The QU-DB cells were treated with different concentrations of BR (12.5-25 µM) and curcumin liposomes (12-20 µm) for 12, 24, and 48 h, and then Annexin-V-Fluos staining kit (Roche, Germany) was used for detection of apoptosis [14] . The treated cell pellets were resuspended in 100 μl of Annexin-V-Fluos solution, and after 15-min incubation at 25 °C, they were analyzed on a FACSCalibur machine (BD Biosciences, CA, USA). Nanoparticle size The DLS of liposome-curcumin is presented in Figure 1. Around 20% of the particles are in the range of 100 nm. More analysis indicating that the remaining particles were about 80 nm in size. Figure 2 illustrates the intensity of liposome-BR, where the diameters of nanoparticles are less than 50 nm. Figure 3 demonstrate the rate of drug release from liposomes. As revealed in the Figure, the extent of drug release of liposome-BR nanoparticles is greater than that of liposome-curcumin nanoparticles at the same time. In particular, during 10 h, 60% of the liposome-BR nanoparticles and only 20% of the liposome-curcumin nanoparticles demonstrated the drug release. Drug entrapment efficiency and stability test Entrapment efficiencies were calculated as 90.89% and 80.41% for the liposome-curcumin and liposome-BR nanoparticles, respectively. Also, the stability of the liposomal suspensions was evaluated after three months of storage at room temperature at 4 °C. The particle size supply, morphology, and drug encapsulation efficiencies of the samples were determined as a function of the storage time. The results indicated that only 15% of the liposomal suspensions efficacy was diminished. The effect of prepared liposomes on cell viability Different concentrations of BR (12.5-25 µM) and curcumin (12-20 µM) liposomes were applied on the QU-DB cells in triplicate for 12, 24, and 48 h, and MTT assay was performed after the treatment. According to Figure 4, the proliferation of cancer cells reduced significantly at the concentrations of 20 to 25 µM for BR and at 16 to 20 µM for curcumin. In addition, the results of our study indicated that BR liposome had a greater antiproliferative effect than curcumin liposome, within the optimum time of 24 hours. Flow cytometry analysis of apoptotic cancer cells The maximum number of apoptotic cells (27.82%) was detected at the concentration of 20 µM for curcumin liposomes, 47.30% at the concentration of 25 µM of BR liposome, and 49.72% in their coadministration by reducing the number of necrotic cells. Data are shown in Figure 5. DISCUSSION The delivery of chemotherapy agents to solid tumors and raising their bioavailability have been a key challenge in the recent biomedical research. Medical imaging and targeted drug delivery using nanotechnology-based tools are rapidly growing to repond this challenge [15,16] . The liposomal curcumin prompted in vitro apoptosis of human pancreatic cells and down-regulated NF-kB machinery. The results of a study on A549/DDP multidrug-resistant human lung adenocarcinoma cells showed that curcumin stimulates apoptosis through a miRNA signaling pathway [17] . Peng et al. [18] have reported increased cell apoptosis via Akt-Bad signaling pathway in U2OS cells by curcumin-loaded nanoparticles. Other researchers have used a combination of 2-hydroxypropyl-γ-cyclodextrin and liposomes to enhance the curcumin bioavailability and aqueous solubility and encapsulation efficiency, in comparison to the sole use of liposome as the drug delivery vehicle. Their results indicated both in vivo and in vitro anticancer capacity for 2-hydroxypropyl-γcyclodextrin/curcumin-liposome complex against KHOS cell line [19] . Koshkina et al. [20] employed 9nitrocamptothecin liposome aerosol against osteo-sarcoma lung metastases in mice. Their results showed a significant decrease in the number of tumor foci and the size of tumor nodules in the lung. An early investigation has shown the usefulness of dopaminergic agonists in treatment of lung cancer [21] . BR acts mainly via D2 receptors, through binding to to adenylyl cyclase and reducing intracellular cAMP. By suppressing the cAMP levels, peptide secretion would be inhibited in a dose-dependent manner [22] . More studies have investigated the effects of dopamine neurotransmitter agonists on cancer cells [23,24] . Previous findings have highlighted a quantitatively significant difference for D2-like dopamine receptor genes expression in the NSCLC, among all kinds of dopamine receptor genes [11,14,25] . Such significant changes could be used to diagnose, treat, and monitor NSCLC [11] . In supplementary studies, cell proliferation and of D2 receptors expression studies were performed before and after treating cancer cells by BR [11,14,25] . It has also been discovered that BR-induced apoptosis in lung carcinoma cells, by activating D2 dopamine receptors and plasma membrane changes, occurs in the early stages of apoptosis [14,25] . Fadok et al. [26] have found that during the expansion of apoptosis, macrophages specifically recognize PS exposed to the surface of lymphocytes. In this case, phagocytosis of cells and apoptotic bodies are performed, and the organisms are protected from inflammation, leading to the exposure of cellular compositions. Therefore, according to the results of our previous studies, we designed and constructed two nanoliposomes, which efficiently were encapsulated in a polymer structure [11,14] . This design led to the enhanced aqueous solubility of the mentioned hydrophobic agents and the bioavailability of drugs. In this research, anti-cancer potency of nanoformulations without cytotoxicity effects on normal cells was confirmed by co-administration of curcumin and BR nano-liposomes in lung cancer cells. Moreover, BR can be suggested as a valuable agent for use in nano-liposomal drugs for medical management of lung cancer. The incidence of cutaneous squamous cell carcinoma in patients receiving voriconazole therapy for chronic pulmonary aspergillosis Voriconazole has been associated with cutaneous squamous cell carcinoma (cSCC) in transplant patients but less is known about the risk in less severely immunosuppressed patients. Our aim was to estimate the incidence of cSCC after voriconazole exposure in patients with chronic pulmonary aspergillosis on a background of chronic lung disease. The notes of patients seen at a tertiary referral centre from 2009 to 2019 with chronic pulmonary aspergillosis were reviewed for the diagnosis of cSCC and voriconazole use documented. Among 1111 patients, 668 (60.1%) received voriconazole for longer than 28 days. Twelve patients received a diagnosis of cSCC; nine had used voriconazole. Mean duration of voriconazole use was 36.7 months. The crude incidence rate was 4.88 in 1000 person/years in those who had voriconazole and 2.79 in 1000 patient/years in those who did not receive voriconazole for longer than 28 days. On Cox regression, age (HR 1.09, 95% CI 1.02–1.16, p = 0.01) and male gender (HR 3.97, 95% CI 0.84–18.90, p = 0.082) were associated with cSCC. Voriconazole use was associated with a slightly increased risk, which was not significant (HR 1.35, 95% CI 0.35–5.20, p = 0.659). Voriconazole use beyond 28 days did not lead to a significantly increased risk of cSCC in a large cohort of patients with chronic pulmonary aspergillosis. Introduction The spectrum of disease caused by Aspergillus encompasses allergic, chronic, saprophytic and invasive disease, depending on the degree of host immunosuppression. Chronic pulmonary aspergillosis (CPA) affects immunocompetent or mildly immunosuppressed patients with chronic lung disease, such as COPD, bronchiectasis, prior pulmonary TB or sarcoidosis (Smith and Denning 2011). Voriconazole is one of the firstline agents in CPA and is often used long-term. Guidelines suggest at least six months of treatment; however, in practice it may be used for longer durations, often for years, as response may be delayed and relapses are common (Denning et al. 2016). Long-term voriconazole treatment is associated with hepatotoxicity, neurotoxicity, photosensitivity and, importantly, a risk of skin cancer. Photosensitivity is common and there is a clear association between its emergence and subsequent development of skin cancer (Cowen et al. 2010). It is hypothesised that voriconazole N-oxide, the primary metabolite of voriconazole and a UVB biproduct, potentiates UVAmediated oxidative DNA damage or inhibits DNA repair (Ona and Oh 2015). Photosensitivity tends to resolve after cessation of the drug; it is not known if this is true of the skin cancer risk. The association between voriconazole and cutaneous malignancy, particularly cutaneous squamous cell carcinoma (cSCC), is well established in severely immunocompromised patients like haematopoietic stem cell transplant (HSCT) or lung transplant recipients, with case reports in patients with renal transplant or HIV (Brunel et al. 2008;Hamandi et al. 2018;Kolaitis et al. 2017;Vanacker et al. 2008;Wojenski et al. 2015). These patients are already at increased risk of non-melanoma skin cancer and voriconazole was shown to increase this risk. Therefore, long-term voriconazole use should be pursued only if the benefit outweighs the risk, and only if frequent skin inspection is possible. Whereas extensive data are available for the risk in the severely immunocompromised patients, the incidence, risk factors and clinical presentation of cSCC in other patient groups treated with voriconazole is not well described. At the National Aspergillosis Centre (NAC), more than 120 patients with CPA are seen yearly and a substantial number will be treated with voriconazole as first-line or salvage therapy, often for longer than 6 months. This large cohort of patients can provide valuable insights into the risk of cSCC in a population other than transplant recipients. The aim of this study was to estimate the incidence of cSCC after voriconazole exposure in a population without severe immune compromise. Methods All patients with the diagnosis of CPA referred to the NAC (2009-March 2019) were included retrospectively. The start of observation was taken as the day voriconazole was started for patients who received voriconazole, and the first attendance in clinic for patients who did not receive voriconazole. Information was obtained from the medical notes for all subsequent visits until March 2019 unless the patient was discharged, lost to follow-up or died, in which case the last observation point was the last clinic visit for patients who did not develop cSCC and the time of diagnosis of cSCC in those who did. Voriconazole exposure was documented as a binary variable; patients receiving voriconazole for less than 28 days were considered non-exposed. We used this cut-off based on the practice to prescribe voriconazole for 28 days initially (as patients have chronic rather than acute disease, and are seen in the outpatient setting). Dosing and frequency of administration were not collected. Age, sex, underlying diseases, photosensitivity, mean of the last five voriconazole levels, location of cSCC, treatment and outcome were recorded for all patients with cSCC. The crude incidence rate and age-specific incidence rate of cSCC in patients who received voriconazole were calculated. Cox regression was performed to compare incidence of cSCC; parameters included were voriconazole use, gender and age. Results A total of 1199 patients with the diagnosis of CPA were identified. Fifty-five had only one attendance with no follow-up and were excluded. For 32 there was no clinical information. One patient had been diagnosed with cSCC after having received voriconazole before referral to NAC and was excluded from further analysis. Of the 1111 patients with follow-up, 668 received voriconazole for 28 days or longer, 120 received |
voriconazole for less than 28 days and 302 never received voriconazole. Mean age was 66.6 years (range 21-96). Female were 475 (42.8%). Mean follow-up duration was 29.3 months (IQR 35.0) for patients who did not have voriconazole and 33.1 months (IQR 42.0) for those who had voriconazole. There were 12 cases of cSCC; nine in patients who received more than 28 days of voriconazole prior to cancer diagnosis (Table 1). All patients were white. All nine patients developed photosensitivity prior to cSCC diagnosis. The mean duration of voriconazole therapy was 36.7 months and the mean time from the start of therapy to cancer diagnosis was 47.2 months. The mean age of patients with cSCC was 65.6 years (SD 12.1) and 10 (83.3%) were male. All were white. The crude incidence rate was 4.88 per 1000 person/years in patients who received voriconazole for longer than 28 days and 2.79 per 1000 person/years in patients who did not receive voriconazole for longer than 28 days. Age-adjusted incidence rates on voriconazole were 13.3 per 1000 person/years in those aged > 74, 5.24 per 1000 person/years in those aged 65-74 and 2.33 per 1000 person/years on those younger than 65. On Cox regression, age (p = 0.01, HR 1.09, 95% CI 1.02-1.16) was significantly associated with risk of cSCC. Male gender (p = 0.082, HR 3.97, 95% CI 0.84-18.90) and voriconazole use (p = 0.659, HR 1.35, 95% CI 0. 35-5.20) were associated with an increased risk of cSCC but did not reach statistical significance. A Kaplan-Meier curve of cSCC risk according to voriconazole status is presented in Fig. 1. Discussion This is the first study, to our knowledge, to estimate the incidence of cSCC in a patient cohort on long-term voriconazole other than HSCT or solid organ transplant recipients. We found a slightly increased risk, although not significantly, of cSCC in patients with CPA treated with voriconazole. Patients with CPA usually have minimal or no immunosuppression; therefore, their risk of skin cancer should be comparable with that of the general population. The number of cases of cSCC observed in this cohort suggests that voriconazole may increase the risk of this cancer in patients without HSCT or solid organ transplant. Half of the patients who developed cSCC had some form of immune compromise, ranging from mild (e.g. 5 mg of prednisolone) to more extensive (e.g. previous chemotherapy for lung cancer). Therefore, it is possible that the risk of cSCC with voriconazole may be increased with any form of immunosuppression, not only the profound immunosuppression associated with transplant. The increased cSCC rate we observed may not be entirely due to voriconazole, and could be attributed to our patients' demographics, too, as CPA mainly affects middle-aged to elderly males, who have a higher incidence of cSCC (Brewster et al. 2007). All patients who developed cSCC after voriconazole use for longer than 28 days were male. cSCC is more common in males, and we observed a striking predominance of males in this series. In addition, chronic inflammation observed in CPA may be another factor predisposing to cancer. Although we did not compare voriconazole levels with those in patients who did not develop cSCC, the mean voriconazole levels were within recommended limits in patients who developed cSCC, suggesting that therapeutic level monitoring is not an effective way to prevent cancer development. cSCC was diagnosed several years after voriconazole initiation in all but one patient, and in no patient was the diagnosis made within 6 months of treatment. This supports the recommendation that voriconazole use for up to 6 months is considered acceptable regarding the malignancy risk. Only one patient (case 8) developed cancer several years after voriconazole was stopped. This suggests that the risk of malignancy is reduced soon after voriconazole is stopped. For the three patients who were treated for only a few days (cases 10-12), cSCC developed several years later; therefore, the association with voriconazole is dubious. In addition, voriconazole-related cSCC described in case reports is multifocal and more aggressive (Vanacker et al. 2008). In our case series, three patients had relapses and one required reconstruction surgery due to extensive disease. Overall, we did not observe a particularly aggressive nature of disease. This study has several limitations. The small number of cSCC cases limits the accuracy of our incidence estimation. We documented voriconazole use as a binary variable and did not correlate duration of treatment with cancer risk. We assumed a duration of less than 28 days as an arbitrary cut-off for long-term voriconazole exposure. This is similar to what was used by Hamandi et al. Using a different cut-off would not have significantly altered the results, as most patients who developed cSCC had either several months or only a few days of treatment (Table 1). In addition, we did not document daily or cumulative doses. The diagnosis of cSCC was done by retrospective review of notes, and therefore some information may have been missing. It is possible that some cases were not recorded by the clinician. Voriconazole use may have led to a more thorough assessment to rule out cancer compared with patients not on voriconazole. Finally, it was not possible to ascertain risk factors such as skin type or sun exposure, as these are not recorded routinely in notes. In summary, in a cohort of patients with CPA, long-term voriconazole use was not associated with a significantly increased risk of cSCC to an extent previously seen in severely immunocompromised patients such as transplant recipients. However, the incidence was higher compared with nonexposed patients. In addition to counselling on the importance of using high factor sunscreen and avoiding sun exposure, regular monitoring for skin toxicity should be undertaken in all patients on voriconazole for longer than six months, particularly when photosensitivity is an issue. Authors' contribution statement Chris Kosmidis conceived project and wrote manuscript, Anna Mackenzie analysed data and wrote manuscript, Chris Harris collected data and revised manuscript, Rola Hashad analysed data and revised manuscript, Fiona Lynch collected data and revised manuscript, David Denning analysed data and revised manuscript. Data availability Not applicable. Compliance with ethical standards Conflict of interest Chris Kosmidis, Anna Mackenzie, Rola Hashad, Chris Harris and Fiona Lynch report no conflicts of interest. DW Denning and family hold Founder shares in F2G Ltd., a University of Manchester spin-out antifungal discovery company. He acts or has recently acted as a consultant to Scynexis, Pulmatrix, Zambon, iCo Therapeutics, Roivant, Biosergen and Fujifilm. In the last 3 years, he has been paid for talks on behalf of Dynamiker, Hikma, Gilead, Merck, Mylan and Pfizer. He is a longstanding member of the Infectious Disease Society of America Aspergillosis Guidelines group, the European Society for Clinical Microbiology and Infectious Diseases Aspergillosis Guidelines group and the British Society for Medical Mycology Standards of Care committee. Ethics approval This study was a retrospective service evaluation and ethical approval was not required, according to UK Health Research Authority guidelines. The study was registered with the audit department of Manchester University NHS Foundation Trust (Audit No. 8056). Consent Informed consent was not required as all data is anonymised. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Distinct Roles of VEGFA and ANGPT2 in Lung Adenocarcinoma and Squamous Cell Carcinoma Background: Vascular endothelial growth factor A (VEGFA) and angiopoietin 2 (ANGPT2) are key mediators in angiogenesis. The expression and clinical significance of VEGFA and ANGPT2 have been investigated in lung cancer, but the results are controversial. The specific roles of VEGFA and ANGPT2 in adenocarcinoma (ADC) and squamous cell carcinoma (SQC) are still not fully understood. To characterize it, we conducted the current study. Materials and Methods: The relationships between clinic-pathological characteristics and the protein expressions of VEGFA and ANGPT2 were analyzed on tissue microarrays by immunohistochemistry (IHC) staining. Then public databases were used to evaluate the association of VEGFA and ANGPT2 mRNA expressions with clinic-pathological parameters and prognosis. Cobalt chloride (CoCl2) was adopted to mimic a hypoxic microenvironment and western blot was used to detect the expression of hypoxia inducible factor-1α (HIF-1α), VEGFA and ANGPT2 in lung cancer cell lines. Results: IHC staining revealed that the expressions of VEGFA and ANGPT2 were enriched in lung cancer tissues compared with normal tissues. Additionally, both VEGFA and ANGPT2 protein levels were significantly associated with the tumor size and lymph node metastasis only in ADC, not SQC. More importantly, increased VEGFA and ANGPT2 protein levels were negatively correlated with overall survival (OS) of ADC individuals. Meta-analyses of 22 gene expression omnibus (GEO) databases of lung cancer implicated that patients with higher VEGFA and ANGPT2 mRNA expressions tended to have advanced stage in ADC rather than SQC. Kaplan-Meier plot analyses further verified that high levels of VEGFA and ANGPT2 mRNA were associated with poor survival only in ADC. Moreover, the combination of VEGFA and ANGPT2 could more precisely predict prognosis in ADC. In hypoxia-mimicking conditions, induced expression of HIF-1α unregulated VEGFA and ANGPT2 proteins abundance. Conclusion: Our results showed hypoxia upregulated the protein levels of VEGFA and ANGPT2 in lung cancer cell lines, and the roles of VEGFA and ANGPT2 were distinct in ADC and SQC. Combined detections of VEGFA and ANGPT2 may be valuable prognostic biomarkers for ADC and double block of VEGFA and ANGPT2 may improve therapeutic outcome. Introduction According to the latest cancer statistics published by American Cancer Society, lung cancer remains the leading cancer-related mortality in the United States for both male and female [1]. Lung cancer is divided into two major classes based on its biology, therapy, and prognosis: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC), the latter accounting for about 80-85% of all Ivyspring International Publisher [2]. NSCLCs consist of several subtypes, mainly including adenocarcinoma (ADC), which accounts for 50% of NSCLCs, and squamous cell carcinoma (SQC), which takes up 30% of NSCLC cases [3]. In the past decade, the discoveries of driver gene mutations, such as epidermal growth factor receptor (EGFR) and kirsten rat sarcoma viral oncogene homolog (KRAS), and corresponding molecule-targeted therapies have dramatically improved the prognosis of a portion of NSCLC patients [4]. But the outcome of most lung cancer patients is still far from satisfactory, which is largely due to lack of effective target, drug resistance and metastasis [5]. The hypothesis "tumor growth is angiogenesis dependent" was first proposed by Folkman in 1971 [6]. Tumor angiogenesis is a complex dynamic process, among which, the vascular endothelial growth factor/vascular endothelial growth factor receptor (VEGF/VEGFR) pathway [7,8] and the angiopoietin (ANGPT)/Tie signal system [9] are the most important elements. VEGFA binds to its receptors VEGFR1 (Flt-1) and VEGFR2 (KDR/Flk-1), thus triggering multiple downstream signaling pathways, such as mitogen-activated protein kinase (MAPK) and phsphoionsitide 3-kinase (PI3Ks). Activated VEGFA signaling pathway can promote proliferation and migration of endothelial cell as well as their survival and vascular permeability [10]. ANGPT1 and ANGPT2 bind with similar affinity to the extracellular domain of Tie2, an endothelial cell tyrosine kinase receptor [11]. ANGPT1 is thought to drive vessel wall stabilization and maturation, mediate the migration, adhesion and survival of endothelial cell. In contrast, as an antagonistic to ANGPT1, ANGPT2, destabilizes vessel assembly, increases vessel permeability, and induces a state of vascular plasticity [12]. Recently, ANGPT2 has been identified as a potent proangiogenic factor which functions in collaboration with VEGFA [13]. The clinical significance of VEGFA and ANGPT2 in lung cancer has been reported in previous studies. Although the results about the relationship among patient's clinic-pathological characteristics, prognosis and VEGFA/ANGPT2 are roughly the same, there are still some contradictions among different groups. An early meta-analysis which included 44 studies indicated the inverse relationship between VEGFA and survival |
in patients with NSCLC and ADC [14]. However, on account of the inadequate studies on SQC (only 2), the author failed to conduct the subgroup analysis on SQC [14]. Zhang et al. conducted an updated meta-analysis about the prognostic impact of VEGFA in patients with NSCLC [15]. Their analyses suggested that high-VEGFA was significantly associated with poor survival in NSCLC patients and the trend was also observed in subgroup analysis of ADC and SQC patients [15]. On the contrary, the results of Pajares and his colleagues indicated that high protein expressions of VEGFA and its receptors were associated with a good prognosis in patients with SQC but not in ADC [16]. Apart from VEGFA, ANGPT2 is another mediator of angiogenesis. It is generally believed that ANGPT2 expression correlates with clinic-pathological features and clinical outcomes as well. Christian et al. observed that a higher ANGPT2 mRNA expression predicates a worse prognosis in primary breast cancer [17]. A meta-analysis conducted by Xuan and his colleagues suggested that high expression of ANGPT2 in tumor tissues was significantly associated with poor survival of NSCLC, but the subgroups analysis about ADC and SQC were not performed [18]. Furthermore, the levels of serum ANGPT2 were also reported to be associated with progression and prognosis in NSCLC [19]. Nevertheless, most researches focused on the role of VEGFA/ANGPT2 in NSCLC while few studies centered on the distinct predictive values of VEGFA and ANGPT2 in ADC and SQC. More evidence demonstrated that ADC and SQC are fundamentally different pathological types with entirely diverse prognosis and therapeutic strategy. For example, Bevacizumab, as the first VEGFA-targeted agent, is approved only for patients with non-squamous NSCLC based on the pivotal study E4599 [20]. To further evaluate the expression and significance of VEGFA and ANGPT2 in ADC and SQC, respectively, tissue microarray (TMA) slides containing different pathological subtypes and large public Gene Expression Omnibus (GEO) databases were utilized. In this study, we showed that the expression of VEGFA and ANGPT2 were significantly associated with progression and clinical outcome of ADC both in mRNA and protein levels. However, the phenomenon was not observed in SQC. Our analysis strongly suggested that treatments targeting to VEGFA and ANGPT2 might be better applied to ADC. Immunohistochemical staining and quantification analysis The specific polyclonal antibodies against VEGFA (19003-3-AP, ProteinTech, 1:200) and ANGPT2 (24613-1-AP, ProteinTech, 1:200) were utilized for IHC on TMA slides with a two-step protocol by the Biossci Biotech, Inc. [21]. The VEGFA IHC image of HLugSqu150Sur01 was provided by the Shanghai Outdo Biobank. To semiquantitative evaluate VEGFA and ANGPT2 density, at least 4 fields at 200×magnification of each spot were selected and the IHC score was assessed by two individuals independently. Scoring was related to two variables: staining intensity and the percentage of positive cells. We applied Fromowitz standard to assess the intensity of staining and the percentage of positive staining tumor cells [22]. The staining intensity was scored as follows: 0 (no staining), 1 (weak staining), 2 (moderate staining), 3(strong staining). The proportion of stained positive tumor cells was divided into four levels: 1 (0%-25% positive cells), 2 (26%-50% positive cells), 3 (51%-75% positive cells) and 4 (76%-100% positive cells). A score ranging from 0 to 12 was calculated by multiplying the intensity with percentage and the median score was defined as cutoff value. Meta-analysis for VEGFA and ANGPT2 mRNA expression on GEO databases The method to perform the meta-analysis was described in our previous meta-analysis on SIX family [23]. The electronic databases obtained from ArrayExpress were used to search for relevant GEO databases of human lung cancer with the mRNA expression of VEGFA and ANGPT2 by using the term "lung cancer". The databases should meet the following criteria: (a) samples in the databases were human normal lung tissue or pathologically diagnosed as ADC or SQC; (b) the mRNA expression value of VEGFA and ANGPT2 were measured in the databases rather than DNA or microRNA; (c) the sample size of the database was more than 50; (d) if the same patient was included in more than one database, only the latest and most complete databases was included in the analysis; (e) the clinic-pathological and prognosis information were showed in these databases, such as grade, tumor size, lymph node metastasis, TNM stage, and clinical outcome. We adopted the median as the cutoff values of mRNA expression. The relationship between clinic-pathological parameters and VEGFA mRNA expression as well as ANGPT2 mRNA expression were assessed by the odds ratio (OR) and its corresponding 95% CI. Heterogeneity of publication bias was assessed by Cochrane Q and I 2 test. We employed random-effect model if heterogeneity was seen between studies (I 2 > 50% or P≤ 0.05). Otherwise, we adopted fixed-effect model (I 2 ≤ 50% or P> 0.05). Finally, a total of 22 independent microarray databases, were enrolled in this meta-analysis ( Table S1). The flow diagram reflecting the selection process of relevant studies was shown in Figure S1. The STATA software package (version 12.0) (Stata Corp LP, College Station, TX, USA) was employed to perform the meta-analysis. Analysis of public microarray mRNA expression datasets of VEGFA, ANGPT2, and hypoxia inducible factor-lα (HIF-1α) for lung cancer were downloaded from the ArrayExpress. GSE68465, containing 443 ADC patients, and GSE4573 with 130 SQC cases were applied to evaluate the mRNA expression level in different histologic grades. GSE31210, an expression profile containing a total of 226 primary ADC patients and GSE32474, including 26 lung cancer cell lines were interrogated to assess the correlation between the mRNA expression of VEGFA, ANGPT2, and HIF-1α. Kaplan-Meier plotter In this paper, we used an online analysis tool to calculate and plot Kaplan-Meier survival curves with hazard ratio (HR) and log-rank P value (http:// kmplot.com/analysis/) [46]. The affymetrix probe ID for VEGFA and ANGPT2 were 210513_s_at and 205572_at, respectively. The follow-up time threshold was 120 months. We used the median expression value to divide the patients into two groups, then the Kaplan-Meier survival curves were downloaded from the website and resized in Adobe Illustrator CS6. Cell culture and treatment Two human lung cancer cell lines (NCI-H1299 and A549) were purchased from American Type Culture Collection (ATCC), and cultured in 1640 medium (HyClone, USA) supplemented with 10% fetal bovine serum (Gibco, USA). All cells were grown in a humidified atmosphere of 5% carbon dioxide at 37°C. CoCl2 was utilized to mimic a hypoxia condition, and cells were exposed to different concentration of CoCl 2 (control, 100µM, 200µM) for 12h. Statistical analysis The Student's t-test was applied to evaluate the differences between groups. A two-tailed P value <0.05 was considered statistically significant. The cumulative survival time was calculated utilizing the Kaplan-Meier method and analyzed with the log-rank test. Statistical analyses were conducted by GraphPad Prism 5.0 and SPSS 16.0. All data were presented as the mean ± standard error of mean (SEM). The expression of VEGFA and ANGPT2 elevated in ADC and SQC compared with normal lung tissues In order to evaluate the protein expression of VEGFA and ANGPT2 in ADC, SQC, and paracancerous tissues, we carried out IHC analysis on four TMAs (two HlugA180Su05, one HLugSqu150Sur01, and one LC642). The VEGFA and ANGPT2 expression with stronger brown staining particles in the cancerous tissues were mainly localized in cytoplasm and cell membrane, and with weaker cytoplasm staining in corresponding adjacent tissues. The representative images of IHC staining for noncancerous and cancers tissues were shown in Figure 1A-D. The IHC scores of tumor tissues were significantly higher than those of matched adjacent tissues (P<0.0001) ( Figure 1A-D). Furthermore, we adopted a meta-analysis to evaluate whether the mRNA expression of VEGFA and ANGPT2 were consistent with the protein abundance. The patients were divided into high and low subgroups based on the median mRNA expression value. Our analysis indicated that the mRNA expression of VEGFA was increased in ADC (OR=3.98, 95% CI: 1.84-8.60, P=0.002 and I 2 =69.2%) ( Figure 1E) when compared with normal lung tissues. The same tendency was seen in ANGPT2 (OR=1.46, 95% CI: 1.05-2.04, P=0.000 and I 2 =79.8%) ( Figure 1G). Analysis of SQC was also proven to have the similar trend (VEGFA: OR=5.09, 95% CI: 2.35-11.03, P=0.439, I 2 =0.0%, Figure 1F; ANGPT2: OR=1.94, 95% CI: 1.01-3.75, P=0.029 and I 2 =71.6%, Figure 1H). In order to deepen our understanding about the expression of VEGFA and ANGPT2 in different histological types, namely ADC and SQC, we compared their mRNA levels between these two pathological types. The combined ORs of VEGFA were 1.73 (95% CI: 1.09-2.76; P=0.000 and I 2 =77.0%) ( Figure S2A), indicating a higher VEGFA expression in ADC. Nevertheless, ANGPT2 just showed a moderate trend without reaching significance (OR=1.01, 95% CI, 0.82-1.25, P=0.651 and I 2 =0.0%) ( Figure S2B). VEGFA expression was associated with cancer progression in ADC, not in SQC Moreover, we investigated the relationship between the protein level of VEGFA and clinical-pathological parameters of ADC and SQC. Patients with stage III showed stronger staining than samples with early stages (stage I-II) (P=0.0036) ( Figure 2A) in ADC. The same trend was also found in ADC patients with larger tumor size. Tissues with bigger tumor size (T3-T4) had increased VEGFA expression than those with smaller tumor size (T 1 -T 2 ) (P=0.0346) ( Figure 2C). We also analyzed the relationship between VEGFA protein level and TNM stage, tumor size in SQC. However, no statistical difference was found ( Figure 2B, Figure 2D). The results of meta-analysis were consistent with the protein abundance, that increased VEGFA mRNA level were significantly associated with advanced tumor stage (OR=1.93, 95% CI: 1.33-2.82, P=0.588, and I 2 =0.0%) ( Figure 2E) and big tumor size (OR=1.70, 95% CI: 1.05-2.74, P=0.385, and I 2 =5.0%) ( Figure 2G) in ADC patients. However, analysis in SQC showed no significance between the VEGFA mRNA expression and TNM stage ( Figure 2F), tumor size ( Figure 2H). Patients with lymph node metastasis expressed more VEGFA protein than those without lymph node metastasis (P=0.0299) ( Figure 3A). But, the VEGFA expression between high differentiation (grade1-2) and low differentiation (grade 3) did not reach a statistical significance (P=0.0741) ( Figure 3C). Analysis conducted on SQC patients also showed no statistical difference between the protein expression of VEGFA and the different lymph node status ( Figure 3B) as well as tumor grade ( Figure 3D). Simultaneously, the meta-analysis suggested that increased VEGFA mRNA level was significantly associated with lymph node metastasis (OR=2.12, 95% CI: 1.59-2.82, P=0.188, and I 2 =31.4%) ( Figure 3E) in ADC patients. The meta-analysis of SQC showed no significance between the VEGFA mRNA expression and N status ( Figure 3F). As the data extracted were not sufficient to conduct pooled analysis for histological grade, GSE68465 including 443 ADC was interrogated to evaluate the mRNA levels of VEGFA in different grades, which showed that ADC patients with high grade expressed more VEGFA than patients with low grade (P<0.0001 and P=0.0009) ( Figure 3G). GSE4573 containing a total of 130 SQC cases was employed to analyze the significance of VEGFA expression in different grades, but no statistical difference was observed ( Figure 3H). Median IHC score 9 was used to divide VEGFA expression into high and low group and the correlation between VEGFA expression and clinic-pathological features of ADC patients in HlugA180Su05 was displayed in Table S2. We found the level of VEGFA protein expression was significantly related to TNM stage (P=0.041), while the correlation was not observed in other clinic-pathological characteristics including age, gender, tumor size, lymph node status and histological grade. ANGPT2 expression was associated with cancer progression in ADC, not in SQC The same analysis was also conducted on ANGPT2. The relationship between ANGPT2 protein expression and TNM stage of ADC was on the verge of statistically significant (P=0.0599) ( Figure 4A). However, the ANGPT2 protein abundance was higher in tumor with big size (T3-T4) than that in small size (T 1 -T 2 ) (P=0.0417) ( Figure 4C). Analysis performed on the relationship between protein level of ANGPT2 in SQC and clinical-pathological features mentioned above showed no statistically significance ( Figure 4B, Figure 4D). The meta-analyses suggested that the correlation between ANGPT2 mRNA expression and TNM stage of ADC hovered around significance (OR=1.44, 95% CI: 0.96-2.14, P=0.415, and I 2 =1.3%) ( Figure 4E) and there was no statistical difference of the ANGPT2 mRNA between |
the big tumor size (T 3 -T 4 ) and small tumor size (T 1 -T 2 ) (OR=1.17, 95% CI: 0.74-1.86, P=0.226, and I 2 =27.8%) ( Figure 4G) in ADC. Statistical differences among the ANGPT2 mRNA expression and TNM stage ( Figure 4F), tumor size ( Figure 4H) in SQC were also not observed. The protein level of ANGPT2 in ADC was correlated with lymph node metastasis (P=0.0076) ( Figure 5A), but we failed to find significant association between tumor grade and the protein abundance of ANGPT2 (P=0.1694) ( Figure 5C). Analysis conducted on SQC patients also showed no statistical difference among the protein expressions of ANGPT2 and the different lymph node status ( Figure 5B), tumor grade ( Figure 5D). The meta-analysis demonstrated that the mRNA level of ANGPT2 was dramatically higher in ADC patients with lymph node metastasis (OR=1.58, 95% CI: 1.18-2.12, P=0.524, and I 2 =0.0%) ( Figure 5E). In contrast, there was no statistical difference between the ANGPT2 mRNA expression and lymph node status in SQC ( Figure 5F). The representative dataset GSE68465 showed that the difference was statistically significant among distinct grade (P=0.0002 and P= 0.0081) ( Figure 5G) in ADC while the bar graph adopted from GSE4573 certified no correlation between the ANGPT2 expression and histologic grade ( Figure 5H) in SQC patients. Division of these patients into ANGPT2-high and low expression groups by median IHC score 8 revealed a strong relationship with lymph node metastasis (P=0.002) (Table S3). Increased expression of VEGFA and ANGPT2 predict poor survival in ADC To explore the prognosis value of VEGFA and ANGPT2 mRNA levels, Kaplan-Meier curves were plotted. The results indicated that patients with higher mRNA level of VEGFA had shorter overall survival (OS) (HR=2.45, 95% CI: 1.91-3.14, P<0.0001) ( Figure 6A Meanwhile, we investigated the association between the protein level and prognosis. Median OS times of patients with VEGFA-low and VEGFA-high were 64.8±6.67 and 35.7±3.98 months, respectively, indicating significant difference of survival (P=0.006) ( Figure 6C) in ADC individuals. The median OS times of the ANGPT2 low group was 68.3±10.56 months, while that of high group was 42.0±6.35 months ( Figure 6I) in ADC patients. By contrast, analysis of VEGFA in SQC subjects did not reach to significance level ( Figure 6F), which was consistent with result adopted from Kaplan-Meier plotter. Univariate Cox regression analysis was used to investigate the correlation between cumulative OS rates and clinic-pathological factors in patients with ADC. As shown in Table 1 We performed a Forward: LR variable selection procedure using these three factors, and the VEGFA expression was identified as an independent predictive factor for the OS in ADC patients (HR=1.745, 95% CI: 1.029-2.959, P=0.039). The same univariate Cox regression analysis was conducted on ANGPT2. ANGPT2 expression, lymph node metastasis, as well as TNM stage were obviously associated with the clinical outcomes of ADC patients (Table S4). VEGFA expression was correlated with ANGPT2 Previous study has indicated that the expression of VEGFA in tumor cells was positively associated with ANGPT2 and predicted poor survival [47]. Herein, lung cancer cell line data reported by Kohn et al. [48], including a total of 26 lung cancer cell lines, was employed to evaluate the correlation between the mRNA expression of VEGFA and ANGPT2. The result displayed that VEGFA mRNA expression was parallel with ANGPT2 (r=0.424, P=0.031) ( Figure 7A). Public dataset GSE31210, containing 226 ADC cases was also interrogated to assess the association between VEGFA and ANGPT2 at mRNA level. As expected, there was a significantly positive association between VEGFA and ANGPT2 (r=0.367, P<0.001) ( Figure 7B). The IHC analysis of VEGFA and ANGPT2 for the same tissue microarray (Hlug-A180Su05) also showed a positive correlation between them (r=0.358, P=0.006) ( Figure 7C), which was consistent with the conclusion draw from the correlation analyses of GSE32474 and GSE31210. The blend Kaplan-Meier curves in GSE31210 showed that patients with low VEGFA and low ANGPT2 at mRNA level had the longest OS and relapse free survival (RFS) time, whereas high VEGFA and high ANGPT2 predicted poorest prognosis ( Figure 7D-E). The same conclusion could be acquired from the IHC score analysis of VEGFA and ANGPT2, that patients with high protein expression of VEGFA and a concomitantly high ANGPT2 expression suffered a dramatic survival reduction (P=0.040) ( Figure 7F), suggesting there is a synergistic effect between VEGFA and ANGPT2. The association among the expressions of HIF-1α, VEGFA and ANGPT2 It is well accepted that VEGFA and ANGPT2 are major angiogenesis factors, and HIF-1α is a transcription factor of VEGFA [49]. Thereby, we tried to explore the regulation effect of HIF-1α on VEGFA and ANGPT2 in NSCLC. We first analyzed the correlation between the HIF-1α and VEGFA/ ANGPT2 at mRNA levels. The results showed that HIF-1α mRNA expression was positively correlated with ANGPT2 both at lung cancer cell lines (r=0.513, P=0.007) ( Figure 8B) and lung cancer tissues (r=0.285, P<0.001) ( Figure 8D). While VEGFA was just parallel with HIF-1α in lung cancer tissues (r= 0.420, P<0.001) ( Figure 8C), not in lung cancer cell lines (r=0.315, P=0.117) ( Figure 8A). As HIF-1α was a major gene response to hypoxia, we used CoCl2 to mimic hypoxia condition [50]. Treatment of NCI-H1299 and A549 cells with CoCl 2 (100 or 200 µM) for 12h induced a significant increase in the protein level of HIF-1α, VEGFA, and ANGPT2 ( Figure 8E) compared with the untreated control cells. At the same condition, hypoxia induced the protein expression of P21. Discussion VEGFA is first discovered as an endothelial cell-specific mitogen and an angiogenesis inducer released by tumor cells in vivo and expressed in human tumors in situ [51]. VEGFA protein has been demonstrated to be highly expressed in several NSCLC cell lines and mediated angiogenesis [52]. ANGPT2, a specific extracellular ligand to Tie2, has also been showed to overexpress in NSCLC tissues in a meta-analysis [18]. In our study, a total of 3388 patients in larger public GEO databases were employed to evaluate the relationship of VEGFA/ANGPT2 in ADC and SQC patients at mRNA level. Additionally, 4 tissue microarrays were used to explore the relationship between ADC/SQC and VEGFA /ANGPT2 at protein level with about 400 patients included. Up to date, we don't find other research integrating so many public databases and evaluating VEGFA/ANGPT2 in ADC and SQC at mRNA and protein levels simultaneously. Our study observed that VEGFA and ANGPT2 expressions in ADC and SQC were significantly higher compared with normal lung tissues both at mRNA and protein levels. Results in our study further indicated that VEGFA was higher in ADC compared with SQC at the mRNA level. However, there was no difference of ANGPT2 in ADC and SQC. Specifically, high VEGFA mRNA level in ADC were associated with advanced stages, large tumor size, positive lymph node metastasis, and poorly tumor cell differentiation, whilst the association was not detected in SQC. We also illustrated that high ANGPT2 was linked to lymph node metastasis both at mRNA and protein levels. However, this phenomenon was not observed in SQC. Coincidentally, a previous research reported that a significant positive association existed between ANGPT2 and lymph node metastasis in breast cancer [17]. Lung cancer represents a highly malignant and particularly aggressive cancer type, with early and widespread metastasis and poor prognosis, thus identifying a potential survival predictor is of great importance. Early in 1996, Ohta et al. reported that 5-year survival rates for NSCLC patients with low-VEGFA and high-VEGFA mRNA level were 77.9% and 16.7%, respectively [53]. The updated metaanalysis involving 74 sets of expression of VEGFA by IHC or enzyme-linked immunosorbent assay (ELISA) in lung cancer was conducted by Zheng et al. [15]. By their analysis, they concluded that the VEGFA overexpression indicated a poor prognosis in patients with NSCLC, ADC, and SQC at protein levels [15]. The disparity between ours and Zheng's may arise from the different detection methods and sample size. In our study, we simultaneously investigated the VEGFA and ANGPT2 in ADC and SQC. Our results indicated that high-VEGFA and high-ANGPT2 were remarkably associated with poor prognosis of ADC, not SQC patients. Tanaka et al. also indicated that the high expression of ANGPT2 was a significant factor to predict a poor postoperative survival in NSCLC. However, they didn't perform subgroup analysis to compare its roles in ADC and SQC [47]. They also demonstrated that the survival of patients with high-ANGPT2 and high-VEGFA was extremely poor, which is in accordance with our KM plotter results. According to our multivariate analysis using Cox regression, the VEGFA overexpression was found to be an independent significant prognostic factor in ADC, which was in agreement with early result reported by Imoto et al. [54]. Their study indicated that VEGFA was an important prognostic factor in completely resected NSCLC, but they did not separate ADC and SQC. Similarly, they thought the VEGFA-positive rate was significantly higher in patients with ADC than in those with SQC (P=0.03). In fact, the difference in genetic changes in histologic type of lung cancer have been reported. For instance, ras mutation are found predominantly in ADC [55], whereas p53 gene mutations are more frequent in SQC compared with ADC [56]. The expression of angiogenic factors, which are activated from mutations such as diver gene ras, may be different in ADC and SQC. Those genes may control other angiogenesis factors through different pathway. HIF-1 is a heterodimer protein complex which is composed of a constitutively expressed HIF-1β subunit and an oxygen-regulated HIF-1α subunit [57]. HIF-1α is a major subunit response to hypoxia, oxidative stress and activates VEGF-induced angiogenesis [58,59]. Previous study has showed that CoCl2 can create a hypoxia-like state in vitro or in vivo [49]. In the present study, we adopted commonly used concentration range of CoCl 2 to create a hypoxia culture mode in two NSCLC cell lines, NCI-H1299 and A549 [49,60]. Our results have confirmed that hypoxia simulated by CoCl 2 can induce HIF-1α expression accompanying by the enhanced protein abundance of VEGFA and ANGPT2. Indeed, anti-angiogenic therapy has been shown responses in many kinds of carcinoma [61]. Bevacizumab is approved only for patients with nonsquamous NSCLC due to frequently life-threatening adverse events such as pulmonary hemorrhage, particularly in patients with SQC [62]. Apart from these safety concerns, patients with squamous NSCLC in several phase III trials could not benefit from the combination of antiangiogenic therapy and chemotherapy compared with chemotherapy alone [3]. Our results showed that further clinical trial targeting VEGFA and ANGPT2 should exclude SQC patients based on the lack of biological impact and prognosis on SQC. ANGPT2 and VEGFA have complementary roles in regulating tumor angiogenesis and synergistic effect on survival, suggesting that dual pathway inhibition is necessary to improve treatment outcomes. A phase I study of single-agent Vanucizumab, a bispecific monoclony antibody (mAb) targeting VEGFA and ANGPT2 showed an encouraging antitumor activity and the further study is expected [63]. Our study confirms that the expressions of VEGFA and ANGPT2 in ADC and SQC are significantly higher than that in normal tissues both at mRNA and protein levels. Furthermore, the relationship between clinic-pathological parameters and expression of VEGFA and ANGPT2 supported their roles in the progression of ADC. VEGFA is positively associated with ANGPT2 in lung cancer cell lines and tumor tissues of ADC. Both VEGFA and ANGPT2 serve as poor prognostic biomarkers, and VEGFA might be an independent prognostic factor of OS in ADC patients, but not in SQC. The prognostic impact of VEGFA in ADC appears strongly associated with a concomitantly high expression of ANGPT2. Therefore, double detection of VEGFA and ANGPT2 could provide precise information for predicting the prognosis of ADC patients. Single-Cell Elastography: Probing for Disease with the Atomic Force Microscope The atomic force microscope (AFM) is emerging as a powerful tool in cell biology. Originally developed for high-resolution imaging purposes, the AFM also has unique capabilities as a nano-indenter to probe the dynamic viscoelastic material properties of living cells in culture. In particular, AFM elastography combines imaging and indentation modalities to map the spatial distribution of cell mechanical properties, which in turn reflect the structure and function of the underlying cytoskeleton. Such measurements have contributed to our understanding of cell mechanics and cell biology and appear to be sensitive to the presence of disease in individual cells. This chapter provides a background on the principles and practice of |
AFM elastography and reviews the literature comparing cell mechanics in normal and diseased states, making a case for the use of such measurements as disease markers. Emphasis is placed on the need for more comprehensive and detailed quantification of cell biomechanical properties beyond the current standard methods of analysis. A number of technical and practical hurdles have yet to be overcome before the method can be of clinical use. However, the future holds great promise for AFM elastography of living cells to provide novel biomechanical markers that will enhance the detection, diagnosis, and treatment of disease. Introduction Many physiologic and pathophysiologic processes alter the biomechanical properties of the tissues they affect. It is well known that muscles get harder with weight training, and skin becomes less resilient with age. Abnormal tissue biomechanics also play a key role in a wide range of diseases such as osteoporosis, osteoarthritis, cystic fibrosis, muscular dystrophy, ventricular aneurysm, and others. Based on the relationship between tissue mechanics and pathology, palpation is used clinically to detect stiff nodules associated with breast cancer and abdominal hardness due to cirrhosis of the liver. In an effort to make such examinations more quantitative, a number of indentation devices have been developed to evaluate the stiffness of soft tissues in vivo [1][2][3][4], though these have yet to achieve wide clinical acceptance. Recently, there has been great interest in a new technique known as elastography [5], which generally refers to any imaging modality that yields information about the mechanical properties of a tissue. Based primarily on ultrasound and magnetic resonance imaging methods, elastographic techniques have demonstrated the ability to detect the size and shape of tumors [5,6], to identify regional anatomic differences in normal tissue stiffness [5,6], to identify abnormal cardiac deformation due to coronary artery disease [7,8], and even have been implemented in a catheter system for intravascular evaluation of atherosclerotic plaques [9]. However, in an elastogram, image contrast is based on regional differences in the response of tissue structures to applied loads, yielding new information not available using traditional medical imaging modalities. Consequently, there is rapidly growing clinical interest in the ability to diagnose disease based on analysis and visualization of regional tissue mechanical properties. It follows that pathophysiologic changes in the mechanical properties of tissues may be manifest at the single cell level. In fact, alterations of cell mechanical properties recently have been reported in certain forms of cancer, arthritis, and cardiovascular disease [10][11][12][13], opening a new window to examine the underlying mechanisms of these pathologies. Moreover, once the normal and abnormal mechanical properties of a given cell type are established, it is enticing to imagine that potential pharmaceutical or genetic treatments might be evaluated by measuring their effects on the mechanical properties of target cells in vitro. Hence, by complementing other evolving single-cell analysis techniques [14,15], the identification of a distinct biomechanical fingerprint of the cell in response to a battery of material tests may offer an important new approach in cell biology. Single cell elastography using atomic force microscopy is a technique with the potential to identify such a mechanical fingerprint. At present, atomic force microscope (AFM) elastography is largely a research tool used by biomedical engineers and biophysicists to study the mechanics of cell function. However, the technique is evolving rapidly to a state where medical applications may be feasible. Therefore, the purpose of this article is to provide a brief introduction to cell biomechanics and its relation to disease; to describe the AFM experiment, including principles of operation and methods of data analysis; to review recent findings in the area of cell mechanics with AFM; and to identify the current limits of the technology and future developments that would enhance transfer to the basic and clinical sciences to aid in the identification of novel cell biomechanical markers that might lead to improved detection, diagnosis, and treatment of disease. Basic cell biomechanics Such a detailed characterization of cell mechanics requires knowledge of the constitutive relation of the cell, which relates cell deformation (i.e., strain) to internal forces and externally applied loads (i.e., stress) acting on the cell. Stiffness is defined as the slope of the force-deformation curve -it depends on geometry and hence on the particular sample studied and the testing device used. Therefore, rather than relating force (F ) and deformation (∆L) directly, it is important to consider the related quantities stress (σ = F/A) and strain (ε = (∆L/L o ) because these are normalized measures (by area, A, and initial length, L o , respectively) independent of size or geometry. That is, the stress-strain constitutive relation reflects an underlying property of the cell. Perhaps the best known and sim-strain, ε stress, σ lo a d in g u n lo a d in g Fig. 1. An idealized linear elastic material (dotted line) is characterized by the Young's modulus obtained from the slope of the stress-strain curve. For most biological soft tissues, the stress-strain relation is nonlinear (solid line) and exhibits viscoelastic hysteresis between loading and unloading segments of the curve. plest constitutive relation for solid materials is Hooke's law, which states that stress is proportional to strain (σ = Eε), where the constant of proportionality, E, is called the Young's modulus. Materials that obey Hooke's law (e.g., rubber, steel, bone) are called linear elastic (Fig. 1). A similar constitutive relation for fluid materials states that the stress is proportional to the rate of strain (σ =: µdε/dt), where the constant of proportionality, µ, is called the viscosity. Such materials (e.g., water, blood plasma) are called Newtonian fluids. Macroscopically, most soft biological tissues are more complex than these simple idealized materials [16]. In addition to being heterogeneous, with mechanical properties varying from one region of the tissue to another, the stress-strain relationship typically is nonlinear (e.g., polynomial or exponential), such that the modulus increases as the tissue is deformed (Fig. 1). Many tissues also have a preferred structural alignment that gives rise to material anisotropy, such that the measured material properties depend upon the axis along which the tissue is tested. Moreover, most soft tissues are viscoelastic materials consisting of solid and fluid (and ionic) components that influence how the tissue responds to mechanical stimulation [17,18]. Hence, the mechanical behavior depends not only on how much the tissue is deformed, but also on how rapidly it is deformed and on the memory of its previous deformation history, resulting in hysteresis between the loading and unloading portions of the stress-strain curve (Fig. 1). Clearly, such tissues are not well characterized by a single Young's modulus, and constitutive equations that combine elastic and viscous properties are required to mathematically model their stress-strain behavior [16]. Single cells appear to share many of the same biomechanical characteristics as macroscopic soft tissues. Even the earliest analyses of leukocyte mechanics recognized the importance of viscoelasticity due to the aqueous gel nature of the cytoplasm [19,20]. More recent studies suggest that actin filaments, intermediate filaments, and microtubules each contribute differently to the viscoelastic properties of fibroblasts and endothelial cells [21][22][23], though the specific roles remain quite controversial [24]. Some anchorage-dependent cells appear to exhibit nonlinear elastic behavior, also called strain hardening [25], but it is unclear whether this is an intrinsic material property or a consequence of the underlying cytoskeletal architecture [26]. Heterogeneity of cell mechanical properties has been associated with the nucleus and other organelles [27,28]. In addition, the major cytoskeletal filaments that determine cell material properties are heterogeneously distributed, preferentially oriented, and dynamic -assembling, disassembling, and reorganizing in response to their mechanical environment [29][30][31][32]. Consequently, reported measurements of the Young's modulus of a cell must be interpreted with caution. Unfortunately, relatively little is known about the more detailed multiaxial, nonlinear, viscoelastic mechanical properties of most cells, and the identification of such represents one of the major challenges in modern biomechanics. Cell mechanics as a disease indicator Nevertheless, the field of cell mechanics has evolved tremendously in the past two decades, and several texts have now been devoted to the topic [33][34][35]. Of particular interest here are the growing number of studies that demonstrate a close association between cell mechanical properties and various disease conditions. For example, cultured myotubes from a dystrophin-deficient rat model of Duchenne muscular dystrophy were only one-fourth as stiff as normal cells [36], and recent evidence suggests that some muscle types are protected from dystrophin deficiency by up-regulating specialized accessory proteins that act to preserve cell stiffness [37]. Chondrocytes isolated from osteoarthritic human cartilage exhibit elevated viscoelastic moduli compared to cells from normal tissue [28], which may underlie the dissimilar responses of these cells to external mechanical stimulation [38]. Differences in mechanical properties between normal hepatocytes and hepatocellular carcinoma cells were restricted to the elastic moduli, while the viscous modulus was unaltered [39]. On the other hand, pressure-overload ventricular hypertrophy specifically increases viscous damping (without affecting elastic stiffness) in passive cardiac myocytes [12]. Therefore, methods of elastographic mapping must be developed that are sensitive to changes in viscous as well as elastic properties of the cell. Erythrocytes from patients with sickle cell disease are stiffer and more viscous than are normal red blood cells [13,40]. These mechanical properties are restored to near-normal values in patients treated with hydroxyurea [13,41], which suggests that measurements of cell mechanics also may be used to monitor the efficacy of therapeutic interventions. Because cell mechanical properties are determined largely by the underlying cytoskeleton, any disease process that alters the composition, organization, kinetics, or crosslinking of the cytoskeleton is likely to be detectable using single-cell elastography. At this time, data on the mechanical properties of different cell types are critically needed to establish methodological criteria and guidelines for comparing measured mechanical properties with a normal population, as is being done for clinical hemorheology [42,43]. Thus, the development of tools for reliable and rapid characterization of cell mechanical properties is essential. Measurement techniques Micropipette aspiration has been used widely to study the mechanical properties of red and white blood cells [20,44,45]. Most studies find that the cytoplasm of these cells behaves like a fluid, with elastic properties attributed to the cell membrane and cortical skeleton. This technique applied to endothelial cells subjected to shear stress revealed greater cytoplasmic stiffness compared to non-sheared cells [46]. While variations of the micropipette method continue to provide important information on cells floating in suspension [11,47], these methods are less well suited to studying adherent cells because the large deformation caused by aspiration may disrupt connections between the cell membrane and the underlying cytoskeleton. In addition, due to the large area aspirated, this technique has limited potential for examining regional variations in mechanical properties. Indentation is an alternative approach for identifying in-plane material properties of biological tissues [48,49]. Indeed, a "cell poker" utilizing pulled glass microfibers has been used to explore the mechanical properties of living cells [27,50,51]. In fibroblasts, stiffness was lower over the nucleus than over the cytoplasm, increased with indentation depth, and decreased after disrupting actin filaments with cytochalasin [27]. However, due to the large size of the probe (diameter ∼2 µm) relative to the cell thickness (typically <5 µm), some of these findings may have been influenced by the rigid substrate to which the cells were attached [52]. A number of alternative specialized techniques such as magnetic twisting cytometry [53], laser tracking microrheology [54], magnetic tweezers [55], and the optical stretcher [56], also have been developed to study the mechanical properties of cells with a well-defined cytoskeleton. However, following its invention in 1986 as a high-resolution imaging tool for investigating semiconductor properties at the atomic scale [57], the AFM rapidly has become one of the most versatile and widely used methods for studying mechanical properties of living cells [58]. Fundamentals of atomic force microscopy The AFM is well suited for cell mechanics applications due to its high sensitivity (sub-nanoNewton), high spatial resolution (sub-micron), and the ability to be used for real-time measurements in a physiologic aqueous cell culture environment. An important advantage of AFM over other cell mechanics techniques is the ability to combine high-resolution scanning with nano-indentation, which allows direct correlation of local mechanical properties with underlying cytoskeletal structures [23,59]. Unlike most other cell imaging techniques, atomic force microscopy is based on a direct mechanical interaction between the probe and the sample. In this sense, the AFM is inherently |
an elastography instrument. Another advantage is that commercial availability of the AFM makes it accessible to a broad range of investigators. Principles of operation In principle, the AFM is a relatively simple instrument that involves laser tracking of the deflection of a microscopic-sized cantilever probe as its tip scans, indents, or otherwise interacts with the sample (Fig. 2). The AFM probe is the transducer of the instrument and typically consists of a rectangular or "V"-shaped cantilever about 100 to 300 microns long and about half a micron thick, microfabricated of silicon or siliconnitride [60]. The physical and geometric properties of the cantilever determine its spring constant, k, which is used to convert the measured cantilever deflection, h, into a contact force, F = k × h. The value of k, which typically ranges from 0.01 N/m to 1.0 N/m for cell mechanics applications, is nominally provided by the manufacturer and may be individually calibrated using a variety of methods [61][62][63]. The standard AFM probe has an integrated pyramidshaped tip with a blunted point having a radius of curvature in the 50-100 nanometer range (Fig. 2b). It is this tip that actually comes in contact with the cell, while the cantilever serves as a soft spring to measure the contact force. The tip dimension determines the spatial resolution of the instrument. Therefore, sharpened pyramids, etched silicon cones, carbon nanotubes, and other high-aspect ratio tips have been developed to scan samples with ultra high resolution [60,64]. However, such tips have been shown to penetrate the cell membrane and cause damage to living cells, whereas the standard pyramid tip apparently does not penetrate the cell membrane [65]. Cell viability has been demonstrated up to 48 hours after AFM scanning [66,67], although significant alterations of cell morphology and transfer of membrane to the probe tip can occur under some conditions [67][68][69]. Modified AFM probes with glass or polystyrene microsphere tips also have been used for some cellular applications to yield a more easily characterized tip geometry, though at the expense of decreased spatial resolution [70,71]. The AFM sensor uses a laser beam reflected off the end of the cantilever probe and onto a four-quadrant photodetector to monitor vertical and lateral deflections of the probe due to contact forces at the tip. AFM probes often are coated with a thin layer of gold to increase reflectivity, especially for cell mechanics applications in which the laser intensity may be attenuated by the phenol red present in standard cell culture medium. The distance between the probe and the photodetector amplifies the laser reflection such that movements of the AFM tip on the order of 0.1 nm can be detected reliably [63]. The actuator that moves the AFM probe in the zdirection toward or away from the sample is a piezoelectric ceramic that deforms in response to applied voltages. The typical z-range is about 6 microns, though custom configurations have achieved a z-range up to about 20 microns [70]. Although piezoelectric materials inherently are nonlinear and hysteretic, these effects can be overcome by software compensation (open-loop design) or direct strain-gauge monitoring (closed-loop design) to yield very precise positioning of the AFM tip with sub-nanometer accuracy in the z-direction. Similarly, piezoelectric positioners are used to control movement in the x-y plane as well, with a maximum scan range typically around 100 × 100 microns. In the standard AFM configuration, the sample is positioned relative to a stationary probe. However, for cell biology applications, it is more convenient to place the entire AFM on the stage of an inverted light microscope to allow simultaneous visualization, including fluorescence microscopy, of the cells [70]. In this configuration, the AFM probe is moved relative to a stationary sample. In addition, whereas the typical setup conceals the petri dish underneath the AFM head, at least one model (the Bioscope from Digital Instruments, Santa Barbara, CA) supports the AFM head from behind. This leaves the sample easily accessible for direct visual inspection and also facilitates the use of other devices such as fluid exchange systems, micromanipulators, and the like that enhance its versatility for cell biology applications. The AFM experiment There are two primary forms of AFM imaging in which the probe is raster scanned over the sample. Tap-ping mode imaging involves oscillating the probe near its resonance frequency and using feedback to maintain the amplitude of the oscillation as the probe encounters different features of the sample. Contact-mode imaging involves simply raster scanning the tip over the sample using feedback to maintain a constant deflection (force) of the cantilever (Fig. 3). In both cases, a topographical image of the sample is constructed from the z-position of the probe at each x-y pixel location. Tapping mode has the advantage of intermittent contact with the sample, virtually eliminating any frictional forces and thus minimizing distortion or damage to the cell [72]. Tapping-mode based elastography techniques such as phase imaging [73,74] and force modulation [75][76][77] yield images with contrast related to local sample stiffness but have not been capable of yielding quantitative estimates of the elastic modulus due to complications with the analysis. Contact mode also has been shown to yield high-resolution images with cell viability sustained for several hours [66]. In addition, contact mode is easier to use than tapping mode and more conducive to switching back and forth between imaging and "force mode," in which nano-indentation is used to obtain quantitative stiffness measurements. For the AFM indentation experiment, the probe is located at a desired position over the sample and is put through an extension-retraction cycle covering a typical z-range of 1 to 3 microns at a typical frequency of 1 to 10 Hz (Fig. 4). As the AFM probe approaches and contacts the sample at position Z o , further extension of the probe (Z − Z o ) is converted into a combination of probe deflection, h, and sample indentation, The indentation response depends on the spring constant of the probe, the geometry of the tip, and the mechanical properties of the sample. One also can vary the rate of indentation to study viscoelastic properties. Thus, by monitoring the z-position and deflection of the probe (the so-called "force curve"), one can obtain an indentation curve of indentation force versus depth that can be analyzed to extract the elastic material properties of the sample, as discussed below. Force mapping is a hybrid combination of imaging and force probing that involves making a series of in- dentations in an array covering a region of interest on the sample and reconstructing an isoforce image from the z-position at which the probe reaches a preset constant deflection (i.e., contact force) [78]. In such images, larger z-values are interpreted as softer regions of the sample because a greater motion of the probe would have been required to achieve the preset force. However, in samples such as living cells, such images are complicated by the highly variable topography of the cell, which also influences the z-position at which a given contact force is achieved. Therefore, it is more accurate to analyze the indentation data and create an image that directly represents the elastic properties obtained at each pixel location. This is the method of AFM elastography. Of course, the second half of the indentation cycle (i.e., the retraction curve) also contains useful information. Differences between the indentation and retraction curves reflect viscoelastic hysteresis of the sample. Upon retraction of the probe, the AFM tip may adhere to the sample and cause negative deflections of the probe. Such retraction events are the focus of experiments on protein unfolding [79], receptor-ligand binding [80], and cell-cell adhesion [81]. With such a range of capabilities, the AFM has proven to be a very versatile tool for applications in cell biology (an extensive review of the state-of-theart can be found in reference [82]), and many of the properties that can be measured with AFM may depend on whether the cell is normal or abnormal. Herein, attention is focused on use of the AFM for mapping cell mechanical properties as an indicator of disease. Although the AFM has yet to achieve its full potential as a tool for measuring the micromechanical properties of cells, there is still much information to be garnered from the standard cell indentation test. As the field of cell elastography advances, other experimental protocols and methods of analysis may be developed, but the classic AFM indentation experiment is likely to remain an important component of any battery of cell mechanical tests. Analysis and visualization of AFM indentation data Since the earliest AFM studies of soft biological samples [83,84], the prevalent method of analyzing AFM indentation data has been application of the so-called "Hertz model" of contact between two elastic bodies. Actually, it was Love who first obtained the widely used solution for indentation with a cone [85]. However, Hertz originally had solved the contact problem for the sphere and other smooth ellipsoidal geometries [86], so herein we refer to this general class of indentation problems as the "Hertz theory." In particular, the equations relating force and depth for indentation with a cone and a sphere, respectively, are given by: where α is the semi-included angle of the cone tip, R is the sphere radius, and ν is the Poisson's ratio that determines the amount of lateral expansion that accompanies axial compression (note that ν = 0.5 for water and other incompressible materials, and this value often is assumed for cells). It is important to appreciate that these solutions are based upon a number of simplifying assumptions, including homogeneous, isotropic, linear elastic material properties; axisymmetry; infinitesi-mal deformations; infinite sample thickness and dimensions; and a smooth sample surface. Therefore, caution must be exercised when such theoretical solutions are applied to the more complex AFM-cell indentation problem. Analysis based on the Hertz theory has been tested on thin films of gelatin, polyacrylamide, and similar substances [71,74,84], and a strong correlation (even numerical equality) between the microscopic and macroscopic elastic properties has been demonstrated in some cases [87,88]. However, because these test materials actually satisfy several key assumptions of the theory (e.g., thick films with homogeneous, isotropic, linear elastic material properties), this agreement does not ensure that the analysis also will be accurate or appropriate for more complex materials such as cells. In particular, whereas the classical analysis assumes that the sample is well characterized by a single Young's modulus, like a piece of rubber or steel, several recent studies suggest that nonlinear material properties may be important at the cellular level [25,71,89,90]. In addition, the classical analysis assumes infinitesimal sample deformation, whereas for soft samples such as cells indented with the standard pyramidal tip, local deformations near the probe always fall into the finite strain regime [91,92], even for small indentations relative to the sample thickness. Interestingly, our recent finite element model studies of AFM indentation suggest that, if the sample is indeed a linear elastic material, then the Hertz theory may yield accurate estimates of the material properties even when applied to finite indentations [91]. However, the linear elastic condition must be demonstrated experimentally and not assumed a priori. Fidelity of the calculated elastic properties also requires accurate identification of the contact point [88], accurate calibration of the probe spring constant [93], and accurate representation of the detailed tip geometry [90,91], each of which can be challenging in practice. To address some of the practical and theoretical limitations of the Hertz theory, alternative approaches for analyzing AFM indentation data have been developed. For example, rather than fitting the entire post-contact data set, Radmacher and coworkers applied the Hertz equation using two post-contact data points to solve for the two unknowns: Young's modulus and contact point [94]. When applied to cells, this method showed that the Young's modulus often depends on the depth range from which the two data points are selected, with the value of E increasing as the points are chosen from deeper indentations [94]. However, data from deeper regions often yielded an inaccurate estimate of the contact point; thus, the resulting modulus values also were questionable. Hoh's technique of force integration to equal limits, or FIEL mapping, overcomes some practical difficulties with the standard analysis of AFM data such as contact point uncertainty, to yield regional maps of relative cell stiffness from the area under the force curve [59]. |
However, this analysis ultimately is founded on the same assumptions as the Hertz model and therefore is subject to the same theoretical limitations. McElfresh and coworkers developed an analysis that explicitly accounts for surface interactions with the cell membrane for AFM indentation of sperm cells, but the approach makes other assumptions, such as single-point contact, that limit applicability to more general cell types [89]. One of the more sophisticated approaches to date is by Mahaffy et al. [71], in which small perturbations upon a large indentation were used to extract frequency-dependent elastic and viscous moduli of polyacrylamide films. When applied to fibroblasts, the method revealed an elastic modulus that increased substantially with indentation depth,possibly indicative of nonlinear elastic cell material properties. However, the potential influence of the underlying rigid substrate was not taken into account specifically. Recently, Demitriadis and coworkers developed an empirical correction for the effects of finite sample thickness on the AFM indentation response [88]. This correction yielded consistent Young's moduli for thin (<5 µm) and thick poly (vinyl alcohol) films, although validation of the method on more complex samples and living cells awaits further study. We recently published a new analysis method whereby computation of an apparent elastic modulus as a function of indentation depth can reveal nonlinearity and heterogeneity of material properties from standard AFM indentation tests [91]. The concept is similar to that of Radmacher et al., mentioned above [94], but in our implementation the contact point is determined independently of the modulus and remains constant for all indentation depths, yielding elastic properties that more properly reflect the material properties of the sample. The form of the governing equation is derived from the Hertz theory: whereẼ is a generalized elastic modulus (equivalent to E/2(1 − ν 2 ) for linear elastic materials, but also may be defined in terms of nonlinear elastic material constants [91]), and φ(D) is a function of the indenter geometry that determines the depth dependence of the indentation response. Since F and D are measured in the AFM experiment (Fig. 4b), and φ(D) is determined by the tip geometry, the above equation may be solved at each force-depth datum to obtain a "point-wise" apparent modulus [91]. Extensive finite element model validation illustrated how depth dependence of the point-wise modulus is sensitive to a number of factors likely to be important for cell indentation, including nonlinear mechanical properties and through-thickness material heterogeneity. The point-wise analysis has been used to distinguish different cell types [90] and also is readily applied to force curves with abrupt features as may occur when probing organelles [95]. Unfortunately, some characteristics of the shape of the curve are not unique -for example, some forms of material heterogeneity look similar to material nonlinearity. Therefore, a more sophisticated battery of tests is required to make such distinctions. Nevertheless, AFM indentation tests clearly have potential to yield more information about cell mechanical properties than a single modulus value, and to ignore this information is to discard potentially critical data on the detailed mechanical properties of the cell. Probing cell mechanics with AFM Due to the unique capabilities of AFM as an imaging tool for scanning the surface of living cells, numerous studies have been conducted comparing such images with alternative modalities. For example, while scanning electron microscopy (SEM) still can resolve greater detail on prepared cell samples [69], some AFM studies have found features of the surface of living cells not observed by SEM [66]. Importantly, AFM also can be used to monitor dynamic cellular processes, including migration and division [96], cytoskeletal reorganization [97], exocytosis [98], and even the response of cancer cells to antitumor drug treatment [99]. Confocal microscopy combined with immunolabeling and fluorescent probes such as green fluorescent protein (GFP) allows imaging the dynamics of living cells with molecular specificity but cannot achieve the spatial resolution of AFM. Therefore, combining imaging modalities with the AFM mounted directly on the stage of an inverted light microscope aids identification of cellular structures observed with AFM [22,23,100]. For example, Fig. 5 shows a human aortic endothelial cell imaged with AFM in contact mode and also with a fluorescent microscope after the cell had been fixed and stained with rhodamine-phalloidin, indicating a strong correlation between the ridge structures observed with AFM and actin stress fibers in the cell. As emphasized above, in addition to imaging the surface topography, AFM is capable of mapping the elastic properties of living cells, which has yielded interesting insights into a number of physiologic cell processes. For example, monitoring the edge dynamics of migrating fibroblasts revealed a thin margin with uniform elastic properties more consistent with an actin-or myosin-based mechanism of extension rather than localized blebbing [94]. Endothelial cells have been observed to get less stiff in the presence of attached monocytes, which indicates a possible mechanism to facilitate monocyte migration, a key process in the inflammatory response in the later stages of atherogenesis [101]. Regional measurements of the mechanical properties of endothelial cells exposed to shear flow show that the upstream edge is earliest to respond with increased stiffening that later extends to the entire cell [102], which suggests localized control of cytoskeletal organization. AFM also has been used to monitor temporal changes in mechanical properties during cell division [103] and revealed an increase in stiffness in the equatorial region that precedes any detectable cleavage furrow by more than two minutes [104]. The AFM has been used to measure characteristic differences in both the elastic and viscous properties of various cell types [69,90,105]. Differences in stiffness also have been observed between the nucleus and cytoplasm of the same cell [59,66,74,102,106,107], and in some cases regions of altered stiffness have been identified by AFM that do not seem to have an obvious anatomical correlate [59]. Substrate-dependent differences in cell elastic properties have been reported as well [108]. To elucidate how cell mechanical properties are related to the structure and function of the underlying cytoskeleton, a number of studies have examined the effects of chemical treatments or genetic mutations that target specific cytoskeletal constituents. In general, the actin cytoskeleton has a dominant effect on cell stiffness measured with AFM [22,23,97,109,110]. Correlation of regional cell mechanics with underlying cytoskeletal components by combining AFM and fluorescent microscopy with immunolabeling showed that actin and intermediate filaments make a major contribution to elastic properties, whereas microtubules make a negligible contribution to cell elastic properties [22,23]. In another study, actin depolymerization by cytochalasin-D decreased elastic and viscous properties of L929 fibroblasts, whereas microtubule depolymerization by nocodazole or colcemid increased elastic stiffness without altering viscous properties [110]. The membrane-associated cytoskeletal binding protein vinculin also was found to play an important functional role in stabilizing focal adhesions such that vinculindeficient mouse embryonic F9 carcinoma cells showed a markedly decreased stiffness compared to wild-type cells [111]. Clearly, cell mechanics is an important indicator of cytoskeletal structure and function [112]. In particular, actin stress fibers are prominent linear structures comprised of actin and myosin [113] that provide a contractile apparatus in many cultured nonmuscle cell types, as well as in vascular endothelial cells in some physiologic conditions [32,114]. AFM force mapping studies show that these structures are very stiff compared to any other cellular component [23]. We have found that, in cultured human aortic endothelial cells, stress fibers were not only stiffer than the surrounding cytoplasm, but the point-wise modulus on the stress fiber increased with indentation depth to maximum values that were large (12-24 kPa) compared with the more uniform values on the surrounding cytoplasm (1-5 kPa) (Fig. 6). When the same region of the cell was examined after 45 minutes' treatment with 4 µM cytochalasin-B, no stress fibers were visible in the AFM image (not shown) and the cell behaved uniformly like a soft, homogeneous material with a constant stiffness (0.5-1.5 kPa). Previous studies have indicated that the cytoskeleton may exhibit strain hardening, with stiffness increasing as deformation increases [25]. Even with linear elastic cytoskeletal elements, such nonlinearity may arise from purely structural considerations, depending on how the elements are interconnected [26]. Our data suggest the cytoskeletal elements themselves may be inherently nonlinear elastic, which is an important distinction for accurately characterizing cellular stresses and for properly identifying the biomechanical fingerprint of the cell. Although the vast majority of AFM elastography studies to date have focused on characterizing normal cells and cell behavior, a few key AFM studies have begun to examine how cell mechanical properties may be altered by disease processes. In particular, a comparison of normal and SV40-transformed human dermal fibroblasts found comparable membrane cortical tension, but the apparent viscosity was 30% lower for the transformed cells, which suggests that such measurements may offer new markers of oncogenic transformation [115]. More recently, Lekka and coworkers have shown that cancerous human bladder epithelial cells have a Young's modulus about one-tenth that of corresponding normal cells [10]. Remarkably, the abnormal modulus values returned toward normal when the cancerous cells were treated with microcrystalline chitosan [116], a drug shown to inhibit glycolytic activity in tumor cells. Such studies may provide critical data for evaluating and understanding candidate therapeutic strategies using single cell analysis, possibly even on a patient-specific basis following biopsy. Taken together with the growing data relating cell mechanical properties to cytoskeletal structure and substrate adhesion, these studies underscore the tremendous potential for AFM elastography of living cells to provide novel biomechanical markers that will enhance the detection, diagnosis, and treatment of disease. Current limitations and future directions for AFM elastography For all of its advantages, the AFM still has a number of limitations that must be overcome to realize the full potential of this unique tool. For instance, accuracy of the cantilever spring constant traditionally has been one limitation of using the AFM for quantitative measurements of mechanical properties. In particular, variations in the thickness [117] and stoichiometry [118] of commercially available AFM cantilevers can result in spring-constant variability of nearly an order of magnitude between batch-produced wafers [61]. This necessitates individual calibration for applications (such as elastography) in which accuracy of the contact force is critical. Fortunately, several methods now exist for nondestructively calibrating AFM cantilevers [61][62][63], and it has been shown that, for a given wafer, springconstant variability of individual cantilevers is within about 10% of the average value for that wafer [93]. Identification of the exact point of contact between the AFM tip and the sample is another source of error in estimating mechanical properties from indentation tests. On stiff samples, the contact point is detected readily as a discontinuity in the slope (first derivative) or a spike in the curvature (second derivative) of the raw force curve [58]. However, this often is ineffective when applied to indentations on soft samples because the transition from pre-contact to post-contact is smooth and obscured by noise in the data. When indenting cells with a pyramidal tip, the problem is exacerbated by the fact that initial contact forces are minimal as the smallest part of the probe tip first contacts the soft cell membrane. Rather than relying on visual inspection of the force curve [88], the contact point may be estimated by including it as an unknown variable when analyzing post-contact data [94], though this is prone to errors with deeper indentations as mentioned above. Alterations in the thermal noise spectrum of the cantilever as it approaches the surface also may reveal contact if the long-range interaction forces are well characterized [119]. Alternative probe-tip geometries or instrumented probes that utilize MEMS technology [120] may allow improved identification of initial cell contact. Another approach is to develop methods of analysis that are insensitive to the precise point of contact, such as FIEL mapping [59]. Ultimately, some independent method for verifying the contact point needs to be developed. Most analysis methods focus on estimating cellular elastic properties and neglect viscoelastic behav- ior. This may be significant particularly since, in some cases, diseased cells may have the same elastic properties as do normal cells but may have altered viscous properties [11,115,121]. In addition to the indentation experiments that have become standard with AFM, it is necessary to develop more sophisticated tests (e.g., creep and relaxation [59,110], frequencydependent microrheology [71]),possibly combining indentation with other modes of cell deformation (e.g., in-plane stretch [91]), to more completely characterize cell mechanical properties. AFM has the flexibility to perform such tests |
but has yet to be fully developed for this purpose. It also is necessary to validate whether micro-scale mechanical properties can be interpreted in the same context as are macro-scale mechanical properties; i.e., to test whether the continuum assumption is valid in the AFM indentation experiment. In particular, Stamenovic and Coughlin recently pointed out that alternative methods for measuring cell mechanical proper-ties (cell poking, micropipette aspiration, and magnetometry) consistently yield Young's moduli of different orders of magnitude [122]. There is an apparent correlation with the length scale over which the cell is interrogated that is consistent with a discrete structural model of the cytoskeleton [122]. On the other hand, these alternative measurement techniques actually may measure different aspects of cell mechanical properties. In fact, a recent study on elastomeric polymer gels showed that, when properly corrected for tip geometry, the Young's modulus determined by AFM using spherical-tipped cantilevers closely matched the results of macroscopic indentation tests [88]. For standard pyramid tips, the microscopic modulus was larger than the macroscopic modulus [88]; but the authors analyzed this data assuming an ideal cone geometry rather than a more accurate blunt cone [91], an error that is known to overestimate Young's modulus values obtained from indentation tests [90,123]. While these studies suggest that some porous polymeric materials behave like a continuum when probed with AFM, similar validation studies must be performed on increasingly complex materials before we can understand fully such measurements of living cells. Another limitation of AFM elastography (and of cell mechanics measurements in general) is that typical experiments are highly time-and user-intensive, so that a limited number of cells can be analyzed in one day. Until the AFM can be used to rapidly evaluate large populations of cells, it will not realize its full potential for basic science and clinical applications, such as screening for disease or evaluating therapeutic treatments. Therefore, an essential modification to expand the use of this technology is to increase throughput. Arrays of multiple probes have been used for rapidly imaging large areas of silicon wafers for the semiconductor industry [124][125][126], though such technology has not been tested yet in cell biology applications. One major challenge is to position automatically such irregular biological samples under the probe tip for consistent and rapid sequential testing. As AFM elastography evolves, the mechanical tests become increasingly sophisticated, and data sets become increasingly large and complex, computational methods and imaging techniques will play a critical role in the analysis and visualization of cell mechanics data. Already, we have identified many limitations when applying the standard Hertz theory, and preliminary finite element models have motivated novel experiments and yielded alternative methods of analysis that promise to increase the information that can be obtained from AFM indentation tests [91]. Finite element models also can accommodate challenging aspects of the AFM indentation problem such as nonaxisymmetry of the tip geometry, inclination angle of the cantilever relative to the cell surface, the irregular topography of the cell, and the more complex cell mechanical properties, including nonlinearity, viscoelasticity, anisotropy, heterogeneity, and even multi-phasic material composition. Such computational methods also will be critical in evaluating alternative theoretical models of the cell, including discrete structurally based models of the cytoskeleton [127]. Recently, the sophistication of such models has increased [92,128], but the AFM indentation problem has yet to be fully characterized. Such detailed computational models also can be used for simulation purposes to better understand how forces at the cantilever tip are manifest as the raw force curve data, as instructional tools to expose a wider population to the AFM, and, ultimately, perhaps as inverse models for extracting cell mechanical properties from individual cell mechanics experiments. Computers also will play an important role in the visualization of increasingly complex elastography data using novel image-processing methods [129]. Summary and perspective In the past decade, the AFM rapidly has become one of the most widely used and versatile tools for studying and physically interacting with living cells. In particular, AFM elastography, which capitalizes on the unique capability of combining mechanical measurements and topological imaging, holds great promise in the field of cell biology. A growing body of data relating cell mechanical properties to cytoskeletal structure and substrate adhesion suggests that single-cell elastography may provide sensitive indicators of the presence of disease. However, it will serve us well to heed the words of Werner Heisenberg who cautioned, ". . . We have to remember that what we observe is not nature in itself but nature exposed to our method of questioning" [130]. A number of technical and practical hurdles remain in the way of obtaining accurate and meaningful cell mechanics measurements with sufficient throughput that they will be practical for reliably examining large populations of cells. Nevertheless, as one powerful method of questioning, the future holds tremendous potential for AFM elastography of living cells to provide novel biomechanical markers that will enhance the detection, diagnosis, and treatment of disease. A Cross-sectional Histo-morphological Study of Vesiculobullous Lesions of Skin Histopathology study of skin biopsies is one of the useful techniques in the investigation of vesiculobullous lesions. Out of the 70 cases of vesicobullous studied, pemphigus vulgaris was the most common type with 44 cases (61.4%) which is followed by 25 cases (35.7%) of pemphigus foliaceus and least percent was found with 1 cases (2.9%) for pemphigus vegetans. Case study showed that disease was predominately observed in male in the ratio of 1.4:1 (25 males and 18 females). It was observed in Pemphigus vulgaris cases that initial lesions involves mucous membranes in 37.2% cases and 13.9% cases involves lesions including both skin and mucosae in 13.9% cases. Mucosal involvement at one time or the other was seen in 31 patients (72.1 %). In despite of this, in pemphigus foliaceus, only 4% cases showed initial mucosal involvement and involvement of both skin and mucous membrane was observed in 12% cases. This establishes that along with clinical correlation, histo-morphological study is also a important tool in the diagnosis of Vesiculobullous skin disorders. Introduction Vesiculobullous disorders represent a heterogenous group of dermatoses with protean manifestations. They have severe economic consequences as well have dramatic impact on the family and health services. These skin diseases have been the subject of broad investigation in recent years [1]. Wide variety of bullous diseases are there, some of which can be very fatal, some bullous lesions may have serious sequele, and required early treatment and intervention to lesser the further morbidity and mortality [2]. Clinical examination of skin bullous lesion provides dermatologist gross morphological finding upon which differential diagnosis can be found out. However histopathological examination is needed for definite diagnosis [3]. A major source of dismay to pathologist is Bullous lesions. Skin biopsie are easily deliberate with precision, direct immunofluorescent microscopy in conjugation with histopathology and it gives the best diagnostic yield in bullous lesions and also make a clear reporting [4]. Case Presentation This study was done at SMS Medical College, Jaipur (Rajasthan). A detailed history with particular reference to the mode of onset, characteristics and distribution of the lesion was taken. Punch biopsy of early vesiculo-bullous lesions were carried out in patients with biopsy punch of 4 mm or 5 mm diameter. The selection of the site for the biopsy is very critical step and lesion should be selected carefully. Histopathological examination may be misleading if lesions is not fully developed and uncomplicated. Lesions as well as normal surrounding tissue both are included in biopsy specimen, so that advancing border of the lesion can be visualized for histopathological examination of vesicles and bullae. It is International Journal of Medical Research and Review Available online at: www.ijmrr.in 393 | P a g e important that the entire lesion be removed intact to permit the study of location of vesicle, the nature of its roof and floor and type, condition and cells present in the blister. The entire specimen should be fixed in formalin and embedded in paraffin [5]. Cytological smear (Tzank smear ) were carried out where it is possible because most of the patient come with older lesions with crusting so it was not possible to take smear in all patients. To make Tzank smear, intact roof of a blister is opened along one side, folded back and the floor is gently scraped. The material thus obtained is smeared on microscopic slide, air dried, and stained with Haematoxylin & Eosin [6]. A vast array of under lying pathologies derive the vesiculobullous lesions, the spectrum of disease encountered ranged from inherited disorders to acquired disease liked drug reactions. Bullous pemphigoid, occurring as multiple, tense bullae of varying sizes commonly in adults, had the highest incidence, followed by pemphigus vulgaris, in which oral lesions were predominant with skin involvement showing flaccid bulla of varying sizes, erythema multiforme, presenting as popular erythematous eruptions caused by a variety of unrelated stimuli, dermatitis herpetiformis, which occurred along with gluten sensitive enteropathy with skin involvement showing vesicles on erythematous bases and pemphigus foliaceus, which had characteristic positive Nikolsky's sign and a few uncommon conditions like superficial pustular dermatoses and Darier's disease were also encountered. Out of the 70 cases of vesicobullous studied, pemphigus vulgaris was the most common type with 44 cases (61.4%) which is followed by 25 cases (35.7%) of pemphigus foliaceus and least percent was found with 1 case (2.9%) for pemphigus vegetans. Male was predominately afflicted with this disease in the ratio of 1.4:1 (25 males and 18 females). Out of the 44 cases of pemphigus vulgaris, 37 patients (88.4%) were of age between 21 and 60 years; and 1 case (4.6%) was below 20 years of age. Only 2 cases (6.9%) were above 60 years of age. Out of the 25 cases of pemphigus foliaceus, 20 patients (80%) were lying in the age group of 21-60 years; while 2 patients (8%) were older than 60 years and 3 patients (12%) were below 20 years. One case of pemphigus vegetans was seen of the age 25 years. It was observed in Pemphigus vulgaris cases that initial lesions involves mucous membranes in 37.2% cases and 13.9 % cases involves lesions both in skin as well as in mucosae. 31 patients were seen to be having mucosal involvement at one time or the other was seen in 31 patients (72.1 %). In despite of this, in pemphigus foliaceus, only 4% cases showed initial mucosal involvement while in 12% cases involvement of both skin and mucous membrane was seen. The nature of the lesions observed during the course of pemphigus is shown in ( Table 1). All cases were found to be suffering from flaccid bullae. Blisters seen arising on nonerythematous skin was seen in 42 cases (95.45%) of pemphigus vulgaris, which spontaneously ruptured to give rise to erosions in 23 cases (52.2%). Crusted lesions, erythematous plaques, vegetations and pustules were present less frequently. Skin lesions were observed on common sites in pemphigus vulgaris were generalised involved in 16 cases (37.2%), on face in 9 Cases (20.9%), scalp in 16 cases (18.6%). On the other hand, in pemphigus foliaceus, lesions were generalised distribute in 8 cases (32%), on face in 7 cases (28%), on scalp in 4 cases (16%), on trunk in 4 cases (56%) and involvement of extremities in 12 cases (48%). In pemphigus foliaceus, blisters arising on erythematous skin were seen in 17 cases (68%), crusted lesions in 10 cases (40%), erosions in 11 cases (44%) and pustules in 8% cases. Of the 44 cases of pemphigus vulgaris, 32 cases (72.7%) showed intra-epidermal suprabasal vesicles and 11 cases (25.0%) showed mid-epidermal vesicles ( Table 2). Acantholysis affecting follicular sheath was seen in 40.9% cases. An inflammatory infiltrate was present in the bulla cavity in 25 cases (56.8%). Neutrophils were predominant in 9 cases (20.9%) and eosinophils 11 cases (25.6%). Midepidermal vesicles which may be due to regeneration of the cells from the floor of the bulla were seen in old bullae. Acantholysis, as groups of cells or single cells within the bulla cavity, was seen in 79.5% cases. Dyskeratosis, basal layer budding and pseudo-epitheliomatous proliferation was not seen in any of the cases. Eosinophilic spongiosis was seen in 1 case (2.2%). This was only known case of pemphigus and had developed new lesions after discontinuation of steroids. In 11 cases (17.6%), the epidermis was lost during the process. |
Previously negative report was observed in 5 of these cases (11.6%) and could be diagnosed on the basis of the above histological features. International Journal of Medical Research and Review Available online at: www.ijmrr.in 394 | P a g e Erythematous plaques 3(06.8%) --6. International Journal of Medical Research and Review Available online at: www.ijmrr.in 395 | P a g e seen in any of the cases (Table 3).Among the 25 cases of pemphigus foliaceus studied, acantholysis was observed in 24 cases (96%) . In 46.5% cases, acantholysis affecting follicular sheath was observed. Subcorneal bulla was seen in 13 cases (52%), subgranular cleavage from middle epidermis in 08 cases (32%). An inflammatory infiltrate was present in the bulla cavity in 23 cases (53.5%). In 3 cases eosinophilic spongiosis was seen (6.9%). Two of them were known cases of pemphigus and it was observed that discontinuation of steriods were followed by development of new lesions . In 8 cases (18.6%), the epidermis was lost during the process. In 9 cases (20.9%) neutrophils were predominant and eosinophils in 11 cases (25.6%). There was negative report was observed in five of these cases (11.6%) and could be diagnosed on the basis of the above histological features. Out of 25 case of pemphigus foliaceus, in 10 cases (83.3%) the infiltrate comprised of polymorphs. Spongiosis was seen in 10 cases (40%) and exocytosis in 8 cases (32%). In a few cases, acanthosis, hyperkeratosis, parakeratosis, papillomatosis and increased pigment formation were also seen. 24 cases (96%) showed acantholysis. Subcorneal bulla was seen in 13 cases (52%), 8 cases were observed to be subgranular cleavage from middle epidermis (32%). Dyskeratosis was seen in only 4 cases (16%).In 12 cases, inflammatory cells were seen in the bulla cavity (48%). Discussion Among the Vesiculobullous, in the pemphigus group, pemphigus vulgaris is the commonest variety [7]. There was 109 cases of pemphigus from Tripoli, libya have observed by Shafi, M et al among that the incidence of pemphigus in Libya is very high, with predominant variant being pemphigys foliaceus [8] . In his study Shafi. M et al also found out that males were predominantly affected than females [8]. In the present study, most of the cases of Pemphigus vulgaris were observed followed by pemphigus foliaceus. A single case of pemphigus vegetans was noticed. Fabbri P et al stated that acantholysis was the major mechanism involved in pemphigus vulgaris and its variant pemphigus vegetans there was suprabasilar clefting, which was the hallmark of these diseases [9]. In pemphigus foliaceus and its variant pemphigus erythematosus, the pathologic alteration of the epidermis occurs superficially i.e in the granular layer [9,10]. The pemphigus group was found to have same observations in the present study, with all the cases of pemphigus group showing acantholysis as the predominant mechanism of bulla formation. It has been International Journal of Medical Research and Review Available online at: www.ijmrr.in 396 | P a g e poorly understood the reason for the intrinsic difference between pemphigus foliaceus and pemphigu vulgaris , but has long been of interest to clinicians as well as the investigators [11]. Funding: Nil, Conflict of interest: None. Permission of IRB: Yes Lansoprazole Upregulates Polyubiquitination of the TNF Receptor-Associated Factor 6 and Facilitates Runx2-mediated Osteoblastogenesis The transcription factor, runt-related transcription factor 2 (Runx2), plays a pivotal role in the differentiation of the mesenchymal stem cells to the osteochondroblast lineages. We found by the drug repositioning strategy that a proton pump inhibitor, lansoprazole, enhances nuclear accumulation of Runx2 and induces osteoblastogenesis of human mesenchymal stromal cells. Systemic administration of lansoprazole to a rat femoral fracture model increased osteoblastogenesis. Dissection of signaling pathways revealed that lansoprazole activates a noncanonical bone morphogenic protein (BMP)-transforming growth factor-beta (TGF-β) activated kinase-1 (TAK1)–p38 mitogen-activated protein kinase (MAPK) pathway. We found by in cellulo ubiquitination studies that lansoprazole enhances polyubiquitination of the TNF receptor-associated factor 6 (TRAF6) and by in vitro ubiquitination studies that the enhanced polyubiquitination of TRAF6 is attributed to the blocking of a deubiquitination enzyme, cylindromatosis (CYLD). Structural modeling and site-directed mutagenesis of CYLD demonstrated that lansoprazole tightly fits in a pocket of CYLD where the C-terminal tail of ubiquitin lies. Lansoprazole is a potential therapeutic agent for enhancing osteoblastic differentiation. The drug repositioning strategy, in which a drug currently used for a specific disease is applied to another disease, has gained increasing attention from both academia and industry in recent years (Abbott, 2002;Rothstein et al., 2005;Bian et al., 2009;Yamamoto et al., 2013;Matsushita et al., 2013;Takamatsu et al., 2014). The advantage of this strategy is that the identified drugs can be readily applied to clinical practice, because the optimal doses, adverse effects, and contraindications are already established. In this study, we identified by the drug repositioning strategy that lansoprazole, a proton pump inhibitor (PPI), upregulates Runx2 and induces osteoblastogenesis. We also proved the effect of lansoprazole in a rat model of femoral fracture in order to explore the possibilities of clinical application. We dissected the effect of lansoprazole and found that lansoprazole induces TRAF6 polyubiquitination, which then activates the noncanonical TAK1-p38 MAPK pathway and upregulates Runx2. Furthermore, in silico analysis predicted and site-directed mutagenesis revealed the binding of lansoprazole to a CYLD pocket and the inhibition of its deubiquitination activity by lansoprazole, which leads to enhanced polyubiquitination of TRAF6. Cell Culture We purchased mouse pluripotent mesenchymal C3H10T1/2 cells and human osteoblastic osteosarcoma (HOS) cells from Riken BioResource Center. We isolated primary bone marrow cells (BMCs) from 6-weekold male Sprague-Dawley (SD) rats as described previously (Takamine et al., 2002). We obtained human BMCs during surgery from 3 patients aged 9 years or younger with idiopathic acetabular dysplasia of the hip or osteonecrosis of the femoral head after appropriate written informed consent was given with prior approval by the ethical review committee of Nagoya University Graduate School of Medicine. We isolated monocyte-enriched fractions from the collected human BMCs by density gradient centrifugation with Ficoll-Paque (GE Healthcare) as described (Kitoh et al., 2004). Mesenchymal progenitors of rat and human were isolated by their binding ability to culture dishes. To seek for ex vivo applicability of lansoprazole in clinical settings, we also expanded human mesenchymal progenitors in StemPro (Gibco) medium. MSCs and mesenchymal progenitors were then subjected to differentiation in osteogenic medium containing 50 μg mL −1 ascorbic acid, 10 mM βglycerophosphate, and 10 −7 M dexamethasone. Additional information is provided in the Supplemental Materials and Methods. Western Blot Analysis and Immunoprecipitation Human MSCs were starved for 1 d in growth medium containing 1% fetal bovine serum (FBS) before addition of 20 μM lansoprazole. After 15 min of incubation, we added 100 ng mL −1 recombinant BMP-2 for 0, 15, and 30 min. Cells were lysed in RIPA lysis buffer (Santa Cruz Biotechnology) supplemented with protease and phosphatase inhibitors. After centrifugation for 15 min at 18,000 ×g, supernatants were subjected to SDS-PAGE, followed by blotting onto PVDF membrane. We collected HOS cells after treatment with or without 20 μM lansoprazole for 2 d. We extracted nuclear and cytosolic fractions using the Nuclear Extraction kit (Cayman) according to the manufacturer's protocols. At the same time, we prepared a whole cell lysate by transferring a part of the cells to RIPA lysis buffer. For immunoprecipitation assay, collected HOS cells were lysed in passive lysis buffer containing 50 mM HEPES, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 1.5 mM MgCl 2 , 1 mM EGTA, 100 mM NaF, and 10 mM sodium pyrophosphate supplemented with protease and phosphatase inhibitors, and incubated for 20 min, followed by centrifugation at 18,000 ×g for 15 min. After a two-fold dilution with the dilution buffer containing 50 mM HEPES, 150 mM NaCl, 0.1% Triton X-100, and 1 mM EDTA supplemented with protease and phosphatase inhibitors, immunoprecipitation was performed by incubation with 2 μg of antibody for overnight, followed by addition of Dynabeads Protein G to the diluted supernatant. The antibodies used are shown in the Supplemental Materials and Methods. In Cellulo Protein Ubiquitination Assay Using Cultured Cells HEK293 cells were transiently transfected with Flag-tagged human TRAF6 cDNA in a CMV-based expression vector, which was a kind gift of Drs. Jun Ninomiya-Tsuji at North Carolina State University and Kunihiro Matsumoto at Nagoya University, using FuGENE 6 (Roche). After 1 d of culture, cells were subjected to complete serum starvation for 1 d, and then pretreated with or without lansoprazole for 30 min, followed by stimulation with or without 100 ng mL − 1 recombinant BMP-2 for the indicated time periods. For immunoprecipitation analysis, cells were washed twice, scraped in ice-cold PBS, and centrifuged at 18,000 ×g for 5 min. Non-covalent protein interactions were dissociated with 1% SDS and boiling for 5 min. Samples were diluted in PBS (1:10) containing 50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% NP-40 supplemented with protease inhibitors, and centrifuged at 18,000 ×g for 15 min. Immunoprecipitation was performed by incubation with an anti-Flag M2 antibody (Sigma-Aldrich) overnight, followed by addition of Dynabeads Protein G (Invitrogen). The resultant immunoprecipitates were subjected to SDS-PAGE, followed by immunoblotting with antibodies against M2 Flag and ubiquitin to visualize TRAF6-associated polyubiquitin chains. In Vitro Protein Ubiquitination Assay in a Test Tube Human ubiquitin, ubiquitin-activating enzyme (E1), and UbcH5c (E2) were purchased from Abcam. Ubc13-Uev1a complex (E2) was purchased from Boston Biochem. Human wild-type CYLD cDNA (Kazusa DNA Research Institute) was cloned into pcDNA3.1(+) vector with a Flag tag at the C-terminal end. An R758A-single-mutant and an R758A and F766A-double-mutant Flag-tagged CYLD expression vectors were constructed using QuikChange site-directed mutagenesis kit (Stratagene). TRAF6 and CYLD were expressed in HEK293 cells and affinitypurified using anti-DYKDDDDK agarose beads (Wako Pure Chemical Ind.). Polyubiquitin chains were synthesized at 30°C for 1 h in a reaction mixture containing 200 ng E1, 100 ng UbcH5c or Ubc13-Uev1a complex, 1-2 μg TRAF6, and 5 μg ubiquitin in buffer B (50 mM Tris-HCl pH 7.5, 10 mM MgCl 2 , 1.5 mM DTT, 5 mM ATP) with the indicated concentrations of lansoprazole. The reaction was terminated by addition of SDS sample buffer (Invitrogen), and subjected to immunoblotting analysis. For deubiquitination reaction, the polyubiquitination chains were synthesized as above but without lansoprazole and the reaction was terminated by adding 10 mM EDTA. The reaction mixture was subsequently pretreated with increasing amounts of lansoprazole for 15 min, followed by incubation with 100-200 nM CYLD in a reaction mixture containing 24 mM Tris-HCl pH 7.5 and 0.7 mM DTT at 30°C for 1 h. Animals All animal care and handling conformed to the Regulations on Animal Experiments in Nagoya University and were approved by the Institutional Animal Care and Use Committee in Nagoya University. We purchased male Jcl:SD rats from CLEA Japan and made a rat model of left femoral fracture. Rats took 7.5 mg kg − 1 d − 1 of lansoprazole (Wako Pure Chemical Ind.) in drinking water, which was about 10 times more than the amount used for humans, and administration of the expected amount of lansoprazole was confirmed by monitoring water intake twice a week. For radiological analysis, 8-week-old rats were administered with lansoprazole or vehicle for 4 weeks (n = 6 per group). For histological analysis, ten-week-old rats were administered with lansoprazole or vehicle for 4 weeks (n = 8 per group). For analyzing an early phase of fracture repair, 8-week old rats were administered with lansoprazole or vehicle for 2 weeks (n = 4 per group). For analyzing a long-term effect of lansoprazole on intact metaphysis, unoperated rats were administered with lansoprazole or vehicle for twelve weeks (n = 4 per group). To label the mineralization front, we subcutaneously injected calcein (Dojindo, 8 mg per kg body weight) in 2% sodium hydrogen carbonate twice before being sacrificed. Rat Model of Fracture Repair All surgical procedures were conducted under sterile conditions. Rats were anesthetized by intraperitoneal injection of 30 mg kg −1 sodium pentobarbital. A lateral incision on the left hindlimb was made to expose the shaft of femur. A 3-mm bone defect was created at the midshaft using a thin oscillating saw (NSK Dental LLC), and then periosteum was stripped circumferentially on both sides of osteotomy to prevent early bone union. Once a 1.2-mm diameter wire (Mizuho Corp.) was inserted from the cross-section into the proximal femoral canal in a retrograde manner, the both ends of osteotomy site were placed in contact with |
each other, and then the wire was reinserted across the osteotomy site into the distal femoral canal to apply stability. All rats were given prophylactic administration of sulfamethoxazole (50 mg kg −1 d − 1 , Sigma-Aldrich) and trimethoprim (10 mg kg −1 d −1 , Sigma-Aldrich), and randomly assigned to receive either lansoprazole or vehicle. At 2 or 4 weeks after fracture, fractured femurs were extracted. Radiological Assessment of Fracture Repair At 1, 2, 3, and 4 weeks after fracture, lateral radiographs of the fractured hindlimb were made using a Multi soft X-ray film shooting apparatus (SOFTEX). Fracture union was defined by the formation of seamless bridging callus on at least one cortex on the lateral radiograph. After extracting femurs, bridging callus formation on 4 cortices was Lansoprazole induces nuclear accumulation of Runx2 and activates its transcriptional activity. (a) Immunoblotting of Runx2 in the nuclear and cytosolic fractions as well as in the whole cell lysate of HOS cells after adding 20 μM lansoprazole for 2 d. The relative expression levels were densitometrically estimated using an essential splicing factor U2AF35 for the nuclear fraction and Gapdh for the cytosolic fraction and the whole cell lysate as a control. Lanso., lansoprazole. (b) Runx2-specific immunofluorescence staining of HOS cells treated with lansoprazole for 3 d. Scale bar, 50 μm. (c) Activity of Runx2-responsive luciferase reporter vector (6xOSE2-luc) in C3H10T1/2 cells. Cells were exposed to 20 μM of lansoprazole for 1 d without adding BMP-2. pCMV6-XL5-Runx2, human RUNX2 cDNA clone. (d) Expression levels of endogenous Spp1 mRNA by real-time RT-PCR in C3H10T1/2 cells. Cells were pretreated with 100 ng mL −1 of recombinant BMP-2 for 3 d, followed by incubation with the indicated concentrations of lansoprazole for 14 d. The relative expression levels were normalized to the mean of vehicle. (e) ALP activity of cell lysates of human bone marrow-derived mesenchymal cells. Cells were cultured in osteogenic medium for 19 d and lansoprazole was added from days 15 to 19. The relative activity was normalized to the mean of vehicle. (f) ALP staining of human bone marrow-derived mesenchymal cells, which were treated with the indicated concentrations of lansoprazole. Cells were cultured in osteogenic medium for 24 d and lansoprazole was added from days 18 to 24. Scale bar, 5 mm. In (c-e), the mean ± SD (n = 6) are indicated. *p b 0.0001 by unpaired t-test (c) and by the Jonckheere-Terpstra trend test over the indicated concentration range of lansoprazole (d and e). (d, e, and f) In the course of our analysis, 5-20 μM lansoprazole exhibits consistent dose-response effects, but data at concentrations less than 5 μM and more than 20 μM have not been omitted under the expectation that these concentrations will be of some help for future references. (e) Expression levels of endogenous RUNX2 mRNA by real-time RT-PCR in human MSCs. The relative expression levels were normalized to the mean of vehicle. Sequential protocol of lansoprazole administration is indicated at the bottom. Lower arrowheads indicate the time points when we passaged cells. (f) Matrix calcium deposition by Alizarin red staining in human bone marrow-derived mesenchymal cells that were induced to differentiate immediately after isolation. Cells were cultured in osteogenic medium for 21 d and lansoprazole was added from days 15 to 21. Scale bars, 5 mm for dishes and 200 μm for magnified images. Note that the temporal protocol of human bone marrow-derived mesenchymal cells in (f) is different from that of human MSCs in (e), because the time courses of differentiation of these cells are different. (g) Matrix calcium deposition by Alizarin red staining in human bone marrow-derived mesenchymal cells that were induced to differentiate after proliferation for 23 d. Cells were proliferated in non-osteogenic medium for 23 d followed by additional culture in osteogenic medium up to day 44, and lansoprazole was added from days 24 to 44. In (f), cells were differentiated immediately after isolation, whereas in (g), cells were first proliferated for 23 d to increase the number of cells. Scale bars, 5 mm. In (a, b, and e), the relative expression levels were normalized to the mean of vehicle, and the mean ± SD (n = 6) are indicated. *p b 0.01 (a), *p b 0.0001 (b), *p b 0.01 (c, days 0-10), *p b 0.001 (c, days 11-17), and *p b 0.002 (c, days 18-24) by the Jonckheere-Terpstra trend test over the indicated concentration range of lansoprazole. In (c and d), the densitometric values were normalized to that of Gapdh and also to vehicle. evaluated on the anteroposterior and lateral radiographs, and the number of seamless bridging calluses in the 4 cortices was counted (0 to 4). In addition, bridging callus area was measured on ImageJ software. Uniting (intramedullary) callus formation was determined by the disappearance of fracture line on the anteroposterior and lateral radiographs and its number was counted (0 to 2). Bone Histological and Morphometrical Analyses Bone specimens were sequentially dehydrated with 70% ethanol, 90% ethanol, and acetone. The specimens were then embedded without decalcification in methyl methacrylate. We prepared 6-μm sagittal sections of the whole femur and 30-μm transverse sections at the Fig. 3. Orally administered lansoprazole accelerates fracture healing in a rat model of femoral fracture. (a) Representative radiographs of vehicle-and lansoprazole-administered rats at 1, 2, 3, and 4 weeks after fracture. Formation of seamless bridging callus (black arrowhead) and incomplete formation of uniting callus (white arrow) are shown. The proportion of rats with the fracture union is indicated in the inset. LAT, lateral view, AP, anteroposterior view. (b) The number of rats showing the indicated number of seamless bridging calluses (left) and complete uniting calluses (right) (n = 6 per group). Bridging callus formation on 4 cortices was evaluated on the anteroposterior and lateral radiographs, and the number of calluses was counted (0 to 4). Uniting callus formation was defined by the disappearance of fracture line on the anteroposterior and lateral radiographs, and the number of calluses was counted (0 to 2). (c) Representative images of new bone formation within the anchoring callus assessed by double calcein labeling at 4 weeks after fracture. *p b 0.03 by unpaired t-test. osteotomy sites with a microtome, and stained them with Villanueva Goldner stain. Bone parameters of the fracture sites were blindly evaluated by computer-assisted histomorphometry using Osteomeasure software (OsteoMetrics). For analysis of bone regeneration in the fracture sites, we compared the area of newly formed bone with that of granulation tissue, and calculated bone volume per granulation tissue volume (BV/TV) without setting definite region of interest (ROI) in the fracture sites. We summed bone histomorphometric values of sagittal and transverse sections to assess osteoblastogenesis in the fracture sites. In the fractured femurs, after excluding pin track areas in the distal femur, we defined the ROI in the metaphysis as the areas between 0.7 mm and 3.0 mm proximal to the growth plate in order to exclude the primary trabecular spongiosa. Cartilaginous callus area was measured using histological sections stained with alcian blue. Total callus area was defined as the sum of bridging callus area and anchoring callus area, and was measured by ImageJ software. Histological images were obtained using a BX51 (Olympus) with an UPlanFL N 20× and 0.50 objective lens (Olympus) fitted with a DXC-390P camera (Sony) and a microscope M60 (Leica) with a 1.0× objective lens fitted on a camera IC80 (Leica). In Cellulo Osteoclast Formation RAW 264.7 cells (Riken BioResource Center), a mouse macrophagelike cell line, were seeded onto a 96-well plate at a cell density of 2 × 10 5 cells per well and cultured for 4 d with or without lansoprazole in the presence of receptor activator of NF-κB ligand (RANKL) (100 ng mL −1 ; Peprotech). Tartrate-resistant acid phosphatase (TRAP)positive cells with more than 3 nuclei were counted as osteoclasts. In Cellulo Bone Resorption Assay The pit formation assay was performed using a calcium phosphatecoated plate obtained from PG research (Kodaira, Japan). RAW 264.7 cells grown in α-MEM medium were cultured in a 48-well calcium phosphate-coated plate (5 × 10 3 cells per well) with or without lansoprazole in the presence of RANKL (100 ng mL −1 ) for 7 d. After removing the cells by washing the plate with 5% sodium hydrochloride solution, digital images of the resorption pits were captured. The pit area was measured following conversion of the original images to binary images with NIH ImageJ software. Statistical Analysis All data are presented as the mean and SD. We evaluated doseresponses using the Jonckheere-Terpstra trend test for the whole range of drug concentrations. Statistical difference between 2 groups was determined by the unpaired t-test. We considered a p value b 0.05 as statistically significant. Lansoprazole Upregulates Runx2 Runx2 is translocated into the nucleus where it binds to the cisacting osteoblast-specific element 2 (OSE2) on promoters of the osteogenic genes including Alpl encoding alkaline phosphatase (ALP) and Spp1 encoding osteopontin (Otto et al., 2003). We screened 1186 FDAapproved compounds for activating the human RUNX2 P1 promoter in mouse pluripotent mesenchymal C3H10T1/2 cells and found that lansoprazole increased the RUNX2 P1 promoter activity in a dosedependent manner (Fig. S1a). We found that lansoprazole enhanced intranuclear accumulation of Runx2 (Figs. 1a, b, S1b, and c), which subsequently stimulated the OSE2 reporter plasmid (Fig. 1c) and upregulated Spp1 mRNA expression (Fig. 1d) as well as ALP activity ( Fig. 1e and f). We also confirmed that lansoprazole induced expressions of Runx2 mRNA and Runx2 protein in mouse osteoprogenitor cells (Fig. 2a and c) and human osteoblastic cells ( Fig. 2b and d). In addition, lansoprazole increased expressions of Runx2 mRNA at all stages of osteogenic differentiation in both human MSCs (Fig. 2e) and rat BMCs (Fig. S1d). We used BMP-2 to differentiate mouse C3H10T1/2 cells into mature osteoblasts. In contrast, we used osteogenic medium to differentiate human and rat primary BMCs, because we hoped to examine the applicability of lansoprazole in clinical settings. In surgical lengthening of limbs, we differentiate MSCs in patient's bone marrow aspirates into osteoblastlike cells ex vivo and inject the differentiated cells into the distraction gap to promote bone formation (Kitoh et al., 2004). We thus added lansoprazole to cultured bone marrow cells that were obtained from pediatric patients undergoing acetabular osteotomy after appropriate written informed consents were given, and found that lansoprazole accelerated the matrix calcium deposition in a dose-dependent manner (Fig. 2f). We next examined the effect of lansoprazole on expanded MSCs, and observed more prominent effects (Fig. 2g) rather than those on the immediately differentiated cells (Fig. 2f). As longer time exposure to 40 μM lansoprazole decreased the number of cells and subsequently reduced the matrix calcium deposition (Fig. 2g), we used 20 μM lansoprazole as a default concentration in the following studies. Lansoprazole Accelerates Fracture Repair and Osteoblastic Differentiation in a Rat Fracture Model To address the clinical relevance of lansoprazole for bone regeneration, we examined the effect of lansoprazole on bone regeneration in a rat model of femoral fracture (Mills and Simpson, 2012). The rats took lansoprazole in drinking water for 4 weeks, which was about 10 times higher than the regular dose used for gastroduodenal ulcer in human. A human equivalent dose (HED) for rodents is 6 to 13 times higher than the human dose (Reagan-Shaw et al., 2008), and the HED of 10 is frequently applied to rodents (Dourson et al., 1996;Foronda et al., 2007;Bian et al., 2009). No significant differences in body weights were observed between both groups at baseline and during the experimental period (Fig. S2). At 3 weeks after fracture, 5 of the 6 rats (83%) achieved fracture union in the lansoprazole-treated group, whereas only 1 of the 6 rats (17%) achieved fracture union in the vehicletreated group. At 4 weeks, fracture union was observed in all of the lansoprazole-treated rats, whereas it was observed in 4 rats (67%) in the vehicle group (Fig. 3a). The total callus areas, however, were not significantly different between the 2 groups at 4 weeks (Fig. S3a). The formation of bridging callus and uniting callus, which succeeds the formation of anchoring callus (Marino et al., 1979), tended to be better in the lansoprazole group than the vehicle group (Fig. 3b). Dynamic analysis of bone formation with calcein double-labeling showed that lansoprazole enhanced new osteoblastic bone formation in the anchoring callus (Fig. 3c). |
Histomorphometric analyses at 4 weeks after fracture demonstrated that compared with vehicle-treated rats (n = 8), lansoprazole-treated rats (n = 8) significantly increased the ratios of bone volume/tissue volume (BV/TV), osteoid surface/bone surface (OS/BS), and osteoblast surface/bone surface (Ob.S/BS) in the gap of the fracture site (Figs. 4a, b, S3b and Table S1). Similarly, at the metaphysis that was penetrated by a wire, lansoprazole significantly enhanced the ratios of osteoid volume/bone volume (OV/BV) and osteoid surface/bone surface (OS/BS), while the bone volume/tissue volume (BV/TV) remained unchanged. In contrast, osteoclast number/bone surface (N.Oc/BS), an osteoclastic parameter, was significantly decreased at the injured metaphysis (Figs. 4c,d,S3b,and Table S2). In addition, at 2 weeks after fracture, when the formation of cartilaginous callus and its subsequent resorption by osteoclasts take place, systemic administration of lansoprazole significantly inhibited osteoclastogenesis but not chondrogenesis at the fracture site ( Fig. S4a-c). Taken together, lansoprazole can facilitate osteoblast differentiation and inhibit osteoclast differentiation around the injured site of the bone such as the anchoring callus and the metaphyses penetrated by a wire. Lansoprazole Activates the Noncanonical BMP-TAK1-p38 MAPK Signaling We next pursued molecular mechanisms of the effects of lansoprazole. We first found that lansoprazole induce expressions of Runx2 gene even when BMP-2 was not added (Fig. 5a), although the effect was less prominent compared to that observed with BMP-2-added C3H10T1/2 cells (Fig. 2a). Native C3H10T1/2 cells have been shown to (c) Effects of BMP signaling-associated protein kinase inhibitors on lansoprazole-induced increases of RUNX2 P1 promoter activity. C3H10T1/2 cells that stably express pGL4-hRunx2P1-luc2 (C3H10T1/2-hRunx2P1 cells) without BMP-2 pretreatment were treated with the indicated kinase inhibitors for 1 h and subsequently treated with 20 μM lansoprazole for 1 d. In addition, we expected future clinical application of lansoprazole, and we decided to use 20 μM in later analyses. The luciferase activities were normalized to those without lansoprazole and the kinase inhibitors and also to the protein amount of cell lysates. BMPRI kinase inhibitor, dorsomorphin (1 μM); Erk1/2 inhibitor, U0126 (20 μM); p38 inhibitor, SB203580 (20 μM); JNK inhibitor, SP600125 (10 μM); and TAK1 inhibitor, 5Z-7-oxozeaenol (1 μM). *p b 0.0001 by unpaired t-test. (d) Phosphorylation of BMP signaling-associated proteins in human MSCs. Relative levels indicate densitometric ratios of phosphorylated to total proteins normalized to the ratio without lansoprazole. (e) Effects of TAK1 and p38 MAPK inhibitors on lansoprazole-induced increases of ALP activity in human MSCs. *p b 0.02 and **p b 0.0003 by unpaired t-test. Human MSCs (LONZA) were cultured in the osteogenic medium with or without 20 μM lansoprazole, 100 nM TAK1 inhibitor (5Z-7oxozeaenol), and 5 μM p38 MAPK inhibitor (SB203580) for 14 d. (f) Lansoprazole activates interactions of Runx2 and CBP. HOS cells were treated with or without 20 μM of lansoprazole for 2 d, and immunoprecipitated with anti-Runx2 antibody. Immunoprecipitated CBP was visualized by immunoblotting. In (a-c), the mean ± SD (n = 6) are indicated. NS, not significant by the Jonckheere-Terpstra trend test (a and b). produce a small but nonnegligible amount of endogenous BMP-2 (Phimphilai et al., 2006). To confirm that lansoprazole required BMPs, we blocked the binding of BMPs to BMP receptors with noggin, and found that noggin abrogated the effect of lansoprazole (Fig. 5b). Accordingly, lansoprazole cannot substitute for the activity of BMPs to induce osteoblastogenesis in C3H10T1/2 cells, but can enhance the effect of BMPs. We thus scrutinized BMP-signaling pathways leading to Runx2 activation using kinase inhibitors. Inhibitors of p38 MAPK and TAK1 markedly attenuated the effect of lansoprazole, whereas inhibitors of BMP type I receptor (BMPRI) and Erk1/2 had little or no effects (Fig. 5c). Similarly, lansoprazole increased phosphorylation of p38 MAPK and TAK1, while phosphorylation of Smad1/5/8 and Erk1/2 remained unchanged (Fig. 5d). Conversely, inhibitors of p38 MAPK and TAK1 significantly suppressed lansoprazole-induced ALP activity in human MSCs (Fig. 5e). We also confirmed that lansoprazole enhanced the association of Runx2 with a transcriptional coactivator CREB-binding protein (CBP) (Fig. 5f). Taken together, lansoprazole specifically activates the noncanonical BMP-TAK1-p38 MAPK signaling pathway by directly potentiating a molecule downstream of BMPR but upstream of TAK1, and enhances osteoblastic differentiation. Lansoprazole consequently reinforced the ability of Runx2 to recruit a transcriptional activator. Lansoprazole Activates TRAF6-anchored Polyubiquitination We next looked for a target molecule that is activated by lansoprazole and is located downstream of BMPR and upstream of TAK1. An ubiquitin ligase TRAF6 directly binds to the TGF-β-BMP receptors, and is an immediate downstream effector of the receptors (Landstrom, 2010). TRAF6catalyzes TGF-β-BMP-induced Lys63-linked autopolyubiquitination of TRAF6 itself, which enhances autophosphorylation of TAK1 and subsequently activates NF-κB (Chen, 2005). Unlike Lys48-linked polyubiquitination, which is operational in ubiquitin-proteasome-dependent protein degradation, Lys63-linked polyubiquitination serves to modulate a signaling activity of the polyubiquitination-bearing molecule. We found that lansoprazole increased the NF-κB-responsive luciferase reporter activity (Fig. 6a). Introduction of a dominant-negative TRAF6 inhibited the effect of lansoprazole by 60.8 ((1.308-0.512)/1.308) ± 13.9 (0.182/ 1.308)% (mean and SD, n = 6) (Fig. 6b). As these experiments pointed to the notion that TRAF6 is likely to be a direct target of lansoprazole, we next performed a ubiquitination assay of TRAF6. Upon overexpression of TRAF6 and BMP-2 induction, lansoprazole enhanced TRAF6-anchored polyubiquitination in HEK293 cells and primary osteoblasts (Figs. 6c and S5a). Furthermore, lansoprazole induced TRAF6-anchored polyubiquitination in a serum-free medium and without BMP-2 induction (Fig. 6d), which suggested that TRAF6 was likely to be a direct target molecule activated by lansoprazole. We also found that lansoprazolemediated TRAF6-anchored polyubiquitination was indeed linked to Lys63 by immunoblotting (Fig. S5b and c). TRAF6 can bind to BMP type II receptor (BMPRII) in the absence of BMPRI. On the other hand, TRAF6 can bind to BMPRI only when a BMPRI-II complex is formed by BMPs (Yamashita et al., 2008). As lansoprazole enhanced polyubiquitination of TRAF6 even in the absence of BMP ligands, we examined if lansoprazole is able to activate TRAF6 tethered on BMPRII. To this end, we added a TRAF6-inhibitory peptide, which functions as a TRAF6 decoy by binding to the TRAF6-binding motif on BMPRI (Poblenz et al., 2007). As expected, we found that the TRAF6-inhibitory peptide did not compromise lansoprazole-mediated activation of RUNX2 P1 promoter activity (Fig. S5d). Lansoprazole is thus likely to operate on BMPRII-engaged TRAF6 in the absence of BMP ligands. Lansoprazole Attenuates Deubiquitination Activity of CYLD In Vitro To prove that the TRAF6 autopolyubiquitination is indeed a target of lansoprazole, we examined the effect of lansoprazole by an in vitro ubiquitination assay (Fig. S6a). Unexpectedly, however, lansoprazole failed to enhance the synthesis of TRAF6-anchored polyubiquitination (Fig. 7a). As binding of a small compound to a specific domain of a target molecule mostly inhibits the target molecule rather than activating it, we searched for a molecule that antagonizes TRAF6 polyubiquitination. We found that a deubiquitination enzyme, CYLD, specifically cleaves Lys63-linked polyubiquitin chains, and downregulates TRAF6mediated signal transduction, which has been characterized in NF-κB activation (Trompouki et al., 2003). As has been previously reported (Xia et al., 2009), CYLD was able to cleave unanchored polyubiquitin chains but not TRAF6-anchored ones in vitro (Fig. S6b). We then examined the effect of lansoprazole on CYLD using unanchored polyubiquitin chains, and found that the cleavage was inhibited by pretreatment of lansoprazole in a dose-dependent manner (Fig. 7b). Lansoprazole Stably Fits in a Pocket of CYLD We next asked if lansoprazole is able to bind to CYLD. An in silico search for ligand-binding sites of CYLD disclosed a unique pocket (Fig. 7c). The pocket was located across which the C-terminal tail of ubiquitin was predicted to lie according to structural alignment of CYLD to homologous HAUSP-USP7 that was crystallized with ubiquitin. Optimization of the docking structures predicted that lansoprazole linked to CYLD by hydrogen bond, σ-п conjugation, and п-п interaction (Fig. 7d). The docked structure model suggests that lansoprazole suppresses deubiquitination activity of CYLD by inhibiting the binding of the C-terminal tail of ubiquitin to CYLD. The active site of CYLD is predicted to be located where the C-terminal tail of ubiquitin ends. To prove that lansoprazole binds to the predicted pocket of CYLD, and inhibits its deubiquitination activity, we next employed the Ubc13-Uev1a complex as an E2 enzyme, which specifically catalyzes unanchored Lys63-linked polyubiquitin chains (Xia et al., 2009). We first confirmed that lansoprazole attenuates wild-type CYLD-mediated cleavage of the polyubiquitin chains (Fig. 7e). As simulation of serial alanine substitutions in the identified CYLD pocket predicted that R758 and R766 were essential residues for the binding to lansoprazole, we engineered an R758A-single-mutant CYLD and an R758A and F766Adouble-mutant CYLD. Each mutant CYLD retained a dose-dependent deubiquitination activity in vitro ( Fig. S6c and d), whereas the single mutant CYLD partly and the double mutant CYLD completely abolished lansoprazole-mediated suppression of the deubiquitination activities ( Fig. 7f and g). Thus, the mutations exclusively affected responsiveness of lansoprazole but not the deubiquitination activity of CYLD in the absence of lansoprazole. These results indicated that specific binding of lansoprazole to the CYLD pocket prevents the C-terminal tail from reaching the active site, which then facilitates TRAF6-mediated polyubiquitination (Fig. 8). Lansoprazole Increases Osteoclast Maturation and Bone Resorption Activity To verify whether CYLD is the molecular target of lansoprazole from other aspects, we examined the effects of lansoprazole on osteoclast differentiation and bone-resorbing function using RAW 264.7 cells, in which a critical role of CYLD and TRAF6 has been shown downstream of receptor activator of NF-κB (RANK) -RANKL signaling Tanaka et al., 2003). Lansoprazole significantly increased the proportion of large osteoclasts at a concentration of 1 to 2.5 μM compared to vehicle without affecting the total number of osteoclasts (Fig. 9A). On the other hand, higher concentrations of lansoprazole more than 10 μM markedly decreased osteoclasts formation compared to vehicle probably due to cytotoxicity (Fig. 9a). Bone resorption activity, evaluated by the pit formation assay, was significantly increased by addition of lower concentrations of lansoprazole less than 1 μM (Fig. 9b). Lansoprazole is thus capable of enhancing osteoclast maturation and bone resorbing activity in cellulo. Lansoprazole Enhances Osteoclast Differentiation and Induces an Osteomalacia-like Condition In Vivo To explore the in vivo effect of lansoprazole-mediated enhancement of osteoclast maturation and sustained activation of Runx2, we examined both acute and chronic effects of lansoprazole on physiological cancellous bone remodeling in rats. To this end, rats were fed with the same amount of lansoprazole as that in the fracture model for 4 and 12 weeks, and unfractured femurs were harvested. No significant differences in body weights were observed between both groups at baseline and during the experimental period (Fig. S7). Systemic administration of lansoprazole for 4 weeks significantly elevated osteoclastic parameters, whereas static and dynamic osteoblastic parameters remained unchanged (Figs. 10a and S3c). In contrast, administration lansoprazole for 12 weeks induced an osteomalacia-like condition such as decreased trabecular thickness Fig. 6. Lansoprazole upregulates TRAF6 polyubiquitination. (a) Activity of NF-κB-responsive-element (RE) luciferase reporter vector in C3H10T1/2 cells. Lanso., lansoprazole (20 μM); TNF-α (1 ng mL − 1 ). *p b 0.0001 by unpaired t-test. (b) Effects of a dominant negative TRAF6 (pDeNy-hTRAF6) on a lansoprazole-induced increase of the RUNX2 P1 promoter activity. A relative luciferase activity was normalized to that of vehicle, and a difference to the vehicle is shown by the mean and SD (n = 6). Lanso., lansoprazole. (c) Lansoprazole enhanced TRAF6-anchored polyubiquitination with BMP-2 induction in HEK293 cells. (d) Lansoprazole induced TRAF6-anchored polyubiquitination without BMP-2 induction in HEK293 cells. and increased osteoid thickness in the absence of enhanced osteoclastogenesis (Figs. 10b and S3d). Discussion Our drug repositioning study revealed that lansoprazole, a proton pump inhibitor, has bone anabolic activity through upregulation of Runx2 and serves as an enhancer of BMP signaling. Using the drug repositioning strategy, Mundy and colleagues identified that statins that are commonly prescribed for hyperlipidemia enhance the BMP2 promoter activity (Mundy et al., 1999). Similar to lansoprazole, statins stimulate bone formation in animal models (Gutierrez et al., 2008). However, statins may (Edwards et al., 2000) or may not (Esposito et al., 2013) increase bone mineral density and prevent osteoporotic fractures in postmenopausal women. BMP-2 is a potent osteoinductive agent that is required for the initiation of fracture healing (Tsuji et al., 2006) and for the development and maturation of osteoblasts (Chen et al., 2012a). BMP-2, however, is also capable of inducing pluripotent mesenchymal cells into multiple lineages such as osteoblasts, chondrocytes, and adipocytes (Date et al., 2004). Controversial effects on bone metabolism of statins may be attributed |
to the diverse effects of BMP-2. On the contrary, Runx2 exclusively induces osteochondroblastogenesis (Komori, 2010), but Runx2 alone is not sufficient to induce the maturation of osteoblasts (Lee et al., 2000). Runx2 is upregulated by BMPs (Phimphilai et al., 2006), and the two molecules orchestrate to induce the development and maturation of osteoblasts (Leboy, 2006). Upregulation of BMPs by statins and upregulation of Runx2 by lansoprazole are expected to coordinately accelerate osteoblastogenesis, but further studies are required. BMP ligands bind to BMPRII, and facilitates the formation of a heterotetrameric receptor complex comprised of BMPRI and BMRPII, which then activates the canonical pathway through phosphorylation of the receptor-regulated Smads as well as the noncanonical MAPK pathways (Katagiri and Tsukamoto, 2013). For Runx2 activation, both canonical and noncanonical pathways play essential roles in mesenchymal precursor cells (Lee et al., 2002). Among the noncanonical pathways, activation of the TAK1-p38 MAPK axis is essential for Runx2 activation in mesenchymal precursor cells (Greenblatt et al., 2010). BMP-mediated formation of a BMPRI-II complex additionally induces accumulation of TRAF6 on the complex. TRAF6 is a ubiquitin ligase as well as a substrate polyubiquitinated by itself. The accumulation of TRAF6 enhances the synthesis of Lys63-linked autopolyubiquitin chains. The polyubiquitin chains confer a scaffold for the TAB-mediated adjacent placement of two TAK1 molecules, which leads to autophosphorylation of TAK1 (Xia et al., 2009;Landstrom, 2010;Sorrentino et al., 2008). The phosphorylated TAK1 then phosphorylates p38 MAPK. The activated p38 MAPK also activates other transcriptional factors that are essential for bone formation including Osterix and distal-less homeobox 5 (Dlx5) (Ulsamer et al., 2008;Ortuno et al., 2010). Dissection of underlying molecular mechanisms of the effect of lansoprazole revealed that lansoprazole has little or no effects on the canonical pathway, but activates the noncanonical TRAF6-TAK1-p38 MAPK pathway. The tumor suppressor CYLD specifically digests Lys63-linked polyubiquitin chains that are attached to the TNF receptor-associated factors, TRAF2 and TRAF6, as well as an essential NF-κB signaling component, NF-κB essential modulator (NEMO) (Brummelkamp et al., 2003;Kovalenko et al., 2003;Trompouki et al., 2003). Among the CYLDcatalyzed molecules, TRAF6 is the only molecule in the TGF-β-BMP pathway (Xie, 2013). In previous in vitro studies using purified proteins, CYLD was able to cleave unanchored polyubiquitin chains, but not polyubiquitin chains conjugated to TRAF6 (Xia et al., 2009). An ubiquitin-conjugating E2 enzyme, UbcH5c, catalyzes both Lys63-and Lys48-polyubiquitin chains either anchored or unanchored to TRAF6, whereas another E2 enzyme, Ubc13, efficiently catalyzes unanchored Lys63-polyubiquitin chains (Xia et al., 2009). We thus examined the effect of lansoprazole on the deubiquitination activity of CYLD using unanchored polyubiquitin chains synthesized with UbcH5c (Fig. 7b) and Ubc13 in vitro (Fig. 7e), and we only used Ubc13 for the subsequent mutagenesis studies of CYLD ( Fig. 7f and g). It is interesting to note that even TRAF6-catalyzed unanchored polyubiquitin chains are capable of activating TAK1 (Xia et al., 2009). Thus, in addition to the induction and acceleration of TRAF6-anchored polyubiquitin chains by lansoprazole ( Fig. 6c and d), stabilization of unanchored polyubiquitin chains by lansoprazole ( Fig. 7b and e) may partly account for the activation of TAK1. TRAF6 is an essential factor for osteoclastogenesis as evidenced by osteopetrotic phenotypes in TRAF6-deficient mice (Lomaga et al., 1999;Naito et al., 1999). TRAF6 is indispensable for signal transduction downstream of RANK-RANKL pathway by binding to the cytoplasmic tail of RANK (Inoue et al., 2000). Dissection of TRAF6 domains demonstrated that different domains of TRAF6 regulate osteoclast differentiation, maturation, and function (Kobayashi et al., 2001). The RING finger domain of TRAF6, which is critical for the ubiquitin E3 ligase activity, is not crucial for osteoclast differentiation but is essential for full maturation and activation of osteoclasts. Likewise, CYLD-deficient mice, in which more polyubiquitinated TRAF6 was accumulated, had enlarged osteoclasts . In addition, bone marrow cells from the CYLD-deficient mice exhibited enhanced expressions of osteoclastic marker genes upon RANKL stimulation. Therefore, lansoprazolemediated increased osteoclast maturation and bone resorbing activity in cellulo, which were also demonstrated at the intact metaphyses of the unfractured femurs in vivo, provide supportive evidence that CYLD and TRAF6 are the molecular targets of lansoprazole. In contrast, in a rat model of femoral fracture, lansoprazole promotes osteoblastic maturation and impairs osteoclastogenesis both at the fracture site and the injured metaphyses, where BMP is induced for bone regeneration. Although the mechanisms underlying the different effects of lansoprazole on osteoclastogenesis between fractured and unfractured bones remain unknown, suppression of osteoclastogenesis at the fracture site by lansoprazole is predicted to be favorable for facture healing. The FDA announced a safety alert regarding the possible association between the risk of fractures and the use of PPIs in aged individuals (Ngamruengphong et al., 2011), although the exact mechanisms remain controversial. The adverse effect may be attributed to decreased absorption of insoluble calcium due to the elevated gastric pH observed in aged women taking PPIs (O'Connell et al., 2005). Along with this observation, mice deficient for Cckbr encoding a gastrin receptor that facilitates gastric acid secretion have modest hypocalcemia (Schinke et al., 2009), and mice deficient for gastric H + -K + ATPase beta-subunit also show decreased bone mineral density (Fossmark et al., 2012). A retrospective study, however, shows that patients taking high doses of PPIs do not have low bone mineral density or increased rate of bone loss (Targownik et al., 2010). We demonstrated that lansoprazole could induce osteopenia and fragility at the metaphyses of unfractured long bones by alteration of the trabecular structures to an osteomalacia-like condition and increased bone resorption of the trabeculae. Previously, the maturational blockage of osteoblasts and osteopenia have been reported in mice overexpressing Runx2 (Liu et al., 2001). PPIs-mediated increased chance of fragility fractures is thus likely to be attributed to sustained activation of Runx2 and partly to enhancement of osteoclast maturation. Lansoprazole and other PPIs have been widely used for more than 20 years in the prevention and treatment of the acid-related disorders such as gastroduodenal ulcers and reflux esophagitis, although the molecular bases of the proton pump-inhibiting effects of lansoprazole remains elusive. In spite of their efficacy, long-term use or high dose of PPIs have been implicated to provoke several adverse effects including increased pneumonia, Clostridium difficile-associated disease, fragility-related fractures, neoplasms, iron deficiency, and acute interstitial nephritis (Moayyedi and Leontiadis, 2012;Chen et al., 2012b). As chronic activation of TGF-β-BMP signal transmission causes detrimental cellular processes including fibrosis, atherosclerosis, tumorigenesis, cancer metastasis, and autoimmune disorders (Akhurst and Hata, 2012), enhancement of the TGF-β-BMP signaling by lansoprazole may partly account for some of these adverse effects. A recent report showed that osteogenic induction of human MSCs in the presence of lansoprazole (10-1000 μM) for a longer period (N14 d) decreased ALP activity (Costa-Rodrigues et al., 2013). In addition, higher concentrations (100-1000 μM) of lansoprazole for 7 d or more attenuated cell proliferation by inducing apoptosis. In contrast, we confirmed increased ALP activity in human bone marrow-derived mesenchymal cells that were treated with 5-40 μM lansoprazole for up to 6 d ( Fig. 1e and f). Therefore, even in fresh bone marrow cells, longer exposure of lansoprazole may inhibit ALP activity at lower concentrations. An offlabel higher dose of lansoprazole in the context of osteogenesis promotion is thus expected to be limited to either short-term or local usage. We have started developing a biodegradable artificial bone that enables a sustained release of concentrated lansoprazole. We believe that the combined use of lansoprazole with biomaterials would lead to the Fig. 10. Systemic administration of lansoprazole enhances osteoclastogenesis and induces osteomalacia-like change at the metaphyses of unfractured rat femurs. (a) Histomorphometric analyses of undecalcified histology in rats that were administered with lansoprazole for 4 weeks. The mean ± SD (n = 4) are indicated. *p b 0.05 (ES/BS) and *p b 0.03 (Oc.S/BS) by unpaired t-test. Lanso., lansoprazole; and NS, not significant. (b) Histomorphometric analyses of undecalcified histology in rats that were administered with lansoprazole for 12 weeks. The mean ± SD (n = 4) are indicated. *p b 0.02 (Tb.Th) and *p b 0.04 (O.Th) by unpaired t-test. Lanso., lansoprazole; and NS, not significant. invention of a unique bone graft substitute that possesses the capacity of osteoinduction in addition to osteoconduction. In addition, lansoprazole along with the osteogenic medium may be able to enhance osteoblastic differentiation of bone marrow derived mesenchymal stromal cells ex vivo. Although detailed absorption, distribution, metabolism, excretion, and toxicity (ADMET) analysis is required for ex vivo application of lansoprazole, the probability of unpredicted adverse effects is expected to be lower than chemical compounds that have never been used in humans. Author Contributions Study design: KM, HK, and KO. Data collection: KM and BO. Structural modeling: TO. Data analysis and interpretation: KM, BO, MI, and AM. Drafting manuscript: KM, HK, and KO. NI and KO take responsibility for the integrity of the data analysis. Catching Up With the HPV Vaccine: Challenges and Opportunities in Primary Care PURPOSE Data confirm that high rates of human papillomavirus (HPV) vaccination have not been achieved despite strong clinician endorsement of the vaccine. We conducted a study of primary care clinicians to assess the broad range of health care delivery, health policy, and attitudinal factors influencing vaccination uptake and opportunities for informed decision making. METHODS We implemented a mixed methods study in RIOS Net, a primary care practice–based research network in New Mexico. We first conducted qualitative, in-depth interviews with primary care clinicians, health policy makers, and immunization experts, and followed up with a confirmatory survey distributed to RIOS Net clinician members. RESULTS Health service delivery challenges emerged as the greatest barrier to HPV vaccination, specifically the lack of capacity to track and distribute reminders to eligible patients. Clinicians also reported variations in counseling approaches attributable to both age and emphasis on the cancer prevention benefits of the vaccine. There was no evidence of sociocultural influences on vaccine decision making, nor did concerns about perceived overprotection emerge. CONCLUSIONS Our findings, based on a long-term program of research, suggest that both patients’ attributes and health system delivery are most influential in HPV vaccination coverage challenges. Interventions targeting innovative communication techniques, as well as health system changes that build on efforts toward coordinated care and utilization of other venues to promote vaccination, will be necessary to address these challenges. INTRODUCTION T he Centers for Disease Control and Prevention (CDC) reports that the human papillomavirus (HPV) vaccine rates trail those of other vaccines for teenagers: tetanus, diphtheria, and pertussis (Tdap), and menningitis. 1 Nationally, 57.3% reported getting the first dose of HPV in 2013, a modest increase from 2012 rates (53.8%). Full coverage (not considering the time frame in which the doses were given) is also less than ideal: among girls aged 13 to 17 years, 37.6% received all 3 doses. 2 National HPV vaccination data for younger eligible recipients (eg, aged 9 to 12 years) are not provided by the CDC. Given new challenges in the health service delivery environment, 3 innovative approaches for increasing HPV vaccination coverage are needed. In the period before the vaccine's release, research was aimed at understanding factors that might predict vaccination uptake. Although identification of such factors varied, there was strong consensus that primary care clinicians would play an important role in HPV vaccine dissemination. [4][5][6] During the prelicensure period, we conducted a qualitative study in primary care settings to better understand and anticipate the challenges and opportunities for HPV vaccination. In this journal, we reported the 4 major domains that emerged from our research, which included (1) the complexity of HPV counseling in the clinical encounter, (2) the recom-mended age, (3) system/compliance issues, and (4) external factors. 7 Given the lower-than-anticipated uptake of the HPV vaccine and associated challenges with promoting shared decision making, we followed up on our original study that examined the conditions into which the HPV vaccine would be introduced. The purpose of this research was to assess the broad range of health policy and health service delivery factors that influenced HPV vaccine in the postlicensure period and to identify opportunities for strategic interventions to enhance both uptake and shared decision making. In this article, we report findings from the current mixed methods study that provides a unique basis from which to understand a range of attitudinal, counseling, and practice-system perspectives about HPV vaccination over an 8-year period. Study Design and Setting We conducted a sequential mixed methods study, including exploratory |
key informant interviews and a subsequent survey designed to confirm and expand upon these qualitative findings. This study (similar to our earlier research) was conducted in RIOS Net, a practice-based network in New Mexico. The University of New Mexico Human Research Protections Office approved the research protocol. Sample and Recruitment In addition to conducting semistructured interviews with RIOS Net clinicians based in academic, community, and school-based health settings, we also purposefully sampled and interviewed key informant stakeholders from state government, professional coalitions, and advocacy organizations focused on immunizations. We utilized a snowball sampling approach for recruitment, first contacting key informants known to the research team for their clinical and/or health policy interest in the HPV vaccine, who then identified other individuals with similar interests and/or involvement in HPV vaccination efforts. 8 Data Collection We conducted interviews from May 2009 to March 2010, mostly in person, with the exception of 2 completed over the telephone. Our semistructured interview guide covered a range of topics, including clinicians' views and experiences implementing the HPV vaccine, perceived attitudes and decision-making influences of parents and caregivers, and health policy implications (ie, mandates for school entry). We ended data collection when we reached consensus that no new information was being provided. We digitally recorded and transcribed all interviews verbatim and offered participants a $50 gift card in appreciation of their time. Data Analysis The research team utilized an iterative analytic process by organizing the transcripts into categories reflecting the roles of stakeholders who participated in the interviews (see above). Each research team member reviewed 2 to 3 transcripts to identify key themes, after which time the group met to compare emergent themes and develop initial coding structures. Subsequently, 4 members of the team (A.L.S., C.M.G., A.S., A.B.) coded the remaining transcripts from each participant category independently and finalized the coding structure by reaching agreement about the application of codes. Transcripts were imported into NVivo 8 (QRS International). We first analyzed the data without reference to the findings from our earlier research. Once this initial period of analysis was completed, we then integrated the 4 domains from our previous work as a starting point from which to compare and contrast emergent themes and identify new areas as relevant. Sample and Survey Development After we analyzed the qualitative data, we developed a survey questionnaire to address the general themes that emerged from the qualitative findings. Included were questions about clinicians' attitudes about the vaccine, their experience with implementing the Advisory Committee on Immunization Practices (ACIP) recommendations, and their views about the influence of health policy (eg, mandatory vaccination) on their practices. We used visual analog scales to indicate strength of agreement (or disagreement), with 0 = strongly disagree and 100 = strongly agree. It was administered in both electronic and paper formats. The electronic version utilized the Web-based application Research Electronic Data Capture (REDCap). The entire questionnaire can be found in Supplemental Appendix 1, at http://www.annfammed.org/ content/13/4/354/suppl/DC1. We took 2 steps to ensure that the questionnaire captured the perspectives of clinicians who administer the HPV vaccine for adolescent patients aged 9 to 18 years. First, we did not send the questionnaire to internal medicine clinician members of the network who were unlikely to order the HPV vaccine for adolescent patients. Second, we included a screening question that invited participants to complete the questionnaire only after confirmation that they currently ordered the vaccine. Data Collection We conducted the survey from May to August 2011. We administered the questionnaire through 4 weekly e-mail solicitations with an embedded survey link. We then sent out 2 more rounds of the questionnaire in paper format to clinicians' at their practice locations. Two randomly selected respondents were drawn from each solicitation round and received $50 gift certificates. Data Analysis The Web-based questionnaires were automatically captured in REDCap; we hand-entered the paper-based questionnaires, double entering 25% of them to ensure data entry reliability. We exported data from REDCap into SAS 9.2 software (SAS Institute) for analysis. We computed descriptive statistics for all items including means and standard deviations of continuous variables and frequencies of categorical items. As a final analytic step, we integrated the survey findings with our qualitative thematic template to assess the degree to which these data supported and/or modified the 4 major domains. RESULTS We conducted key informant interviews with 25 individuals and a total of 98 of 158 eligible RIOS Net clinicians responded to the survey (62%). Demographic characteristics are displayed in Table 1. We provide a comparison of the findings described below with our prelicensure findings in Table 2 and a full summary of survey results in Tables 3 and 4. HPV Counseling in the Clinical Encounter Clinicians reflected on their challenges about discussing HPV infections with their adolescent patients, sim- Clinicians reported emphasizing the cancer prevention benefits of the vaccine to avoid what they perceive as a potentially uncomfortable discussion about adolescent sexual behavior, particularly with parents in the examination room. As one clinician stated, "I talk about it as preventing cervical cancer. If they want to discuss the sexual aspects, they're going to have to bring it up, because I won't." Survey respondents strongly endorsed the statement, "In presenting the HPV vaccine, I emphasize its cancer prevention benefits" (mean = 87.2). An adolescent immunization advocate strongly promoted not explicitly linking the vaccine with sexual activity, saying, "We don't want this to be the sex vaccine because it's not. It's a cervical cancer-preventing vaccine or a HPV vaccine or a warts vaccine…." Recommended Age Clinicians differentiated between having discussions about the vaccine's benefits and offering the vaccine to adolescent patients. Survey respondents reported that a discussion about the vaccine is useful to initiate sexual risk counseling for older adolescents, as exemplified by response to the question, "In presenting the HPV vaccine, I see it as an opportunity to open up a discussion about adolescent sexual activity (mean = 73.9)." In the qualitative interviews, clinicians expressed a preference to have these discussions with older adolescents (eg, those aged 15 to 18 years) because the conversation was more in line with their developmental status. As one clinician stated, "To me, 11 and 12 is a little bit young…because [for] some, their body is barely starting to change. And lots of times they're even afraid to ask. And it's like, 'What do you mean; I've got to protect myself from cervical cancer? I've never even had a period.'" The average age at which clinicians reported offering the HPV vaccine to adolescent patients was 10.7, however, even younger than the earliest ACIPrecommended age (11 to 12 years). When asked at what age they felt most comfortable first offering the vaccine, the vast majority stated that they preferred younger ages: 21.4% chose 9 to 10 years, whereas 65.3% chose the recommended age of 11 to 12 years; only 13.3% selected categories that included adolescents aged 13 years and older. System and Compliance Issues Clinicians strongly endorsed the role of system-level barriers to HPV vaccination, noting that opportunities to offer the vaccine to their adolescent patients are limited, especially in the context of visits addressing other acute health issues or behavioral or mental health needs. Exactly one-half of clinician respondents Note: Responses are on a scale from 0 to 100, in which 0 = strongly disagree and 100 = strongly agree. indicated that they do not schedule appointments to offer the second and third doses. Clinicians responding to the questionnaire estimated that only 9% of their patients received all 3 doses in the recommended 6-month time period, and that 28% of their adolescent patients aged 9 to 18 years had received the first dose. Few clinicians (22%) reported having reminder systems in their clinics. As one family physician noted, "The burden's on the patients. To be honest, I don't have the personnel to be calling the patients." External Factors: Sociocultural Beliefs, Media, and Policy Despite well-publicized concerns that adolescents might feel overprotected from HPV (and even other sexually transmitted diseases) upon vaccination in the prelicensure period, clinicians reported that this fear did not play a major role in actual HPV vaccination decisions. Consistent with the literature, most clini-cians reported that parents had not raised such concerns during discussions about the vaccine, 9 and recent findings from a national sample of girls aged 12 to 18 years old confirm that incidence of sexually transmitted infections does not increase with HPV vaccination. 10 When asked on the questionnaire to anticipate whether "the HPV vaccine will lead females aged 9 to 18 years to feel overprotected and more likely to engage in risky sexual behavior," the mean response was 23.2. Further, despite concerns that diverse cultural considerations (eg, religious or social conservatism) would influence vaccination uptake, clinicians did not report variations in receptivity to the vaccine across the different populations that they serve (mostly Hispanic and American Indian). DISCUSSION In this follow-up to our original study, we triangulated data from a mixed methods approach that led us to examine the range of implementation challenges which affect current HPV vaccination practices and to compare and contrast these circumstances with those identified in our previous research. Based on this longitudinal program of research situated in a Southwestern practice-based research network, we offer recommendations about how this information can be used to develop contextually appropriate interventions. Age-Based Counseling Strategies Similar to the findings of Vadaparampil et al, 11 we found that primary care clinicians continue to struggle with many, though not all, of the challenges anticipated from our earlier study. Perhaps the most prescient concern was that clinicians recognized the likelihood of framing the HPV vaccine differently based on patient age. Indeed, this prediction emerged as a central theme as clinicians in this current study expressed a preference for introducing the vaccine through a cancer prevention lens to younger girls (aged 9 to 12 years), while opportunistically integrating sexual risk behavior counseling with older adolescents (aged 15 to 18 years). This finding is consistent with other recently published qualitative research that found lower self-reported vaccination rates when clinicians considered the timing of sexual activity as compared with clinicians who emphasized a cancer prevention message. 12,13 Further, a recent (2013) CDC report found that 3-dose completion rates are 25.8% at 13 years and 48.2% by 17 years, 2 and results from 2 national surveys of physicians found that clinicians were more likely to offer HPV vaccines to older female adolescents 10 and reported more vaccine refusals among parents of younger adolescents. 14 This pattern is especially trou- bling, given that the antibody response is more robust in younger adolescents (aged 9 to 15 years). 13 The data we report show there are subjective interpretations of what constitutes the most appropriate strategy, and interventions that provide guidance to health care professionals might include criteria for choosing one approach rather than another. It may be necessary to tailor counseling strategies in such a way that emphasize cancer prevention or sexual risk reduction based on consideration of patient and contextual attributes (eg, age, cultural group, and level of parental involvement). 15 Furthermore, parents and their vaccineeligible adolescent children may be attracted to interactive media (computer kiosks) or applications for electronic devices that would help them to independently review information and make decisions before they reach the examination room. 16 Although clinicians did not detect differences in vaccine receptivity based on ethnicity and socioeconomic status, our published research 17 and that of other researchers 18 show that even within ethnic groups, parents and adolescents have different decision-making approaches. System Issues In our current study, health system delivery issues, including the lack of flexible tracking and reminder capabilities, were identified as the most important barriers to achieving broader HPV vaccination coverage. These barriers lead to missed opportunities, because adolescents come to the clinic most often for acute, episodic care. 11 Most clinicians in this study reported having an electronic medical record system, though most respondents did not rely on this system to encourage vaccination visits. Limitations in tracking can lead to further problems, including potential overvaccination and avoiding mixing the 2 available vaccines, as it is difficult to verify vaccination history. Implications for Intervention A major focus of our work is to address the tension between increasing vaccination uptake and ensuring informed decision making. Two strategies emerged from the research: improved communication and health systems |
change. Improved Communication Experience has shown that messages disseminated through various social media channels for teenagers, starting with the ACIP-recommended age and older, can be helpful. Social marketing and Internet-based interventions, however, can also be designed to support brief discussions between primary care clinicians with parents and teenagers in the clinical encounter, as well as dissemination of preventive health recommendations. 19 Health System Changes Health system delivery barriers, more than any other single factor, were identified by participants as the greatest limitation to achieving shared goals of increased vaccination uptake and informed decision making. Given the strong endorsement of the HPV vaccine among both clinicians and the general public, efforts to enhance vaccination may be best served by interventions aimed at improving delivery mechanisms rather than those strategies that focus on individual behavior change. Optimizing newly required electronic medical systems to enhance tracking and generate timely reminders has been shown to increase vaccination rates. 20 Other promising health system approaches include the creation of dedicated nurse appointments for the second and third dose of the HPV vaccine. There may be trade-offs to consider in such approaches, however, as these efforts to increase vaccine coverage may decrease opportunities for patient-clinician communication. Given that very few adolescents complete the 3-dose series in the recommended 6-month period, our previous research suggests that many adolescents and parents may need a basic refresher about the HPV vaccine, and it will be important for staff to be comfortable with these counseling needs. Another approach to increasing compliance with the vaccine would be to incorporate the HPV vaccine as part of an adolescent immunization platform (Tdap, meningococcal vaccines). [21][22][23][24] The infrequency of adolescent visits to primary care clinics during the ages of 13 to 17 years makes it all the more important to bundle immunizations as a package to ensure initiation and multidose completion. Finally, further research is needed to reduce the burden on the primary health care system by identifying other venues for vaccination, such as school-based health centers. Students in these settings are easier to access, and the school year is long enough to complete vaccination in a multidose series. 24,25 Limitations Given that this research reflects the views and experiences of clinicians and immunization experts who work with underserved ethnic minority and rural populations, it is possible that these findings are relevant to these circumstances and not to conditions of vaccine delivery across the country. Our findings, however, are widely consistent with those reported in the broader literature while offering perspectives that may be uniquely relevant to primary care clinicians and researchers working in similar contexts. In our earlier study, before the approval and release of the HPV vaccine, we sought to identify the range of issues relevant to implementing the vaccine in primary care settings. Now that the vaccine has been available for more than 8 years, persistent gaps in vaccination coverage show the need to overcome both implementation and informed decision-making barriers. Contextually sensitive interventions will be needed to address these challenges. Findings from this current study, based in a long-term program of research examining these evolving dynamics, provide guidance toward that goal. Optimization of Enzymatic-assisted Extraction of Polysaccharides from Roxburgh Rose Pomace and its Antioxidant Activity In this experiment, at first the roxburgh rose juice was extracted and then roxburgh rose residue was taken as raw material. We have used the enzyme assisted method to study the extraction process of polysaccharides from roxburgh rose pomace. The effects of mesh number, the concentration of the enzyme, temperature and time, pH and solid-liquid ratio on the polysaccharides yield were explored by single factor experiments. And then orthogonal experiment was designed to study the optimal techniques on extracting of polysaccharides from pomace. The in vitro antioxidant activity of the obtained polysaccharides was studied. The results showed that the optimal condition of extracting polysaccharides was as follows: enzyme concentration 2.5%, enzymatic hydrolysis at 60° for 40 min, pH 4.0, mesh number were 100 and solidliquid ratio was 1:25 based on the ratio of cellulase and pectinase was 2: 1. The average polysaccharides yield of enzymatic extraction method reached (4.79±0.07) % under the optimal condition. The antioxidant activity assays in vitro revealed that polysaccharides from roxburgh rose pomace can be used as natural antioxidants in functional foods and pharmaceutical industries. Introdution Plant polysaccharides are polymers of monosaccharides units joined together, which are widely found in plant cell wall and animal cell membranes in nature [1,2]. Studies on polysaccharide activity have found that it involved in many complex bioactivity and functions such as cell recognition, growth, differentiation, metabolism and so on [2][3][4]. It also has anti-tumor, antioxidation, regulating serum lipid and other bioactivities. As one of the main nutritional or pharmacological components of roxburgh rose, polysaccharide of roxburgh rose has many bioactivities, such as antioxidation, immunomodulation, delaying the aging process and so on [5][6][7]. It has good value of development and utilization in the human body. At present, the cultivation of roxburgh rose is mainly concentrated in the southwest of China, the output of roxburgh rose in Guizhou ranks first in China [8]. Therefore, it has great significance to develop local characteristic resources. Recent studies on roxburgh rose have shown that it contains vitamin C, organic acids, polysaccharides and other active components [6,7], which is widely used in the fields of health products and medicine [7][8][9]. Yet, with the development of juice processing in China, roxburgh rose residue which after roxburgh rose juicing nearly 50% each year [8,9], it is also used as cattle feed or directly disposed as waste. The polysaccharides in the residue were discarded, resulting in the waste of resources. Therefore, it has certain social values as well as some economic benefits to discuss the utilization method of residue resources from roxburgh rose. Therefore, there has been increasing interest in developing and utilizing natural, effective and safe antioxidants, such as plant polysaccharides, to protect the human body from chronic. Enzymatic assisted extraction of polysaccharides has the advantages of mild conditions, easy removal of impurities and high yield [10,11]. In this study, the processing waste residue of roxburgh rose in Longli County of Guizhou province was used as raw material, and then orthogonal experiment was designed to study the optimal techniques on extracting of polysaccharides from pomace by cellulase and pectinase. The in vitro antioxidant activity of the obtained polysaccharides was studied. It not only reduced waste and pollution to the environment, but also provides the theoretical basis for the comprehensive utilization of the roxburgh rose residue resources so as to achieve the purpose of high value conversion and utilization. Materials and Methods The roxburgh rose residue was collected in the Longli County (Guizhou province, China) in September 2016, which after roxburgh rose juice was taken to microwave vacuum, drying, and stored at -20℃ until it was used for the purpose of experiment. Extraction of crude polysaccharide from roxburgh rose residue The dried roxburgh rose residue was crushed and sifted, 2 g residue was extracted with cellulase and pectinase according to the parameters set by single factor experiment [10]. Then later on it was precipitated by the addition of ethanol and kept for the whole night, the supernatant was discarded after centrifugation at 6000 rpm for 20 min, it was washed with absolute ethanol, then centrifuged, and the ethanol washing step was repeated twice, the crude polysaccharide A after freezedried. The crude polysaccharide was re-dissolved in distilled water for degreasing and removing protein by add 3 times volume of petroleum ether (3:1 v/v) and sevag (Chloroform: n-butanol =4: 1) method [10,12,13], then precipitated by the addition of ethanol overnight, the supernatant was discarded after centrifugation at 6000 rpm for 20 min, it was washed with absolute ethanol, then centrifuged, and this ethanol washing step was repeated twice, the crude polysaccharide B after freeze-dried. Determination of polysaccharide content The absorbance of each sample was determined by sulfuric acid-phenol method [14,15]. Orthogonal Experiments The yield of crude polysaccharides from roxburgh rose pomace was taken as the index on the basis of single factor experiment, and the optimal techniques on extracting of polysaccharides from pomace by orthogonal experiments with six factors and three levels L18(3 6 ), the factor level was shown in Table 1. Analysis of antioxidative properties of polysaccharide from roxburgh rose pomace The crude polysaccharide A and B was re-dissolved in distilled water for antioxidant activity. Reduction power Making accurate shift polysaccharide samples from roxburgh rose pomace, followed by adding 2.5mL 0.2mol/L phosphate buffer solution and 2.5mL 1% six potassium dicyanate solution, and rapid cooling after reaction 20min at 50℃, then adding 2.5mL 10% three trichloroacetic acid and centrifugation at 3000 rpm for 10 min, the supernatant was added in 2.5mL distilled water and 0.5mL 0.1% trichloride solution, and then the absorption was measured at the wavelength of 700 nm after statically placing 10min. Scavenging hydroxyl radical(·OH)activity 2 mL 9 mmol/L FeSO 4 , 2 mL 9 mmol/L salicylic acid and 2 mL sample solution were successively added to the tube with stopper. The absorptivity was measured at 510nm after water bath reaction for 30 minutes at 37 ℃ by adding 2mL 8.8 mmol/L H 2 O 2 [16]. Scavenging activity of DPPH radical 0.1984g 2,2-diphenyl-1-picrylhydrazyl (DPPH) was accurately weighed and dissolved with anhydrous ethanol to 50 mL as reserve liquid, then 2mL DPPH reserve solution was dissolved in absolute ethanol. 2 mL DPPH solution and 2 mL ethanol as the test tube, the absorptivity was determined at 517nm after light avoidance reaction [17]. Statistical analysis Values are expressed as mean +/-standard deviation. Statistical analyses were performed by one-way analysis of variance with the Duncan's multiple range test (SPSS 17.0, Illinois, USA). P values of less than 0.05 were considered statistically significant. Determination of optimal proportion of enzymes Taking 2.00 g dry and mesh 80 of roxburgh rose pomace, adjust solution pH 5, then adding distilled water according to 1: 20 g/mL, and 2% the ratio of cellulase and pectinase enzyme solution was 1:1, 1:2, 2:1, respectively. As can be seen from figure 1, when the ratio of cellulase to pectinase is 2: 1, the polysaccharide yield is significantly higher than that of the other two groups(p<0.05). Therefore, in the following experiment, the compound enzyme was used to assist the extraction of polysaccharides from pomace at the ratio of cellulase and pectinase is 2: 1. Values with different superscripts within a column are significantly different (P<0.05). All measures were performed in three independent samples. Effect of enzymatic hydrolysis time on the yield of polysaccharides from roxburgh rose pomace In the reaction process of enzymatic hydrolysis, the yield of polysaccharides changed with the continuous extension of extraction time, reaching up to equilibrium. As it is shown in figure 2(a), along with the increase of reaction time, the yield of polysaccharides increased gradually. The content of polysaccharides reached the maximum when the system was heated for 60 min. The cellulase and pectinase tend to lose their bioactivity for a long time heating, moreover the dissolution of polysaccharides tends to balance after reaching the highest, which leads to the decrease of polysaccharide content. Yet, the reaction system may be basically balanced after 80 minutes, and there is no significant difference between the latter three groups(p>0.05). Effect of enzymolysis temperature on the yield of polysaccharides from roxburgh rose pomace With the increase of enzymatic hydrolysis temperature, the content of polysaccharides increased gradually. As the figure below shows clearly, it is not conducive to the dissolution of polysaccharides when the temperature is lower than 45 ℃ ; when the enzymatic hydrolysis temperature reached 55 ℃ , the content of polysaccharides ranked highest, it maybe the high temperature which might promote the movement rate of molecules. Shortly the content of polysaccharide decreased, yet the difference was not distinguished(p>0.05). Since the temperature will affect the activity of enzyme in a certain temperature range, the reaction rate of enzyme will increase with the increase of temperature, thus accelerating the decomposition of substrate. When the heating temperature is too high, the enzyme activity will be weakened or even lost, the reaction rate will be slowed down and the polysaccharide will decompose due to its thermal instability, which will lead to the decrease of polysaccharide yield. Effect of solid-liquid ratio on the yield of polysaccharides from roxburgh rose pomace From figure 2(c), the |
content of polysaccharides extracted from pomace increased with the increase of the solid-liquid ratio; the content of polysaccharide increased significantly when the solid-liquid ratio is 1:30(p<0.05). The yield decreased at radio of 1: 35, but the difference was not significant(p>0.05). It is suggested that when the solid-liquid ratio is too low, it is not conducive for the full dissolution of roxburgh rose pomace. Only when the solid-liquid ratio is appropriate, the swelling ability of the raw material cells get enhanced, it promoted the permeation of compound enzymes and solvents to the roxburgh rose pomace effectively, resulted the yield of polysaccharide increased [3]. The solid-liquid ratio too high may lead to the dissolution of other impurities, thus affecting the yield of polysaccharides, but also lead to the follow-up enrichment process cumbersome. Effect of mesh on yield of polysaccharides from roxburgh rose pomace the content of polysaccharides got ascent first and then fell with the increase of mesh. When the degree of comminution reached 100 mesh, the polysaccharide content was significantly higher than others (p<0.05), then the yield decreased sharply (p<0.05). Therefore, 100 mesh can be obtained as the optimum for extraction. Effect of enzymatic concentration on yield of polysaccharides from roxburgh rose pomace Accordingly, cellulase and pectinase can effectively destroy or decomposition cell wall of roxburgh rose pomace, which is helpful to improve the extraction rate of polysaccharide. Just as what is shown below in figure 2(e), the content of polysaccharides raised with the increase of the concentration of enzyme. When the concentration was 1.5%-2.0%, the yield was significantly higher than that of the other groups (p<0.05); then the yield significantly decreased (p<0.05), due to a certain substrate concentration in this experiment has reached saturation. Effect of pH on yield of polysaccharides from roxburgh rose pomace When pH was 4-5, the polysaccharide content was significantly higher than that of other groups (p<0.05); the extraction rate of polysaccharides tends to decrease with the pH value of the solution continues to increase (p<0.05). This may be due to the pH restricted the enzyme activity, and cellulase and pectinase activity is stronger under the condition of partial acidity, which is helpful to the extraction of polysaccharides. When the solution was neutral or alkaline, the enzyme was inactivated, and the polysaccharide content was significantly lower than that in other groups (p＜0.05). Orthogonal experiment The effects of mesh number, the quantity of the enzyme, temperature and time, pH and solid-liquid ratio on the polysaccharides yield were explored by single factor experiments. And then orthogonal experiment was designed to study the optimal techniques on extracting of polysaccharides from roxburgh rose pomace. The orthogonal optimization test is carried out according to the three optimal levels determined by the single factor experiments, and the design is shown above in table 1. The experimental results and intuitive analysis are shown below in table 2, affecting the content of polysaccharides which was extracted from pomace by enzyme assisted method, that is, enzymatic concentration > enzymatic hydrolysis time > enzymatic hydrolysis temperature > solid-liquid ratio > pH > mesh; and the optimization of enzymatic-assisted extraction of polysaccharides from roxburgh rose pomace were A1B3C1D3E1F3: the enzymatic concentration was 2.5%, the enzymatic hydrolysis at 60℃ for 40 min, the solidliquid ratio is 1:25, pH was 4 and mesh 100. Verification experiment The optimum enzymatic-assisted extraction conditions obtained from orthogonal experiment, the average yield of polysaccharides extracted from roxburgh rose pomace reached (4.79±0.07)% by cellulase and pectinase. Antioxidant analysis of polysaccharide from roxburgh rose pomace The antioxidant activity of polysaccharides from the roxburgh rose residue was evaluated by the color of the solution. From figure 3(a), the scavenging rate of hydroxyl radicals was not more than 30% in each group, and the crude polysaccharide B was significantly higher than the other groups. It shows that the polysaccharide from the residue has the ability of scavenging hydroxyl radical, but it is inferior to the polysaccharide from soybean curd residue [18], guarana powder [19] and fungal [20]. The ability of scavenging DPPH free radicals is commonly used to measure the antioxidant ability of fruits and vegetables [21]. From figure 3(b), as the concentration of polysaccharide increased gradually, the free radical scavenging rate of DPPH was significantly enhanced and not less than 60%, which was better than polysaccharide from ginkgo biloba [22] and dendrobium candidum [23], lower than guarana powder [19]. The inhibition rate of DPPH free radical in the 0.1 mg / mL crude polysaccharides from the roxburgh rose residue was similar with 0.5 mg/mL(p>0.05). But the scavenging ability of crude polysaccharides B at high concentration was significantly higher than that of crude polysaccharide A(p<0.05). From figure 3(c), it can be drawn that different concentrations of crude polysaccharides from the roxburgh rose residue have a certain reduction power, but with the concentration of crude polysaccharides A increased, the reducing power has little change. The reduction power increased with the crude polysaccharide B concentration increased(p<0.05). But both of them were significantly lower than roxburgh rose residue group, and superior to ginkgo biloba polysaccharides [21]. It is suggested that the electron provided by crude polysaccharides can not only reduce Fe 3+ to Fe 2+ , but also form inert compounds with free radicals, thus avoiding self-oxidation chain reaction. However, the roxburgh rose residue group has strong reducing ability as a result of polyphenols, flavonoids and other bioactive substances. Conclusion In this experiment, roxburgh rose juice was extracted which after roxburgh rose residue was taken as raw material. Later on we had studied the extraction process of polysaccharides from roxburgh rose pomace by enzyme assisted method. The single factor and L 18 (3 6 ) orthogonal experiment were designed to study the optimal techniques on extracting of polysaccharides by cellulase and pectinase (2: 1). The optimization of enzymatic-assisted extraction of polysaccharides was as follows: mesh 100 roxburgh rose residue, add distilled water at the solid-liquid ratio 1:25, 2.5% enzymatic and adjust pH was 4, hydrolysis at 60℃ for 40 min, mesh 100, the average yield of polysaccharides reached(4.79±0.071)%, indicating that the enzyme assisted extraction increased the yield of polysaccharides significantly. The antioxidant experiments showed that the radical scavenging rate on DPPH and ·OH of 1 mg/mL polysaccharides after degreasing and deproteinization were 86.1% and 28.4%, respectively, and reducing power was 0.625. Compared with the roxburgh rose residue, the crude polysaccharide has stronger scavenging ability of DPPH free radical, yet its hydroxyl radical scavenging ability is not as good as other polysaccharides such as soybean curd residue, guarana powder. It not only provides a good idea for the rational use of roxburgh rose residue, but also provides experimental basis for the industrial production from residue polysaccharides, Therefore, it provides a new way for making full use of characteristic resources in Guizhou province. However, further researches are needed to study the relationship between its composition and antioxidant. Potential Diagnostic Hemorheological Indexes for Chronic Kidney Disease in Patients With Type 2 Diabetes Many studies have demonstrated that an alteration in hemorheological properties is closely correlated with diabetic microcirculatory diseases. However, most of these studies have been limited to animal studies or used a small number of clinical samples, due to a lack of effective point-of-care (POC) devices to measure such properties within clinical environments. Owing to recent developments in microfluidic technology, several hemorheological POC devices have been designed that allow for the possibility of conducting extensive clinical studies using hemorheological measurements. Here, we reviewed recent clinical studies of diabetic kidney disease (DKD) associated with hemorheological parameters. We found that RBC deformability alone did not show a significant difference according to the degree of DKD, whereas critical shear stress (CSS) was found to be closely related to the ratio of albumin to creatinine and glomerular filtration rate. We also reviewed studies that alteration of hemorheological properties are associated with the development of DKD, which showed that CSS could be considered as a potential index to diagnose other diabetic complications as well as DKD. INTRODUCTION The exponential increase in the incidence of diabetes is a major global healthcare issue (Bommer et al., 2017). In 2015, it is estimated that 8.8% (415 million) of the world's population between the ages of 20 and 79 are suffering from diabetes, and the incidence of such diabetes is expected to increase by 1.5 times in 2040 (Baik, 2019). Diabetic vascular complications are a major cause of death in type 2 diabetic patients. Moreover, such diabetic complications lead to kidney failure, blindness, and non-traumatic limb amputation in adults (Afkarian et al., 2016). Diabetic complications in small blood vessels consist of diabetic eye disease known as DR, DKD, and DPN. These diabetic microangiopathies are closely correlated with macro-angiopathy such as coronary artery disease and stroke (Forbes and Fotheringham, 2017). Therefore, early detection of diabetic microangiopathy including DKD is crucial to prevent a progression to various complications and reduced quality of life of patient (Park, 2014). Stages of kidney disease can be determined by the measured or estimated GFR. Stage 1 indicates normal value of GFR (>90 ml/min/1.73 m 2 ) and stage 2 is only minimally reduced in GFR (60-89 ml/min/1.73 m 2 ). The characteristics of stage 3 include the presence of microalbuminuria, a moderately reduced GFR range (30-59 ml/min/1.73 m 2 ), and elevated blood pressure. Stage 4, known as clinical nephropathy, involves focal glomerular sclerosis and macroproteinuria, with a deterioration of the GFR (15-29 ml/min/1.73 m 2 ). At stage 5, the GFR continues to decline (<15 ml/min/1.73 m 2 ) and end-stage renal disease may develop (Levey et al., 2002). Without treatment, DN has been reported to develop undetected, until noticeable symptoms occur. If DN is diagnosed early and managed appropriately, the progression of DN can be prevented or delayed (Shlipak, 2010). Several methods for screening DN have been introduced, with the urinary albuminto-creatinine ratio (uACR) being the most commonly used clinical method to detect early DN. A uACR range between 30 mg/g and 300 mg/g of creatinine confirms a diagnosis of microalbuminuria, and has been reported to be excellent predictor for developing DN (Mogensen, 1984). Even though the uACR is relatively simple to use, inexpensive, and convenient for making determinations, it has several limitations as a screening tool for DN, including daily fluctuations in creatinine levels and being strongly influenced by other comorbidities such as infection, fever, menstruation, and exercise within 24 h, independent of kidney damage. Therefore, efforts continue to determine a more stable, complementary screening tool for early DN detection. There have been numerous studies on alteration of hemorheological properties associated with DM since diabetic vascular diseases have shown close correlations with hemorheological indexes (Cho et al., 2008;Lee et al., 2018). Typical hemorheological indexes include deformability and aggregation of RBCs, ESR, CSS, yield stress and whole blood viscosity, These indexes had been only measurable in the laboratory for a long time. However, due to limitations in point-of-care technology (POCT) and the relevant instruments, the number of clinical patients involved in previous studies has not been large, and differing results have been reported. Owing to recent developments in POCT concerning hemorheological measurements, several more targeted clinical studies have been conducted within the last 5 years. However, recent advances in leading-edge technologies including microfluidics have made the measurement of these hemorheological indexes possible in clinical settings. Thus, clinical studies have been undertaken to confirm an association between hemorheological alteration and diabetesrelated kidney diseases Chung et al., 2018). These clinical study focused on RBC deformability and CSS among hemorheological properties, since these two indexes had been the most potential diagnostic indexes of DM associated microangiopathy (Razavian et al., 1992;Brown et al., 2005). Therefore, this review aimed to assess the recent clinical studies concerning DN that had used hemorheological measurements. This is the first review that evaluates the association of CSS and DKD. IMPACT OF DIABETIC MELLITUS ON HEMORHEOLOGY Diabetes-related micro-vascular diseases have been shown to be closely associated with alteration of hemorheological properties such as increased blood viscosity (Monica et al., 2019), reduced cellular deformability (Brown et al., 2005), and increased cellular aggregation (Foresto et al., 2000). In addition, hemorheological alterations are frequently observed in the early onset of diabetes (MacRury and Lowe, 1990), and it has been suggested that these altered hemorheological properties may precede the growth of diabetic vascular complications (Martinez et al., 1998). Prior to focusing on a specific diabetic disease, it |
would be better to review the impact of DM on RBC rheology. Impaired RBC Deformability in Diabetic Mellitus When insulin control fails, there is a consequent increase of glucose level in the blood plasma, which leads to hyperglycemia. Hyperglycemia is a major metabolic perturbation that results in microvascular damage in patients with diabetes (Friedman, 1981;Porte and Schwartz, 1996). However, the mechanism through which hyperglycemia leads to DN is not well understood. It has been reported that non-enzymatic glycation and oxidation (glycoxidation) forms AGEs, which is closely related to the pathogenesis of DKD (Makita et al., 1991;Fukami et al., 2008;Singh et al., 2014). Also, glycated hemoglobin (HbA1c), which represents glycation of RBCs, showed inverse correlation with RBC deformability in DM (Keymel et al., 2011) and DN (Shin et al., 2007b). Figure 1 depicts schematically the influence of diabetes-related factors on RBC deformability and aggregation. Reduced RBC deformability is frequently observed in T2DM. The reduction of RBC deformability is associated with the denaturation of the cell membrane through the effects of glycation and oxidative stress. An abnormal blood glucose level results in biochemical changes within RBC membrane proteins and membrane lipids. Oxidative stress, associated with highlevel glucose concentrations, results in denaturation of the RBC membrane lipids and proteins (Resmi et al., 2005). Glycosylation of membrane proteins, which is closely related to reduced membrane fluidity (Watala et al., 1985), is the main cause of the reduction of RBC deformability (Schwartz et al., 1991;Symeonidis et al., 2001). Also, the reduction of RBC deformability is associated with increased viscosity of glycosylated internal fluid. Many studies have concluded that denaturation of the RBC membrane and increased internal viscosity of erythrocyte induce abnormal viscoelastic properties of erythrocyte membranes and may contribute to the onset of diabetic vascular diseases (Watala et al., 1992;Linderkamp et al., 1999). Consequently, the adverse alteration of RBC deformability can degrade the primary functions of RBCs, including the transportation of metabolites through the blood vessels (Caimi et al., 1993;Gyawali et al., 2016). Increased RBC Aggregation in Diabetes Mellitus Increased RBC aggregation is commonly encountered in patients with T2DM, which accompanies the change of composition in the erythrocyte membrane. RBC membrane is slightly negative charged due to the sialic acid of glycoproteins on RBC membrane. The negative-charge on the RBC membrane result in electrostatic repulsion between RBCs and consequently reduces erythrocyte aggregation (Rogers et al., 1992). However, many studies reported that there was a significant reduction in the negative charge on the membrane surface for patients with T2DM (Gambaro et al., 1988;Budak et al., 2004), leading to an increased RBC aggregation (Rogers et al., 1992). Higher plasma fibrinogen levels have been found in patients with T2DM than in healthy controls (Mahendra et al., 2015). Fibrinogen production and its concentration in plasma increase in patients with insulin-resistant T2DM. Increased fibrinogen level in plasma is an important risk factor of cardiovascular disease in patients with T2DM. When hyperfibrinogenemia occurs alongside decreased albumin levels, RBC aggregation is further increased (Angelkort, 1999). Moreover, if high levels of fibrinogen correspond with impaired RBC deformability in patients with T2DM, there is likely to be a synergistic effect on the tendency of RBCs to aggregate (Xue et al., 2013). Increased RBC aggregation is considered to be the main cause of diabetic vascular complications since strongly aggregated RBCs cannot flow through capillaries (Grigoleit et al., 1973). More studies have provided supporting data on the relationship between hyperaggregation of RBC and hemodynamic resistance in vivo (Yalcin et al., 2004). Furthermore, patients with T2DM who had increased RBC aggregation showed a greater prevalence of peripheral vascular diseases than those without diabetes. Thus, it has been proposed that increased RBC aggregation is directly associated with this pathological results (Osmundson et al., 1990;Maser et al., 1991). HEMORHEOLOGY Hemorheology is the study of deformation and flow of blood and its cellular components. Since erythrocytes comprise the major components in blood, many hemorheological studies have focused on RBC rheology, such as deformability and aggregation of RBCs. Owing to the deformation and aggregation of RBCs, whole blood exhibits shear-thinning non-Newtonian viscosity. Viscosity of whole blood decreases with shear rate. This shearthinning characteristic is mainly due to RBC aggregation as well as RBC deformability. RBC aggregation involves the autoassembly of RBCs at low shear rates, whereas RBC deformability relates to shape changes under shear stress in liquid flow. When whole blood is under a high shear stress field, RBCs tend to disperse and deform. When shear stress gradually decreases, elongated RBCs return to their original shape and dispersed RBCs tend to aggregate at a certain shear stress, with the minimum stress required to prevent RBC aggregation known as the CSS of whole blood. Determinants of RBC Deformability Deformability is a key feature of RBC that allows RBCs to pass through smaller vessels than themselves. At the cellular level, the major determinants of RBC deformability are (i) protein phosphorylation in membrane and cytoskeletal organization, (ii) control of ion and water of intracellular fluid (iii) surface area-to-volume ratio and (iv) the metabolism and integrity of hemoglobin (Kim et al., 2015;Huisjes et al., 2018). These factors are significantly altered in sickle cell disease and sepsis (Baskurt et al., 1998;Alapan et al., 2016). In addition, these factors are also closely related to adenosine triphosphate levels and the redox state (Karger et al., 2012;Kuhn et al., 2017). Combinations of these factors within hereditary and environmental conditions determine RBC deformability. This deformability regulates the blood's life span as well as the efficiency of oxygen transport. A slight reduction in RBC deformability may result in decrease of blood flow into capillaries and subsequently result in various microvascular diseases (Shin et al., 2007b). Measurement Techniques A filtration method has been in use since the early days of measuring RBC deformability. This method involves assessing the capability of RBCs to pass through either small pores or microfluidic channels with applying positive or negative pressure (Reid et al., 1976;Bransky et al., 2007). RBC deformability (or rigidity) can be evaluated by measuring the passing time of a certain volume of RBCs through a filter. Even though the filtration method is simple, it has various problems including the blockage of filters by rigid leukocytes and the aggregates of activated platelets. Because of these limitations, filtration methods have produced variations in experimental results. There has been an attempt to mitigate these limitations using a mesh processed with micro-fabrication - (Odashiro et al., 2015). Laser diffractometry is another key method for measurement of cell deformability, because of its precision and reproducibility. There exist three commercially available laser diffractometers, adopting different shearing flow geometries. Their geometries are cop-and-bob, plate-plate and Poiseullie flow channel, respectively (Tomaiuolo, 2014;Kim et al., 2015). Among them, a microfluidic ektacytometry, which is designed for POC test, has been applied to measure RBC deformability for the diagnosis of patients with potential microcirculatory diseases (Shin et al., 2007b). The technique is performed on a shear stress-scanning laser-diffractometer (RheoScan), which measures an EI over an appropriate range of shear stresses [0.1-20 pascal (Pa)], as shown in Figure 2 (Shin et al., 2007a;Baskurt et al., 2009a). Microfluidic ektacytometry provides a maximum elongation index (EI max ), an elongation index at 3 Pa (EI 3Pa ), and half shear stress (τ 1/2 ). The half shear stress is the value corresponding to shear stress where the EI is 50% of EI max (Baskurt et al., 2009b). These parameters (EI 3Pa , EI max , τ 1/2 ) are still being tested for their usefulness in diagnosis of various deformability-associated disorders. Table 1 summarized advantages and drawbacks of each measurement technique with operating principles. RBC Aggregation Similar to RBCs from most mammalian animals, human RBCs tend to aggregate forming of linear stacking shapes, which is often called rouleaux (Chien and Jan, 1973;Baskurt et al., 2012). One-dimensional forms of rouleaux can grow and form three-dimensional (3-D) networks. While these 3-D network aggregates grow exponentially from flow to stasis, a shear flow or mechanical shear can break aggregates into smaller aggregates and disperse them into individual cells (Shin et al., 2004). RBC aggregation and disaggregation are reversible processes that can occur in response to ambient shear forces, and aggregates will re-occur when external shear force is decreased (Shin et al., 2006). Thus, the degree of aggregation is determined by the balance of forces between the promoting aggregation and the opposing aggregation. Whereas the forces promoting aggregation are only partially understood through two conflicting theories, namely, polymer bridging and depletion layer theory (Neu and Meiselman, 2002). The factors affecting RBC aggregation can be divided into cellular and plasmatic factors (Rampling et al., 2004). Cellular Factors Cellular factors involve the intrinsic tendencies of RBCs to form aggregates, whereas plasmatic factors concern the nature of a suspending medium. RBC aggregation substantially differs from RBC aggregability, even though both terms are frequently conflated. RBC aggregability reflects the intrinsic tendency of RBCs at a cellular level to form aggregates, whereas RBC aggregation refers to the overall tendency of RBCs to aggregate. Since RBC aggregability is a cellular property promoting RBC aggregation, plasmatic factors such as the concentration of fibrinogen should be excluded in accessing RBC aggregability. An assessment of RBC aggregability in different samples can be undertaken replacing autologous plasma with a standard medium (Meiselman, 2009). Water-soluble high molecular weight polymers, such as polyethylene glycol, polyvinylpyrrolidone, or dextran, are recommended as standard suspending media. Seventy kilodalton (kDa) dextran is frequently used for RBC aggregability comparison testing (Meiselman et al., 2007). Such measurements exclude the effects of donordependent plasma protein on RBC aggregation and thus distinguish cellular factors affecting RBC aggregation. Typical cellular factors are surface charge density (Nash et al., 1987) and membrane strain (Meiselman, 1993). Plasmatic Factors Concerning the plasmatic factors, there are various relevant plasma proteins and osmolality-affecting components. Plasma proteins affecting RBC aggregation have been found to be fibrinogen, albumin, and globulin (Yamamoto, 1986;Armstrong et al., 2004). Among them, fibrinogen has been shown to be the most important factor of RBC aggregation (Schechner et al., 2003). However, it remains unclear how these plasma proteins interact with each other and in combination in relation to RBC aggregations. Some studies have reported synergetic effects on RBC aggregation in solutions comprising both fibrinogen and albumin (Ben-Ami et al., 2003;Lee et al., 2016b). Measurement Techniques Three commercially available aggregometers for the measurement of RBC aggregations include the Myrenne Aggregometer (Myrenne GmbH, Roetgen, Germany), the LORCA (R&R Mechatronics, Hoorn, Netherlands), and the Determinants Measurement techniques Deformability Intrinsic factors -ratio of surface area to volume) -rheological properties of hemoglobin -membrane protein phosphorylation -cytoskeletal integrity. metabolism (Kim et al., 2015;Huisjes et al., 2018) Experimental alteration -variations in osmotic pressure, calcium, NO and temperature -membrane protein/lipid modification -Hb-Membrane interaction, -ATP-depletion (Singh and Shin, 2009) Filtration -membrane (Reid et al., 1976): -microfluidic filter (Bransky et al., 2007) -micromesh (Odashiro et al., 2015) -clogging, laboratory, low reproducibility RheoScan-A (Rheomeditech, Seoul, Korea). These aggregometers adopt the same principle of analyzing the syllectogram during the aggregation process but use different geometries (Baskurt et al., 2009c). The Myrenne Aggregometer consists of a lower cone and an upper plate with infrared diode and detector (Martínez et al., 1996). Prior to testing, the sample is sheared at 500 s −1 for 10 s and RBC aggregates are dispersed. Then, the initial shear processing is abruptly ceased and transmitted light is measured for 10 s and the results are analyzed as stasis aggregation index, M. Another aggregation index (M1) can be measured at a low shear rate (3 s −1 ) instead of the stasis. The LORCA system consists of a rotating cup and a stationary bob with laser diode and two photodiodes built into the bob (Hardeman et al., 2001). A laser beam is applied onto the sample, and the backscattered light is measured with the photodetectors. The dispersing shear rate is 600 s −1 . After abrupt stop of the shearing, backscattered light from the sample is recorded for a 120 s and then analyzed. The RheoScan-A involves the use of a microchip consisting of a circular test chamber and a small metal stir bar. Owing to the disposable test chip, a test can be completed within 3 min (Shin et al., 2009b). When the stir bar rotates at 1000 RPM for 10 s, pre-existing RBC aggregates can be completely dispersed. After abrupt stoppage of the rotating stirrer, the transmitted light is recorded for |
120 s and the light intensity-time curve is analyzed. The amplitude (AMP) and half-time (T 1/2 ) parameters can be determined in a manner similar to other devices. The M index is the area below the syllectogram during the first 10 s and the AI refers to the ratio of the area below the syllectogram to total area during the first 10 s (Shin et al., 2009b). Critical Shear Stress Conventional aggregation indexes use individually specific measurement parameters or use units of measurement that are arbitrary, such as M and AMP. Since these indexes are strongly dependent on the characteristics of specific instruments, quantitative comparisons between studies are essentially precluded. To overcome these difficulties of comparison, several studies have introduced CSS to disaggregate or disperse RBC aggregates completely. Considering various magnitudes of shear flow of in vivo blood circulation, the reversible dynamics of RBC aggregation and disaggregation would occur repeatedly. RBCs are disaggregated in arteries except in cases of ischemia, but RBC aggregates are easily observed in venules. Some studies have measured the critical shear rate (CSR) as the minimum shear rate to disperse RBC aggregation (Hardeman et al., 2001). However, the CSR is strongly affected by hematocrits, which effect should be corrected. Measurement Techniques Given the limitations of the CSR, CSS has been used for measurement as it is hematocrit-independent FIGURE 2 | Schematic of microfluidic ektacytometry to measure red blood cell deformability (reproduced with permission from Shin et al., 2007a). FIGURE 3 | Schematic of microfluidic aggregometry to measure critical shear stress. (Razavian et al., 1992). Similar to the CSR, CSS is the minimum shear stress to disperse RBC aggregation. To measure CSS, one must monitor RBC aggregation while simultaneously varying a wide range of shear stresses. Due to the complex and difficult technical requirements, the CSS measurements has been delayed. Owing to development of microfluidics and optics, a transient microfluidic aggregometer was successfully developed to measure the CSS directly within 20 s (Hou and Shin, 2008;Shin et al., 2009b). The microfluidic aggregometer was further developed as a commercial instrument, named as the RheoScan-D300 (Rheomeditech, Seoul, South Korea). The disposable chip of the RheoScan-D300, as shown in Figure 3, is composed of a sample chamber holding a whole blood sample, a microchannel, and a waste chamber with a rubber lid. The whole blood is stored in a sample chamber and flown by the pressure difference through the microchannels. While the sample flows through the micro-channel, the backscattered light is measured with the photodiodes and pressure data are recorded over time. The light intensity gradually increases and then decreases. When the light intensity reach the maximum, the corresponding shear stress is defined as the critical shear-stress (τ c ). The operating principles to measure CSS have been reported elsewhere (Shin et al., 2009a). Factors Affecting CSS Measurements Similar to conventional aggregation indexes, cellular and plasmatic factors affect CSS. In a previous study (Xue et al., 2013), CSS was examined with varying fibrinogen concentration as well as RBC deformability. With increasing concentrations of glutaraldehyde (GA) (0.001-0.005%) or heat treatment (HT) at 49 • C, over increasingly extended time intervals (0-7 min), the RBC deformability gradually decreased, resulting in a proportional increase in CSS. The effect of cell deformability on CSS was even greater with higher concentration of fibrinogen (2-6 g/L). This indicated that there was a synergetic amplification of CSS in the presence of both reduced deformability and elevated fibrinogen levels. Both conditions of reduced RBC deformability and increased fibrinogen levels are commonly observed in diabetic patients who have microcirculatory diseases (Singh and Shin, 2009). The Advantages of CSS as an RBC Aggregation Index CSS is the minimum shear stress required to disperse RBC aggregates, and appears to be an excellent index to represent RBC aggregation. First, CSS is hematocrit-independent so hematocrit correction is not required, whereas for the CSR, the AI increases and the threshold shear-rate decreases when the hematocrit increases (Shin et al., 2009b). Second, CSS holding dimensional unit such as the millipascal (mPa) is a more physical value to compare directly with those measured with other devices than any other aggregation indexes. Most conventional aggregation indexes such as the M-index use arbitrary unit; therefore, the results obtained cannot be compared across studies. Furthermore, CSS also reflects cellular factors as well as plasmatic factors. Therefore, CSS appears to be a potentially useful index to represent RBC aggregation (Kim et al., 2013). CLINICAL ASSOCIATION OF HEMORHEOLOGY WITH DN Many studies have examined the clinical association of hemorheological alterations with DN. The clinical association of hemorheological properties with diabetic mellitus are summarized in Table 2. Since the present study considered clinical studies for human subjects, animal studies were excluded in this review. A filtration method involving meshes has been used and typical pore sizes have ranged from 3 to 5 µm. However, clinical results have varied widely depending on pore size (Tomaiuolo, 2014). Zimmermann et al. (1996) reported that RBC deformability decreased with the severity of DN and showed a high correlation with DN (n = 58). They found that impairment of RBC deformability associated with the concentrations of glycosylated hemoglobin in patients with T2DM. Kikuchi et al. Diabetic mellitus [Deformability] -Impaired deformability in T2DM -Hyperglycemia-induced glycation and oxidation (Resmi et al., 2005) -formation of advanced glycation end-products (AGEs) including HbA1c -increased internal fluid viscosity & reduced membrane fluidity (Watala et al., 1985;Linderkamp et al., 1999) [Aggregation] -Increased aggregation in T2DM -reduced charges in RBC membrane (sialic acid moieties of glycoproteins) -increased fibrinogen level and decreased albumin in T2DM lead to synergistic increase of RBC aggregation - (Angelkort, 1999;Vayá et al., 2011;Li et al., 2015;Mahendra et al., 2015) Diabetic Nephropathy -significantly higher CSS in patients with DN than in those without DN (p < 0.001) -CSS cut-off value: 312.67 mPa -moderate sensitivity (60.2%), specificity (60.3%) and AUC (0.635) -No significant difference of RBC deformability alone to the degree of DN (1982) also reported impaired RBC deformability in patients with kidney failure (n = 74). Sotirakopoulos et al. (2004) examined RBC deformability in patients with CKD on hemodialysis or peritoneal dialysis (n = 88). They found impaired erythrocyte deformability in dialysis patients and that RBC deformability was more severe immediately post-dialysis. Two studies have been conducted using more systematic analyses. Brown et al. (2005) examined the association of impaired RBC deformability and DN using a 3 µm-pore mesh (n = 57). They reported that there were distinctive differences in RBC deformability between patients with diabetes mellitus without nephropathy and those with DN (p < 0.01). They also found a close correlation between RBC deformability and creatinine level in plasma. Saito et al. (2011) adopted the same filtration method using nickel mesh. They reported that RBC filterability for patients with diabetes was much lower than that for controls. However, erythrocyte filterability did not show any significant correlation with either the respective DN nor antidiabetic treatments. Some studies of DN have used a microfluidic ektacytometer (RheoScan), which is suited to clinical environments. Shin et al. (2007b) investigated the association of RBC deformability with DN and DR (n = 212) and reported that RBC deformability of the diabetic patients with diabetes mellitus showed a significantly lower than that of control group. Moreover, for patients with complications including CKD, end-stage renal failure, DR, and for those with a combination of DR and DN, RBC deformability was further reduced when compared to patients without such diabetic complications. Lee et al. (2015) examined various hemorheological parameters in patients with T2DM at different stages of CKD (n = 105), and found significant decrease in AI, deformability (EI), CSS, fibrinogen, and the ACR (all p < 0.05). Also, the deformability at 3 Pa (EI @3Pa ) was found to be an independent predictor of GFR in multiple regression analysis. Recently, Lee et al. (2018) reported that significantly different values metabolic and hemorheological parameters were observed according to the progression of DKD, as listed Table 3. Among them, the 'fibrinogen × ESR/EI' showed a significant difference at the moderate CKD and severe CKD stages. With this parameter, the prevalence of microalbuminuria was diagnosed with the moderate sensitivity (74.5%) and specificity, (63.1%), respectively. Chung et al. (2018) had conducted a similar clinical study using the same instrument as Lee et al. (2018). However, they reported somewhat different results in relation to DN (n = 421). CSS was much higher in patients with DN than in those without DN (317.43 ± 125.11 vs. 385.22 ± 182.89, p < 0.001) ( Table 4). After considering some factors including age, sex and diabetic duration, the highest tertile of CSS showed three folds the risk of DKD compared to compared to the lowest CSS tertile. With a CSS cut-off value (312.7 mPa), the estimated GFR (eGFR) was also moderately predicted with 60.3% of sensitivity and 59.6%, specificity. Also, uACR was diagnosed with 60.2% of sensitivity and 60.3% of specificity. In these two recent studies (Chung et al., 2018;Lee et al., 2018), RBC deformability did not show a significant difference according to the degree of DN. This was an unexpected result given that RBC deformability has been proposed as a potential index of diabetic microangiopathy for more than two decades. Lee et al. (2015) had earlier proposed the use of a fibrinogen EI, and later reported that a fibrinogen × ESR/EI showed slightly higher sensitivity than the fibrinogen EI in predicting DN among various hemorheological markers . Fibrinogen, which is elevated in patients with diabetes, has been associated with RBC aggregation (Vayá et al., 2011). The ESR, commonly known as an indicator of inflammation, has also been found to associate with RBC aggregation in diabetic patients (Li et al., 2015). Therefore, the fibrinogen × ESR/EI, as a combined index of hemorheological factors, has been recently proposed in diagnosing DN . However, in Chung et al. (2018), the fibrinogen EI was marginally higher for the eGFR in a group aged < 60 years old, whereas CSS alone showed a significant difference between patients with diabetes with and without DN. As noted, CSS is significantly affected by fibrinogen concentration and cell deformability (Xue et al., 2013). Thus, CSS may provide the same information as the fibrinogen × ESR/EI. CSS is another index of RBC aggregation; however, CSS has a crucial advantage in that it does not need to adjust hematocrit, unlike the conventional aggregation indices (Shin et al., 2009a). Also, CSS holds the dimensional unit such as mPa, which enables to direct comparison of the measured CSS values The P-value represents the analysis of variance P for the baseline measures among the groups. Data are presented as the mean (standard deviation). a p < 0.05 between the non-CKD group and CKD group 1. b p < 0.05 between the non-CKD group and CKD group 2. c p < 0.05 between the non-CKD and CKD group 3. d p < 0.05 between the non-CKD group and CKD groups 4,5. e p < 0.05 between CKD group 1 and CKD groups 4,5. f p < 0.05 between CKD group 2 and CKD groups 4,5. g p < 0.05 between CKD group 3 and CKD groups 4,5. CKD, chronic kidney disease; CSS, critical shear stress; EI, elongation index at 3 Pa; ESR, erythrocyte sedimentation rate (reproduced with permission from Lee et al., 2018). between different instruments (Lee et al., 2016a). Similar to whole blood viscosity, CSS values vary with temperature (Lim et al., 2010). Therefore, CSS may be a potential index to diagnose DKD. CONCLUSION AND FUTURE PERSPECTIVES Diabetes mellitus is a growing burden on global healthcare. Diabetes-related complications have led to an increasing mortality rate. DKD is a main diabetic complication, occurring in patients with a long duration of diabetes. DKD occurs due to the dysfunction of glomerular microvasculature of the kidney. DKD is a clinical syndrome characterized with persistent albuminuria, which can be identified through observing a continuous decline in the GFR. The uACR is a common method to screen for DN; however, the uACR is influenced in relation to various factors including a patient's general dietary condition, medications, comorbidities, and improper urinary sampling (Wu et al., 2014). It is suggested to confirm the diagnosis of microalbuminuria through repeated urine tests over a period of 3-6 months (Nazar, 2014). Hemorheological alteration has been suggested as a potential biomarker of DN for more than three decades. Most studies |
have reported that cell deformability is generally reduced in patients with DKD, but findings suggest that it is not sufficiently precise to use for patient screening. However, more recent studies have reported that cell deformability alone did not show a high correlation with DKD. Instead, two other parameters have been proposed to diagnose DKD or DN. One involved a combination of hemorheological parameters, namely, the fibrinogen × ESR/EI , and the other used CSS (Chung et al., 2018). Both parameters represent reduced deformability, increased aggregation of RBCs, and elevated fibrinogen concentration. Furthermore, the effects of these parameters has been considered in relation to whole blood, and also showed a close correlation with DN (Monica et al., 2019). However, whole blood viscosity is complex, with hemorheological parameters affected due to various factors non-related to diabetes mellitus. A careful combination of hemorheological parameters directly related to DN would more likely lead to a successful diagnosis of DN. CSS is not only associated with diabetes-related microvascular complications but also diabetes-related macro-vascular complications. Park et al. (2018) reported that CSS was strongly correlated with acute myocardial infarction in patients with T2DM. CSS has also been found to correlate highly with hyperlipidemia (Gyawali et al., 2015a), chronic inflammatory indexes, and oxidative stress (Gyawali et al., 2015b). Furthermore, Gyawali et al. (2014) reported that CSS alone could be used to diagnose metabolic syndrome (MetS) with higher sensitivity and specificity than other conventional indexes such as the ankle brachial pressure index. The cut-off value of CSS for diagnosing DKD was 310 mPa (Chung et al., 2018), whereas the mean CSS in healthy controls was 200.5 mPa (Shin et al., 2009a). Cho et al. (2014) reported a 30% increase in CSS in acute coronary syndrome: 265 mPa in patients with stable angina, 338 mPa in patients with unstable angina, and 324 mPa in patients with acute myocardial infarction. Therefore, CSS can be considered as a potential index to diagnose not only DKD but also other diabetes-related complications. More research can be valuable in confirming the relationship between elevated CSS and diabetic vascular complications and finding each relevant cut-off value to use as a screening tool. AUTHOR CONTRIBUTIONS HL, WN, SL, CA, JM, KW, and SS contributed to the literature survey, brainstorming, writing, and critical review of the manuscript. HL, WN, and SS edited the manuscript. A Study on Vitamin D Level in Critically Ill Children Admitted in PICU at A Tertiary Care Hospital Department of Pediatrics, Silchar Medical College and Hospital, Silchar, Assam, India *Corresponding Author Somrita Chakravarty Department of Pediatrics, Silchar Medical College and Hospital, Silchar, Assam, India Abstract Introduction: Vitamin D is critical in bone metabolism and calcium homeostasis,as well as an important regulator in metabolic processes, cardiovascular and immune systems. It is crucial for humans to have sufficient vitamin D levels to maintain bone health and improve response to infection. Vitamin D deficiency (VDD) represent a potentially modifiable risk factor for greater illness severity and clinical outcome during critical illness. The objective of this study is to assess Vitamin D status in critically ill children admitted in PICU and to determine the frequency of VDD amongst them and its association with disease severity and important clinical outcomes. Methods: It was a prospective observational study conducted over a period of 6 months in the PICU of tertiary care hospital on 100 critically ill children aged 1 month –12 years. Serum Vitamin D levels were estimated within 24 hours of admission to PICU. Severity of illness was assessed using Pediatric Index of Mortality III (PRISM III) score. The need for catecholamines and mechanical ventillation, length of PICU stay and mortality were also recorded. Results: The prevalence of VDD among critically ill children was 49%.Vitamin D level was significantly lower in VDD group as compared to ‘no deficiency’ group (16.3±2.5 vs 38±5.9, p value 0.01). There was also a trend toward increased disease severity, higher need of catecholamines and mechanical ventillation in VDD group. The length of PICU stay (pvalue0.004) and mortality rate (pvalue0.006) was also significantly higher in VDD group. Conclusion: VDD is common in critically ill children and associated with greater illness severity, longer PICU stay and higher mortality. Introduction Vitamin D is an essential hormone that is very important for optimal health. It is a fat soluble vitamin with key endocrine functions. 1 This pleiotropic hormone plays a major role in regulation of calcium metabolism, maintenance of bone integrity, cardiovascular system, and several pathways of immune system in general and innate immune system in particular. 2 It activates toll-like receptors in leukocytes →leading to induction of antimicrobial peptides that kill micro organisms (predominantly cathelicidin). 1,25(OH)2D3 exerts its molecular effects by binding to the vitamin D http://jmscr.igmpublication.org/home/ ISSN (e)-2347-176x ISSN (p) 2455-0450 DOI: https://dx.doi.org/10.18535/jmscr/v8i1.138 receptor .Vitamin D and vitamin D receptor (VDR) expression and activity are associated with immunity against various infections and autoimmune diseases. 3 Levels of 25-hydroxy vitamin D (25(OH)D) are used to assess the adequacy of vitamin D stores in our body. Patients with levels less than 20 ng/mL are commonly categorized as vitamin D deficient 2,4 and treatment is initiated in children in order to prevent rickets 4 . Various studies across the world including India have estimated the prevalence of VDD in healthy children to be in the range of 10-90%. 4,5,6 Such a high prevalence of VDD may be the result of inadequate dietary intake of vitamin D, inadequate exposure to sunlight because of an indoor lifestyle, a high level of skin pigmentation. 7,8 Any critical illness or injury disturbs the internal milieu of the body and may adversely affect the reserves of vital nutrients and minerals of the body. In addition, nutrition is often suboptimal in these group of patients and this, in turn, may further aggravate any deficiencies. Vitamin D deficiency has been reported to be widely prevalent in critically ill children from ICUs worldwide. [9][10][11][12] So this study has been undertaken to estimate the prevalence of vitamin D deficiency in critically ill children admitted in PICU and also to investigate any association between VDD and important clinical outcomes including illness severity, requirement for ventilation and catecholamines, length of PICU stay and mortality. Materials and Methods Design and Settings This was a prospective observational study conducted over a period of 6 months (July 2019 -December 2019) in children admitted to the pediatric intensive care unit (PICU) of our tertiary care hospital. The severity of illness was measured using the Pediatric Risk of Mortality III (PRISM III) score at admission to the PICU .The patients were followed up during their PICU stay and the following parameters were recorded: the need for mechanical ventilation and catecholamines, development of multiple organ dysfunction syndrome (MODS), septic shock, hypocalcaemia, the length of PICU stay and any mortality. Standard definitions were used for sepsis, septic shock and MODS. 13 Participants A total of 100 critically ill children aged 1 month to 12 years admitted in PICU were eligible for inclusion as cases. Exclusion Criteria Children who were already on vitamin D supplementation, or had received large doses for rickets or any documented vitamin D deficiency in the past 1 year, those on corticosteroids and children with any underlying chronic liver or kidney disease or had recent renal stones were excluded from the study. Baseline demographic data such as age, gender, weight, cause of PICU admission, signs suggestive of severe acute malnutrition (SAM) (as defined by WHO) 14 were recorded Ethics The study was approved by the Institutional Ethical Committee .After obtaining informed written consent from the parents, the eligible children was enrolled in the study. Vitamin D measurement Venous blood samples (2.0 ml) were collected from all cases (within 24 hours of admission to the PICU) for analysis of serum 25(OH)D in the Central Composite Laboratory under Biochemistry Department of the institution. VDD was defined as a level of 25(OH)D <20 ng/ml. 15 All the pediatric cases included in the study were divided into two groups as Vitamin D deficiency group (25[OH]D <20 ng/ml) and 'no deficiency ' group(25[OH]D ≥20 ng/ml). Outcome 1. The primary outcome measure was to estimate the prevalence of VDD in critically ill children admitted in the PICU of our tertiary care centre. 2. Secondary outcomes were to compare vitamin D-deficiency group and no deficiency group for severity of illness (PRISM III score), shock, MODS, need for mechanical ventilation and catecholamines, length of PICU stay and mortality. Statistical Analysis Data entry and statistical analyses were performed using Microsoft Excel 2010 and SPSS software version 20. Demographic variables were recorded as percentages, means, standard deviations (SD), medians and interquartile ranges (IQR), as applicable. Dichotomous outcomes were compared by using the x2 test. Continuous variables were compared by Student's t-test. A P value <0.05 was taken as significant. Results A total of 150 children admitted to the PICU during the study period were selected for the study. Out of them, 44 children were excluded due to prespecified exclusion criteria and 6 children as their parents refused to give consent. So the final outcome was measured in 100 critically ill children. The study flowchart and recruitment are shown in Figure 1. Baseline demographic and clinical data are described in the subsequent tables. The mean age of study population was 4 years (SD = 2.9).In terms of gender, there was a male preponderance (68%). Thus, in our study, the prevalence of vitamin D deficiency among critically ill children was found to be 49%. (Table 3) Vitamin D levels were significantly lower in vitamin D deficiency group as compared to 'no deficiency 'group. (p=0.01). The severity of illness as assessed by PRISM III score was significantly higher in vitamin D deficient group (p value 0.003). More cases with vitamin D deficiency had SAM compared to not deficient cases, but the difference was not statistically significant. Also, the study showed an increased requirement of catecholamines in vitamin D deficient group as compared to 'no deficiency group', but the difference was not statistically significant. There was also a trend towards increased occurrence of septic shock, hypocalcaemia and MODS in VDD cases, but the difference was statistically insignificant. (Table 4 Table 5) Lastly, when considering the outcome between the two groups, mortality rate was found to be significantly higher in those with vitamin D deficiency (Table 6) 19 Few studies have documented significant association of vitamin D deficiency with poor outcomes such as longer duration of ICU stay 10,19 , increased requirement of inotropes 11,15 and higher admission illness severity scores 9,10 and death rates, increased rates of ventilator-associated pneumonia, blood culture positivity and an increased incidence of organ dysfunction Our study revealed a higher illness severity (higher PRISM III SCORE) in VDD group as compared to 'no deficiency' group, and the difference was statistically significant. A few studies found an inverse correlation between VDD and various indices of illness severity 9,10 while others found no such association. 11, 15 Madden K et al 9 noted that the median PRISM-III raw score on admission day was 5 (IQR 0-11.5), and it was inversely correlated with 25(OH)D level (r = -0.23, P value <.0001) .In a study conducted by Ebenezer K et al 12 conducted in South India, higher PIM score was found to be associated with low vitamin D levels (r s = −0.29, p v= 0.04). Ponnarmeni S et al16 also found a higher illness severity (higher PRISM III score) in VDD group as compared to no deficiency group (17 (13-22) vs 14 (13)(14)(15)(16)(17)(18)(19)(20)(21)(22), p value =0.26). The current study also revealed that children with VDD had an increased incidence of shock and an increased requirement of catecholamines, although the difference was not statistically significant when compared to the 'no deficiency' group. Madden et al. 9 in their study observed that patients who received vasopressors had lower levels of 25(OH)D than those who did not (median 19 16 Lastly, percentage of mortality was also higher in VDD children than those not deficient(30.6% vs 25.4%, p value 0.006).Similarly, Ponnarmeni S et al 16 also noted a higher mortality in VDD group (15.9% vs 14.7%),though the difference was not statistically significant. The higher mortality percentage in our study might be possibly due to increased illness severity at presentation, late referral from the periphery ,increased prevalence of malnutrition and infections or any gap in health care delivery. Strength |
and Limitations Our current study reveals a high prevalence of vitamin D deficiency in critically ill children and its association with greater illness severity and longer duration of PICU stay. A well defined eligibility criteria and systematically done prospective data collection are the strengths of the study. The limitations of the study are as follows. 1) The study was done for a very small period of 6 months only. 2) Certain factors which may have a influence on vitamin D levels could not be investigated: serum phosphorus, alkaline phosphatase and parathyroid hormone. 3) Also, the effect of vitamin D supplementation could not be assessed. Conclusion From the present study, we observed a high prevalence of Vitamin D deficiency in critically ill children in our study population. Vitamin D deficiency was found to be associated with greater illness severity, longer duration of PICU stay and higher mortality. Thus, in future, further large scale interventional studies are needed to be undertaken to study the outcomes of vitamin D supplementation in early stages of critical illness. Screening and Identification of Lujo Virus Inhibitors Using a Recombinant Reporter Virus Platform Lujo virus (LUJV), a highly pathogenic arenavirus, was first identified in 2008 in Zambia. To aid the identification of effective therapeutics for LUJV, we developed a recombinant reporter virus system, confirming reporter LUJV comparability with wild-type virus and its utility in high-throughput antiviral screening assays. Using this system, we evaluated compounds with known and unknown efficacy against related arenaviruses, with the aim of identifying LUJV-specific and potential new pan-arenavirus antivirals. We identified six compounds demonstrating robust anti-LUJV activity, including several compounds with previously reported activity against other arenaviruses. These data provide critical evidence for developing broad-spectrum antivirals against high-consequence arenaviruses. Introduction Lujo virus (LUJV), the first highly pathogenic arenavirus identified in Africa in over 40 years, was recognized after a cluster of severe viral hemorrhagic fever (VHF) cases in Southern Africa in September and October of 2008 [1,2]. The index case was identified in Zambia, and four subsequent nosocomial cases were detected in associated healthcare workers in South Africa. The infections initially presented as a non-specific febrile illness, but progressed in severity over 10-13 days to respiratory distress, neurological signs, and circulatory collapse [1]. To date, no additional cases of LUJV infection have been reported. However, the lack of a defined etiology or source of exposure for the index case; the apparent ease with which primary, secondary, and tertiary contacts were infected; and the unusually high case fatality rate (80%) caused significant alarm. The family Arenaviridae encompasses three genera: Mammarenavirus, Reptarenavirus, and Hartmanivirus [3]. With the exception of the bat-borne Tacaribe virus, rodents are the natural hosts of the mammalian-specific mammarenaviruses, with each viral species closely associated with a single rodent species reservoir [4,5]. Infected rodents develop a persistent infection and continually shed the virus in body secretions such as urine and feces; human transmission occurs via contact with these secretions, either directly or in contaminated food products [6]. Mammarenaviruses are divided into two groups based on the geographic, genetic, and epidemiological relationship with their respective rodent hosts: Old World (or lymphocyte choriomeningitis/Lassa complex) species and New World (or Tacaribe complex) species [7]. Several Mammarenavirus species in both these groups can cause severe VHF disease in humans. Highly pathogenic Old World complex viruses such as LUJV and Lassa virus (LASV) are found in Africa, whereas the New World pathogens such as Junin (JUNV), Guanarito (GTOV), Machupo, and Chapare viruses are found in Rescue of Recombinant LUJV Recombinant reporter LUJV expressing the green fluorescent protein ZsGreen1 (rLUJV/ZsG) was generated using a previously described reverse genetics system [12]. Briefly, to rescue recombinant LUJV, T7 promoter-based plasmids containing full-length antigenomic sense LUJV L and S genome segments (prototypic Zambian strain; Genbank NC_012777.1 and NC_012776.1, respectively) were transfected into BSR-T7/5 cells. The reporter rLUJV/ZsG was generated by substituting the full-length antigenomic sense S segment with a modified version encoding ZsG. At 5 days post-transfection, cell culture supernatants were harvested, clarified by low-speed centrifugation, and used to infect Vero-E6 cells. Virus stocks harvested 3-4 days post-infection were quantified in Vero-E6 cells by tissue culture infective dose 50 (TCID 50 ) assays, using either immunofluorescence (wt virus) or ZsG expression (reporter virus) [13]. Next-Generation Sequencing and Bioinformatics RNA was extracted from cell culture supernatants (clarified by low-speed centrifugation) using the MagMAX-96 Total RNA Isolation Kit (Thermo Fisher Scientific, Waltham, MA, USA) on a 96-well Roche MagMAX extraction platform with a DNase-I treatment step according to manufacturer's instructions. RNA sequencing was performed using KAPA RNA HyperPrep kit (Roche, Basel, Switzerland) for library preparation and analyzed on the MiniSeq system (Illumina, San Diego, CA, USA). Data Analysis ZsG fluorescence values for compound-treated infected cells were normalized to mocktreated (DMSO only), infected cells and used to fit a 4-parameter equation to semi-log plots of the concentration-response data. From this, the compound concentrations inhibiting ZsG expression by 50% (EC 50 ) were interpolated. Cell viability was similarly calculated in compound-treated, mock-infected cells to determine the 50% cell cytotoxicity concentration (CC 50 ) of each compound. The selectivity index was calculated by dividing CC 50 by EC 50 . Data analysis was performed using GraphPad Prism v9 (GraphPad, San Diego, CA, USA). Suitability for high-throughput screening was determined using the Z prime (Z ) score, a measure of statistical effect size, calculated as Z = 1 − (3 × (σ P + σ N )/|µ P − µ N |, where σ P = standard deviation of positive control, σ N = standard deviation of negative control, µ P = mean of positive control, and µ N = mean of negative control. Z values between 0.5 and 1.0 were considered acceptable. Signal-to-noise ratio was determined as (µ P − µ B )/σ B , where µ P = mean signal of positive control, µ B = mean background signal, and σ B = standard deviation of background signal. Significance was calculated using a one-sample t-test. Generation of a Recombinant LUJV Expressing ZsGreen1 Fluorescent Protein LUJV contains two ambisense genome segments termed large (L) and small (S) with respect to their relative nucleotide length ( Figure 1A). Each genome segment encodes two genes: RNA-dependent RNA polymerase (RDRP) and Z protein on the L, and the nucleoprotein (N) and glycoprotein precursor (GPC) on the S ( Figure 1B). Transcription of monocistronic viral mRNA is initiated at the untranslated region (UTR) and terminates at the intergenomic region (IGR). Previous attempts to generate mammarenaviruses expressing reporter plasmids have either incorporated additional genome segments, as done with tri-segmented recombinant lymphocytic choriomeningitis virus (LCMV) [14], or introduced additional coding sequences (CDS) into the S genome segment, as in a recombinant reporter LASV [15,16]. Here, to generate a recombinant LUJV expressing the fluorescent protein ZsG, we constructed a modified S genome segment in which the ZsG coding sequence was inserted upstream of the N coding sequence, with a porcine teshovirus-1 2A peptide linker sequence (P2A) inserted between ZsG and NP ( Figure 1B). This genetic format transcribes a single viral mRNA, but the two separate proteins are expressed via a ribosomal skipping event during P2A translation [17]. The modified S genome segment was incorporated into the previously described reverse genetics system for LUJV [12]. The recombinant reporter LUJV, able to express all parental viral proteins alongside the inserted ZsG reporter, was successfully rescued using this system and was termed rLUJV/ZsG. To establish the suitability of rLUJV/ZsG as a surrogate for wt LUJV, we first investigated viral growth kinetics. Huh7 or Vero-E6 cells were infected with either wt LUJV or rLUJV/ZsG at MOI 0.1, and titers were determined at 12, 24, 48, 72, and 96 hpi. In both cell lines, growth kinetics of wt LUJV and rLUJV/ZsG were comparable at each time point assessed. In cells infected with rLUJV/ZsG, strong ZsG expression was observed in monolayers of both cell types at 96 hpi ( Figure 1C). To assess the stability of the ZsG insert, we passaged rLUJV/ZsG 10 times in Vero-E6 cells (in triplicate). At each passage point, strong ZsG expression was detected in the monolayer, indicating that the ZsG-P2A-NP modification in the S genome segment was stable ( Figure 1D). To further assess the genome stability, we used next-generation sequencing (NGS) to analyze the viral RNA from passages 1, 5, and 10, and compared the sequences to those of wt LUJV passaged concurrently ( Figure 1E). The ZsG CDS was retained over 10 passages and no mutations were seen in any of the replicates. Similarly, no mutations were detected in the L genome segment over 10 passages. In the S genome segment, no mutations were seen over 5 passages in either wt LUJV or rLUJV/ZsG. At passage 10, 2 of the 3 replicate samples of both viruses had accumulated one non-synonymous, non-identical mutation. Suitability of the LUJV Reporter Virus For Use in an In Vitro Screening Assay To determine the suitability of rLUJV/ZsG for in vitro screening assays, potential LUJV inhibitors were screened using rLUJV/ZsG in Huh7 cells, which have been previously used in the discovery and evaluation of antiviral compounds for other VHF pathogens [16,18,19]. The assays were optimized using ribavirin, a well-documented broad-spectrum antiviral. The Z score (0.89) and signal-to-noise ratio (190:1) were calculated using 50 µM ribavirin as the positive control and DMSO vehicle only as the negative control and indicated a robust assay. Ribavirin dose-response curves demonstrated a concentration-dependent reduction in ZsG fluorescence, with a calculated EC 50 value of 11.13 ± 1.4 µM and CC 50 of >50 µM ( Figure 1F). These values are broadly similar to those reported for other arenaviruses, including LASV (2.47 µM) and JUNV (18.3 µM) [16,20], indicating that rLUJV/ZsG is suitable for a screening assay that uses reduction in ZsG fluorescence to indicate inhibition of viral replication. Antiviral Compound Screening To identify anti-LUJV therapeutics, a wide range of compounds was identified from published data on arenavirus inhibitors, broad-spectrum antivirals, and preliminary inhouse screens using an FDA-approved library of compounds. Virus inhibition was initially determined for each of 83 candidate compounds at three concentrations (5000, 500, and 50 nM), with cell viability confirmed concurrently (Supplementary Table S1). Twenty compounds demonstrated robust anti-LUJV activity (ZsG fluorescence reduced to <30% of control) while maintaining >80% cell viability. These compounds were further evaluated using an extended concentration curve ( Table 1). Six of the twenty compounds had SI ≥ 10, demonstrating promising efficacy and applicability for therapeutic use; efficacy was further confirmed against wt LUJV. The 20 compounds demonstrating robust anti-LUJV activity can be broadly grouped based on known or predicted modes of antiviral action as: (1) inhibitors of viral replication; (2) inhibitors of viral entry, intracellular trafficking, or virus egress; (3) protein kinase inhibitors; (4) selective estrogen receptor modulators (SERMs); or (5) miscellaneous compounds not falling into the four previous groups. Additional Compounds with Anti-rLUJV/ZsG Properties Four of the compounds selected from the initial screen did not fit into the four previous groups based on predicted mode of action ( Figure 4B). Three of these compounds are FDAapproved compounds: benztropine mesylate (EC 50 5.14 ± 1.52 µM, CC 50 > 25, SI > 4.9) is a neurotransmitter inhibitor (anticholinergic); clemastine fumarate (EC 50 3.75 ± 1.79 µM, CC 50 13.12 ± 2.33 µM, SI = 3.5) acts primarily as an H1 histamine antagonist; and loperamide HCl (EC 50 3.67 ± 0.99 µM, CC 50 12.29 ± 3.57 µM, SI = 3.3) is an opioid receptor that targets µ-opioid receptors in the large intestine and acts as an antidiarrheal agent by decreasing intestinal movement. All three were able to moderately inhibit rLUJV/ZsG and had SI values between 3 and 5. The final compound, obatoclax (EC 50 0.36 ± 0.1 µM, CC 50 1.5 ± 0.33 µM, SI = 4.2) inhibits the Bcl-2 family of proteins and has been investigated as an experimental anti-cancer drug. It inhibited rLUJV/ZsG at sub-micromolar EC 50 value, but was cytotoxic at a relatively low concentrations, which reduced SI to 4.2. Additional Compounds with Anti-rLUJV/ZsG Properties Four of the compounds selected from the initial screen did not fit into the four previous groups based on predicted mode of action ( Figure 4B). Three of these compounds are FDA-approved compounds: benztropine mesylate (EC50 5.14 ± 1.52 μM, CC50 > 25, SI > 4.9) is a neurotransmitter inhibitor (anticholinergic); clemastine fumarate |
(EC50 3.75 ± 1.79 μM, Confirmatory Screening with wt LUJV To confirm that the antiviral activity of the compounds was not specific to the ZsGexpressing recombinant virus, the top six candidates (SI ≥ 10) were selected to be screened using wt LUJV; these compounds were 2-dFC, apilimod, AVN-944, brequinar, BX-795, and favipiravir. These six drugs were tested using a shortened concentration curve (5 dilutions centered around their calculated EC 50 value), with cell viability determined concurrently. All compounds evaluated inhibited wt LUJV titers in a concentration-dependent manner, and all compounds reduced viral titers by approximately 4 logs while retaining >75% cell viability ( Figure 5). These results further confirm the suitability of using a reduction of ZsG fluorescence in rLUJV/ZsG-infected cells as a convenient readout in place of wt virus infection. Discussion Only one outbreak of LUJV has been recorded to date, with four out of five patients succumbing to fatal disease. The effectiveness of existing antivirals in treating LUJV patients is unknown. Ribavirin was administered to the sole surviving patient; however, based on this single treated case, it is not possible to determine the effect of treatment on the outcome. Furthermore, given our knowledge of ribavirin treatment in other arenavirus infections, in which it is only partially effective and associated with significant side effects, investigations into uncovering effective and safe arenavirus therapeutic options are badly needed [7]. High-throughput screening assays utilizing recombinant reporter viruses as surrogates for the wild-type parental virus have previously been successful in Figure 5. Ability of the most potent anti-LUJV compounds to inhibit wild-type virus replication. Titer reduction assays using wt LUJV were performed in Huh7 cells treated with varying concentrations of indicated compound 1 h prior to infection (MOI 0.1). For each compound concentration, inhibition of rLUJV/ZsG and cell viability were concurrently assessed. Four days post-infection, cell culture supernatants were collected and LUJV titers (TCID 50 /mL) were determined in Vero-E6 cells by immunofluorescence assays. For each concentration, LUJV titers (black) and cell viability (blue) are shown. Dotted lines represent the limit of detection for the TCID 50 assay (red) and the mean wt LUJV titer in mock-treated (DMSO only) cells (green). Each point represents the mean of quadruplicate wells, with error bars indicating standard deviation; graphs are representative of at least 3 independent experiments. ZsG fluorescence in rLUJV/ZsG-infected Huh7 cells at each compound concentration is also shown (10× magnification). Discussion Only one outbreak of LUJV has been recorded to date, with four out of five patients succumbing to fatal disease. The effectiveness of existing antivirals in treating LUJV patients is unknown. Ribavirin was administered to the sole surviving patient; however, based on this single treated case, it is not possible to determine the effect of treatment on the outcome. Furthermore, given our knowledge of ribavirin treatment in other arenavirus infections, in which it is only partially effective and associated with significant side effects, investigations into uncovering effective and safe arenavirus therapeutic options are badly needed [7]. High-throughput screening assays utilizing recombinant reporter viruses as surrogates for the wild-type parental virus have previously been successful in identifying several anti-mammarenavirus compounds [15,16,46,47]. Here, using a novel recombinant reporter LUJV, we were able to identify six compounds with high efficacy against LUJV that warrant further evaluation. Given the genetic diversity among species, the large number of different arenavirus host reservoirs, and the geographic range of these pathogens, the emergence of new arenaviruses pathogenic to humans remains an ongoing public health concern [7]. Importantly, several of the compounds identified here as effective against LUJV (i.e., favipiravir, 2-dFC, and brequinar) have also been shown to inhibit other arenavirus species. The nucleoside analogs favipiravir and 2-dFC are both known to inhibit other arenaviruses including LASV, JUNV, Machupo virus, and GTOV [16,48,49]. The most promising inhibitor identified here was the dihydroorotate dehydrogenase inhibitor brequinar (EC 50 0.11 ± 0.02 µM, CC 50 > 12.5 µM, SI > 115), which has previously been shown to inhibit LASV, JUNV, and LCMV at similar sub-micromolar EC 50 values [50]. These broad-spectrum antivirals inhibit a wide range of highly pathogenic arenavirus species, presumably by targeting conserved viral processes, and are thus promising candidates for treating current and future emerging arenaviruses. It should be noted that inhibitors of pyrimidine biosynthesis can exhibit potent antiviral activity in cell culture, but be ineffective in animal models, which is likely due to pyrimidines in the animals' diet overcoming the antiviral effect [51]. While not yet approved, obatoclax is an inhibitor of the Bcl-2 family of proteins and an experimental anti-cancer drug that has undergone several phase II clinical trials; it has been shown to inhibit LCMV [28] as well as multiple other virus species including Rift Valley fever virus and HSV-2 [30]. The ability to rapidly screen large libraries of FDA-approved compounds or small molecules for antiviral efficacy is a huge advantage of our reporter virus systems, since comparable studies using wt virus are difficult to adapt for highthroughput screening and are prohibitively time-consuming, costly, and labor-intensive. The six most promising candidates with both high efficacy and low cytotoxicity fell within three broadly defined groups based on modes of action: inhibiting viral replication; viral entry and trafficking; or protein kinases. The majority (four out of six) of these with the highest SI values (and therefore the most promising therapeutic candidates) were inhibitors of viral replication. Several other compounds that were identified demonstrated potent anti-LUJV activity, but had associated cytotoxicity, leading to low SI values and a narrow therapeutic window. Although this would likely make these compounds unattractive as single-drug treatment options, combination therapy using compounds with distinct modes of action may synergistically inhibit LUJV [19,27,55,56]. This would allow these more cytotoxic compounds to be used at far lower concentrations while maximizing their antiviral potential. Here we describe the development of a recombinant LUJV engineered to express the fluorescent protein ZsGreen1 and validate its use in high-throughput screening assays for large or complex screens containing multiple compounds at various concentrations. We identified several highly efficacious anti-LUJV compounds with multiple predicted modes of action. Future studies are planned to investigate if potential synergistic actions increase the effectiveness of these compounds, as well as to advance the most promising candidates to studies using the available animal models of disease to evaluate in vivo effectiveness. Given the genetic diversity of mammarenaviruses and the high-consequence nature of their infections, it is promising to report that newly emerging species such as LUJV are similarly inhibited by several previously reported compounds. Identifying pan-arenavirus inhibiting compounds serves as a key step in the preparedness against new and emerging pathogenic threats. Variable Effects of PD-Risk Associated SNPs and Variants in Parkinsonism-Associated Genes on Disease Phenotype in a Community-Based Cohort Genetic risk factors for Parkinson's disease (PD) risk and progression have been identified from genome-wide association studies (GWAS), as well as studies of familial forms of PD, implicating common variants at more than 90 loci and pathogenic or likely pathogenic variants at 16 loci. With the goal of understanding whether genetic variants at these PD-risk loci/genes differentially contribute to individual clinical phenotypic characteristics of PD, we used structured clinical documentation tools within the electronic medical record in an effort to provide a standardized and detailed clinical phenotypic characterization at the point of care in a cohort of 856 PD patients. We analyzed common SNPs identified in previous GWAS studies, as well as low-frequency and rare variants at parkinsonism-associated genes in the MDSgene database for their association with individual clinical characteristics and test scores at baseline assessment in our community-based PD patient cohort: age at onset, disease duration, Unified Parkinson's Disease Rating Scale I-VI, cognitive status, initial and baseline motor and non-motor symptoms, complications of levodopa therapy, comorbidities and family history of neurological disease with one or more than one affected family members. We find that in most cases an individual common PD-risk SNP identified in GWAS is associated with only a single clinical feature or test score, while gene-level tests assessing low-frequency and rare variants reveal genes associated in either a unique or partially overlapping manner with the different clinical features and test scores. Protein-protein interaction network analysis of the identified genes reveals that while some of these genes are members of already identified protein networks others are not. These findings indicate that genetic risk factors for PD differentially affect the phenotypic presentation and that genes associated with PD risk are also differentially associated with individual disease phenotypic characteristics at baseline. These findings raise the intriguing possibility that different SNPs/gene effects impact discrete phenotypic characteristics. Furthermore, they support the hypothesis that different gene and protein-protein interaction networks that underlie PD risk, the PD phenotype, and the neurodegenerative process leading to the disease phenotype, and point to the significance of the genetic background on disease phenotype. Genetic risk factors for Parkinson's disease (PD) risk and progression have been identified from genome-wide association studies (GWAS), as well as studies of familial forms of PD, implicating common variants at more than 90 loci and pathogenic or likely pathogenic variants at 16 loci. With the goal of understanding whether genetic variants at these PD-risk loci/genes differentially contribute to individual clinical phenotypic characteristics of PD, we used structured clinical documentation tools within the electronic medical record in an effort to provide a standardized and detailed clinical phenotypic characterization at the point of care in a cohort of 856 PD patients. We analyzed common SNPs identified in previous GWAS studies, as well as low-frequency and rare variants at parkinsonism-associated genes in the MDSgene database for their association with individual clinical characteristics and test scores at baseline assessment in our community-based PD patient cohort: age at onset, disease duration, Unified Parkinson's Disease Rating Scale I-VI, cognitive status, initial and baseline motor and non-motor symptoms, complications of levodopa therapy, comorbidities and family history of neurological disease with one or more than one affected family members. We find that in most cases an individual common PD-risk SNP identified in GWAS is associated with only a single clinical feature or test score, while gene-level tests assessing low-frequency and rare variants reveal genes associated in either a unique or partially overlapping manner with the different clinical features and test scores. Protein-protein interaction network analysis of the identified genes reveals that while some of these genes are members of already identified protein networks others are not. These findings indicate that genetic risk factors for PD differentially affect the phenotypic presentation and that genes associated with PD risk are also differentially associated with individual disease phenotypic characteristics at baseline. These findings raise the intriguing possibility that different SNPs/gene effects impact discrete phenotypic characteristics. Furthermore, they support the hypothesis that different gene and protein-protein interaction networks that underlie PD risk, the PD phenotype, and the neurodegenerative process leading to the disease phenotype, and point to the significance of the genetic background on disease phenotype. INTRODUCTION Parkinson's disease (PD), the second most common neurodegenerative disease, has an insidious onset and a long pre-symptomatic and symptomatic course. Four cardinal features that include resting tremor, bradykinesia, rigidity, and postural instability define the motor aspects of the disease. The constellation of clinical symptoms however is variable both in terms of symptom combination and temporal profile. This variability has led to phenotypic classification according to different disease characteristics. A commonly accepted classification is based on motor symptoms: disease subtypes include a tremor-predominant, akinetic/rigid, and mixed subtype (1). More recently, additional classifications have emerged based on different clinical features such as non-motor features, disease progression, a combination of motor and non-motor features, combination of clinical features and comorbidities, multimodal imaging and genetic burden. More specifically, Sauerbier et al. (2) in their review proposed the existence of a distinct non-motor subtype (NMS) of NMS-dominant PD based on the burden of non-motor symptoms in early PD including cognitive dysfunction, anosmia, anxiety, depression, sleep disorders, and autonomic dysfunction observed either alone or in varying combinations. Simuni et al. (3) reported that, for the Primary Progression Markers Initiative (PPMI) PD cohort, higher baseline non-motor scores were associated with female sex and a more severe motor phenotype. Longitudinal increase in non-motor score severity was associated with older age and lower CSF aβ1-42 at baseline. Lawton et al. (4) identified four phenotypic clusters in their cohort: (1) fast motor progression, (2) mild motor and non-motor disease, (3) severe motor disease, poor psychological |
well-being and poor sleep with intermediate motor progression, and (4) slow motor progression with tremordominant unilateral disease. Mollenhauer et al. (5) in their analysis of the De Novo Parkinson (DeNOPA) cohort, reported that baseline predictors of worse progression of motor symptoms included male sex, orthostatic blood pressure drop, diagnosis of coronary artery disease, arterial hypertension, elevated serum uric acid, and CSF neurofilament light chain. A variable temporal profile of motor symptom appearance and progression has been reported in different cohorts that have been followed longitudinally for different lengths of time and identified predictors of disease progression and phenotypic clusters. In the DeNOPA cohort, predictors of cognitive decline in PD included previous heavy alcohol abuse, current diagnoses of diabetes mellitus, arterial hypertension, elevated periodic limb movement index during sleep, decreased hippocampal volume by MRI, and higher baseline levels of uric acid, C-reactive protein, high density lipoprotein (HDL) cholesterol, and glucose. In their cohort, risk markers for faster disease progression included cardiovascular risk factors, deregulated blood glucose, uric acid metabolism and inflammation. In the PPMI cohort, Aleksovski et al. (6) reported that the postural instability gait disorder (PIGD) subtype, compared to the tremor-predominant subtype, was characterized by more severe disease manifestations at diagnosis, greater cognitive progression, and more frequent psychosis (5). In the PPMI cohort, Latourelle et al. (7) found that higher baseline MDS-UPDRS motor score, male sex, and increased age, as well as a novel Parkinson's disease-specific epistatic interaction, were indicative of faster motor progression. In their retrospective review of a cohort of 100 autopsy confirmed PD cases, Pablo-Fernandez at al. (8) reported that the presence of autonomic dysfunction defined as autonomic failure on autonomic testing or the presence of at least two symptoms such as urinary symptoms, constipation, orthostatic hypotension, or sweating abnormalities was associated with a more rapid progression and shorter survival. Other classifications of disease subtypes have been proposed in addition to motor, non-motor symptom and disease coursebased classifications. Inguanzo et al. (9) employed a radiomics and hybrid machine learning approach to identify mild, intermediate and severe disease subtypes based on a combination of dopaminergic deficit by imaging and escalating motor and non-motor manifestations. In the last two decades, genome-wide association studies (GWAS) of common genetic variants and dissection of the low frequency and rare variants contributing to familial forms of PD has implicated an increasing number of genetic loci in disease risk and severity. This has cemented the view that PD is a complex and heterogeneous genetic disorder, with variants at many genes impacting disease phenotype and course. We are just beginning to understand whether PD-risk variants are differentially associated with baseline features or disease subtype. Tan et al. (10) performed a GWAS of motor and cognitive progression in PD and reported that ATPBB2, a phospholipid transporter related to vesicle formation, is associated with motor progression, and that variants at APOE drive cognitive progression, whereas there was no overlap of variants associated with PD risk and PD age-at-onset with disease progression. Iwaki et al. (11) demonstrated sex-specific SNP associations with features of the PD phenotype: female patients had a higher risk of developing dyskinesias and a lower risk of developing cognitive impairment. Periñán et al. (12) reported an association of the TT genotype at the PICALM SNP rs3851179 with a decreased risk of cognitive impairment in PD. GBA variants have been associated with PD and generally are associated with faster progression and more severe phenotypes (13,14). Blauwendraat et al. (15) reported that in a large PD patient cohort, GBA risk variants decrease age at onset in PD. Genetic factors that increase the risk of PD and genetic factors that affect disease severity and progression are not necessarily identical. Furthermore, individual genetic factors that influence disease severity and progression may not have an immediately identifiable impact in the clinical practice setting. It is therefore important to consider the predictive ability and significance of the impact of genetic variation on individual phenotypic characteristics and parameters that are clinically relevant and may have treatment implications (16)(17)(18). If one or a set of genetic variants contribute differentially to a particular phenotypic characteristic, it will be challenging to discover them using GWAS or gene-level association tests in a genome-wide screen since phenotypically well-characterized cohorts are typically modest in size, making it unlikely to discover genome-wide significant associations. We have therefore taken a focused approach, choosing to evaluate possible associations with SNPs that have been previously demonstrated to show significant associations with PD using large GWAS and low frequency and rare variants at parkinsonism-associated genes identified in the MDSgene database (19), hypothesizing that these genetic variants may differentially contribute to baseline clinical parameters/symptoms. Under this hypothesis, evaluating their association in a smaller cohort of subjects where individual clinical symptoms and objective test scores are obtained at baseline using structured clinical documentation support (SCDS) tools embedded in the electronic medical record (EMR) in a routine clinical practice setting (20) could allow for the discovery of significant associations. This would not be possible in the context of a case-control GWAS. Indeed, we find that common SNPs from PD-risk genes identified in GWAS are individually associated with a range of clinical features: family history of dementia, the presence of hallucinations, bradykinesia, depression, orthostatism, disease subtype, and complications of levodopa therapy. When lowfrequency and rare variants at PD-risk genes and parkinsonismassociated genes are analyzed in gene-level tests, associations with clinical characteristics such as presence of bradykinesia, depression, autonomic symptoms (orthostatism, constipation) UPDRS motor scores, mentation, complications of therapy scores, H&Y stage, and a family history of dementia are identified. All of the associations we report survive Bonferroni correction and some approach or reach genome-wide significance. It is interesting to note that the gene associations identified from the analysis of individual common SNPs do not always overlap with those identified in gene-level tests using low-frequency and rare variants suggesting an important role of the genetic background on the phenotypic manifestations. Subjects and Clinical Information Eight hundred and fifty-six subjects with clinically definite or clinically probably Parkinson's disease (Bower criteria) (21) enrolled in two previously described patient cohorts [Molecular Epidemiology of Parkinson's Disease, MEPD (22), N = 201; DodoNA (23), N = 655] were included in this study. All patients in these cohorts had a diagnosis of PD at study entry and were residents of Cook and Lake Counties in Illinois, USA. Though both cohorts include individuals with diverse ancestries, the filtering described in the following section restricted the analysis to 786 individuals of European ancestry: 504 males, 282 females. Blood samples were collected in the majority of cases at an initial baseline visit or within a 3-month window following the initial visit. Data on clinical parameters were obtained from SCDS developed to standardize clinical assessment and retained within the EMR as described (20,23). Given the community-based practice setting, our cohort included both de novo and previously diagnosed PD patients. The following phenotypic characteristics were analyzed in our cohort: initial motor and non-motor symptoms as reported by the patient, as well as motor and non-motor symptoms identified by the clinician at their baseline encounter. Objective clinical assessment at the baseline encounter included scores on the Mini-mental Status Evaluation (MMSE) / Montreal Cognitive Assessment (MoCA) or Short Test of Mental Status (STMS) (24)(25)(26). Due to copyright limitations, cognitive status was assessed initially using the MMSE, at a later timepoint the MoCA, and finally the STMS. The individual test scores on the MoCA and STMS were converted to MMSE scores using established normograms prior to analysis (26,27). Objective clinical assessments at the baseline encounter also included scores on the Unified Parkinson's Disease Rating Scale (UPDRS) (28) [I -Mentation, Behavior and Mood; II -Activities of Daily Living; III -Motor Examination; IV -Complications of Therapy; V -Hoehn &Yahr stage; VI -Schwab & England Activities of Daily Living Scale], Epworth sleepiness scale (ESS) (29) and Geriatric Depression scale (GDS) (30), information on family history of PD, dementia, stroke, epilepsy, multiple sclerosis, and neuropathy, as well as information on comorbidities including diabetes, cardiovascular disease, migraine, schizophrenia, anxiety, depression, peripheral neuropathy and sleep apnea. Supplementary Table 1 presents the list of clinical parameters and descriptive statistics for these parameters. Treatment details including medical and surgical therapy were collected but not included in the analysis presented here. Genotyping and Quality Control Measures Blood samples were stored at −80 • C until DNA was extracted. Genotypes were obtained by interrogating an Affymetrix Axiom TM genome-wide human array containing 531,674 variants that included custom content, specifically variants at genes associated with PD and other neurological disorders. Prior to imputation using IMPUTE2 (31) against the 1,000 Genomes Phase 3 CEU genome, subjects were filtered in PLINK 1.07 (32) or 1.9 (33) for low overall genotyping rates (<95%) and sexdiscordance, and variants with >5% missing calls were removed from the analysis. Imputed SNPs were retained only if R 2 ≥ 0.90. Only subjects with European ancestry were retained by using principal components one and two (PC1 and PC2) from a principal components analysis (PCA) with 103 ancestry informative markers (AIMs). For association tests with single variants, variants were also filtered by Hardy-Weinberg test statistic (1 × 10 −4 ) and to have a minor allele frequency (MAF) > 1%. Association Tests Genes and variants initially identified for testing association with clinical parameters were selected based on a prior demonstrated association with PD/parkinsonism or disease progression [MDSgene.org; (10)(11)(12)(34)(35)(36)(37)(38)], or because the gene harbors pathogenic variants that cause PD/parkinsonism [for review, see (39, 40)]. Of 168 variants with a previously reported association and a MAF > 1%, 138 variants (Supplementary Table 2) were present in our data after filtering as described above. These were tested using PLINK for association applying logistic regression for binomial variables if at least 3% of subjects (N = 24) displayed the clinical parameter, or linear regression with standardized (mean 0, standard deviation 1) scaled variables and reverse scoring the MMSE so that worse scores indicate poorer performance. Associations were evaluated for both sexes jointly and for each sex separately. Sex, age-at-encounter and, since our community-based cohort includes both de novo and previously diagnosed patients, years-from-diagnosis were included as covariates for associations evaluated in both sexes, age-at-encounter, and years-from-diagnosis as covariates for associations evaluated in just one sex, and years-of-education added as an additional covariate for tests of association with cognitive measures (MMSE). Protein-Protein Interaction Network Evaluation To evaluate whether the genes whose variants exhibited significant associations with clinical parameters identify protein products that are members of a functional protein-protein interaction network, those genes were entered into the Search Tool for the Retrieval of Interacting Genes, STRING, v.11 (43). RESULTS We hypothesized that SNPs which have been previously demonstrated to show significant associations with PD-risk using large GWAS and low frequency and rare variants at parkinsonism-associated genes identified in the MDSgene database (19) differentially contribute to discrete baseline clinical parameters/symptoms. To test this hypothesis, we evaluated their association in two well-characterized patient cohorts [MEPD (20) and DodoNA (23)] where individual clinical symptoms and objective test scores were obtained at baseline using SCDS tools embedded in the EMR. The findings are presented in the following two sections. Single SNP Association Analyses We initially evaluated whether common SNPs that have been previously associated with PD-risk in large GWAS are also associated with distinct binomial clinical phenotypic features of PD at their baseline presentation. We find significant associations that are at times sex-specific, and that the significant SNPs are typically located in non-overlapping genes/regions ( Table 1). Using an additive model, female PD patients carrying the minor allele (T) at SNP rs429358 at APOE, or having an APOE ε4 allele have an ∼8-fold increased risk of having a positive family history of more than one family member with dementia. Individuals with the minor allele (T) at SNP rs3431186 at TMEM175, which encodes a potassium channel that regulates lysosomal membrane potential and pH stability in neurons (44), are about twice as likely to have reduced arm swing, a manifestation of bradykinesia. Male PD patients with the minor allele (T) at SNP rs5396167 in KPNA1, which encodes importin α5 and is involved in lysosomal biogenesis and autophagy (45,46), have a 2.8-fold increased risk to have hallucinations at baseline. Males with the minor (T) allele at SNP rs12528068 108.6 kb from the RIMS1 gene, which encodes one of four isoforms of presynaptic scaffolding |
proteins involved in synaptic transmission (47), have a 2.1-fold increased risk of a history of essential tremor. Individuals carrying the minor allele (G) at SNP rs186798 in ELOVL7 have a 3.8-fold increased risk to also have a prior diagnosis of peripheral neuropathy. The ELOVL7 gene is a PD risk factor that also confers regional vulnerability, i.e., it is a Braak stage-related gene with an altered expression pattern in the brains of PD cases, with down regulated expression in endothelial cells and oligodendrocytes (48) ( Table 1). Additional associations are identified using the GENO-2DF model, which considers both additive and dominance effects. SNP rs2280194 in BIN3 and rs10253857 in an intergenic region near SNX13 are associated with a family history of dementia. SNP rs2074404 in WNT3 is associated with a family history of stroke. SNP rs2694528 in NDUFAF2, which is near ELOVL7, is associated with the presence of neuropathy. SNPs rs8192591 in NOTCH4 and rs1293298 in CTSB are associated with bradykinesia as an initial motor symptom. SNPs rs117615688 in CRHR1 and rs382940 in SLC44A1 are associated the nonmotor symptoms of insomnia and restless leg syndrome (RLS), respectively ( Table 1). We also identified significant associations between common SNPs conferring risk of PD in GWAS and test scores that reflect an objective assessment of the PD patient ( Table 2). The minor allele (T) at SNP rs12528068 in an intergenic region 108.6 kb from RIMS1 that is associated with a history of essential tremor in males is also associated with increased dyskinesia scores in females. SNPs rs113343 and rs6497339 at SYT17, which encodes synaptotagmin-17, are associated with higher GDS scores. SNP rs12283611 at DLG2, which functions in the clustering of receptors, ion channels and associated signaling proteins, is associated with lower UPDRS-VI scores. We included years-from-diagnosis as a covariate in the above analyses since our community-based cohort includes previously diagnosed patients. It is interesting that some results that trended toward significance survive Bonferonni correction if this measure of disease duration is not included as a covariate (Supplementary Tables 4, 5). Individuals with the minor (A) allele at the SNP rs9468199 in an intergenic region 3.2 kilobases (kb) from LOC1005071, an uncharacterized non-coding RNA, are twice more likely to present with the tremor-predominant PD subtype and not the akinetic/rigid or mixed disease subtype. The minor (C) allele at SNP rs12813102 in GPR19, which encodes a proton-sensing G-protein coupled receptor abundant in skin and brain (49), has a relatively strong effect on higher H&Y stage (β ∼1.7 on standardized H&Y scores) in both sexes or just males. In contrast, other SNPs have less strong effect sizes (β range 0.24-0.47 on standardized scores). The presence of the minor allele (C) at SNP rs823118 in NUCKS1, which is involved in homologous recombination DNA repair (50), is associated with higher MMSE baseline scores only in males. Its small effect is not unexpected given that early in the disease process, cognitive impairment is not prominent in typical PD. Finally, the minor allele (T) at SNP rs224750 located 167.5 kb from PARD3 is associated with higher UPDRS-IVc scores only in females. PARD3 is a gene involved in the regulation of cellular junction formation in ependymal cells, cilia, tumor suppression (49). It will be useful to evaluate these variants in longitudinal follow-up studies. In summary, these results collectively demonstrate that some of the PD-risk SNPs identified in case-control GWAS are also associated with the differential presentation of PD and discrete phenotypic characteristics at baseline. Gene-Level Association Analysis We employed gene-level association tests (sequence kernel association tests) to evaluate whether the set of rare (MAF < 1%) or both rare and less common (MAF < 5%) variants present in the PD-associated genes of our cohorts also exert differential effects on baseline clinical features. Significant findings from these gene-level association tests in our cohorts are presented in Table 3. The following findings are notable: LRRK2 is associated with a prior diagnosis of essential tremor (ET). NUCKS1, a gene that shows allele-specific gene expression in the human brain (51), is significantly associated with UPDRS-III motor scores and with UPDRS-V (H&Y stage). Of note, the PD-risk SNP rs823118 in the same gene was associated with higher MMSE scores in males when disease duration was not included as a covariate (Supplementary Table 5). TOX3, a transcriptional co-activator (52) previously associated with periodic leg movements during sleep (53), and SULT1C2, a cytosolic sulfotransferase (54), are associated with the UPDRS-IV total score. TRIM40, a gene whose protein product may function as a E3 ubiquitin-protein ligase (55) and inhibit NF-kB activity, is associated with UPDRS-VI (Schwab & England score). SNCA and FAM184A are associated with dyskinesias at baseline encounter, and SNCA is also associated with cognitive impairment. CHD9, which encodes a transcriptional activator (56) and GPNMB, which encodes a transmembrane glycoprotein (57), are associated with the initial motor symptoms of micrographia (bradykinesia manifestation) and rigidity, respectively. GPNMB, demonstrating genome-wide significance, is also associated with bradykinesia at baseline, as are THSD4, which attenuates TGFβ signaling, and MCCC1, which is used in NFκB signaling (58). STK39, which encodes a protein kinase that may mediate stress-activated signals (51), is associated with RLS. Certain comorbid conditions often seen in PD patients are associated with the different genes. BIN3, which encodes a protein involved in cytokinesis (59) is associated with anxiety disorder. VAMP4 (60) is associated with sleep apnea. PET117, which encodes a mitochondrial protein homolog (61), is associated with traumatic brain injury. Similar results are obtained when sequence kernel association tests are performed without including sex, age at encounter, disease duration and, for MMSE, years of education, as covariates (Supplementary Table 6). In these analyses, different measures of complications of levodopa therapy are associated with some of the genes described above: TOX3 and SULT1C2 are associated with the UPDRS-IVa-Dyskinesia subscore; SULT1C2, MCCC1, TOX3, and BAG3 are associated with the UPDRS-IVb-Fluctuations subscore; and STBD1 is associated with the UPDRS-IVc-Other subscore. In summary, these results demonstrate that variants in PDassociated genes are differentially associated with the following phenotypic features: history of essential tremor, initial motor and non-motor symptoms, test scores, motor and non-motor symptoms at baseline study entry, family history of essential tremor and of dementia, and comorbidities including anxiety, sleep apnea, and traumatic brain injury (TBI). These associations raise the possibility of underlying links between PD, essential tremor, mood disorders, and TBI. Protein-Protein Interaction Network Analysis The protein products of the genes included in this analysis s are involved in many different cellular processes implicated in neurodegeneration. To assess whether the significant associations between SNPs/genes with baseline clinical parameters identified here reflect functional interactions between the genes, we entered all of the genes identified as having significant associations with a phenotypic feature (i.e., all genes listed in Tables 1-3, Supplementary Tables 4-6) into the Search Tool for the Retrieval of Interacting Genes (STRING) and evaluated their participation in protein-protein interaction (PPI) networks. The network shown in Figure 1 was obtained using 0.4 as the minimum required interaction score (medium confidence) and allowed up to 20 second-shell interactions to reveal indirect interactions among these proteins. The network contains 69 nodes and 113 edges (cf. 46 expected) with an average node degree of 3.28 and an enrichment p-value of 1.10 × 10 −16 . Sixteen nodes are unconnected to the protein-interaction network. The top 30 Gene Ontology (GO) processes in which these genes and their interactors are implicated are shown in Table 4, with the genes having significant SNP or gene-level associations highlighted in bold. It is interesting to note that this analysis reveals three interaction patterns: one in which proteins encoded by genes such as APOE, KPNA1, LRRK2, TMEM175, MCCC1, FAM49B, and SNCA are members of closely interacting networks, a second one in which genes such as PARD3 or NUCKS1 are members of more remotely interacting networks, and a third one in which genes such as ELOVL7, GPR19, LCORL, FAM184A, and BIN3 are not nodes in these protein-interaction networks. Several genes occupy central nodes in the protein network: APOE occupies a central node in the protein network and in our analysis is associated with the family history of dementia. APOE is a well-established AD risk factor (62) with an important role in normal brain function (63) and the APOE e4 allele has been associated with cognitive decline in PD (10,64,65). LRRK2 is also occupying a central node in the protein network: LRRK2 has a dual role as a PD risk factor and a gene involved in PD pathogenesis (66,67) and encodes a protein kinase involved in autophagy. SNCA also occupies a central node in the PPI and is a key player in PD pathogenesis (68). Taken together with the results of the association analyses, these results are consistent with the hypothesis that genetic variation that affects the functioning of protein-protein interaction networks can contribute to the differential presentation of PD symptoms. In addition, it is important to note that a number of these genes are members of known networks and hubs, whereas others are not. DISCUSSION Here we present the results of association analyses of baseline clinical features in PD with genetic variants that have been shown to be significant in prior case-control GWAS to confer PD risk or have been identified as PD-associated genes in the MDSgene database. We analyzed discrete clinical phenotypic features and test scores in a two-pronged approach: in the first, we evaluated their association with individual common SNPs that have been demonstrated in case-control GWAS to confer PD-risk; in the second we used gene-level tests to evaluate the association of these phenotypic features with low frequency (1-5% MAF) and rare (<1% MAF) variants in both pathogenic PD genes and the genes conferring PD-risk identified by casecontrol GWAS. The rationale of this approach is based on the hypothesis that individual discrete phenotypic characteristics may be differentially affected by the action of individual SNPs that tag a particular PD-risk haplotype, and/or multiple variants at a particular gene. Furthermore, the observed associations may reflect the effects of variants with different MAF. The alternative to this hypothesis is that the single SNPs and variants within a gene that confer PD-risk affect groups of clinical features or test scores more uniformly. Our results support this hypothesis: individual common SNPs conferring PD risk are associated with phenotypic traits mostly in a non-overlapping manner, and gene-level tests reveal associations with individual clinical features and test scores that are often differentially affected, though at times have overlapping effects. This raises the intriguing possibility that individual phenotypic characteristics of a neurodegenerative disease such as PD that are associated with a specific gene may be related with the same phenotypic characteristic in a different neurodegenerative disease/syndrome. This may allow for the development of a "polyphenic" risk score to complement polygenic composite risk scores that already have been developed for Alzheimer's disease and other diseases (69). It is interesting to point out certain associations that may hint to pathogenetic links between PD and other disorders. The relationship between PD and ET has long been a matter of debate (70). In our cohort, gene-based tests reveal an association between LRRK2 and history of essential tremor. This finding suggests that genetic variation at LRRK2 may provide a link between long-standing ET and the development of PD at least in some cohorts. The presence of neuropathy in our cohorts is associated with variants in the ELOVL7 and NDUFAF2 genes that are located in the same region on chromosome 5. Clinically, peripheral neuropathy has been reported in PD, however, its cause remains unclear, potentially reflecting medication adverse effects (71). Another striking association in our cohort is that of SNCA with cognitive impairment. The role of common variants at SNCA as PD risk factors, as well as rare gene variants as pathogenic mutations has been clearly demonstrated over the last two decades. Our findings suggest that multiple, less common variants at SNCA, not necessarily pathogenic variants, may affect cognition in PD patients. The reported prevalence and incidence estimates in PD show a 1.5:1 male to female ratio (72). Here we find that sex often differentially affects an association with a particular phenotypic trait, either in the form of a symptom or a test score: some of the associations are significant for males or females, whereas others in both sexes. This suggests that sex may have a differential effect on the phenotypic manifestation of |
genetic PD risk. As would be expected from our current understanding of the genetic mechanisms underlying PD, protein-protein interaction network analysis demonstrates that about two-thirds of the genes with significant associations are members of previously identified networks. However, about a third of the genes appear unconnected to these networks. This raises the interesting possibility that as yet unidentified gene networks and connections may be implicated in phenotypic manifestations, in either a deleterious or protective role. It is important to stress that the analyses presented here are based on patient-reported initial symptoms and symptoms at baseline encounter, as well as objective test scores determined at the baseline encounter. Longitudinal evaluation of this and other cohorts through a standardized assessment at annual intervals will enable the extension of this analysis to determine whether the impact of the genotypes on the clinical phenotype and test scores is among other factors dependent on disease subtype, severity and duration. It also will be informative to undertake additional analyses that cluster individual symptoms and analyze their associations with genetic risk factors. One limitation to our study is the inclusion of both de novo and previously diagnosed patients. Therefore, our cohort is likely more heterogeneous than an exclusively de novo cohort such as the PPMI cohort. However, given that the study participation originates in a community-based cohort, it is likely more representative of the phenotypic spectrum that is typically observed in clinician practices. Furthermore, the PD diagnosis in our cohort according to published diagnostic criteria (21) is ascertained at the baseline visit and can also be reliably ascertained at annual intervals using the EMR-based SCDS, thus providing high clinical diagnostic accuracy. In addition, the use of SCDS allows for detailed and accurate clinical data collection in a routine clinical practice, thus more accurately reflecting the clinical course. A second limitation of this study is that the sample size of our cohort limits its power to detect associations. While none of the associations with common PD-risk SNPs reach genome-wide significance (∼5 × 10 −8 ) ( Tables 1, 2, Supplementary Tables 4, 5), gene-level tests using rare variants identify four associations with baseline clinical features that approach or reach significance for the number of mapped genes (2.81 × 10 −6 ): TOX3 and SULT1C2 with UPDRS IV-total score, GPNMB with bradykinesia, CATSPER3 (73) with anxiety ( Table 3, Supplementary Table 6). It is important to point out in this context that the genes included in this analysis have been previously clearly associated with PD-risk in case-control GWAS. Nevertheless, given the size of our cohort, it will be informative to evaluate the reproducibility of our findings in other cohorts. In summary, our analysis shows that common SNPs conferring PD-risk, as well as low-frequency and rare variants in genes implicated in PD/parkinsonism are associated with distinct phenotypic characteristics at baseline presentation in our PD cohorts, supporting the hypothesis that the genetic background significantly affects disease presentation and raising the possibility that it also affects disease course and severity. The associations observed are often, but not always, dependent on sex. It is conceivable that this is related to the observed PD prevalence and incidence estimates that point to PD-risk differences based on sex. Finally, this analysis identifies different patterns in protein interaction networks that may underlie disease phenotype and pathogenesis. Longitudinal studies of this and other PD cohorts using this approach can provide insights on the impact of genetic risk factors on disease severity and progression, and enhance our understanding of the underlying pathogenetic mechanisms contributing to PD. DATA AVAILABILITY STATEMENT The datasets presented in this article are not readily available because this would jeopardize patient confidentiality. Additional information can be made available to qualified researchers after completing a material transfer agreement that maintains patient confidentiality with NorthShore University HealthSystem. Requests to access the datasets should be directed to the corresponding author. ETHICS STATEMENT The studies involving human participants were reviewed and approved by NorthsShore University HealthSystem Institutional Review Board. The patients/participants provided their written informed consent to participate in this study. AUTHOR CONTRIBUTIONS KM designed the study and wrote the manuscript. BC performed data analysis and contributed to the writing. KM, DM, and RF designed clinical instruments used in the study. KM, DM, APP, BS, and NK provided the clinical assessment. AP, LG, and RV provided research assistance. JW, AE, and HY processed genomic and clinical data. KM, BC, RF, and DM edited the manuscript. All authors contributed to the article and approved the submitted version. Vascular Resection for Pancreatic Cancer: 2019 French Recommendations Based on a Literature Review From 2008 to 6-2019 Introduction: Vascular resection remains a subject of debate in the management of Pancreatic Ductal Adenocarcinoma (PDAC). These French recommendations were drafted on behalf of the French National Institute of Cancer (INCA-2019). Material and Methods: A systematic literature search, with PubMed, Medline® (OvidSP), EMBASE, the Cochrane Library, was performed for abstracts published in English from January 2008 to June 2019, and identified systematic reviews/metaanalyses, retrospective analyses and case series dedicated to vascular resections in the setting of PDAC. All selected articles were graded for level of evidence and strength of recommendation was given according to the GRADE system. Results: Neoadjuvant treatment should be performed rather than direct surgery in borderline and locally advanced non-metastatic PDAC with venous and/or arterial infiltration (T4 stage). Patients who respond or those with stable disease and good performance status should undergo surgical exploration to assess resectability because cross-sectional imaging often fails to identify the extent of the remaining viable tumor. Combining vascular resection with pancreatectomy in these cases increases the feasibility of curative resection which is still the only option to improve long-term survival. Venous resection (VR) is recommended if resection is possible in the presence of limited lateral or circumferential involvement but without venous occlusion and in the absence of arterial contact with the celiac axis (CA; cephalic tumors) or the superior mesenteric artery (SMA; all tumor locations) (Grade B). The patients should be in good general condition because mortality and morbidity are higher than following pancreatectomy without VR (Grade B). In case of planned VR, neoadjuvant treatment is recommended since it improves both rate of R0 resections and survival compared to upfront surgery (Grade B). Due to their complexity and specificities, arterial resection (AR; mainly the hepatic artery (HA) or the CA) must be discussed in selected patients, in multidisciplinary team meetings in tertiary referral centers, according to the tumor location and the type of arterial extension. In case of invasion of a short segment of the common HA, resection with arterial reconstruction may be proposed after neoadjuvant therapy. In case of SMA invasion, neoadjuvant therapy may be followed by laparotomy with dissection and biopsy of peri-arterial tissues. A pancreaticoduodenectomy (PD) with SMA-resection is not recommended if the frozen section examination is positive (Grade C). In case of distal PDAC with invasion of the CA, a distal pancreatectomy with CA-resection without arterial reconstruction may be proposed after neoadjuvant therapy and radiologic embolization of the CA branches (expert opinion). Conclusion: For PDAC with vascular involvement, neoadjuvant treatment followed by pancreatectomy with venous resection or even arterial resection can be proposed as a curative option in selected patients with selected vascular involvement. Material and Methods: A systematic literature search, with PubMed, Medline ® (OvidSP), EMBASE, the Cochrane Library, was performed for abstracts published in English from January 2008 to June 2019, and identified systematic reviews/metaanalyses, retrospective analyses and case series dedicated to vascular resections in the setting of PDAC. All selected articles were graded for level of evidence and strength of recommendation was given according to the GRADE system. Results: Neoadjuvant treatment should be performed rather than direct surgery in borderline and locally advanced non-metastatic PDAC with venous and/or arterial infiltration (T4 stage). Patients who respond or those with stable disease and good performance status should undergo surgical exploration to assess resectability because cross-sectional imaging often fails to identify the extent of the remaining viable tumor. Combining vascular resection with pancreatectomy in these cases increases the feasibility of curative resection which is still the only option to improve long-term survival. Venous resection (VR) is recommended if resection is possible in the presence of limited lateral or circumferential involvement but without venous occlusion and in the absence of arterial contact with the celiac axis (CA; cephalic tumors) or the superior mesenteric artery (SMA; all tumor locations) (Grade B). The patients should be in good general condition because mortality and morbidity are higher than following pancreatectomy without VR (Grade B). In case of planned VR, neoadjuvant treatment is recommended since it improves both rate of R0 resections and survival compared to upfront surgery (Grade B). Due to their complexity and specificities, arterial resection (AR; mainly the hepatic artery (HA) or the CA) must be discussed in selected patients, in multidisciplinary team meetings in tertiary referral centers, according to the tumor location and the type of arterial extension. In case of invasion of a short segment of the common HA, resection with arterial reconstruction may be proposed after neoadjuvant therapy. In case of SMA invasion, neoadjuvant therapy may be followed by laparotomy with dissection and biopsy of peri-arterial tissues. A pancreaticoduodenectomy (PD) with SMA-resection is not recommended if the frozen section examination is positive (Grade C). In case of distal PDAC with invasion of the CA, a distal pancreatectomy with CA-resection without arterial reconstruction may be proposed after neoadjuvant therapy and radiologic embolization of the CA branches (expert opinion). Conclusion: For PDAC with vascular involvement, neoadjuvant treatment followed by pancreatectomy with venous resection or even arterial resection can be proposed as a curative option in selected patients with selected vascular involvement. INTRODUCTION-RESECTABILITY OF PANCREATIC DUCTAL ADENOCARCINOMAS (PDAC) Selection of patients for vascular resection is based on the probability of obtaining complete surgical resection (R0), because unlike R1 resection, this can result in prolonged survival or even be curative [Level of Evidence (LE) 3] (1-7). The presence and extent of vascular involvement are determined on high-quality thin-section images, with an anatomical basis for the classification of tumors as "borderline resectable" or "locally advanced" but not metastatic (8,9). Many classifications have been used to define the extent of PDAC, which is based on the relationship between the tumor and the venous or arterial axes (10, 11) (LE 3). The most common system is the National Comprehensive Cancer Network's (NCCN) classification, updated in November 2018 (12) (LE 2) (Supplemental Material 1). The notion of a "borderline" tumor has recently changed to take into account the anatomical classification, the probability of a histologically incomplete resection (R1), the patient's clinical status (general condition, co-morbidities, performance-status, "fragility syndrome") and the "biological" status of the disease (LE 2) (13)(14)(15)(16). The International Consensus on the definition of "borderline" tumors recommends to use a threshold CA 19-9 rate ≥ 500 units/mL for the latter (14) (LE 3) (Supplemental Materials 2, 3). A recent study (17) has shown that a standardized pathological protocol R0-resection based on 1 mm clearance was rarely achieved after upfront venous resection due to microscopic involvement of the SMV-groove (LE 4). It is important to note that patients considered to be at high risk of R1 resection and/or those with an unfavorable clinical and/or "biological" status are now candidates for neoadjuvant therapy (18)(19)(20)(21)(22)(23)(24)(25)(26) (LE 3). In one North American study (27) the benefits of neoadjuvant therapy were found to be significant in the presence of "unilateral" venous involvement (Ishikawa II-III) (LE 3). The PV patency ratio and its improvement under treatment are new prognostic indicators for PDAC treated with preoperative chemo-radiotherapy (28) (LE 4). For borderline resectable PDAC, several more recent studies including two meta-analyses (29, 30) (LE 3), one phase II trial (31) and one randomized controlled trial (32) (LE 2), have confirmed that survival was improved after neo-adjuvant therapy followed by surgery than after upfront surgery followed by adjuvant therapy, even in an intent-to-treat analysis. The NCCN recommendations version 1.2019 (November 8, 2018) state that: "Immediate" resection of borderline tumors is no longer recommended (unlike 2016 recommendations), despite the absence of a randomized trial (neoadjuvant therapy vs. "immediate" surgery) and the definition of the best therapeutic protocol to use" (12,15) The purpose of neoadjuvant therapy is to increase the rate of patients candidates for potentially curative secondary resection. A systematic review published in 2017 (33) compared the pathological |
data in patients who underwent "upfront" surgery to those who underwent surgery after "neoadjuvant treatment." A significant reduction in the relative risk (RR) of R1 resection (RR = 0.66) and other negative predictive factors (tumor size, lymph node metastases, perineural extension, and lymphatic emboli) were observed after neoadjuvant treatment (LE 3) (Supplemental Material 4). Due to the high prevalence of "borderline resectable" and "locally advanced" PDAC (around 15 and 25% respectively) and the lack of consensus about the treatment of theses entities, our aim was to establish recommendations regarding the treatment of PDAC with vascular involvement based on the existing literature. METHODOLOGY The National Institute of Cancer (INCa) commissioned these Guidelines in January 2017 and appointed a guideline leader (chair A.S.) who invited selected authors, all involved in the management of PDAC, to participate in the project development (May 2017). The key questions were prepared by the coordinating team and then approved by the other members. The coordinating team formed task-force subgroups, each with its own leader (J.R.D. for surgery), and divided the key topics among these task forces (October 2017). Process and steps taken to reach the final recommendations were illustrated in Table 1. The INCa team independently performed systematic literature searches, with PubMed, Medline R (OvidSP), EMBASE, the Cochrane Library, and the internet for abstracts published from January 2008 to December 2017. Each task force also performed a systematic literature search. The literature search was restricted to abstracts published in English. Searches were updated every 3 months until June 2019. The search focused on fully published randomized controlled trials (RCTs), metaanalyses, prospective series and national and international guidelines and consensus. However, the literature search concerning vascular resections identified no RCT, only 4 systematic reviews/metaanalyses on venous resection and 2 systematic reviews/meta-analysis on arterial resection; thus, retrospective analyses and case series were also included. Conversely all case reports were excluded. Manuscripts from abstracts containing relevant data were included. A summary of each reviewed manuscript was completed and summarized in literature tables for each key topic to prepare evidence-based and well-balanced statements on the assigned key questions for each task force. All selected articles were graded by the level of evidence and strength of recommendation according to The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system (Supplemental Material Methodology). Each task force developed a draft and proposed statements on their assigned key questions, which were discussed on 4 plenary meetings (from November 2017 to September 2018). Recommendations were formulated based on the available evidence. All recommendations included a Grade rating based on the quality of evidence and strength of recommendation (Supplemental Material Methodology). A first synthesis of the work from different groups was completed in January 2019. Thereafter, a combined document with all recommendations was created, which was reviewed and approved by all the group leaders, finalized and submitted to a national external review by 70 physicians, oncologists and surgeons (out of 126 solicited by the INCa), selected by regional cancer networks and 17 scientific societies collaborating in this project. The final manuscript was drafted after taking into account all comments and answering questions from the external validation group (April 2019) (Supplemental Material Methodology). All authors agreed on the final draft of the manuscript containing the recommendations. After agreement of all group members during a final plenary session held in Paris in May 2019, the guidelines was published online in November 2019 (https://www.e-cancer.fr). These Guidelines will be considered for review every year or sooner if new and relevant evidence becomes available. An update will be done every 3 years after the publication of the recommendations. Any updates to the Guidelines in the interim will be noted on the INCa website. These Guidelines are an official statement of the French Association of Hepatobiliary and Pancreatic Surgery and Liver Transplantation (ACHBT). It provides practical advice on how to manage pancreatic adenocarcinoma. VASCULAR RESECTION FOR PANCREATIC CANCER Venous Resection "What Are the Indications of Venous Resection?" Recommendations Venous resection associated with pancreatectomy is recommended if resection is possible in the presence of limited lateral or circumferential involvement but without venous occlusion and in the absence of arterial contact with the celiac trunk (cephalic tumors) or the superior mesenteric artery (all tumor locations) (Grade B). The patients should be in good general condition because mortality and morbidity are higher than in pancreatectomy without venous resection (Grade B). In the case of a planned venous resection, neoadjuvant treatment is recommended since it improves the rate of R0 resections and survival (Grade B). Comments Performing venous resection (VR) followed by reconstruction of the mesenteric-portal venous axis during pancreatic resections for PDAC may allow "en-bloc" resection facilitated by the superior mesenteric artery (SMA) "first" approach (34,35). Pancreaticoduodenectomies (PDs) are associated with VR in up to 25% of cases in France and Europe (less frequently in the US and more frequently in Japan). Distal pancreatic resections are associated with VR in 5-35% of cases (36)(37)(38)(39) (56)], a segmental resection followed by direct end-to-end anastomotic reconstruction with a "growth factor" (type 3 VR) or a "long" resection. In the latter setting (type 4 VR), if mesenteric root mobilization and lowering of the right liver are insufficient to compensate for the length of the VR (57), interposed graft reconstruction may be used, including an autologous venous or peritoneal (58), a cryopreserved homologous (59), a heterologous (60), or a prosthetic (61-65) graft. A recent study (66) reported the prognosis after reconstruction in 229 VRs (LE 4) and the median benefit to survival with segmental VR followed by a end-to-end anastomosis (usually planned). In this study, 129 patients underwent lateral VR followed by a direct suture (Group 1: 56%), 64 underwent a segmental VR followed by end-to-end anastomosis (Group 2: 28%) and 36 underwent VR followed by interposed graft reconstruction (Group 3: 16%). The surgical morbidity and mortality were comparable in all 3 groups. However, median survival was significantly different in the three groups: 27.6 months, 18.8 months and 13 months in groups 2, 1, and 3, respectively (66). If the venous splenoportal confluence is resected, the splenic vein territory is at risk of: (a) segmental portal hypertension (SPH) with gastric congestion; (b) varicose veins at the gastrojejunal anastomosis and pancreaticojejunal sites and esophageal varices with a late risk of upper GI bleeding; and (c) splenomegaly and thrombocytopenia, in case of prolonged survival (67, 68) (LE 4). In case of gastric congestion, reimplantation of the splenic vein is possible in the inferior mesenteric vein (IMV) (69) or the left renal vein (44, 54) (LE 4). However, reimplantation is not necessary if the resection has preserved the confluence between the splenic and left gastric veins and/or the IMV (70, 71) (LE 4). 5. Impact of the reconstruction technique on the long-term permeability of venous reconstructions. A recent meta-analysis has shown that reconstruction with interposition grafts (IG) influences the long-term permeability of venous reconstructions but not survival (65). This meta-analysis of 14 studies including 257 VRs with IG and 570 VRs without, showed that when venous reconstruction was performed with an IG, postoperative morbidity, mortality, and survival at 1, 3, and 5 years were comparable to those observed with other reconstruction techniques. Finally, acute thrombosis is very rare in the immediate post-operative period. In a multicenter study of 406 VRs, only 7 patients developed acute thrombosis (1.7%) (38) (LE 4) and in a Japanese study of 197 VRs, only 3 patients developed acute thrombosis (1.5%) (75) (LE 4). Overall, the 1-year permeability rates ranged from 82 to 93% (61-63, 75) ( Table 3) (LE 4). Conversely, late thrombosis is frequent, often associated with recurrence (75% after a median of 15 months), and accompanied by portal hypertension and ascites in 75% of cases (61,72,75,76) Finally, it is difficult to intraoperatively distinguish V+ from adventitious fibrotic adhesions secondary to peritumoural inflammation (96), particularly after neoadjuvant treatment. Although a desmoplastic reaction will result in negative pathologic examination of the resected vein, the benefit of a neoadjuvant strategy exceeds the risk of incomplete resection (15) (LE 2). In all studies, the survival of V-patients is comparable to that of patients with "standard" resection (97) (LE 4). However, a matched comparative study on small samples (98) (19 PD+VR with V-: 10 "upfront" VR/9 after chemoradiotherapy vs. 19 patients in the control group: 11 "upfront" VR/8 after chemoradiotherapy) reported that survival was better in patients who underwent VR and whose vein was V-than in patients who underwent standard PD, suggesting the benefit of systematic venous resection in the absence of any venous contact (LE 3). (77). Moreover, the N+ rate was higher (80 vs. 65%) and the prognosis was poorer than that in patients with "clearly" resectable tumors (median cancer-specific survival: 14.4 vs. 24.4 months and median recurrence-free survival: 12 vs. 16.5 months; p = 0.0038). Survival was correlated with the severity of venous involvement observed on preoperative CT scan (Nakao Types B, C, or D: median specific survival 26, 12, and 16 months, respectively). Post-operative chemotherapy had a positive impact on cancer-specific survival regardless of the type of venous extension (Nakao Type B: 26 vs. 13 months; Type C: 27 vs. 8.6 months, p < 0.0001; Type D: 20 vs. 9.6 months, p < 0.0052), but compliance to treatment at 3 and 6 months was lower in case of venous involvement (57 and 45% vs. 73 2. In case of SMA invasion, neoadjuvant therapy is recommended, followed by laparotomy with dissection and biopsy of peri-arterial tissues in case of tumor stability or tumor response. If the frozen section examination is positive, a PD with arterial resection is not recommended (Grade C). 3. In case of distal PDAC with invasion of the celiac axis, neoadjuvant therapy is recommended. In case of stabilization or tumor response, a distal pancreatectomy with celiac axis resection without arterial reconstruction may be proposed after radiologic embolization of the CA branches (expert opinion). (LE 3). A study in Japan Since this meta-analysis, several retrospective monocentric studies including small numbers of patients have reported AR results (mainly common HA or celiac axis, more rarely SMA) (44, 117-119) (LE 4). A recent systematic review (115) (2000-2016) of 13 studies including 70 patients undergoing pancreatectomy with SMA resection, which is rarely performed (out of 10,726 undergoing pancreatectomy) concluded that there was no evidence to support SMA resection. Indeed, in the 25 patients with available individual patient-level outcome data, perioperative morbidity ranged from 39 to 91%, the mortality rate was 25% and median survival was only 11 months (LE 4). However, the increasing use of neoadjuvant therapy protocols has increased the pool of selected patients who are candidates for "secondary" resection despite an initial suspected arterial invasion (116). On the other hand, the Heidelberg group reported that they described as a more conservative approach to the major arterial axes, in particular the SMA: -a first study published in 2016 (122) (123) described radical tumor removal by sharp dissection along the CA and the SMA with complete dissection of all soft tissue between both arteries and superior mesenteric/portal vein ("TRIANGLE operation"). In case of positive frozen section(s) of the arterial sheaths, "nonresection" and palliative treatment were indicated. This study included a consecutive series of 15 patients. The R0 resection rate (1 mm) was 40% (6/15) in patients who had pancreatectomy with "arterial sparing" resection (LE 4). 2. Three additional situations can be distinguished and in each of these 3, AR must be planned: 2.1 First, anatomical variants of HA: "Right" HA arising from the SMA during a planned PD for a "clearly" resectable tumor: a) The HA may be "accessory": recent data suggest that preoperative embolization by interventional radiology followed several days later by "en bloc" resection may be performed with no significant risks of liver/biliary ischemia due to development of intrahepatic arterial shunts (124) (LE 4). A systematic review has shown the feasibility and lack of morbidity and mortality of this strategy (125) (LE 4). This strategy avoids opening the accessory tumor/HA interface, which can be exposed with the risk of tumor spillage in case of RHA preservation (126) (LE 4). b) The HA can perfuse the "total liver" (Michell type 9: 1-5%): in this rare setting, it requires reconstruction of any type [direct anastomosis, by an interposed "reversed" saphenous vein graft (44,53,117) (LE 4) or reversal of the splenic artery (127) (LE 4)] to ensure vascularization |
of the biliary tree and the hepatico-jejunostomy following PD. HA should be reconstructed before continuing pancreatic resection to avoid any liver ischemia, particularly when an associated venous resection is needed. This procedure is contraindicated when the CA is invaded at its origin on the aorta or if the GDA is invaded (132-134) (LE 4). Indeed, DP-CAR requires a tumor-free and patent GDA to ensure "reverse flow" vascularization of the liver and bile ducts from the SMA through the PDA and GDA (12) (LE 2) and, for many authors, the use of preoperative occlusion of CHA or, at the best, the 3 branches of the CA to favor development of arterial collaterals thus reducing the risk of bile ducts and gastric ischemia (135,136) (LE 4). Embolization should be performed 1-2 weeks before resection (137,138) (LE 4). This procedure, which avoids any arterial reconstruction, remains controversial. Some authors prefer reconstruction in case of insufficient flow during an intraoperative "Doppler" control (139,140) (LE 4). Embolization is not effective in the case of "total liver" HA arising from the SMA, which requires reconstruction (137,138) (LE 4). Monocentric studies: Several monocentric, mainly Japanese, studies with small groups of patients, have been published in the past 10 years and were included in two recent systematic reviews, with reported median survivals ranging from 10 to 26 months and 5-year survival rates of 20% (97) (LE 4). Another recent retrospective study (141) compared the outcomes of patients receiving (n = 11) or not (n = 9) various regimens of neoadjuvant chemotherapy (mainly GEM-nab-PTX). Despite the small number of patients, those who received neoadjuvant chemotherapy had significantly less arterial invasion (p = 0.025), lymphatic invasion (p < 0.0001), and vascular invasion (p = 0.035) with significantly higher recurrence-free and overall survival rates. -A retrospective study in Japan (142) (144). Twenty patients who underwent DP-CAR ("modified" Appleby; 2.4%) recruited in 16 centers (obviously a limitation for this study) were compared to 172 patients who underwent DP who were matched for age, sex, BMI, albumin blood level, ASA score, pancreatic consistency, main pancreatic duct diameter, and pathology (60% PDAC). The procedure was longer for DP-CAR (median 276 vs. 207 min; p < 0.01) and the rates of postoperative acute renal failure (10 vs. 1%; p < 0.03) and 30-day mortality ( ARs are therefore very rarely indicated, often associated with venous resection (24, 97, 120, 121) (LE 3), and must be "planned" since they require routine neoadjuvant treatment and frequent preoperative arterial embolization. Surgery should always begin with an "artery first" approach (34,35) to accurately evaluate any persistent arterial involvement confirmed by frozen section examination (34,35,116). Forty (121) to 70% (24, 119) of patients do not receive postoperative chemotherapy (primarily single agent gemcitabine) due to postoperative morbidity and prolonged recovery time and there are few data on patient quality of life. Interestingly, a recent study from the MD Anderson Cancer Center (149) including 127 patients who received neoadjuvant treatment before PD (including vascular resection in 58 (46%); VR = 44, AR = 3, both = 11) reported that: (a) all patients experienced at least a transient skeletal muscle, visceral fat and subcutaneous fat loss; but (b) a relative increase in skeletal muscle (HR = 0.50) and albumin (HR = 0.57) during the first postoperative 12-months were associated with improved overall survival. This suggests that persistent postoperative skeletal muscle loss may represent an early marker of poorer outcomes. CONCLUSIONS PD with venous resection improves survival compared to no resection, especially with R0 resection. Mortality and morbidity are higher in PD with venous resection than in PD without vascular resection. PD with upfront venous resection has a poorer oncological results (increased risk of R1 resection, poorer survival) than PD with venous resection after neoadjuvant treatment. PD with arterial resection is associated with increased morbidity and mortality (compared to PD with venous resection) and has not been shown to be beneficial. A distal splenopancreatectomy with celiac axis resection is associated with increased morbidity and mortality and the oncological benefit of this approach has not been clearly demonstrated. Today, literature provides more support for neoadjuvant therapy in the management of pancreatic cancer. Waiting for RCTs results including clearly resectable tumors, neoadjuvant therapy and a complete R0 resection in all patients who require planned vascular resection with (or without) reconstruction should be the goal. Such patients should be treated by an experienced team in both preoperative/neoadjuvant therapy and vascular resection at the time of pancreatic resection. Such expertise is not available at every centers, which makes another strong case for the regionalization of complex cancer care that involves multiple treatments. AUTHOR CONTRIBUTIONS JD and AS contributed equally to the design and implementation of the research, to the analysis of the results, and to the writing of the manuscript. HOMA-IR Values are Associated With Glycemic Control in Japanese Subjects Without Diabetes or Obesity: The KOBE Study Background: Several studies have reported that insulin resistance was a major risk factor for the onset of type 2 diabetes mellitus in individuals without diabetes or obesity. We aimed to clarify the association between insulin resistance and glycemic control in Japanese subjects without diabetes or obesity. Methods: We conducted a community-based cross-sectional study including 1083 healthy subjects (323 men and 760 women) in an urban area. We performed multivariate regression analyses to estimate the association between the homeostasis model assessment of insulin resistance (HOMA-IR) values and markers of glycemic control, including glycated haemoglobin (HbA1c), 1,5-anhydroglucitol (1,5-AG), and fasting plasma glucose (FPG) levels, after adjustment for potential confounders. Results: Compared with the lowest tertile of HOMA-IR values, the highest tertile was significantly associated with HbA1c and FPG levels after adjustment for potential confounders, both in men (HbA1c: β = 1.83, P = 0.001; FPG: β = 0.49, P < 0.001) and women (HbA1c: β = 0.82, P = 0.008; FPG: β = 0.39, P < 0.001). The highest tertile of HOMA-IR values was inversely associated with 1,5-AG levels compared with the lowest tertile (β = −18.42, P = 0.009) only in men. Conclusions: HOMA-IR values were associated with markers of glycemic control in Japanese subjects without diabetes or obesity. Insulin resistance may influence glycemic control even in a lean, non-diabetic Asian population. INTRODUCTION Insulin resistance is a clinical condition characterized by a decreased sensitivity to insulin in peripheral tissues and is strongly associated with metabolic diseases, such as type 2 diabetes mellitus and obesity. [1][2][3] Prospective cohort studies in subjects without diabetes have also revealed that increased insulin resistance worsened glycemic control and contributed to the development of type 2 diabetes mellitus. [4][5][6] However, in all previous reports, the average body mass index (BMI) of the subjects was high (28-33 kg/m 2 ), and >50% of the subjects were obese. [4][5][6] Thus, it is unclear whether insulin resistance affects glycemic control in subjects without obesity or diabetes. For assessing glycemic control, temporal variations in the indicative parameters are more important than values obtained at a single point in time. Glycated hemoglobin (HbA1c) and 1,5-anhydroglucitol (1,5-AG) levels are generally used to evaluate glycemic control in clinical practice. HbA1c levels are the gold standard marker of glycemic control in patients with diabetes, and they reflect average plasma glucose levels during the past 2-3 months. 7 In contrast, 1,5-AG levels are used as an index that reflects glycemic control during the past few days or weeks and glycemic control fluctuations. 8,9 Thus, it is necessary to use several indices in various time periods to evaluate glucose metabolism. However, to the best of our knowledge, no earlier studies have investigated the association between insulin resistance and glucose metabolism using multiple markers of glycemic control. Therefore, in the present study, we aimed to investigate the impact of insulin resistance on glucose metabolism of Japanese subjects without diabetes or obesity. We used three markers that are commonly used to evaluate glucose metabolism in Japanese populations: HbA1c, 1,5-AG, and fasting plasma glucose (FPG) levels. METHODS Subjects We used data from the baseline survey in the Kobe Orthopedic and Biomedical Epidemiological (KOBE) study. The KOBE study is a population-based prospective cohort study of risk factors for cardiovascular disease or worsening of quality of life in Kobe City, a major urban area in Japan, that has been ongoing since 2010. The KOBE study has been described in detail elsewhere. 10 The present study was approved by the Ethics Committee of the Institute of Biomedical Research and Innovation (Committee approval number: [11][12]. Written informed consent was obtained from all participants. A total of 1118 subjects (342 men and 776 women) participated in the baseline survey from July 2010 to December 2011. None of the participants had past history of cardiovascular disease or cancer, and none were under therapy with medications for hypertension, dyslipidemia, or diabetes at the time of the survey. We excluded 34 participants who were diagnosed with diabetes or obesity on the basis of FPG level of ≥7.0 mmol/L (n = 8) and/or HbA1c level of ≥6.5% (n = 22) or BMI of ≥30 kg/m 2 (n = 4) at baseline. A participant with missing data (n = 1) was also excluded. We ultimately analysed data of 1083 subjects (323 men and 760 women) without diabetes or obesity in this study. Measurements Each subject completed a self-reported questionnaire to assess past medical history and lifestyle factors, such as smoking status, alcohol consumption, and regular exercise habits, and trained researchers directly confirmed the responses to the questionnaire. Waist circumference was measured at the level of the umbilicus in a standing position. Height and body weight were measured with patients wearing socks and light clothing, and BMI was calculated by dividing weight in kilograms by the squared height in meters. Fasting blood samples were drawn from all participants after they had fasted for at least 10 hours. Blood samples were transported to a single commissioned clinical laboratory centre (SRL Inc., Tokyo, Japan) for measurements. Plasma glucose levels (mmol/L) were determined using the glucose oxidase method. 1,5-AG levels were measured using an enzymatic method. HbA1c levels were measured using high-performance liquid chromatography and were expressed as National Glycohemoglobin Standardization Program units and International Federation of Clinical Chemistry and Laboratory Medicine values for the current analysis. 11 Serum immunoreactive insulin (IRI) levels (pmol/L) were determined using the chemiluminescence enzyme immunoassay (CLEIA) method, and homeostasis model assessmentinsulin resistance (HOMA-IR) values were calculated using the following formula: HOMA-IR = IRI × glucose/22.5. 12 Estimated glomerular filtration rate (eGFR) was calculated using the following formula: eGFR (mL/min per 1.73 m 2 ) = 194 × creatinine −1.094 × age −0.287 (× 0.739 if female), 13 and chronic kidney disease (CKD) defined as eGFR of <60 mL/min per 1.73 m 2 . High-molecular-weight adiponectin (HMW-adiponectin) levels were measured using the CLEIA method. Statistical analysis Gender-specific analyses were performed in light of observed gender differences in HOMA-IR distribution. HOMA-IR values were divided into tertiles to compare the characteristics. Data were presented as means (standard deviations [SDs]) or medians (interquartile ranges) for continuous variables, or numbers (percentages) for categorical variables. We used one-way analysis of variance for continuous variables and the chi-square test or Fisher's exact test for categorical variables to compare the characteristics among the groups. Multiple adjustments were performed with linear regression models to estimate the association between HOMA-IR values and markers of glycemic control, such as FPG, 1,5-AG, and HbA1c levels. We also performed multivariate logistic regression analysis to clarify the association between HOMA-IR values and any of the higher percentiles (80th or 90th percentile) of HbA1c levels, lower percentiles (10th or 20th percentile) of 1,5-AG levels or higher percentiles (80th or 90th percentile) of FPG levels. Multivariable analyses were adjusted for potential confounders in the following steps: (1) age; (2) BMI, regular exercise habits, current smoking, current alcohol drinking, CKD, and HMW-adiponectin levels, in addition to the variables in step 1; and (3) waist circumference substituted for BMI in step 2. The adjusted coefficient of determination (adjusted R 2 ) was also calculated. Two-tailed P values of <0.05 were considered statistically significant. All analyses were performed using STATA SE 11 data analysis and statistical software (Stata Corp LP, College Station, TX, USA). Table 1 shows the characteristics of the participants according to HOMA-IR category by gender. The mean (SD) age was 60.8 (9.0) and 58.0 (8.7) years in men and women, respectively. Participants in |
higher HOMA-IR categories had higher HbA1c and FPG levels, both in men and women, and only men had lower 1,5-AG levels. Participants in higher HOMA-IR categories also had higher BMI and waist circumference, as well as lower HMW-adiponectin levels, both in men and women. Association between HOMA-IR values and markers of glycemic control The association between HOMA-IR values and markers of glycemic control, such as HbA1c, 1,5-AG, and FPG levels, in the multivariate linear regression analysis are shown according to gender in Table 2 (men) and Table 3 (women). HbA1c and FPG levels were significantly higher in the highest tertile group of HOMA-IR values than in the lowest tertile group, both in men and women. 1,5-AG levels were significantly lower in the highest tertile group of HOMA-IR values than in the lowest tertile group in men but not in women. The association between HOMA-IR values and markers of glycemic control was unchanged after adjusting for potential confounders, including BMI, waist circumference, and HMW-adiponectin levels. We performed multiple linear regression analysis to estimate the association between HOMA-IR values and markers of glycemic control according to BMI and gender. The results showed that the absolute values of coefficient were larger in the group with high BMI than in the group with low BMI in men (eTable 1), which suggested a strong association between HOMA-IR and these glycemic control parameters; however, these findings were not clearly observed in women (eTable 2). We also performed multivariate logistic regression analysis to clarify the association between HOMA-IR values and any of the higher percentiles of HbA1c levels, lower percentiles of 1,5-AG levels, or higher percentiles of FPG levels, both in men (eTable 3) and women (eTable 4). Compared with the lowest tertile group of HOMA-IR values, the highest tertile group had significantly higher odds ratios for any of higher percentiles of HbA1c levels, lower percentiles of 1,5-AG levels, or higher percentiles of FPG levels, both in men and women, after adjusting for potential confounders. DISCUSSION This is the first report, to the best of our knowledge, to assess the relationship between HOMA-IR values and several indices of glucose metabolism, obtained at various time points, in Japanese subjects without diabetes or obesity. As a result, we found that HOMA-IR values were significantly associated with all indices of glucose metabolism in men, as well as with HbA1c or FPG levels in women. HOMA-IR is generally considered an index of insulin resistance in the liver. 12 Insulin suppresses the elevation of plasma glucose levels by promoting glucose uptake into cells and by inhibiting glucose release from the liver. However, when insulin resistance increases, the regulatory mechanism fails and blood glucose levels remain elevated. [14][15][16] In the present study, our results indicate that increased insulin resistance further deteriorates glucose metabolism in patients with not only type 2 diabetes mellitus and obesity but also in those without diabetes or obesity. The relationship between insulin resistance and several indices of glucose metabolism was maintained after adjusting for confounding factors, such as BMI or waist circumference. Multivariate regression analysis in this study revealed that BMI and waist circumference were not correlated with the indices of glucose metabolism. However, in the multivariate regression model that excluded HOMA-IR as an independent variable, BMI (eTable 5) and waist circumference (eTable 6) maintained significant correlation with HbA1c and FPG levels. Therefore, these results indicated that insulin resistance might regulate glucose metabolism downstream of BMI and waist circumference. The present study showed that HOMA-IR values were not significantly associated with 1,5-AG levels in women. 1,5-AG is monosaccharide excreted in the urine. Approximately, 99%-100% of the excreted 1,5-AG is reabsorbed in the renal tubules, and a constant level is maintained in subjects with normal glucose tolerance. When blood glucose levels reach the threshold at which urinary glucose appears, 1,5-AG levels decrease remarkably because 1,5-AG reabsorption is inhibited. 8,17,18 In other words, 1,5-AG levels do not change if blood glucose levels do not reach the urinary glucose excretion threshold. Considering these mechanisms, it is suspected that most of the participants, especially women, had normal glucose tolerance, although participants both with normal glucose tolerance and mild glucose intolerance were included in the present study. In male participants, HOMA-IR values were weakly correlated with 1,5-AG levels compared to the correlations of HOMA-IR values with HbA1c and FPG levels. Thus, it is possible that the suspected high prevalence of normal glucose tolerance influenced this result. A cohort study of the general Japanese population revealed that metabolic syndrome increased the risk of onset of type 2 diabetes mellitus, suggesting that insulin resistance contributed to the onset of type 2 diabetes mellitus in the Japanese population. 19 Another cohort study of the general Japanese population showed that both a decrease in insulin secretion ability and an increase in insulin resistance contributed to the onset of type 2 diabetes mellitus. 20 In the present study, a correlation between an index of insulin resistance and several indices of glucose metabolism was observed in Japanese subjects without obesity and with low insulin resistance. These findings suggest that insulin resistance mainly contributed to the onset of type 2 diabetes mellitus in Japanese subjects. Lifestyle interventions, such as healthy diet and regular exercise, have been shown to be effective for the improvement of insulin resistance but not impaired insulin secretion capacity. 21 Thus, we recommend lifestyle interventions for the prevention of onset of type 2 diabetes mellitus not only in obese subjects but also in nonobese subjects. In the present study, we also found that the association between HOMA-IR values and each marker of glycemic control was stronger in subjects with high BMI than in those with low BMI. Therefore, even in non-obese (BMI <30 kg/m 2 ) Asians, we suggest that the impact of lifestyle intervention on the onset of type 2 diabetes mellitus is larger among subjects with high BMI than among those with low BMI when impaired insulin secretion capacity is suspected. This study has several limitations. At first, we used HOMA-IR, which is an indirect index for the evaluation of insulin resistance. Although the glucose clamp technique is necessary for direct evaluation, 22 we were unable to apply this test in our subjects. However, a previous study reported that the use of HOMA-IR was appropriate to assess insulin sensitivity in subjects without diabetes. 23 Second, we used a self-reported questionnaire in the present study; thus, recall bias might have affected the evaluation of physical activity. Finally, we could not evaluate postprandial hyperglycemia accurately in this study because we did not measure the blood glucose afterload. We used HbA1c levels as the diagnostic criteria of diabetes in this study; thus, subjects having marked postprandial hyperglycemia were excluded. In conclusion, the present study showed that HOMA-IR values were significantly associated with several indices of glucose metabolism in Japanese subjects without diabetes or obesity and that insulin resistance prescribed glucose metabolism downstream of BMI or waist circumference. These findings suggest that insulin resistance may mainly influence glycemic control even in non-diabetic subjects without obesity. ONLINE ONLY MATERIALS eTable 1. Associations between HOMA-IR values and markers of glycemic control divided by median BMI in men (n = 323). eTable 2. Associations between HOMA-IR values and markers of glycemic control divided by median BMI in women (n = 760). eTable 3. Associations between HOMA-IR values and higher percentile of HbA1c or FPG or lower percentile of 1,5-AG. eTable 4. Associations between HOMA-IR values and higher percentile of HbA1c or FPG or lower percentile of 1,5-AG. eTable 5. Associations between BMI and markers of glycemic control in multivariate regression model that excluded HOMA-IR. eTable 6. Associations between waist circumference and markers of glycemic control in multivariate regression model that excluded HOMA-IR. Abstract in Japanese. Interaction of Eye Protein Kinase C and INAD inDrosophila Drosophila eye-specific protein kinase C (eye-PKC) is involved in light adaptation and deactivation. eye-PKC, NORPA (phospholipase Cβ), and transient-receptor-potential (TRP) (calcium channel) are integral components of a signal transduction complex organized by INAD, a protein containing five PDZ domains. We previously demonstrated the direct association between the third PDZ domain of INAD with TRP in addition to the carboxyl-terminal half of INAD with the last three residues of NORPA. In this work, the molecular interaction between eye-PKC and INAD is defined via the yeast two-hybrid and ligand overlay assays. We show that the second PDZ domain of INAD interacts with the last three residues in the carboxyl-terminal tail of eye-PKC, Thr-Ile-Ile. The association between eye-PKC and INAD is disrupted by an amino acid substitution (Ile-700 to Asp) at the final residue of eye-PKC. In flies lacking endogenous eye-PKC (inaC p215), normal visual physiology is restored upon expression of wild-type eye-PKC, whereas the eye-PKCI700D mutant is completely inactive. Flies homozygous for inaC p209 andInaD p215, a mutation that causes a loss of the INAD-TRP association, were generated. These double mutants display a more severe response inactivation than either of the single mutants. Based on these findings, we conclude that the in vivoactivity of eye-PKC depends on its association with INAD and that the sensitivity of photoreceptors is cooperatively regulated by the presence of both eye-PKC and TRP in the signaling complex. In Drosophila, visual transduction is a G-protein-coupled phospholipase C-mediated process that leads to depolarization of photoreceptors (see Refs. 1-3 for reviews). In this signaling pathway, rhodopsins are activated by light and catalyze the GDP/GTP exchange in the heterotrimeric G-protein, Gq. The ␣-subunit of Gq interacts with the phospholipase C, NORPA, that catalyzes the hydrolysis of phospholipids to generate inositol 1,4,5-trisphosphate and 1,2-diacylglycerol (4). After the activation of NORPA, two cation channels, TRP 1 and TRPL (TRP-like), are opened, leading to membrane depolarization (5). The mechanisms by which these two channels are gated remain elusive. One hypothesis is that the TRP calcium channel is controlled by depletion of internal calcium stores initiated by inositol 1,4,5-trisphosphate binding to its receptor (6), similar to the store-operated channel present in nonexcitable cells (7). However, a recent report suggests that an inositol 1,4,5-trisphosphate receptor is not involved in visual signaling (8). In addition calcium release from internal stores has not been directly implicated in visual excitation (9,10). Despite the uncertain role of inositol 1,4,5-trisphosphate in photoreceptors, diacylglycerol generated by NORPA is a potent activator of protein kinase C (PKC). eye-PKC encoded by the inaC locus is known to be involved in the negative feedback regulation of visual transduction in Drosophila photoreceptors (11,12). The importance of PKC as a regulator of the visual response is shown by the defective visual electrophysiology of the inaC mutants missing the eye-specific isoform of PKC. The inaC mutants display slow deactivation of the light response and are unable to adapt to different light intensities (11). In addition, inaC photoreceptors undergo light-dependent retinal degeneration (12). eye-PKC has been shown to be a component of a signal transduction complex by immunoprecipitation and affinity chromatography assays (13)(14)(15)(16). This complex is organized by an adaptor protein, INAD, that also interacts with two key components of the visual cascade, TRP and NORPA (13)(14)(15)(16)(17)(18). The formation of this signaling complex is crucial for a functional visual signaling process, as flies lacking INAD show mis-localization of eye-PKC, TRP, and NORPA and display a drastically reduced response to light (16). INAD appears to act as a scaffold protein that assembles with these signaling proteins into a multiprotein complex. Formation of this macromolecular complex may be essential for the fast kinetics of signal transduction in Drosophila photoreceptors (14,16). Since the gating mechanism of TRP is not known, investigations on the protein-protein interactions in this complex may shed light on events leading to depolarization. We previously reported a point mutation in INAD that eliminates the TRP interaction and brings about a slow recovery of the visual response (17,19). Thus the INAD/TRP interaction appears to regulate the TRP calcium channel (16,17). We also investigated the functional role of the NORPA/INAD association by generating transgenic flies expressing a modified NORPA that lacks the INAD interaction. The norpA transgenic flies display a unique electroretinogram recordings (ERG) characterized by delayed activation and slow deactivation. This phenotype of slow kinetics is also observed in InaD 2 that contains a missense mutation in the fifth PDZ domain leading to a loss of the NORPA association (16). In this report, we define the respective binding domains in * This work was supported by National |
Institutes of Health Grant EY09743. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ An Established Investigator of the American Heart Association. To whom correspondence should be addressed: Dept. of Pharmacology, 402 Medical Research Bldg I., Vanderbilt University, Nashville, TN 37232-6600. E-mail: [email protected]. Tel.: 615-343-0441; Fax: 615-343-6532. 1 The abbreviations used are: TRP, transient receptor potential; PKC, protein kinase C; NORPA (norpA), no-receptor-potential; INAD (InaD), inactivation-no-afterpotential D; INAC (inaC), inactivation-no-afterpotential C; ERG, electroretinograms; PCR, polymerase chain reaction. eye-PKC and INAD and demonstrate the functional consequence of this interaction in vivo. Our findings suggest the essential role of the eye-PKC and INAD association in regulating visual transduction. Potential substrates for eye-PKC that are involved in deactivation and desensitization of the visual response are discussed. Molecular Biological Techniques Standard molecular biology methods were used in subcloning, polymerase chain reaction (PCR), and nucleotide sequencing according to published procedures (20). DNA sequencing was performed using the Sequenase 2.0 kit (Amersham Pharmacia Biotech). Site-directed mutagenesis was carried out by the PCR-based overlap extension method. The Yeast 2-Hybrid System The yeast two hybrid system used in this work is detailed in Golemis et al. (21). The yeast strain used was EGY191 (MATa, trp1, his3, ura3, 2ops-LEU2); the LacZ reporter plasmid was pSH18 -34; the LexA fusion plasmid was pEG202, and the activation domain fusion plasmid was pJG4 -5. Methods for yeast transformation, testing protein-protein interactions, and checking protein expression were from Golemis et al. (21). Antibodies used to assay the level of fusion protein expression were polyclonal rabbit anti-HA (Zymed Laboratories Inc.) and anti-LexA (Invitrogen). eye-PKC Fusion Protein Constructs PKC Full-length-Primers flanking codons 1-381 of inaC were used to amplify 5Ј sequence by PCR. The primers included 5Ј BglII/EcoRI restriction sites and 3Ј BamHI/XhoI restriction sites. BglII-XhoI restriction fragment was cloned into BamHI-XhoI sites of pGEX-4T-1 (Amersham Pharmacia Biotech). Then the NsiI-XhoI restriction fragment from pPKC (pPKC is a full-length inaC generated by RT-PCR) was cloned in to recreate inaC in pGEX-4T-1. The EcoRI-BamHI (blunt, vector-encoded) restriction fragment from pPKC was cloned into EcoRI-XhoI(blunt) sites of pJG4 -5 before the EcoRI-EcoRI restriction fragment from the pGEX-4T-1 PKC clone was ligated into the EcoRI site of this construct. PKC 675-700 -For correct framing, the HindIII-XhoI restriction fragment from pPKC was cloned into HindIII-XhoI sites of pBluescript II SK(Ϫ), and the resulting EcoRI-XhoI restriction fragment was subcloned into pEG202. 20 Amino Acids (10 Amino Acids)-The last 20 (10) codons were cloned by annealing and ligating complementary oligonucleotides into the BamHI-XhoI sites of pEG202. Amino acid substitutions are detailed in Table II. 4 Amino Acids and 3 Amino Acids-The final 4 and 3 codons of eye-PKC were cloned into the BamHI-XhoI sites of pEG202 using complementary oligonucleotides. PDZ4A-A PCR product encompassing codons 485 to 577 of InaD with 5Ј EcoRI and 3Ј XhoI sites was subcloned into pJG4 -5. PCR fragments and sequences across the junction of fused proteins were verified by DNA sequencing. Unless otherwise indicated, duplicate constructs were isolated and tested in triplicate. Western Blotting, Ligand Overlay Assay, and Overexpression of Fusion Proteins in Bacteria These three procedures were carried out as described previously (17). Genetic Crosses Fly stocks were maintained at 25°C in a 12-h dark/12-h light cycle. Recombination between marked InaD p215 (InaD, sp) and inaC p209 (inaC, px, sp) chromosomes was performed using standard techniques. If recombination occurs, double mutants will have spotted (sp) marker, but not plexus (px). Potential double mutants were further evaluated for the presence of the inaC p209 and InaD p215 mutations by biochemical methods. P-element-mediated Germ-line Transformation-The mutant and wild-type eye-PKC cDNA were subcloned into a modified pCaSpeR 4 vector (24) that contains the Drosophila hsp70 promoter. The P-element construct and a transposase plasmid were injected into the inaC p209 embryos to generate transgenic flies (25). Flies with the transgene integrated into the second or third chromosome were selected and made homozygote for further analysis. ERG ERG recordings were carried out as described (19). Experimentally, flies were anesthetized by carbon dioxide and immobilized. Glass electrodes were filled with physiological saline (0.7% NaCl). Light stimulation was delivered by a fiber optic light source (Oriel). Signals were amplified by means of a WPI Dam 60 preamplifier (WPI) and digitized using Superscope II (GW Instruments). RESULTS Use of the Yeast Two-hybrid System to Investigate the Association between eye-PKC and INAD-We employed the yeast two-hybrid method using the LexA system (21), with eye-PKC (Fig. 1A) and INAD fused with the LexA DNA binding domain and a transcription activation domain, respectively. A positive interaction between the two proteins will activate transcription of two reporter genes. Leu2 activation results in growth of yeast on plates lacking leucine, and LacZ gives colonies with a blue color on plates containing 5-bromo-4-chloro-3-indolyl -D-galactopyranoside (Xgal). We used the expression of LacZ as a means for quantitation. As a negative control, SNF1, a yeast serine/threonine protein kinase fusion (23), was shown not to interact with INAD (Fig. 1B). The amino and carboxyl terminus domains of eye-PKC contain amino acid sequences unique to this PKC (12) and may be the basis for the interaction with INAD. We initially found that INAD interacts with both full-length eye-PKC and PKC-18 (which contains residues 18 -700) as shown by LacZ reporter activity (Fig. 1B). To assess the role of the amino terminus in interaction, we tested an amino terminus fragment (PKC-NH 2 , Fig. 1A) and found that it did not interact with INAD (Fig. 1B). Although the carboxyl terminus (PKC 562-700 , Fig. 1A) was the obvious determinant for association (Fig. 1B), we attempted to identify the scope of interaction by deleting eye-PKC to within 70 (represented as PKC 631-700 ), 26 (PKC 675-700 ), 20, 10, 4, and finally 3 amino acid residues of the carboxyl terminus without a loss of association with INAD (Fig. 1). The last three residues in the carboxyl-terminal tail of eye-PKC, Thr-Ile-Ile (TII), fit the general motif for a PDZ domain interaction. PDZ domains typically interact with the carboxylterminal tail of proteins that terminate (0 position) with a hydrophobic residue (such as Val, Ile, or an aromatic residue, Phe) and have Ser or Thr or Asp at the Ϫ2 position (26). Amino acid substitutions were made to determine the critical residues required for the association. The final amino acid residue, Ile, when changed to Val, slightly decreases the interaction with INAD when tested in a protein fusion that contains the last 10 amino acid residues of eye-PKC (Fig. 1B). In contrast, the Asp or Asn substitution markedly reduces the association (Fig. 1B). We also examined the Asp substitution in the context of PKC-18 (PKC-18 TID). PKC-18 TID fails to associate with full-length INAD (Fig. 1B) nor does it interact with the second PDZ domain of INAD (results not shown). Mapping the eye-PKC-interacting Sequence in INAD-INAD was first predicted to contain two PDZ domains (19), but further analysis suggests that there are five PDZ domains in the protein (16,27,28). These PDZ domains are named sequentially PDZ1 to PDZ5 in the text. The PDZ domains of INAD in Drosophila and Calliphora, the blowfly, (13) are aligned in Fig. 2A to show the similarity of each individual PDZ domain from these two dipteran species. Although these five regions share the general characteristics of PDZ domains, the individual sequences (PDZ1 to PDZ5) are distinct and divergent. This is consistent with the notion that they interact with different proteins. To map the region of INAD that interacts with eye-PKC, a series of INAD constructs were fused to the transcription activation domain and tested against PKC-18. These INAD fusion proteins are named according to the PDZ domains they encompass (Fig. 2B). We show strong association between PKC-18 with PDZ1-2 and PDZ2 (Fig. 2C). We also observed a weak yet detectable interaction between PKC-18 and PDZ1. However, no eye-PKC association with either PDZ2-5 or PDZ2-3 was detected (Fig. 2C). These results, observed consistently in several independent experiments, indicate that eye-PKC interacts with PDZ1 and PDZ2, with PDZ2 showing a stronger interaction (See Fig. 3C). To account for the lack of association between eye-PKC and PDZ2-3 or PDZ2-5, which also contain PDZ2, we propose that PDZ3 may interfere with PDZ2 binding to eye-PKC. Consistent with this hypothesis, is the observation that a weaker interaction between full-length INAD and eye-PKC was detected. A recent report suggests that eye-PKC interacts with PDZ4 (16). We therefore paid special attention to this region of the protein. Three additional constructs that encompass PDZ4 (Fig. 2B) were tested, including PDZ4A, which contains exactly the same region of INAD as was reported to interact with eye-PKC (16). All of these PDZ4 fusion proteins fail to associate with eye-PKC. The expression level of these fusion proteins is similar to or higher than that of PDZ1-2 and should be sufficient to allow detection of an interaction with PKC-18 (Fig. 2C). Mutational Analysis of PDZ1 and PDZ2 Interaction with eye-PKC-To confirm that eye-PKC interacts with INAD via the PDZ domains, amino acid substitutions were introduced in the PDZ domain to modify the structure. It is known from the crystal structure of a peptide-PDZ domain complex that the PDZ domain binding to the amino-terminal hydrophobic residue is dependent on a "carboxylate binding loop" (29,30). Residues in this loop of PDZ1-PDZ5 are shown boxed in Fig. 2A. eye-PKC appears to show a weak interaction with PDZ1 ( Fig. 2C and 3C). We mutated PDZ1 by changing Lys-22 to Ser, Gly-29 to Glu, Ile-30 to His, and Ile-32 to Thr in PDZ1*. These nonconserved substitutions are likely to bring about a change in the structure of PDZ1. These mutations eliminated the interaction between PDZ1 and eye-PKC ( Fig. 2C and 3C). We analyzed a modified PDZ2 (PDZ2*) that contains the following substitutions: Arg-254 to Ser, Leu-260 to Gln, Gly-261 to Glu, and Leu-262 to His. We observed a marked reduction of the interaction between the modified PDZ2 and PKC-18 ( Fig. 2C and 3C). Taken together, our findings provide evidence for a strong interaction between the second PDZ domain of INAD and the last three residues in the carboxyl-terminal tail of eye-PKC. Further Evidence of the Involvement of PDZ2 in the PKC Interaction-To further confirm the results obtained above, we employed ligand overlay assay. This method was used successfully to define the interaction between the third PDZ domain of INAD and TRP (17). PDZ2, PDZ1-2, and PDZ4 were radiolabeled and probed in filters containing retinal extracts and bacterial extracts containing either an eye-PKC fusion protein or T7 gene 10, respectively. Fig. 4 shows that both PDZ2 and PDZ1-2 display strong and specific association with retinal eye-PKC and PKC fusion protein. In contrast, PDZ4 recognizes neither retinal PKC nor PKC fusion. Interestingly, in PDZ1-2, the presence of PDZ1 somehow potentiates the interaction of PDZ2 with its targets, whereas PDZ1 alone only weakly associates with PKC (data not shown). These findings agree with those obtained by the yeast two-hybrid assay, providing further evidence that PDZ2 interacts with eye-PKC. Electrophysiological Analysis of Transgenic Flies Expressing eye-PKC I700D in the inaC p209 Genetic Background-To gain insight into the functional significance of the eye-PKC/INAD association in vivo, we generated transgenic flies lacking the interaction and examined the visual electrophysiology. We investigated a point mutation in eye-PKC, eye-PKC I700D , that eliminates the interaction of the kinase with INAD (Fig. 1B) without the modifications in those domains required for the predicted kinase activity (12). Transgenic flies expressing eye-PKC I700D under the control of a Drosophila heat-inducible promoter (hsp70) were generated in an eye-PKC null (inaC p209 ) (12) genetic background. After heat shock treatment, the expression of eye-PKC I700D was comparable with that of eye-PKC in wild-type flies (Fig. 5A). The function of eye-PKC I700D was analyzed by ERG (Fig. 5B), showing that eye-PKC I700D displays a phenotype indistinguishable from that of inaC p209 (Fig. 5B). This abnormal phenotype of inaC p209 is characterized by a slowly decaying receptor potential during the light pulse (desensitization) and delayed deactivation upon cessation of light stimulation (Fig. 5B). By contrast, transgenic flies expressing wild-type eye-PKC rescue the defective visual physiology of inaC p209 (Fig. 5B). We conclude that disruption of the eye-PKC-INAD interaction in eye-PKC I700D |
results in a complete loss of the eye-PKC activity in vivo. Electrophysiological Analysis of Double Mutants-InaD p215 flies express a mutant INAD protein that does not interact with TRP (17). To determine how the phenotype of inaC p209 flies is modified in the absence of the INAD/TRP interaction, we generated inaC p209 ,InaD p215 double mutants. The lack of eye-PKC was confirmed by Western blotting, and the presence of mutant INAD was confirmed by a loss of TRP binding in an overlay assay (Fig. 6). ERG recordings of the inaC p209 ,InaD p215 double mutant were compared with those of wild-type and single mutant flies. In the experimental paradigm used, flies were dark-adapted for 90 s then given two 5-s pulses of the orange light with a 6-s interval. As shown in Fig. 7, wild-type flies respond to light with fast kinetics and show desensitization during the light pulse. Importantly, the response to a subsequent stimulation of the same light intensity is not inactivated in wild-type flies, such that a second light pulse triggers a response almost the same as the initial one. Neither of the single mutant flies (inaC p209 or InaD p215 ) shows response inactivation under the experimental conditions (Fig. 7). In contrast, the double mutants display a marked reduction in this response to a second pulse of light. The peak amplitude of the second light response for wild-type, inaC p209 and InaD p215 flies was approximately 90% that of the first light response (wild-type, 91 Ϯ 6.2%; InaD, 90 Ϯ 5.0%; inaC, 95 Ϯ 5.1%, n ϭ 5). For the inaC p209 ,InaD p215 double mutant, the peak amplitude of the second response was greatly reduced (44% Ϯ 7.5%, n ϭ 10) (Fig. 7). Double mutants also show abnormal deactivation similar to inaC. We measured the rate of deactivation as the time required for the light response to decay to 50% upon cessation of the light pulse. For wild-type and InaD p215 flies, deactivation kinetics were very rapid (wild-type, 27.4 Ϯ 3.0 ms; InaD, 29.1 Ϯ 2.7 ms, n ϭ 10). For the inaC p209 ,InaD p215 double mutant, the rate of deactivation was at least 80-fold slower than wild type and is similar to that of inaC flies (inaC, 2372 Ϯ 640 ms; inaC, InaD, 2430 Ϯ 430 ms, n ϭ 10) (Fig. 7). DISCUSSION We used the yeast two-hybrid system to map the regions of eye-PKC and INAD that interact and to determine the type of interaction. We found that the carboxyl-terminal tail of eye-PKC associates with predominantly the second PDZ domain of INAD. A fusion protein containing only the last three carboxylterminal residues of eye-PKC, Thr-Ile-Ile, interacts with INAD. This terminal sequence is consistent with the tripeptide motif, (S/T/D)X(V/I/F) (X is any amino acid) that specifies the association with a PDZ domain (26). Substitution of the ultimate isoleucine with an aspartic acid or an asparagine disrupts the interaction with INAD. To gain insight into the in vivo function that the eye-PKC-INAD interaction specifies, we generated transgenic flies expressing eye-PKC I700D that will lack the INAD association. Significantly, these transgenic flies in the inaC p209 genetic background display an ERG phenotype of abnormal deactivation and desensitization similar to that of inaC p209 . This indicates that eye-PKC I700D fails to rescue the mutant phenotype. Thus, a lack of PKC interaction with INAD results in a complete loss of in vivo eye-PKC activity. This may be due to mis-localization of this enzyme, in turn rendering eye-PKC unable to phosphorylate specific substrates involved in regula-tion of the visual response. To investigate how eye-PKC and INAD regulate the visual response, we characterized the electrophysiology of a In addition to slow deactivation inaC, InaD double mutants also display the inactivation phenotype (reduced peak amplitude of the second light response). inaC p209 ,InaD p215 double mutant. We found that the double mutants display a more severe response inactivation, whereas inaC and InaD flies show no inactivation phenotype under the same experimental conditions. Response inactivation may result from depletion of limiting factors such as internal messengers and/or precursors, leading to exhaustion of the excitatory process (11). It is also likely that inactivation is due to defects in re-setting the proteins to the resting state such as conversion of metarhodopsin to rhodopsin (31,32). Our finding suggests that the presence of both eye-PKC and TRP in the complex cooperatively contributes to regulation of sensitivity of photoreceptors. One potential function of the signaling complex is to ensure eye-PKC and TRP are in close proximity for efficient regulation of the light response. A simple interpretation of the interdependence of eye-PKC and TRP is that eye-PKC regulates an intermediate, also associated with the signaling complex, that acts on TRP. Alternatively, eye-PKC may directly regulate TRP or regulate another molecule(s) via TRP. Such a model for eye-PKC regulation via the signaling complex has been proposed (14), and our analysis of the inaC,InaD double mutant provides the supporting evidence. Our finding that the response inactivation is more pronounced in the double mutant suggests that this regulatory step is partially activated in the absence of eye-PKC. It is possible that compensatory mechanisms differ during the development of the single and double mutants. It is also possible that this activity is constitutive or is modulated by another effector(s) in addition to eye-PKC. Candidate substrates for eye-PKC include TRP, as mentioned, NORPA, INAD, or another unidentified molecule(s) present in the signaling complex. There is no evidence that either NORPA or TRP are phosphorylated during the light response. Light-induced phosphorylation has been demonstrated in vitro for Calliphora INAD (13). Thus INAD is a possible substrate of eye-PKC. INAD contains no domains with an obvious catalytic function, and if phosphorylation of INAD is part of the eye-PKC regulatory process, it may achieve its effect by altering the protein binding properties of INAD. This could change the composition or function of molecules within the signaling complex. Our yeast two-hybrid and ligand overlay results both indicate that predominantly the second PDZ domain of INAD associates with eye-PKC, whereas no interaction was detected with PDZ4. This result is different from a previous report in which interaction of eye-PKC with the fourth PDZ domain of INAD was detected by affinity chromatography (16). We tested a total of five constructs that contained the fourth PDZ domain without any indication of this interaction. These included a fusion protein that contained exactly the same region as was tested before (16). One possible explanation for these conflicting results is that the different assay systems are measuring different types of association between eye-PKC and INAD. INAD may bind and cluster eye-PKC to the signaling complex, and it can also act as a substrate for the kinase activity (13). We showed that amino acid substitutions made in the second PDZ domain of INAD disrupted the eye-PKC binding. None of these amino acid changes were near serine or threonine residues that are putative PKC phosphorylation sites. Furthermore, mutations in the carboxyl-terminal tail of PKC abolish its binding to the second PDZ domain. Thus the interaction we describe for eye-PKC/INAD is a typical carboxyl-terminal tail/PDZ domain association. The basis of the reported interaction with the fourth PDZ domain remains to be determined. Another provocative explanation could be that eye-PKC may bind different PDZ domains of INAD during different physiological conditions. For example, phosphorylation of INAD may change the relative affinity of the interaction in PDZ2 and PDZ4. Clarification of the role of these two eye-PKC/INAD interactions will require analysis of transgenic flies expressing modified InaD in which these PDZ domain are mutated. Copolymer micelles function as pH-responsive nanocarriers to enhance the cytotoxicity of a HER2 aptamer in HER2-positive breast cancer cells Efficient delivery of nucleic acids into target cells is crucial for nucleic acid-based therapies. Various nucleic acid delivery systems have been developed, each with its own advantages and limitations. We previously developed a nanoparticle-based delivery system for small chemical drugs using pH-responsive PEG8-PDPA100-PEG8 polymer micelles as carriers. In this study, we extend the application of these pH-responsive micelle-like nanoparticles (MNPs) to deliver oligonucleotides. We demonstrate that the MNPs efficiently encapsulate and deliver oligonucleotides of different lengths (20–100 nt) into cells. The cargo oligonucleotides are rapidly released at pH 5.0. We prepared MNPs carrying a Texas red-fluorescently labeled anti-human epidermal growth factor receptor 2 (HER2) aptamer (HApt). Compared to free HApt, the HApt-MNPs resulted in significantly better cellular uptake, reduced cell viability, and increased apoptosis in SKBR3 breast cancer cells, which overexpress HER2. Moreover, HApt-MNPs were significantly less cytotoxic to MCF7 breast cancer cells, which express low levels of HER2. After cellular uptake, HApt-MNPs mainly accumulated in lysosomes; inhibition of lysosomal activity using bafilomycin A1 and LysoTracker Red staining confirmed that lysosomal activity and low pH were required for HApt-MNP accumulation and release. Furthermore, HER2 protein expression declined significantly following treatment with HApt-MNPs in SKBR3 cells, indicating that HApt-induced translocation of HER2 to lysosomes exerted a potent cytotoxic effect by altering signaling downstream of HER2. In conclusion, this pH-responsive and lysosome-targeting nanoparticle system can efficiently deliver oligonucleotides to specific target cells and has significant potential for nucleic acid-based cancer therapies. Introduction In recent decades, nucleic acid-based therapeutic agents have been developed and used in the clinic to treat multiple diseases, including cancer, 1-4 diabetes, 4-6 cardiovascular diseases, 7,8 and infectious diseases. [9][10][11] However, the large size, high anionic charge density, and hydrophilic properties of DNA and RNA molecules significantly limit their cellular uptake. [12][13][14][15] Hence, a plethora of nucleic acid carriers have been developed in an attempt to overcome these barriers. In general, nucleic acid delivery systems can be classified into two major groups: viral and nonviral. [16][17][18][19] Viral vectors, such as retrovirus or adenovirus-based systems, have high transduction efficiencies. However, their application is limited by high production costs and biosafety and immunogenicity concerns. 18,20,21 Nonviral systems, such as lipidor polymer-based approaches, have gained much attention due to their low cytotoxicity, weak or lack of immunogenicity, and ease of production. [22][23][24][25][26] Among the nonviral delivery systems, polymeric nanoparticles are an attractive option for gene therapy due to their unique properties including self-assembly behavior, ability to condense and protect nucleic acids, cell association, efficient cell transfection, and low cytotoxicity. 23,[25][26][27][28] Recently, the addition of stimuli-sensitive functions has enabled polymeric nanoparticles to specifically respond to pathological or externally applied "triggers" (eg, temperature, pH, enzymatic catalysis, and light or magnetic fields) and further extended their potential applications. [29][30][31][32] For example, pH-responsive polymeric nanoparticles capable of protecting nucleic acids in the blood circulation and actively releasing their cargo in the tumor microenvironment and/or inside the target tumor cells have received overwhelming interest in the context of cancer therapy. We previously developed a novel pH-responsive delivery system for small chemical anticancer drugs from amphiphilic triblock copolymers composed of poly [2-(diisopropylamino)ethyl methacrylate] (PDPA) and methoxy-poly(ethyleneglycol) segments (PEG 8 -PDPAn-PEG 8 ; n=30, 50, or 100). 29,33 The PEG shell of the micelle-like nanoparticles (MNPs) prolongs the circulation of MNPs in the blood by reducing nonspecific associations with plasma proteins and tissues. 34,35 The PDPA core serves as a pH sensor, which is hydrophobic at physiological pH (pH . 7.4) and becomes hydrophilic at acidic conditions due to protonation of di-isopropylamin. 36,37 Thus, the pH-responsive MNPs are stable at pH 7.4 but swell and disaggregate at a pH ,6.0. MNPs composed of PEG 8 -PDPA 100 -PEG 8 have been proven to possess a range of desirable properties, such as low cytotoxicity, superior efficiency to deliver small chemical drugs, and considerable lysosomal accumulation after cellular uptake. 29 Human epidermal growth factor receptor 2 (HER2), a member of the epidermal growth factor-related protein (ErbB) family of receptor tyrosine kinases, is overexpressed on the plasma membranes of tumor cells in a variety of cancers compared to normal cells. Overexpression of HER2 is associated with the development and progression of a multitude of malignancies and is a major therapeutic target in several cancers. 38 Downregulation of HER-2 expression attenuates HER2 signaling and promotes cell cycle arrest and apoptosis by altering a number of downstream signaling pathways. 39,42,43 Recently, a 42 nucleotide (nt) trimeric anti-HER2 aptamer (HApt) derived from degradation of HER2 in lysosomes was reported to kill HER2-overexpressing N87 gastric cancer cells and SKBR3 breast cancer cells. 40,41 The anti-HApt exerts a cytotoxic effect by inducing crosslinking of HER2 on the cell |
surface, which results in the translocation of HER2 from the plasma membrane to cytoplasmic vesicles (mainly in lysosomes), where HER2 is thought to be degraded by proteases. 41 In this study, we extended the application of our PEG 8 -PDPA 100 -PEG 8 MNPs to the delivery of oligonucleotides. The MNPs could efficiently package oligonucleotides of varied lengths (20-100 nt) and release them in a pH-dependent manner. We prepared MNPs encapsulating fluorescently labeled HApt and demonstrated that, compared to the free HApt, HApt encapsulated within MNPs was more efficiently taken-up and exerted more potent cytotoxic effects in SKBR3 breast cancer cells, which overexpress HER2. Preparation of polymeric Peg 8 -PDPa 100 -Peg 8 MNPs MNPs without oligonucleotides (unloaded MNPs) were prepared using a solvent evaporation method, as described previously. 29 Briefly, 3 mg of copolymer was dissolved in 1 mL of tetrahydrofuran (THF). Phosphate-buffered saline (PBS, 1 mM, pH 7.4) was added dropwise under high-speed stirring, and polymeric micelles were obtained after complete evaporation of THF. The final concentration of copolymer was adjusted to 0.3 mg/mL. Oligonucleotide-loaded MNPs (oligo-MNPs) were prepared by adding oligonucleotides 539 ph-responsive nanocarrier enhances her2 aptamer cytotoxicity dissolved in PBS in a dropwise manner to polymeric PEG 8 -PDPA 100 -PEG 8 THF solution under high-speed stirring. The final concentration of oligonucleotides was adjusted to 500 nmol/L. The oligo-MNPs were transferred into dialysis tubing (molecular weight cutoff, 50,000 g/mol) and dialyzed against PBS to remove free oligonucleotides. The size and morphology of the MNPs were examined by DLS and transmission electron microscopy (TEM; JEM-1400; JEOL, Tokyo, Japan). circular dichroism analysis Free and MNP-encapsulated HApt (1 μM) in PBS were analyzed using a Chirascan spectrometer (Chirascan; Applied Photophysics Ltd, Surrey, UK). Circular dichroism (CD) spectra were obtained from 300 to 200 nm using a 2 mm quartz cuvette at 25°C. The background CD spectrum of unloaded MNPs was obtained and subtracted from the CD spectrum of the HApt-MNPs. rates of ph-dependent release of hapt from micelles The release profile of HApt from HApt-MNPs was studied at various pH values using a fluorescence spectrophotometer (F-4500; Hitachi, Tokyo, Japan) at 610 nm. Briefly, 2 mL of Texas red-labeled HApt-loaded micelles was sealed in a dialysis tube (molecular weight cutoff, 50,000 g/mol), placed in a beaker containing 50 mL of PBS (10 mmol/L, pH =5.0, 6.5, or 7.4) at 37°C, and the concentration of fluorescent HApt in the PBS solution was measured after different time points. Accumulative HApt release rate (%) was calculated using A1/A2×100, where A1 is the absorption intensity of Texas red-labeled HApt at different time points and A2 is the intensity of the total amount of Texas red-labeled HApt (including free and encapsulated HApt). Quantification of MNP uptake by hela cells Human cervical carcinoma Hela cells were plated in 24-well plates (10 6 cells/well) and incubated for 24 h at 37°C in a 5% CO 2 atmosphere. MNPs containing Texas red-labeled oligonucleotides (500 nM) were added to each well and incubated for 8 h, and then, cells were washed twice with ice cold PBS to remove excess MNPs, harvested, and resuspended in PBS. Flow cytometry was performed on a fluorescence activated cell sorting (FACS) Calibur flow cytometer (BD Biosciences, San Jose, CA, USA), and data were analyzed using FlowJo 7.6.1. Uptake of fluorescent oligonucleotides by HeLa cells To assess aptamer-mediated killing of breast cancer cells, SKBR3 or MCF7 cells were incubated in 24-well plates with free HApt or HApt-MNPs at an aptamer concentration of 100, 125, or 150 nmol/L containing McCoy's 5A or DMEM without FBS for 8 h. Next, cells were cultured in fresh complete media for 16 h and then stained with 2 μM calcein acetoxy-methylester and 4 μM propidium iodide (PI) (Beyotime Biotechnology Co., Ltd., Nantong, People's Republic of China) for 30 min. Cells were observed under a fluorescence microscope; live cells were stained green, and dead cells were stained red. apoptosis assays An Annexin V-PI apoptosis detection kit (Signalway Antibody Co., Ltd., College Park, MD, USA) was used to assess apoptosis. SKBR3 cells cultured in 12-well plates were treated with free HApt, MNP-encapsulated HApt, or NCApt (125 nmol/L HApt or NCApt) for 8 h and then incubated in fresh complete media for 16 h. Cell apoptosis was analyzed using the FACSCalibur flow cytometry. confocal imaging of hapt-MNPs in sKBr3 cells SKBR3 cells were incubated with Texas red-labeled HApt-MNPs for 8 h (HApt concentration of 125 nM), followed by fresh complete media for 16 h. Cells were washed three times with PBS, fixed in 3.7% paraformaldehyde in PBS for 15 min at room temperature, washed, permeabilized with 0.5% Triton X-100 in PBS, washed, and then incubated in 1% bovine serum albumin in PBST for 30 min at room temperature. Nuclei were counterstained with 4,6-diamidino-2-phenylindole (Sigma-Aldrich Co., St Louis, MO, USA) or Hoechst 33342 (Sigma-Aldrich Co.) for 10 min at room temperature. HApt fluorescence was examined by confocal microscopy (TCS SP8; Leica Microsystems, Wetzlar, Germany). Texas red fluorescence intensity was measured by calculating the corrected total cell fluorescence (CTCF) using ImageJ. 44,45 CTCF = Integrated density -(area of selected cell × mean fluorescence of background readings) her2 immunoblotting SKBR3 or MCF7 cells were cultured in six-well plates (10 6 cells/well) in complete media for 24 h. Free HApt or HApt-MNPs were added (HApt concentration of 125 nM) and incubated for 24 h, and then, cells were collected, transferred to microcentrifuge tubes, and lysed in RIPA buffer (Pierce, Appleton, WI, USA) for 30 min on ice; protein concentrations were estimated using the bicinchoninic acid (BCA) assay (Bio-Rad Laboratories Inc., Hercules, CA, USA). An equal volume of sample buffer (125 mM Tris, pH 6.8, 4% sodium dodecyl sulfate (SDS), 10% glycerol, 0.006% bromophenol blue, and 1.8% β-mercaptoethanol) was added to each sample and boiled for 35 min. Samples (15 μg total protein) were loaded onto protein precast gels (Bio-Rad Laboratories Inc.) and electrophoresed at 120 V for 90 min, and proteins were transferred to polyvinylidene fluoride (PVDF) membranes at 300 mA for 115 min in transfer buffer. The membrane was immediately placed into blocking buffer (5% nonfat dry milk, 10 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Tween 20) for 1 h at room temperature and then incubated with primary anti-HER2 (rabbit; 1:10,000) or anti-β actin (rabbit; 1:5,000) antibodies (Abcam, Cambridge, UK) overnight at 4°C. The membrane was washed twice in washing buffer (10 mM Tris, pH 7.5, 100 mM NaCl, 0.1% Tween 20) and then incubated with alkaline phosphatase-conjugated antirabbit IgG secondary antibody (1:3,000; Abcam) diluted in 5% nonfat dry milk solution at room temperature. The membrane was washed three times, and protein signals were developed using enhanced chemifluorescence substrate (EMD Millipore, Billerica, MA, USA) and visualized using a Tanon 5500 (Tanon Science & Technology Co., Ltd., Shanghai, People's Republic of China). The relative amounts of each protein were determined by quantification of the band density values using ImageJ. statistical analysis The Student's t-test was used to compare data; P-values ,0.05 were considered statistically significant. Physicochemical properties and encapsulation efficiency of oligo-MNPs The physicochemical properties of the oligo-MNPs are summarized in Table 2. The diameter of the oligo-MNPs decreased as the length of encapsulated oligonucleotide increased. Encapsulation efficiency increased as oligonucleotide length increased from 20 to 80 nt. The polydispersity index values obtained using DLS revealed the size uniformity of the oligo-MNPs. At pH 7.4, the oligo-MNPs were negatively charged and their zeta potentials exhibited small fluctuations between -32.1 and -40.1 mV, suggesting that the oligo-MNPs are electrically stabilized by forming colloidal dispersions at pH 7.4 in PBS. In acidic solution (pH ,6.0), the PEG 8 -PDPA 100 -PEG 8 copolymers were positively charged due to protonation of the tertiary amine groups on the PDPA residues. However, the zeta potentials became negative when the pH was changed from acidic to alkaline (starting at pH ~6.5) due to deprotonation of the PDPA residues. 29 Since both the empty MNPs and oligo-MNPs were prepared in PBS at pH 7.4, they were expected to display a negative surface charge and the MNPs exhibited a more negative charge after oligonucleotide encapsulation. We were unable to determine exactly how the MNPs encapsulate oligonucleotides, though the hydrophobic interactions between nucleic acids and PDPA may play a major role. MNP-encapsulated hapt is structurally similar to free hapt The diameter and charge of HApt-MNPs in pH 7.4 PBS were 156.8±9 nm and -35.9 mV, compared to 184.8±16 nm and -10.6 mV for unloaded MNPs (Table 2). TEM showed that the HApt-MNPs had a regular round shape and were uniformly dispersed ( Figure 1A). We used CD spectroscopy to investigate the structural integrity of HApt in HApt-MNPs; structural integrity is thought to be required for HApt to bind to HER2 expressed on the cell membrane. 41 CD spectra of free HApt and HApt-MNPs (both at 1 μM HApt) were obtained from 300 to 200 nm. The CD spectrum of empty MNPs was subtracted from the CD spectrum of HApt-MNPs. Both free HApt and MNP-encapsulated HApt exhibited positive peaks at 265 and 210 nm and a negative peak at 245 nm ( Figure 1B). The similarity of the CD spectra for free HApt and HApt-MNPs suggests that the conformation of HApt was preserved after encapsulation. Furthermore, the peak values were much larger for HApt-MNPs than free HApt. Since the same concentrations of HApt were analyzed for free HApt and HApt-MNPs, the larger peak values for HApt-MNPs may be caused by condensation of HApt within the MNPs. [46][47][48] The CD spectra of the free and MNP-encapsulated NCApt were similar ( Figure S1). The rate of oligo-MNPs uptake is influenced by the length of oligonucleotide encapsulated To assess the cellular uptake of oligo-MNPs, we incubated HeLa cells with Texas red-labeled oligo-MNPs (0.15 mg/mL MNPs, equivalent to 125 nM oligonucleotide). After the encapsulation reaction, the oligo-MNPs were purified to remove free oligonucleotides. Cellular uptake of the fluorescently labeled oligonucleotides was determined by flow cytometry. Oligonucleotides of the same length with different sequences exhibited comparable cellular uptake rates However, as shown in Table 1, the uptake rate declined as the oligonucleotide length increased .80 nt, suggesting that longer oligonucleotides (.80 nt) negatively affect the uptake of oligo-MNPs. The fluorescent signals were mainly localized inside the cells ( Figure S2). We suspect that larger portions of longer oligonucleotides will be exposed to the outside of the micelles, which may prevent oligonucleotide condensation and limit cellular uptake. hapt is rapidly released from hapt-MNPs at ph values under ph 5.0 The cumulative levels of Texas red-labeled oligonucleotide were measured in vitro under different pH conditions at 37°C to examine pH-responsive release of oligonucleotides from oligo-MNPs. Under mildly alkaline conditions (pH 7.4), HApt was released slowly from HApt-MNPs, which reached 9.7%±5.0% after 47 h. Under weakly acidic conditions (pH 6.5), the release rate increased to 27%±4.0% after 47 h. However, at pH 5.0, HApt was rapidly released from as early as 12 h incubation and the level of fluorescence was nearly 60% after 26 h (Figure 2). These results confirm that HApt is released from HApt-MNPs in a pH-responsive manner. HApt was not fully released at pH 5.0, consistent with our previous study. 29 The hydrophobic PDPA core may become gradually protonated and became swollen but not collapse under acid conditions, trapping some HApt with the core. At the same time, small amounts of HApt release may also partly explain the absorption of HApt on the dialysis tubing. HApt-MNPs are more efficiently taken up by sKBr3 cells than free hapt We used SKBR3 breast cancer cells as a cellular model of HER2 overexpression and MCF7 cells as a model of normal/ low HER2 expression. 49 HER2 mRNA and protein levels were examined using quantitative real-time polymerase chain reaction and Western blotting, respectively. HER2 mRNA expression was 11.3-fold higher in SKBR3 cells than in MCF7 cells ( Figure S3A). Accordingly, HER2 protein was abundantly expressed in SKBR3 cells but barely detectable in MCF7 cells ( Figure S3B). ph-responsive nanocarrier enhances her2 aptamer cytotoxicity SKBR3 cells were incubated with the same aptamer concentration (125 nM) of free HApt or HApt-MNPs. Confocal fluorescence microscopy showed the Texas red signals were much stronger in SKBR3 cells incubated with HApt-MNPs than free HApt at 8 h ( Figure 3). Moreover, after 16 h incubation, fluorescent signals were observed in distinct clusters in SKBR3 cells incubated with |
HApt-MNPs compared to the weaker, diffuse signals in cells incubated with free HApt. This clustering pattern suggests that the HApt-MNPs were taken up into vesicular compartments after binding to HER2 on the cell membrane. 38,41 The intensity of the cellular fluorescent signals was 1.72fold higher in SKBR3 cells incubated with HApt-MNPs (CTCF =1,497,627.0) than cells incubated with control NCApt-MNPs (CTCF =871,819.2) at the same aptamer concentration (125 nM), suggesting that cellular uptake of the HApt-MNPs partially depends on HER2-mediated endocytosis. As nanoparticles up to several hundred nanometers can enter cells via endocytosis in membrane-bound vesicles, a certain amount of HApt-MNPs and NCApt-MNPs must also have been taken up by SKBR3 cells via HER2-independent endocytosis ( Figure 3). The importance of the HApt-HER2 interaction on the uptake of HApt-MNPs was also supported by flow cytometry analysis of SKBR3 and MCF7 cells. After incubation with HApt-MNPs, a lower percentage of MCF7 cells, which express low levels of HER2, was fluorescent (29.1%) compared to HER2-overexpressing SKBR3 cells (56.8%; Figure S4). Moreover, we performed competitive assays of HApt-MNP uptake and cell viability. After 24 h preincubation of SKBR3 cells with free HApt (0.5 or 1.0 μM), the media were refreshed and the cells were incubated with HApt-MNPs (containing 100 nM HApt) for 24 h. Pretreatment with free HApt significantly decreased the efficiency of HApt-MNP uptake ( Figure S5A) as well as the mortality of the cell compared to cells treated with only HApt-MNPs ( Figure S5B). Preincubation with the control NCApt had no apparent effect on HApt-MNP uptake or cell viability ( Figure S5A and B). These results indicate that the uptake of HApt-MNPs and the resulting cell death are, at least to some extent, dependent on the HApt-HER2 interaction. Unloaded MNPs exhibit low cytotoxicity toward breast cancer cell lines To evaluate the cytotoxicity of the unloaded MNPs toward breast cancer cells, HER2-overexpressing SKBR3 cells and HER2-underexpressing MCF7 cells were incubated with The CCK-8 assay showed that both cell lines remained highly viable. The percentages of viable SKBR3 and MCF7 cells were ~75 and .85%, respectively, even when incubated with the highest concentration of unloaded MNPs (200 μg/mL; Figure 4A and B). hapt-MNPs exert a potent cytotoxic effect in her2-overexpressing sKBr3 cells Next, we evaluated the cytotoxicity of HApt-MNPs in SKBR3 and MCF7 cells. In the CCK-8 assay, no significant cytotoxicity was observed in either SKBR3 or MCF7 cells treated with MNPs carrying the control NCApt ( Figure 5A and B) or unloaded MNPs ( Figure 4A and B). In contrast, when the aptamer concentration was $125 nM, at least 60% of HER2-overexpressing SKBR3 cells were killed by HApt-MNPs, whereas only 10% of MCF7 cells were killed ( Figure 5A and B). This suggested that the cytotoxicity of HApt-MNPs depends on the expression of HER2 in the target cells and was confirmed by live-dead cell staining of SKBR3 cells treated with free HApt or HApt-MNPs ( Figure 5C) and apoptosis assays of SKBR3 cells incubated with free or MNP-encapsulated HApt or NCApt ( Figure 5D). The LD 50 of HApt-MNPs in SKBR3 cells was 115 nM, while the LD 50 of free HApt was .5 μM ( Figure S6). These data indicate that the interaction between HER2 and HApt enables HApt-MNPs to exert a specific cytotoxic effect in HER2overexpressing cells. hapt-MNPs accumulate in lysosomes and induce degradation of her2 To demonstrate the HApt-MNPs primarily accumulate in lysosomes after cellular uptake, a lysosome tracker (LysoTracker Green) was used to stain SKBR3 cells. The cells were cultured with Texas red-labeled HApt-MNPs or free HApt for 8 h at the same HApt concentration (125 nM), followed by fresh complete media for 16 h, and then stained. The cellular Texas red signals were much stronger for HApt-MNPs ( Figure 6A, top panel) than free HApt ( Figure 6A, bottom panel). HApt-MNPs appeared to be distributed in a clustered pattern within SKBR3 cells, suggesting that the nanoparticles were located in the vesicular compartments after uptake. Merging the Texas red and LysoTracker signals indicated that the nanoparticles were colocalized to lysosomes. Furthermore, stronger lysosomal signals were observed in cells incubated with HApt-MNPs, suggesting more lysosomes accumulated after cellular uptake of HApt-MNPs. These results demonstrate that HApt-MNPs were more efficiently delivered to lysosomes than free HApt. To determine whether the increased HER2 degradation and cytotoxic effects observed in SKBR3 cells treated with HApt-MNPs can be attributed to altered lysosomal activity, we examined whether cell viability and HER2 protein expression were affected when lysosomal function was compromised using bafilomycin A1, a macrolide antibiotic that passively permeates into lysosomes and increases lysosomal pH, which disrupts lysosomal activity. 50,51 Figure 6C indicates that pretreatment with bafilomycin A1 followed by co-incubation with HApt-MNPs increased cell viability (by ~25%) and upregulated HER2 (by ~38%) compared to cells treated with the same concentration of HApt-MNPs without bafilomycin A1 to block lysosomal function ( Figure 6D). The increase in pH inside cells treated with bafilomycin A1 was confirmed by LysoTracker Red staining ( Figure S7). Together, these results confirm that HApt-MNPs were released inside lysosomes at low pH. Discussion HER2, a member of the epidermal growth factor-related protein (ErbB) family, is commonly overexpressed on the plasma membrane in a variety of tumors, including gastric and breast cancers, mainly due to amplification of the erbB2 gene. Overexpression of HER2 on the cell surface promotes tumor progression and metastasis. Monoclonal antibodies targeting HER2 (eg, Herceptin/Trastuzumab) are clinically used to treat HER2-overexpressing metastatic gastric and breast cancers. Stimulation of the immune system (eg, ADCC) is critical for the cytotoxic of monoclonal antibodies. However, the resulting immune reactions also lead to several side effects. 52 The trimeric version of the HApt used in this study was initially developed by Mahlknecht et al, 41 who demonstrated that HApt promoted translocation of HER2 from the cell surface to the cytoplasm in HER2-overexpressing N87 gastric cancer cells, which was associated with lysosomedependent clearance of HER2 protein. Lee et al 40 reported that HApt exerted a cytotoxic effect in HER2-overexpressing SKBR3 breast cancer cells. HApt has been shown to induce cross-linking of HER2 on the cell surface, resulting in the translocation of HER2 to cytoplasmic vesicles for lysosomal degradation. Furthermore, HApt-mediated HER2 degradation triggered G0/G1 phase cell cycle arrest and cell death in ph-responsive nanocarrier enhances her2 aptamer cytotoxicity SKBR3 cells. 40,41 Therefore, HApt does not exert a cytotoxic effect by directly stimulating the immune system. Based on these previous reports, we hypothesized that our previously reported pH-responsive nanocarrier 29 would be ideally suited to deliver HApt to HER2-overexpressing cells. The MNPs are pH-responsive nanocarriers that encapsulate nucleic acids, which could facilitate cross-linking and thus internalization of HER2, and disassemble under acidic conditions, which may increase targeted degradation of HER2 in lysosomes. In this study, we confirmed that compared to free HApt, HApt-carrying nanoparticles (HApt-MNPs) increased HApt uptake and lysosomal transport in HER2-overexpressing SKBR3 cells (Figures 3 and 6A). Endogenous HER2 protein expression decreased significantly in cells treated with HApt-MNPs compared to cells treated with free HApt ( Figure 6B). When lysosome activity was blocked, cell viability and HER2 protein expression increased in cells treated with HApt-MNPs compared to cells treated with the same concentration of HApt-MNPs alone ( Figure 6C and D). Cell viability and apoptosis assays showed that HApt-MNPs exerted a more potent cytotoxic effect in SKBR3 cells than free HApt ( Figure 5A, C, and D). Collectively, these data demonstrate that HApt-MNPs exert a specific cytotoxic effect in HER2-overexpressing SKBR3 cells. Several factors may contribute to the more potent cytotoxic effects of HApt-MNPs than free HApt. First, the MNP nanoparticles increased the delivery of HApt into HER2-overexpressing cells compared to free HApt. Second, the pH-responsive release property of HApt-MNPs ensured that the encapsulated HApt cargo was released in endosomes/lysosomes. Third, the MNPs increase the local concentration of HApt: aggregation of HApt on MNPs would increase the ability of HApt to cross-link to HER2 on the cell surface. The ability of MNPs to remain stable at normal physiological pH is beneficial, as it prolongs their retention time in the blood circulation and thereby greatly reduces the adverse effects of the drug cargo in normal tissues. 37,39 Moreover, pH-sensitive release of HApt from MNPs in the endosomal or lysosomal compartments (pH 4.0-6.5) would result in rapid intracellular release of HApt in the target cells. 53,54 Our lysosome activity inhibition experiments confirmed that the release of HApt inside the cell was dependent on low lysosomal pH. Furthermore, HApt must maintain the correct structure during loading and unloading within MNPs to retain its cytotoxic activity. We used CD spectroscopy to show that the conformation of HApt is likely to be preserved after encapsulation within MNPs. Together these properties indicate that MNPs are a promising candidate for effective delivery of HApt to HER2-overexpressing cancer cells. Further studies are needed to evaluate the cytotoxic effect of HApt-MNPs in vivo. These findings also indicate that MNPs could be used for the delivery of a range of oligonucleotides. Oligonucleotides ranging in length from 20 to 100 nt were efficiently encapsulated, with optimal encapsulation at 50 to 80 nt. The size of the MNPs declined as the length of encapsulated oligonucleotide increased. This indicates large single-stranded nucleic acids coil back on themselves due to complementary base pairing (hairpin structure) to form highly compacted supercoiled molecules. 55,56 Further studies are necessary to determine whether supercoiling of long oligonucleotides suppresses the cellular uptake of oligo-MNPs. Conclusion Our pH-responsive and lysosome-targeting nanoparticle system can efficiently deliver a therapeutic oligonucleotide in vitro. The HApt-MNPs were translocated to lysosomes, where HApt was rapidly released at low pH. Therefore, MNPs may have the potential to improve the delivery and efficacy of nucleic acid-based cancer therapies. Notes: (A) HER2 mRNA expression was quantified by quantitative reverse transcription polymerase chain reaction. Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA), reverse transcribed using PrimeScript RT reagent Kit (TaKaRa, Dalian, People's Republic of China) according to the manufacturer's instructions, and amplified using SYBRPremix Ex Taq (TaKara). Pcrs were performed in triplicate with the following conditions: 95°c/30 s, 40 cycles of 95°c/5 s, 60°c/15 s, and 72°c/10 s on a stratagene MXP3000 cycler (stratagene, la Jolla, ca, Usa) and repeated at least three times. relative mrNa levels were calculated using the -ΔΔct method using β-actin as a control and expressed as 2 -ΔΔct . The primer pairs were as follows: β-actin-f/β-actin-r: cTgggacgacaTggagaaaa/ aaggaaggcTggaagagTgc; HER2-f/her2-r: gcagcTTcaTgTcTgTgcc/acagagacTcagacccTggc. Mean ± sD values for three independent experiments are presented. Significant differences were determined using the Student's t-test, ***P,0.001. (B) her2 protein expression was analyzed by Western blotting. β-actin was used as a protein loading control. Preincubation with the control NCApt had no significant effects on HApt-MNP uptake or cell viability. Abbreviations: hapt, human epidermal growth factor receptor 2 aptamer; MNPs, micelle-like nanoparticles; Ncapt, negative control aptamer; Ta, Texas red. Publish your work in this journal Submit your manuscript here: http://www.dovepress.com/international-journal-of-nanomedicine-journal The International Journal of Nanomedicine is an international, peerreviewed journal focusing on the application of nanotechnology in diagnostics, therapeutics, and drug delivery systems throughout the biomedical field. This journal is indexed on PubMed Central, MedLine, CAS, SciSearch®, Current Contents®/Clinical Medicine, Journal Citation Reports/Science Edition, EMBase, Scopus and the Elsevier Bibliographic databases. The manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. Visit http://www.dovepress.com/ testimonials.php to read real quotes from published authors. International Journal of Nanomedicine 2018:13 submit your manuscript | www.dovepress.com Dovepress Dovepress Dovepress 553 ph-responsive nanocarrier enhances her2 aptamer cytotoxicity Figure S6 Viability of sKBr3 cells after treatment with different concentrations of free hapt or Ncapt. Notes: cells were incubated with 0.2-5 μM of free hapt or free Ncapt for 8 h, followed fresh complete media for 16 h. Then, cell viability was assessed using the ccK-8 assay. Mean ± sD values for three independent experiments are presented. approximately 55.1% of sKBr3 cells treated with 5 μM free hapt were nonviable, compared to only 13.9% of cells treated with 5 μM free Ncapt. Abbreviations: ccK-8, cell counting Kit-8; hapt, human epidermal growth factor receptor 2 aptamer; Ncapt, negative control aptamer. Figure S7 LysoTracker Red staining of bafilomycin Al-treated cells. Notes: SKBR3 cells cultured in a 24-well plate were incubated with McCoy's 5A media containing 50 nM bafilomycin A1 (Baf) (TOCRIS, Bristol, UK) dissolved in DMSO for 1 |
h at 37°c, followed by 50 nM lysoTracker red (YeaseN Biotech co., ltd., shanghai, People's republic of china) for 30 min, washed three times with PBs, and then examined by fluorescence microscopy (AMG EVOS, Mill Creek, WA, USA). For control experiments, 1% DMSO was added to the medium. In the control group, the intensity of red-fluorescence was rather strong but faded sharply after the addition of bafilomycin A1, indicating an increase in lysosomal pH. Scale bars are 200 μm. Abbreviations: DMsO, dimethyl sulfoxide; PBs, phosphate-buffered saline. Insights Gained and Future Outlook From scRNAseq Studies in Autoimmune Rheumatic Diseases Autoimmune rheumatic diseases have a major impact on public health as one of the most common morbidities, and many of these disorders involve both local and systemic manifestations with severe consequences for patient health and quality of life. However, treatment options for many of these diseases remain inadequate for a substantial portion of patients, and progress in developing novel therapeutics has been slow. This lack of progress can be largely attributed to an insufficient understanding of the complex mechanisms driving pathogenesis. Recently, the emergence of single-cell RNA sequencing (scRNAseq) has offered a powerful new tool for interrogating rheumatic diseases, with the potential to assess biological heterogeneity and individual cell function in rheumatic diseases. In this review, we discuss the major insights gained from current scRNAseq interrogations of human rheumatic diseases. We highlight novel cell populations and key molecular signatures uncovered, and also raise a number of hypotheses for follow-up study that may be of interest to the field. We also provide an outlook into two emerging single-cell technologies (repertoire sequencing and spatial transcriptomics) that have yet to be utilized in the field of rheumatic diseases, but which offer immense potential in expanding our understanding of immune and stromal cell behavior. We hope that scRNAseq may serve as a wellspring for the generation and interrogation of novel hypotheses regarding autoreactive lymphocytes and tissue infiltration patterns, and help uncover novel avenues for therapeutic development. INTRODUCTION Autoimmune rheumatic diseases have a major impact on public health as one of the most common morbidities, affecting some 3% of the general population (1). Many of these disorders involve both local and systemic manifestations in multiple organs, leading to severe health consequences and compromising patient quality of life. Treatment options for these disorders are largely limited to immunosuppressants, in the form of conventional synthetic drugs such as methotrexate, or monoclonal antibodies against specific targets (biologics). While many of these treatments are effective at first pass in alleviating patient symptoms, a substantial portion of patients are refractory towards treatment (2). Others may still suffer from periodic flares in disease activity, and/or loss of drug efficacy over time. However, the development of new therapeutics for improved treatment has been painstakingly slow, with no treatment having been approved for primary Sjogren's syndrome, and very few for systemic lupus erythematosus, leading to major difficulties in prescribing more precise clinical guidelines tailored for individual patients (3,4). This lack of progress can be largely attributed to an insufficient understanding of the complex mechanisms driving pathogenesis. In particular, the what/ where/how questions regarding the cell types causing disease progression have only been incompletely addressed, and higherlevel knowledge of how disparate cell types may interact with each other and their environment in these diseases is lacking. Attempting to answer these big-picture questions using older low-throughput technologies has proven to be a laborious endeavor. Fortunately, the emergence of new high-throughput technologies for assessing biological heterogeneity may provide fruitful approaches for tackling these questions. Since its advent from relatively low-throughput applicability, single-cell RNA sequencing (scRNAseq) has evolved into a powerful tool for assessing biological heterogeneity and individual cell function in complex environments at extremely high throughput (5,6). Unlike previous bulk sequencing studies where unique characteristics of a particular population of cells might be hidden by complex changes in the overall average, scRNAseq can be used to efficiently classify changes in individual cell types with simultaneous knowledge of global and local structure. As such, scRNAseq can identify rare populations and their functional characteristics in an unbiased manner in physiological and pathological conditions. Furthermore, trajectory inference based on minute transcriptome changes can be used to accurately predict differentiation and/or developmental hierarchies within and between cell populations (7). Novel bioinformatics tools have also enabled more detailed dissection of the pathways and molecules contributing to trajectory progression (8), as well as communications patterns between cells (9). In the context of rheumatic diseases where the overall composition, differential hierarchies, and functional characteristics of cell types contributing to disease remains incompletely understood, scRNAseq thus holds considerable promise for as a key method for unlocking novel, unbiased insights. These insights may further serve as foundational resources for uncovering new avenues for translational research and therapeutic development. In this review, we first summarize major findings obtained from current scRNAseq studies in rheumatoid arthritis, systemic lupus erythematosus, and primary Sjogren's syndrome. We also highlight a number of provocative questions raised by the results obtained from these studies, and which may be of relevance towards furthering current knowledge in the field. We further discuss the potential value for integrating other forms of single-cell omics data into these mapping studies and beyond, to preview some forms of potential knowledge to come from future studies. RHEUMATOID ARTHRITIS Rheumatoid arthritis (RA) is an autoimmune disease characterized by erosion of the protective collagen layer in joints across the body under inflammatory stress (10). In a significant majority of cases, this stress is directly provided by the infiltration of large numbers of immune cells into the synovial membrane, although a leuckocyte-poor presentation of RA has also been described (11). Infiltrating immune cells have been demonstrated to secrete a wide range of pro-inflammatory cytokines to stimulate local inflammation and osteoclast differentiation, and may also help to promote excessive proliferation of fibroblast-like synoviocytes (FLS) (12). The resulting increase in synovial fluid production is a key driver of the joint swelling observed in RA, and may also promote the formation of a complex joint microenvironment. Dissecting the full complexity of this microenvironment has proven to be difficult. However, the advent of scRNAseq has allowed for more comprehensive census of the cell types present and their functional states, as reviewed below. A number of efforts have been made to generate a cellular atlas of rheumatoid arthritis in both peripheral circulation and in afflicted joints. Two early single-cell mapping studies using a homebrew microfluidic system (13) or the plate-based CEL-seq2 method (14,15) both reported striking heterogeneity among synovial FLS, characterizing a novel population expressing Thy1 (CD90) and lacking expression of CD55. In contrast to the CD55+ FLS distributed in the synovial membrane lining, Thy1+ FLS localize more deeply within synovial tissue. These Thy1+ FLSs also appear to express key chemokines, such as CXCL12, and might contribute to the recruitment of immune cells to synovial tissue. Unfortunately, the relative proliferative potential of these FLS subtypes has not been examined, and it is unclear if these FLS will become a dominant population as RA progresses. It is also not fully clear if these FLS retain the ability to produce synovial fluid, and/or if the composition of the fluid produced is distinct between subtypes. Interestingly, a portion Thy1+ FLS may also express the MHC-II molecule HLA-DRA. These results raise an interesting possibility that differentiation of synovial FLS towards a proinflammatory state may occur as an early event during RA pathogenesis, with this conversion potentially contributing to the recruitment of immune cells to synovial tissue. The potential of synovial FLS derived from RA patients to act as additional antigen-presenting cells to activate T cells has been previously demonstrated in several reports (16,17). However, these previous reports all included an additional interferon-gamma sensitization step, leading to some questions regarding the true relevance of these in vitro assays. In light of the scRNAseq results, it may be interesting to re-evaluate the antigen-presentation potential of synovial FLS by focusing only on the Thy1+ subpopulation. If these FLS are indeed capable of antigen presentation without the need for additional cytokine stimulus, and can easily localize near infiltrating T cells in synovial tissue, targeting this population may prove to be essential for preventing further relapse in RA. A preliminary clinical trial (TRAFIC, ISRCTN36667085) has already been initiated to target this FLS in patients refractory to methotrexate, and results from these studies may yield additional insights into their function (18). Recently, another mapping-focused study has been performed using samples from patients with or without anti-citrullinated peptide antibodies (ACPA, a type of autoantibody observed in roughly two-thirds of patients with established RA) in a cohort with minimal treatment intervention (19). Interestingly, the study reported a significant decrease in the expression of the MHC-II molecule HLA-DRB5 on a portion of plasma cells in the peripheral blood of ACPA+ patients compared to both ACPA-counterparts and healthy controls, and with remarkable similarity to naïve B cells. As these HLA-DRB5+ plasma cells do not appear in the concurrently sampled synovial specimens, it is possible that this population may represent a protective phenotype against RA, or might otherwise be composed of B cells with high affinity for antigens unrelated to RA progression. Variation in MHC-II genes has been previously reported to be significantly associated with RA from gene association studies, with variation in HLA-DRB1 having been long identified to be correlated with levels of ACPA, a finding reproduced in a number of cohorts using different technologies (20,21). It may be interesting to evaluate if HLA-DRB5 may also be a concurrent or independent risk allele for ACPA levels, and if expression changes in HLA-DRB5 occur independently/ concurrently with expression changes in other MHC-II genes and/or molecules involved in antigen presentation and co-stimulation. At the same time, the study also noted a substantial increase in the expression of chemokines from the CCL family in both dendritic cells and macrophages of ACPA-patients compared to ACPA+ counterparts. If these increases are indeed independent of current patient disease severity, this result may suggest an alternative, more innate-centric model for disease progression in ACPA-patients, compared to a more lymphocyte-driven reaction in ACPA+ counterparts. Such a divergence in mechanism may partly explain clinically observed differences in disease progression (22). Further investigations may help to shed light on these possibilities. Other recent studies have focused on a specific population of cells to investigate them in greater detail. For instance, a study focused on the macrophage compartment in RA synovial tissue emphasized the existence of CD206+MerTK+ macrophages with a unique regulatory signature in RA patients experiencing clinical remission, and correlated with remission maintenance (23). This population matches up well with a population of residential joint-protective macrophages identified in mice (24). Since CD206+ has been traditionally used to identify regulatory M2 macrophages in multiple contexts, and stimulation of MerTK signaling has been shown to favor M2 differentiation, these results are in one respect unsurprising (25,26). Interestingly however, the study also indicated that a primary ligand for MerTK, Gas6, was abundantly produced by sublining Thy1+ FLS, and that this production occurred simultaneously with that of the chemokine CXCL14 in patients with active disease and experiencing remission. Notably, MerTK+ macrophages were largely located in the lining layer, and were thus not in direct contact with the Thy1+ FLS, while CD68+MerTK-macrophages were present in the sublining. This latter result suggests that apparently pro-inflammatory FLS may also serve as a key driver of a homeostatic regulatory circuit by promoting differentiation of M1 macrophages towards M2 phenotypes. Additional work on this homeostatic circuit and the identification of biological pathways to promote M2 conversion and Thy1+ FLS production of Gas6 may thus be of considerable therapeutic interest. SYSTEMIC LUPUS ERYTHEMATOSUS Systemic lupus erythematosus (SLE) is a severe autoimmune disorder typically involving systemic manifestations capable of causing major, and oftentimes chronic, organ and tissue damage (27,28). The clinical progression of SLE frequently involves repeated series of flares in disease activity followed by period of remission, for reasons that remain incompletely understood. From an immunological standpoint, it is commonly appreciated that autoantigens derived from cell nucleus components (both proteins and double-stranded DNA) are the key drivers of disease, and high titres of circulating anti-nuclear antibodies (ANA) typically correspond with elevated disease activity (29). These autoantibodies have been shown to form large deposits in the kidneys and other organs, and are a key predictor of renal injury, presumably via antibody-mediated damage |
(30,31). As such, the search for novel therapeutics against SLE has emphasized the identification of drugs capable of restraining B cell activity and antibody production. However, by virtue of its systemic nature, a large range of immune cell types are likely to be involved in SLE disease progression. In order to map this systemic disease, large cohort studies have been conducted to characterize the immune populations present in peripheral circulation of SLE patients. The first of these studies focused on identifying populations of immune cells especially responsive to interferon stimulation (32). These analyses led to the identification of IL-1b+ monocytes, CD11c+ B cells, and FCGR3A+ NK cells as expressing interferon-stimulated genes (ISGs) at high levels. These increases appeared to be systemic among the patients sequenced, and some of these populations appear to be enriched among patients with higher disease activity scores, although the actual correlations are not fully clear. Notably, despite longstanding concerns regarding patient heterogeneity in rheumatic diseases being a major confounding factor, the actual heterogeneity in regards to cell type changes appears to be quite small, and at a level controllable by batch normalization. More importantly, these data emphasize the important influence of interferon signaling in the context SLE, and demonstrate that pervasive interferon signaling may significantly reshape the differentiation processes of circulating immune cell populations. Of note, a similar ISG signature has also been observed in lowdensity granulocytes (33) and in dendritic cells (34), further demonstrating its reach. While a large number of interferon genes exist, there has long been evidence suggesting that IFNa may be the one most strongly correlated with disease activity in SLE, and treatment with IFNa may provoke SLE-like side effects in some patients (35,36). Intriguingly, a monoclonal antibody against the IFNa receptor has also emerged in early clinical trials as a candidate with significant promise in patients with moderate-to-severe SLE, although results in phase III have not yet been clarified due to differing choices in primary endpoint designation between two trials (NCT01438489, NCT02446912, NCT02446899), with significance relative to placebo being found using one metric, but not the other (37)(38)(39). Further investigations may help clarify if the abundance of ISGhi immune cells is indeed predominantly caused by signaling from IFNa, and the suitability of targeting this pathway for treating a broad population of SLE patients. For instance, one possibility may be that patients with the highest ISG activity may be more sensitive to intervention through IFNa compared to patients with lower ISG activity. At the same time, additional explorations of other biological pathways and processes that are highly correlated with progression towards ISGhi signatures may provide additional insights into the mechanisms by which these cell populations exert pathological functions. In particular, it is unclear if continued expression and stimulus of interferon receptors is required for the maintenance of these ISGhi cell states, or if they represent a more stable end-state. Trajectory analyses that reconstruct the differentiation hierarchies between cell states may be particularly useful in this context for identifying potential novel forks in differentiation caused by the elevated levels of interferon observed. The inclusion of additional data from currently unpublished efforts to study signatures in patients with acute flares and with other unique pathological features may help to further clarify this possibility. Beyond the peripheral transcriptome, other studies have focused on evaluating regional differences brought about by SLE pathology in the kidneys and skin. By isolating immune cells from renal biopsies, one study was able to capture a complex landscape of renal-infiltrating immune cells broadly similar to the peripheral atlas, including several populations of T and B cells defined by an ISG signature (40). Part of this signature could also be recapitulated in urine, raising the potential for a noninvasive method for assessing lupus nephritis (41). Similarly, a study of lesional and non-lesional skin from SLE patients also reported increases in ISG signature in T cells (42). Interestingly, the renal landscape also recovered an increase in ISG signature in tissue resident macrophages expressing the transcription factor BLHLE41, consistent with the hypothesis that the function and transcriptome of residential immune cells may also be significantly altered as a result of interferon signaling in SLE. However, substantial numbers of T and B cells have also been demonstrated to be present in the kidneys under physiological conditions (43). It is unclear how these other residential populations might be impacted by interferon signaling, or if they might be displaced by the large numbers of infiltrating cells. After all, these T and B cell populations also include significant proportions of naïve B cells as well as T cells with a follicular-helper phenotype, raising a possibility that T and B cells may even localize together into germinal center-like structures in the kidney. While histological studies of lupus nephritis have previously found associations between these structures and worsened clinical outcomes, it remains unclear if these structures can indeed facilitate affinity maturation of B cells. Additional immune repertoire focused sequencing may help to clarify this question. PRIMARY SJOGREN'S SYNDROME Primary Sjogren's syndrome (pSS) is characterized by excessive dryness at mucosal membranes due to insufficient fluid production from secretory glands (particularly salivary and lacrimal glands) (44). Histological examination of these glands in patients with pSS will typically identify immune foci formed by infiltrating T and B cells, and these cell types are generally understood to contribute significantly to disease pathogenesis. A significant portion of patients with pSS will also experience chronic fatigue, and symptomatic manifestations in a number of organs have been reported, although the underlying mechanisms driving systemic effects and possible disease subtypes are unknown at this time (45). pSS has also proven to be especially difficult to treat, with current therapy often limited to alleviation of symptoms (such as artificial tears for sicca) (46). This intractability is due to traditional immunosuppressants failing to demonstrate durable clinical efficacy either alone or in combination in larger randomized clinical trials. However, a number of early stage trials evaluating novel compounds and/or biologics have yielded some promising early results. For instance, sustained reductions in disease activity through 24 weeks via treatment with an antibody targeting the BAFF receptor has been recently reported in a phase 2b trial (NCT02962895) (47). Further research is urgently needed to clarify if these biologics and/or other drugs do in fact represent viable therapeutic options for pSS. Thus far, one preliminary scRNAseq study has been performed examining the peripheral blood of patients with pSS as compared with healthy controls (48). The authors reported an increase in CD4+ T cells with a potentially cytotoxic phenotype in pSS patients, with this population appearing to be transcriptionally similar to CD8+ CTLs from the dimension reduction provided. At the same time, follow-up flow cytometry data presented raises confusion as to the expected percentages of this population within CD4 T cells. Interestingly, earlier histological examinations of minor salivary glands in pSS patients have reported the existence of high percentages of CD4 CTLs in SS lesions (49). In addition, CD4 CTLs in other contexts have been demonstrated to be capable of directly killing target cells based on recognition mediated by MHC-II (in contrast to MHC-I recognition in conventional CD8+ CTLs) (50). It may be interesting to investigate if peripheral blood CD4 CTLs in pSS are clonally and functionally associated with the lesional CD4 CTLs seen in salivary glands, and if these cells can perform MHC-II mediated targeted killing of glandular exocrine cells. The establishment of a clonal relationship between peripheral and gland-infiltrating T cell populations may be especially important in the context of pSS due to the complexity of the peripheral signature of pSS. A number of large-cohort studies using bulk RNAseq of peripheral blood have found remarkably few differences between the pSS patients and healthy controls, with most of the variation being concentrated in the interferon signaling pathways (51)(52)(53). As such, one interpretation may be that a few interferon-sensitive populations might be substantially related to disease progression. It is unclear if interferon signaling may also modulate the presence and/or function of CD4 CTLs, or of the other populations observed. Two preliminary single-cell repertoire sequencing studies of gland-infiltrating T cells has already been performed, and have reported clonal expansion of select T helper populations (54,55). Additional work, ideally with concurrent sampling of multiple afflicted glands, will be necessary to further clarify the immune cell populations contributing to pSS pathogenesis, and potentially identify novel therapeutic strategies for this condition. SUMMARY AND OUTLOOK FROM CURRENT scRNAseq STUDIES Atlas-focused mapping studies have been conducted in each of the three rheumatic diseases reviewed above, and have helped provide a broad contour of the general immune cell landscape in each of these diseases, in both circulation and in afflicted tissues. These studies have also made some inroads into identifying some potentially pathogenic and protective cell types that have been previously underappreciated. A summary of the primary findings reached thus far is included as Table 1. In the case of SLE, which has the largest number of patients sequenced thus far, interferon signaling appears to play the biggest role in differentiating single-cell profiles during both active disease and remission. Multiple immune cell types appear to have skewed differentiation, and this skew extends to tissue infiltrating immune cells. The pervasiveness of this signature raises hope that disruption of the interferon signaling axis, whether via monoclonal antibodies or other means, may serve to effectively restrain disease activity in SLE. scRNAseq studies in SLE have thus emphasized a direction for translational research with the establishment of a dominant signaling pathway and the characterization of unique cell subtypes. For instance, if additional mappings are performed to gauge therapeutic response in SLE patients to a particular treatment, it may be possible to utilize these signatures as a key reference point. Assessment of these signatures might also help as a leading indicator to predict relapse or remission. Importantly however, these assessments will require validation in much larger cohorts of patients than those typically sequenced in scRNAseq studies in order to reach sufficient power. As such, developing other assays that can capture the nuanced insights obtained from scRNAseq data will be necessary to further translational research into SLE. In the context of RA, less is known about the full spectrum of immune cells involved and the essential signaling mechanisms involved. While several sequencing studies of the synovium in RA have been performed, these studies have largely emphasized more local alterations in fibroblast and myeloid signatures in joints. In contrast to SLE, these alterations are not necessarily reflected in peripheral circulation. However, from the clinical correlation analyses reported thus far, it appears that the presence of some of these populations is positively associated with increases in systemic scores of disease activity and inflammation (such as ESR, CRP, and overall DAS28). It is currently unclear how these local changes in one joint may propagate to others, or otherwise influence RA-involved conditions in other tissues. Extensive future investigations are necessary to gain a clearer picture of the relevant molecular mechanisms and immune response patterns in RA, and how these mechanisms might enable autoimmune reactions across the body. The landscape of gland-infiltrating immune cells in pSS is similarly underexplored. As reviewed above, only a single scRNAseq study of circulating immune cells in pSS has been performed thus far, identifying a potential increase in CD4 CTLs. However, the relationship between circulating phenotypes and glandular damage in pSS remains unclear. Although there have been some indications of interferon signaling playing an important role in pSS based on transcriptome differences relative to healthy controls, it is currently unclear if strong ISG signatures can be found in both circulating and gland-infiltrating immune cell populations on a single-cell level in a manner akin to SLE. Detailed mappings of both peripheral blood and infiltrated glands in pSS patients will be necessary to clarify these and other possibilities. Besides the information already obtained from atlas studies, scRNAseq also has immense potential for uncovering additional biological insights by being integrated into other forms of omics information. In recent years, other sequencing-based omics technologies have become increasingly mature and are now available through established protocols. In the following section, we highlight single-cell immune repertoire sequencing and spatial transcriptomics as two technologies with particular potential to be applied in assessing rheumatic diseases and answer key questions regarding their pathological progression. CHALLENGE 1: LINEAGE TRACING AND IDENTIFICATION OF CLONALLY-PROLIFERATING AUTOREACTIVE T AND |
B CELL POPULATIONS Current scRNAseq studies of rheumatic diseases have elucidated a significant number of distinctive T and B cell populations as reviewed above. However, classification of these T and B cells has been driven by transcriptome-level information on expression profiles, and not based on their antigenic specificity, thus neglecting a critical component of T and B cell identity. Antigen specificity of T and B cells is determined by the sequence of unique recognition receptors [T cell receptor (TCR) and B cell receptor, respectively] encoded as specific sequences generated by genetic recombination (56,57). Unfortunately, in 3' biased sequencing data, it is not possible to directly recover these sequences. As such, the scRNAseq studies of autoimmune rheumatic diseases have thus far been unable to make direct determinations regarding the clonal status of T and B cells. Fortunately, new approaches for library construction have made it increasingly possible to capture full-length sequences, including in the TCR and BCR regions. In plate-based workflows, utilization of unique enzymes and/or amplification sequences (such as template-switching oligonucleotides with Moloney murine leukemia virus reverse transcriptase, or notso-random primers) generate full-length information and capture entire TCR and BCR sequences (58,59). In dropletbased sequencing, 5' library construction is now available through commercial platforms, and the addition of a second workflow for targeted amplification of TCR and BCR sequences can capture sequences in the complementarity determining region 3 (CDR3). As such, it is now possible to perform paired single repertoire sequencing along with scRNAseq to simultaneously profile the clonal identity and transcriptome of a given cell. This advance has opened up new possibilities, as the importance of autoreactive T and B cells clones in mediating rheumatic diseases has long been appreciated. Early studies in animal models raised the hypothesis of there being 'forbidden clones' driving autoimmune diseases (60), and subsequent findings in patients have helped to demonstrate the clear and common occurrence of autoreactive clones in rheumatic diseases. For instance, antibodies against double-stranded DNA (dsDNA) and other cell nucleus components (ANA) is a hallmark of SLE (61), while anti-Ro/anti-La (62) and ACPA (63) antibodies are observed at high frequencies in patients with SS and RA, respectively. Positive detection and circulating titers of these autoantibodies is now routinely used as a key parameter during clinical assessments of disease severity. In the context of SLE and SS in particular, deposits of these autoantibodies can often be found in afflicted tissues (particularly in renal glomeruli in SLE and salivary glands in SS) and contribute directly to disease pathology. Local stromal cells, such as salivary epithelial cells, might also promote the survival of these autoantibody-producing B cells as part of a vicious cycle (64). Similarly, autoreactive T cells have been reported for rheumatic diseases, although their presence is not assessed as part of traditional clinical evaluations (65,66). Autoreactive CD4+T cells can produce large amounts of cytokines to maintain inflammatory environments, while autoreactive CD8+T cells may directly lyse their target cells. These and other ancillary pathogenic characteristics have led to the view that targeting of autoreactive clones may be an efficient therapeutic strategy for rheumatic diseases. However, it is also important to note that autoreactive T and B cell clones will only represent a small portion of the total immune repertoire of any individual at a given time, particularly in circulation. Furthermore, positive serum detection of autoantibodies has been reported to precede clinical disease in SLE, SS, and RA, and a high percentage of people who do not develop rheumatic diseases may also have circulating autoantibodies (67). In the context of SLE, it has been reported that ANA-reactive B cells can be found among the plasma cells in peripheral circulation of lupus patients, but that these clones may only represent less than 1% of circulating plasma cells. It is unclear at this time whether autoreactive clones will be more highly concentrated in B cells of a particular phenotype/maturation status, or if they will be distributed in line with unrelated clones. It is also unclear if autoreactive clones might have differences in developmental trajectory compared to unrelated counterparts. Given the fact that patients with rheumatic diseases typically maintain significant circulating titers of autoantibodies, one might assume that autoreactive clones may tend to occupy a higher percentage of class-switched memory and plasma cells which can directly secrete large amounts of antibody. However, several studies that have inspected the proportion of ANAreactive B cells by stage in SLE patients have found only a surprisingly small increase in circulating ANA-reactive plasma cells compared to controls (68,69). Similarly, percentages of RF and ACPA B cells may only represent 1% of circulating B cells in patients in RA (70,71). While a single clone at 1% representation may stand out as being highly expanded in repertoire sequencing analysis, it may not be clearly dominant amongst all clones and be difficult to identify. Furthermore, a single autoantigen may trigger oligoclonal expansion, in which multiple clones expand concurrently, such that no single clone clearly emerges as dominant. As such, it can be surmised that the large majority of B cells reported in scRNAseq studies of peripheral blood do not belong to autoreactive clones and will not produce autoantibodies, and that distinguishing the autoreactive clones may be a difficult endeavor. It may also be possible that the expression profiles and trajectory behavior of autoreactive clones are indistinguishable from that of unrelated clones that have been exposed to the same inflammatory environment. Alternatively, it is possible that a more precise definition of B cell status derived from scRNAseq, when analyzed together with clonal identity, might help identify subtypes of plasma cells highly enriched for autoreactive clones. Such enrichment may also be more prominent when analyzing a more specific context, such as during clinical remission or abrupt flares. Furthermore, it is critical to remember that B cells can undergo somatic hypermutation and affinity maturation, wherein B cells can reshape their BCR sequence to increase their affinity for a given antigen in germinal centers under T cell help (72). As such, it is possible that many of the apparently non-autoreactive B cells may mature into autoreactive clones once their affinity is raised to a sufficiently high level. Focused evaluation of one particular antigen then, such as dsDNA (73), instead of the broader spectrum of ANAs, may also yield more direct information on specific autoreactive clones. Through currently available computational approaches that can generate B cell clonal lineage trees, it may be possible to gain additional insight into autoreactive clones should enough BCR sequencing and antigen affinity data become available (74). Since T cells lack the ability to undergo repeated rounds of affinity maturation and class-switch recombination, the TCR sequence of any given clone can be expected to stay stable, allowing for easier analysis. However, due to technical difficulties with generating tetramer libraries against possible autoantigens, even less is known about the behavior and frequencies of autoreactive T cell clones in rheumatic diseases at this time. In the context of pSS, and cloned T cells lines recognizing SSA have been generated (75), and autoreactive Th17 cells producing IL-17 following recognition of a peptide in the M3 muscarinic acetylcholine receptor have also been reported (76). Similarly, a number of studies have remarked that T cells isolated from the synovial fluid of patients with RA show signs of oligoclonal expansion, and may be found at higher frequencies than in peripheral blood, leading to a tempting inference that these expanded clones might be autoreactive (77)(78)(79). Some validation studies have also identified specific autoantigens, such as cartilage proteoglycan (80) and citrullinated tenascin-C (81), which may be recognized by these autoreactive clones. Unfortunately, because synovial fluid of RA patients may also contain high concentrations of T cell-recruiting chemokines such as CXCL10 and CCL5, nonautoreactive memory cells expressing cognate chemokine receptors may also migrate to joints. As such, further work will be necessary to distinguish the identities of autoreactive and bystander memory clones. A summary of how inclusion of repertoire information may help guide analysis is depicted in Figure 1. Incorporation of TCR and BCR repertoire data into scRNAseq thus offers an exciting new lens for analyzing and understanding the pathogenesis of rheumatic diseases, and may shed light on the true behavior of autoreactive clones in rheumatic diseases. CHALLENGE 2: SPATIAL MAPPING OF scRNAseq DATA AND THE TISSUE SPECIFICATION OF PATHOGENIC IMMUNE CELL POPULATIONS The recent advent of spatially-aware transcriptomics has made it possible to gain a concurrent understanding of the tissue localization and transcriptomes of individual cells (82). While pioneering studies in spatial transcriptomics were somewhat limited in point resolution to small clusters of cells, novel improvements in barcode generation and seeding have now made single-cell analysis possible. In the context of rheumatic diseases, this technology holds considerable promise for enabling comprehensive analyses of the function and origin of tissueinfiltrating immune cell populations, and unveiling the identity and status of the stromal cells that they interact with. Spatial transcriptomics may prove to be especially useful for elucidating the mechanisms driving multi-organ involvement seen in patients with severe disease. For instance, SLE is known to cause systemic manifestations in an extraordinarily wide range of tissues spanning the entire body, and current studies of skin and renal manifestations. Single-cell mapping studies have further demonstrated that significant numbers of ISGhi immune cells may enter into tissue sites. However, current analyses of these infiltrating immune cells have been limited to transcriptome information without spatial location. A spatial mapping might be able to resolve the context in which these immune foci form, as well as the contents of each foci. One question of particular interest is whether all of these foci are indeed dependent on interferon signaling, and which communication signals from surrounding stromal tissue may serve to help/hinder their formation. For instance, immune foci have been found to be preferentially localized near certain types of cells, but not others, in the context of tumors (83,84), but it is currently unclear if similar biases in cell type distribution can be observed in rheumatic diseases. Resolving these questions, as highlighted in the FIGURE 1 | Single-cell Lineage Tracing of Autoreactive T and B cell Clones. Expanded T cell clones emerge from a pool of naive T cells (blue) with similar phenotypes and a highly diverse TCR repertoire (double bars of different colors representing distinct TCRs). Extrapolation from sequencing of peripheral blood has yielded estimates of 10E8 distinct TCRs among naive T cells in circulation alone. Under resting conditions, these naive T cells are expected to only undergo homeostatic proliferation, such that the numbers of cells belonging to a particular clone should not substantially increase. However, when these naive T cells encounter a presented cognate antigen, they can undergo activation, differentiation, and clonal proliferation (orange and red TCRs). As a consequence, formerly naive T cells will adopt novel functional characteristics (different cell colors representing distinct phenotypes). Notably, a single clone can differentiate into multiple different functional states that coexist at the same point in time, a process in which asymmetric cell division has been proposed to be critically involved. Some differentiated characteristics are may be somewhat unique to CD4+ (B cell help) or CD8 + (targeted killing) T cells, although these functions are not absolutely exclusive. Similar to T cells, B cells will also undergo activation/maturation, differentiation, and clonal proliferation from a pool of naive progenitors with high BCR sequence diversity. However, B cells also have the unique ability to undergo somatic hypermutation during the process of affinity maturation to modify their BCR sequence (different BCRs represented by unique, but similar, colors of the immunoglobulin symbol). As a result, a single BCR clone observed in plasmablasts may encompass multiple unique BCR sequences in naive B cells that have subsequently converged. Additional bioinformatics tools for BCR clustering and lineage tree tracing have been developed to help identify clones potentially linked by somatic hypermutation events. During an immune reaction, both circulating and tissue-resident cell populations will play significant roles. The formation of an inflammatory environment in tissue with high levels of cytokines can induce rapid recruitment of circulating immune cells. These recruited cells may then infiltrate into the tissue to carry out their disparate functions. These tissue-infiltrating immune cells will also interact with local residential populations of stromal and immune cells. Tissue-resident populations may also egress into circulation to carry out functions elsewhere in the body (with perhaps the most clearly |
defined example being the migration of tissue-resident dendritic cells bearing antigens into lymph nodes to initiate adaptive immune responses there). Animal model studies have demonstrated that both the recruitment of immune cells from circulation to a specific site and the egress of site-resident populations to distal tissues may begin in a manner of minutes following pathogen exposure. Complex interactions between resident and newly-infiltrating immune cells may then lead to the formation of histologically-observable structures, such as immune cell foci, in which T and B cells typically converge in an ordered manner to form a structure similar to germinal centers observed in secondary lymphoid organs (spleen and lymph nodes). Leading edges of immune cell infiltration may also be observed, in which some types of immune cells appear to be restrained from infiltrating into some portions of a tissue, or otherwise highly enriched near some stromal cell populations, but not others. These edges may form as a result of modulation by the local environment, in the form of inhibitory signals or other factors. Existing histological studies have widely documented the occurrence of immune cell foci in autoimmune rheumatic diseases, with the presence of these foci being a common diagnostic criterion for pSS. Leading edges have more commonly been observed in the context of tumors as a result of spatial transcriptomics analyses, but have yet to be clarified in the rheumatic diseases due to the lack of similar datasets thus far. schematic shown in Figure 2, may help answer key questions concerning stromal-immune cell interplay in rheumatic diseases. As reviewed above, histology-guided analyses in RA have already helped to identify differences in spatial distribution between Thy1+ and CD55+ FLS, and have found preferential accumulation of immune cells near Thy1+ FLS. A preliminary spatial transcriptomic study of RA synovium has also been conducted, and has suggested co-localization between CD8+ T effector memory cells and class-switched B cells, and segregation between T cells and plasma cells (85). Unfortunately, the spatial context of other immune and stromal cell types in synovial tissue and in the joint microenvironment remains unclear. For instance, stromal populations such as articular chondrocytes that maintain the cartilage matrix may also be involved in local inflammatory processes during RA progression (86), and osteoblasts have been shown to be receptive of immune cell stimuli (87). A fuller characterization of the spatiotemporal heterogeneity of these and other stromal cell types on a singlecell level might help to greatly expand current understandings of RA pathogenesis. Furthermore, spatial analyses may also help to compare newly-infiltrating immune cells with long-lived residential immune cells that can persist in many tissues. For example, the skin under normal conditions is expected to contain large numbers of Langerhans cells and memory T cells. These cells provide an important barrier function as a first line of defense against external pathogens, and have been conditioned to survive and function in the dermal environment. These residential cells can thus be expected to express unique transcriptome profiles that enable local adaptation, and can more readily respond to local insults by virtue of their localization (88). Traditionally, residential populations are expected to react against known pathogens, but autoreactive clones have also been shown to adopt residential characteristics in autoimmune diseases, particularly in the skin (89,90). However, it is not completely understood how or if the distribution and function of these residential cells may be modified by further infiltration of nonresidential immune cells. It is also unclear how cells with elevated ISG signatures may be distributed in the skin, or if immunosuppressant treatment to treat SLE will have a differential impact on these residential populations as compared to newly-infiltrating ISGhi cells as a result of differences in location and drug bioavailability. These results will also have significant ramifications for understanding side effects associated with immunosuppressant use, such as risk of infection. A spatially-guided study in these and other related scenarios may be of great use for answering these questions and clarify the true functional relevance of immune cell structures in afflicted tissues. CONCLUSION The application of scRNAseq into human rheumatic disease holds great promise for answering foundational questions regarding how autoimmune reactions are instigated and maintained in complex environments, as well as identifying novel targets for therapeutic intervention. Current efforts to map the atlas of immune cell types involved in each disease have already identified a number cell types and pathways as being of notable interest, particularly in the context of SLE. These mapping studies have also identified notable modifications in regional, tissueresident populations of both immune and non-immune cells as a result of disease progression that point to larger-scale remodeling caused by immune cell activity. If recently emerging technologies for repertoire and spatial mapping can be integrated into scRNAseq studies, a much more complete picture of auotreactive lymphocytes and tissue infiltration patterns can also be obtained. Combination of these assays with other emerging single-cell technologies in each sample, such as ATAC-seq (91) and proteomics (92), might also assist in generating even more comprehensive multi-omics data. Future studies taking advantage of these existing resources and newer technologies may help bring about breakthroughs in treatment for rheumatic diseases. Patterns of bronchial challenge testing in Canada Background Bronchial challenge testing (BCT) measures airway hyperresponsiveness; asthma guidelines recommend using BCT when symptoms manifest despite normal spirometry. Improper application of these guidelines commonly results in the misdiagnosis of asthma. Yet, statistics concerning BCT remain largely obscure. The current paper addresses this gap and explores how various health variables may elucidate adherence to asthma guidelines and patterns of BCT across Canadian provinces. Methods Using the Access to Information Act, medical financial claims for BCT (or equivalent procedures) were requested from each of the Canadian provinces and territories. Based on the available information (from provinces only), correlations between frequency of BCT claims and medical demographics (e.g., prevalence of respirologists, health expenditures) are reported. Results Controlling for population or for people with asthma, physicians from Québec claim four times more BCT per year than those in other provinces; physicians from Alberta close to eight-fold fewer. The number of respirologists per capita and BCT per capita correlated moderately, r(132) = 0.582, p < 0.001, [95% CI 0.421, 0.716]. Excluding “outliers” (i.e., British Columbia, Alberta, and Saskatchewan) greatly strengthened this correlation, r(87) = 0.930, p < 0.001, [95% CI 0.883, 0.958]. Discussion These findings demonstrate that provinces vary in their use of BCT. This result seems to stem, at least in part, from differences in the prevalence of respirologists. Interestingly, geographic region appears to wield a strong influence; in the correlation between number of tests and number of respirologists, physicians from Western provinces (i.e., Alberta, Saskatchewan, and British Columbia) administered fewer tests than their Eastern colleagues. Given the association between inadequate application of BCT and misdiagnosis of asthma, physicians should pay special attention to the Canadian guidelines when considering an asthma diagnosis. INTRODUCTION Existing asthma guidelines [1][2][3][4][5][6] recommend that physicians confirm the diagnosis of asthma with objective pulmonary function testing (PFT) measures [7], including, but not limited to, spirometry [8] and nonspecific bronchial challenge testing (BCT) [9]. This practice helps to reduce misdiagnosis and its societal and personal costs [10]. Without PFT, physicians risk over-diagnosing or under-diagnosing asthma by approximately 30% [11][12][13][14][15][16] and 20% [15][16][17][18], respectively. Consequently, an editorial in the Canadian Medical Association Journal [19] stated that "Failure to make the diagnosis of asthma objectively is unacceptable… Physicians who do not use spirometry for their asthma patients should not be managing asthma," (p. 1099), although some disagree [20]. And yet, the little available documentation concerning frequency of PFT suggests that Canada scantily meets its asthma guidelines [2]. One study [21] reported only 54% of individuals diagnosed with asthma ever had PFT; only 49% of general practitioners and 46% of physicians reported using PFT. In a study looking at over-diagnosis [12,22], only 49% of patients went through spirometry on the initial diagnosis of asthma. Two other studies revealed that only 43% and 52% of diagnosed individuals received PFT in Ontario [23] and Nova Scotia [13], respectively. Overall, these findings propose that Canadian physicians resort to PFT in a suboptimal fashion [24]. In the presence of symptoms but normal spirometry, Canadian guidelines [2] recommend measuring airway hyper-responsiveness via BCT [9]. A rough estimate of the percentage of people requiring BCT can be formed from the number of patients that show normal spirometry despite reporting respiratory symptoms (although alternative methods besides BCT can also be used to investigate asthma given normal spirometry, such as bronchodilator response [25], peak flow monitoring and trial of therapy [26], and sputum induction and fractional exhaled nitric oxide [27]). The NHANES III study [28], based on a US sample of more than 16,000 individuals, estimated that more than 85% of adults older than 24 years of age with respiratory symptoms (including diagnosis of asthma) will have normal lung function as measured by spirometry. Moreover, when specifically investigated for asthma, between 55% [29] and 67% [30] of children and approximately 34% [31] of adults older than 50 years will typically show normal spirometry. In one study [32] of adults with symptoms of obstructive pulmonary disease, 63% had normal spirometry (80% excluding chronic obstructive pulmonary disease); BCT later identified asthma in 47% of individuals with normal spirometry. Despite these abovementioned data concerning normal spirometry, to the best of our knowledge, the only source documenting BCT is a survey that relied on a small convenience sample [33]. This survey estimated the application of BCT in the province of Québec to approximate 5150 tests per year. Otherwise, no systematic data are readily obtainable on the use of BCT within Canada. This paper presents patterns of BCT across Canada, speculates on potential factors influencing its use, and contextualizes these trends in relation to the Canadian asthma guidelines. This work will contribute to bring a unique perspective to better understand, and eventually address, the misdiagnosis of asthma in Canada. METHOD Using the Access to Information Act, medical financial claims for BCT (or equivalent procedures; see Supplementary File 1 1 for an overview of how each province labels BCT) were requested from each of the Canadian provinces and territories (see Supplementary File 2 1 for exclusions and details on data format and handling). Time-series data were obtained from 9 provinces: spanning the most recent year on record (2014)(2015) and going back to the earliest year reporting was available (2000)(2001). Based on the available information (see Supplementary File 3 1 ) an exploratory effort was performed to: (i) examine national and provincial trends and frequencies of (insurance) reimbursement records for BCT; (ii) normalize numbers of BCT claims per 10,000 people based on population estimates from Statistics Canada [34] and per 1000 individuals with asthma [35]; and (iii) correlate the number of BCT claims with the prevalence of professionals (i.e., respirologists, allergists, internists, and general practitioners) based on estimates from the Canadian Medical Association Masterfile [36]. The correlation between the number of BCT claims and various health variables were also investigated (e.g., population with asthma, location of physician training, and health expenditures). For all computations, SPSS version 24 and bootstrapping (resampled 2000 times) were used to obtain confidence intervals for all correlations. Bootstrapping is a commonly used technique that simulates multiple potential samples based on the available data to provide estimations that do not rely on the classical assumptions of statistical inference. Could BCT be over-prescribed? From consultations with asthma experts who harbour decades-long operational insights into the diagnosis of asthma in Canada To this aim, an informal, "back-of-the-envelope" calculation was performed using simplified assumptions-to be taken with a pinch of salt. To form a rough provincial estimate of the percentage of asthmatics who had BCT over the past 15 years (the period of available data), the total number of claims throughout 2000-2014 was divided by the number of people with asthma in 2014 (the most comprehensive figures available from Statistics Canada) [35] for each individual province. In sum, this procedure highlights the total number of BCT claims as a percentage of the number of asthmaticsa rough index to the number of asthmatics who received BCT. Figure 1 displays data for the number of medical claims for BCT. Two provinces markedly contrast with the others when controlled for population or alternatively for population with asthma: physicians from Québec perform four times more tests per year (the highest ratio), whereas physicians from |
Alberta perform nearly eight-fold fewer tests (the lowest ratio). Table 1 depicts correlations and confidence intervals of BCT claims per capita. Correlation analyses revealed a moderate relationship between number of respirologists and BCT claims (Model 1 in Table 1 and Figure 2). A visual inspection of Figure 3 suggests that Alberta, British Columbia, and Saskatchewan differ considerably from the rest of the other provinces in terms of their BCT and respirologist ratios. A second, exploratory correlation analysis without these provinces was therefore performed (Model 2 in Table 1), which greatly strengthened the correlation between BCT claims and respirologists. In general, these "outliers" seem to cluster on the West coast ( Figure 4). Geographic region only seems to meaningfully modulate the correlation between BCT claims and respirologists (Table 1). Finally, Figure 5 shows clear differences among provinces in the estimated percentage of asthmatics with BCT over the past 15 years. DISCUSSION Compared with other provinces, the number of tests in Québec is four times higher; in Alberta, it is eight-fold lower-acknowledging population size and asthma rates. The higher rate of Québec respirologists (per capita) may partly explain this higher rate of BCT. However, Canada's geography seems to moderate the correlation between the number of tests and respirologists: there were fewer tests per respirologist in the West (i.e., Alberta, Saskatchewan, and British Columbia, with the exclusion of Manitoba). Other variables-including numbers of allergists and internists, location of completion of medical training, population, percentage of urban population, population density, and health expenditures-also correlate with the number of tests, albeit more weakly. Moreover, the current interpretation of these findings alongside the ratios of BCTs to asthmatics, as illustrated in Figure 5, suggest that Québec and Ontario aside, Canada may be under-utilizing BCT. How many tests? Published reports document that between roughly 60% [28] and 90% [32] of adults show normal spirometry despite respiratory symptoms. Assuming Canada manages asthma in congruence with the asthma guidelines [2], the proportion of asthmatics getting BCT should loosely match the percentages of these reports. Between 2000 and 2014 it is estimated (see Figure 5) that the equivalent of nearly 60% of asthmatics went through BCT in Québec, whereas in Ontario this number was just a hair over 30%. These two provinces contrast with the rest of Canada that all lie under the 25% threshold. In addition, medical specialists are more likely to recommend PFT compared with general practitioners [23,24,42]. This paper shows that the use of BCT relates to the number of respirologists, allergists, and internists, but not to the number of general practitioners; however, most Canadian provinces seem to shy away from BCT. Overall, this trend echoes an overarching tendency to discount the guidelines of the Canadian Asthma Consensus: for example, to confirm the diagnosis of asthma objectively given that almost 50% of patients reported having never received a lung function test [21]. Ignoring asthma guidelines may have important consequences. One study [10] estimated that the improper diagnosis of asthma due to a lack of objective PFT could cost Canadians more than $275 million over a 50-year period. There is also some evidence that provinces differ in their misdiagnosis rates. Data from a recent survey of 10 Canadian cities [12,22] suggest that over-diagnosis of asthma may be higher in western provinces (Manitoba, 49%; British Columbia, 48%; Alberta, 44%) and lower in eastern provinces (Ontario, 39%; Nova Scotia, 29%; Québec, 24%). 2 Furthermore, this regional distribution seems to follow the general pattern identified earlier regarding the relation between respirologists and BCT ( Figure 4). These findings raise the possibility that regional variation in BCT translates to similar variations in misdiagnosis rates. Potential reasons for variation in BCT in Canada Why provinces differ in their use of BCT remains unclear; a wide array of complex factors and interactions is likely at play (see Table 1 for examples). For instance, to obtain financial compensation in Québec, workers need to confirm the diagnosis of occupational asthma with objective tests [43,44]. Moreover, Québec reportedly possesses a higher concentration of occupational and exercise-induced asthma specialists (Dr. Sandra written communication, July 2015)-although this does not mean they necessarily disagree with the objective confirmation of asthma using alternative methods such as spirometry and peak flow measurement. As a case in point, the use of spirometry depends on whether physicians believe it necessary to pose an accurate diagnosis [24]. This paper shows that completing medical training outside of Canada relates to fewer BCT claims; Québec physicians are correspondingly much more likely to obtain their medical training in Canada than physicians from other provinces [35]. Perhaps the influence of prominent Canadian experts (e.g., Freddy Hargreave [45,46] and his protégées [47,48], and the Montréal group, including the Meakins-Christie Laboratories [49]) lingers as a factor still governing the higher BCT rates in eastern Canada and Québec today. Limitations The following overarching caveats apply to this study. (i) Nova Scotia and Newfoundland and Labrador provided partial or complete omissions of data (see Supplementary file 2 1 ). (ii) The term "BCT" is hardly uniform across provinces (see Important limitations also apply to the procedure for estimating the number of asthmatics who had BCT. First, provinces with a greater proportion of individuals followed for occupational asthma (e.g., Québec) may have inflated numbers, assuming these individuals undergo BCT more than once. Furthermore, the total number of tests only covers the past 15 years, whereas asthmatics may have undergone BCT before this period, hence this analysis likely under-estimates the actual number of asthmatics who received BCT. However, the procedure used estimates from Statistics Canada for the number of people who report having received a diagnosis of asthma by their physician [38]-these individuals may or may not have received BCT. This last limitation may actually provide more support for this estimation because including cases in which BCT excluded the diagnosis of asthma would further decrease the percentages of suspected asthmatics who had BCT, and reinforce the hypothesis that most provinces under-use BCT. And yet, this procedure provides but a rough estimation of the number of asthmatics who received BCT. CONCLUSION This paper shows substantial interprovincial variations in the use of BCT. These findings expand our understanding of the diagnosis of asthma in Canada by highlighting the critical role of respirologists and the moderating effect of geographic region. A tentative analysis also suggests that certain provinces may be under-prescribing BCT, thereby increasing risks for misdiagnosis of asthma and associated health care costs. Overall, this investigation complements earlier research on the misdiagnosis of asthma [10][11][12][13][14][15][16][17][18][21][22][23][24] and contributes to a more systematic understanding of the use of objective testing in asthma. Future research should address the possibility of under-reporting in the current study, as well as the reliance on a rough approximation to determine how many individuals diagnosed with asthma underwent BCT. To detect and address the misdiagnosis of asthma specifically, the respiratory medicine community would stand to benefit, over and beyond documenting PFT, from a more precise and thorough assessment of clinicians' adherence to the Canadian guidelines when diagnosing and managing asthma. InbR, a TetR family regulator, binds with isoniazid and influences multidrug resistance in Mycobacterium bovis BCG Isoniazid (INH), an anti-tuberculosis (TB) drug, has been widely used for nearly 60 years. However, the pathway through which Mycobacterium tuberculosis responds INH remain largely unclear. In this study, we characterized a novel transcriptional factor, InbR, which is encoded by Rv0275c and belongs to the TetR family, that is directly responsive to INH. Disrupting inbR made mycobacteria more sensitive to INH, whereas overexpressing inbR decreased bacterial susceptibility to the drug. InbR could bind specifically to the upstream region of its own operon at two inverted repeats and act as an auto-repressor. Furthermore, InbR directly bind with INH, and the binding reduced InbR’s DNA-binding ability. Interestingly, susceptibilities were also changed by InbR for other anti-TB drugs, such as rifampin, implying that InbR may play a role in multi-drug resistance. Additionally, microarray analyses revealed a portion genes of the inbR regulon have similar expression patterns in inbR-overexpressing strain and INH-treated wild type strain, suggesting that these genes, for example iniBAC, may be responsible to the drug resistance of inbR-overexpressing strain. The regulation of these genes by InbR were further assessed by ChIP-seq assay. InbR may regulate multiple drug resistance of mycobacteria through the regulation of these genes. Scientific RepoRts | 5:13969 | DOi: 10.1038/srep13969 Staphylococcus aureus 9 , and BmrR in Bacillus subtilis 10 . These regulators inhibit or stimulate the expression of their target genes to contribute to bacterial drug resistance. TetR is a large family of transcriptional regulators that contains a conserved helix-turn-helix DNA-binding domain and a C-terminal ligand regulatory domain. These regulators usually serve as repressors and are widely distributed among bacteria 11 . The most frequently characterized function of TetR proteins is regulation of efflux pumps and transporters, which are involved in antibiotic resistance and toxic chemical compound tolerance [12][13][14][15] . For example, the TetR of E. coli controls expression of the gene encoding a tetracycline efflux pump responsible for drug resistance 16 . TetR binds to the promoters of efflux pump genes and is regulated by a plethora of ligands that can cause protein conformational changes and eradicate protein binding, thereby relieving its repression of transcription 17 . In M. tuberculosis, the transcription factor Rv3066, which belongs to the TetR family, was recently found to bind specific co-activator drug molecules (ethidium) and to regulate bacterial drug resistance 18 . Rv3066 can directly bind ethidium and can de-repress the expression of a multidrug transporter operon, mmr 19 . The genomes of both M. tuberculosis and M. bovis BCG encode a large group of TetR family regulators 20,21 . However, transcription factors that can directly bind the first-line anti-TB drugs remain uncharacterized, and the molecular network through which the bacteria respond to the drugs remain largely unclear in M. tuberculosis and related mycobacterial species. M. bovis BCG is a vaccine strain 21 that has been used as a model strain for studying gene regulatory mechanisms in mycobacterial species, including the pathogenic strain M. tuberculosis. In this study, using M. bovis BCG as a model strain, we screened and characterized InbR, the first INH-binding transcriptional factor that regulates mycobacterial susceptibility to multiple drugs. The results showed that InbR functions as a repressor, and while its overexpression decreased bacterial susceptibility to INH, and its disruption led to supersensitivity of M. bovis BCG to INH. InbR was found to regulate the expression of multiple genes, including the iniBAC operon. Furthermore, we proposed an INH-inducible sequential signal cascade, in which InbR functions as a master regulator and plays an important role in the regulation of mycobacterial susceptibility to multiple anti-TB drugs. InbR positively regulates INH resistance in M. bovis BCG. Only a few transcription factors have been reported to contribute to mycobacterial drug resistance to date. To identify potential regulators that contribute to M. tuberculosis INH resistance, we screened a transcriptional factor library by spotting recombinant M. bovis BCG strains, in which the corresponding transcriptional regulator was overexpressed by the constitutive strong promoter hsp60 21 , on plates containing INH (2 μ g/ml). First, all the annotated putative transcriptional regulators (approximately 300 ORFs) of M. tuberculosis were cloned in a batch into the overexpressing plasmid pMV261. Secondly, each recombinant strain was spotted onto 7H10 agar plates that contained 2 μ g/ml INH. As a result, those strains that were more resistant to INH were able to grow and thus were identified as primary candidates. To avoid eventual random mutations that may confer drug resistance, the assays were repeated three times for primary candidates, and finally the drug susceptibility of recombinant strains were attributed to the overexpression of the candidate genes. A TetR family transcription factor encoded by Rv0275c, designated as InbR, was isolated. As shown in Fig. 1A, we measured the growth of inbR-overexpressing and pMV261 empty vector M. bovis BCG strains on the surface of a solid agar medium with or without INH. When a gradient of different concentrations of mycobacterial strains was spotted on the surface of a solid agar medium without INH, similar bacterial lawns were observed for both the inbR-overexpressing and pMV261 empty vector strains (Fig. 1A, left panel). By contrast, while the same concentration gradient of mycobacterial strains were spotted on a plate containing 2 μ g/ml INH, the bacterial lawn for the pMV261 empty |
vector BCG strain was smaller than that for the inbR-overexpressing strain, indicating that the strain overexpressing inbR was more resistant to INH than the wild-type strain (Fig. 1A, right panel). This finding suggested that InbR was potentially involved in the regulation of INH-drug resistance in M. bovis BCG. Orthologs of Rv0275c (InbR) were identified based on sequence similarity and the conservation of adjacent genes. Strikingly, InbR and its orthologs were found to be transcribed divergently from a hypothetical protein (Fig. 1B). The Rv0275c region is highly conserved within M. tuberculosis H37Rv, M. tuberculosis H37Ra, and M. bovis BCG (100% amino acid identity over the entire length of the protein), but not in Mycobacterium smegmatis (53% amino acid identity). The gene inbR encodes a 241-residue protein containing a typical TetR_N superfamily domain within an AcrR domain (Fig. 1C), which suggests that InbR belongs to the TetR/AcrR family of transcription factors. We further assayed the regulatory effect of InbR on the growth of M. bovis BCG in response to INH by determining mycobacterial growth curves. Prior to this assay, the M. bovis BCG inbR-deleted mutant strain (BCG/Δ inbR) was obtained (Fig. S1), together with the complementary strain (BCG/Δ inbR comp). As shown in Fig. 2, no obvious difference was observed in the growth of the pMV261 empty plasmid and inbR-overexpressed BCG strains in 7H9 medium in the absence of drugs ( Fig. 2A, left panel). However, compared with the pMV261 empty plasmid strain, the inbR-overexpressed BCG strain grew significantly better than the pMV261 empty plasmid strain in 7H9 medium containing 1 μ g/ml of INH ( Fig. 2A, right panel; p < 0.05). Without INH, the growth of the inbR-deleted strain (BCG/Δ inbR) and wild type (BCG/WT) have similar growth curves (Fig. 2B, left panel). With 0.1 μ g/ml of INH, the growth of the inbR-deleted strain was significantly inhibited compared with that of the wild type (Fig. 2B, right panel). Additionally, this type of inhibition can be complemented in the complemented strain (Fig. 2C). Moreover, overexpression of inbR decreased the INH susceptibility of the M. tuberculosis H37Ra strain as well (Fig. S2). These results are consistent and indicated that InbR positively regulates INH resistance in M. bovis BCG. InbR recognizes a palindromic motif and specifically binds to its promoter as an auto-repressor. In mycobacteria, many TetR family transcriptional factors possess an auto-regulating function. We used an electrophoretic mobility shift assay (EMSA) to examine the binding of the InbR (Rv0275c) protein to the upstream region of its own operon in vitro. As shown in Fig (Fig. 3B, lane 2). By contrast, the pre-immune serum failed to precipitate significant amounts of DNA (Fig. 3B, lane 3). In addition, Rv3430cp, the promoter of an unrelated gene, used as negative control, could not be recovered with InbR antiserum. These findings strongly suggested that InbR could bind with its own promoter region. By using β -galactosidase assays, we further characterized that InbR functions as an auto-repressor (Fig. S3). We characterized the DNA binding motif of InbR by Dye primer-based DNase I footprinting assays. As shown in Fig. 3C, when increasing amounts of InbR protein (0-2 μ M) were co-incubated with DNaseI, the region around TGCCGCTAATTATGGAAACACCTGTATCCTGATATTGGCCGG was obviously protected on the coding strand. The protected DNA region extended from position − 72 to − 30 in the DNA strand (Fig. 3C). A palindromic motif formed by two inverted repeats partially matched, which was separated from each other by two nucleotides, was found in this region. Further EMSA assays confirmed the significance of the motif for specific recognition by InbR. DNA substrate mutants were synthesized ( Fig. 3C) and EMSA assays were conducted (Fig. 3D). As shown in Fig. 3D (right panel, Lane 6-10), InbR lost the ability to bind with Rv0275cp4 in which the two inverted repeats were replaced by random sequences. By contrast, replacement of either part of the repeat or the interspaced sequence did not abolish their interaction (Fig. 3D, lane 6-15, lane [21][22][23][24][25], although the binding was a little bit weaker compared with that of inherent Rv0275cp1. These results suggested that the binding of InbR may be not very precise and a flexible and partial mismatch is allowed. In conclusion, InbR is an auto-repressor and the auto-regulation of InbR relies on a palindromic sequence motif. InbR directly binds INH and the binding represses its DNA-binding activity. As far as we know, overexpression of inbR increase INH resistance. On this basis, we further examined whether INH induced the expression of inbR in M. bovis BCG by quantitative RT-PCR (qRT-PCR). M. bovis BCG strains were grown until the logarithmic growth phase (OD 600 = approximately 0.6) and various concentrations of INH (0.5 μ g/ml, 1 μ g/ml, and 2 μ g/ml) were added to the medium. Cells were harvested 24 h later and qRT-PCR was performed. We found that inbR induction was increased by 1.2-fold, 1.67-fold, and 4.08-fold under INH concentrations of 0.5 μ g/ml, 1 μ g/ml, and 2 μ g/ml, respectively (Fig. S4). The results implied that high concentrations of INH will significantly induce the expression of inbR in vivo and interactions between InbR and INH are possibly present. EMSA assays were subsequently conducted to check the possible interaction between INH and InbR. As shown in Fig Consistently, no response was observed when the same amount of unrelated small molecules, such as guanosine-5′ -triphosphate (GTP) or cyclic diguanylate monophosphate (c-di-GMP), was passed over the His-InbR-immobilized NTA chip (Fig. 4B, right panel). The results showed that InbR directly binds INH. In addition, SPR experiments were conducted with an immobilized promoter DNA on a chip and InbR in different conditions. InbR promoter DNA was immobilized on the SA chip, and proteins with or without small molecules were passed over. As shown in Fig. 4C, when increasing concentrations of the InbR protein (0.5 μ M to 2 μ M) were passed, corresponding increases in response values were observed (Fig. 4C, left panel). By contrast, unrelated Rv0135c protein did not show any response when passed over the chip. In addition, when InbR was treated by increasing concentrations of INH (20 μ M to 80 μ M INH co-incubated with 0.4 μ M InbR) prior to use, corresponding decreases in response values were observed (Fig. 4C, right panel). By contrast, with identical treatment by an unrelated small molecule GTP, the response value did not change (Fig. 4C, right panel). These results jointly indicated that InbR binds INH and the binding represses its DNA-binding ability. The function of InbR is not INH-specific and the mode of action is complicated. Although InbR does not bind either EMB or RIF, relationships between InbR and the drugs still exist. Minimal inhibitory concentrations (MIC) of wild type, inbR-overexpressing, inbR-deleted and complementary strains were tested with INH, RIF, EMB and mitomycin C (MMC). The MICs of the inbR-deleted strain were all lower compared with those of the wild-type strain (Table 1). By contrast, the MICs of the inbRoverexpressing strain were all higher than that of the wild-type (Table 1). Additionally, growth of the inbR-deleted strain was significantly inhibited by either EMB or RIF at a low concentration in which the wild type strain grew very well (Fig. S6). That is to say, disrupting the inbR gene made the M. bovis BCG strain more sensitive to multiple drugs, whereas overexpressing inbR decreased the susceptibility, and the results suggested that the function of InbR is not INH-specific. To further elucidate the mechanism by which InbR regulates drug resistance, we performed microarray analyses on inbR-overexpressing and INH-treated wild type M. bovis strains. While comparing the results with that of the non-treated wild type strain, many genes that had consistent expression profiles were identified (Table 2 and Table S5). On the one hand, ribosomal proteins, including S18, L9, S19, L22, S3, L16, L29, S17 and L30, iniBAC and several hypothetical proteins were upregulated in both inbR-overexpressing and INH-treated strains. On the other hand, a large number of metabolic enzymes, including pqqE, lldD1, echA7, gltA1, fadE12, accA2, accD2, gltA1, narG, narH and narJ, and regulatory proteins such as sigI, pfkB, devR were all downregulated. Moreover, there were also many genes with expression profiles that are different in inbR-overexpressing and INH-treated strains. For example, argJ, argB, argD, argE and argR are only upregulated in the former strain. These results suggest InbR uses a complex network to conduct multiple levels of regulation. We performed qRT-PCR assays to verify the differential expression of several important genes in the inbR-overexpressing and the inbR-deleted strains. On the one hand, expression of Rv0081 and dosR were downregulated (0.3-fold or 0.02-fold) in the inbR-overexpressing strain (Fig. 6A), and upregulated (2.5-fold or 3.8-fold) in the inbR-deleted strain (Fig. 6B). On the other hand, the expression of the groEL1, groEL2 and iniBAC operons was upregulated in the inbR-overexpressing strain (Fig. 6A), and downregulated in the inbR-deleted strain (Fig. 6B). These qRT-PCR results were consistent with the results of the microarray and showed that whenever a gene is a direct or indirect target, there is regulation by InbR. In summary, InbR regulates bacterial susceptibility to multiple anti-TB drugs in M. bovis BCG, via regulation of a large number of genes. Discussion The molecular network through which M. tuberculosis responds to anti-TB drugs and the intrinsic regulatory mechanism underlying mycobacterial INH resistance remain largely unclear. In the present study, we report a TetR family regulator; i.e., InbR, which interacts directly with the first-line anti-TB drug INH in M. bovis BCG. Overexpression of inbR decreased mycobacterial INH susceptibility, whereas disrupting inbR made the mycobacteria supersensitive to multiple anti-TB drugs. Most interestingly, we provide evidence that INH can directly bind to InbR and negatively affects the regulator's DNA-binding ability. Thus, we have uncovered a novel mechanism underlying regulation of mycobacterial susceptibility to INH. The TetR/AcrR family regulators usually function as repressors and are widely distributed among many bacteria 11 . Most of these proteins are involved in the regulation of drug resistance, biosynthesis of antibiotics, osmotic stress, and bacterial pathogenicity 11 . The AcrR operon of E. coli contains three genes; namely, acrR, acrA, and acrB, the last two of which are multidrug resistant efflux pumps 22,23 . By comparison, InbR has a typical AcrR domain but, unlike in E. coli, is encoded in a single operon. Targets of InbR were, therefore, going to be elucidated. In the present study, we provided evidence to show that InbR acts as an auto-repressor and regulates the expression of a large number of genes. Among these genes, many overlapping genes of InbR regulon genes and INH responsive genes were identified (Table 2 and Table S5). INH responsive genes such as iniBAC, have been shown to be involved in tolerance to multiple anti-TB drugs 5 . Therefore, similar expression profiles for these genes may also give multiple drug resistance to inbR-overexpressing strains. In the InbR regulon, some are direct targets, while the others are indirect targets. Many genes are not drug specific genes in mycobacteria but play roles in multiple stress adaptation. In addition, a ChIP-seq assay revealed that direct targets of InbR are enriched in the GO term small molecule binding. This result implied that the binding of small molecules play an important role in InbR's mode of action. Therefore, other types of small molecules may be preventing targets of InbR regulon genes as well. Additionally, this could be an acceptable explanation for InbR INH-nonspecific functions. As has been revealed by microarray analysis and qRT-PCR results, InbR could strongly induce the expression of the operon iniBAC (Table 2 and Fig. 6). The iniBAC operon encodes transport-related genes in mycobacteria and confers multiple anti-TB drug tolerance to M. tuberculosis and M. bovis BCG 5 . It is believed that the effect of InbR on multidrug resistance in M. bovis BCG are, mainly or partially, due to the overexpression of the iniBAC operon. Interestingly, a subsequent ChIP-seq assay revealed a high quality peak (qvalue = 1.4E5) downstream of iniBAC. Therefore, the regulation of iniBAC is distinct; for example, by antisense RNA. Moreover, InbR may also regulate iniBAC indirectly. For example, five regulators were reported as regulators for iniBAC in TBDB (http://TBDB.org, Rv0081, Rv0967, Rv1353c, Rv1956 and Rv2250c 37,38 ), in which Rv1956 and Rv1353c were the direct targets of InbR (ChIP-seq peaks found upstream, Table S5), while Rv0081 and Rv0967 were the indirect targets of InbR |
(down-and upregulated in inbR-overexpressing strain, respectively, Table S5). Although the details were not very clear, it is logical to conclude InbR may regulate iniBAC expression through direct and/or indirect pathways. One interesting finding is that InbR could regulate susceptibilities of multiple drugs. As we know, drug resistance in M. tuberculosis results primarily from acquisition of chromosomal mutations in genes encoding the drug target proteins, such as katG and inhA [24][25][26] . Nonetheless, gene expression changes were also thought to introduce drug resistance. For example, downregulation of katG was found to be highly associated with isoniazid resistance in M. tuberculosis 27 . Moreover, whiB7 was believed to be one of the main causes of mycobacterial intrinsic drug resistance 28 . In general, the affection for a transcriptional regulator to drug resistance is quite different from an enzymatic gene such as katG. The effect of katG follows a very simple rule: the activation of pro-drug INH. Inactivation of katG leads to defects Continued in INH activation thus introducing drug resistance. By contrast, the effect of a transcriptional regulator would be much more complex. In living cells, regulators set up a network and work jointly, which is flexible and stable. Omitting a single regulator that is not lethal may not affect the function of such a network. In this study, we found InbR could bind INH and positively regulate drug resistance in mycobacteria. Molecular mechanisms were also investigated and several clues were found; however, the biological role for this novel regulator InbR is still not fully understood and further studies are needed. In conclusion, this study showed that the TetR-family transcriptional regulator InbR binds isoniazid and influences multidrug resistance in M. bovis BCG. Experimental Procedures Strains, plasmids, enzymes and reagents. E. coli BL21 (λ DE3) cells and pET28a were purchased from Novagen (Darmstadt, Germany) and were used to express proteins. Restriction enzymes, T4 ligase, modification enzymes, DNA polymerase, dNTPs, and all antibiotics were obtained from TaKaRa Biotech (Shiga, Japan). PCR primers were synthesized by Invitrogen (Carlsbad, USA). Ni-NTA (Ni 2+nitrilotriacetate) agarose was purchased from Qiagen (Hilden, Germany). 7H9 and 7H10 broths were purchased from Becton, Dickinson Company (New Jersey, USA). Antibodies were obtained from the Wuhan laboratory animal center of CAS (Wuhan, China). Table 2. Expression patterns of 20 featured gene clusters in inbR-overexpressed and INH induced strains. * log 2 transformed expression values in microarray analysis. ** ChIP-seq peaks identified in inbR overexpressed strain. Up, upstream of the operon or gene; Dn, downstream of the operon/gene; In, inside of a gene; NA, peak is not available. Peaks are visualized in Figure 5 and Figure S7. with the recombinant plasmid, were grown in 200 ml of LB medium up to OD 600 of 0.6. Protein expression was induced by the addition of 0.3 mM isopropyl β -D-1-thiogalactopyranoside (TaKaRa). Harvested cells were resuspended and sonicated in binding buffer (20 mM Tris-HCl, pH 8.0; 100 mM NaCl; and 10 mM imidazole), and the lysate was centrifuged at 10,000 × g for 30 min. The cleared supernatant was loaded onto the affinity column. The column-bound protein was washed with buffer (20 mM Tris-HCl, Table 2) that have been listed in Table 2 ChIP-PCR and ChIP-seq assays. Chromatin immunoprecipitation (ChIP) was performed as described previously 29 with modifications. M. bovis BCG cells were grown in 100 ml 7H9 medium, fixed with 1% formaldehyde, and stopped with 0.125 M glycine. Crosslinked cells were harvested and resuspended. The sample was sonicated on ice and the average DNA fragment size was determined to be approximately 0.5 kb. A 100 μ l sample of the extract was saved as the input fraction, whereas the remaining 900 μ l was incubated with 10 μ l of antibodies against corresponding proteins or preimmune serum under rotation for 3 h at 4 °C. The complexes were immunoprecipitated with 20 μ l 50% protein A agarose for 1 h under rotation at 4 °C. The immunocomplex was recovered by centrifugation and resuspension in 100 μ l TE (20 mM Tris-HCl, pH 7.8; 10 mM EDTA; and 0.5% SDS). Crosslinking was reversed for 6 h at 65 °C. The DNA samples of the input and ChIP were purified, resuspended in 50 μ l TE, and analyzed by PCR with Platinum Taq (Invitrogen). The amplification protocol included one denaturation step of 5 min at 95 °C, then 32 cycles of 1 min at 95 °C, 1 min at 60 °C, and 1 min at 72 °C. For the ChIP-seq assay, ChIP-enriched DNA was obtained similarly, except that the fragment size was approximately 300 bp, which is the desired size for Illumina short DNA library construction. Sequencing libraries were constructed following the manufacturer's instruction and then subject to Illumina HiSeq2000/2500 instruments (BGI, Shenzhen, China). Short reads were aligned using Bowtie2 30 and peaks were called by MACS 31 . Peaks were annotated using Bioconductor toolbox (http://bioconductor.org). Dye primer-based DNase I footprinting assay. The DNase I footprinting assay was performed as previously described 32 . A 420-bp fluorescently labeled DNA fragment that encompassed bases − 200 to + 200 of the translational start site of Rv0275c was generated by PCR amplification. The fluorescently labeled probe was subjected to the same binding reaction as in EMSA. Then, 0.0025 U of DNase I was added and incubated for 5 min at room temperature. The digested DNA fragments were purified. The samples were analyzed with the 3730 DNA analyzer coupled with a G5 dye set using an altered default genotyping module that increased the injection time to 30 s and the injection voltage to 3 kV. The 420-bp fragment was sequenced using special primers in the Thermo Sequenase Dye Primer Manual Cycle Sequencing Kit (USB, Inc., Cleveland, OH, USA) following the manufacturer's instructions. Electropherograms were analyzed and aligned using the GENEMAPPER software (version 4.0, Applied Biosystems). Microarray analysis. Microarrays used in this study consisted of 15,744 60-mer probes, which were synthesized in situ by Agilent Technologies. The probes were designed based on the genome sequences of M. bovis BCG Pasteur_1173P2_uid58781 (GenBank accession numbers: NC_008769) and covered 3934 ORFs. Each probe was repeated thrice on the array. The inbR-overexpressing M. bovis BCG strain, M. bovis BCG wild-type strain, and INH-treated strain (M. bovis BCG wild-type strain grown on exponential phase OD 600 ≈ 0.8 and treated with 0.5 μ g/ml INH for 24 h) grown on exponential phase OD 600 ≈ 1.2 were harvested. Total RNA was extracted and purified using an RNeasy mini kit (Cat. #74106, QIAGEN, GmBH, Germany) following the manufacturer's instructions. RNA integrity was determined by utilizing RNA integrity number (RIN) generated using an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, US). Total RNA was amplified and labeled by Low Input Quick Amp Labeling Kit, One-Color (Cat. #5190-2305, Agilent Technologies) following the manufacturer's instructions. Labeled cRNA (complementary RNA) were purified using the RNeasy mini kit. Each slide was hybridized with 600 ng Cy3-labeled cRNA using a Gene Expression Hybridization Kit of Agilent Technologies (Cat. #5188-5242) according to the manufacturer's instructions. After 17 h of hybridization with 15744 60-mer probes, slides were washed in staining dishes (Cat. #121, Thermo Shandon, Waltham, MA, US) with Gene Expression Wash Buffer Kit (Cat. #5188-5327, Agilent Technologies) following the manufacturer's instructions. Slides were scanned using an Agilent Microarray Scanner (Cat. #G2565CA) with default settings; Dye channel: Green; Scan resolution = 5 μ m; and PMT = 100% and 10%, 16 bit. Data were extracted with Feature Extraction software (ver. 10.7, Agilent Technologies). The raw data were normalized using the Quantile algorithm in the Gene Spring software (ver. 11.0, Agilent Technologies). Normalized microarray expression data deemed significant (P ≤ 0.05) from the InbR-overexpression M. bovis BCG or BCG exposed to INH were selected, and the genes with fold change > 2.0 were selected for further analysis. Quantitative real-time PCR. Isolation of mRNA and cDNA from mycobacterial strains was performed as described previously 33 . For real-time PCR analysis, gene-specific primers (Table S4) Surface plasmon resonance (SPR) analysis. SPR analysis was carried out in a Biacore 3000 instrument (GE Healthcare) with nitrilotriacetic acid (NTA) and SA sensor chips as described previously 35,36 . The assays were performed at 25 °C. For the binding of INH with proteins, a His-tagged protein was immobilized onto NTA chips at densities of approximately 1,200 response units (RU). INH was used as the ligand and was diluted in HBS buffer (10 mM HEPES, pH 7.4; 150 mM NaCl; 50 μ M EDTA; 5 mM ATP; and 0.005% BIAcore surfactant P20) at concentrations of 8 nM, 40 nM, and 200 nM, and injected at 10 μ l/min for 5 min. GTP was substituted for INH in the negative control. An overlay plot was produced using BIAevaluation 3.1 software to depict the interaction between INH and proteins. To assess the binding of DNA with proteins, biotinylated Rv0275cp probes were immobilized onto streptavidin (SA) chips at densities of approximately 200 (RU). His-tagged Rv0275c, Rv0135c or His-tagged Rv0275c-INH, and Rv0275c-GTP, were diluted in HBS buffer at concentrations of 1, 2, 4, or 4 μ M protein + 0.2, 0.4, and 0.8 μ M INH and injected at 10 μ l/min for 5 min. GTP was substituted for INH in the negative control. Determination of the MIC of anti-TB drugs. MIC determination was performed as previously described 2 . Briefly, M. bovis BCG/pMV261, inbR-deleted mutant strain and inbR-overexpression strain were grown to OD 600 of 1.0 and diluted to approximately 1 × 10 7 cfu·ml −1 . Then, 0.05 ml of the dilution was used to inoculate 5 ml of Middlebrook 7H9 media containing various concentrations (0-1.28 μ g ml −1 ) of four anti-TB drugs, namely, INH, RIF, EMB, and MMC. The cultures were incubated while shaking at 37 °C for 2 weeks. The MIC was calculated as the concentration of each drug that inhibited bacterial growth. Determination of mycobacterial growth curves and the effect of antibiotics. To determine mycobacterial growth curves and the effect of antibiotics, the recombinant strains were grown for a week in Middlebrook 7H9 media (supplemented with 10% albumin dextrose catalase, 0.05% Tween-80, and 0.2% glycerol) containing 30 μ g/ml Kan. Cells were cultured to an OD 600 between 1.5 and 2.0, and each culture was diluted (4:100) in 100 ml of fresh 7H9 broth. The cultures were then allowed to grow further at 37 °C with shaking at 200 rpm. When cells entered a log growth phase (OD 600 of approximately 0.4), the indicated concentration of each antibiotic was added. The cultures were then allowed to grow further at 37 °C with shaking at 120 rpm. Aliquots were obtained at the indicated times, and the cultures were plated on 7H10 medium (supplemented with 0.2% glycerol) to determine colony-forming units 33 . Demystifying “drop-outs” in single-cell UMI data Many existing pipelines for scRNA-seq data apply pre-processing steps such as normalization or imputation to account for excessive zeros or “drop-outs." Here, we extensively analyze diverse UMI data sets to show that clustering should be the foremost step of the workflow. We observe that most drop-outs disappear once cell-type heterogeneity is resolved, while imputing or normalizing heterogeneous data can introduce unwanted noise. We propose a novel framework HIPPO (Heterogeneity-Inspired Pre-Processing tOol) that leverages zero proportions to explain cellular heterogeneity and integrates feature selection with iterative clustering. HIPPO leads to downstream analysis with greater flexibility and interpretability compared to alternatives. Introduction Droplet-based single cell RNA-sequencing (scRNA-seq) methods have changed the landscape of genomics research in complex biological systems [1][2][3][4] by producing single-cell resolution data at affordable costs. In the state-of-the-arts protocols, a step called barcoding unique molecular identifiers (UMI) has been introduced to remove amplification bias and further improve data quality [5]. Some literature [6][7][8] suggests that barcoding leads to a different data structure from read count data structure but many tools remain to not acknowledge the difference between the count data produced with and without barcoding. Many pipelines have been built for scRNA-seq UMI data analysis. Despite subtle differences in these pipelines, the general order of a scRNA-seq analysis is as follows: quality control (filtering), cleaning (normalization, imputation, de-noising, batch-correction, etc.), feature selection which often involves dimension reduction, and downstream analysis such as clustering and lineage analysis [9]. In this paper, we do not discuss filtering and focus on the later three steps. First, the challenge of scRNA-seq data cleaning has |
led to the development of a wide array of tools. Some methods adjust for sequencing depths using size factors [10,11]. Some impute the reads directly using a zero inflated model, to reduce the noise from drop-outs [12]. Some try to de-noise the entire data set by fitting parametric models, where one example is sctransform that uses the residuals from negative binomial regression [13] and another example is SAVER that uses Poisson LASSO regression [14]. Despite the diversity of proposed methods, the general consensus has been reached to use one of the following distributions to model the counts: Poisson, negative binomial, or zero-inflated negative binomial distribution. Second, methods for feature selection have been less controversial. Most tools use some form of gene variance to mean ratio to identify genes that are highly dispersed, where the dispersion level is interpreted as a signal of biological heterogeneity [6,10,13]. Another less recognized approach is to use the zeros in the read or UMI counts; genes with inflated zeros are interpreted as biologically important signals [15]. Lastly, after data cleaning and feature selection, the pre-processed data will then be piped into downstream analysis tools for clustering analysis [16][17][18], trajectory inference [19,20], or differential expression analysis [21,22]. Currently, pre-processing and downstream analysis have been mostly considered as separate and consecutive steps [10,14,16,23]. Here, we present extensive analyses of publicly available UMI data sets that challenge most existing pre-processing tools' assumption, mainly that pre-processing is a necessary step before feature selection and downstream analysis. Our results suggest that clustering, or resolving the cell heterogeneity, should be the foremost step of the scRNA-seq analysis pipeline, not as part of the downstream analysis. Normalizing or imputing the data set before resolving the heterogeneity can lead to adverse consequences in downstream analysis. Adding to the arguments that the UMI data is much cleaner than the read count data [6,8], our analyses demonstrate that the simple Poisson distribution is sufficient to fully leverage the biological information contained in the UMI data if the cell-type heterogeneity has been appropriately accounted for. As a result, we provide a new perspective on scRNA-seq data analysis by integrating the pre-processing step and clustering, which was classified as part of the down-stream analysis. The proposed procedures have been implemented in software HIPPO (https://github.com/tk382/HIPPO). HIPPO leverages zero proportions to detect different levels of cell-type heterogeneity in each gene and can be particularly useful for low UMI data sets with excessive zeros, such as typical datasets generated from 10X protocols. Demystifying drop-outs We started by exploring zero detection rates in three UMI datasets generated by 10X protocols for both homogeneous and heterogeneous cell populations. Taking a subset of data in Zheng [3] as created in Freytag [17] as an example, we computed zero proportions, defined as the proportion of cells with zero counts per gene, across 15,568 genes, in CD19+ B cells, CD4+/CD25 regulatory T cells, and combined. The obtained statistics were plotted against gene-level average count and were compared with expected zero proportions under the Poisson, negative binomial, and zero-inflated negative binomial distribution, respectively (Fig. 1a). For a homogeneous cell population, we observe most genes align well with the expected curve under the Poisson assumption. Few genes can benefit from using the negative binomial model to account for extra dispersion from the Poisson, but our results strongly suggest that to model the drop-outs by introducing an extra zero-inflation component by the zero-inflated negative binomial distribution is Fig. 1 a Comparisons of zero proportion, gene variance, and CV as indicators for cellular heterogeneity in different UMI data sets. b Distributions of p values from likelihood ratio test for over-dispersion and zero-inflation. c Comparisons feature selection using likelihood ratio test and zero proportions. LR test tends to select the genes that have mean UMI count close to 0. d t-SNE plots of CD34+ cells in Zheng data, and relationship between zero proportions and gene means before (black) and after (colors) clustering of CD34+ cells. e Distributions of zero inflation in different PBMC data sets. The x-axis labels represent gene types from GENCODE annotations and the number of genes within each type unnecessary. For example, in Zheng dataset, 257 genes out of 5,568 genes would benefit from negative binomial modeling (p values pass Bonferroni criterion in likelihood ratio test at 0.05 type I error level), but no gene would benefit from extra zero inflation parameter. The p values are not calibrated to the uniform distribution because there are many genes that have UMI count of 1 in one cell and 0 in everywhere else, in which case p value is close to 1 (Fig. 1b). This result shows that drop-outs are within the range of natural Poisson sampling noise in UMI data for a homogeneous cell population, and they do not introduce excessive zero inflation, which is contradictory to prevalent opinions [11,16,24,25]. Extra zero inflation can be measured by comparing observed zero to expected zero counts under Poisson distribution within a homogeneous cell population (Methods). Through the following analysis, we show zero proportions are as effective measures for cell-type heterogeneity as other widely used alternatives, gene variance, coefficient of variation (CV), or dispersion parameter in negative binomial distribution [15]. It provides simplicity and interpretability in particular for data sets with low UMI counts and reasonable number of zeros, as zero-inflation is meaningless when no zero is observed. Analysis in multiple UMI data sets shows that zero proportions in most genes can be effectively modeled by the Poisson distribution, as more than 95% of absolute z values (Methods) are below 2. For mixed cell types, zero proportions considerably deviate from expected values under the Poisson model, as only less than 30% of the genes have z values below 2. This shows that the zero inflation test is an effective way to find genes that contribute to cellular heterogeneity. On the contrary, gene variance of mixed cell types does not always surpass those of a single cell type. In Zheng data, 62% of the genes had higher variance in pure naive cytotoxic cells than in mixed PBMC cells. On average, gene variance is similarly distributed for homogeneous and heterogeneous cell populations (Additional file 1: Table S2 and Figure S8). Therefore, the gene variance is rather more of a gene-specific characteristic while being less informative about the characteristics of the entire cell population. CV, on the other hand, suffers from an inherent numerical instability issue when gene mean is close to 0, because when mean is close to 0, CV estimates have high variability. Another popular option is to conduct model selection to assign genes to one of three candidate distributions of Poisson, NB, and ZINB, but measuring overdispersion also suffers from a similar problem in selecting biologically meaningful genes [6,26]. When we used statistics from likelihood ratio test and select top genes from the resulting statistics, the selected genes were different from those selected when we used zero proportion (Fig. 1c). For three data sets of [3,17,27] (median sequencing depth of 4371, 1298, and 2393.5 respectively), the likelihood ratio test selects genes that are overly focused on those with mean close to 0. Intuitively, the dispersion parameter scales with the square of the ratio of gene mean to the gene variance and in nature very similar to CV. These genes with very low mean are likely to have little information about the cells. Still, dispersion parameter can be more useful than zero-inflation statistic when the data set has high UMI counts, so deviance has been implemented in HIPPO as an alternative feature selection method (Additional file 1: Figure S9, S10). We expand the data to study all 68,579 cells from Zheng dataset [3]. When we aligned the zero proportions with the expected Poisson curve according to the provided cell type labels: CD14+ monocyte, CD19+ B, CD34+, CD4+ T helper2, CD4+/CD25 T Reg, CD45RA+/CD25-naive T, CD4+/CD45RO+ memory, CD56+ NK, CD8+ cytotoxic T, CD8+/CD45RA+ naive cytotoxic, and dendritic cells. Most of these cell types look relatively homogeneous. However, one cell type, CD34+, was particularly noisy with very high zero proportions, indicating cellular heterogeneity (Fig. 1d). Based on the diagnosis from t-SNE plots, we identified three subtypes within the CD34+ cells. The alignment of zero proportions against the Poisson curve was immediately improved according to the inferred subtype labels. This indicates the effectiveness of zero proportions as metrics to evaluate cellular heterogeneity and their potentials to discern cell types. We further checked how zero proportions could be dispersed from the Poisson distribution for genes with various functional annotations across all PBMC datasets (Fig. 1e). Specifically, we calculated the difference between observed zero proportions and expected proportions (under Poisson) for each functional group using reference data from GENCODE of GRCh38 [28]. The vast majority of genes are categorized as "proteincoding genes." Their zero proportions cover a wide range from 0 to 0.7, but centered at 0, indicating variability in zero proportions but no systematic inflation of zero proportions. In contrast, immune-related genes are consistently zero-inflated, with the interquartile range as high as 10 to 20%. The enrichment analysis (Additional file 1: Table S4) shows that immune-related genes have significantly higher proportion of zero-inflated genes compared to genes that are not related to immune function. The top-ranked annotations for zero inflation include IG C genes, TR C genes, and HLA genes. IG C genes are immunoglobulin genes of the constant (C) region, while TR C genes are T cell receptor genes of the constant region, and hence, both gene types are deeply connected to immune system. HLA genes are genes in human leukocyte antigen system that is responsible for the regulation of the immune system. Genes involved in immune functions are expected to be inherently heterogeneous [29]. For example, HLA genes are highly polymorphic than others, and TR-C genes go through VDJ recombinations that lead to more diverse sequences across cells. This result corroborates with the notion that cellular heterogeneity is the main driver of zero inflation. Higher level of heterogeneity in immune genes explain the past studies' results that even within one cell type, there are zero-inflated genes [30]. The high heterogeneity of cells in certain genes suggests that it is difficult, to say the least, to fully account for the cell type for every gene; as the cells are finely clustered, a point will be reached where the number of cells left in each cluster is not enough to do statistically meaningful analysis. Therefore, we use the following stopping criteria for iterative clustering where the procedure stops when the number of genes with zero-inflated is less than a certain threshold. This criterion is designed to take into account the remaining granular biological heterogeneity in certain genes that cannot be fully resolved through cell-type. Zero inflation test for cellular heterogeneity Based on the above observations, we propose a new feature selection strategy that uses detected zero proportion of a given gene as the statistic to test for cellular heterogeneity. Under the null hypothesis, where complete cellular homogeneity is assumed, the proportion of zeros is equal to the expected zero proportion under Poisson distribution. Under the alternative hypothesis, zero proportion is inflated, as if the count data follows mixture of Poisson (Methods). Formally, our framework can be presented as follows: where g is gene index and λ g is the mean UMI count for gene g. The above testing framework is based on an assumption that whether UMI count being 0 follows the Bernoulli distribution. Test statistic z g follows a standard normal under the null hypothesis (Methods). Genes with rejected null hypotheses will be selected for downstream analysis. For example, the CD34+ cell population within Zheng2017 dataset [3] has 2.7% of the genes with significant zero inflation at 5% type I error level after Bonferroni correction. But after clustering into subtypes, each subtype had 1.3%, 0.3%, and 1.2% of genes with zero inflation respectively. We demonstrate the intuition of this test procedure in Table 1 using gene PPBP as an example. PPBP was identified with a high zero proportion of 26% and an average mean UMI count of 25.89 within CD34+ cells, indicating very high zero inflation with z-score greater than 10 6 when the proposed test is applied. After we separate CD34+ cells into three subtypes, the test within each subtype is no longer statistically significant. We observe PPBP is highly expressed in |
subtypes 2 and 3 and is almost unexpressed in subtype 1. This shows how cellular heterogeneity can drive excessive zeros and how zero proportions can be used to discern cell types. The proposed framework significantly differs from existing ones in several ways. First of all, only the proportion of zeros (p g ), but not that of other non-zero count values, is used in the test. We empirically show that this statistic is sufficient for cellular heterogeneity analysis in many data sets with low UMI counts. This allows us not to search for a particular parametric distribution to fit all non-zero values, which can be computationally more burdensome. In terms of clustering, analysis shows that deviance and zero-inflation both lead to feature selection with similar performance (Additional file 1: Figure S11, S12); our software can use deviance test for feature selection when a data set has high UMI counts and zeros alone do not hold enough information. Secondly, this framework allows each gene to have different grouping structure across cells. Most existing methods select one set of genes to cluster all the cells [10,18,31,32], which implicitly assume cell types can be well-defined biologically by a common set of genes. This is not realistic given the fact that each gene's heterogeneity level varies with its function. For example, housekeeping genes are expected to behave similarly in all cells but immune-related genes, known to have more diverse genetic profiles with highly polymorphic nucleotides [33,34], might be more finely differentiated among the sampled cells. Our approach acknowledges this type of variability. Finally, our approach provides a much more optimistic view of the UMI data analysis. No complicated modeling is needed for resolving the cellular heterogeneity. We observe that the droplet-based data-generating process in the 10X protocols affect UMI counts in different cell types and even across datasets in a similar fashion (Additional file 1: Figure S1, S2). Regardless of samples or cell types, all cells show the same distribution of zero proportions with respect to mean UMI count. This means the technical noise affects each and every cell fairly, and hence, biological heterogeneity alone can largely explain the zero inflation phenomenon. Once the heterogeneity is accounted for, without any other pre-processing steps, zero proportions of UMI data closely follow the expected curve under a Poisson distribution. These observations urge us to re-evaluate some widely used pre-processing methods under this scope. Inappropriate pre-processing introduces unwanted noise in the downstream analysis One of the most popular method for normalization is to divide UMI counts by a cellspecific scaling factor so that total UMI counts are equal across cells [11]. This strategy implicitly assumes sequencing depth effects are purely technical. Total UMI count needs to be carefully corrected in case of data integration. Data sets collected from different protocols and batches have different distribution of UMI counts. However, when there is no batch effect, the total UMI count has valuable biological information. Here, we show sequencing depths are confounded with cell types and size factor-based adjustment can obscure biological information. For a given gene, dividing UMI counts by cell-specific factors does not change its zero proportion across cells but changes its mean. As a consequence, zero proportions across genes no longer follow the expected curve under a Poisson distribution (Fig 2a), and the two curves from each cell type are separated from one another. For example, in 6 PBMC data sets (Azizi and Zheng) [3,27], monocytes have lower UMI counts than B cells. The median UMI counts for monocytes and B cells, respectively, are 787 and 1180 for Zheng data for 68,000 PBMC cells, 4831 and 5575 for Azizi 2018 data for breast cancer tumor patient 9 (replication 1), 4891 and 5372 for patient 10, and 5093 and 5722 for patient 11 (replication 1). When they are forced to match the median UMI count of all cells, the counts for the monocytes are inflated while those for the B cells deflated. In addition, cell types are stratified on the zero proportion plot after adjustment, indicating that total UMI counts of each cell contain valuable information Fig. 2 a Scatterplots between gene means and zero proportions across genes calculated from raw UMI data, clustered data, and data after sequencing-depth normalization, respectively. Fitted line is negative binomial curve. b, c Evaluations of pre-processing in Sctransform. b Distributions of sequencing depths across cells in raw UMI data vs. data cleaned by SCtransform. c Comparisons of three monocyte markers in raw UMI data vs. data cleaned by SCtransform. d, e Evaluations of pre-processing in DCA. d Log fold changes and log p values from differential expression analysis using the same data set but imputed by DCA as heterogeneous and homogeneous cell populations, respectively. e The p values comparisons between two different imputation strategies show the general deflation of biological signals of DCA when applied to heterogeneous cells about its cell type (Fig. 2a, Additional file 1: Figure S14). The UMI counts for different cell types do not need to be consistent across different tissues or organisms. For example, in Zhang2019 data, B cells and monocytes have similar total UMI counts (Additional file 1: Figure S14), but fibroblast has more UMI counts than all other types. Forcing fibroblasts to have the same UMI counts as other types would reduce the signal strengths of the markers for fibroblasts. Sctransform is one recent influential UMI analysis method [13]. The key idea of sctransform pre-processing is to remove sequencing depth effects by introducing logscale sequencing depth as a covariate and regressing it out from each cell under a negative binomial model. Similarly, this approach destroys the natural Poisson structure for zero proportions. We show in Fig. 2b, c an example of how normalization can further interfere with detection of biological signals. Across all 6 PBMC datasets mentioned above, we observe B cells always have more UMI counts than monocytes before pre-processing. Applying sctransform barely modifies the sequencing depths of monocytes but shrinks the UMI counts of B cells to match those of monocytes. Due to the artificial shrinkage, biological markers for B cells, such as MS4A1, CD79A, and CD79B, lose their power to discern B cells and monocytes [35]. This suggests cell type differences could be potentially compromised due to excessive cleaning from sctransform. Another popular pre-processing step is to apply deep learning based de-noising tools such as Deep Count Autoencoder (DCA) and SAVER, which de-convolute the technical effects from biological effects and impute zero accounts due to dropouts at the same time. DCA implements deep neural network with flexible parametric options for noise distributions. Similarly, we observe DCA blurs the distinction among cell types, because denoising methods essentially regularize each cell to resemble one another. We illustrate its negative impacts on downstream analysis by comparing differential expression analysis results using two imputation strategies. We selected naive T cells and regulatory T cells from Zhengmix8eq for the experiments, which clustering algorithms often struggle to differentiate because of their similarity. In the first experiment, we imputed naive T cells and regulatory T cells together. In the second, we imputed naive T cells and regulatory T cells separately. Then, we performed DE analysis on imputed data sets using edgeR's likelihood ratio test [22]. We observe much greater log fold change values between naive T and regulatory T cells from data imputed separately than data imputed together. Overall, the signal strength of DE analysis is greatly compromised across all the genes if imputing two cell types together (Fig. 2d-e). Using type I error level of 0.05, 320 genes pass the Bonferroni criterion if clustering is performed first, while only 156 does if imputation is performed first. Known markers including CD4, CTLA4, FOXP3, and IL2RA [35] lost significant amount of biological signals by showing weaker log-fold change (Fig. 2d, Additional file 1: Figure S15). When the cells were first clustered and then imputed, the p values were 1e−04, 8e−04, 2e−07, and 4e−11 respectively for those genes. When the cells were imputed first through DCA, the p values were 3e−01, 4e−02, 4e−02, and 6e−07. Hence, three of the 4 genes lost statistical significance at a very liberal p value threshold of 0.05. This analysis suggests imputing the UMI data without resolving cell heterogeneity can lead to loss of important biological information. HIPPO: Heterogeneity-Inspired Pre-Processing tOol The above analyses suggest the first and foremost step in pre-processing is to account for the cellular heterogeneity. Imputation or normalization before resolving the cellular heterogeneity may lead to inevitable loss of biological signals. We implement this new perspective into a computational tool called HIPPO, where we integrate the proposed zero inflation test into a hierarchical clustering framework. Specifically, we first selected genes with strong indication for cellular heterogeneity. We use a cutoff of 2 on z score for selection of genes. The selected features were then used to cluster the cells into 2 groups using PCA + K-means. Then, each cluster was evaluated with their intra-variability using the mean Euclidean distance from the centers of K-mean algorithm. The group with the highest intra-variability was selected and assigned for next round of clustering. The feature selection and clustering steps are iteratively repeated until one of the two ending criteria are met: K round of clustering for pre-determined number of clusters K, or the number of zero-inflated genes is less than a certain percentage of the genes. The former one can be difficult to set in real practice without any prior knowledge and the later one offers a more natural stopping criterion. HIPPO is computationally cheap because fewer and fewer features will be left for the next round of clustering, and the Poisson-based test statistic has closed-form expression (Fig. 3a, b). In Fig. 3c, we show the results from each iteration of HIPPO on Zhengmix8eq data. HIPPO successfully identifies monocytes, natural killer cells, B cells, and T cells in the respective order. Then, it further separates naive cytotoxic cells, memory T cells, and naive T cells from a group of regulatory T cells and helper T cells. However, when forced to separate into one more group, instead of clustering the remaining T cells, it created another subgroup of natural killer cells. Meanwhile, Seurat and Sctransform fails to separate the memory T cells, regulatory T cells, and helper T cells, grouping them as one cluster. (Fig. 3d). The adjusted rand index for the three methods show that HIPPO performs the best throughout the different K specification (Fig. 3e). When the selected features' characteristics were studied through CV, gene variance, and zero proportion, Seurat and Sctransform selected more features (2000 and 3000 respectively while HIPPO selected 950), but they are highly concentrated where gene means are near 0. This is because their feature selection focuses on coefficient of variation which becomes numerically unstable as gene mean becomes near zero. HIPPO selects fewer but more relevant genes by using the zero proportion as the selection metric (Fig. 3f ). This result is repeated in a different data set from muscular heart tissue in Fig. 4a. Genes selected by both methods are those with non-zero mean UMI counts, but Seurat selects extra number of genes that have mean count very close to 0. These genes are likely to add noise instead of contributing to real biological signals detection. HIPPO's iterative procedures naturally offer strong interpretability through sequential visualization of the analysis at each round of clustering. We use HIPPO results on an unlabeled 10X UMI data set of 10K E18 mouse heart cells for illustration. Sequential feature selection can be monitored through the visualization of the changing relationships between zero proportions and gene means. As cells are clustered into finer distinct groups, or as more cellular heterogeneity is resolved, regression lines between zero proportions and gene means get more closely aligned with the expected Poisson curve (Fig. 4b). Simultaneously, we can use a heatmap to visualize top features that contribute most at each round of clustering (Fig. 4c). In addition to biomarkers identified based on zero inflation, HIPPO also implements a differential expression test based Fig. 3 HIPPO framework applied to Zhengmix8eq data. a Computing time for each method using LAMBDA QUAD workstation with Intel Xeon W-2175 processor sequentially (non-parallel). b Computing time for HIPPO using different k. c HIPPO's sequential clustering results for K = 3, · · · , 8. d t-SNE plots for clustering |
results from three methods: HIPPO, Seurat, and SCTransform, compared to true labels. Seurat and SCTransform cannot differentiate helper T/regulatory T and memory T cells. e Clustering results comparisons using Adjusted Rand Index. f Comparisons of features selected by different methods for their gene mean, CV, and variance. Seurat and SCTransform use CV as the selection criteria, and hence, their features weigh heavily on genes with small mean expression and variance on all count values to extract more features (Methods, Fig. 4d). The differential analysis can be viewed together with a t-SNE plot constructed with the same color code (Fig. 4e). Discussion We have provided a new perspective on the analysis of single-cell UMI data sets of multiple tissues and protocols (Additional file 1: Figures S1, S2, S3, S4, S7). Extensive analyses confirm the claims of recent literature [6] that different tool must be applied to the UMI data set from the tools for read count data set; UMI data set is free from amplification bias, so the level of technical noise is much lower. The results also show that cell-type heterogeneity must be tackled as the first step of analysis for more reliable downstream Fig. 4 HIPPO framework applied to 10K E18 mouse heart cells. a Distributions of means across gene features selected by Seurat, Hippo, both, or none. b Sequential feature selection visualizes how genes gradually align closer to the expected Poisson line as more heterogeneity accounted for. c Heatmaps of top features selected at the first 5 rounds of clustering. d Top differential expression genes obtained at each round of clustering. e Visualization of sequential clustering using t-SNE plots analyses. Moreover, through a streamlined feature selection method that reflects the dynamic nature of cellular process, the proposed method provides a computationally and mathematically simple analysis tool with great interpretability. There are remaining challenges that are important in the future development of single cell UMI data analysis. First, lack of labeled data restricts the analyses in certain protocols such as Drop-seq. There is strong evidence for our method in 10X data sets. Supplementary Figures also show that the claims hold in all 10X data, Tung2018 data that uses Hi-Seq 2500 [25] and Baron2016 data that uses in-Drop [36]. In Drop-seq, the noise level was too high to assume the zero proportions follow the exponential curve relative to the gene mean (Additional file 1: Figure S7). It is either that Drop-seq data sets have different noise structure from the 10X data sets, or in particular Macosko data [2] of muscular retina cells have excessively high cellular heterogeneity [1]. Future new Drop-seq data could help resolve the discrepancy between 10X and Drop-seq. Second, although HIPPO is computationally simple compared to existing tools, the current computational bottleneck is the principal component analysis, which could be slow for large cell numbers. In that case, advanced computing techniques such as sub-sampling or more rigorous filtering should be applied. In addition, HIPPO is implemented as evolving modular software. In the current release, we include two different feature selection methods, deviance test and zero-inflation test, and two differential expression detection methods. Alternative dimension reduction or clustering methods can be easily incorporated into the framework. We focus on the pre-processing with resolving cellular heterogeneity in our analysis tool, but this novel perspective on the noise structure of UMI data can be extended to other steps of analysis pipeline. Batch correction, lineage analysis, or trajectory inference can all benefit from the simpler noise structure not only computationally but also by avoiding unnecessary normalizing steps that can introduce unwanted bias and noise. Datasets Throughout the analysis, we used publicly available single cell UMI sequencing data from various protocols. Most analysis in the main text is focused on SRP073767 which is also available in 10x Genomics, and it sequences 68,000 PBMC cells using Cell Ranger 1.1.0 [3]. We use different subsets of this data sets, namely Zhengmix4eq, Zhengmix4uneq, and Zhengmix8eq as defined in Duo (2018) [31]. Other data sets used in the main text are GSE111108 [37] and GSE115189 [17], and GSE114724 [27]. Supplementary data includes more data sets from 10X including 5k Cells from a combined cortex, hippocampus and subventricular zone of an E18 mouse (v3 chemistry), 1k Brain Cells from an E18 Mouse (v2 chemistry), and 10k Heart Cells from an E18 mouse (v3 chemistry). We also use Tabula Muris data from various mouse tissues [38]. We also use GSE84133 [36] as an example of in-Drop, GSE63473 [2] as an example of Drop-seq, SDY998 [39] as an example of CEL-seq2, and GSE77288 [25] as an example of Hi-Seq. All the data sets were analyzed after their own filtering process (Table 2). Benchmarked methods In Fig. 3, we benchmark Seurat 3.0.0 [40] and SCTransform version 0.2.0 that is integrated with Seurat platform. Seurat was implemented following its guided tutorial https://satijalab.org/seurat/v3.1/pbmc3k_tutorial.html, and SCTransform through a vignette https://rawgit.com/ChristophH/sctransform/master/inst/doc/seurat.html. All parameters were selected through software's default except resolution parameter for clustering to generate results for various number of clusters. Seurat used in Fig. 4 A was also the same version with the default parameters for feature selection. In Fig. 3a, the t-SNE plots were created using the features selected by the first round of HIPPO because they reflected the division of true cell labels the most accurately. DCA was installed through Conda and imputation was performed following the tutorial on https://github.com/theislab/dca. In one experiment, we first divide the data set into correct labels, and then impute them separately using DCA (imputing homogeneous Poisson mixture model Consider a gene by cell matrix if UMI counts X for gene g = 1, · · · , G and cell c = 1, · · · , C. To understand the behavior of the zeros for each gene, the first step is to reduce the information from each gene to the proportion of zeros across the cellŝ which is an estimator for the true zero proportion of gene g: p g . We study its relationship against the mean expression for the set of cells, because p g would decrease as the expression level increases. With the test statistic above, we test a one-sided hypothesis for each gene g, whether the zero proportion is higher than the expected rate under the Poisson model. For the alternative hypothesis, we believe that UMI counts follow finite Poisson mixture. The hypotheses for each gene g are formally specified below. In practice, we re-frame the hypotheses as H 0 : K g = 1, H A : K g > 1 when p = K g k=1 π k e −λ kg . In other words, zero inflation indicates there is cell heterogeneity across the samples. If the cell population is truly homogeneous, the count data follows Poisson data with expected zero proportion e −λ g . Chen (2018) and Sarkar (2020) demonstrates that most genes in UMI data follow Poisson distribution [6,7] while other noisy genes follow negative binomial or zero-inflated binomial distribution. Such model, although fundamentally different, is closely tied to the Poisson mixture model because negative binomial is the limiting distribution of Gamma-Poisson. If λ cg for each cell is drawn independently from the gamma distribu- While negative binomial assumes a continuous mixture of Poisson, the proposed model assumes a finite mixture of Poisson, which is simpler and more directly addresses the source of zero inflation. In practice, we do not explicitly estimate π k , but instead simply test if observedp g is larger than expected p with estimated gene mean λ. (H A : p g > e −λ g ). It might seem counterintuitive that this test statistic does not fully leverage the specification of the alternative hypothesis; mixture parameters π k are not estimated. Alternatively, for example, one might suggest that we can conduct a likelihood ratio test of Poisson versus Poisson mixture. The main strength of the proposed reduced test statistic is its robustness to the modeling assumptions. Table 3 shows that the proportion of zeros are always larger than expected under different alternative hypotheses. Under the proposed alternative, mixture of Poisson, the proportion of zeros under the null hypothesis would be e −λ where λ is the weighted mean of the gene mean for each cell-type. Due to Jensen's inequality, p under alternative hypothesis is always greater than that under the H 0 (Additional file 1: Figure S13) ( Table 4). In the first row, the right column is larger than the left column due to Jensen's inequality. For negative binomial, the dispersion parameter r is constructed so that the variance is λ 2 r + λ, so that Poisson is a special case of negative binomial with r = ∞. The zero-inflated negative binomial distribution is parameterized as π 0 δ 0 + (1 − π 0 ) NB(λ, r) Feature selection and Inference We provide two ways to select features: zero-inflation test and deviance test. The clustering performance is similar using both methods. For zero-inflation test, HIPPO defines the observed zero proportionp g and expected zero proportion e −X . For gene g with count data X gc for cells c = 1, · · · , C, consider an estimate for the proportion of zerosp g aŝ The gene mean is estimated as the average UMI countsX g = 1 C C c=1 X gc and is treated as a fixed number. Then, The test statistic z-score for gene g is as below. To note, the gene mean e −X is also a random variable that follows a log-normal distribution, whose inference is not trivial (further discussion can be found in Supplementary Text 1). However, this feature selection method works well in practice and intuitively interpretable. Users can also use deviance measure to select the top features [8]. The deviance threshold is required from the users to select cutoffs to select the features. Hierarchical clustering Algorithm 1 outlines the iterative procedure of HIPPO's hierarchical clustering. Several stopping criteria can be determined by the user: the maximum number of clusters K, the feature selection statistic threshold z, and outlier gene proportion o. The algorithm first computes the number of outlier genes to allow, G × o. For example, if there are 30,000 genes in total and o is specified as 1% = 0.01, then the algorithm allows 300 features to have zero inflation. During the clustering procedure, HIPPO terminates in either scenarios: there are K identified clusters or if there are less than G × o genes that exceed the specified z value threshold. HIPPO takes all the cells and select the features whose zero inflation statistic z exceeds the threshold. Only zero-inflated features are then log-transformed (log(X)+1) and sent Algorithm 1 Cell-Type Hierarchical Clustering K: upper limit of cluster number z-threshold: threshold for feature selection = 1 for k = 2, · · · , K do if Less than the designated number of genes exceed z threshold then stopping criterion; terminate algorithm else update the matrix by selecting new features log transformation + centered/scaled PCA + Kmeans divide cells into two groups, one with label and another with label k log transformation + un-centered/un-scaled PCA update intra-cluster variation by taking sample variance of un-scaled PCs update = cluster with the highest intra-cluster distance end if end for return cluster labels for each k into principal component decomposition with scaling and centering. HIPPO then clusters the cells into two groups using the dimension-reduced cell embeddings through K-means, which is performed multiple times (user-defined, default 10) for stability. Meanwhile, during the hierarchical clustering, HIPPO keeps track of intra-cluster variation by performing unscaled, uncentered PCA. HIPPO computes the first 10 dimensions (user-defined) of cell embeddings and sum up the sample variance of each component. The PCA for recording the intra-cluster variability is not scaled to avoid potential bias due to the cell population size. Since a subset of cells are considered for clustering at each round, fewer and fewer cells are used for the dimension reduction. If scaled, their dimension-reduced cell embeddings would be artificially more far apart compared to when more cells were considered for clustering. The intra-cluster variance is the criterion for selecting the cell group to be further clustered in the next round. Differential expression testing HIPPO provides two methods to identify differentially expressed genes, one by performing a t-test on means, and the |
other by performing a Poisson likelihood test. For both methods, the cells are separated into two groups as follows. t =X C 1 g −X C 2 g Oral Bioavailability-Enhancing and Anti-obesity Effects of Hydroxysafflor Yellow A in Natural Deep Eutectic Solvent Hydroxysafflor yellow A (HSYA), a primary active component in Carthami Flos, has been extensively applied in the treatment of cardiometabolic diseases. In this study, a natural deep eutectic solvent composed of glucose and choline chloride with 10% (v/v) of water (90% GCH) was evaluated to enhance the oral absorption of HSYA. Compared with HSYA in water, the relative oral bioavailability of HSYA in 90% GCH was increased to 326.08%. Furthermore, 90% GCH was demonstrated to decrease the mucus viscosity and increase the absorption rate constant of HSYA in the jejunum by 2.95 times. A pharmacodynamic study revealed that HSYA in 90% GCH was more effective in reducing body weight and correcting steatohepatitis and dyslipidemia in high-fat diet-induced obese rats. Serum metabolomics results showed that the correction of serum aromatic amino acid disorder may contribute to the anti-obesity effect of HSYA in 90% GCH. In conclusion, 90% GCH could be a delivery carrier for HSYA against obesity. INTRODUCTION Carthamus tinctorius L. belongs to the compositae family and has been used as a food additive and natural pigment for thousands of years. Carthami Flos (the dried florets of C. tinctorius) is widely applied for treating cardiometabolic diseases. In Carthami Flos, hydroxysafflor yellow A (HSYA) is a representative water-soluble quinochalcone C-glycoside pigment, possessing multiple attractive pharmacological activities against cardiometabolic diseases. 1,2 According to the biopharmaceutics classification system, HSYA belongs to class III drugs with high solubility and low permeability. 3 It has been reported that the oral bioavailability of HSYA in rats was only 1.2% together with a short half-life. Meanwhile, HSYA was highly metabolized in the gastrointestinal tract and liver, resulting in high first-pass extraction, thereby lowering bioavailability. 4,5 In Chinese clinics, safflower yellow injection (HSYA ≥70%, intravenous drip), a yellow pigment extract from Carthami Flos, is one of the top Chinese patent medicines in sales. However, there are still many adverse reactions, such as allergic reactions. 6 It is necessary to design new oral preparation and improve the in vivo kinetic behavior of HSYA to make it more effective and safer. Natural deep eutectic solvents (NADES) are liquid supermolecules formed by natural primary metabolites such as sugars, organic acids, amino acids, and amines as hydrogen bond donors and acceptors through intermolecular interactions, especially hydrogen bonds, and have shown excellent promise in drug delivery applications. 7 Choline-based NADES are emerging as outstanding drug/natural compound carriers due to their biocompatibility, availability, and cost-effectiveness. 8 It has been reported that a NADES composed of glucose and choline chloride with 10% (v/v) of water (90% GCH) was able to stabilize safflower pigments (including HSYA, carthamin, and cartormin) in part due to the strong hydrogen bond interactions between pigments and NADES' molecules. 9,10 Recently, Tong et al. demonstrated that after oral administration of HSYA in Lproline-acetamide (L-Pro-Am), the relative oral bioavailability (Fr, to water extract) of HSYA was 183.5%, suggesting that NADES could be a delivery carrier for HSYA. 11 In our previous study, oral HSYA could reduce body weight and fat accumulation and alleviate insulin resistance in high-fat diet (HFD)-induced obese mice. 12 Solely, choline-based NADES were reported to treat HFD-induced obesity effectively. 13 Therefore, it is intriguing to investigate whether 90% GCH could improve the oral absorption and anti-obesity effects of HSYA. In this study, the oral bioavailability of HSYA in 90% GCH in vivo was evaluated. Second, the effects of H 2 O and 90% GCH on intestinal absorption of HSYA were studied by using an in situ single-pass intestinal perfusion model. In addition, an HFDinduced obese rat model was employed to evaluate the antiobesity effect of HSYA in 90% GCH followed by serum metabolomics to explore the potential anti-obesity mechanism. Overall, this work illustrates the promise of using 90% GCH as an oral delivery carrier for HSYA against obesity. MATERIALS AND METHODS 2.1. Chemicals and Materials. Rutin was used as an internal standard (IS, purity: >98.0%) and purchased from Shanghai Yuanye Biotechnology Co., Ltd. (Shanghai, China). D-Glucose, choline chloride, pancreatin, pepsin (from porcine gastric mucosa), and mucin type II were purchased from Sigma (St. Louis, MO, USA). The HFD (D12492, 60% calories from fat) was purchased from Research Diets Inc. (New Brunswick, NJ, USA). Mass spectrum-grade methanol and acetonitrile were purchased from Honeywell Burdick & Jackson (Muskegon, MI, USA). Liquid chromatographic-grade formic acid was obtained from Merck (Darmstadt, Germany). Deionized water was purified using a Millipore Milli-Q system (Bedford, MA, USA). All other reagents and solvents were of analytical grade. 2.2. HSYA Preparation. About 50 g of Carthami Flos was immersed at 60°C three times in 600 mL of deionized water for 1 h. The water extract was filtered and concentrated at 50°C under vacuum. The concentrated solution was purified over a column of AB-8 macroreticular resin (Shaanxi Lebo Biochemical Technology Co., Ltd., China), eluted with 0, 30%, 50%, 70%, and 95% ethanol. Furthermore, the 50% eluted part was concentrated and purified by Sephadex LH-20 (GE Healthcare Bio-Sciences AB, Sweden), analyzed in HPLC, and freeze-dried into powder (HSYA purity: >95.0%). By the way, HPLC analysis was performed on a Waters 2695 system equipped with a 2998 PDA detector. A Waters SunFire C18 column (4.6 mm × 150 mm, 5 μm) was used, and the column temperature was set at 25°C. The mobile phase consisted of methanol (A) and 0.7% phosphate acid water (B). The gradient program was as follows: 10−40% A at 0−2 min, 40−42% A at 2−5 min, 42−45% A at 5− 7 min, 45−10% at 7−9 min with a flow rate of 1.0 mL/min. The injection volume was 10 μL, and the wavelength was set at 403 nm. 2.3. 90% GCH Preparation. First, GCH solvent was prepared by mixing glucose and choline chloride at a 2:5 mole ratio with mild heating and stirring at 50°C. The 90% GCH was prepared by diluting a certain volume with deionized water. 2.4. Solubility Tests. Solubility tests were carried out by saturating solvents (90% GCH and H 2 O) with an excess of HSYA in an Eppendorf (EP) tube, stirred at room temperature (20°C) and 50°C for 2 h, settled for 3 h, and centrifuged at 12,000 rpm for 20 min. The supernatant was diluted with deionized water and filtrated by 0.22 μm microporous filters. All the samples were tested by triplicate. The HPLC method mentioned in Section 2.2 was used for quantitative analysis of HSYA. 2.5. In Vivo Pharmacokinetic Studies. 2.5.1. Instrumentation and Chromatographic Conditions. Chromatographic analysis was carried out using a Waters ACQUITY I-Class UPLC system (Milford, MA, USA), equipped with a binary pump solvent delivery system, an automatic sampler, and an on-line degasser. A UPLC BEH C18 column (100 mm × 2.1 mm, 1.7 μm) was used for separation. The mobile phase consisted of acetonitrile (A) and water containing 0.1% formic acid (B). The gradient program was as follows: 0.0−1.0 min, 5% A; 1.0−2.5 min, 5−25% A; 2.5−5.0 min, 25−55% A; 5.0−5.5 min, 55−95% A; 5.5−6.0 min, 95−5% A; 6.0−7.5 min, 5% A. The flow rate was 0.30 mL/min. The column temperature and the automatic injector temperature were set at 30 and 10°C, respectively. MS detection was performed using a Waters Xevo TQ-XS mass spectrometry system (Milford, MA, USA) equipped with an ESI source in negative mode. The ionization source conditions were set as follows: capillary voltage 2.6 kV, desolvation temperature 400°C, cone gas flow 150 L/h, and desolvation gas flow 1000 L/ h. Argon (99.99%) was used as the collision gas. The detection and quantification of the analytes were performed using multiple-reaction monitoring mode. The cone voltage and collision energy were optimized for HSYA (40 and 26 eV) and IS (14 and 34 eV). The dwell time was automatically set by Waters Masslynx v 4.2 software. 2.5.2. Method Validation. Blank plasma samples were obtained from six different male Sprague-Dawley (SD) rats. The chromatogram comparison of blank plasma and blank plasma added with HSYA or IS was used to determine the specificity of assay. The calibration curve of HSYA was based on the ratio of peak area of HSYA to IS and nominal concentration of calibration standards, which was in line with linear regression y = ax + b, where x stands for the plasma concentrations of HSYA in rats and y stands for the peak area ratio of HSYA to IS. The range of concentrations of the calibration curve was from 12.5 to 10,000 ng/mL. The lower limit of quantification (LLOQ) was considered as the lowest concentration on the calibration curve with precision variation <20%. Quality control (QC) samples at low, medium, and high concentrations of HSYA were analyzed for three consecutive validation days to assess the intraday and interday precision. The precision and accuracy were expressed as relative standard deviation (RSD) and relative error (RE), respectively. The extraction recovery was used to analyze QC samples with low, medium, and high concentrations of HSYA, which was the ratio of the average peak area of HSYA added before extraction to that after extraction. The stability of QC samples containing low, medium, and high concentrations of HSYA was investigated under the conditions of standing at room temperature for 6 h, repeated freeze−thaw three times, and storage at −20°C for 6 days. 2.5.3. Pharmacokinetic Study. Ten male SD rats (230−250 g) were provided by Chengdu Dossy Experimental Animals Co., Ltd. (Chengdu, China) and housed in clean plastic cages with free access to water and food in ambient temperature with 40%− 60% humidity under a 12 h light and dark cycle for 1 week. These rats were further divided into two groups at random (i.e., the H 2 O group and 90% GCH group) and fasted for 12 h with free access to water prior to drug administration. The dosage preparation was made by dissolving appropriate amounts of HSYA in H 2 O and 90% GCH (40 mg/mL) and then giving a single dose (100 mg/kg) by gastric gavage. About 300 μL blood samples were immediately collected in heparinized polyethylene tubes at 0.083, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 10, and 12 h after dosing from the fossa orbitalis vein of the rats in both groups. After centrifugation at 5000 rpm for 10 min, plasma was finally obtained and stored at −80°C until analysis. The animal welfare and experimental procedures were strictly in accordance with the guide for the Care and Use of Laboratory Animals (US National Research Council, 1996) and the related ethics regulations of the Ethical Committee of Shaanxi University of Chinese Medicine. Sample Preparation. A 100 μL plasma sample was added with 10 μL of IS and 190 μL of methanol in a 1.5 mL EP tube. The mixture was vortexed for 30 s and then centrifuged at 14,000 rpm for 10 min. The supernatant was transferred to a fresh 1.5 mL EP tube and centrifuged at 14,000 rpm for another 10 min. At last, 1.5 μL of the supernatant was injected for UPLC-MS/MS analysis. 2.5.5. Pharmacokinetic Parameter Calculation. To determine the pharmacokinetic parameters, concentration−time data were analyzed with a non-compartmental method using DAS 2.0 software. The maximum plasma concentration (C max ) and peak time (T max ) of HSYA were determined directly from the mean plasma concentration−time (C−T) curve. The areas under the C−T curve from time 0 to t (AUC 0−t ) and from time 0 to infinity (AUC 0−∞ ) were calculated using the linear trapezoidal rule. The terminal half-life (t 1/2 ) was calculated as 0.693/kel, where kel was the apparent elimination rate constant of HSYA from plasma. The AUC 0−t of HSYA in 90% GCH versus H 2 O was used to calculate the relative bioavailability (Fr). 2.6. Mucus Rheology Studies. The biological similarity mucus was modified on a previously reported method. 14 Specifically, 0.75% (w/v) polysorbate 80 was added to 0.5% carboxymethylcellulose sodium and 5% mucus was added to make simulated mucus under stirring conditions. A total of 10 μL of 0, 3.13, 6.25, 12.5, 25, and 50% v/v of 90% GCH was added to 200 μL of simulated mucus, and viscosity was measured with a rheometer in the shear rate range of 1−200 1/s at |
37°C. 2.7. In Situ Single-Pass Intestinal Perfusion. It has been reported that HSYA is mainly absorbed in small intestinal segments. 15 Herein, the effects of 90% GCH on the absorption of HSYA in small intestinal segments (i.e., duodenum, jejunum, and ileum) were studied. The male SD rats were divided into two groups (n = 3). Before the operation, the rats were fasted for 12 h and free to drink water. The preparation of Krebs-Ringer (K-R) buffer was done according to a previous literature study. 16 Rats were anesthetized with 10% (v/w) chloral hydrate, and normal body temperature was maintained with an infrared lamp. After confirming the disappearance of pain reflex, the duodenum, jejunum, and ileum (approximately 10 cm) were carefully pulled out and small cuts were made at the beginning and the end, and silicone tubes were inserted and ligated. The inlet was filled with a well-weighed EP tube equipped with a test solution, and another well-weighed EP tube was placed at the exit to collect the effluent. After the operation, cotton pads soaked in saline were used to cover the wound. The intestinal contents were rinsed with normal saline, and the preheated K-R buffer and a drug-containing perfusate were equilibrated at 37°C for 30 min at a flow rate of 0.2 mL/min. After 30 min balance, a 5 mL EP tube with known weight containing 1 mg/mL HSYA in H 2 O or 90% GCH was immediately used for intestinal perfusion. A 5 mL EP tube with known weight was used to receive the effluent at the exit and was replaced every 15 min for a total of 90 min. At the end, the perfused intestinal segments were cut off. The length and circumference of the intestinal segments were measured under normal saline at 4°C to ensure that the intestinal segment was not stretched. For the determination of HSYA, 500 μL of the effluent was taken, added with 500 μL of methanol, and centrifuged at 12,000 rpm under 4°C for 10 min. Finally, 10 μL of the supernatant was injected for HPLC analysis according to the method mentioned in Section 2.2. The absorption rate constant (K a ), the effective permeability coefficient (P eff ), and the absorption percentage (F ab ) of HSYA in the small intestine were calculated according to the following mathematical expression: 17,18 where V in and V out are the volumes of the perfusate entering and leaving the intestine, respectively; Q is the perfusion flow rate; C in and C out are the concentration of the drug in the perfused inlet and outlet, respectively; L is the length of the intestinal segment (cm), and r is the radius of the intestinal segment (cm). 2.8. Stability Study in Simulated Gastric and Intestinal Juices. The simulated gastric and intestinal juices were prepared according to Chinese Pharmacopeia. 19 HSYA (1 mg/mL) in H 2 O and 90% GCH was added to simulated gastric and intestinal juices (v/v, 1:7.5) and incubated at 37°C for 0, 1, 2, or 4 h, respectively. The juice (200 μL) was sampled at each time point and centrifuged at 12,000 rpm for 10 min. Finally, 10 μL of supernatant was injected for HPLC analysis according to the method mentioned in Section 2.2. 2.9. Pharmacological Study. 2.9.1. Animals and Experimental Design. The male SD rats were divided into three groups (n = 6) and fed with an HFD (D12492, 60% calories from fat) for 5 weeks. Two of the groups were orally dosed daily for 5 weeks with 100 mg/kg HSYA in H 2 O and 90% GCH. The other group of rats was treated with pure water. Throughout the trial, the body weight was recorded weekly. All experiments were terminated at the end of the fifth week. 2.9.2. Oral Glucose Tolerance Test (OGTT). After a 6 h fasting period, blood glucose was determined with a glucose meter (Abbott FreeStyle Optium, Witney, UK) using the blood collected from the tip of the tail vein before and 30, 60, 90, and 120 min after D-glucose (2.0 g/kg, 50% solution; Hubei Kelun Pharmaceutical Co., LTD.) challenge. Cumulative changes in blood glucose response were quantified by the incremental area under the curve (AUC). 2.9.3. Collection of Serum and Tissue Samples. At the end of the fifth week, fasting for 12 h was performed under 10% chloral hydrate anesthesia and 4 mL blood samples were taken from the abdominal aorta. Blood was centrifuged at 3000 rpm for 10 min under 4°C. Then, the serum samples were collected and stored at −80°C. The epididymal fat, liver, stomach, duodenum, jejunum, ileum, lung, spleen, heart, and kidney of each rat were immediately excised, and the liver was weighed. All samples were stored in 4% paraformaldehyde for follow-up analysis. The liver index was calculated as a percentage of liver weight to body weight. 2.9.4. Biochemical Measurements. Alanine aminotransferase (ALT), aspartate transaminase (AST), blood urea nitrogen (BUN), creatinine, total triglycerides (TG), high-density lipoprotein cholesterol (HDL), and low-density lipoprotein cholesterol (LDL) were measured using kits for rats from Royta life and Analytical Sciences, Shenzhen, China. Serum total cholesterol (TC) and fasting serum glucose (GLU) were measured using kits for rats from Nanjing Jiancheng Bioengineering Institute, Nanjing, China. All kits were used according to the manufacturers' instructions. 2.9.5. Histopathological Analysis. Tissue samples were fixed in 4% paraformaldehyde for 24 h and embedded in paraffin. Tissues were cut into sections (each 4 μm-thick) and stained using hematoxylin and eosin (H&E). Histological sections were analyzed and photographed under a light microscope (Nikon Eclipse E100, Nikon, Japan). 2.10.2. Sample Preparation. Serum (100 μL) was placed in a 1.5 mL EP tube and fully vortexed with precooled 400 μL of 80% methanol. Then, the sample was incubated on ice for 5 min and centrifuged at 15,000g at 4°C for 20 min. The supernatant (400 μL) was diluted to final concentration containing 53% methanol and then transferred to a fresh 1.5 mL EP tube and centrifuged at 15,000g at 4°C for 20 min. The final supernatant was injected for UHPLC−MS/MS analysis. 2.10.3. Data Processing and Analysis. The raw MS files were processed using Compound Discoverer 3.1 (CD 3.1, Thermo-Fisher). Peak intensities were normalized to the total spectral intensity. The normalized data was used to predict the molecular formula based on additive ions, molecular ion peaks, and fragment ions. Then, the peaks were matched with the mzCloud (https://www.mzcloud.org/), mzVault, and MassList databases to obtain the accurate qualitative and relative quantitative results. Metabolites were annotated using the KEGG (https:// www.genome.jp/kegg/pathway.html), HMDB (https://hmdb. ca/metabolites), and LIPIDMaps (http://www.lipidmaps.org/) databases. Principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) were performed at metaX (BGI Tech, China). T-tests were used to calculate the statistical significance (P-value). The metabolites with variable importance in projection (VIP) > 1 and P-value < 0.05 as well as fold change (FC) > 1.2 or FC < 0.833 were considered to be differential metabolites. Pathway analysis was performed with the MetaboAnalyst 5.0 pathway analysis module (http://www. metaboanalyst.ca). 2.11. Statistical Analysis. Experimental data were analyzed using SPSS 25.0. Differences between multiple groups were analyzed using one-way ANOVA with Tukey's multiple comparisons test. For comparisons between two groups, Student's t-tests were used for parametrically distributed data and Mann−Whitney tests for nonparametrically distributed data. All data were presented as mean ± SD. When P < 0.05, it was considered statistically significant. RESULTS 3.1. HSYA Solubility in H 2 O and 90% GCH. The solubilities of HSYA in H 2 O were 258.814 ± 14.519 mg/mL at 20°C and 321.877 ± 6.756 mg/mL at 50°C. Due to the high viscosity of 90% GCH, the solubilities of HSYA were reduced to 153.254 ± 7.054 mg/mL at 20°C and 176.550 ± 7.921 mg/mL at 50°C. 3.2. 90% GCH Enhanced the Oral Bioavailability of HSYA. With the aim of investigating if 90% GCH could increase the oral bioavailability of HSYA, we established and validated a UPLC-MS/MS method to determine the plasma concentration of HSYA. The retention times of HSYA and IS were 3.03 and 3.51 min, respectively, with no observed endogenous interference ( Figure 1A). The calibration curve for HSYA was y = 0.00150532x − 0.0126754, with r 2 = 0.9992. The LLOQ for HSYA was regarded as 12.5 ng/mL. The average extraction recoveries were ranged from 93.1 to 104.9%, while the average matrix effects were between 93.7 and 102.2% (Table S1 in the Supporting Information), indicating that the pretreatment method was appropriate. The intra-and interday precision variations of HSYA at each QC level were within 1.7−6.4%. The RE of accuracy was within ±5.7% (Table S2 in the Supporting Information). The stability of HSYA was acceptable at room temperature for 6 h, three freeze−thaw cycles, and at −20°C for 6 days. All RSD values were less than 14.3%, and RE values were within ±9.9% (Table S3 in the Supporting Information). These results indicated that the developed UPLC-MS/MS method is satisfactory to determine the plasma HSYA concentration. The C−T curves of HSYA in H 2 O and 90% GCH groups are shown in Figure 1B. Meanwhile, the major pharmacokinetic parameters are listed in Table 1. Specifically, HSYA in both groups was absorbed rapidly and reached C max within 1 h and then eliminated in 12 h. Nevertheless, the C max of HSYA in the 90% GCH group was 5.02-fold as that in the H 2 O group. The AUC values were also significantly increased (P < 0.05), resulting in the Fr of HSYA in 90% GCH at 326.08%. 3.3. 90% GCH Decreased the Simulated Mucus Viscosity. The mucus barrier lining the gastrointestinal tract poses a significant barrier to drugs' oral delivery, 20 so the rheology of simulated mucus treated with 90% GCH was assessed. The viscosity of the simulated mucus treated with 90% GCH showed a notable drop throughout the entire measured shear range. At a shear rate of 50 1/s, the mean viscosity of untreated simulated mucus was measured as 10.07 mpa·s. The addition of 3.13, 6.25, 12.5, 25, and 50% v/v of 90% GCH decreased the mean simulated mucus viscosity to 9.08, 8.03, 8.93, 9.02, and 8.34 mpa·s, respectively, with the minimum corresponding to 6.25% v/v of 90% GCH. 3.4. 90% GCH Increased Small Intestinal Absorption of HSYA In Situ. The small intestine is the primary site of drug absorption in the gastrointestinal tract. The in situ single-pass intestinal perfusion model was used to explore the effects of H 2 O and 90% GCH on the small intestinal absorption of HSYA. As shown in Table 2, K a values in the jejunum were 2.95 times higher than that in the H 2 O group. Meanwhile, the P eff and F ab values of jejunum and ileum increased significantly (P < 0.05). Surprisingly, the P eff and K a values of jejunum in 90% GCH group were significantly higher than those of duodenum and ileum (P < 0.05). 3.5. 90% GCH Increased the Stability of HSYA in Simulated Gastric and Intestinal Juices. We also considered whether 90% GCH could prevent HSYA degradation from gastric and intestinal juices. In Table 3, HSYA in both groups in simulated intestinal juice was more stable than simulated gastric juice. In simulated gastric juice, the degradation of HSYA in 90% GCH was slower than that in the H 2 O group within 1−2 h. The degradation of HSYA in 90% GCH in simulated gastrointestinal (gastric × intestinal) juice within 2 h was slower than that in the H 2 O group but increased rapidly on 4 h. 3.6. 90% GCH Enhanced the Protective Effect of HSYA against HFD-Induced Obesity. The average body weight and liver index of rats in the 90% GCH group was significantly decreased compared with the HFD group (Figure 2A,C, P < 0.05 or P < 0.01). Meanwhile, the rats in the 90% GCH group ate less food than HFD-fed rats ( Figure 2B). The AUC of OGTT was significantly decreased in the 90% GCH group compared with H 2 O and HFD groups ( Figure 2D,E, P < 0.05 or P < 0.01), suggesting that 90% GCH could reinforce the positive effect of HSYA on impaired glucose tolerance induced by HFD. The following serum biochemical indicators were also measured: ALT, AST, BUN, Creatinine, GLU, TC, TG, HDL, and LDL ( Figure 2F−N). Among them, AST in the 90% GCH group was significantly lower than that in |
the other two groups ( Figure 2G, P < 0.05), indicating that 90% GCH may strengthen the effect of HSYA on steatohepatitis induced by HFD. In addition, HDL in the 90% GCH group was higher than that in the other two groups ( Figure 2M, P < 0.05 or P < 0.01), whereas LDL in the 90% GCH group was lower than that in the other two groups ( Figure 2N, P < 0.05). These results revealed that 90% GCH could improve the treatment effect of HSYA on dyslipidemia. No significant differences in serum ALT, BUN, Creatinine, GLU, TC, and TG levels were observed among the three groups. Moreover, the 90% GCH group reduced the epididymal fat area and liver lipid droplet size compared with the HFD group. No morphological abnormalities were observed in the stomach, duodenum, jejunum, and ileum, indicating that 90% GCH did not impact the native mucus gel structure. Also, H&E staining images of the lung, spleen, heart, and kidney were identical in all groups ( Figure S1). Above all, 90% GCH could effectively and safely enhance the treatment effect of HSYA against HFDinduced obesity. We further analyzed the perturbation of endogenous serum metabolites using UHPLC−MS/MS. PCA analysis showed that the QC samples were clustered, suggesting that the experimental data were well controlled. The metabolic pattern of the 90% GCH group tended to be separate from that of the HFD group and H 2 O group ( Figure 3A). The PCA and PLS-DA analysis were further performed in the HFD group and 90% GCH group, The data is shown as mean ± SD. *P < 0.05, **P < 0.01, compared with the H 2 O group. The data is shown as mean ± SD; *P < 0.05, compared with the H 2 O group; # P < 0.05, compared with jejunum. a The data is shown as mean ± SD; *P < 0.05, compared with the same group in simulated gastric juice; # P < 0.05, compared with the 90% GCH group in simulated gastric juice on the second hour. and the metabolic patterns of the two groups were also significantly different ( Figure 3B,C). Subsequently, we screened the differential metabolites by VIP > 1, FC > 1.2 or FC < 0.833, and P value < 0.05. A total of 61 potential differential metabolites were identified between H 2 O and HFD groups ( Figure 3D). Importantly, compared with the HFD group, 90 increased metabolites and 129 decreased metabolites were identified in the 90% GCH group ( Figure 3E). The detailed information is shown in Table S4 of the Supporting Information. Hierarchical clustering analysis results showed that the metabolic patterns in HFD and 90% GCH groups were reversed ( Figure 3F). The metabolic pathway analysis further confirmed that arginine Figure 3G). DISCUSSION The oral bioavailability of HSYA is extremely low, mainly due to its good water solubility and poor lipid solubility, which made it difficult to penetrate the intestinal mucosa. Considering the good biocompatibility of choline-based NADES, 90% GCH was evaluated to enhance oral absorption of HSYA. In vivo pharmacokinetic results showed that 90% GCH significantly increased the oral absorption of HSYA in rats with an Fr at 326.08%. 90% GCH could affect not only the absorption rate constant but also the extent of absorption. The AUC values in the 90% GCH group were significantly increased compared with the H 2 O group, indicating that 90% GCH promoted the absorption extent of HSYA. Compared with the L-Pro-Am solvent, 11 the oral bioavailability of HSYA in 90% GCH was increased by 1.78 times, suggesting that 90% GCH was more effective. L-Proline is a non-essential amino acid of protein synthesis in the human body, whereas acetamide is clinically used for organic fluorine pesticide poisoning and belongs to class 2B carcinogen. Although Tong et al. reported L-Pro-Am as a drug delivery vehicle, its safety remains to be studied. 11 90% GCH is composed of choline chloride and glucose. Choline chloride can be used in the treatment of fatty liver and cirrhosis and as a feed additive. The content of choline in 90% GCH is lower than those used in dietary supplements. Therefore, we do not anticipate choline in 90% GCH to produce any significant biological effects. At the same time, our study demonstrated that 90% GCH is non-toxic and has good biocompatibility during the experiment. A previous study has reported that choline chloridebased NADES are eco-friendly, non-toxic, and biodegradable organic compounds, which maintain the biological activity of the target compound and are currently one of the most studied NADES in drug delivery applications. 21 Moreover, glucose and choline chloride have more hydroxyl groups, which provide strong stabilization ability originating from the formation of strong hydrogen bonding interactions between HSYA and themselves. 22 In addition, the viscosity of the simulated mucus treated with 90% GCH showed a notable drop throughout the entire measured shear range, suggesting that 90% GCH could be a potential mucus-modulating agent for improving the mucus penetration of HSYA. The in situ single-pass intestinal perfusion experiment further confirmed that the absorption rate constant of HSYA in the jejunum of the 90% GCH group was increased by 2.95 times, thereby increasing the oral absorption of HSYA. Consistent with the pharmacokinetic results, the stability experiments showed that the degradation of HSYA in 90% GCH in the simulated gastrointestinal juice within 2 h was slower than that in the H 2 O group, which may partially contribute to the high blood concentration maintained within 2 h. Previous studies demonstrated that choline-based NADES reduced mucus viscosity without significantly impacting the native mucus gel structure, suggesting that it would assist in mucus penetration in vivo. 8,23 Based on this information and our data, we believe that 90% GCH protects HSYA from enzymatic degradation, assists in the transport of HSYA through the mucous layer, and mediates the penetration of HSYA through the intestinal mucosa and increases the drug concentration into the blood. In addition, the absorption of HSYA is affected by the efflux transporter such as P-gp and BCRP in intestinal epithelial cells, which is also an important reason for the low bioavailability of HSYA. 15 Moreover, the intestinal flora is rich in enzyme systems, which participate in the catabolic process of HSYA in vivo. 24,25 The high viscosity of 90% GCH may protect HSYA from intestinal flora metabolism and lead to its increased oral bioavailability, but more evidence is needed. Based on serum biochemical and multiorgan histological analyses, HSYA in 90% GCH was more effective in alleviating body weight, steatohepatitis, and dyslidemia in obese rats, and no morphological abnormalities were observed in any other tissues. Therefore, we do not anticipate choline in 90% GCH to produce any significant biological effects, although additional studies are needed to fully understand the potential role of 90% GCH. The mechanism findings based on serum metabolomics revealed that the major metabolites affected by HSYA in 90% GCH were related to amino acid metabolism, which is closely related to obesity. 26 Generally, amino acids serve as the basic constituents of proteins and can also enter the tricarboxylic acid cycle through transamination or oxidative deamination to provide energy when glucose and lipids cannot be effectively utilized in diabetic or insulin-resistant states. 27 High levels of aromatic amino acids (AAA) are known risk factors for obesity and cardiometabolic diseases, which refers to phenylalanine, tyrosine, and tryptophan. Phenylalanine is a metabolic precursor to tyrosine via phenylalanine hydroxylase in the liver, which further metabolizes to neurotransmitters. 28 L-Tyrosine is one of the amino acids believed to be related to body mass index and is positively correlated with obesity in humans, and its level has been shown to decrease after weight loss. 29 Tryptophan is generally degraded by the kynurenine pathway. 30 AAA could convert to phenylpyruvic acid, 4-hydroxyphenylpyruvic acid, and indole-3-pyruvic acid by AAA aminotransaminase-mediated transamination. 31 In addition, tryptophan is also degraded and metabolized by gut bacteria into 3-methylindoles. 32 In this study, after HSYA in 90% GCH intervention, AAA and their metabolites, i.e., L-tyrosine, phenylpyruvic acid, and 3methylindoles were significantly decreased in obese rats. Moreover, histidine is an important amino acid in the body, which can be utilized to produce glucose through gluconeogenesis. 33 Thus, changes in histidine metabolism may lead to an imbalance of energy and glucose homeostasis. A recent study reported that obese individuals have a rich histidine metabolism. 34 Collectively, HSYA in 90% GCH can partially correct the amino acid metabolism disorders in HFD-induced obesity. CONCLUSIONS In this study, 90% GCH was developed to increase the oral bioavailability of HSYA in vivo with an Fr at 326.08%. The viscosity of the 90% GCH-treated simulated mucus was decreased significantly. Further analysis by in situ intestinal perfusion showed that 90% GCH significantly enhanced the small intestinal absorption rate constant of HSYA, in which the jejunum absorption rate was increased by about 2.95 times. The pharmacodynamic results showed that HSYA in 90% GCH effectively and safely attenuated steatohepatitis, corrected blood lipids, and improved glucose intolerance induced by HFD. Serum metabolite spectrum and pathway enrichment analysis revealed that HSYA in 90% GCH could partially restore the disorder of amino acid metabolism. In conclusion, our findings suggest that 90% GCH could be a promising oral drug delivery carrier for HSYA against cardiometabolic diseases. Matrix effect of HSYA and rutin (IS) in rat plasma (Table S1); precision, accuracy, and recovery for the analysis of HSYA in rat plasma (Table S2); stability of HSYA in rat plasma (Table S3); differential metabolites in serum samples between 90% GCH and HFD groups (Table S4); and representative hematoxylin and eosin staining images of the epididymal fat, liver, stomach, duodenum, jejunum, ileum, lung, spleen, heart, and kidney in different groups ( Figure S1) Punicalin Attenuates Breast Cancer-Associated Osteolysis by Inhibiting the NF-κB Signaling Pathway of Osteoclasts Background: Breast cancer bone metastasis and osteoporosis are both severe diseases that seriously threaten human health. These diseases are closely associated with osteolytic lesions. And osteoclasts are the key targets of this pathological process. Given the lack of effective preventive or treatment options against these diseases, the exploitation of new pharmacological agents is critically required. Method: We assessed the efficacy of punicalin on receptor activator of nuclear factor-κB ligand (RANKL)-mediated osteoclast formation, F-actin ring formation, gene expression, bone resorption, nuclear factor-κB (NF-κB) as well as on mitogen-activated protein kinase (MAPK) signaling pathways and molecular docking in vitro. The impact of punicalin on breast cancer-induced osteoclastogenesis, breast cancer cell proliferation, and apoptosis were examined. Transwell assays were also performed. Moreover, we evaluated in vivo effects of punicalin in postmenopausal osteoporosis models and breast cancer bone metastasis model by micro-CT scanning and histomorphometry. Results: Punicalin inhibited osteoclast formation, F-actin ring formation, bone resorption, as well as osteoclast-related gene expression by suppressing the NF-κB signaling pathway. In vitro, punicalin also suppressed the breast cancer-induced osteoclastogenesis, and proliferation, migration as well as invasion of MDA-MB-231 cells and dose-dependently promoted their apoptosis. In vivo, punicalin significantly suppressed breast cancer-induced osteolysis, breast cancer-associated bone metastasis, and ovariectomized (OVX)-mediated osteoporosis by repressing osteoclast and breast cancer cell. Conclusion: Punicalin is expected to offer a novel treatment for the prevention of osteolysis diseases, including osteoporosis and breast cancer-associated osteolysis. INTRODUCTION Globally, among women, the prevalence of breast cancer is the highest and breast cancer is one of the tumors with the highest morbidity and mortality rates (Jemal et al., 2007). The bone is prone to invasion by breast cancer, and nearly 75% of advanced breast cancer patient develop bone metastasis (Costa and Major 2009;Casas et al., 2013). Although bone metastases themselves rarely directly cause death in breast cancer individuals, serious complications of bone metastasis, such as hypercalcemia, chronic pain, incontinence and paralysis caused by spinal cord compression due to pathological fractures, all seriously affect life quality and breast cancer prognostic outcomes (Coleman, 1997;Coleman et al., 1998;Coleman, 2001). In recent years, progress in cancer treatment has increased life expectancy and, on the contrary, increased the risk of bone metastases (Nakai et al., 2019). There is still a lack of ideal therapeutic methods for breast cancer-associated bone metastasis. The current therapeutic approach involves surgical removal of the bone metastatic lesion in combination with drug therapy (Zhang et al., 2020). However, the incidence of perioperative complications is high, and the outcomes are not entirely satisfactory (Wegener et al., 2012). During breast cancer |
bone metastasis, osteoclasts, rather than tumor cells, are responsible for osteolysis (Taube et al., 1994). Receptor activator of nuclear factor-κB ligand (RANKL), which belongs to the tumor necrosis factor (TNF) superfamily, is a vital role in differentiation of osteoclast in the bone microenvironment and is a competitor for osteoprotegerin (OPG) when binding receptor activator of nuclear factor-κB (RANK) on osteoclast precursor surfaces, leading to further induction of TRAF6 activation Nakai et al., 2019). The downstream signaling pathway involved in TRAF6-induced osteoclast differentiation, comprising activation of nuclear factor-κB (NF-κB) as well as mitogen-activated protein kinase (MAPK) signaling pathways, upregulate expression levels of osteoclastogenesis-related transcription factors in the nucleus (He et al., 2014;Koga et al., 2019). Metastatic breast cancer cells cause increased expression of RANKL by releasing cytokines, which leads to overactivation of osteoclasts and pathological osteolysis (Roodman, 2004). Some cytokines are released during osteolysis, which enhance breast cancer cell growth, leading to a "vicious cycle" (Roodman, 2004). Therefore, activation of RANKL signaling pathway plays a central function in breast cancer bone metastasis. Blocking any link of this signaling pathway could be a possible therapeutic strategy for breast cancer-associated bone metastases, which can effectively prevent and cure osteolysis caused by breast cancer. Antiosteoclast bisphosphonates and denosumab can slow down the progression of metastatic breast cancer-induced osteolysis (Stopeck et al., 2010;Van Poznak et al., 2011). Unfortunately, these drugs may cause adverse reactions, including renal impairment and jaw osteonecrosis. Bisphosphonates are not ideal for all breast cancer patients (Marx et al., 2005). In addition, anti-bone resorption treatment is palliative and does not inhibit the invasion of tumor cells. Chemotherapy can suppress cancer cell metastasis, however, the lack of antiosteoclast activities is one of its defects (Valkenburg et al., 2013). Therefore, to inhibit breast cancer-induced osteolysis as well as bone metastasis, the current approach mainly uses combination therapy . Development of a drug with the ability to suppress both osteoclast differentiation as well as breast cancer progression will promote the treatment and prevention of breast cancer bone metastasis. We found that PNC suppresses RANKL-associated osteoclastogenesis (in BMMs, splenocytes, and RAW 264.7 cells), bone resorption, F-actin ring formation, as well as osteoclast-specific gene expressions by suppressing the NF-κB signaling pathways in vitro. PNC also inhibited MDA-MB-231 human breast cancer cell proliferation, migration, invasion and tumor cell-mediated osteoclast development. Moreover, it promoted their apoptosis in vitro. In ovariectomized (OVX) and breast cancer bone metastasis mouse models, treatment with PNC markedly prevented the breast cancer-induced osteolysis and OVX-mediated osteoporosis in vivo. Thus, PNC is a potential treatment option for osteoporosis and metastatic breast cancer-induced osteolysis. Materials, Reagents and Cell Culture PNC (>98% pure, Figure 1H) was acquired from Chengdu MUST Biotechnology (Sichuan, China) and dissolved in alpha minimum essential medium (a-MEM; Gaithersburg, MD, United States) to establish a 62.5 mM solution maintained at 4°C. Bovine bone slices we use are purchased from Shanghai Lushen Biotechnology Co., Ltd. Soluble mouse recombinant macrophage colony-stimulating factor (M-CSF), RANKL, fetal Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 789552 3 bovine serum (FBS), and penicillin as well as streptomycin were acquired from R&D Systems (Minneapolis, MN, United States). DMSO and TRAP staining kits were acquired from Sigma-Aldrich (St. Louis, MO, United States) while cell counting kits (CCK-8) were procured from KeyGen Biotech (Nanjing, China). The remaining reagents were bought from Sigma Aldrich, except if stated otherwise. RAW 264.7 cells, an osteoclast precursor, were bought from American Type Culture Collection (ATCC; Rockville, MD, United States) and cultured in α-MEM medium with 1% penicillin/streptomycin and FBS (10%). For induction of osteoclast differentiation in vitro, genetically pure C57BL/6 female mice (4-6 weeks in age) were procured from Hunan Silaikejingda Experimental Animal Co., Ltd. (Changsha, China) and used to extract 1) primary splenocytes from ground spleen tissues and 2) primary bone marrow macrophages (BMMs) from tibias as well as femurs. The primary splenocytes as well as the BMMs were incubated in α-MEM with M-CSF (30 ng/ml). MDA-MB-231 cells were procured from the American Type Culture Collection (Rockville, MD, United States) and incubated in Dulbecco's modified Eagle's medium (DMEM) that had been supplemented with 10% FBS and antibiotics. All cells used in this study were kept under sterile conditions and a persistent high-humidity, 5% CO 2 atmosphere at 37°C. Removal of nonadherent cells was done prior to each passage. Cell Viability We detected PNC-associated cytotoxic effects on three osteoclast precursor cell types (RAW 264.7 cells, splenocytes, BMMs) and MDA-MB-231 cell proliferation by conducting the CCK-8 assays. In Vitro Osteoclastogenesis Assessment To adequately investigate inhibitory effects of PNC on RANKLmediated osteoclast formation, three osteoclast precursor cell types (RAW 264.7 cells, BMMs, and splenocytes, 3×10 3 cells per well) were cultured in 96-well plates after which they were exposed to non-toxic PNC concentrations, ranging from 0 to 125 μM. Stimulation of preosteoclasts was done for up to 7 days using RANKL (50 ng/ml). For evaluation of PNC-mediated attenuation of MBA-MD-231 human breast cancer cellmediated formation of osteoclasts, we first mixed RAW264.7 cells and MDA-MB-231 cells to contain 2 × 10 3 RAW 264.7 cells and 1 × 10 3 MDA-MB-231 cells per 100 ul. Then the complete medium containing two kinds of cells was dropped into the 96well plate to 100 ul per well. Meanwhile, these cells are treated with PNC (0, 32.2, 62.5, or 125 μM), co-cultured for a total of 7 days after which TRAP staining was performed. After osteoclastogenesis, cell fixing was done in paraformaldehyde (PFA, 4%) and branded by staining with tartrate-resistant acid phosphatase (TRAP). In this study, osteoclasts with ≥3 nuclei were categorized as TRAP-positive, implying mature, differentiated multinucleated osteoclasts. Bone Absorption Pit Evaluation Culture of RAW 264.7 cells (3,000 cells/slice) was done on sterile bovine bone slice surfaces in 96-well culture plates and subjected to PNC at diverse concentrations (from 0 to 125 µM). Furthermore, for osteoclast formation, RAW 264.7 cells were treated with RANKL (50 ng/ml). After the formation of mature osteoclasts, bone slice-adhered cells were eliminated through mechanical agitation as well as sonication. Finally, investigation of osteolytic absorption pits was done by scanning electron microscopy (SEM, FEI Quanta 250). Quantitative analyses (%) of resorbed bone surface areas were done by the ImageJ software (National Institutes of Health, Bethesda, MD, United States). Extraction of RNA and Quantitative Polymerase Chain Reaction Analysis For analysis of osteoclast-specific gene expressions, inoculation of RAW 264.7 cells was done in a six-well plate and treated with PNC (0, 31.2, 62.5 or 125 μM). Then, total RNA extraction was conducted using the RNeasy Mini Kit (Qiagen, Valencia, CA, United States) as directed by the manufacturer. cDNA synthesis was done by a reverse transcriptase kit from TaKaRa Western Blotting To evaluate the impact of PNC on signaling pathways, RAW 264.7 cells that had been inoculated in 6-well plates (5 × 10 5 cells/ well) in complete α-MEM supplemented with 1% streptomycin/ penicillin, 10% FBS were incubated. After verification of the appearance of osteoclasts, cells were incubated for 4 h with or without 125 μM PNC treatment followed by incubation with RANKL (50 ng/ml) for 0, 5, 10, 20, 30, or 60 min. The medium was removed and rinsed twice with phosphate buffer. Radioimmunoprecipitation assay (RIPA) lysis buffer (Well Biology, Changsha, China) and a protease inhibitor cocktail were used to extract total proteins from the cultured cells. Then, centrifugation of the lysate was done for 15 min at 12,000 rpm after which the supernatant with the protein was obtained. Measurement of protein concentrations was done using the bicinchoninic acid (BCA, Thermo Fisher, MA, United States) assay. A specific amount of every cell lysate (30 mg) was dissolved in sodium dodecyl sulfate polyacrylamide gel (10%). Products were transferred to polyvinylidene difluoride membranes (Millipore; Bedford, MA, United States) that were later blocked for 2 h using 5% non-fat milk powder in TBS-Tween (TBS: 0.05 M Tris, 0.2% Tween-20, 0.15 M NaCl, and pH 7.5). Overnight incubation was done at 4°C in the presence of primary antibodies. Then, TBS-Tween was used to wash the membranes followed by incubation for 2 h in the presence of IRDye 800CW (molecular weight 1,162 Da)-conjugated secondary antibodies. Culturing of MDA-MB-231 cells was done in 6-well plates (density: 5 × 10 5 cells/well) with DMEM, antibiotics as well as 10% FBS. For 2 days, they were treated with PNC (0, 62.5, or 125 μM). Following treatment, C-caspase-3 activities were measured by the Caspase Colorimetric Assay Kit (KeyGen Biotech) as instructed by the manufacturer. Briefly, cell lysates (about 0.1 mg total protein) were supplemented to the reaction mixture, including caspase-3 colorimetric substrate peptides. Incubation for 4 h was done at 37°C. Finally, visualization of proteins on the blots was done by an Odyssey infrared imaging system (LI-COR, Nebraska, United States). Band intensities were measured and quantified using ImageJ software. NF-κB Luciferase Reporter Gene Assay For determination of if PNC disturbed NF-κB signaling pathway activation in RAW 264.7 cells, an NF-κB luciferase reporter construct was transfected in cells. RAW 264.7 cells (1 × 10 5 cells per well) were inoculated in a 24-well plate, followed by 24 h of incubation and treated for 1 h with PNC (0, 31.2, 62.5, or 125 μM) prior to incubation with RANKL (50 ng/ml). Then, cell lysates were incubated for 2 min at room temperature using substrate (Promega, Madison, WI, United States). We used the Promega Luciferase Assay System (Promega) to measure luciferase activities as instructed by the manufacturer. The control group was used to normalize luciferase activities. Experiments were done in triplicates. Molecular Docking Analyais Using the structure of IκBα as a template, mouse IκBα kinase domain homology models were constructed using Modeler 9.12. Based on AutoDock and AutoDock Vina (Trott and Olson, 2010;Ouyang et al., 2019), PROCHECK was utilized to demonstrate stereochemical architectures of IκBα, while creation of the link between IκBα and kinases was done using the Lamarckian genetic algorithm. Molecular docking figures showing binding were established using a PyMOL Visualization Software (Schrödinger LLC, New York, NY, United States). Micro-CT Scanning After euthanasia of the mice treated with PNC, the tibia and fibula were obtained and fixed for 48 h in 4% paraformaldehyde. Then, tibia and fibula of the mice were analyzed by high-resolution micro-CT. The scanning parameters were: minimum resolution of 10 μM, 80 kV scanning voltage, and scanning current of 80 mA (Ouyang et al., 2019). After the region of interest in the proximal tibia near the tibial plateau was measured, the data were obtained, including relevant bone surface/volume ratio (BS/BV), bone mineral density (BMD), trabecular number (Tb.N), and trabecular space (Tb.Sp), bone volume/tissue volume (BV/TV), trabecular thickness (Tb.Th). Histological and Histomorphometric Analyses Mouse tibia and femur samples undergoing micro-CT scanning analysis were decalcified for 5 weeks using 10% EDTA and embedded in paraffin. Tissue sections were used for TRAP as well as H&E staining. High-resolution microscopy was performed for imaging and specimen examination. Finally, the abundance of TRAP-positive multinucleated osteoclasts in each specimen was determined. The ratio of the trabecular osteoclast number to bone surface (Oc.N/BS, N/mm) and the percentage of osteoclast surface to bone surface (Oc.S/BS,%) were measured on trap stained sections at 200-fold magnification (Sakata et al., 1999). Under microscopy (Leica image analysis system, Q500MC), ImageJ (NIH, United States) was used for bone static histomorphometric evaluation of tumor area, Oc.N/BS and Oc.S/BS. Apoptosis Assay PNC-mediated apoptotic outcomes on MDA-MB-231 cells were assayed by the Vybrant Apoptosis Assay Kit #2 (Invitrogen). After treatment with PNC (0, 62.5, 125, and 250 μM) for 48 h, MDA-MB-231 cells were rinsed twice using sterile PBS after which they were pelleted. Supernatants remained unused and were discarded while cells resuspended in 1X Annexin-binding buffer. To evaluate early apoptosis, cell staining was done using propidium iodide and Alexa Fluor 488 Annexin V using the Vybrant Apoptosis Assay Kit #2 (Invitrogen). Fluorescenceactivated cell sorting (FACS) was carried out by FACScan flow cytometry (Becton-Dickinson, Sunnyvale, CA, United States). Data were obtained using the CELL Quest software. Invasion and Migration Assay BioCoat ™ Matrigel ™ Invasion Chambers (BD Biosciences) with 24-well chambers and pore polycarbonate filters (8-µm) as well as Transwell ® permeable supports (Corning, Inc., Acton, MA, United States) with 24-well chambers and pore polycarbonate filters (8-µm) were utilized in invasion and migration assays as instructed by the manufacturers. The MDA-MB-231 cells (4 × 10 4 ) were seeded in serum-free medium (100 µl) with varying concentrations of |
PNC (0, 62.5, 125, and 250 µM), and 500 µl of complete medium was seeded in the lower wells. After cultivation for 24 h, MDA-MB-231 cells in the top chambers were eliminated using a cotton swab. Invading and migrating cells at the bottom of the filter were fixed for 30 min in paraformaldehyde (4%) and stained for 6 min using crystal violet (1%). Pictures were obtained by Olympus inverted microscopy. Invaded and migrated MDA-MB-231 cells were counted by ImageJ software. Ovariectomized Mouse Model Animal experimental procedures involved in this study were all permitted by the Animal Care Committee of Central South University. Female C57BL/6 mice (8 weeks old, female, n 30) used in the experiment were acquired from Hunan Silaikejingda Experimental Animal Co., Ltd., (Changsha, China). The mice were placed in plastic cages in specific pathogen-free (SPF) conditions for 1 week to acclimate to the new environment before undergoing OVX surgery. Mice (n 20) were anaesthetized by intraperitoneal injection of pentobarbital (40 mg/kg, Sinopharm Shanghai Co., Ltd.) preoperatively, followed by bilateral ovariectomy (Ouyang et al., 2019). After OVX resection, mice were randomly divided into three groups: the sham non-OVX mice, the vehicle-treated OVX mice, PNCtreated OVX mice administered with 5 mg/kg/day (PNC intraperitoneally), with n 10 per group. Concentrations of PNC were determined after a preliminary screening in vivo by referring to other studies (Lin et al., 1999). Sham and vehicle mice were injected with sterile PBS. The injected mice that were closely observed for 4 weeks were euthanized at the end of the experiments. Finally, the right tibia of the mice was harvested for micro-CT scanning. Intratibial Xenograft Model of Breast Cancer Bone Metastasis Female BALB/c-nu/nu mice (6 weeks old, n 30) used in the protocol were acquired from Hunan Silaikejingda Experimental Animal Co., Ltd., (Changsha, China) and kept in plastic cages under pathogen-free (SPF) environments for a week. After acclimating to the new facility and environment, the mice involved in the study were randomized into three groups: placebo control group (sterile PBS, n 10), vehicle group (sterile PBS, n 10), and PNC group (5 mg/kg/day, intraperitoneally, n 10). The concentration of PNC used was determined after a preliminary screening in vivo. Sham and vehicle mice were injected with sterile PBS for 28 days. The PNC-treated and vehicle-treated mice were administered with MDA-MB-231 (10 7 cells/ml) in the tibial plateau of the right leg to construct a bone metastasis model of breast cancer. Finally, injected mice were nurtured for 28 days and euthanized, and the right tibia of the mice was harvested for micro-CT scanning. Statistical Calculations All experiments were separately carried out in replicates of three or more. Data are shown as mean ± SD. Determination of significance was done by the Student's t-test in SPSS 22.0 software (SPSS, Inc., United States). p < 0.05 was considered significant. Punicalin Suppressed Osteoclast Function and Formation at Noncytotoxic Concentrations In Vitro To confirm PNC-associated effects on osteoclasts rather than a toxic effect, we first determined the nontoxic concentrations of PNC against preosteoclasts through the cell counting kit-8 (CCK-8) test. Figure 1A shows that there were no cytotoxic effects on the three osteoclast precursor cell types (RAW 264.7 cells, BMMs, and splenocytes) when PNC concentrations were ≤125 μM. Based on the above experimental results, nontoxic concentrations of 31.2, 62.5, and 125 μM were selected to carry out the following anti-osteoclastogenesis experiments. We observed that, in the control group, despite significant osteoclastogenesis showing round, large, and red-stained multinucleated osteoclasts, addition of PNC dose-dependently diminished the area and Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 789552 6 number of the mature TRAP-positive osteoclasts from RAW 264.7 cells, BMMs, and splenocytes ( Figures 1B,C), indicating that a nontoxic concentration of PNC could inhibit osteoclastogenesis. In addition, well-shaped F-actin rings are essential for osteoclast function, so we investigated the impact of PNC treatment on the formation of F-actins. In the control group, RAW 264.7 cells produced polarized actin rings after RANKL stimulation, while PNC treatment decreased the number and size of the F-actin rings, suggesting that PNC influenced osteoclast function ( Figures 1D,E). This finding is consistent with osteolysis on bone slices as detected by scanning electron microscopy ( Figures 1F,G), showing that deep and big bone resorption pits were present in RANKL stimulated control group. The addition of PNC reduced the size and number of bone resorption pits. The above results indicate that the function and formation of osteoclasts are effectively inhibited by nontoxic concentrations of PNC and encouraged us to investigate potential effects of PNC on the formation of osteoclasts. Punicalin Inhibited Osteoclast-Specific Gene Expression Levels In Vitro Expressions of the specific genes involved in differentiation of osteoclast were activated after stimulation using RANKL. The effect of PNC on the formation of osteoclasts was done by examining RANKL-induced mRNA expressions of Calcitonin FIGURE 2 | PNC impaired RANKL-mediated expressions of specific genes in osteoclastogenesis. (A) Diverse concentration of PNC (0, 32.2, 62.5, 125 μM) mitigated the transcription of osteoclast-specific genes, including CTR, Cathepsin K, TRAP, V-ATPase-d2, NFATc1, and c-fos. (B) with or without 125 μM PNC, for 0, 1, 3, or 5 days separately. Expressions of specific genes associated in osteoclast formation were determined through real-time PCR and normalized to β-actin. **p < 0.05, **p < 0.01 compared to the control group. Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 789552 receptor (CTR), NFATc1, TRAP, Cathepsin K, V-ATPase-d2, and c-fos. In the control group, expressions of all genes were induced by RANKL, while PNC treatment time-and dosedependently downregulated osteoclast gene expression (Figures 2A,B). The above experimental data further demonstrated that PNC at nontoxic concentrations inhibited specific gene expressions in osteoclast formation in vitro. PNC Inhibited RANKL-Associated NF-κB Activation During Osteoclast Differentiation To investigate specific inhibitory effects of PNC on osteoclasts, we evaluated two important pathways involved in the formation of osteoclasts, namely NF-κB and MAPK signaling pathways (Hu et al., 2020). Therefore, we tried to establish the impact of PNC on expressions of NF-κB as well as MAPK signaling pathways through Western blots to confirm the mechanism through which PNC suppresses osteoclastogenesis. As shown in Figures 3A,C, the addition of PNC increased IκBα expression and reduced the ratio of phosphorylated IκBα compared with those in the RANKL stimulation group. IκBα is used to stabilize NF-κB. This molecule binds to cytoplasmic p65 of NF-κB, allowing NF-κB to remain stable (Hu et al., 2020). Therefore, the NF-κB pathway was inhibited by PNC treatment. From the above experimental results, we determined the probable binding sites between PNC and IκBα by molecular docking. As expected, PNC established molecular bonds with multiple amino acids, including Ala-102, Gln-107, Gln-111, Asp-136, Asn-105, Gln-154, Gly-155, Leu196, Gly197, and Ile-198 ( Figure 3B). These findings imply that PNC may suppress RANKL- mediated NF-κB activation, thereby reducing the formation of osteoclasts. The inhibitory impacts of PNC on the NF-κB signaling pathways were confirmed by NF-κB luciferase reporter gene analysis ( Figure 3D). As shown in Figure 3A, extracellular signal-regulated kinase (ERK) as well as c-Jun N-terminal kinase (JNK) signaling pathways seemed to be partially suppressed, and the p38 signaling pathways seemed to be partially activated, but the results obtained through statistical analysis were not statistically significant ( Figure 3C). In summary, PNC suppresses the downstream NF-κB signaling pathway of RANKL by targeting IκBα binding during osteoclast formation. Punicalin Attenuated Bone Loss by Suppressing Osteoclast Formation in Ovariectomized Mice We created an OVX mouse model to investigate suppressive effects of PNC on osteoclasts in vivo. Through Micro-CT scanning, it was revealed compared to the sham group, osteoporosis was found in proximal tibia of the vehicle group, with severe bone loss as well as incomplete bone trabecular structure ( Figure 4A). In contrast, a slight bone loss as well as complete trabecular structures were observed in the PNC treatment group. Subsequently, we analyzed the bone parameters of bone trabeculae specimens in the proximal tibia, and we found that compared to the vehicle group, trabecular number (Tb.N.), bone mineral density (BMD), trabecular thickness (Tb.Th.), as well as bone volume/tissue volume ratio (BV/TV) were markedly elevated in the PNC treatment group, while trabecular spacing (Tb.Sp) and bone surface/volume ratio (BS/BV) was decreased ( Figure 4B). The osteoprotective effect of PNC was further confirmed by histological morphological analysis. Figures 4C,D shows that TRAP positive cells in the vehicle group were increased in number and area, with severe trabecular structure impairment and obvious bone loss. In the PNC treatment group, the number as well as osteoclast areas were significantly reduced, and bone resorption were less severe, which was comparable to micro-CT scan results. According to the above results, PNC effectively alleviated bone loss by inhibiting osteoclasts in OVX mice. Punicalin Inhibited Breast Cancer Cell Proliferation, Migration, Invasion, Breast Cancer-Mediated Osteoclast Differentiation, and Promoted the Apoptosis of Breast Cancer Cells Figure 5A shows that after PNC treatment for 24 h, only PNC at concentrations of 1,000 μM showed toxic effects on tumor cells, but with increasing culture time, both 500 and 1,000 μM PNC showed toxic effects. This finding provides a potential range of options for us to delineate the therapeutic effect of PNC in breast cancer. Figures 5B,C shows that the early apoptotic effect of 62.5 μM PNC on MDA-MB-231 cells was observed compared to the control group. Moreover, MDA-MB-231 cell early apoptosis were elevated at higher concentrations of PNC (125 and 250 μM). The effect on late apoptosis was also observed after PNC treatment. The above studies confirmed that PNC at a nontoxic concentration can effectively promote human breast cancer cell apoptosis. To further study whether PNC has suppressive effects on osteoclast differentiation induced by MDA-MB-231 cells, we cocultured MDA-MB-231 cells and RAW 264.7 cells. Figures 5D,E shows that in the control group without PNC, after co-culture in the presence of MDA-MB-231 cells, the RAW 264.7 cells differentiated into osteoclasts. In contrast, after 31.2 μM PNC treatment, the induction effect was significantly inhibited, and the abundance of TRAP-positive cells was suppressed. With increasing PNC concentrations, inhibitory effects became more obvious. According to the microscopic osteoclast count, significant differences occurred between the control and treatment groups at the three concentrations of PNC, 31.2, 62.5 and 125 μM. To illustrate the underlying mechanisms of PNC-mediated apoptosis, we investigated the functions of antiapoptotic Bcl-2 protein, an important Bcl-2 family member (Cory and Adams, 2002). Protein expression levels of Bcl-2 were decreased after PNC treatment at concentrations of 62.5 and 125 μM ( Figures 5H,I). Thus, we concluded that PNC modulates Bcl-2 expressions in MDA-MB-231 cells. Caspase-3 is a terminal executor of apoptosis belonging to the caspase cascade. Once cleaved, caspase-3 activates and triggers programming cell death (Emoto et al., 1995;Xue et al., 1996;Porter and Jänicke, 1999). To establish the function of the caspase-3 cascade in PNC-mediated apoptosis, we evaluated cleaved caspase-3 (active form of caspase-3) levels by western blots after PNC treatment. Figures 5H,I shows that PNC increased caspase-3 cleavage at 125 μM, while the lower concentration groups (62.5 μM) had no significant increase compared to the control group. Thus, in our study, the abundance of apoptotic cells was found to be increased after cells had been treated with 62.5 and 125 μM PNC. Punicalin Prevented Osteolysis in MDA-MB-231 Breast Cancer Cell-Derived Tumor-Bearing Mice In Vivo After the dual activity of osteoclast differentiation and tumor metastasis was inhibited by PNC in vitro, we further studied murine models of breast cancer bone metastases to establish the therapeutic effects of PNC. MDA-MB-231 cells were inoculated from the tibial plateau into vehicle-treated or PNC-treated nude mice to create a model of breast cancer bone metastasis. Four weeks after the injection of PNC (5 mg/kg/day), we performed micro-CT scanning near the proximal tibia of the mice and found severe bone erosion in the proximal tibia of tumor-bearing mice Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 789552 of the vehicle group without PNC treatment, while PNC treatment significantly relieved the osteolytic lesions near the proximal tibia ( Figure 6A). Next, bone parameters, such as BV/ TV, Tb.Sp, BMD, Tb.Th., BS/BV as well as Tb.N., were analyzed ( Figure 6B). Compared to the vehicle group, bone mass in the PNC treatment group was significantly higher, while the BS/BV and Tb.Sp values of the vehicle group were higher. The dual activity of |
PNC in inhibiting osteoclast differentiation and tumor metastasis was further demonstrated by histological section staining. Figure 6C shows that bone cortex of proximal tibia of the tumor-bearing mice in the vehicle group was traversed by tumors, and the metaphysis was also damaged, which led to tumor invasion into the luminal growth of the knee joint. The addition of PNC protected the metaphyseal and cortical integrity and prevented the tumor from growing into the articular cavity of the knee joint. TRAP staining showed that compared with vehicle-treatment, PNC treatment significantly reduced the number of osteoclasts. Bone morphologic analysis of the proximal tibia, including the tumor area, Oc.N/BS and Oc.S/ BS, further confirmed that PNC prevented breast cancer-induced pathological bone destruction ( Figure 6D). DISCUSSION We determined that PNC can suppress RANKL and tumormediated osteoclastogenesis, suppressed breast cancer cell migration, invasion, and proliferation, induce their apoptosis, and exert a systemic and local anti-osteoclast effect in vivo, which is mainly mediated through the NF-κB signaling pathway. Therefore, the application of PNC might be a novel approach for preventing and treating osteolysis-associated diseases, such as osteoporosis and metastatic breast cancer-induced osteolysis. To rule out drug toxicity effects on osteoclasts, we first tried to find a safe concentration of PNC. Studies have shown that PNC is safe at concentrations of 0-100 μM and has a slight inhibitory effect on mouse mononuclear macrophage J774A.1 cells at 150 μM (Shen et al., 2021). Consistent with the results, our study found that the nontoxicity concentration of PNC on preosteoclasts (BMMs, RAW 264.7 cells, and splenocytes) was lower than 125 μM ( Figure 1A). However, the effects of PNC on differentiation of osteoclasts has not been clearly established. Therefore, we focused on its effect on the formation of osteoclasts. In vitro, PNC inhibited RANKL-mediated osteoclast formation as well as bone resorption. To confirm the suppressive impact of PNC on the differentiation of osteoclasts, we studied the specific genes and signaling pathways related to osteoclasts. As major regulators of osteoclast differentiation, NFATc1 and c-fos led to expressions of osteoclast-associated genes, including V-ATPase-d2, Cathepsin K, TRAP, and CTR, which are also involved in bone resorption activities of osteoclasts (Boyle et al., 2003;Crotti et al., 2006;Yavropoulou and Yovos, 2008). Expression levels of these genes were suppressed. In addition, the process of osteoclast formation involves NF-κB as well as MAPK signaling pathways activation, the latter of which includes JNK, ERK, and p38 (Hu et al., 2020). After stimulation with RANKL, extracellular stimulation was transduced to the nucleus through NF-κB, ERK, JNK, and p38 phosphorylation, which led to differentiation into mature osteoclasts. It has been reported that PNC has suppressive effects on both NF-κB as well as MAPK signaling pathways (Afaq et al., 2005). We found that PNC suppresses NF-κB signaling pathway but has a minimal impact on MAPK signaling pathways. This difference may be due to different cell lines. A luciferase assay confirmed the suppressive effects of PNC on NF-κB signaling pathway. Then, we identified the underlying binding sites between PNC and NF-κB pathway through molecular docking analysis. It was found that PNC was embedded in the IκBα binding pocket by establishing molecular binding with certain amino acids. Then, we assayed the systematic effects of PNC in an OVX mouse model. Micro-CT, TRAP and histological staining assays all showed that PNC significantly reduced bone loss in the OVX mice by suppressing osteoclastogenesis. Therefore, suppressive effects of PNC on osteoclasts was verified in vitro as well as in vivo. Due to the "vicious cycle" in breast cancer bone metastasis (Roodman, 2004), we evaluated the effects of PNC on breast cancer cell-mediated osteoclastogenesis and on breast cancer cells. PNC inhibited osteoclast formation induced by human breast cancer MBA-MD-231 cells and that PNC could suppress the proliferation, invasion, migration of MBA-MD-231 cells and promote their apoptosis in a dose-dependent manner. Breast cancer cells can activate osteoclasts by producing RANKL or by stimulating osteoblasts to produce RANKL (Roodman, 2004;Ikeda and Takeshita, 2016;Ouyang et al., 2018). Our study showed that the inhibition of osteoclasts by drugs is due to suppression of the NF-κB pathway by PNCs, which is downstream of the RANKL pathway in osteoclast differentiation. Therefore, PNC might also inhibit osteoclast differentiation by inhibiting downstream signaling of RANKL secreted by breast cancer and osteoblasts. Consistent with a previous study, in the coculture system, we found that osteoclasts were fewer in number and smaller in volume than those of the osteoclast differentiation previously induced by RANKL ( Figures 1B, 5D), which may be because our coculture system did not have osteoblasts, thus resulting in less osteoclastogenesis (Zhai et al., 2014). From the above experimental findings and the theory of a "vicious cycle," we hypothesized that PNC could prevent bone metastasis in murine tumors by suppressing osteoclast formation and function as well as the activity of MBA-MD-231 cells. Therefore, we established a bone metastatic model of breast cancer in nude mice to further verify whether PNC has a bone protective effect. Micro-CT scans established that, compared to the vehicle group, bone volumes of tumor-bearing mice in the PNC group were significantly higher. Furthermore, histological staining of the tumor-bearing mice showed that compared to the vehicle group, tumor growth was limited in bone marrow cavities and osteoclast formation near the binding site was suppressed in the PNC treatment group. Combined with the above experimental results in vitro as well as in vivo and the theory of a "vicious cycle," we believe that PNC can inhibit bone metastases by inhibiting osteoclast formation, function as well as MBA-MD-231 cell proliferation, migration, and invasion and promoting their apoptosis. Previous drugs can inhibit osteoclasts Frontiers in Pharmacology | www.frontiersin.org November 2021 | Volume 12 | Article 789552 to prevent bone metastases in breast cancer, while PNC may be more effective because it targets both osteoclasts and breast cancer cells (D'Oronzo et al., 2019;Nakai et al., 2019). However, our investigation still has many limitations. First, an ideal breast cancer bone metastasis model should be able to reproduce all tumorigenesis as well as bone metastasis stages, but a mouse model of immunodeficiency might omit immune system regulation of the metastatic stage of breast cancer. Studies have used the clinically important 4T1-derived syngeneic mouse models of breast cancer bone metastases (Bidwell et al., 2012;Lee et al., 2014); however, there is a remarkable difference between the mouse and human microenvironments. In this paper, bone metastases were reproducibly generated in breast cancer bone metastasis animal models (Mundy 2001;Siclari et al., 2014), rarely occurred in other sites, which is valuable for our study. In addition, after PNC treatment of animal models, local histology revealed that there were no significant adverse effects or deaths in the PNC or other groups. However, the systemic side effects of PNC were not investigated. Therefore, in-depth studies involving more parameters of tissue specificity and biocompatibility are needed. Bone metastasis mechanisms are complex and are related to interactions among metastatic breast cancer cells, osteoblasts, and osteoclasts (Chen et al., 2010;Futakuchi et al., 2016). However, we have not yet studied whether PNC affects osteoblasts. Therefore, more studies should be conducted on PNC effects on osteoblasts in breast cancer bone metastasis. The formation and activation of osteoclasts has a relationship with increased reactive oxygen species (ROS) in osteoclasts (Ha et al., 2004;Lee et al., 2005;Yip et al., 2005). PNC has been found to act on the ROS pathway (Shen et al., 2021). However, the ROS pathway was not studied in our experiment. Therefore, we also need to further investigate the effect of PNC on ROS in osteoclasts. In addition, since we mainly found that drugs have effects on osteoclasts, we chose to study only the signaling pathway of osteoclasts and not the signaling pathways related to breast cancer cells, indicating the need for further studies to investigate PNC effects on breast cancer cellassociated signaling pathways. At present, the suppressive mechanisms of PNC on breast cancer cells have not been thoroughly explored. Some metastasis-associated genes are related to bone-homing (CXCR4), osteoclast recruitment (IL11), invasion (ADAMTS1 and MMP-1), and angiogenesis (CTGF and FGF5) (Casimiro et al., 2012). Abilities of metastasis-related genes to regulate metastasis have not been established, and more studies should investigate the effects of PNC on these genes. Taken together, our work demonstrates for the first time that PNC can inhibit breast cancer cell invasion as well as migration and promote apoptosis in vitro, as well as attenuate osteoclast differentiation as well as function in vitro by blocking the NF-κB pathway. In vivo, PNC reduced breast cancer cell-mediated pathologic osteolysis in tumor-bearing mice, and decreased bone resorption in postmenopausal osteoporosis mice. Based on these findings, PNC may be a potential strategy for treatment of osteoporosis and breast cancer-associated osteolysis. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author. ETHICS STATEMENT The animal study was reviewed and approved by The Animal Care Committee of Central South University. Outcomes of endovascular interventional therapy for primary Budd-Chiari syndrome caused by hepatic venous obstruction To date, interventional therapy for patients with Budd-Chiari syndrome (BCS) due to hepatic vein obstruction (HVO) has not been standardized in China. In Western countries, BCS primarily occurs due to thrombosis and the majority of patients receive thrombolysis. In China, BCS is mostly caused by the membranous occlusion of the HV or IVC. The present retrospective study evaluated the efficacy of recanalization techniques in patients with primary BCS due to HVO. The data of 69 patients with BCS due to HVO, who underwent endovascular therapy at 2 centers in China between December 2010 and December 2012, were analyzed. All of the patients underwent balloon angioplasty. In addition, 14, 6 and 5 patients received thrombolysis, endovascular stent and thrombolysis + endovascular stent, respectively. The overall technical success rate was 95.7% (66/69), and was comparable among the treatments. The HV pressure after the treatments was significantly lower compared with that prior to the procedures (23.3±6.9 vs. 46.5±8.6 cmH2O; P<0.001). The mean follow-up duration was 75 months (range, 60–84 months). During the 5-year follow-up, 10 patients (15.2%) had developed a recurrence of BCS-associated symptoms, of which 7 were successfully treated. The cumulative survival rates at 12, 36 and 60 months after endovascular interventional therapy (balloon angioplasty or combined treatment) were 98.5, 98.5 and 93.9%, respectively. After treatment by endovascular therapy, the patients with BCS caused by HVO had high survival rates and low recurrence rates in the short- and mid-term. Introduction Budd-Chiari syndrome (BCS) is a clinical disorder caused by the obstruction of the hepatic vein (HV) outflow system, anywhere from the hepatic venules to the cavoatrial junction (1)(2)(3). BCS may be caused by obstruction of the HV and/or inferior vena cava (IVC) (1,3,4). Patients with BCS suffer from liver injury secondary to the obstruction of venous outflow, leading to progressive symptoms and even liver cirrhosis if not treated in a timely manner. Patients with BCS in the acute phase may succumb to hepatic failure. In chronic BCS, patients develop hepatocirrhosis leading to various complications, including gastrointestinal bleeding, refractory ascites and hepatocellular cancer (1,3,5). The treatment of BCS has evolved considerably over the past few decades, and the overall 5-year survival rate has risen to 80-90% (4,(6)(7)(8). Various treatment options for BCS include the following: i) Anti-coagulant therapy; ii) endovascular decompression therapy, including thrombolysis, stent-graft placement, angioplasty and transjugular intrahepatic portosystemic shunt (TIPS); and iii) orthotopic liver transplantation. Due to differences in the etiopathogenesis of BCS between China and western countries, treatment options vary widely. In western countries, BCS mostly occurs due to thrombosis and most patients receive thrombolysis, TIPS or liver transplantation (9,10). However, in China, BCS is mostly caused by membranous occlusion of the HV or IVC. Therefore, most patients with BCS in China undergo angioplasty (11)(12)(13). Formerly, BCS in China was thought to be due to IVC obstruction only. However, recent Chinese studies have indicated that most patients with BCS either present with an obstruction of the HV alone, or of the IVC and HV combined (11,14). Therefore, in China, the endovascular treatment of HV obstruction (HVO) associated with BCS has become a new challenge in clinical practice. While endovascular interventional therapy for patients with BCS due to IVC obstruction has been standardized (15,16), it is still evolving for patients with BCS due to HVO. Western countries advocate the use |
of TIPS as a primary treatment for HVO in BCS (3,9,10), while Chinese physicians prefer recanalization (4,13,17). The present retrospective study assessed the efficacy of recanalization in 69 consecutive patients with BCS due to HVO. Materials and methods Patient data. All procedures were performed in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5). All patients provided written informed consent prior to undergoing the procedure. The prospectively maintained data of patients diagnosed with BCS and treated at Anhui Provincial Hospital (Hefei, China) or Affiliated Hospital of Xuzhou Medical University (Xuzhou, China) and December 2010 and December 2012 were reviewed. For inclusion in the present study, patients were required to conform to the following: Primary BCS due to HV stenosis or HV occlusion as confirmed by magnetic resonance imaging ( Fig. 1) or digital subtraction angiography, and symptoms associated with portal hypertension. Patients with any of the following were excluded: Hepatic sinusoidal obstruction syndrome; IVC thrombosis; recurrent BCS; or BCS due to any other cause, including cancer, cysts or parasites. Patients who were lost to follow-up within 5 years were also excluded ( Fig. 2). Recanalization of the HV via transjugular and/or femoral vein approach. The venography of the IVC was performed first, via the internal jugular and/or femoral vein, in order to identify the ostium of the HV and correlate it with the pre-operative images. In patients with HV stenosis, a 5F Cobra catheter and an ultra-slip wire were used to explore the ostium of the target HV. For patients with complete HVO (left, middle or right), a single-bend catheter and a self-made single-bend needle ( Fig. 3A and B) were used to puncture the major HV (left, middle or right that was most affected by the obstruction) trunk, branch and/or traffic branch (i.e., the connecting vessel between two major HVs). After successful cannulation, HV angiography was performed, and the HV pressure was measured in order to assess luminal obstruction of the affected major HV and its branches. After complete assessment, balloon angioplasty (Figs. 4-6) and endovascular stenting were performed. The patients with BCS who underwent balloon angioplasty only, were included in balloon angioplasty group. The patients who received balloon angioplasty in addition to other methods of angioplasty, including thrombolysis and/orendovascular stenting, were included in the combination therapy group. Recanalization of HV via percutaneous transhepatic and IVC approach. After successful percutaneous puncture of one of the branches of the HV, the guide wire was passed through the obstructed segment of the HV and then drawn out from the internal jugular or femoral vein. Subsequently, recanalization of the HV was performed in the opposite direction through the IVC using an internal jugular or femoral vein approach. Recanalization of HV with thrombus. A 5F thrombolytic catheter was placed in the HV via a transjugular approach and thrombolysis was performed with urokinase (100,000 U; 4-6 times daily). Radiography was performed every 3 days for review, and the catheter was adjusted so that the catheter's lateral orifice was within the thrombus. Recanalization of the HV was performed after the thrombosis was completely dissolved, or if two consecutive reviews did not indicate any progression of the thrombosis. TIPS. Through the transjugular route, angiography was performed to identify the major HVs. If the major HV could not be identified by angiography of the HV, IVC and portal vein, angiography was performed using ultrasound-guided percutaneous liver puncture through an accessory HV. The puncture site of the HV or IVC and portal vein was selected based on the angiography results and TIPS was performed (9,10). Second-stage treatment. For patients with BCS who did not undergo HV angioplasty for the first time, a different treatment plan was selected based on the patient's liver function (Child's score) (18). If the patient's Child's score was less than 12 points, medical treatment (including diuresis and liver protection) was given. If medical treatment was effective, HV angioplasty was performed again following 6 months of medical treatment. If the patient's Child's score exceeded 12 points or medical treatment was not effective, TIPS or liver transplantation was considered. Post-angioplasty treatment. All patients received subcutaneous low-molecular-weight heparin (5,000 IU; twice daily) for 3 days, and then oral warfarin (5 mg; daily) for 12 months after treatment. The dose of warfarin was adjusted such that the prothrombin time was maintained at 20-25 sec. Success criteria for angioplasty. After interventional therapy, if the HVs featured a smooth blood flow and a transmembrane pressure difference of <4 cmH 2 O, the procedure was considered successful. Follow-up. The two centers routinely followed the patients up by telephone and through outpatient services every week for the first month, monthly for the next 3 months, every 3 months thereafter. Significant clinical events were recorded, including clinical deterioration, new radiographic signs on the liver and new BCS-associated interventions. Clinical deterioration was defined as re-admission after discharge, occurrence of new symptoms, recurrence of massive ascites, venous dilation over the trunk, leg edema, variceal bleeding or hepatic encephalopathy. The deadline for the follow-up was July 2017, the time-point of death or the time after which the patient was lost to follow-up. One assigned clinician was responsible for collecting the data for each of the participating patients. The recorded data included socio-demographic features, clinical manifestations, radiology results, interventional treatments and outcomes. Another clinician checked and assessed the data monthly. Categorical dataare presented as n (%) and quantitative data as the mean ± standard deviation. Quantitative data that conformed to a normal distribution (e.g., venous pressure) were analyzed using Student'st-test. Quantitative data that did not conform to a normal distribution were analyzed using the Wilcoxon rank-sum test, and qualitative data (e.g., ascites grade) were analyzed using the χ 2 test. Kaplan-Meier curves were drawn to analyze patient survival and the log-rank test was used to analyze differences in survival rates. P<0.05 was considered to indicate a statistically significant difference. Results Patient enrolment. From December 2010 to December 2012, 350 patients with BCS were treated at the two centers (Fig. 2). In accordance with the exclusion criteria, patients with secondary BCS, patients with BCS were diagnosed prior to admission and those with involvement of IVC were excluded. Of the remaining 80 patients with BCS with HVO only, 11 were excluded from the study due to insufficient follow-up. Thus, 69 patients (43 male, 26 female; mean age, 43 years; mean duration of symptoms, 98 months) were included in the analysis. The baseline clinical characteristics and laboratory tests of patients with BCS are presented in Tables I and II). Success rate of endovascular therapy. Among the 69 patients, primary HV recanalization was successful in 63 cases. Primary therapy failed in 6 patients due to extensive HVO. Of these patients, 2 with severe symptoms received TIPS (post-operatively, the symptoms of one patient resolved and the other one succumbed to liver failure). The remaining 4 patients with mild symptoms received conservative treatment (3 successfully received second-stage recanalization 6 months later). The cumulative technical success rate for HV recanalization was 95.7% (66/69; Fig. 7). No serious complications, including pericardial tamponade or ruptured blood vessels, were encountered. A total of 66 patients successfully underwent angioplasty. Recanalization was performed in 43, 14 and 9 patients via the transjugular, femoral vein and percutaneous transhepatic approach, respectively. In 35 and 13 patients, the target veins for recanalization were 1 or >2 major HVs. In 7 patients, the target veins were an accessory HV and in 11 patients a major HV and an accessory HV were recanalized. All of the patients underwent balloon angioplasty. In 41 patients, balloon angioplasty constituted the sole treatment (balloon angioplasty group). In addition, 14, 6 and 5 patients in the combination therapy group respectively received thrombolysis, endovascular stenting and a combination of thrombolysis and endovascular stenting. The baseline characteristics (including sex, age, duration of symptoms, albumin levels, incidence of ascites and cancer antigen-125) of the balloon angioplasty and combination therapy groups are presented in Table III. The differences of incidence of ascites, albumin levels, cancer antigen-125 levels and duration of symptoms between the two groups were statistically significant (Table III). Clinical efficacy. On admission, 3, 20 and 43 patients presented with mild, moderate and massive ascites, respectively. After endovascular therapy, resolution of ascites was achieved in 53 patients, while 13 patients had a small amount of residual ascites. The difference in the grade of ascites prior to and after therapy was statistically significant (χ 2 =122.250, P=0.001). The mean HV pressure after HV recanalization was significantly lower compared with that prior to the procedure (23±7 vs. 47±9 cmH 2 O; t=17.979, P=0.001; Table IV). The symptoms of 61 patients were completely relieved after HV recanalization, while those of 5 patients were partially relieved. Survival rate. The median follow-up period was 75 months (range, 60-84 months). At 12, 36 and 60 months, the cumulative survival rate of patients subjected to endovascular interventional therapy was 98.5, 98.5 and 93.9%, respectively, while that in the balloon angioplasty group was 97.6, 97.6 and 97.6%, respectively, and that in the combination therapy group was 96.0, 96.0 and 88.0%, respectively. No significant differences in all cumulative survival rates between the balloon angioplasty group and combination therapy group were identified (χ 2 =2.387; P=0.122; Fig. 8). Discussion In theory, angioplasty of an obstructed HV may reduce congestion of the liver, decrease pressure of the HV and the portal vein, and restore the patient's liver function, thereby making it an ideal treatment for BCS caused by HVO (7,8). However, endovascular therapy remains a challenging procedure in these patients due to the anatomical characteristics of the HV (13,19) and the difficulty in restoring HV flow. However, in the present study, the technical success rate of the operation in the Chinese BCS patients with HVO was high (95.7%) and the post-operative asymptomatic survival rate (80.3% at 5 years) was also high after angioplasty of the obstructed HVs. In each patient, the first task was to determine which approach to use for recanalization of the obstructed HV. At present, the 3 most commonly used therapeutic approaches are the transjugular, transfemoral and percutaneous transhepatic approach (4,7). In the majority of patients with BCS due to HVO, the obstruction is in the proximal HV, resulting from stenosis of the ostium or membranous occlusion (4,14,20). These patients are ideally treated by recanalization via the jugular or femoral vein. The transjugular approach has a high success rate, as the angle between the HV and the proximal IVC is usually relatively large, and the guide wire may easily access the HV via the jugular vein (4,17,21). If recanalization of the HV via the transjugular fails, the femoral vein may be used, but this is technically more difficult. In certain patients with an accessory HV that intersects the IVC at an obtuse angle, the transfemoral approach may be more appropriate (4,11,19,21). The percutaneous transhepatic approach may be used in patients with mild ascites, if the jugular and femoral approaches have failed (4,13). For patients with ascites whose extent is more than mild, diuretics and other conservative treatments should be offered initially, and the percutaneous transhepatic approach could be employed once the ascites has significantly reduced. The second task is to select the appropriate target HV for recanalization. Ideally, the obstructed vein should be the first choice. The right HV drains a large area of the liver, has a smaller angle with the IVC and is the preferred vein for interventional therapy (7,13,19-21). If all 3 major HVs are obstructed, it may be attempted to reopen the right HV first. If the diameter and drainage range of the HVs selected for recanalization are sufficiently large, reperfusion may be effective for relieving the portal hypertension (4,17,21). However, if the diameter of the HV is small, multiple HVs may be considered for recanalization during the same procedure (4,7,11,13,18,21). If all of the HV segments are occluded, leading to failure of the interventional therapy, recanalization of the accessory HV may be attempted to relieve portal hypertension (13,19). The final task was to determine what type of angioplasty was suitable for recanalization of the obstructed HV. In the present study, it was observed that the symptoms completely disappeared if the transmembrane pressure difference after percutaneous transluminal angioplasty was 4 cmH 2 O. Therefore, it is |
indicated that the transmembrane pressure difference is a reliable predictor of successful treatment, although specific indicators require to be verified (7). The choice of treatment depends on the location and extent of the obstruction, as well as the characteristics of the lesion (3,4,8,13,17,21,22). For thrombus-free obstruction, balloon angioplasty has been recommended, with the diameter of the selected balloon based on the diameter of the vessel being treated, typically 12-20 mm (23). If the transmembrane pressure difference is <4 cmH 2 O, the treatment is considered successful. If the transmembrane pressure difference remains >4 cmH 2 O, an endovascular stent is placed. For the thromboembolic HV, thrombolytic therapy has been recommended, followed by balloon angioplasty and endovascular stent (8,17). For patients with treatment failure and mild clinical symptoms, conservative medical treatment prior to the second stage of medullary treatment, performed after the formation of collateral circulation in the compensated liver, has been suggested (3,6,8). In the present study, 3 patients were successfully subjected to second-stage treatment. In patients with a severe condition, TIPS was required to relieve portal hypertension and increase the chance of survival (8)(9)(10)24). For certain cases, liver transplantation has also been recommended (5,8). In the present study, no significant differences in symptom-free survival rates were identified between the balloon angioplasty and combination treatment groups. This may be due to the small sample size. Furthermore, in the present cohort was not randomized regarding the interventional treatment, but it was selected based on the pathological features of each patient, which may have introduced selection bias. Therefore, even if the post-operative asymptomatic survival rate of patients receiving a specific treatment was higher than that of patients receiving other treatments, this does not confirm the superiority of the treatment. Re-obstruction of the vein is a post-operative complication and the underlying cause of recurrence of BCS symptoms (4). According to various studies, HV stenting may reduce the incidence of HV re-obstruction (3,8,16,25). However, stents are permanent devices that, increase the difficulty of a subsequent interventional procedure in the case of a recurrent stenosis, and should therefore be used with caution. A large multicenter study indicated that the patency rates of covered and bare stents in patients with BCS were comparable (6). In a patient with BCS due to HVO, a covered stent affected the formation of a compensatory collateral circulation and further aggravated hepatic ischemia (12). Therefore, the use of a covered stent for recanalization of BCS caused by HVO may not be recommended. BCS associated with HVO may be classified as stenosis, membranous or segmental occlusion, thrombosis or wide occlusion (1,3,4). For stenosis or membranous/segmental occlusion, angioplasty has been effective and has a high post-operative asymptomatic survival rate (4,13,17). In the present study, the majority of patients recovered after endovascular interventional therapy, and the cumulative survival rate at 60 months post-operatively was 93.9%. The cumulative survival rate at 60 months in the balloon angioplasty group (97.6%) was higher than that in the combination therapy group (88.0%). This is probably due to the fact that most of the patients who received combination therapy had more severe pre-operative symptoms such as ascites and complicated baseline conditions (lower serum albumin), which were associated with a relatively poorer prognosis. However, since the sample size was relatively small, the difference between the 2 groups was not statistically significant, and further study with a larger population is required. For patients with extensive obstruction, it is more difficult to recanalize via angioplasty, with high rates of failure and low rates of symptom-free survival. These patients are better treated by TIPS, which directly reduces portal pressure, relieves sinus pressure and has a higher efficacy (3,9,10). In the present study, the post-operative symptom-free survival rates at 12, 36 and 60 months were 98.5, 98.5 and 92.4%, respectively. These are slightly higher than the survival rate (97.7, 92.2, and 90.0%) reported by Cui et al (4). In the present study, as most patients with BCS had HV stenosis or membranous or segmental occlusion, the outcomes of recanalization were good. However, only 29-41% of Western patients with BCS have membranous or segmental occlusion of the HV (4,26,27). Therefore, percutaneous recanalization is not applicable to most Western patients with BCS (4,26,27), and TIPS is the recommended treatment (3,9,10). The putative 6-to-120-month survival rates of Western patients with HV-associated BCS after TIPS were 72-97% (10,26), which is slightly lower than the 12-, 36-and 60-month survival rate determined in the present study. This may be due to the condition being more complicated in most Western patients with BCS receiving TIPS, and therefore, their survival rate is relatively low. These differences between Eastern and Western patients with HV-associated BCS may explain for the differences in the reported rates of recanalization. However, regardless of these differences, in the present study, the mid-term asymptomatic survival rate was high. This indicates that recanalization was effective for the treatment of HV-associated BCS in Chinese patients. The present study is limited by its retrospective nature and small sample size. In addition, the patients were not randomized regarding the mode of endovascular treatment, but the treatment was selected based upon the pathological features of the patients, which may have introduced bias. Hence, infuture, large sample prospective studies are required to validate the results of the present study. In conclusion, in the Chinese cohort of the present study, the majority of patients with BCS had segmental or membranous HVO. In the current study, the endovascular interventional treatment of BCS patients with HVO that was applied based on the characteristics of HV vascular lesions. These patients experienced high symptom-free survival rates and low recurrence rates after endovascular therapy. Larger prospective studies are required to confirm the efficacy of endovascular therapy for BCS caused by HVO. of this study and revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript. Ethical approval and consent to participate All procedures were performed in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5). Informed consent was obtained from all patients for being included in the study. Patient consent for publication Not applicable. Identification of Attractive Drug Targets in Neglected-Disease Pathogens Using an In Silico Approach Background The increased sequencing of pathogen genomes and the subsequent availability of genome-scale functional datasets are expected to guide the experimental work necessary for target-based drug discovery. However, a major bottleneck in this has been the difficulty of capturing and integrating relevant information in an easily accessible format for identifying and prioritizing potential targets. The open-access resource TDRtargets.org facilitates drug target prioritization for major tropical disease pathogens such as the mycobacteria Mycobacterium leprae and Mycobacterium tuberculosis; the kinetoplastid protozoans Leishmania major, Trypanosoma brucei, and Trypanosoma cruzi; the apicomplexan protozoans Plasmodium falciparum, Plasmodium vivax, and Toxoplasma gondii; and the helminths Brugia malayi and Schistosoma mansoni. Methodology/Principal Findings Here we present strategies to prioritize pathogen proteins based on whether their properties meet criteria considered desirable in a drug target. These criteria are based upon both sequence-derived information (e.g., molecular mass) and functional data on expression, essentiality, phenotypes, metabolic pathways, assayability, and druggability. This approach also highlights the fact that data for many relevant criteria are lacking in less-studied pathogens (e.g., helminths), and we demonstrate how this can be partially overcome by mapping data from homologous genes in well-studied organisms. We also show how individual users can easily upload external datasets and integrate them with existing data in TDRtargets.org to generate highly customized ranked lists of potential targets. Conclusions/Significance Using the datasets and the tools available in TDRtargets.org, we have generated illustrative lists of potential drug targets in seven tropical disease pathogens. While these lists are broadly consistent with the research community's current interest in certain specific proteins, and suggest novel target candidates that may merit further study, the lists can easily be modified in a user-specific manner, either by adjusting the weights for chosen criteria or by changing the criteria that are included. Introduction Several strategies exist for the pursuit of drugs to treat neglected tropical diseases. Major approaches can generally be classified as: (A) label extension, extending the indications of existing drugs for other conditions to tropical diseases; (B) piggyback discovery, in which the discovery of new drugs is focused on one or a few classes of well-studied and validated targets; and (C) de novo drug discovery [1]. These strategies collectively seek to exploit two possible sets of drug targets: those that have been validated in other organisms and diseases, and those that have not -perhaps because they are unique to neglected-disease pathogens -but that nevertheless have potential as novel sites of action. Since experimental investigations of possible drug targets are time-consuming and expensive, it is worthwhile to conduct in silico analyses [2][3][4][5][6][7][8] to identify the proteins most worthy of experimental follow-up. These analyses consider traits commonly thought to be desirable in a drug target, including essentiality, druggability (whether drug-like molecules are likely to interact with the target), assayability, specificity/selectivity (potential for inhibiting the pathogen without harming the host), and importance in life-cycle stages of the pathogen relevant to human health. Inferring these traits from experimental data is a nontrivial task. For example, guesses at a target's essentiality can be made from gene knockout experiments with the pathogen of interest [9] or related organisms [3,6], from naturally occurring gene deletions in clinical isolates [10], from microarray and/or proteomic data [11], and/or from metabolic chokepoint (flux balance) studies [12,13]. Since the best choices are partly a matter of opinion, there is a clear need for databases that are flexible enough to integrate datasets from different sources and to filter these datasets based on the preferences of individual researchers. To facilitate target-focused analyses for pathogens prioritized by the World Health Organization's Special Programme for Research and Training in Tropical Diseases (TDR), TDRtargets.org [14] was created as a central repository of target-related data. The database may be used for two general scientific tasks: (A) analysis of individual proteins, finding information that relates to their potential as drug targets; and (B) genome-level analysis, sorting and ranking multiple proteins as drug target candidates according to user-specified criteria. The latter task is the main focus of this paper. TDRtargets.org is designed to facilitate multiple approaches to target prioritization. Users can browse target lists that others have posted (http://tdrtargets.org/published), generate their own lists from standard criteria offered by the database, and/or extend the criteria used to rank prospective targets by uploading files representing additional published or unpublished data. A previous publication [14] has outlined the user interface and concepts underlying the possible queries. In this study, we provide examples of whole-genome prioritization of targets, focusing on key issues for the specific diseases covered. We use these prioritization tools to generate lists of promising drug targets for TDR organisms -lists which provide useful starting points for target characterization in these organisms, as well as illustrate the general utility and versatility of TDRtargets.org in identifying and ranking targets. Database Infrastructure We have previously described the construction of the TDRtargets.org database, as well as the formulation of searches (queries) to identify proteins meeting criteria of interest and the viewing, saving, and exporting of search results [14]. Since then, while the overall workflow of the database has remained the same, additional genomes and datasets have been included (see below), and several improvements have been implemented on the user interface side of the database. Although users have always been able to perform ''weighted union'' queries, with different weights (point values) assigned to different user-specified criteria, formulating these queries and viewing and adjusting their results has recently been made more convenient. To construct a weighted union query from the website's target search page, a user (1) selects a pathogen (e.g., P. falciparum), (2) selects a criterion (e.g., functional category = enzyme) with which to query the pathogen genes, (3) enters a name and a weight for the query in the ''Run this query'' sub-menu at the bottom of the page, (4) clicks the ''Next Query'' button, and (5) repeats steps 2 to 4 until the last criterion is entered, at which point the user selects ''Run this query'' rather than ''Next Query.'' The search results are displayed on a page where users have the option of changing the previously entered weights for |
each criterion (Figure 1). (These results are archived on the user's history page, where he/she can combine different subsets of previous queries with the Union function to obtain new ranked target lists.) The presentation of ranked lists has also been revised to display the criteria met by each protein (Figure 1). Further flexibility in data analysis is provided by an option to export the results to a dynamic spreadsheet so that proteins' fulfillment of individual criteria can be viewed and the weights of the criteria can be adjusted offline. Using External Data in TDRtargets.org The TDRtargets.org web application lets users take advantage of datasets obtained externally or in-house. Lists of genes matching user-defined criteria may be saved as text files (each containing a column of gene identifiers -one per line -plus an optional second column for point values, if the targets have been ranked outside of TDRtargets.org) and uploaded at the user's history page. Uploaded lists can be combined with other gene sets from the same organism using any of the history page tools, including ranking by weighted union. In the present work, a number of target lists meeting different criteria were obtained from external resources, uploaded into TDRtargets.org, and used in various prioritization strategies (see Results), as follows. (A) T. cruzi genes with proteomic evidence of expression in amastigotes (at least 2 mass spectra/peptides mapped to the protein) were obtained from TriTrypDB.org [15]. (B) S. mansoni genes with evidence for expression at the transcript level (i.e., genes with mapped expressed sequence tags derived from the ''egg,'' ''schistosomula,'' and ''adult worm'' cDNA libraries) were taken from SchistoDB.net [16]. (C) Drosophila melanogaster genes associated with abnormal phenotype tags (i.e., ''lethal'' and ''neurophysiological defect'') were taken from FlyBase.org [17]. This list was converted into a list of the corresponding S. mansoni orthologs (available from OrthoMCL.org [15]) before uploading into TDRtargets.org. Genome Data and Functional Datasets The current version of the database includes genome data for ten different pathogens (Brugia malayi, Leishmania major, Mycobacterium leprae, Mycobacterium tuberculosis, Plasmodium falciparum, Plasmo- Author Summary In cell-based drug development, researchers attempt to create drugs that kill a pathogen without necessarily understanding the details of how the drugs work. In contrast, target-based drug development entails the search for compounds that act on a specific intracellular target-often a protein known or suspected to be required for survival of the pathogen. The latter approach to drug development has been facilitated greatly by the sequencing of many pathogen genomes and the incorporation of genome data into user-friendly databases. The present paper shows how the database TDRtargets.org can identify proteins that might be considered good drug targets for diseases such as African sleeping sickness, Chagas disease, parasitic worm infections, tuberculosis, and malaria. These proteins may score highly in searches of the database because they are dissimilar to human proteins, are structurally similar to other ''druggable'' proteins, have functions that are easy to measure, and/or fulfill other criteria. Researchers can use the lists of highscoring proteins as a basis for deciding which potential drug targets to pursue experimentally. Identification of Drug Targets In Silico www.plosntds.org dium vivax, Schistosoma mansoni, Toxoplasma gondii, Trypanosoma brucei, and Trypanosoma cruzi) and one endosymbiont bacterium (Wolbachia, endosymbiont of B. malayi). The depth of data coverage in various functional datasets (searchable at http://tdrtargets.org/ search) varies for different organisms; wherever possible, gaps in coverage are compensated for by mapping relevant information from orthologous proteins in other organisms. (For example, protein structure data available for P. falciparum proteins were mapped to P. vivax proteins.) Ortholog identification on whole genomes was carried out using tools available from OrthoM-CL.org [18]. Data recently added to TDRtargets.org include curated data on production of recombinant proteins and activity assays from BRENDA [19]; three-dimensional models of proteins from B. malayi and its endosymbiont Wolbachia, M. leprae, and S. mansoni, obtained from ModBase [20]; and phylogenetic information on Arabidopsis thaliana (so that users can search for proteins with or without orthologs in plants). Ranking Target Genes via Weighted Unions TDRtargets.org has a flexible ranking system for prioritizing target proteins. In multi-criteria searches, it is possible to take a Boolean intersection of the criteria so that only those proteins with all of the desired traits (e.g., essentiality AND druggability AND assayability, etc.) are selected. However, a protein may lack one or more preferred properties and still be the target of an effective drug (Table 1). Therefore the prioritization queries presented below are devised as weighted unions (see ''Database infrastructure'' above), in which each criterion is assigned a subjective weight (point value) and targets earn points for each criterion they meet. (Less important and undesirable criteria are given small and negative weights, respectively.) These queries return ranked lists of all potential targets, ordered by cumulative score. Target lists can then be re-ranked, if desired, by changing the weights and/or adding additional criteria (see ''Database infrastructure'' above). Overview of Queries Presented in This Paper The criteria used in generating the lists presented below are summarized in Figure 2. As a starting point, a basic set of criteria of general interest were chosen to frame a ''standard'' query for identifying targets in L. major (see Query 2 in Figure 2). In compiling this basic set of criteria, we included most datasets that are commonly available for organisms with complete genomic information so that the standard query could be easily applied to different pathogens. Queries 3, 4, and 5 of Figure 2 are examples of extending the standard query. Queries 6, 7, 8, and 9 of Figure 2 are framed in a pathogen-specific manner to prioritize target proteins from a particular metabolic pathway, subcellular location, or life-cycle stage. These queries make use of criteria based on external datasets uploaded to TDRtargets.org. (Readers can explore the upload tool at http://tdrtargets.org/history.) Queries 10 and 11 of Figure 2 were based heavily on data obtained by manual curation of the literature [21] and homology/orthology analysis for protein-specific information, illustrating how even incompletely annotated genomes are amenable to target identification. Additional details of these queries are noted below. Searching for Candidate Drug Targets in Leishmania An example of the weighted-union approach to target prioritization (see Methods) is shown in Query 2 of Figure 2, which covers the Leishmania major genome. In this example, points are awarded for many of the criteria covered in Table 1, plus some additional conditions. From these criteria a list of prioritized targets is generated ( Table 2). Such a list is hardly the final word in Leishmania target selection, however. The researchers who generated the list in Table 2 may subsequently decide that, since essentiality data for Leishmania genes are very limited, they will In general, the following might be considered desirable target traits: a low molecular weight and a lack of transmembrane (TM) domains (to favor expression and solubility of recombinant protein), existence of 3D crystal structures and ModBase models (for structure-based drug design), absence of orthologs from humans (to favor selectivity), high druggability and compound desirability scores (0-to-1 scale), and a precedent for assayability. consider the presence of an essential ortholog in at least one other organism to be an acceptable predictor of essentiality. Orthologous proteins usually have the same function [22], and several studies indicate that having essential orthologs is predictive of essentiality [23,24]. The researchers could then amend their initial query so that, for example, 50 additional points are awarded to targets whose orthologs are essential in C. elegans, E. coli, M. tuberculosis, and/or S. cerevisiae (the four organisms for which genome-wide essentiality data are available in TDRtargets.org). Such a revision can easily be made by running a new query using the ''Any evidence of essentiality in any species'' option within the Essentiality subsection of the Search For Genes/Targets page and then using the query history page to find the union of this query and the previous one. The results are similar to but distinct from the previous results (Table 3). Now consider a more drastic revision of the Leishmania search: application of the previous criteria ( Figure 2, Query 3) to two other pathogens, namely P. falciparum and M. tuberculosis. This too is readily done within TDRtargets.org -there is a ''Change species'' option on the Query History page -again highlighting the ease of modifying previous searches. While use of exactly the same criteria to prioritize targets in different species might seem naïve, the results (Tables 4 and 5) are instructive. First of all, the top-ranked proteins of each species are rather different, showing that this search strategy is sensitive to species differences, as opposed to being unalterably biased toward the same proteins in every species. Second, many of the top-ranked targets in each species appear to be appealing options. For example, the three top-scoring targets from each species -dihydrofolate reductase/thymidylate synthase, enoyl-ACP reductase, and triose-phosphate isomerase in Table 2. Preliminary genome-wide prioritization of Leishmania major targets. Ranking Gene_ _name Gene product Weight Top targets according to the criteria shown in Query 2 of Figure 2. Complete genome-wide rankings for this example and all other examples discussed in the paper (Tables 3-11) are available online at http://www.tdrtargets.org/published/browse/366. Please note that multiple targets often receive the same total weight, and that the order in which these ''tied'' targets are displayed has no significance. doi:10.1371/journal.pntd.0000804.t002 Figure 2. A summary of the multiparameter search queries presented in this study. Ten different queries (Queries 2-11) are listed as individual columns for which the criteria are shown on the left. For each criterion, the number of qualifying proteins from a given pathogen is shown in black and the associated weight is shown in red within parentheses. Symbols: (#) enzymes were selected by combining searches by EC number and by functional category, except for Queries 10 and 11, which were based only on EC number; (&) the conserved-in-taxon criterion refers to the presence of orthologs in L. major, T. brucei, and T. cruzi (Tables 2 and 3), P. falciparum and P. vivax (Tables 4 and 7), M. tuberculosis and M. leprae (Table 5), and L. major and T. cruzi (Table 8); (") druggability and compound desirability scores were queried using respective cutoff values of $0.6 and .0.3 (Tables 2 to 5), $0.4 and .0.2 (Tables 6 and 7), and $0.5 (druggability scores only; P. falciparum and enoyl-ACP reductase (InhA), glutamine synthetase, and 5-enolpyruvylshikimate-3-phosphate synthase (AroA) in M. tuberculosis -have all attracted interest as proven or prospective targets [25][26][27][28][29][30]. It is interesting that legitimate candidates such as these rise to the top of the target rankings despite certain quirks of this ''one set of criteria fits all species'' example. In the M. tuberculosis prioritization, for instance, many of the top-ranked targets are essential even though the genome-wide mutagenesis data available for this species were not queried. Thus, although these lists are imperfect, they generally suggest that rational choices of criteria lead to plausible and informative rankings of target desirability across species. T. brucei and P. falciparum: Metabolic Pathway-and Organelle-Specific Targets While TDRtargets.org integrates numerous datasets relevant to target prioritization, it cannot possibly anticipate every possible prioritization strategy that could be used by any given researcher. Accordingly, users can upload (weighted or unweighted) lists of targets meeting any criteria for which they have relevant data; these may then be combined with other queries covered by TDRtargets.org. Supplementation of standard TDRtargets.org criteria with a user-provided criterion is illustrated in the following example. Researchers specializing in the T. brucei glycolytic pathway are convinced that this pathway is essential in these parasites and wish to rank the enzymes within this pathway for their suitability as drug targets. Since they already assume the pathway to be essential and know glycolysis is also present in host cells, they may not address these issues in their search criteria, but may instead award points as listed in Query 6 of Figure 2. The query shown there combines criteria addressing integral TDRtargets.org data (such as availability of structural models) with a usergenerated list of ''bonus points'' to some T. brucei enzymes in proportion to their relative control over the glycolytic flux [31]. The rationale for such a scoring might be that the greater an enzyme's flux control, the less completely it must be inhibited for flux through the entire pathway to be affected (and thus the better a target it is). In this example, the inclusion of flux control as a |
criterion lifts the two glycosomal orthologs of glyceraldehyde-3phosphate dehydrogenase, the enzyme with the highest control Table 3. Revised L. major rankings after incorporating an essential-in-other-species criterion. Ranking Gene name Gene product Weight Top targets according to the criteria shown in Query 3 of Figure 2. Italicized targets are those that were not top-ranked in the list shown in Table 2 Table 6). The recent genetic validation of this enzyme [32] likewise identifies it as a possible target of interest. Interestingly, hexokinase was thought to have a much lower control coefficient [31] but may also have promise as a drug target [33]. The next scenario also employs a user-provided list, which in this case permits scrutiny of a specific organelle rather than a specific metabolic pathway. Consider a newly independent crystallographer with a special interest in Plasmodium apicoplasts, which are absent from the human host and thus are likely to contain many appealing drug targets [34]. The PlasmoAP algorithm [35] predicts that 485 proteins are localized to the apicoplast; the user can download this list from PlasmoDB.org [36], manually delete proteins that seem unlikely to reside in the apicoplast, and then upload the modified list to TDRtargets.org. In sorting through the ,400 proteins likely to reside in the apicoplast, the user may decide to minimize competition with labs already working on apicoplast biology by penalizing well-studied proteins (e.g., subtracting 100 points from targets whose 3D structures have already been solved) while rewarding other desirable characteristics such as those discussed above (likely essentiality, lack of orthologs in humans, etc.). Finally, a previous publication [37] has convinced the hypothetical user that a low molecular weight and low isoelectric point (pI) improve the odds of successful expression of soluble Plasmodium proteins, so those factors are weighted accordingly (Query 7 of Figure 2). The most highly ranked proteins in this example (Table 7) include some proteins (e.g., pseudouridine synthetase and cysteine desulfurase) that are rarely mentioned in the Plasmodium literature, consistent with this researcher's desire to explore truly novel target options. Trypanosoma cruzi: Candidate Targets Associated with an Intracellular Lifestyle Unlike the bloodstream trypomastigotes of African Trypanosomes (Salivaria), the T. cruzi (Stercoraria) bloodstream forms do not replicate, and instead invade cells. In this parasitic strategy, Table 4. Application of standard search criteria to P. falciparum. Ranking Gene name Gene product Weight which is shared with Leishmania spp., the replicative amastigotes are the intracellular parasite forms that persist and maintain the infection. Given the early evolutionary divergence of Salivarian trypanosomes [38] and the different strategies used by Salivarian and Stercorarian parasites to mount and maintain an infection, these groups of parasites may exhibit numerous instances of (A) gene loss and (B) gene duplications followed by neofunctionalization [39]. Proteins that are orthologous between T. cruzi and Leishmania but that lack T. brucei counterparts may represent proteins vital to intracellular survival and/or growth, which could be excellent targets for drug development. To look for such proteins, we used a general strategy similar to that used for Leishmania (see Query 3 of Figure 2) but now focused on T. cruzi, with an extra phylogeny-based restriction: additional weight was added to proteins that have been conserved in Leishmania and T. cruzi but that have been lost or substantially changed in T. brucei. The attributes and weights used in this query are shown in Query 8 of Figure 2. The strategy also relies on proteomic evidence of expression in intracellular amastigotes [40]. However, because the proteomic data have a low coverage of the proteome, only a moderate weight has been assigned to this criterion. (This illustrates users' ability to assign relative weights based not only on which characteristics they consider predictive of target desirability, but also on their confidence that available experimental datasets accurately reflect those characteristics.) The results of this prioritization of T. cruzi targets are shown in Table 8. Because of the hybrid nature of the strain used to sequence the genome of T. cruzi, the list is somewhat redundant: most single copy genes appear twice in all genome databases. The top 32 targets include representatives of validated pathwaysergosterol biosynthesis, as represented by sterol C-24 reductase, and glycolysis, as represented by glucokinase -and other interesting alternatives for drug development. As suggested above, glycolysis is an essential pathway in trypanosomes, and the glycosome-localized glucokinase has attracted interest as a possible target since it was discovered in the sequenced Leishmania and T. cruzi genomes [41]. On the other hand, the top-ranked sterol C-24 reductase provides a good example of the attractiveness of the phylogenetic criteria used in this strategy. The ergosterol biosynthesis pathway is also present in T. brucei, although it is not essential for the bloodstream forms, which scavenge sterols from the host [42]. This highly ranked C-24 reductase belongs to the OrthoMCL ortholog group OG4_16908 (OrthoMCL version 4), which contains orthologs from the genomes of T. cruzi, L. major, and yeast (ERG4). However, this enzyme is apparently absent in the genomes of T. brucei TREU927, T. brucei gambiense, T. vivax, and Another top-ranking target in Table 8 is the T. cruzi serine acetyltransferase (TcSAT), involved in the de novo synthesis of cysteine, which is present in Leishmania and T. cruzi and absent in T. brucei [43]. Cysteine in these parasites is important for the biosynthesis of polyamines and for antioxidant metabolism based on trypanothione, the trypanosome equivalent of glutathione. Inhibitors of the E. coli SAT enzyme have recently been shown to inhibit the growth of Entamoeba histolytica, another pathogen that is highly sensitive to oxidative stress [44]. Other interesting targets in this list include a putative amine oxidase (Tc00.1047053511277.600) which further analysis shows is conserved in several Leishmania species but absent in sequenced T. brucei subspecies; a putative phosphatidate cytidylyltransferase (Tc00.1047053509073.70) that belongs to an ortholog group with a very restricted phylogenetic distribution (OG4_29276), with members from many Leishmania species with complete genomes, Entamoeba histolytica (another pathogen), and two non-pathogenic species (Thalassiosira pseudonana and Aquifex aeolicus); a protein kinase (Tc00.1047053509287.20) whose yeast orthologs regulate endocytosis through the organization and function of the actin cytoskeleton; and a tyrosine protein phosphatase (Tc00.1047053506839.60) that also shows an unusual phylogenetic distribution, being almost exclusively present in T. cruzi, Leishmania spp., and metazoa. Mycobacterium tuberculosis: Exploiting Previous Prioritizations Previous target prioritization efforts [2][3][4][5][6][7][8] raise the question of how these efforts should be viewed in relation to TDRtargets.org. We consider TDRtargets.org to be complementary to others' prioritization work, and anticipate that it can be used to combine Table 6. Prioritization of glycolytic enzymes in T. brucei. Ranking Gene name Gene product Weight and apply the ranking methods of other target identification efforts. For instance, a recent paper on M. tuberculosis by Hasan and colleagues [4] provided an excellent synthesis of experimental data to rank targets by persistence in dormant stages. These data (available in [4] as Supplemental Dataset S1, and also at http:// tdrtargets.org/published/browse/379) can be easily interrogated and combined with other queries using TDRtargets.org. For example, while Hasan et al.'s rankings considered proteins essential for growth on defined medium in vitro [45,46], they did not reward proteins thought to be essential for growth in macrophages or in the infection of mice [47,48], which may well be very relevant to human infection. In addition, because Hasan et al. awarded points to proteins with solved crystal structures, it seems apt to give points to proteins whose structures have been solved during the four years that have elapsed since the original analysis was published. TDRtargets.org was therefore used to make a few modifications to one set of Hasan et al. 's rankings: the set that emphasized targets' likely importance in persistent-stage disease. We uploaded a modified version of this list that excluded points for PDB structures, then gave additional points to all genes represented in the Protein Data Bank of crystal structures [49] as of April 2010. To these subtotals, we added points based on an analysis of latentstage infections by Murphy & Brown [7]. In that analysis, genes were given upregulation and downregulation scores based on their expression in various models of dormancy, thus offering a distinct estimate of genes' importance during latency, and ''attenuation'' scores based on the effect of gene knockouts on growth in various contexts, including the macrophage and mouse studies noted above. (See ''Additional file 1'' from [7]; see also http://tdrtargets. org/published/browse/383.) The combined input of the two previous studies was thus used to create a ''consensus list'' ( Table 9) that might be considered superior to either one alone. Combining the two previous analyses could also be done off-line using spreadsheets, but performing these operations within TDRtargets.org is considerably faster and facilitates retrieval of TDRtargets.org-compiled information on each individual target. Naturally, our ''consensus list'' reflects the limitations of the previous analyses, e.g., the low rankings of important persistent-stage proteins such as Rv0470c (mycolic acid synthase, PcaA) and Rv2583c (GTP pyrophosphokinase, RelA), as discussed by Hasan et al. [4]. Helminths: The Importance of Homology Since many valuable helminth datasets are only starting to emerge, our attempts to prioritize helminth targets required some analysis beyond the standard TDRtargets.org queries. For example, B. malayi and S. mansoni proteins are not yet scored for druggability in TDRtargets.org, so we assessed their druggability by comparing their amino acid sequences to those of known drug targets in the StARLITe/ChEMBL database [50]. The sequence similarity analysis was performed using BLAST; a helminth protein was considered druggable if (A) it is $80% of the length of the corresponding druggable target, (B) it has an amino-acid sequence that aligns with $80% of the druggable target, and (C) the BLAST expectation value of the alignment is less than 10 210 (database size: 11,508 genes for B. malayi, 13,331 genes for S. mansoni). In addition, proteins' functional importance in helminths was inferred from knockout data taken from their orthologs in C. elegans and D. melanogaster (see Materials and Methods and Queries 10 and 11 of Figure 2). Being able to connect the helminth proteins to similar proteins in other species was thus critical in allowing us to evaluate their potential as drug targets. Our strategy of relying heavily on orthology and sequence similarity to rank helminth targets is broadly similar to those used by Kumar et al. [6] to rank Brugia targets and by Caffrey et al. [3] to rank Schistosoma targets. However, these authors sought targets that met each of several desired criteria (Boolean ''AND''); for example, Kumar et al. only considered Brugia proteins with orthologs in C. elegans but not in humans, and whose absence causes deleterious phenotypes (according to RNAi of C. elegans orthologs). In contrast, we again used the ''weighted union'' approach to avoid premature elimination of any proteins from consideration as targets. Kumar et al. also took a distinct approach to druggability, rewarding proteins with domains targeted by compounds obeying the Lipinski ''Rule of 5'' [51] and having EC numbers associated with druggability. Additionally, Kumar et al. penalized proteins for hydropathicity (which reduces the ease of recombinant expression) and rewarded them for being expressed (according to a small dataset of expressed sequence tags, or ESTs, encompassing 250 genes); in contrast, we gave additional points to all proteins having EC numbers (and therefore presumed to be enzymes), 3D structural models, and/or bibliographic references. A comparison of our helminth prioritizations (Tables 10 and 11 Table S1 of [6], also available at http://tdrtargets. org/published/browse/282). This lack of overlap is likely due in part to (A) our emphasis on druggability, as inferred from sequence similarity against targets in the ChEMBL database, and (B) the fact that we didn't penalize proteins with human orthologs (see Discussion subsection ''No List is Canonical''). By adding two conditions to the weighted union to penalize proteins with orthologs in human and in mouse (with weights 240 and 220, respectively), some overlap between both lists can be observed: among our top 200 targets, 32 are also among the top 200 as ranked by Kumar et al. One unique aspect of our list is that it includes four tRNA synthetases among the top 39 proteins. These enzymes have been proposed as drug targets in Brugia, and are also considered good drug target candidates in other parasites such as trypanosomes [52], since they must be essential but often have major structural differences with respect to the human orthologs. The list of 57 recommended Schistosoma targets generated by |
Caffrey and colleagues (see Table S1 of [3], also available at http:// tdrtargets.org/published/browse/247) includes 18 targets they considered to be of the highest priority because they are druggable, are expressed in relevant life-cycle stages, yield deleterious phenotypes, and are homologous to proteins with solved crystal structures including co-crystallized ligands. Of these 18 targets, eight rank within our 170 top Schistosoma targets. An obvious difference between the two lists is that ours includes nine tubulins among the top 23 proteins. The prominence of the tubulins is consistent with beta-tubulin's validation as a helminth drug target [53]. A number of ATPases also appear among our top targets. The top-ranked target in our list is the alpha (catalytic) subunit of a Na + / K + ATPase (Smp_015020), which in mammals (and probably also in schistosomes) is the target of ouabain and other cardiac glycosides [54]. This target does not appear in the list of 57 targets published by Caffrey et al.; however, the beta subunit of this or a closely related Na + /K + ATPase (Smp_124240) is ranked #52 in this study. Other attractive targets include a putative extracellular-signal-regulated kinase (ERK, Smp_142050), and a putative HMG-CoA reductase (Smp_138590), which is the target of cholesterol-lowering drugs like mevinolin [55]. Stability of Ranked Lists A relevant question for any ranked list of targets using any strategy is: how different would this list be if the weight given to a certain attribute is changed? Using the M. tuberculosis queries whose results are in Table 5, we analyzed the robustness of the final ranked list, by selecting one attribute at a time and changing its weight from a very low (negative) score to a very high (positive) score. To assess the change observed in the ranked list we counted the number of curated targets (i.e., those with some level of validation) observed within the top 200 targets in the ranked list and used this value as our objective function (see panel B in Figure 3). Using this measure, we observed that a high score is obviously needed for those attributes that are enriched in validated targets (see panel A in Figure 3) in order to find well-known targets at the top of the list. This is also true for attributes that are not independent of these ''good'' attributes (e.g., availability of 3D structures). In contrast, changing the weight of attributes that are not expected to be enriched in validated drug targets (e.g., low molecular weight) does not affect the final result. In these cases, the final lists are all different, but they are consistent in having the highest ranks of the list being enriched in validated targets. In general, of course, targets' rankings within a list can be increasingly stabilized by including more and more relevant criteria in the prioritization. Old Targets Versus New Targets In analyzing target candidates, we often wonder what sort of mix of well-studied and not-so-well-studied proteins might be most ''desirable'' at the top of a ranked list. On the one hand, having well-known targets or targets of known drugs at the top of our lists offers some assurance that our search strategies are reasonable (i.e., they serve as ''positive controls'' of the strategy). On the other hand, a method that only identifies well-established targets would not serve the important purpose of suggesting novel targets, so the presence of novel (even ''hypothetical'') targets near the top of a list is also welcome. With the deliberate exception of Table 7, our lists reflect a desire to spotlight both previously validated and newly emerging targets. In addition to trying to achieve a mix of new and established targets in prioritization lists, users need also to robustly consider which established targets they should classify as ''successful.'' Some targets enjoy long-held high esteem within the research community in the absence of any clinical validation, while other proteins, particularly for the organisms being studied here, are targets of clinically used drugs whose product profiles are unlikely to be acceptable in future drug development programs. False Negatives Previous bioinformatic analyses of drug targets [4] have suggested that certain established targets never rank highly unless given artificial boosts in points for that specific purpose. Examples of these ''false negatives'' are also apparent in the lists presented here. For instance, cytochrome b is the known target of the antimalarial drug atovaquone [56], yet it ranks in the bottom 25% of targets represented by Table 4 because it has transmembrane domains (making recombinant expression difficult), is not easy to assay in isolation, lacks a known crystal structure, and so on. Likewise, certain targets of antihelminth drugs -such as the acetylcholine and GABA receptors, glutamate-gated chloride channel, and SLO-1 potassium channel [57,58] -do not appear near the top of our helminth lists. There are several possible (nonexclusive) explanations for this. First, some drugs were found through phenotypic screens and their targets do not meet many of the criteria required in a target-based approach, and thus might not be expected to rank highly. Second, current target prioritization strategies are generally based on the assumption that drugs will cause a loss-of-function phenotype, but most existing helminth drugs lead to gain-of-function phenotypes [57]. Ranking proteins according to their potential as gain-of-function targets might be a fruitful direction of future work. Finally, it is conceivable that the total number of viable drug targets vastly exceeds the number that have been clinically validated, such that the position of many nonvalidated targets ahead of some validated ones is appropriate. False Positives The failure of some validated targets to be highly ranked in our lists is not particularly surprising or troublesome, as discussed above. A more interesting issue is that of ''false positives,'' i.e., proteins that do rank highly but have not been validated as drug targets despite considerable effort. For instance, the Leishmania adenosine kinase ranks among the top 25 proteins in Tables 2 and 3, yet turns out to be nonessential in promastigotes [59]. Similarly, the Plasmodium enoyl-ACP reductase (FabI) ranks 2 nd in Table 4, yet is nonessential for blood-stage growth [60]. Among M. tuberculosis proteins, pantothenate kinase (PanK or CoaA) is in the top 100 of the Query 5 rankings (though not among the top 24 and thus not shown in Table 5), yet screens targeting this enzyme yielded no leads active against wild-type M. tuberculosis cells (C. E. Barry, personal communication). PanK activity in vivo appears to be so far in excess of what is required for growth that killing M. tuberculosis cells by inhibiting this enzyme is virtually impossible. Although such examples can be seen as discouraging, we can use them to ask whether the incidence of false positives can be reduced through the use of additional datasets and search strategies. The nonessesentiality of the Plasmodium FabI during erythrocyte stages is perhaps suggested by the fact that expression of the enzyme is neither high nor tightly regulated during the erythrocyte life-cycle stages [61]. While TDRtargets.org does not currently offer a metric for the periodicity of gene expression in blood-stage Plasmodium, this could be added to future versions of the database. No List Is Canonical The target rankings presented here are meant to be illustrative rather than definitive. The lists presented here were sent to experts on relevant neglected diseases for evaluation, and, predictably, we encountered numerous reasonable differences of opinion. For helminths, arguments were made both for and against penalizing proteins with orthologs in humans. The presence of human orthologs suggests an increased likelihood of toxicity in the host; on the other hand, several existing drug targets do have human orthologs. For M. tuberculosis, it was noted that existing drugs tend to target information-processing enzymes (DNA and RNA polymerase, DNA gyrase) rather than metabolic enzymes, so searches for new drugs might pay special attention to that area. Generally applicable suggestions included penalties for proteins that are part of macromolecular complexes, since they are hard to study in isolation, and for proteins of unknown function, since they are hard to study with biochemical or biophysical methods. In addition to legitimate differences of opinion among researchers, the relative appeal of individual targets will continue to change as additional data are gathered. Fortunately, the infrastructure of TDRtargets.org is flexible enough to accommodate different individuals' interests (as seen especially in the lists focused on T. brucei glycolysis and Plasmodium apicoplasts) and the incorporation of new data (most prominent in the rankings for the helminths and for M. tuberculosis persistence). We therefore see TDRtargets.org as a tool that individual scientists may use to explore new research directions, rather than as a final arbiter of proteins' potential as drug targets. As noted, target prioritization with TDRtargets.org or any other computational method is probably most useful as a prelude to (rather than a replacement of) laborious experimental follow-up work. Experimental characterization of promising targets often requires chemical inhibitors of target activity; therefore lists of target-specific inhibitors would be of great value to the research community. Though TDRtargets.org currently includes a preliminary dataset of such inhibitor-target associations, future editions of the database should offer major expansions and refinements of this dataset. Repair of a lateral malleolus defect with a composite pedicled second metatarsal flap The patient was a 26-year-old man who fell while riding a motorcycle and friction led to defects in the lateral malleolus and soft tissue of the ankle. Although the wound surface healed with scarring and skin grafting, the patient had symptoms of ankle joint instability 4 months after the injury. Using a second metatarsal composite tissue flap with a dorsalis pedis artery pedicle, we repaired the soft tissue defect of the ankle and reconstructed the lateral malleolus. The head of the metatarsal bone was used to reconstruct the lateral malleolus and the flap was used to cover the wound surface. The distal fibula and metatarsus were completely healed 36 months postoperatively. The ankle had maintained stability at this time, with equal limb length and only a mild limitation of dorsal flexion in the ankle joint. The patient could walk, jog, and walk up and down stairs without limitations. There was no pain or limitation in activity at the donor site. Our findings suggest that the second metatarsal composite tissue flap with a dorsalis pedis artery pedicle is an effective option in reconstruction of the adult distal fibula. ankle joint. We successfully treated a patient with a traumatic complete lateral malleolus defect. We repaired the soft tissue defect of the ankle and reconstructed the lateral malleolus using a second metatarsal composite tissue flap with a dorsalis pedis artery pedicle. This procedure has seldom been used in adults. Case presentation The patient was a 26-year-old man who fell while riding a motorcycle. Friction between the lateral malleolus and the pavement led to defects in the lateral malleolus and soft tissue of the ankle. Wound debridement and dressing changes were conducted at another hospital. The wound surface healed with scarring and skin grafting. However, the patient had symptoms of ankle joint instability 4 months after the injury due to the complete absence of the lateral malleolus ( Figure 1A). The study protocol was approved by Changzheng Hospital Ethics Committee. The patient provided written informed consent. Investigations Radiography showed slight valgus malformation of the ankle joint ( Figure 1B). Treatment To simultaneously repair the defects of the lateral malleolus and the soft tissue, we took the dorsalis pedis artery as the vascular pedicle, and used the second metatarsal bone and skin flap. The head of the metatarsal bone was used to reconstruct the lateral malleolus and the flap was used to cover the wound surface. We created a fusiform flap along the vertical axis of the dorsalis pedis artery, identified the dorsalis pedis artery between the extensor hallucis longus and extensor digitorum longus, and retained the deep peroneal nerve. We dissociated the dorsalis pedis artery from proximally to distally to form a second metatarsal composite tissue flap with a dorsalis pedis artery pedicle, including the extensor pollicis brevis and the dorsal skin of the foot. The flap was then transferred to the recipient site through a subcutaneous tunnel towards the proximal end. We made the fibular articular surface of the second metatarsus fit the talus, and fixed the fibula and metatarsus with steel plates to reconstruct the lateral malleolus. The extensor pollicis brevis was used to repair the |
lower tibiofibular ligament and the flap was used to cover the wound surface of the lateral malleolus ( Figure 2). At the donor site, the transverse metatarsal ligament was sutured in the dorsum of the foot, and the wound was directly sutured. Outcome and follow-up Postoperatively, the foot was fixed in a plaster cast for 4 weeks. Ankle joint exercise was started at 4 weeks postoperatively, partial weightbearing exercise at 8 weeks postoperatively, and full weightbearing exercise at 12 weeks postoperatively. The ankle joint received continuous physical therapy for 12 weeks. At 12 months postoperatively, radiography showed that the distal fibula and metatarsus were completely healed. A follow-up examination at 36 months postoperatively showed that the ankle had maintained stability, with equal limb length and only a mild limitation of dorsal flexion in the ankle joint ( Figure 3). The patient could walk, jog, and walk up and down stairs without limitations. The American Orthopedic Foot and Ankle Society score was 90. 3 There was no obvious pain or limitation in activity at the donor site. Discussion Reconstruction of the lateral malleolus defect is generally recommended when the defect is causing valgus instability of the ankle joint. 1,2,[4][5][6][7] The current methods used to repair a lateral malleolus defect include a pedicled vascularized fibular head graft, 4 iliac crest graft, 5 bone flap from the lateral eminence of the scapula, 6 and allograft bone transplantation. 7 However, there is no gold standard for lateral malleolus defect repair. The metatarsus can be easily cut and removed, and thus it has been widely used for reconstruction of the finger bones, and metacarpal, ulna, and temporomandibular joints. [8][9][10][11] The metatarsus has two arterial sources of the dorsal metatarsal artery from the dorsalis pedis artery and the plantar metatarsal artery from the posterior tibial artery. 12 These two vessels form a wide arterial network in the head of the metatarsal bone, 12 and provide the main blood supply to the metatarsal base. 13 The first dorsal metatarsal artery approaches the first metatarsus in the first intermetatarsal space, and its distal end enters the first dorsal metatarsal muscle and the second medial metatarsal head to supply the muscle and bone. The third and fourth metatarsal arteries supply the third, fourth, and fifth metatarsal bones. 14 Blood supply to the second metatarsus is abundant. [12][13][14] Therefore, the dorsal end of the dorsalis pedis artery and the first and second metatarsal arteries can be retained when the second metatarsal flap is cut. In the process of repairing the lateral malleolus with the second metatarsus, retention of these metatarsal arteries is conducive to fracture healing. There are some main points for success when using a composite pedicled second metatarsal flap to repair a lateral malleolus defect, as follows. (1) According to the degree of soft tissue defect of the lateral malleolus, the designed flap must enable tension-free wound repair after flap transfer. (2) When dissociating the dorsal vessels, care must be taken to avoid damaging the first and second dorsal metatarsal arteries, which enter the second metatarsus from the dorsalis pedis artery. When the head and flap of the metatarsal bone are cut, the soft tissue connection between the head and the flap must be protected, and they cannot be separated. This is because microvascular branches of the dorsalis pedis artery enter the flap. (3) The dorsalis pedis artery and vein must be dissociated as far as possible towards their proximal ends to ensure that the vascular pedicle of the flap has sufficient length. This is performed to avoid tension on the vessels after the tissue flap is rotated. (4) The lower tibiofibular ligament must be inspected, repaired, and reconstructed as much as possible. (5) The wound at the donor site can be sutured directly. If the defect area is large and cannot be sutured directly, a rotational flap can be used to repair the wound at the donor site. In 2008, Hu et al. 15 reconstructed the lateral malleolus with a pedicled second metatarsal flap for a 4-year-old boy. After 7 years of follow-up, their patient had adequate ankle function and stability. We repaired a soft tissue defect of the ankle and reconstructed the lateral malleolus using a second metatarsal composite tissue flap for a 26-year-old man. Our results are similar to those of Hu et al. 15 These findings suggest that the second metatarsal composite tissue flap with a dorsalis pedis artery pedicle is a useful option for reconstructing the distal fibula. Using a composite pedicled second metatarsal flap to repair a lateral malleolus defect has advantages and disadvantages. The advantages of this method are as follows. (1) The anatomical positions of the vessels and bone tissues of the donor site are constant and are close to the recipient site. (2) Local transfer of the vascular system can be used to simultaneously repair defects of the lateral malleolus and the soft tissue. The technically easy method, abundant blood supply, and good healing properties minimize the occurrence of complications, such as an anastomotic stoma. (3) The skin over the area of the repaired lateral malleolus retains feeling, and the shape is close to normal. The skin is conducive to wearing shoes and has good resistance to abrasion. The disadvantages of using a composite pedicled second metatarsal flap to repair a lateral malleolus defect are as follows: (1) one main vessel of the ankle is damaged; and (2) the integrity of the transverse arch is damaged to varying degrees. The plantar fascia should be tightened and sutured to reduce morphological changes of the arch. Declaration of conflicting interest The authors declare that there is no conflict of interest. The Added Benefit of Opicapone When Used Early in Parkinson's Disease Patients With Levodopa-Induced Motor Fluctuations: A Post-hoc Analysis of BIPARK-I and -II Introduction: Opicapone (OPC) was efficacious in reducing OFF-time in two pivotal trials in patients with Parkinson's disease (PD) and end-of-dose motor fluctuations (BIPARK-I and -II). Post-hoc analyses of these trials evaluated the efficacy of OPC following pre-defined segmentation of the wide spectrum of motor fluctuations in PD. Methods: Data from matching treatment arms in BIPARK-I and -II were combined for the placebo (PLC) and OPC 50-mg groups, and exploratory post-hoc analyses were performed to investigate the efficacy of OPC 50 mg vs. PLC in subgroups of patients who were in “earlier” vs. “later” stages of both their disease course (e.g., duration of PD <6 years vs. ≥6 years) and levodopa treatment pathway (e.g., number of daily levodopa intakes <4 vs. ≥4). Efficacy variables included changes from baseline in absolute OFF-time and total ON-time. Results: The Full Analysis Set included 517 patients (PLC, n = 255; OPC 50 mg, n = 262). OPC 50 mg was significantly more effective than PLC in reducing OFF-time and increasing ON-time in the majority of subgroup analyses (p < 0.05). Moreover, patients in “earlier” stages of both their disease course and levodopa treatment pathway experienced numerically greater efficacy when using OPC 50 mg, in comparison with those in “later” stages. Conclusion: OPC 50 mg was efficacious over the whole trajectory of motor fluctuation evolution in PD patients. There was also a signal for enhanced efficacy in patients who were earlier vs. later in their disease course and levodopa treatment pathway. INTRODUCTION More than 50 years since its introduction, levodopa (L-DOPA) remains the most efficacious treatment for Parkinson's disease (PD) (1). The long-term success of L-DOPA is compromised by the development of motor complications, but recent studies have shown that delaying the initiation of L-DOPA results in a reduced quality of motor control that is not offset by longer-term benefits (2-5). Indeed, longer disease duration at the start of L-DOPA therapy is an independent and important risk factor for the development of motor fluctuations and dyskinesias, as is the dose (but not the duration) of L-DOPA used (6,7). It has been proposed that the emergence of response fluctuations and drug-induced dyskinesias in the course of sustained treatment with L-DOPA results from discontinuous drug delivery and pulsatile stimulation of striatal dopamine receptors, which result in downstream changes in the basal ganglia (8,9). Furthermore, response fluctuations are attributed to increasing loss of buffering capacity in progressively diminishing neurons (9). Hypothetically, improving bioavailability and steadiness of exogenous L-DOPA may result in a more extended ON-time period and less troublesome dyskinesia in patients in early stages of PD when the pulsatile stimulation of the system is not yet severe, and the priming effect is less profound compared to patients with more advanced disease. Once established, such motor complications can be difficult to treat, but a variety of pharmacological and nonpharmacological interventions have shown efficacy in clinical trials (10,11). A common initial approach to wearing-off effects is to modify the administration of L-DOPA, often by using smaller, more frequent doses of L-DOPA, increasing the total dose of L-DOPA, or switching to controlled-release or modified-release L-DOPA preparations. In most patients, these strategies are at best successful for a year or two (12,13). Prolongation of the clinical effect of L-DOPA by co-administering with a long-acting dopamine agonist (DA) (14) or catechol-O-methyltransferase (COMT) inhibitor (15), or by preventing dopamine degradation in the brain with a selective monoamine oxidase inhibitor (MAO-BI) (16), are other effective strategies. COMT inhibitors extend the half-life and bioavailability of L-DOPA and may lead to a more continuous delivery of L-DOPA to the brain (15). Opicapone (OPC) is a third-generation, oncedaily COMT inhibitor developed to fulfill the need for a more potent, longer-acting COMT inhibitor, with a well-established safety profile (17)(18)(19)(20). OPC has been shown to be generally welltolerated and efficacious in reducing OFF-time in two pivotal trials in patients with PD and end-of-dose motor fluctuations (BIPARK-I and -II) (21,22). On the basis of these trials, OPC was first approved in the European Union as adjunctive therapy to preparations of L-DOPA/dopa decarboxylase inhibitors in adult patients with PD and end-of-dose motor fluctuations who cannot be stabilized on those combinations (23). Presently, it is also approved and marketed in the USA, Japan, South Korea, Australia, and other countries. We have now conducted exploratory post-hoc analyses of data from the BIPARK-I and -II trials (21,22) to evaluate the efficacy of OPC following a pre-defined segmentation of the wide spectrum of motor fluctuations in PD, based on baseline diseaseand therapy-related characteristics. Study Design BIPARK-I and -II were Phase III, multicenter, randomized, double-blind, placebo (PLC)-controlled trials of OPC as an adjunct to L-DOPA in patients with PD with end-of-dose motor fluctuations, details of which have been published previously (21,22). The trials had similar designs (Figure 1), eligibility criteria, and methods. In BIPARK-I, patients were randomized to treatment with OPC (5, 25, or 50 mg once daily), PLC, or entacapone (200 mg with every L-DOPA intake) for 14-15 weeks (21). In BIPARK-II, patients were randomized to treatment with OPC (25 or 50 mg once daily) or PLC for 14-15 weeks (22). In both trials, the primary efficacy endpoint was change from baseline to endpoint in absolute OFF-time vs. PLC, based on patient diaries (21,22). In the current study, data from matching treatment arms in BIPARK-I and -II were combined for the PLC and OPC 50mg groups and exploratory post-hoc analyses were performed to investigate the efficacy and safety/tolerability of OPC 50 mg vs. PLC in patients who were divided on the basis of baseline diseaseand therapy-related characteristics into representative subgroups of patients who were in "earlier" or "later" stages of both their disease course and L-DOPA treatment pathway, within the motor fluctuations spectrum of PD. Study Population In BIPARK-I and -II, eligible patients were male or female, aged 30-83 years, with a ≥3-year diagnosis of idiopathic PD, Hoehn and Yahr (H&Y) 1-3 at ON-state, who were receiving L-DOPA treatment for ≥1 year and experiencing end-of-dose motor fluctuations. Details of the full inclusion/exclusion criteria from the trials have been published previously (21,22). These post-hoc analyses included all patients treated with OPC 50 mg and PLC in BIPARK-I and -II. Study Assessments Baseline characteristics, efficacy, and safety/tolerability were assessed for each patient pairwise baseline subgroup, defined on the basis of a putative segmentation of the motor fluctuations spectrum, for both disease-and therapy-related characteristics. Disease-related characteristics comprised duration of PD (<6 years vs. ≥6 years; <7 years vs. ≥7 years; <8 years vs. Efficacy variables consisted of absolute OFF-time, total ONtime, and ON-time with troublesome |
dyskinesia, evaluated in patients treated with OPC 50 mg or PLC. Safety/tolerability is not addressed here as it is planned to publish this separately. Statistical Analyses Patient disposition and demographic/baseline characteristics were assessed for the Safety Set, which included all patients who received at least one dose of study drug. Efficacy assessments were conducted for the Full Analysis Set (FAS), which included all randomly assigned patients who took at least one dose of study drug and had at least one post-baseline efficacy assessment. Subgroup analyses were performed via an analysis of covariance (ANCOVA) that modeled the change of each efficacy variable from baseline to endpoint as a linear fixed-effect model of study and geographical area as factors and baseline respective pairwise variables as covariate in the FAS. Each pairwise comparison was analyzed separately, so multiple comparison correction was not required. Ninety-five percent confidence intervals and matching p-values were derived for the least square (LS) mean estimates and their differences. The last observation carried forward (LOCF) was applied to handle missing diary data. Forest plots are presented to visually assess differentiation for each pairwise subgroup. Study Population In total, 535 patients were randomized to receive PLC or OPC 50 mg in BIPARK-I and -II (Figure 2). The Safety Set included 522 patients (PLC, n = 257; OPC 50 mg, n = 265) and the FAS included 517 patients (PLC, n = 255; OPC 50 mg, n = 262). In the overall OPC 50 mg Safety Set, 60.4% of patients were male, mean (standard deviation [SD]) age was 64.5 (8.8) years, mean (SD) duration of PD was 7.6 (4.3) years, mean (SD) time since onset of motor fluctuations was 2.7 (2.9) years, mean (SD) H&Y staging at ON was 2.4 (0.5), mean (SD) absolute OFF-time at baseline was 6.2 (2.0) h, mean (SD) L-DOPA dose at baseline was 698.4 (322.1) mg/day, and mean (SD) duration of L-DOPA therapy was 6.3 (4.4) years. Baseline characteristics of the overall PLC Safety Set were similar to the OPC 50 mg Safety Set (24). Baseline characteristics by OPC 50 mg and PLC subgroups are summarized in Supplementary Tables 1, 2, respectively. Efficacy OPC 50 mg was significantly more effective than PLC in reducing OFF-time from baseline in the majority of subgroup analyses (p < 0.05), the exceptions being patients who received ≥6 L-DOPA intakes (p = 0.0623), patients with L-DOPA treatment duration ≥7 years (p = 0.1352), patients with L-DOPA treatment duration ≥8 years (p = 0.2309), and patients treated with ≥700 mg/day L-DOPA (p = 0.0640) (Table 1; Figure 3). Moreover, patients who were in "earlier" stages of both their disease course and L-DOPA treatment pathway experienced numerically greater efficacy when using OPC 50 mg, in comparison with those in "later" phases. OPC 50 mg demonstrated greater efficacy vs. PLC in each pairwise subgroup, with the following two exceptions: patients who received <5 L-DOPA intakes vs. ≥5 L-DOPA intakes (−57.5 vs. −60.8 min) and patients who received L-DOPA without an MAO-BI vs. those who received L-DOPA plus an MAO-BI (−58.6 vs. −63.7 min) (Table 1; Figure 3). Nevertheless, the OPC 50 mg magnitude of effect for these two exceptions was greater in each pairwise subgroup of patients who were in "earlier" phases of their motor fluctuation trajectory. OPC 50 mg was also significantly more effective than PLC in increasing total ON-time from baseline in the majority of subgroup analyses (p < 0.05), excluding the following: patients with duration of PD ≥8 years (p = 0.0541), patients with onset of motor fluctuations >2 years previously (p = 0.0527), patients who received ≥6 L-DOPA intakes (p = 0.0767), patients with L-DOPA treatment duration ≥7 years (p = 0.4855), and patients with L-DOPA treatment duration ≥8 years (p = 0.4902) (Supplementary Table 3). As for OFF-time reduction, patients who were "earlier" regarding both their disease course and L-DOPA treatment pathway experienced numerically greater efficacy when using OPC 50 mg, in comparison with those in "later" phases. OPC 50 mg demonstrated enhanced efficacy vs. PLC in each pairwise subgroup, except for patients who received L-DOPA without an MAO-BI vs. those who received L-DOPA plus an MAO-BI (59.8 vs. 77.7 min) (Supplementary Table 3). Nevertheless, the OPC 50 mg magnitude of effect even for this exception was greater in the pairwise subgroup of patients who were in "earlier" phases. Increases from baseline in ON-time with troublesome dyskinesia were not significantly greater for OPC 50 mg in comparison with PLC in all subgroup analyses (p ≥ 0.05), with the following exceptions: patients who received ≥5 L-DOPA intakes (p = 0.0095), patients with L-DOPA treatment duration ≥4 years (p = 0.0295), and patients with L-DOPA treatment duration ≥6 years (p = 0.0148)-all in the pairwise subgroups of patients who were in "later" phases (Supplementary Table 4). Moreover, differences between OPC 50 mg vs. PLC in the increase from baseline in ON-time with troublesome dyskinesia were less in the majority of subgroups of patients who were "earlier" vs. "later" in both their disease course and L-DOPA treatment pathway, the exceptions being the following: patients with PD duration <9 vs ≥9 years (11.1 vs. 10.7 min), patients with L-DOPA treatment duration <8 vs. ≥8 years (12.4 vs. 8.8 min), patients whose daily L-DOPA amount was <700 vs. ≥700 mg (13.0 vs. 9.5 min), and patients who received L-DOPA without a DA vs. those who received L-DOPA plus a DA (12.0 vs. 11.1 min) (Supplementary Table 4). Nevertheless, none of the differences were more than 5 min between each pairwise subgroup. DISCUSSION These exploratory post-hoc analyses of BIPARK-I and -II demonstrated that OPC 50 mg is efficacious over the whole trajectory of motor fluctuation evolution in PD patients, with similar effect sizes in subjects with recent onset of wearingoff effects and those in more advanced stages. OPC 50 mg was significantly more effective than PLC in reducing OFF-time and increasing ON-time for nearly all the subgroups that were analyzed (p < 0.05). The patients with shorter disease duration and duration of motor fluctuations, and those who were relatively early in their L-DOPA treatment pathway, experienced greater efficacy when using OPC 50 mg than those with later PD stages. Even in the "early" subgroups for which statistical significance was not demonstrated, there was a trend toward superiority of OPC 50 mg over PLC (p-values between 0.05 and 0.1). Changes in ON-time with troublesome dyskinesia were small and did not differ significantly from PLC for nearly all subgroup analyses (p ≥ 0.05). Furthermore, differences between OPC 50 mg vs. PLC in the increase from baseline in ON-time with troublesome dyskinesia were less in most of the subgroups of patients who were "earlier" in both their disease course and L-DOPA treatment pathway. These findings not only indicate that OPC is efficacious across all stages of development of motor fluctuations in PD patients, but also that patients who are at an early stage of their disease course may especially benefit from its introduction. L-DOPA is the most effective symptomatic treatment for PD from the early stages of the disease (2-4). The priority of treatment is therefore to obtain clinically meaningful benefit from each L-DOPA intake by facilitating its delivery to the brain. Optimization of the peripheral metabolism of L-DOPA through COMT inhibition is a rational first approach. When OPC is used, this also has the advantage of allowing a simplified drug regimen, since, unlike entacapone and tolcapone, OPC is administered once daily (20). In the prospective, multicenter, open-label OPTIPARK study, a total of 495 patients were treated with OPC 50 mg for 3 (Germany) or 6 (UK) months, in addition to their current L-DOPA and other anti-Parkinsonian treatments, and 393 (79.4%) patients completed 3 months of treatment (25). After 3 months, 71.3% of patients showed improvement on the Clinician's Global Impression of Change (primary endpoint) and 76.9% experienced improvement on the Patient Global Impressions of Change (25). These findings complement existing evidence from BIPARK-I and -II (21,22), by demonstrating that the efficacy of OPC 50 mg observed in the clinical trials was also experienced by PD patients with motor fluctuations treated in everyday routine clinical practice. The current study was exploratory in nature and involved a post-hoc analysis. The BIPARK trials were not powered for the subgroups included in the analysis and low patient numbers in some subgroups may have led to insufficient statistical power to detect differences. Moreover, there were differences in the magnitude of effect of both OPC and PLC between subgroups and it is therefore important to consider not only the overall treatment difference for OPC vs. PLC but also the magnitude of effect of both OPC and PLC when interpreting the findings for individual subgroups. Further analysis is planned to try to identify patient profile(s) that might particularly benefit (or not benefit) from OPC therapy. In summary, this study supports the efficacy of OPC 50 mg, in comparison with PLC, across the entire trajectory of motor fluctuation development in PD, from very early fluctuation to those with more advanced stages. It also indicates that patients who were in "earlier" stages in relation to their disease duration and the time since first occurrence of motor fluctuations may have enhanced efficacy when using OPC; further work is required to establish this. The pathophysiological basis for this remains unclear but may relate to less advanced nigrostriatal denervation and less severe pulsatile stimulation of the system compared to later disease stages. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. ETHICS STATEMENT The studies involving human participants were reviewed and approved by Institutional Review Boards at the participating sites (see Supplementary Material for full list). The patients/participants provided their written informed consent to participate in this study. AUTHOR CONTRIBUTIONS All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication. FUNDING The study, data analysis, and manuscript preparation were funded by Bial -Portela & C a , S.A. On $G$-sequential connectedness Recently, Cakalli has introduced a concept of $G$-sequential connectedness in the sense that a non-empty subset $A$ of a Hausdorff topological group $X$ is $G$-sequentially connected if there are no non-empty, disjoint $G$-sequentially closed subsets $U$ and $V$ meeting $A$ such that $A\subseteq U\bigcup V$. In this paper we investigate further properties of $G$-sequential connectedness and prove some interesting theorems. Introduction The concept of connectedness and any concept related to connectedness play a very important role not only in pure mathematics but also in other branches of science involving mathematics especially in geographic information systems, population modeling and motion planning in robotics. In [1], Connor and Grosse-Erdmann have investigated the impact of changing the definition of the convergence of sequences on the structure of sequential continuity of real functions. Cakalli [2] extended this concept to topological group setting and has introduced the concepts of G-sequential compactness and G-sequential continuity; and has investigated some results in this generalized setting (see also [3]). One is often relieved to find that the standard closed set definition of connectedness for topological spaces can be replaced by a sequential definition of connectedness. That many of the properties of connectedness of sets can be easily derived using sequential arguments has also been, no doubt, a source of relief to the interested mathematics instructor. Recently, Cakalli [4] has defined G-sequential connectedness of a topological group and investigated some results in this generalized setting. The purpose of this paper is to develop some further properties of G-sequential connectedness in metrizable topological groups, and present some interesting results. Preliminaries Before giving some results on G-sequential connectedness we remark some background as follows. Throughout this paper, N denotes the set of all positive integers and X denotes a Hausdorff topological group written additively satisfying the first axiom of countability. We use boldface letters x, y, z, ... for sequences x = (x n ), y = (y n ), z = (z n ), ... of terms of X. s(X) and c(X) respectively denote the set of all X-valued sequences and the set of all X-valued convergent sequences of points in X. A sequence (x k ) of points in X is called to be statistically convergent to an element ℓ of X if for each neighborhood U of 0 lim n→∞ |
1 n |{k ≤ n : x k − ℓ / ∈ U }| = 0, and this is denoted by st − lim n→∞ x n = ℓ. Statistical limit is an additive function on the group of statistically convergent sequences of points in X (See [17] for the real case and [18], [19], [20] for the topological group setting and see [21], and [22] for the most general case, i.e., topological space setting). A sequence (x k ) of points in X is called lacunary statistically convergent to an element ℓ of X if for every neighborhood U of 0 where I r = (k r−1 , k r ] and k 0 = 0, h r : k r − k r−1 → ∞ as r → ∞ and θ = (k r ) is an increasing sequence of positive integers. For a constant lacunary sequence θ = (k r ), the lacunary statistically convergent sequences in a topological group form a subgroup of the group of all X-valued sequences and lacunary statistical limit is an additive function on this space (see [18] for topological group setting and see [23] and [24] for the real case). Throughout this paper, we assume that lim inf r kr kr−1 > 1. By a method of sequential convergence, or briefly a method, we mean an additive function G defined on a subgroup c G (X) of s(X) into X [2]. A sequence x = (x n ) is said to be G-convergent to ℓ if x ∈ c G (X) and G(x) = ℓ. In particular, lim denotes the limit function lim x = lim n x n on the Clearly if f is G-sequentially continuous on X, then it is G-sequentially continuous on every subset Z of X, but the converse is not necessarily true since in the latter case the sequences x' s are restricted to Z. This was demonstrated by an example in [1] for a real function. The notion of regularity introduced above coincides with the classical notion of regularity for summability matrices and with regularity in a topological group for limitation methods. See [16] for an introduction to regular summability matrices, [25] for an introduction to regular limitation (summability) methods and see [26] for a general view of sequences of reals or complex. We recall the definition of G-sequential closure of a subset of X from [2] and [3]. If A ⊂ X and G and say that a subset A is G-sequentially closed if it contains all of the points in its G-closure, It is clear that φ G = φ and X G = X for a regular method G. If G is a regular method, then Note that even for regular methods, it is not always true that ( Even for regular methods, the union of any two G-sequentially closed subsets of X need not be a G-sequentially closed subset of X as seen considering Counterexample 1 in [3]. Ç akallı has introduced the concept of G-sequential compactness and proved that G-sequentially continuous image of any G-sequentially compact subset of X is also G-sequentially compact [2, Theorem 7]. He has defined G-sequential continuity and obtained further results in [3] (see also [27], [28], [29], [30], [31], [32], [33] for some other types of continuities which can not be given by any sequential method and see [34] for some kinds of continuities which coincide with uniform continuity when the domain of the function is connected). A function f : X → X is G-sequentially continuous at a point u if, given a sequence (x n ) of points in X, G(x) = u implies that G(f (x)) = f (u). Recently Mucuk and Sahan [35] have investigated further properties of G-sequential closed subsets of X. They have modified the definition of open subset to G-sequential case in the sense that a and obtained that the union of any G-sequentially open subsets of X is G-sequentially open. From the fact that for a regular sequential method G, G-sequential closure of a subset of X includes the . If a function f is G-sequentially continuous on Huang and Lin [36]; and Fedeli and Donne [37] have investigated sequential connectedness. Cakalli [4] has recently introduced the concept of G-sequential connectedness as follows. Definition 1. A non-empty subset A of a topological group X is called G-sequentially connected if there are no non-empty, disjoint G-sequentially closed subsets U and V of X meeting A such that A ⊆ U V . Particularly X is called G-sequentially connected, if there are no non-empty, disjoint G-sequentially closed subsets of X whose union is X. The following Theorem is given in [4, Corollary 1]. Theorem 1. If G is a regular sequential method and A is a G-sequentially connected subset of X, We recollect the following definition. Definition 2. [4, Definition 2] Let Here we remark that a subset B We note that C x G coincides with the ordinary sequential connected component of x when G = lim. We write π 0 X G for the set of G-sequentially connected components of all points in X and similarly write π 0 A G for the set of all G-sequentially connected components of all points in a subset A. Lemma 1. Let x, y ∈ X. If x and y are in a G-sequentially connected subset A of X, then x and y are in the same G-sequentially component of X. Proof: Let x and y be contained in a G-sequentially connected subset A of X. Then x, y ∈ A ⊆ C x G and x, y ∈ A ⊆ C y G . So C x G ⊆ C y G and C y G ⊆ C x G . Therefore C x = C y . ✷ Lemma 2. The G-sequentially connected components of X form a partition of X. Proof: It is obvious that G-sequentially connected components form a cover of X. We prove that for x, y ∈ X if the components C x G and C y G intersect, then C Proof: Let f : A → B be a G-sequentially homeomorphism. Define a map induced by the function f : A → B. Since f is G-sequentially continuous, the image of a Gsequentially connected subset is G-sequentially connected [4,Theorem 1]. Hence if y ∈ C x G then by Lemma 1, f (x) and f (y) are in the same G-sequentially connected component and so , then G-sequentially components of x and y are same, i.e, C x G = C y G and therefore π 0 (f ) is any regular sequential method G. Proof: Since the G-sequentially connected component C G x is G-sequentially connected, by For a regular sequential method G, the G-sequentially connected component of the identity is a G-sequentially closed, normal subgroup of X. Proof: Write K for the G-sequentially connected component of the identity point 0. By Theorem 5, K is G-sequentially closed. To prove that K is a subgroup, we prove that K − K ⊆ K, where K − K is the set of all points x − y for x, y ∈ K. Since K is a G-sequentially connected subset, for each x ∈ K, the set x − K is G-sequentially connected as the image of G-sequentially connected subset K under a G-sequentially continuous function. Then by Theorem 2 is G-sequentially connected as a union of G-sequentially connected subsets including 0. But the largest G-sequentially connected subset including 0 is K. Therefore K − K ⊆ K, i.e, K is a subgroup. Further for any a ∈ X the function f a : X → X, x → a + x − a is G-sequentially continuous. So f a (K) = a + K − a is a G-sequentially connected subset including 0 ∈ X. Therefore a + K − a ⊆ K, which implies that K is normal. ✷ Let X be a topological group and U a symmetric neighbourhood of the identity 0. We say that X is generated by U , if each element of X can be written as a sum of some elements in U . Theorem 7. Let G be a regular sequential method. If X is generated by a G-sequentially connected and symmetric neighbourhood U of the identity, then X is G-sequentially connected. Proof: We prove that every point of X is in the G-sequentially connected component of 0. Let x ∈ X. Since X is generated by U , the point x can be written as x = x 1 + x 2 · · · + x n for some x i ∈ U . So x ∈ (U + U + · · · + U ) which is G-sequentially connected by [4,Lemma 7] and includes the identity point 0. Therefore by Lemma 1, x and 0 are in the same G-sequentially connected component. This proves that X has only one G-sequentially connected component, i.e, X is G-sequentially connected. ✷ In the following theorem we prove that a G-sequentially connected topological group is generated by any G-sequentially open neighbourhood of the identity point of X. Theorem 8. Let G be a regular sequential method. If X is G-sequentially connected, then X can be generated by any G-sequentially open and symmetric neighborhood of the identity. Proof: Let U be a G-sequentially open and symmetric neighbourhood of the identity point 0 ∈ X and let X U be the subgroup of X generated by U . Hence each x ∈ X U can be written as a sum x = x 1 + · · · + x n of the points in U and so x ∈ (U + U + · · · + U ). Therefore X U can be written as a union of the sums U + U + · · · + U . Here by [35,Theorem 15] each sum U + U + · · · + U is G-sequentially open and includes 0. By [35, Theorem 3] X U becomes G-sequentially open and therefore by Theorem 3, X U is also G-sequentially closed. Since X is G-sequentially connected it follows that X U = X. ✷ In the following definition of G-sequentially locally connectedness by a G-sequentially connected neighbourhood of a point x ∈ X, we mean a G-sequentially connected subset which contains a G-sequentially open subset including x. Definition 4. Let G be a sequential method on X. We call X as G-sequentially locally connected, if for any G-sequentially open neighbourhood U of x, there is a G-sequentially connected Proof: Let X be G-sequentially locally connected. Let A be a G-sequentially open subset of X, C a G-sequentially connected component of A and x ∈ C. Since X is G-sequentially locally connected, there is a G-sequentially connected neighbourhoof U x of x such that U x ⊆ A. But since the largest G-sequentially connected subset of A containing x is C, we have that x ∈ U x ⊆ C. Conclusion The present work contains further results on G-sequentially connectedness in first countable and Hausdorff topological groups. So that one may expect it to be more useful tool in the field of topology in modeling various problems occurring in many areas of science, geographic information systems, population modeling and motion planning in robotics. It seems that an investigation of the present work taking 'nets' instead of 'sequences' could be done using the properties of 'nets' instead of using the properties of 'sequences'. As the vector space operations, namely, vector addition and scalar multiplication, are continuous in a cone normed space so cone normed spaces are special topological groups, we see that the results are also valid in cone normed spaces (see [38] for the definition of a cone normed space). For further study, we also suggest to investigate the present work for the fuzzy case. However, due to the change in settings, the definitions and methods of proofs will not always be analogous to those of the present work (see [39] for the definitions in the fuzzy setting). Abstracts from FEDELAT 2016 meeting: Latin American Federation of IASP chapters s from FEDELAT 2016 meeting: Latin American Federation of IASP chapters Alfredo Covarrubias-Gómez, Elias Atencio |
We are pleased to share the abstracts of presentations from the recent XI Congreso Latinoamericano de dolor FEDELAT, June 15 to 17, 2016 in Costa Rica. The Latin American Federation of IASP Chapters (FEDELAT, in Spanish) was founded in 1990 by the presidents of the Mexican, Colombian, Argentinian, and Puerto Rico’s IASPChapters. To date FEDELAT is composed of 18 IASP chapters. The main purpose of the Federation is to encourage education and networking relations among chapters strengthening regional collaboration. Over the last 14 years FEDELAT has maintained a close relation with the Spanish and Portuguese IASP chapters. Together, the 3 institutions, conducted the Ibero-American Reunion (RIAD, in Spanish). This event takes place every year, gathers professionals from both Europe and Latin-America, and exposes the regional advances on diverse pain issues. FEDELAT had an important collaboration in events that promote education in Latin America. The former president, Dr Fabián Piedimonte, fromboth FEDELAT and the Argentinian IASP chapter represented the regional committee that organized the IASP World Pain Congress in Buenos Aires. FEDELAT’s president is Dr Elias Atencio. Dr Atencio and the directive board have continued this work and over the last 2 years had organised the following events: (1) The Hands-On Course on interventional techniques for pain control which took place in Buenos Aires (Argentina) from April 20 to 30, 2016; another will take place in Panama City (Panama) from November 10 to 20, 2016. (2) The Annual RIAD occurred in Sevilla, Spain in 2015 and in San Jose, Costa Rica in 2016. In Costa Rica most lectures were centered on “articular pain” according to the IASP International Day Against Pain. (3) FEDELAT is collaborating with the Mexican IASP Chapter to promote and organize the IASP Pain Management Camp. This second event in Latin America will include lectures about articular pain, cancer pain, acute pain, and opioids pharmacology and will take place October 10 to 15, 2016 in Mexico City, Mexico. (4) Also the federation is developing research protocols to assess epidemiological aspects of chronic pain and conducted Practice Guidelines adapted to Latin American needs. These guidelines were published in a peer reviewed medical journal indexed in LILIACS. During the Annual RIAD 4 general modules were discussed cancer pain and palliative care, research and education to improve pain curricula in Latin America, pharmacological pain relief, and interventional techniques for pain relief. In conclusion, the Federation and regional IASP Chapters are working together, harder than ever, in order to achieve a Latin America Without Pain. Alfredo Covarrubias-Gómez, MD Pain and Palliative Medicine http://www.covarrubias-gomez.org Federación Latinoamericana de Asociaciones para el Estudio del Dolor (FEDELAT) Latin-American Federation of Associations for the Study of Pain Secretary Mexican Association for the Study and Treatment of Pain: An International Association for the Study of Pain Chapter Past president National Centre for the Study of Pain CEO and founder Department of Pain and Palliative Medicine, Mexican NHI on Internal Medicine Attending physician/Associate professor Acknowledgement:We thank Dr Melissa Garcia-Pons for her invaluable help with translation of some of the abstracts. We are pleased to share the abstracts of presentations from the recent XI Congreso Latinoamericano de dolor FEDELAT, June 15 to 17, 2016 in Costa Rica. The Latin American Federation of IASP Chapters (FEDELAT, in Spanish) was founded in 1990 by the presidents of the Mexican, Colombian, Argentinian, and Puerto Rico's IASP Chapters. To date FEDELAT is composed of 18 IASP chapters. The main purpose of the Federation is to encourage education and networking relations among chapters strengthening regional collaboration. Over the last 14 years FEDELAT has maintained a close relation with the Spanish and Portuguese IASP chapters. Together, the 3 institutions, conducted the Ibero-American Reunion (RIAD, in Spanish). This event takes place every year, gathers professionals from both Europe and Latin-America, and exposes the regional advances on diverse pain issues. FEDELAT had an important collaboration in events that promote education in Latin America. The former president, Dr Fabián Piedimonte, from both FEDELAT and the Argentinian IASP chapter represented the regional committee that organized the IASP World Pain Congress in Buenos Aires. FEDELAT's president is Dr Elias Atencio. Dr Atencio and the directive board have continued this work and over the last 2 years had organised the following events: (1) The Hands-On Course on interventional techniques for pain control which took place in Buenos Aires (Argentina) from April 20 to 30, 2016; another will take place in Panama City (Panama) from November 10 to 20, 2016. (2) The Annual RIAD occurred in Sevilla, Spain in 2015 and in San Jose, Costa Rica in 2016. In Costa Rica most lectures were centered on "articular pain" according to the IASP International Day Against Pain. (3) FEDELAT is collaborating with the Mexican IASP Chapter to promote and organize the IASP Pain Management Camp. This second event in Latin America will include lectures about articular pain, cancer pain, acute pain, and opioids pharmacology and will take place October 10 to 15, 2016 in Mexico City, Mexico. (4) Also the federation is developing research protocols to assess epidemiological aspects of chronic pain and conducted Practice Guidelines adapted to Latin American needs. These guidelines were published in a peer reviewed medical journal indexed in LILIACS. During the Annual RIAD 4 general modules were discussed cancer pain and palliative care, research and education to improve pain curricula in Latin America, pharmacological pain relief, and interventional techniques for pain relief. In conclusion, the Federation and regional IASP Chapters are working together, harder than ever, in order to achieve a Latin America Without Pain. Alfredo Attending physician/Associate professor 6 mgME "per cápita" (mgME, milligrams in morphine equivalence) while in Canada and USA report an average of 722.7 mgME and 717.8 mgME "per cápita" respectively. Opioid therapy for cancer-related pain management poses 3 major issues: (1) limited opioid use promoting an inefficient pain control, (2) rightful concerns about inappropriate opioid use even in the cancer population, and (3) presence of chronic pain in cancer survivors. Pharmacological management includes: (1) Non-steroid Anti-inflammatory Drugs (NSAIDs): These drugs had shown insufficient analgesia in moderate and severe pain intensity, an analgesic ceiling effect without dose-dependent relation, and a dose-dependent occurrence of adverse effects. (2) Opioid therapy includes Fentanyl, Oxycodone, Tapentadol, Methadone, Buprenorphine, and Morphine. These drugs are effective and safe for cancer-related pain management. However some of them pose issues related to the strength of the evidence (Methadone and Tapentadol). Transcutaneous Fentanyl had been studied in at least 4 meta-analyses for breakthrough pain. Non-pharmacological management includes: (1) Neurolytic celiac plexus block: This approach reported a persistent benefit in a long-term follow-up beyond 3 months. (2) Transcutaneous Electrical Nerve Stimulation (TENS): This technique failed to provide sufficient evidence to determine its effectiveness in cancer-related pain. (3) Acupuncture: This approach failed to provide sufficient evidence to determine its effectiveness in cancer-related pain. (4) Massage: Despite being weak, the available evidence suggests a beneficial effect of massage techniques for cancer pain relief. (5) Traditional herbal medicine: This approach combined with conventional therapy might be efficacious as an adjunctive therapy for patients with cancer pain. Pain in cancer survivors is observed in 30% to 60% of survivors. Although there is not any specific guideline opioids may be considered for moderate to severe pain that has been unresponsive to non-opioid therapies and non-pharmacologic approaches and when the chronic opioid therapy is likely to possess equivalent or better risk-to-benefit ratios. Treatment with ozone is increasing worldwide as a result of its simplicity of application, high efficiency, good tolerance and almost no side effects. In 1785, Dr Martin van Marum, notes that a gaseus substance with characteristic odor and oxidative properties is produced as a result of a strong electrical discharge in the air. In 1840 Prof Cristian Schobein correlated the described changes with the formation of a gas called ozone. Composed by 3 oxygen atoms, the ozone is an unstable gas, denser and more soluble in water than oxygen. It is a powerful oxidant that can be produced by chemical electrolysis, electric discharge and ultraviolet radiation. Ozone cannot be stored and must be used immediately after its production because of its instability and rapid decomposition. At the concentrations used in clinical practice, ozone is believed to have immunomodulatory, anti-inflammatory, antibacterial, antiviral, fungicidal and analgesic characteristics. The application of ozone for treatment of pain is based on its potent analgesic and anti-inflammatory effect, resulting from decreased production of inflammatory mediators and oxidation (inactivation) of mediators of pain, improvement of microcirculation and tissue oxygenation, elimination of toxins and activation of endogenous analgesic mechanisms. Ozone therapy has several clinical indications in the area of pain, especially in low back pain, herniated discs, disc-radicular disease and osteo-articular pathology. It can be applied locally or parenterally, and may be used more than one pathway in combination with a synergistic effect between them. Inhalation is contraindicated because it is highly toxic to the lungs. Further, direct intravenous injection of ozone is not recommended because of the risk of embolism. Ozone can be administered locally to intervertebral disc, paravertebral or epidural spaces, intramuscular, intraarticular, subcutaneous, in trigger points, vagina, bladder, urethra and rectum. Another way is parenteral auto-hemotherapy, consisting of a mixture of ozone and blood withdrawn from the patient by a specific device. Blood is then reinfused, without contact with the external environment. Maria A. Rico, MD Anestesiologist/Pain Specialist, Clinica Alemana Santiago, Chile Introduction: In recent years there is a growing interest of patients and their families in the potential benefit derived from cannabis sativa use, to relieve cáncer related pain and other symptoms, and other chronic diseases. The existence of an endocannabinoids system and specific receptors are known since 1990. Mechanism of action: Agonism of CB1 receptors modulate synaptic transmission of excitatory and inhibitory circuits. Effects depend on the specific neural area stimulated, acting upon analgesia, memory process, appetite, muscle tone, etc. Activation of CB2 receptors may influence the activity of cytokines in the immune response to inflammation. Scientific evidence of cannabinoids in medicine: Nausea/ vomiting in patients with cancer and chemotherapy, using synthetic derivatives as dronabinol and nabilone, showed significant benefit until year 2000. The introduction of antiemetic agents, such HT3 blockers, leave cannabinoids as a third line indication. The effect of increased appetite with THC derivatives is significant only in HIV patients; it has not been seen in cancer patients. Anxiolytic effects and improved night sleep, although they are popular, they are not support in the literature. Benefits in epilepsy and multiple sclerosis has been stated. The use in refractory epilepsy, is becoming very popular, but it is still anecdotal. Cannabinoids as analgesics: Comparative studies of cannabinoids and opioids demonstrate an efficacy equivalent to a weak opioid as dihydrocodeine, with limitation of adverse effects in increasing doses (Dizziness, dry mouth, fatigue, drowsiness, euphoria, agitation). Non oncologic pain: Systematic reviews show studies in a wide variety of pathologies and doses. Cannabinoids were administrated as natural and synthetic forms, via inhaled, oral and transmucosal route. The results are generally modest, with a relative greater efficacy in neuropathic pain, but always the benefits are limited by its non-negligible adverse effects when increasing doses. Then, the evidence of benefits is rather weak. In palliative care: For pain control, it seems beneficial as an adjuvant to opioids treatment, allowing for reduction of their doses, and not as a single analgesic. It has also been pointed out that cannabinoids can contribute to a better quality of life, related to reduced anxiety, better night time sleep, appetite and are generally well appreciated by patients and families. Conclusion: Cannabinoids in pharmaceutical forms could be useful in a selected group of patients, for the management of some symptoms. Dr Marco Antonio Narváez Tamayo Attending Specialist, Study and Treatment of Pain Unit at the Hospital Obrero No. 1 (Hospital materno-infantil), Caja Nacional de Salud, La Paz, Bolivia Introduction: Minimally invasive percutaneous techniques have drastically changed the therapeutic paradigm of conventional surgery. Current training, skills, instruments, image accuracy among other factors have brought us forward, achieving access to what was unthinkable many years, so we now can inhibit, control or solve problems that detract greatly the quality of life of our patients with refractory chronic pain. Objective: Determine the indications of interventional techniques in the refractory chronic pain management. Material and method: A non-systematic review of the literature was conducted, identifying articles on the basis of the following MeSH terms: indications, |
interventional treatment, technologies, blockade, radiofrequency, refractory chronic pain or neurolysis. Those who presented more than 3 terms were included in the search. Considerations included place or site where the study has developed, and the number of cases reviewed. Different techniques commonly used in specialized units of treatment pain were reviewed, selecting the most frequent and those enrolled with the strongest evidence available to date. These include thermal radiofrequency of the medial branch of dorsal roots, cervical epidural analgesia, percutaneous epidural neuroplasty with Catheter Racz, radiofrequency of the sacroiliac joint, vertebroplasty, kyphoplasty, cementing the hip in mode femoro-cementoplasty, disc nucleoplasty by radiofrequency, neurolysis of stellate ganglion, selective inhibition of the splanchnic nerves, celiac plexus neurolysis, superior hypogastric plexus and ganglion Walther (odd node), radiofrequency of the dorsal root ganglion, foraminoplasty, spinal infusion in chronic cancer pain, among the most studied. Results: In order to specifically review the indications for minimally invasive treatment, they were classified into 4 major groups: infusion, modulation, radio frequency and neurolysis. In this regard we show the results of some of the most common techniques of this particular chapter. Analysis also included procedures like control or relief of chronic pain in patients undergoing bone cementingm, vertebral cementation and cementation of long bones (femur, humerus, tibia). A common denominator for most of minimally invasive procedures, has been the low level in the pain intensity during the procedure, as well as in the post-operative period. Nevertheless, the biggest difference is represented by the lower risk, low complication rate and substantially a minimum percentage of adverse effects in this research group. Conclusion: The increasing development of minimally invasive techniques is proving in recent years, encouraging results in relation to the relief of chronic pain refractory to conservative measures. These techniques are safe when performed in specialized centers experience with interventional pain techniques and the necessary guidelines for the quality of the process are respected. Alicia Alonso Cardañ o, MD Hospital Universitario de Leon, Altos de Nava S/N, Leon, Leon 24071, Spain Opioid analgesics are effective tools for the treatment of moderate to severe pain of all individuals that are subject of the right and dignity of relief unnecessary suffering, but it is important to develop parallel tools to prevent and reduce the dramatic increase in opioid addiction behaviors. Comparison of opioid prescriptions in North America vs Europe might reveal factors involved in the so-called "opioid epidemic." The main controversy is about opioids for pain management of non-oncological origin or benign processes, in which long term therapies might be relevant. Bibliographic review of recent medical and scientific literature most recent published in journals of impact, in order to establish the current situation, with the target set in the extraction of predisposing factors, possible existence of predictive factors, investigate different ways of dealing with opioid prescription and even different ways of addressing similar conditions in different continents. It was found that prescribing opioids may contribute to the development of a misuse of opioids, which has led to the development of clinical guidelines for recommendations and specific programs by various associations both medical and governmental, plus the manifestations promoted by the group of patients and their corresponding associations scattered across different states. In conclusions, strategies to minimize addiction to treatments seem to be effective, although a longer time to determine these observations analysis will be needed, while the controversy over the use or misuse of opioid prescription still will coexist in parallel. The selection of one type of drug or another to address the pain of nonmalignant origin should be considered in the field of each chronical pain syndrome in question, taking into account the concurrent diseases of each patient, contraindications relating to the treatment, the response that the individual has shown to pretreatments, its beneficial effects, and even the preferences in this regard that the patient can present having in mind other possibilities of treatment for his/her syndrome in particular, as are other nonpharmacological measures. Evidence in favor of neurolytic blocks for cancer pain Prof Joã o Batista Santos Garcia, MD, PhD, Pain and Palliative Care Service-Cancer Hospital of Maranhao, Sa˜o Luis, Brazil. Pain is one of the most common symptoms of cancer, affecting over 90% of the patients during the course of their treatment. In advanced cases pain is referred to as severe in half the cases, and it originates from multiple factors. In order to address this issue, the World Health Organization (WHO) proposed an approach to the treatment of cancer pain patients using a 3 ladder stair system in which opioids are viewed as the cornerstone to the treatment, but also highlighting the importance of adjuvants. This strategy has been considered effective in controlling pain, reducing pain in 70% of patients, however, 30% of cancer patients still end up developing refractory pain. For these cases, interventional techniques have been proposed as a fourth step in the analgesic ladder, nonetheless pain specialists have considered this approach earlier in the course of treatment in order to lessen unnecessary suffering. Interventional therapy for cancer can be obtained through several techniques, including the injection of neurolytic substances. Neurolysis of the celiac plexus (NCPB) is often indicated for patients with pancreatic cancer with severe pain, although it may be helpful to other visceral tumors such as lower esophagus, gastric, hepatobiliary system, and intestinal to the proximal half of the transverse colon. NCPB improves pain, reduces opioid consumption, and reduces rates of side effects, including constipation, nausea, and vomiting, as compared with standard analgesic therapy. Neurolysis of the superior hypogastric plexus may be used to control pelvic cancer pain. And ganglion impar neurolysis is effective for rectal pain from rectal cancer or from radiotherapy complications (actinic rectitis). Intercostal nerve block is effective to manage pain arising from metastatic lesions to the ribs. One study showed that 80% of the patients noted greater than 50% improvement in pain and 56% reduced their analgesic use after the procedure. Nevertheless, evidence for those blocks is weak as it comes mainly from small case series or retrospective studies. Intrathecal neurolysis is a radical therapy indicated to patients suffering from perineal refractory pain. Since this procedure can produce significant morbidity, including bowel and/or bladder dysfunction, motor weakness, neuritis, and paresthesias, it is usually reserved for patients with low life expectancy and poor performance status. Although evidence for this neurolysis is favorable, it is also derived from poor quality studies. To conclude, interventional techniques may be used to promote better pain control for patients refractory to opioid therapy, or who are limited by side effects. Even so, since these procedures are invasive and are not free from treatment-related complications, clinicians must weigh possible benefits and risks before proceeding with the intervention. Patricia Gomez Department of Anaesthesia and Pain, Colombian National University, Bogotá, Colombia Their analgesic, and peripheral and central sensitization prevention actions make them the most formulated and used in self-medication drugs on a worldwide level. PubMed search of review articles with the keywords: Nonsteroidal anti-inflammatory & NSAIDs. Several adverse events has been reported: (1) Gastrointestinal (GI): these are one of the most common adverse events related to drugs. The relative risk of ulcers, perforations, and bleeding is of 5.3 higher in comparison to non-NSAID consumers. It is strongly recommended to identify GI risk factors and prophylaxis (including age above 60, presence of helicobacter, use of anti coagulants and other factors). The COX2 selective inhibitors are associated with the lower rate of GI complications and symptomatic ulcers. Nonetheless, the use of COX2 inhibitors can increase cardiovascular risks. (2) Cardiovascular (CV): NSAIDs use is associated with a little increase but consistent CV events like myocardial infarction risk, affected basically in part by the COX2 power inhibition; although all the NSAIDs except aspirin can be associated with a potential thrombotic risk. (3) Renal events: Renal toxicity is infrequent, but its risk is increased in those patients with renal or hepatic dysfunction, nephrotic syndrome, old age, diabetes, hypertension, cardiac congestive insufficiency or dehydration. (4) Allergic Reactions: Respiratory diseases exacerbation induced by NSAIDs, especially by non-specific NSAIDs are the most frequent hypersensitive drugs reactions. (5) Haematological events: Agranulocytosis and aplastic anemia are low frequency events. Due to the fact that agranulocytosis is a pharmacogenetic problem, local studies have to be made. Some additional myths and realities: (1) The selective COX2 are not more efficacious than non-selective NSAIDs. (2) intravenous or IM administration of NSAIDs do not produce less GI adverse events and are not more efficacious. (3) concomitant use of 2 or more NSAIDs does not increase their efficacy, but does increase their toxicity. (4) The COX2 inhibitors can be used in low CV risk patients for short periods of time. (5) Agranulocytosis due to dipyrone is uncommon in Latin America, Spain and other countries. According to Cochrane, a Mexican, Latin America, Spanish and Polish consensus, dipyrone is an efficient drug with a very favorable cost/benefit and risk/benefit profile. Complications of corticosteroid epidural injections Elías Atencio, MD, FIPP Attending Specialist, Department of Pain Medicine at the Caja de Seguro Social, Complejo Hospitalario Metropolitano, Panama City, Panama The complications associated with the injection of corticosteroids are rare, but a recent report (2012) noted an increase in fungal infections in the US, reporting a total of 650 cases with 39 deaths associated with this type of infection. Apart from the toxic effects of the corticosteroids in the intrathecal space, they are few the serious complications reported from this technique. A recent retrospective study examined a total of 4265 epidural corticosteroid injections in 1857 patients along 7 years. There were 161 cervical interlaminar injections, 123 lumbar, 17 caudal and 3964 transforaminal injections. No serious complications were identified, 103 minor complications (2.4%) were reported, including increased pain (1.1%), pain in the puncture site (0.33%), persistent heaviness (0.14%) and others (0.8%). The complications were less common with the transforaminal injections (2.1%) than with the interlaminar injections (6%). In general overall complications can be grouped as follows: (1) Neurotoxicity-arachnoiditis and aseptic meningitis due to intrathecal injection. The first is expressed as constant and burning low back and leg pain, increased urinary frequency and incontinence, muscle spasms in the back and legs, variable sensory disturbances and motor dysfunction. The second is usually a benign condition that causes signs of neurological irritation including burning pain in the legs, headache, meningismus and in severe cases seizures. It is difficult to determine which component of the steroid is the neurotoxic agent, Nelson suggests that polyethylene glycol is the causative agent. There is no definitive treatment for arachnoiditis or for aseptic meningitis; symptomatic treatment is the mainstay. (2) Other potential damages include cord injury by the injection needle, injection into a spinal artery with embolization. (3) Pharmacological effects such as cushingoid effects and impaired glucose tolerance. Anatomy of the trigeminal system, trigeminal neuralgia and its interventional treatment The revised international classification of headache suggests 3 variants of trigeminal neuralgia (TN): (1) classic or idiopathic trigeminal neuralgia; (2) trigeminal neuralgia with persistent concomitant facial pain; and (3) symptomatic trigeminal neuralgia, caused by a distinct structural lesion other that vascular compression. Nevertheless, in a clinical aspect, the proposal by Burchiel has a wider and more comprehensive sense, differentiating the trigeminal pain in 7 subtypes based on their characteristics and causes. In terms of its pathophysiology, current opinion suggests that the classic TN is caused by a proximal compression of the trigeminal nerve root close to the brainstem through a tortuous blood vessel (an artery or a vein), leading to a process of secondary demyelination probably mediated by ischemic damage by microvasculature impairment. TN is an infrequent disease; analysis of some of the available studies suggest that the prevalence in the general population ranges between 0.01% and 0.3%, and the gender ratio of women to men is about 3:2. It can occur at any age, but over 90% of the cases occur above age 40, being its peak of onset between 50 and 60 years old. The incidence of TN in multiple sclerosis is higher than in the general population. Pain is unilateral, with only 3% of bilateral involvement; V3 and V2 branches are the most commonly affected individually or combined, while V1 is rarely affected. The medical treatment is based on the use of antiepileptic drugs (AEDs). The first-line therapy, according to current treatment guidelines based on evidence, should be carbamazepine (200-1200 mg/d) and oxcarbazepine (600-1800 |
mg/d). The second-line treatment, with less evidence, consists of adding lamotrigine (400 mg/d) or other anti-epileptics and baclofen. Failure of medical treatment should lead to surgical procedures such as (1) microvascular decompression, (2) percutaneous damage to the Gesserian ganglion by injecting drugs (glycerol), thermocoagulation by radiofrequency, or balloon compression, and (3) radiosurgery, by Gamma Knife. Monetary cost associated to pharmacological treatment for patients with chronic non-oncological pain in Mexico Perceptions of the COVID-19 pandemic in Japan with respect to cultural, information, disaster and social issues A questionnaire survey was distributed via the Internet to 600 respondents. Preliminary results revealed that most Japanese people regularly washed their hands and had low resistance to wearing masks even before the COVID-19 pandemic. Internet news was the most common source of information. Half of the respondents said they would “stay at home evacuation” if a disaster occurred during the COVID-19 pandemic, reflecting the strategy promoted to reduce crowding in evacuation shelters. If a state of emergency must be reinstated, one-third of respondents said they could bear it for a few months and another one-third for a few weeks. A brief timeline of the COVID-19 outbreak in Japan Although news related to COVID-19 in China started to be broadcast in Japan in January, another critical news item was the cruise ship "Diamond Princess," which reported positive cases of COVID-19 in February 2020 [1]. A total of 712 people (as of 17 March 2020) out of 3711 became infected [2]. The number of new positive cases subsequently started increasing due to tourists from China visiting Japan during the Chinese New Year holiday in February 2020 and Japanese people returning to Japan from countries with severe outbreaks in March 2020. The situation in Japan became so severe that the government decided to announce a state of emergency on 7 April 2020 in seven prefectures (Tokyo, Chiba, Saitama, Kanagawa, Osaka, Hyogo, and Fukuoka); this was expanded to the whole country on 16 April 2020 [3]. The first peak in newly confirmed cases was 708 on 10 April 2020 [3]. The state of emergency was canceled in 39 prefectures on 14 May 2020 and for the whole country on 25 May 2020 when the daily number of new confirmed cases dropped below 50-100 for several days [3]. Many businesses gradually reopened, and the number of daily new confirmed cases increased until the second peak of 1595 on 7 August [3]. The third peak started at the beginning of November and reached 2508 cases per day on 21 November 2020 [3]. A summary of the number of daily new positive cases and the total number of severe cases in Japan is shown in Fig. 1, together with significant incidents. The second state of emergency was announced on 7 January 2021 in ten prefectures (Tokyo, Chiba, Saitama, Kanagawa, Gifu, Aichi, Osaka, Kyoto, Hyogo, and Fukuoka) when the daily number of new confirmed cases reached over 7000 cases per day. Experiencing and living with disasters during the COVID-19 outbreak in Japan Since March 2020, the global public opinion and press focused their attention on COVID-19, but this has not been the only crisis ongoing in the world. Countries such as Japan experienced the concurrence between the pandemic and other hazards, including the 2020 Kyushu floods that occurred on 4 July, causing more than 50 deaths [7]. The multiple events Progress in Disaster Science 10 (2021) 100158 happening at the same time created new challenges for emergency management, that were considered for developing the data collection of this research. The research followed the principle included in the last United Nations' Global Assessment Report on Disaster Risk Reduction, according to which "surprise is the new normal," and nonlinear events derive from series of shocks more than primary triggers alone [8]. Societal resilience to interacting, interconnected, and cascading risk begins with the identification of their common root causes and the cross-cutting aspects between their mitigation measures [9]. For example, in the early stage of the COVID-19 outbreak in Japan, academics and practitioners discussed how to improve the management of disaster warnings and disaster evacuation shelters considering emergency scenario where other hazards could have challenged the country during the pandemic [4][5][6]. This topic becomes an operational priority after the floods in July 2020. One of the main ideas was to reduce crowding in evacuation shelters and to encourage residents not to evacuate (stay at home evacuation) if they were actually safe (i.e., based on hazard maps) or to choose alternative evacuation destinations (i.e., friends' or relatives' houses, hotels, and department stores) [10]. Several guidelines for managing evacuation shelters have been published by both the central government [11] and local governments [12,13], and scientific articles and reports have also been produced [14,15]. The United Nations Office for Disaster Risk Reduction (UNDRR) and United Nations Development Programme (UNDP) developed guidelines for tsunami evacuation during the COVID-19 pandemic that reflected recent disasters, disaster drills, and evacuation guidelines in Japan [16]. However, it is essential to investigate and understand better how compound and interacting events could orient evacuation in the context of the future developments of COVID-19 pandemic waves. Social and economic impacts of COVID-19 on Japan Similar to other countries in the world, Japan is facing social impacts from the COVID-19 outbreak. The Japanese government decided to use the epidemic cluster-based approach because of the limited availability of testing, the impractical information of the technology-based tracing approach, and the economic system. Communities and individuals followed the government's request to adopt telework, flexible work schedules, and school closure and to reduce eating out. It should be noted that these restrictions during the stay home period under the state of emergency were not accompanied by penalties, as in other countries. This helped prevent the serious spread of COVID-19 in Japan [17] but caused severe economic impacts and induced some social problems. The government implemented economic support campaigns, such as a 100,000 yen stimulus per person and cash support for restaurants and students to help them continue their studies, in late April 2020. One source [18] reported that the 100,000 yen cash support increased consumption expenditure to almost normal levels, but it had fallen again to −10% by September 2020. Other economic impacts, such as reduced total sales from eating out and fewer numbers of tourists, have also been reported [18]. A series of "Go-To" campaigns were launched, such as "Go-To Travel," which started in July 2020 and provided discounts and subsidies of up to 50% on travel within Japan for Japanese citizens and foreign residents, and "Go To Eat," which began in October 2020 and provided a discount at restaurants. It was hoped that this campaign would support hotels, restaurants, and transportation businesses. Nevertheless, at least 713 companies declared bankruptcy (restaurants (108 cases), hotels (66 cases), apparel and miscellaneous goods stores (47 cases), and construction firms (46 cases)) [19]. In addition, transportation-related businesses were also seriously damaged. Regarding the household impact, one source [20] found that stress under the state of emergency increased anger, to a greater degree, for husbands than for wives. An increase in anger from staying indoors is thought to cause domestic violence. The stay-at-home measures prevented the spread of the infection but essentially kept victims (wives and children) trapped in situations of domestic violence. Such domestic violence may lead to divorce, and the 100,000 yen cash handout was meaningless for most victims of domestic violence, as the cash was sent to the head of the household [21]. A study [22] confirmed that this virus has led to the progression of dementia among elderly residents in care facilities, as they have been restricted from seeing their families prevent infection. Fortunately, in Japan, the rate of outbreaks in elderly care facilities has been minimal compared to that in Europe. However, the Ministry of Health announced that programs such as daycare services and elderly care must be discontinued to protect immunocompromised elderly people from infection [23]. Research objectives Based on the abovementioned background, this research has the main objective of investigating the overall impact of COVID-19 on Japanese society at the start of the 3rd wave of the outbreak from the following perspectives. 1) The COVID-19 pandemic in Japan and cultural perspectives on the outbreaks 2) COVID-19 information sources and types 3) Disaster evacuation under the COVID-19 pandemic 4) The impact of COVID-19 on daily life and other social considerations 5) Future perspectives on recovery and living with COVID-19 Preliminary results from this research will highlight recent research topics for further investigation, and more detailed consideration based on the social backgrounds of the respondents as well as disaggregated data will be performed in the future. Data and method Three study areas were selected in this research based on exposure to natural hazards. First, Miyagi and Iwate Prefectures are the two areas that were hit the hardest by the 2011 Great East Japan Earthquake and Tsunami. Fukushima Prefecture was not selected, as answers might also be influenced by the Fukushima Daiichi nuclear accident. Second, the capital area (i.e., Tokyo metropolitan area, Saitama Prefecture, Kanagawa Prefecture, and Chiba Prefecture) was selected as having less experience of recent disasters in Japan. Third, the Kyushu region was selected based on recent experience with the 2016 Kumamoto earthquake and the 2020 Kyushu floods during the COVID-19 pandemic. To quantitatively assess the current situation in Japan, a questionnaire survey was performed for speedy data collection and to control the attributes and quality of the answers. The survey was conducted by Rakuten Insight during 5-9 November 2020 among a total of 600 respondents, 200 for each study area. The ratio of male and female is 50%. The average age is 45.8 years old; is respectively 9.3% in their 20s, 22.3% in their 30s, 29.2% in their 40s, 25.7% in their 50s, and 13.5% in their 60s. Among them, we found that 48.7% graduated from university, 39.8% graduated from high school, 61.3% were employed, and the remaining 14.5% were housewives. Respondents received Rakuten points. All questions were presented in the Japanese language. The questionnaire has a total of 50 questions, including 15 questions on personal demographic information (i.e., age, gender, education level, and residence area). Some parts of the questionnaires (i.e., information, risk perception, and social impact) were based on a previously conducted questionnaire survey performed by the European Commission's Joint Research Centre (EU-JRC) and University College London (UCL) to facilitate international comparisons in the future. In the following section, the preliminary results (without consideration of detailed personal information) from three study areas will be shown and discussed based only on simple tabulation. Further detailed investigations will be performed using the same data in the future. The COVID-19 pandemic in Japan and cultural perspectives on the outbreaks The questionnaire results show that the state of emergency (23%) and the cluster on the Diamond Princess were the two main things that raised emergency awareness. The increase in positive cases in Japan (15.8%) and in the respondents' prefecture (14.2%) had a more significant impact on emergency awareness than the outbreak in China (5.3%) or the outbreak on a global scale (4.7%). Regarding the reason for the prolonged spread of the virus in Japan, the results show that international travel (tourism and business) was thought to be the primary cause (60%), and less awareness (37.3%), the infectivity/death rates of the virus (37.2%) and improper measures for restricting going out (31.0%) were other presumed causes. The situation in Japan has been much better than in Europe and the U.S. with regards to the number of new confirmed cases and the number of deaths. However, the comparison might be inaccurate due to the country's cluster-based approach and the reasons stated in section 1.3. The results show that the habit of washing hands frequently (70.2%) and a greeting culture that does not involve kisses or hugs (68.5%) were the two most common reasons for this, but another cultural practice, the removal of shoes when entering the house (34.3%), was rated as highly, as shown in Fig. 2. Other health-related reasons (the nutressentialalance of Japanese foods and the obesity rate) were also not important contributors from the Japanese respondents' point of view. Only less than 9% of respondents reported shaking hands or hugging when greeting others, which supported the previous answers. Japan has a very low mortality rate attributable to safe water, sanitation, |
and hygiene, which may be similar to other countries [24]. It was found that only 69% had a hand-washing habit before the COVID-19 outbreak, and 4.7% answered that they still had no such habit even at the time of answering the questionnaire. These results are comparable with the findings of a study in European countries where hand-wash habits range around 60-80% [25]. Japan is one of the few countries in the world in which wearing surgical masks (to protect the wearer from pollen, the common cold, or influenza or for women not wearing makeup) was common practice before the COVID-19 outbreak. Thirty-nine per cent of the respondents said that they did not mind wearing surgical masks for the whole day, 46.8% said that they had no problem wearing surgical masks for a certain period, and 14.2% said they were still uncomfortable wearing masks. The same study in India [26] reported that 89% of respondents wore masks outside. It can be interpreted that wearing surgical masks is now becoming common even in countries that, unlike Japan, had no prior mask-wearing culture. Such cultural issues can be further investigated by comparing the responses to the same questions of people in other countries to see if they genuinely influence the COVID-19 outbreak level. COVID-19 information sources and types Yahoo! News was the most popular source of information (64%), followed by NHK (42.2%) and other TV news outlets (45.5%). A freeware application for instant communication on electronic devices called "LINE" was the most popular application used in Japan, outpacing similar applications, such as "WhatsApp," that are popular in other countries. LINE news was also a popular information source (25.0%), and official surveys related to COVID-19 were also conducted via LINE. Unsurprisingly, the Internet (89%) was the most common method of obtaining information, followed by TV (66.8%) and newspapers (27.7%), as shown in Fig. 3. Although social media is a valuable source of information for investigating trends in COVID-19 perception [27] and has been identified in many studies as a potential disaster communication tool [28], social media was not the primary source of information in this sample. Regarding the quality and balance of the news and information related to COVID-19, respondents felt that there was irrelevant news for advisories and physical happiness and security for children and other vulnerable persons but sufficient information on countermeasures for stopping the spread of the virus (physical distancing, wearing a surgical mask and handwashing). Other news and information (i.e., number of new positive cases, social event restrictions, the status of daily services, and economic support) were evaluated as somewhat sufficient. The most critical information was news related to the business hours of supermarkets and pharmacies (49.2%), followed by interruptions to public transportation (34.0%) and information on protection goods (i.e., surgical masks and gloves) (31.3%). Disaster evacuation in the context of the COVID-19 pandemic Approximately 45% of the respondents had experienced a disaster (earthquakes (91.1%), typhoons (38.4%), tsunamis (19.2%), and floods (13.7%)) before the COVID-19 outbreak. Of these, 49.8% decided to evacuate, and evacuation on foot was the most common (65.2%), followed by evacuation via a personal vehicle (27.4%). Their evacuation goals were schools (34.1%), other (31.1%), community centers (11.9%), and private facilities (11.1%). Regardless of past disaster experience, 63.8% reported that they would evacuate on foot and 30.3% by private car if a disaster occurred during the COVID-19 pandemic. These numbers are not significantly different from those reported for disasters that occurred before the COVID-19 outbreak. In addition, schools (29.7%), community centers (12.8%), and private facilities (9.0%) were still preferred evacuation destinations, but 51.7% of respondents said they would "stay at home evacuation," as shown in Fig. 4. These results indicate that we must ensure that residents know whether they are truly safe from each type of disaster and understand that evacuation to save one's life has a higher priority than staying at home if one is at risk during a disaster. The preparation of infection prevention equipment (63.3%) was the most significant concern when deciding on the evacuation method and goal, followed by the shelter space (58.2%), the availability of private space at the shelter (49.5%), and the safety of the shelter (46.3%). In other words, it is essential to test the information needed for scenarios of concurrent hazards to address the typical organizational and behavioral vulnerabilities acting as escalating factors during a complex crisis. Only 37.8% of respondents thought that past disaster experiences helped them better prepare for a disaster during COVID-19. Based on Table 1, the same trend can be seen in the capital area, Kyushu region, and Iwate Prefecture. Only respondents from Miyagi Prefecture believed that their past disaster experience was helpful (52% for men and 60% for women). Although Miyagi and Iwate Prefectures were hardly hit by the 2011 Great East Japan Earthquake and Tsunami, perceptions from these two prefectures differed. In addition, the first confirmed positive case in Iwate Prefecture occurred as late as 29 July 2020. Therefore, the impact of disaster experience should be investigated in more detail in future studies. Similarly, there was no significant difference in preparation for floods/electric blackouts or preparation for other emergencies based on the experience of a state of emergency. Therefore, there must be other influencing factors, in addition to the Table 1 Answers to the question regarding whether past disasters are useful for disaster response during the COVID-19 pandemic. fact that the characteristics of natural hazards and infectious diseases are different. The impact of COVID-19 on daily life and other social and economic considerations Levels of satisfaction with work, mental health, and the economy decreased, but satisfaction with family increased (12% good and very good), likely because people tended to stay at home more. One percent of the respondents experienced domestic violence during the state of emergency, and most of them (85.7%) could not seek help. During the state of emergency, as expected, the usage of food and utilities (water, electricity, and gas) and communication technology (phone and the Internet) increased, and the usage of public transportation (45.5%), private transportation (41.5%), medical services (35.5%) and financial services (20.7%) decreased. Respondents were mainly concerned about the impact on their health (79.7%), the situation of medical facilities (41.5%), and the difficulty of maintaining physical distance, as shown in Fig. 5. Future perspectives on recovery and living with COVID-19 At present, the most significant concern is the next wave of infection (52.5%), which has since occurred, as the 3rd wave emerged in the middle of November. If the state of emergency has to be reinstated, 34% of the respondents said they could accept it for a few months, whereas 36.2% could accept it for only a few weeks. Most preferred resuming activities step by step with COVID-19 countermeasures (44.2%) rather than waiting until the end of the outbreak (26.0%) and resuming activities as early as possible, even under the risk of disease spread (16.2%). Most respondents thought that recovery priority should be given to education and working conditions (43.8%), access to economic support systems (35.5%), and access to social/cultural/entertainment facilities (34.5%). Especially for social and economic support systems, the respondents believed that support for basic needs (i.e., goods and other utilities) should be prioritized (74.8%), followed by support for lowincome persons (52.7%) and support for persons who own their own business (39.5%), as shown in Fig. 6. Conclusions and recommendations A questionnaire survey of 600 persons in Japan was performed via the Internet to quantitatively assess the situation just before the 3rd wave of the COVID-19 outbreak in Japan. Several main conclusions and recommendations for future research can be proposed based on preliminary results obtained from a simple tabulation. − New cases from the Diamond Princess cruise ship and the state of emergency were the two factors that most increased awareness of COVID-19 in Japan. − Approximately 70% of Japanese regularly washed their hands before the COVID-19 pandemic, and less than 9% of respondents shake hands or hug as a form of greeting. This issue is worth clarifying as a cultural reason for the lower level of infection in Japan compared to those in countries in Europe and the U.S. − The Internet and TV were the two primary sources of information. Surprisingly, social media was not a significant information source. The most important information was news related to the operation of supermarkets and pharmacies during the state of emergency. − Approximately 50% of respondents said they would stay at home if the next disaster occurred during the COVID-19 pandemic. Thus, the government must ensure that residents know whether they are truly safe from each type of disaster and understand that evacuation to safety has a higher priority than staying at home (if they are at risk during a disaster). Further work is urgently needed to understand scenarios of concurrent hazards, identifying the common behavioral vulnerabilities, and prioritizing the information needed in case of emergency. − Only approximately 40% thought that the experience of past disasters helped them better prepare for disasters during the COVID-19 Fig. 5. The issues of most significant concern during the peak of COVID-19. pandemic. Surprisingly, Iwate Prefecture had no confirmed cases for many months, but 70% of respondents from Iwate believed that past disaster experience would not be helpful. However, approximately 50-60% of people in Miyagi did believe that past disaster experience would be beneficial. Future research could focus on the difference between the two prefectures and explore the reason for this difference and the months-long lack of confirmed cases in Iwate Prefecture. − Levels of satisfaction with work, mental health, and the economy decreased, but satisfaction with family increased, probably because of a greater tendency to stay at home. − The use of many facilities, including hospitals, declined. It is therefore vital to keep monitoring long-term heath situations, as the finding of the number of infected cases could be delayed because of medical check avoidance. − If a state of emergency has to be issued again, 34% of respondents said they could handle it for a few months, and another 36% said they could handle it for a few weeks. − Recovery priority should give to education and working conditions, social/cultural/entertainment facilities, and social and economic support systems. CRediT authorship contribution statement Declaration of Competing Interest The authors would like to declare that this research has no competing Interest. Urogenital Schistosomiasis and Intestinal Parasitosis Coinfection among School Age Children in Adim Community, Nigeria The prevalence of urogenital schistosomiasis, intestinal parasitosis and their co-infection was carried out among the school age children in Adim Community from August to November, 2015. Urine and stool samples were collected from each of the subjects selected by simple random sampling method and processed using standard bacteriological and parasitological techniques. Of the 200 subjects examined, 42(21%) were infected with Schistosoma haematobium, 88(44%) with intestinal parasites and 21(10.5%) were co-infected. Subjects aged 5-10 years had the highest prevalence of infection (30%) with S. haematobium, while subjects’ aged16-20 years had the highest prevalence of infection (80%) with intestinal parasites. The difference was statistically significant (p=0.001). Males recorded the highest prevalence of infection (30%) with S. haematobium, for intestinal parasites (50%) and for coinfection (15%) while females had (17.1%), (41.4%) and (8.5%) respectively and the difference was statistically significant (p=0.001). Hookworm (45.5%) had the highest frequency among the helminthes while Entamoeba histolytica (4.6%) was the only protozoan detected. This work confirmed a high prevalence of urogenital schistosomiasis, intestinal parasitosis and their co-infection among the school age children in Adim community. Introduction Intestinal parasitic diseases and urogenital schistosomiasis keep on constituting a noteworthy general wellbeing and formative test particularly among the school-matured youngsters in Nigeria. These contaminations have been represented as infections of neediness and underdevelopment since they have been connected to the absence of satisfactory unhygienic condition, provision of safe water and of uncalled for individual cleanliness (WHO, 2002a). These parasitic sicknesses deny the destitute individuals of good wellbeing, adding to financial insecurity and social minimization; and the needy individuals of immature countries encounter a cycle where under nourishment and rehashed contaminations prompt to a high bleakness from era to generation. These parasitic ailments deny the needy individuals of good wellbeing, adding to monetary flimsiness and social underestimation; and the needy individuals of immature countries encounter a cycle where under sustenance and rehashed diseases prompt to a high grimness |
from era to era (Mehraj et al., 2008). All around, two billion people are tainted with helminths, out of the greater parts living in asset poor settings (Noyer and Brandt, 1999;WHO, 2002). The event of helminthic diseases is related with financial, ecological and different variables like numbness of straightforward wellbeing advancing components and congestion, restricted access to clean water, tropical atmosphere and low elevation (WHO, 2002b). School-aged children are one of the groups at high risk for intestinal parasitic infections and these have detrimental effects on the survival, appetite, growth and physical fitness, school attendance and cognitive performance of school age children (Nokes & Bundy, 1993;Stephenson et al., 2000;de Silva et al., 1997;Hadidjaja et al., 1998). This study was at determining the prevalence of urogenital schistosomiasis and intestinal parasitosis and coinfection among the subjects in Adim community in Cross River State, Nigeria. MATERIALS AND METHODS Study Area This review was directed in Adim Community situated in Biase L. G. A. in C. R. State, Nigeria. Adim is an average rustic group found 110 kilometers toward the North of Calabar, the Cross River State capital. There is no pipe borne water in the group and the occupants depend for the most part on three new water streams to be specific Ibeteuroma, Egboga and Ogamenah for their household, monetary and recreational exercises (Ejezie et al., 1991). The area is situated in the tropical rain forest belt with an average annual rainfall of 1500-2000 millimeters. The main occupations of the residents of this community include hunting, fishing and farming. The principle crops grown by the farmers are rice (Oryza sativa), yam (Discorea prachincilis) and cassava (Manihot utilissima). Also there are 3 nursery schools, 2 primary schools, 1 secondary school and 1 health center in the community. Subjects and Ethical consideration The target population was children between 5 and 20 years. The consent and ethical approval were sought and obtained from the school authority and the ethical committee of the Cross River State Ministry of Health. The Onun of Adim (village head) was briefed on the significance of the study and the level of involvement of the communities before its commencement. The procedures, significant benefits, and the harmless nature of the study were also explained to all the people in the community. Study design The sequence of activities were followed: visiting the village head of Adim and the headmasters of the schools to brief them about the study, informing consent parents of pupils, administrating questionnaire, collecting and processing the urine and stool samples for parasitological survey. Collection of Urine Samples Two clean universal containers for collection of urine sample were issued to each subject selected for the study. The samples were collected between 10.00 am and 12.00 pm when maximum egg excretion occurs in the urinary bladder (Chem and Mott, 1989). The people were allowed to do a little exercise in order to get a maximum egg shield. Each bottle was labeled according to the code number of the subjects. Same number of subjects also submitted their stool samples. Examination of Urine for Haematuria and Proteinuria The appearance of the urine was observed and recorded as whether clear, cloudy, presence or absence of visible blood were observed and recorded accordingly. Heamaturia was detected soon after collection of urine sample using dipstick (Ames: Bayer Diagnostic Brussels, Belgium) (Inyang-Etoh et al., 2010). It was carefully dipped into the bottle containing the urine for 5 seconds. The resulting change in color of the strip was compared with the manufactures color chart to estimate the amount of blood in the urine. The same method was also used for the examination of proteinuria. Haematuria was reported as 5-10 ery/µl (+), 50ery/µl (++), 250ery/µl (+++). Proteinuria was reported as 10mg protein/dl indicating trace proteinuria 30mg/dl(+), 100mg/dl(++), and 500mg/dl(+++) (Inyang-Etoh et al., 2010). Examination of Urine for Schistosoma haematobium Ova An aliquot of 10ml of the urine was moved into a widespread compartment holding 5ml of 1% fluid arrangement of carbol fuchsin for recoloring and conservation of ova (Ejezie, et al., 1991). The examples were protected along these lines until the season of filtration. The adjusted filtration framework for the recognition of ova of Schistosoma haematobium was received for the review (Useh and Ejezie, 1999). Diagnosis for intestinal parasites. A. Diagnosis by macroscopy The stool was examined for the appearance (colour), consistency, presence of blood, mucus and segments of adult worms. B. Diagnosis by direct wet mount microscopic examinations were performed for detection of intestinal parasites; A drop of saline was placed on one end of a clean grease free slide while a drop of iodine was placed on the other end leaving enough space at the edges to handle the slide. A small amount of faeces was then placed on the microscope slide using an applicator stick and mixed in the drop of saline and iodine respectively using the edge of a coverslip then covered with the coverslip. Finally the samples were examined microscopically using x10 and x40 objective lens with the iris condenser sufficiently closed to give good contrast. Careful attention was paid during the identification of the ova and larvae of intestinal parasites in the stool samples. C. Diagnosis by formol ether concentration technique. Procedure; Ten milliliters of 10% formalin was added to approximately 1g of stool in a universal container and stirred using an applicator stick until the suspension was slightly cloudy. A funnel with a gauze filter inside was placed into a centrifuge tube and the faecal suspension passed through the filter until 7ml were accumulated. The filter was then removed and discarded with the lumpy residues. Thereafter 3ml of ether was added to the filtrate and mixed well for one minute then centrifuged at 2000rpm for 2 minutes. The fatty plug (debris) was loosened using an applicator stick and the supernatant discarded by quickly inverting the centrifuge tube. The sediment was resuspended by tapping and a drop transferred to a clean grease free slide, covered with a coverslip and examined microscopically using x10 and x40 objective lens with the iris condenser sufficiently closed for identification of ova, cysts and larvae of intestinal parasites. Data Analysis Statistical analysis was carried out using SPPS version 16.0 for windows and Microsoft Excel Tool Pak. The Chi-squared (X 2 ) test was used to test for the difference in prevalence of infection according to the age and gender. ANOVA was used to compare more than two groups. The probability level at 0.05 was used. Table 1 shows the prevalence of urogenital schistosomiasis and intestinal parasites according to the age group of subjects examined. Children aged 5-10 years had the highest prevalence of S. haematobium infection (30%) while subjects aged 16-20 years recorded the least infection rate (0%) and the difference was statistically significant (X 2 = 31.2, (df)2, p=0.001). For the infection by intestinal parasites, subjects aged 16-20 years recorded the highest prevalence (80%) while subjects aged 11-15years had the lowest infection rate (41.3%). Of the 200 samples examined, 21(10.5%) subjects were coinfected with urogenital schistosomiasis and intestinal parasites. There was no statistically significant difference for age groups among the three diferent types of infections (F=0.459, df(259) p=0.632). The prevalence of infection according to the gender is shown in table 2. Higher prevalence was recorded in males for both the infections of S. haematobium (30%), intestinal parasites (50%) and for both infections (15%). Significant difference was detected in the prevalence of S. haematobium infection between the genders of the subjects (X 2 =3.841, (df)2, p=0.001) but no statistically significant difference was found in the prevalence of the three types of infections between the genders (F=0.186, df(259) p=0.666). Table 3 shows the distribution of parasites detected according to educational status of subject's parents. Subjects whose parents had Primary as their educational level had the highest prevalence rate of infection with S. haematobiium (29%), intestinal parasite (59%) and coinfection (14%) while those with tertiary level of education had the lowest levels of these infections (5%), (10%) and (5%) respectively. There was no statistically significant difference in the prevalence of the three types of infections according to level of education (F=1.793, df(259) p=0.169). Figure 1 shows the distribution of intestinal parasites detected. Among the helminthes, hookworm infection had the highest prevalence rate (45.5%) while Trichuris trichiura had the lowest (11.4%). Among the intestinal protozoan parasite only Entamoeba histolytica (4.6%) was detected. In this study, female subjects were more infected than the males with a prevalence rate of 30% against 17.1% in males. The reason for the higher prevalence among the female is presumably due to higher water contact activities particularly in washing and bathing in cercariae-infested rivers. This is because in this part of the world females are more involved in domestic activities than their male counterpart. The age group 5-10 years had the highest prevalence of infection with S. haematobium (30%) while subjects aged 16-20 years recorded the highest prevalence of infection with intestinal parasites (80%) and coinfection (80%). This may probably be due to the fact that children in this age bracket are likely to ignore the awareness created about the disease while they continue with their water contact activities. (Uneke et al., 2007). This could be due to high level of soil contact activity and low personal hygiene. However, the age dependent patterns of infection prevalence are generally similar among the major helminthes species, exhibiting a rise in childhood to a relatively stable in adulthood (Naish et al., 2004). RESULTS This study further showed that four intestinal parasites were detected including Ascaris lumbricoides, Trichuris trichiura, Hookworm and Entamoeba histolytica. However, Hookworm and Ascaris lumbricoides occurred more frequently than the protozoan parasites. The protozoan parasite; Entamoeba histolytica was seen in two subjects only. The transmission of Ascaris/Trichuris infections are known generally to occur more in rural areas than in urban or semi-urban areas however, in urban slums, the transmission is probably related to poor sanitary conditions or contaminated water supplies and perhaps Trichiuris and other soil-transmitted helminthes cannot successfully complete their life cycle in the absence of a more soil rich rural environment and may be less adapted to conditions in urban or semi-urban areas for successful transmission (Mehraj et al., 2008). Conclusion: It can be concluded that there has been a slight reduction in the prevalence of Schistosoma haematobium in the area and that children aged 5-10 years had the highest prevalence rate of the infection and female recorded the highest. From the result, children aged 15-20 years were co-infected with intestinal parasites and urogenital schistosomiasis and female still recorded the highest. The prevalence of co-infections has not shown to be gender related but more research works are required to elucidate this. It is hoped that the findings of the investigation will contribute to effective disease control planning and implementation in the area of study and others in Nigeria and other tropical countries with similar heavy burdens of infectious disease. Conflict of interest: The authors declare that there is no conflict of interest regarding the publication of this work. Fatigue in Systemic Lupus Erythematosus: An Update on Its Impact, Determinants and Therapeutic Management Fatigue is a complex and multifactorial phenomenon which is often neglected by clinicians. The aim of this review was to analyze the impact, determinants and management of fatigue in patients with Systemic Lupus Erythematosus (SLE). Fatigue is one of the most prevalent symptoms in SLE, reported by 67% to 90% of patients. It is also described as the most bothersome symptom, considering that it may impair key aspects of health-related quality of life, while also leading to employment disability. It is a multifactorial phenomenon involving psychological factors, pain, lifestyle factors such as reduced physical activity, whereas the contribution of disease activity remains controversial. The management of fatigue in patients with SLE should rely upon a person-centered approach, with targeted interventions. Some pharmacological treatments used to control disease activity have demonstrated beneficial effects upon fatigue and non-pharmacological therapies such as psychological interventions, pain reduction and lifestyle changes, and each of these should be incorporated into fatigue management in SLE. Introduction Fatigue is a universal symptom experienced by nearly everyone in the general population. However, we lack a consensual definition of fatigue. Fatigue can be described as a subjective unpleasant sensation of exhaustion with physical and mental components, which interferes with individuals' ability to function at |
their normal capacity. Fatigue impairs quality of life, and may lead to irritability, inability to concentrate, and poor motivation [1,2]. In chronic conditions such as Systemic Lupus Erythematosus (SLE), but also in other autoimmune diseases such as Sjögren's syndrome or systemic sclerosis, the experience of fatigue seems to differ from 'everyday tiredness', as being more frequent, unpredictable and typically unresolved by rest [3]. This symptom remains a complex, multidimensional and poorly understood concept, often neglected by clinicians who prefer to focus on objective manifestations. The aim of this review was to report upon the impact, determinants and management of fatigue in patients with SLE. The Most Frequent and Disabling Symptom Fatigue is recognized as one of the most prevalent symptoms in SLE, reported by 67% to 90% of patients, depending on the series [4]. In a 2020 survey analyzing the burden of SLE from the patients' perspective in European countries [4], fatigue was described as the most common symptom (affecting 85.3% of the 4375 respondents) and the most bothersome symptom, which is consistent with previous studies. Fatigue is reported as severe in intensity in more than a third of SLE patients [4]. Fatigue may impair several key aspects of the patient's quality of life, with repercussions on both physical and mental health. Indeed, SLE patients report that fatigue has a negative impact on emotions, cognition, work, activities of daily living, leisure activities, social activities and family activities. They describe physical impairment, with walking and exercising difficulties. Emotional consequences of fatigue, such as frustrations and stress due to being unable to accomplish tasks, sadness or loss of motivation, are also common [5][6][7][8][9]. Fatigue in SLE has a significantly negative impact on work ability and work productivity, as it can lead to limitations in workplace activities by affecting endurance, mobility, concentration, or interactions with employees and coworkers. Fatigue in SLE is also associated with a higher risk of absenteeism and unemployment [10]. Altogether, fatigue is an important determinant in the perception of SLE impact upon patients' daily living, even for those in remission. A Multifactorial Manifestation Altogether, fatigue is a highly multifactorial manifestation (Figure 1), caused by a complex interplay between disease itself, psychosocial, behavioral and personal variables. A recent study from our group described 3 main clusters of fatigue in SLE patients: (1) the most frequent profile (67.5% of the patients) was represented by patients with moderate fatigue, low disease activity and low anxiety and depression; (2) a quarter of the patients had very high fatigue, high depression and anxiety but low disease activity; and (3) less than 10% of the patients had high levels of fatigue, with high disease activity, low anxiety and no depression [11]. This suggests that the mental health status is an important predictor of fatigue in SLE, that disease activity plays a weaker role in SLE fatigue, and that, most of the time, other factors contribute to fatigue in SLE. Lupus-Related Determinants The association between fatigue and disease activity in SLE has been widely studied and debated for a long time, with controversial results [11][12][13][14][15][16][17][18][19]. Type I interferons, which are key cytokines in SLE, are associated with fatigue and may provide a clue towards a pathogenic explanation for fatigue in SLE. Disease activity seems to play a role in the genesis of fatigue but it cannot fully explain fatigue by itself. Indeed, a study based on an inception cohort of adult patients with SLE found that fatigue and disease activity followed distinct trajectories over 10 years [20]. Additionally, in our recent FATILUP studies [21], the association between fatigue and disease activity was significant, but weak (OR: 1.05 (95% CI: 1.00-1.12) per 1 point increase in SELENA-SLEDAI score). Therefore, it is likely that disease activity has a complex and potentially indirect contribution to fatigue, such as by influencing other major determinants of fatigue, for example pain and psychological factors. Some specific organ involvements such as neurological impairment and painful manifestations such as arthritis or oral ulcers have been found to be associated with fatigue in some studies [14,21,22]. Pain has been reported to have a specific role in SLE fatigue, and chronic pain treatment is essential to the management of fatigue in SLE [12,14,17,23]. Organ damage, especially renal or cardiac failure, can also be important causes of fatigue in SLE patients [24][25][26][27]. Furthermore, the use of glucocorticoid has been shown to be independently associated with fatigue in SLE [21]. Psychological Determinants Mental health status, emotional and functional wellbeing have shown to be major determinants of fatigue in SLE patients. Depression and anxiety appear to be among the strongest predictors of fatigue in SLE patients [11][12][13]21,23,28,29]. There is a clear association between fatigue and depression in general, and scales assessing depression often include fatigue-related items. Depression affects both physical and mental dimensions of fatigue in SLE. Additionally, depression is frequent in SLE patients (between 17 and 75% of patients), and some authors have mentioned that SLE contributes to depression through its neurological involvement, an autoimmune effect, and the emotional consequences of pain and disability [30][31][32]. Stress, which is a subjective negative perception of life events, which may be influenced by sociological and psychological factors and SLE burden, seems to mediate the relationship between depression and fatigue over time in SLE patients. Decline in stress has been associated with a meaningful improvement in fatigue in SLE [31]. Sleep disorders have also been shown to be common and significant predictors of fatigue, occurring in more than half of SLE patients [28,29,33,34]. SLE may contribute to sleep disorders because of pain and inflammation, and steroid use has been associated with sleep disorders [35]. In addition, helplessness (a state in which a person remains passive in negative situations), coping disability (difficulties in facing problems in an adequate manner) and abnormal illness-related behavior have been associated, although not independently, with fatigue in SLE in some series [14,15,19]. The role of psychological determinants is therefore major in SLE fatigue. Consequently, it is crucial to suggest a thorough psychological assessment of SLE patients reporting severe fatigue, especially for those with no or low disease activity, since mood disorders are frequent in patients with SLE [4,21,[30][31][32] and multifactorial. Comorbidities Fibromyalgia is a major predictor of fatigue in SLE [14,18,36]. In a study conducted by Touma et al., trajectories with higher fatigue scores were associated with a higher prevalence of fibromyalgia [20]. Fibromyalgia is common in SLE patients (from 6.2% to 30% of patients) but may be underdiagnosed by physicians [21,37]. Consequently, the role of fibromyalgia should be considered in SLE patients who complain about fatigue and widespread pain. Other frequent comorbidities such as anemia, hypothyroidism, or adrenal failure are risk factors in fatigue. Vitamin D insufficiency was associated with fatigue in SLE in some but not all studies [38,39]. SLE patients have a high risk of vitamin D deficiency because of photoprotection as well as in case of renal failure. Behavioral and Socio-Demographic Features Reduced levels of physical activity and aerobic capacity significantly increase fatigue in SLE. SLE patients have many actual and perceived barriers to exercise. It has been shown that, compared to sedentary controls, SLE patients have reduced levels of aerobic fitness, reduced exercise capacity and reduced muscle strength, which further leads to a reduced ability to perform physical activity. Furthermore, SLE patients are limited by arthralgia, anemia, and other SLE organ involvements. For all of those reasons, SLE patients often have limited physical activity and assume a sedentary lifestyle [40][41][42]. Obesity and smoking are other potential behavioral determinants of fatigue in this population [14,43]. The role of sociodemographic features is contradictory, but some studies found higher levels of fatigue in SLE patients with low annual income, low education level, or difficulty in accessing health care [14,15]. In some studies, a low level of perceived social support was also associated with fatigue [12]. Interventions to Improve Fatigue Recently, an increasing number of interventional studies focused on fatigue in SLE, and some pharmacologic and non-pharmacologic therapies have demonstrated beneficial effects on fatigue. Improving disease activity is associated with significant reduction in fatigue in randomized controlled trials of belimumab, blisibimod, and hydroxychloroquine [44][45][46]. This effect is likely to be observed with any treatment improving disease activity in SLE, although this has not been formally proven. N-acetyl-cysteine has also been shown to improve fatigue in SLE. A double-blind, placebo controlled, randomized trial found that a 2.4 g/day dose of N-acetyl-cysteine is effective for reducing fatigue and improving disease activity, and is safe and well-tolerated [47]. Vitamin D supplementation also seems to have positive effects on fatigue in SLE patients. An observational study found significantly lower fatigue scores after vitamin D supplementation in 80 SLE patients, and a randomized double-blind placebo-controlled trial showed a decrease in fatigue in juvenile-onset SLE patient receiving vitamin D supplementation [48,49]. Physical activities such as supervised training, home training, and appropriately prescribed graded aerobic exercise, have been associated with favorable improvements in patient-reported fatigue in different studies. Importantly, exercise was reported to be safe and well tolerated, with rare adverse effects, and no reported deleterious effects on disease activity or inflammation [50][51][52]. Physical activity should therefore be generally recommended for the management of fatigue in SLE patients, especially since it also leads to less pain interference, better physical function, cardiovascular risk reduction, and even positive impact on anxiety and depression. A trial conducted by Davies et al. indicates that a low glycemic index diet and a low-calorie diet were both associated with reduction in fatigue in SLE, indicating the role of weight loss in the improvement in fatigue [53]. Different psychosocial interventions have been associated with significant improvement in fatigue in SLE: cognitive behavioral therapy, psychoeducation, psychotherapy, relaxation and self-management. Those interventions focus on coping ability improvement, cognitive restructuring and perceived social supports [52,[54][55][56]. Even if the effect in reducing fatigue has been shown to be weak in most of these studies, such interventions can decrease psychological distress and pain and therefore might be integrated into the general management of SLE patients. Some interventions targeting pain have also shown their ability to improve fatigue in patients with SLE. A randomized trial found a significant decrease in fatigue in SLE patients receiving transcutaneous auricular vagus nerve stimulation [57]. Additionally, a randomized controlled trial indicates benefits of acupuncture in reducing fatigue in patients with SLE [58]. The Need for a Personalized Management At this time, there is no validated recommendation for the management of fatigue in SLE. Since fatigue may be influenced by a variety of factors and because of the diverse profiles of fatigue in SLE, the management of fatigue should rely upon an individualized person-centered approach (Figure 2). Women with SLE have reported the need for fatigue acknowledgement by clinicians, as well as conversations about fatigue, with information about coping strategies [59]. Fatigue management in SLE would start with an assessment of the intensity and the characteristics of fatigue using validated scales, enabling an individual follow-up of fatigue over time. In recent years, there has been an increasing interest in using Patient Reported Outcomes (PROs), because they place the patients at the center of their health management and help to establish a trusting physician-patient relationship. The most commonly PROs used to evaluate fatigue in SLE are the Fatigue Severity Scale (FSS), the FACIT-fatigue score, which we use in clinical practice, the Fatigue-VAS, which are unidimensional scales measuring fatigue intensity, and the Multi-dimensional Fatigue Inventory (MFI), which analyze general fatigue, physical and mental components of fatigue as well as the reduction in activities and motivation [60]. A personalized investigation of fatigue predictors is needed, with evaluation of disease activity, search for intricate causes (major organ damage, chronic pain, anemia . . . ) and psychosocial factors, assessment of life habits (physical activity, quality of sleep, smoking, obesity . . . ). Common medical causes of fatigue, such as pregnancy, infections, metabolic diseases or drug-induced fatigue must not be forgotten. Finally, optimal management of fatigue for patients with SLE should be based on providing targeted interventions, according to the patient profile [61]. Conclusions According to patients, fatigue is the most common and disabling symptom in SLE, and this may impair patients' physical and mental health and reduce patients' quality of life by impacting upon their emotions, work, and daily life activities. Fatigue must therefore be adequately assessed and managed in SLE. It is |
a complex and multifactorial phenomenon, possessing many patterns. Psychological factors seem to be the most important fatigue predictors in SLE patients. Pain and fibromyalgia are also major fatigue determinants, along with lifestyle, especially reduced physical activity. Disease activity seems to have a complex contribution to fatigue, and its role remains controversial. Consequently, the management of fatigue in patients with SLE should rely upon a person-centered approach, with a personalized assessment, and targeted interventions. Some pharmacological treatments used to control disease activity, such as Belimumab, have demonstrated beneficial effects on fatigue. Non-pharmacological therapies, such as psychological interventions, pain reduction and lifestyle changes should be integrated into fatigue management in SLE. In recent years, the scientific community seems to have increased their understanding of the importance of fatigue management in SLE, and we can hope for a better understanding and treatment of fatigue in patients with SLE in the future. Funding: This study was supported by the EU-funded (ERDF) project INTERREG IV and V Rhin and the Centre National de Référence des Maladies Autoimmunes Systémiques Rares Est Sud-Ouest (RESO). A Mutant of SWAP-70, a Phosphatidylinositoltrisphosphate Binding Protein, Transforms Mouse Embryo Fibroblasts, Which Is Inhibited by Sanguinarine SWAP-70, a phosphatidylinositol trisphosphate (PtdIns(3,4,5)P3) binding protein, has been suggested to be involved in transformation of mouse embryo fibroblasts (MEFs) as well as membrane ruffling after growth factor stimulation of the cells. A mutant, SWAP-70-374, was found to be able to bind to F-actin in vitro, whereas wild-type SWAP-70 failed to do so. This mutant was present at the plasma membrane without any stimulation while the wild-type protein was present only in the cytosol unless cells were stimulated with EGF. Expression of this mutant in MEFs resulted in morphologic transformation, fast growth, and loss of contact inhibition, suggesting that SWAP-70 with this mutation can transform the cells. ERK1/2 was activated in SWAP-70-374-transformed cells. Use of MEK inhibitors revealed that the ERK1/2 pathway does not affect the cell growth of MEFs but is responsible for loss of contact inhibition. To investigate the function of SWAP-70 further, drugs that can inhibit SWAP-70-dependent cell responses were screened. Among various drugs, sanguinarine was found to inhibit transformation of MEFs by SWAP-70-374. This drug was able to inhibit SWAP-70-mediated membrane ruffling as well, suggesting that its effect was closely related to the SWAP-70 signaling pathway. These results suggest that SWAP-70-374 can activate some signaling pathways, including the ERK1/2 pathway, to transform MEFs. Introduction SWAP-70 is a phosphatidylinositol trisphosphate (PtdIns (3,4,5) P 3 ) binding protein, that has been implicated to play a role in the formation of cancer. Strong expression of SWAP-70 is often seen in human B-cell neoplasms [1]. SWAP-70 has been also shown to be expressed in higher levels in malignant gliomas than in lowgrade ones or normal brain tissue [2]. These results suggest that SWAP-70 may be closely related to formation of malignant tumors in vivo. Involvement of SWAP-70 in transformation of the cells has also been suggested in vitro. We have shown that SWAP-70 is required for the anchorage-independent growth of v-Src-transformed MEFs [3]. In addition, growth of the cells lacking SWAP-70 has been shown to be slower than that of the cells expressing SWAP-70 [3]. These results suggest that SWAP-70 may be involved in regulation of cell growth in some way. It has been suggested that SWAP-70 is important for cell motility and invasion of the tumor cells [2,4]. Although how SWAP-70 contributes to the formation of cancers is largely unknown, some of SWAP-70's activities at the biochemical or cell biological level have been revealed. SWAP-70 contains a pleckstrin homology (PH) domain, which is responsible for PtdIns(3,4,5)P 3 binding, in the central part and a coiled-coil domain in the carboxyl-terminal half [5,6]. In addition, carboxyl-terminal region of SWAP-70 has been shown to bind to non-muscle F-actin in vitro [7]. One of the well-studied cell responses related to actin rearrangement is membrane ruffling, which is closely related to cell motility. SWAP-70 translocates from the cytoplasm to the membrane ruffles upon PtdIns 3-kinase activation after growth factor stimulation and co-localizes with Factin in adherent cells such as MEF or Cos7. Cells lacking SWAP-70 show impaired membrane ruffling after growth factor stimulation, suggesting that SWAP-70 may play a crucial role in induction of membrane ruffling [5]. SWAP-70 lacking the F-actin binding domain has been shown to act as a dominant negative reagent for membrane ruffling, suggesting that this actin-binding activity is important for membrane ruffling [7]. Binding of SWAP-70 to activated Rac1, which has been shown to regulate actin rearrangement including membrane ruffling, has been also detected [7]. Taken together with the fact that SWAP-70 binds to PtdIns(3,4,5)P 3 , a product of PtdIns 3-kinase, that has been also suggested to be essential for membrane ruffling, it is likely that SWAP-70 is an important molecule that may put the functions of PtdIns(3,4,5)P 3 , F-actin, and Rac1 together. Supporting these findings, SWAP-70 has been shown to be essential for proper homing of B cells to lymphoid organs, which may require F-actin rearrangement [8]. Because F-actin rearrangement is likely to be related to cell transformation, these findings support the idea that SWAP-70 contributes to tumor formation in some way. Sanguinarine, a benzophenanthridine alkaloid, has been shown to exhibit anti-cancer activity in vivo and in vitro [9,10,11,12,13,14,15]. For instance, sanguinarine exhibits antiproliferative and antiangiogenic effects in melanoma and prevention activity of occurrence of skin cancers. There are also a number of reports suggesting that sanguinarine inhibits growth of tumor cell lines and induces apoptosis. Recently, it has been suggested that sanguinarine interacts with DNA and histones, which might be the mechanism for its anti-tumor activity [16]. However, this does not completely explain the fact that sanguinarine is effective only for certain tumor cell lines. In this paper, we demonstrate that a mutant of SWAP-70 can transform mouse embryo fibroblast and further suggest that an anti-cancer drug, sanguinarine inhibits SWAP-70-dependent cell responses. Cells and culture conditions Mouse embryo fibroblasts (MEFs) were cultured from a 129/ SvEMS strain in Dulbecco's modified minimal essential medium (DMEM) supplemented with 10% fetal bovine serum. The culture was maintained carefully and established as an immortalized cell line: this was named as MEF clone 18. However, MEFs are usually mixtures of cells derived from various origins: thus cells can give various phenotypical backgrounds. For this reason, when cell lines expressing some gene are produced, each line could have a different background. To deal with this problem, cells were isolated by limiting dilution method and grown from single cells. One of these cells, 18-2, was used in this study [3]. In this way, phenotypic background should be identical among the clones. 70-5 is a MEF cell line that expresses wild-type SWAP-70 [3]. Antibodies and Western blotting Anti-SWAP-70 antibody was raised against bacterially expressed full-length human SWAP-70 fused to glutathione-S transferase as described before [3]. This antibody does not recognize mouse SWAP-70 but does recognize human SWAP-70. Anti-ERK1/2 and anti-phospho-ERK1/2 antibodies were purchased from Cell Signaling Technology Inc. (Danvers, MA). Antib-actin antibody was from Sigma-Aldrich (St. Louis, MO). Western blotting was done as described before [3]. Briefly, blocking was done with 1% skim milk, and the protein was visualized by the ECL system. Determination of cell growth Cells were plated on the 6-cm dishes at a density of 1610 5 cells per dish and maintained with medium change every other day. The numbers of the cells per dish were counted. For examination of cell survival, cells were plated at a density of 2610 5 cells per 6cm dish and maintained without medium change. The cell numbers shown in the figures are the means of three independent experiments. Sanguinarine was purchased from Extrasynthese (Genay Cedex, France), PD98059 and U0126 were from Wako Co Ltd. F-actin cosedimentation assay Binding of SWAP-70 WT/374 to F-actin was tested as described before [7]. Briefly, human non-muscle actin (Cytoskeleton Inc., Denver, CO) was polymerized in F-buffer (50 mM KCl, 2 mM MgCl 2 , 0.2 mM ATP, and 0.2 mM DTT in 2 mM imidazole, pH 7.1) at room temperature for 30 min and incubated with purified recombinant wild-type His-SWAP-70 or His-SWAP-70-374 for 1 hr at room temperature. After ultracentrifugation at 109,0006g for 30 min at 25uC, the proteins in the supernatant or the pellet were analyzed by SDS-PAGE, which was followed by Coomassie Blue staining. The amount of SWAP-70 in each lane was quantified by densitometric analysis using ImageJ 1.43 u (National Institutes of Health, USA). The ratio of amount of SWAP-70 in the pellet to that in the supernatant was calculated in each sample. Then the value of the sample with F-actin was normalized to that of the sample without F-actin. The results show the relative amount of SWAP-70 specifically bound to F-actin. The numerical formula to obtain the results is: (Relative affinity) = {(amount of SWAP-70 in the pellet in the experiment with Factin)/(amount of SWAP70 in the supernatant in the experiment with F-actin)}/{(amount of SWAP-70 in the pellet in the experiment without F-actin)/(amount of SWAP70 in the supernatant in the experiment without F-actin)}. Confocal microscopy Cells were fixed with 3.7% formaldehyde for 5 min at room temperature. After permeabilization of the cells with PBS containing 0.1% Triton for 5 min, F-actin was stained with TRITC-phalloidin (Sigma-Aldrich). Cells were analyzed using a confocal fluorescence microscope (Fluoview 300, Olympus, Tokyo). Proteins bound to the beads were eluted with an SDS sample buffer and analyzed by Western blotting with anti-GFP antibody. Soft agar colony formation assay 1610 5 cells were suspended in 2 ml of a medium containing 0.2% agarose and plated onto a basal agarose layer consisting of a medium containing 0.5% agarose in 6-cm dishes and incubated for 2 weeks. 2 ml of a medium containing 0.2% agarose was added onto the culture every week. Serum requirement for the cell growth Cells were plated at a density of 5610 4 cells per 6-cm dish and cultured for 4 days in the medium containing low serum with a medium change on day 2. Growth of the cells were observed under the microscope. SWAP-70-374 localizes at the plasma membrane We introduced various mutations in SWAP-70 to find out interesting mutants. Especially, charged amino acids were targeted because they might affect the conformation of the protein. Among them, the 374 (K374A/K375A) mutation, which resides just downstream of the central PH domain, showed an interesting phenotype. This mutation did not affect the PtdIns(3,4,5)P 3 binding activity of the protein (Fig. 1A). Cellular localization of the protein was examined using green fluorescence protein (GFP) as a marker. Cos7 cells were transfected with the expression vectors for wild-type and mutant SWAP-70s fused to GFP. When the proteins are in the cytosol, the signals of GFP get gradually stronger toward the nuclei and the nuclei look almost transparent. In contrast, when the proteins are located at the plasma membrane, the strength of the signals is even and the nuclei look turbid. By these criteria, localization of SWAP-70s was observed under the microscope (Fig. 1B). The wild-type SWAP-70 was found mostly in the cytosol in unstimulated cells: but it moved to the plasma membrane and accumulated at the membrane ruffles after stimulation with EGF as described before (Fig. 1B). In contrast, a mutant, SWAP-70-374, was found at the plasma membrane in most of the cells even when cells were not treated with EGF (Fig. 1B). The protein moved to the membrane ruffles after stimulation with EGF, suggesting that this mutant protein is functional (Fig. 1B). The percentages of the cells with SWAP-70 located at the plasma membrane are shown in Fig 1C. This was confirmed by examining the X-Z image of the cells. As shown in Fig. 1D, wild-type protein found in the cytosol in unstimulated cells moved to the membrane after stimulation by EGF. The mutant, 374, also localized at the plasma membrane even without EGF-stimulation. It should be noted that this mutant co-localized with F-actin. Biochemical properties of SWAP-70-374 Because SWAP-70-374 co-localized with F-actin, its binding activity to F-actin was analyzed. As shown previously, SWAP-70 exhibits actin binding ability in a co-sedimentation assay with Factin only when the C-terminal portion containing actin binding domain was used. In contrast, full-length SWAP-70 does not exhibit this ability [7]. Considering the fact that SWAP-70 colocalizes with F-actin only when the cells are stimulated, these results suggest that the actin binding |
domain of full-length SWAP-70 may be masked and may be exposed only when activated. Because SWAP-70-374 co-localizes with F-actin under unstimulated conditions, we asked whether SWAP-70-374 bound to Factin in its full-length form in vitro. The co-sedimentation assay was performed with non-muscle F-actin using two purified recombinant proteins: His-SWAP-70-WT and 374. As shown in Fig. 2A, a considerable amount of His-SWAP-70-374 co-sedimented with Factin, whereas only very small amount of His-SWAP-70-WT cosedimented. To quantify the difference between the F-actin binding ability of the two proteins, the intensity of the bands of SWAP-70s was measured and the scores that indicate relative affinity of the proteins to F-actin were calculated (Fig. 2B). His-SWAP-70-374 showed much higher affinity to F-actin than His-SWAP-70-WT. The small amount of His-SWAP-70-WT co-sedimented with F-actin could be attributed to the nonspecific binding with F-actin through PH domain as previously reported [7]. These results suggest that SWAP-70-374 may take the activated conformation capable of binding to F-actin. Elements required for translocation of SWAP-70-374 to the membrane How SWAP-70-374 localized at the plasma membrane was examined using the mutants containing additional mutations. As shown in Fig. 3, SWAP-70-374 was found at the plasma membrane in most of the cells as mentioned above. Because translocation of SWAP-70 to the plasma membrane after growth factor stimulation requires binding of SWAP-70 to PtdIns(3,4,5)P 3 through its PH domain [19], dependence of localization on the PH domain-PtdIns(3,4,5)P 3 interaction was examined using a mutant SWAP-70-374m1 carrying an additional mutation in the PH domain, which abolishes the PtdIns(3,4,5)P 3 binding activity (Fig. 1A). SWAP-70-374m1 was still found at the plasma membrane, suggesting that the binding activity to PtdIns(3,4,5)P 3 was not required for translocation of the protein carrying the 374 mutation. In contrast, the SWAP-70-374 (1-564) protein, which lacked the F-actin binding domain at the very carboxyl-terminal region [7], failed to translocate to the plasma membrane, suggesting that the F-actin binding activity was driving the protein to the plasma membrane. These observations were reproducibly seen in all of the cells expressing the SWAP-70 mutants. Morphologic transformation of MEFs by expression of the mutant SWAP-70 Because SWAP-70 is required for full transformation of MEFs by the v-Src oncogene [3], it is possible that activated SWAP-70 can contribute to transformation of MEF 18-2. To examine this possibility, the mutant protein, SWAP-70-374, was expressed in MEFs. SWAP-70-374m1 was also used. Expression vectors for the SWAP-70s were introduced in the MEFs by electroporation and the stable transformants were selected by hygromycin. As shown in Fig. 4A, MEF clones expressing mutant SWAP-70s were obtained. When cells were plated in a low density, morphology of the cells expressing SWAP-70-374 was shorter and thicker than that of the wild-type cells, suggesting that expression of SWAP-70-374 weakens the ability of the cells to attach to the substratum as is the case with transformation by the src oncogene (Fig. 4B) [3]. In contrast, cells expressing SWAP-70-374m1 were flatter than the wild-type cells, suggesting that they are strongly attached to the substratum (Fig. 4B). Expression of SWAP-70 mutants affect the cell growth Cell growth of the transformants was monitored. MEFs expressing SWAP-70-374 grew faster than the parental cells or the cells expressing wild-type SWAP-70 ( Fig. 5A and B). The growth curve differed from time to time depending on the slight difference in the numbers of the cells plated on the cells, reflecting the delicate nature of the cell lines. However, in any case, growth rates of SWAP-70-374-expressing cells were higher than those of control cells. Also, saturation density of the cells was much higher in the cells expressing SWAP-70-374 (compare Fig. 5A and B). Although the growth rate of the cells was greatly enhanced and it appeared that cell number would increase without limit, these cells never piled up but rather formed a dense monolayer, unlike many other transformed cells, such as Src-transformed MEFs. In contrast, those expressing SWAP-70-374m1 grew more slowly than the parental cells with low saturation density (compare Fig. 5 A and C). These results agree with the fact that cells were flatter than the parental ones and suggest that SWAP-70-374m1 has an inhibitory effect on cell growth. Whether expression of SWAP-70-374 induces oncogenic transformation was also tested by the soft agar colony formation assay. As shown in Fig. 5D, the SWAP-70-374-expressing cells formed colonies in soft agar, whereas parental cells did not. Many of the malignant transformants can grow under the lowserum conditions. This was also tested. As shown in Fig. 5E, when the serum concentration was decreased to 0.5%, parental cells did not grow. In contrast, the growth of the SWAP-70-374-expressing cells was not affected. Growth rate in the medium containing 0.5% serum was about 80% of that in the medium containing 5% serum. These results suggest that SWAP-70-374 can induce malignant transformation. The focus formation did not give clear results. Because SWAP-70-374-transformed cells grow in monolayer, it was very hard to detect the foci. The SWAP-70-374transformed cells just pushed other cells to expand their area (data not shown). As shown in Fig. 4A, the expression levels of the exogenous SWAP-70 were always low in SWAP-70-374 and relatively higher in SWAP-70-374m1. The reason for this is not known. The low expression of SWAP-70-374 could be due to its toxicity or its instability. MEFs expressing SWAP-70-374 are sensitive to nutrient starvation Generally, transformed cells grow faster than the normal cells, pile up and do not exhibit contact inhibition. Therefore, whether MEFs expressing SWAP-70-374 show contact inhibition and resistance to nutrient starvation was tested. The transformants were cultured without medium change for 2 weeks. During the cultivation, parental and wild-type SWAP-70-expressing MEFs reached confluence and survived more than one week ( Fig. 6A and D). In contrast, those expressing SWAP-70-374 grew faster but when nutrients were used up, they died quickly (Fig. 6B, E, F). These results suggest that MEFs expressing SWAP-70-374 are not able to stop growing by contact inhibition and lose resistance to nutrient starvation. Those expressing SWAP-70-374m1 survived under the condition of nutrient starvation (Fig. 6C). Therefore, expression of SWAP-70-374 gives some characteristics of cell transformation to MEFs. Sanguinarine inhibits overgrowth of SWAP-70-374transformed cells as well as membrane ruffling induced by stimulation with EGF To understand how SWAP-70-374 transforms MEFs, drugs that inhibit overgrowth of SWAP-70-374-transformed cells under nutrient starvation conditions were screened. Among 300 drugs that include inhibitors for signaling molecules, proteins required for secretion systems, and various kinases, 5 drugs exhibited such an effect. To find which of these drugs target SWAP-70 itself or something very closely related to the SWAP-70 pathway, another assay was performed with these drugs. It has been shown that SWAP-70 is required for membrane ruffling induced by EGF, which is inhibited by wortmannin, a PtdIns-3 kinase inhibitor (Fig. 7A). The selected drugs were added to this system. Only sanguinarine inhibited membrane ruffling of Cos7 cells induced by EGF stimulation (Fig. 7A). Therefore sanguinarine inhibits two functions of SWAP-70, which is membrane ruffling and cell transformation. These results suggest that this drug targets SWAP-70 itself or another factor which is closely related to the SWAP-70 pathway. Sanguinarine greatly inhibited growth rate of SWAP-70-374-transformed cells by sanguinarine but not that of parental MEFs (Fig. 7B). These results support the idea that sanguinarine blocks the SWAP-70-related signaling pathway. ERK1/2 is activated in SWAP-70-374-transformed cells To see what signaling pathway is involved in transformation of MEFs by SWAP-70-374, activation of ERK1/2, which was often seen in transformed cells, was examined. As shown in Fig. 8A, ERK1/2 was activated in SWAP-70-374-transformed cells. In contrast, this was not observed in non-transformed cells. Therefore, the effect of MEK inhibitors on growth of SWAP-70-374-transformed cells was examined. As shown in Fig. 8B, the MEK inhibitors, PD98059 and U0126, blocked death of SWAP- 70-374 expressing MEFs after starvation with nutrients as was the case with sanguinarine (Fig. 8B). In contrast, the growth rate of these cells treated with the MEK inhibitors remained unchanged until they reached confluence (Fig. 8C). This was true with the parental cell line (Fig. 8C). However, these drugs caused contact inhibition of SWAP-70-374-transformed cells, even though saturation density was higher than that of parental cell lines ( Fig. 8B and C). Figure 8D shows that the MEK pathway was indeed inhibited by these inhibitors. These results suggest that activation of the ERK1/2 pathway partially contributes to transformation of MEFs by SWAP-70-374. Discussion It has been shown that SWAP-70 is closely related to malignancy of cancer. Overexpression of SWAP-70 has been often seen in malignant tumor and in various tumor cell lines [1,2]. In this paper, we demonstrate that a mutant of SWAP-70, SWAP-70-374, can morphologically transform, enhance cell growth, and abolish the ability to stop growing by contact inhibition including resistance to nutrient starvation in MEF cells. SWAP-70-374 may be a constitutively activated mutant. SWAP-70-374 was capable of binding to F-actin in vitro, whereas the wild-type protein was not. Corresponding to this finding, SWAP-70-374 co-localized with Factin at the plasma membrane constitutively, when the wild-type protein did so only when cells were stimulated with growth factors. SWAP-70-374 is likely to take a conformation similar to the activated form of SWAP-70. The finding that SWAP-70-374 can still translocate to membrane ruffles after growth factor stimulation suggests that the conformation of SWAP-70-374 may not be abnormal (Fig. 1B). ERK1/2 was found to be activated in SWAP-70-374 expressing cells, suggesting that SWAP-70-374 can activate some signaling pathways. The localization of SWAP-70-374 to the plasma membrane was dependent on the F-actin binding activity. The PtdIns(3,4,5)P 3 binding activity may not be required for this translocation because an additional mutation that should abolish the PtdIns(3,4,5)P 3 binding activity did not alter the membrane localization of the protein. It has been shown that translocation of the wild-type protein to the plasma membrane from the cytosol requires the PtdIns(3,4,5)P 3 binding activity of the protein. Therefore, the mechanism of membrane localization of SWAP-70-374 may be different from that of the wild-type protein stimulated with growth factors. These results suggest that SWAP-70-374 forms a unique complex that produces signals for abnormal growth of the cells. Although the PtdIns(3,4,5)P 3 binding activity was not required for membrane translocation of the protein to the plasma membrane, the PtdIns(3,4,5)P 3 binding activity appears to be still required for transformation of MEFs. Interestingly, SWAP-70-374m1, which is incapable of binding to PtdIns(3,4,5)P 3 , inhibited growth of MEFs. Expression of this protein slows down the growth of the cells and makes them flatter than the parental cells. Saturation cell density was also lower than that of the parental cells. It is likely that SWAP-70-374m1 may be in an activated conformation that can interact with other proteins in the signaling pathway: but because of SWAP-70-374m1's inability to bind to PtdIns(3,4,5)P 3, the signal may be stopped at SWAP-70-374m1 level. Sanguinarine has been shown to exhibit anti-tumor activity. It has been shown that sanguinarine induces apoptosis on various tumor cell lines [10,11,12,13,14]. It has been suggested that sanguinarine interacts with chromatin to modulate transcription [16]. In this paper, sanguinarine was found to inhibit transformation of MEFs by SWAP-70-374. Only SWAP-70-374-dependent acceleration of growth rate was inhibited by sanguinarine. Growth rate of 18-2 was not affected by sanguinarine. This can be explained as follows. SWAP-70-374 may activate not only the ERK1/2 pathway but also some other signaling pathways. One of the unknown pathways may be responsible for acceleration of the growth rate (Fig. 9). This unknown pathway may not be activated in 18-2 cells. In such a case, sanguinarine cannot inhibit growth of 18-2 cells but can inhibit that of SWAP-70-374-transformed cells (Fig. 7B). Unlike sanguinarine, MEK inhibitors did not slow down the growth of SWAP-70-374-transformed cells (Fig. 8C). However, MEK inhibitors caused restoration of contact inhibition of these cells (Fig. 8C). These results suggest that the ERK1/2 pathway regulates only contact inhibition of these cells and that there may be at least one other pathway that stimulates cell growth (Fig. 9). Characteristics of MEFs may differ from lines to lines. Whether the regulation of contact inhibition by the ERK1/2 pathway is common to all the MEF lines should be tested in the future. were cultured in the absence or presence of sanguinarine (0.8 mM), PD98059 (10 mM), or U0126 (10 mM) without medium change for 7 days. In the absence of the drugs, the cells died whereas they survived |
in the presence of these drugs. C, 18-2 cells or SWAP-70-374-2 cells were seeded at a cell density of 2610 5 cells per 6-cm dish. Cells were grown in the presence or absence of PD98059 (10 mM) or U0126 (10 mM) with medium change on every other day. Open circles, SWAP-70-374-2 cells without drugs; open triangles, SWAP-70-374-2 cells with PD98059; open squares, SWAP-70-374-2 cells with U0126; closed circles, 18-2 cells without drugs; closed triangles, 18-2 cells with PD98059; closed squares, 18-2 cells with U0126. Error bars indicate standard deviation. D, Confluent cultures of SWAP-70-374-2 cells were replated to fresh dishes and cultured for 1day in the presence or absence of PD98059 (10 mM) or U0126 (10 mM). The cells were lysed and analyzed for phosphorylation of ERK1/2. Left half, probed with an anti-phospho-ERK1/2 antibody. Right half, probed with an anti-ERK1/2 antibody. doi:10.1371/journal.pone.0014180.g008 Sanguinarine also inhibited membrane ruffling of Cos7 cells after growth factor treatment, which is another function of SWAP-70. Therefore, it is possible that this drug may inhibit SWAP-70mediated signaling pathways. In addition to sanguinarine's role in the nucleus, our new finding may supply information important for understanding of the function of sanguinarine. It should be noted that the concentration of sanguinarine required for inhibiting SWAP-70-374-induced transformation was much lower than that required for inhibition of chromatin related reactions. Because PtdIns(3,4,5)P 3 binding activity of the SWAP-70 mutant is required for transformation of MEFs, it is likely that the PtdIns 3-kinase pathway is also involved in the transformation of MEFs by the mutant SWAP-70. However, which signaling pathways other than the ERK1/2 pathway are required for cell transformation are not known. Further examination may be required to understand the whole picture of signaling pathways activated by the SWAP-70-374-F-actin complex. Determining the status of non-transferred embryos in Ireland: a conspectus of case law and implications for clinical IVF practice The development of in vitro fertilisation (IVF) as a treatment for human infertilty was among the most controversial medical achievements of the modern era. In Ireland, the fate and status of supranumary (non-transferred) embryos derived from IVF brings challenges both for clinical practice and public health policy because there is no judicial or legislative framework in place to address the medical, scientific, or ethical uncertainties. Complex legal issues exist regarding informed consent and ownership of embryos, particularly the use of non-transferred embryos if a couple separates or divorces. But since case law is only beginning to emerge from outside Ireland and because legislation on IVF and human embryo status is entirely absent here, this matter is poised to raise contractual, constitutional and property law issues at the highest level. Our analysis examines this medico-legal challenge in an Irish context, and summarises key decisions on this issue rendered from other jurisdictions. The contractual issues raised by the Roche case regarding informed consent and the implications the initial judgment may have for future disputes over embryos are also discussed. Our research also considers a putative Constitutional 'right to procreate' and the implications EU law may have for an Irish case concerning the fate of frozen embryos. Since current Medical Council guidelines are insufficient to ensure appropriate regulation of the advanced reproductive technologies in Ireland, the report of the Commission on Assisted Human Reproduction is most likely to influence embryo custody disputes. Public policy requires the establishment and implementation of a more comprehensive legislative framework within which assisted reproductive medical services are offered. Introduction Although the status of the 'unborn' has been debated since Aristotle [1,2], the search for clarification has never been more urgent. In delivering judgment on a dispute over the control of non-transferred embryos, Irish courts confront this ancient question in a way that will yield answers aspiring to be definitive, but will probably not receive universal approval. They may look to philosophi-cal debate in order to make an informed decision about the status of the non-transferred embryo compared to the status of the naturally conceived (in vivo) embryo and the subsequent protections afforded to it. This judicial landscape is not entirely new: the abortion debate also pivoted on this question, as did the issues of whether the contraceptive pill was ethically acceptable. Irish courts could also be influenced by the debates surrounding the (ulti-mately successful) Eighth Amendment to the Constitution Bill [3], as this reflected the prevailing electoral opinion on reproductive issues in 1983. While selecting conception as the beginning of human life has intrinsic appeal due to its simplicity as a discrete, identifiable event, that very characteristic draws criticism [4,5]. This approach means that human embryos conceived either naturally (in vivo) or with medical assistance (in vitro) would be accorded the same legal rights and protections. However, the non-transferred IVF embryo (in contrast to the in vivo embryo) is not necessarily 'on the way to being born'. Indeed, the embryo derived from IVF treatment still requires a significant amount of skill and intervention in order to provide it with the potential to be born. If this is not provided by medical science, and if the embryo is not transferred to the uterus, then the nontransferred embryo will not be born. If this view of the moral status of the embryo were strictly followed, then 'any research or other manipulationthat damages any embryo or interferes with its prospects for transfer to a uterus and subsequent development is ethically unacceptableonce conceived, the being was recognised as man because he had man's potential. The criterion for humanity thus was simple and all-embracing: If you are conceived by human parents, you are human' [6]. The potentiality argument assigns legal rights to the embryo derived from IVF on the basis that it has the potential to develop into a human being. This approach admits the uncertainty of pinpointing the exact time when life can be said to begin, and instead advocates that the embryo deserves entitlements because it has the potential for human life. But how should this potentiality position be applied to non-transferred embryos? In one sense, it fails to reconcile the distinction between what is and what could be. For example, 'the bare fact that something will become X (even if it will inevitably become X, which is far from being the case with the fertilised egg and the adult human being) is not a good reason for treating it now as if it were in fact X.' For example, anyone now reading these words is potentially dead, but that is hardly justification for treating those individuals as if they were already dead [7]. In the UK, the Warnock Report [4,8] concluded that the human embryo obtained from IVF does not have the potential to develop into a human merely because it occupies a petri dish in the laboratory. From this application of the potentiality principle, the IVF embryo has an undeclared (i.e., yet to be determined) moral value and may be stored, used for research purposes, or discarded. This position is supported by the recognised high loss rate for human embryos throughout the early stages of pregnancy in nature. Should Irish Courts adopt this potentiality prin-ciple in their quest to define the term 'unborn' for the purposes of the Eighth Amendment/Article 40.3.3 of the Constitution, then it will still have to define when the embryo has the potential to become a human being. Considering the contraveiling philosophical position holding that IVF embryos have no moral status that makes them worthy of protection, it has been argued that 'the most widely held view of the status of the embryo takes an intermediate position. It holds that the embryo deserves respect greater than that accorded to other human tissue, because of its potential to become a person and the symbolic meaning that carries for many people' [9]. This recognises that the embryo is a living entity deserving some respect, although not at the same level of protection as human persons. In other words, the IVF embryo before transfer should be afforded some protection due to its potential to develop into a human being and the symbolic significance it has [10]. Irish Courts might adopt this intermediate position when ascertaining the beginning of life in the context of a dispute over the fate of stored embryos. Regulatory dynamics in Ireland and the United Kingdom Over the past 20 years, modern fertility treatments including IVF have become more widely available to medical consumers. Although the advanced reproductive technologies are now more culturally accepted than before, how these treatments are regulated in Ireland creates difficulty for both consumers and providers, if a couple has stored embryos and their personal or family situation changes. Given the persuasive authority of the UK legal system in Ireland, legislators in Ireland might consider the UK's legal response to assisted reproduction as they develop here. Regulation of IVF in the UK is based on the Human Fertilisation and Embryology (HFE) Act 1990 [11]. This Act established one organisation, the Human Fertilisation and Embryology Authority (HFEA), to supervise and license all fertility clinics throughout the UK It also provides information and advice to the British government about embryos and treatment services governed by the Act, and endeavours to ensure that the whole area of reproductive technology is practiced in a transparent manner. But because no similar unifying agency exists in Ireland, on matters of communication, control, and public accountability, the regulation of IVF in Britain is in stark contrast to that in Ireland. Although regulation of the advanced reproductive technologies in Ireland is evolving, it remains fragmented and less sophisticated than in the UK The Medical Council was, by default, the only statutory body having any authority over IVF until recent years, when at least two other quasi-government bodies (the Irish Medicines Board and the Commission on Patient Safety and Quality Philosophy, Ethics, and Humanities in Medicine 2009, 4:8 http://www.peh-med.com/content/4/1/8 Assurance) began to take an interest in regulating some aspects of assisted reproductive procedures. Originally, the Medical Council exerted control over IVF because an agency with a specific remit for the advanced reproductive technologies was lacking. This circumstance necessitated IVF being regulated by the same board as all other branches of medicine. As of 2009, Ireland still does not have a single organisation mandated to coordinate monitoring and regulatory efforts for IVF, analogous to the HFEA in the UK Although the Institute of Obstetricians & Gynaecologists is the advisory body to the Medical Council on 'all matters relating to obstetrics and gynaecology in Ireland', when approached for direction or approval on providing specific IVF applications, the Institute issued a statement that it 'does not believe that it is within its remit to do so' [12]. The Medical Council was established under the Medical Practitioners Act [13] with responsibility for registering all physicians and setting professional practice standards. This body periodically publishes A Guide to Ethical Conduct and Behaviour [14] to provide 'a set of principles which doctors must apply in each situation, together with their judgement, experience, knowledge and skills'. The most recent Medical Council guidelines (March 2004) concerning IVF reminded doctors 'of their obligation to preserve life and to promote health. The creation of new forms of life for experimental purposes or the deliberate and intentional destruction of in vitro life already formed is professional misconduct' [14]. The Medical Council guidelines also state that IVF should only be used after thorough investigation has failed to reveal a treatable cause for infertility. Prior to fertilisation of an ovum, extensive discussion and counselling is essential. Any fertilised ovum must be used for normal implantation and must not be deliberately destroyed. Cryopreservation techniques have allowed greater flexibility in scheduling embryo transfers with no adverse effect on reproductive outcome, so IVF clinics can freeze surplus (non-transferred) embryos and remain compliant with Medical Council guidelines. In response to EU Cell and Tissue Directive 2004/23/EC [15], the Irish Medicines Board (IMB) acquired a role in regulating IVF laboratories due to the designation of such facilities as 'tissue establishments'. The IMB exists to assure compliance with international standards regarding quality and safety for the donation, procurement, testing, processing, preservation, storage and distribution of human tissues and cells. The task of accreditation of all Irish IVF programmes was therefore taken up by the IMB with a view to develop a system for notification of adverse events and reactions, to organise inspections and control measures, as well as |
to ensure data protection and confi-dentiality. The IMB also focused on tissue traceability and record maintenance that could interface with a single European coding system. In contrast to the broad nature of the Medical Council guidelines, the IMB approach to 'tissue establishments' is more narrowly defined. As 'tissue establishments', IVF providers in Ireland are required to secure a specific license from the IMB for human embryology laboratory operations. This IMB involvement in the arena of the advanced reproductive technologies, while perhaps appearing highly circumscribed, actually manifests a powerful regulatory presence over IVF since the embryology laboratory plays such a crucial role in the provision of IVF. Indeed, any IVF centre's laboratory licensure status as determined by the IMB also by extension determines the fate of that particular clinical IVF programme. Yet another entity with regulatory input regarding assisted reproductive treatments is the Commission on Patient Safety and Quality Assurance (CPSQA). This commission was chartered to "develop clear and practical recommendations to ensure quality and safety of care for patients". It concluded its work in mid-2008, and proposed regular clinical audits and a three-year licensing scheme for IVF clinics as part of the 'Building a Culture of Patient Safety' initiative [16]. The government of Ireland also chartered a Commission on Assisted Human Reproduction (CAHR) to guide approaches for regulating the advanced reproductive technologies here [17]. The absence of any Irish legislation dealing specifically with assisted reproduction left the CAHR free to consider the question ab initio. This multi-disciplinary panel issued a comprehensive report in 2005 addressing social, ethical and legal factors to be taken into account in determining public policy [18]. As the most recent expert examination of reproductive treatments in Ireland, the CAHR report highlighted the fact that Medical Council guidelines alone have insufficient depth to ensure appropriate regulation of the advanced reproductive technologies. Having examined current legislation in the UK and other countries, the CAHR report concluded that regulation by statute was desirable in Ireland [18]. Constitutional considerations in Ireland According to the Eighth Amendment/Article 40.3.3 of The Irish Constitution [19], 'The State acknowledges the right to life of the unborn and, with due regard to the equal right to life of the mother, guarantees in its laws to respect, and, as far as practicable, by its laws to defend and vindicate that right.' It is unclear, however, whether the protection stipulated in the Eighth Amendment/Article 40.3.3 applies from the instant of fertilisation or from some subsequent point in the developmental process. In the modern age of IVF, this dilemma thrusts the most ancient of questions, 'when does life begin?', to the practical intersection of medicine and law [2]. It is not merely a philosophical exercise; indeed it must be posed in order to determine when Constitutional protections are afforded to the embryo [5,10,20]. When the question is settled, it will establish what might be done with the embryo and also who decides in the event of any dispute. The fact that this matter has not been addressed by legislative or judicial process has significant implications for patients seeking fertility treatment in Ireland. Clarification can only be obtained from a substantive pronouncement from the higher Courts or by way of Constitutional referendum. This very controversy emerges in the matter of Roche v. Roche et al [21], a case certain to impact the future of fertility treatment in Ireland. The dispute involves a woman who wished to use embryos created by her eggs and her husband's sperm during their marriage in an attempt to become pregnant. After the couple had separated, the embryos remained in storage at an IVF centre in Dublin. Subsequently, her estranged husband refused to give consent for the embryos to be transferred. Mrs. Roche contends that the embryos had a 'right to life' by virtue of the Eighth Amendment/Article 40.3.3 of the Irish Constitution, and accordingly the embryos must be transferred in order to give them an opportunity to grow and develop in the womb and be born. The central challenge in the Roche case is establishing a definition of the term 'unborn' for the purposes of Constitutional law [21]. The vagueness of the term 'unborn' in Ireland is not new, and has been illustrated by legal difficulties in other contexts [22]. Constitutional protection for the 'unborn' or 'nontransferred'? An analysis of the campaign surrounding the 1983 Amendment to the Irish Constitution suggests that supporters of the amendment were satisfied that the term 'unborn' provided protection from the time of conception. The then-Deputy Prime Minister Richard Spring, said that 'It is clear that the word 'unborn' is likely to be interpreted by the Supreme Court as the moment at which the human ovum is penetrated by a sperm -the moment when human life commences' [23]. Yet the legally nebulous nature of the term 'unborn' makes it difficult to determine from which specific point in the fertilisation process the Constitution begins to protect life. [24] concluded that the right to life of the unborn applied from the moment of conception. Indeed, the Green Paper on Abortion [25] concurred with the ruling that conception (fertilisation) is the starting point for statutory protection of the 'unborn'. But a close reading of the Court's decision raises the possibility that the event of fertilisation may have been erroneously equated with the related (but physiologically distinct) process of implantation, which, for IVF patients, is conditional upon undergoing embryo transfer. Indeed, it was asserted that the human embryo either in vivo (in the body) or in vitro (in the laboratory) is entitled to constitutional protection from the moment of conception [1,[26][27][28]. If the embryo were to be guaranteed Constitutional protection from the instant of conception, then this could bring significant complications to the provision of reproductive medicine in Ireland. For example, if Constitutional protection were applied to all IVF embryos prior to in utero transfer, then would assisted embryo hatching (or any other beneficial/therapeutic procedure performed on an embryos) become assault? Certainly no action could be taken to allow IVF embryos to perish. Some authors [29] have expressed the view that 'it cannot be said with certainty whether the protection afforded by the Eighth Amendment/Article 40.3.3 to the "unborn" applies from the moment of fertilisation, the moment of the implantation or from some later date.' Nevertheless, in resolving a dispute over non-transferred embryos, Irish Courts may consider this judgment as it is the only substantive Irish precendent addressing the Constitutional status of the 'unborn' to date. The case of Attorney General (Society for the Protection of Unborn Children (Ireland) Ltd) v. Open Door Counselling Ltd Attorney General v. X [30] concerned an application by the State to prevent a fourteen year old girl (who was pregnant as a result of rape) from travelling to the UK for an abortion. There was evidence of 'real and substantial risk' that the girl would take her own life unless she left Ireland for the procedure. Mr. Justice Costello granted the injunction, but the Supreme Court reversed this decision on appeal and held that there was a real and substantial threat to the life of the girl, and that in such circumstances, the termination of pregnancy was constitutionally permissible [30]. Mr. Justice McCarthy added, 'I think it reasonable...to hold that the people when enacting the [Eighth Amendment to the Constitution] were entitled to believe that legislation would be introduced so as to regulate the manner in which the right to life of the unborn and the mother could be reconciled ' [30]. This statement highlights the fact that, if the Court attempts to provide any definition of the 'unborn' in a dispute over the fate of non-transferred human embryos, then the wishes of the electorate who enacted the Eighth Amendment may require interpretation. Given the novelty of IVF in 1983, it will be difficult for the Court now to ascertain public sentiment now based on electoral data collected then. The Court may consider that if the electorate were aware of the various contingencies associated with the advanced reproductive technologies, then they would have perhaps wanted the embryo to be protected from the earliest time possible, i.e. from the moment of fertilisation. In a dissenting opinion [30], Mr. Justice Hederman commented that distinctions could not be made between the phases of unborn life before birth. It was his position that Constitutional protection was not restricted by any condition the unborn was required to attain: 'The Eighth Amendment establishes beyond any dispute that the Constitutional guarantee of the vindication and protection of life is not qualified by the condition that the life must be one which has achieved an independent existence after birth. The right of life is guaranteed to every life born or unborn. One cannot make distinctions between individual phases of the unborn life before birth or between unborn and born life' [30]. This case generated intense debate throughout Ireland regarding the socially charged issue of abortion, and highlighted the inconsistencies in the Constitutional provison governing the protection of the 'unborn' [31]. The Constitutional Review Group (CRG) [32] subsequently focused on this point and recognised that the lack of clarity surrounding the definition of the unborn was creating significant legal and moral difficulties: 'There is no definition of "unborn" which, used as a noun, is at least odd. One would expect "unborn human" or "unborn human being". Presumably the term "unborn child" was not chosen because of the uncertainty as to when a foetus might be properly so described. Definition is needed as to when the "unborn" acquires the protection of the law. Philosophers and scientists may continue to debate when human life begins, but the law must define what it intends to protect. "Unborn" seems to imply "on the way to being born" or "capable of being born". Whether this condition is obtained from fertilisation of the ovum, implantation of the fertilised ovum in the womb, or some other point has not yet been defined' [32]. The expression 'on the way to being born', as recommended by the CRG, supports the argument that the nontransferred embryo should not be guaranteed protection by the Eighth Amendment/Article 40.3.3 of the Constitution. This is because without transfer and subsequent implantation in the uterus, the embryo cannot be assured to be ' on the way to being born' or 'capable of being born'. Indeed, many fertilised eggs are lost naturally and never implant in the uterus. As the All-Party Oireachtas Committee on the Constitution in its Fifth Progress Report on Abortion stated: 'great numbers of fertilised ova are lost in the natural course of things and never become implanted in the uterine wall. As a result some argue that implantation is the decisive event in the development of unborn life' [33]. An alternate interpretation offered in the CAHR report suggested that new human life exists once the process of fertilisation is complete; the embryo is more than simply a cluster of cells -for once destroyed, it is impossible to recreate that particular life [34]. From this position, it seems evident that any court decision concluding that a non-transferred embryo from IVF has the same 'right to life' as the embryo formed without assistance would have a profound effect on the provision of IVF in Ireland. Such a definition would certainly have implications for any couple engaged in litigation over the fate of stored embryos, since the non-transferred human embryo would legally acquire the classification as an 'unborn' for the purposes of the Eighth Amendment/Article 40.3.3. By extension, if such a definition were to be legally instituted, in order to 'defend and vindicate' its 'right to life' the State would seem obliged to have some legal mechanism to require the embryo to be transferred to a uterus. But would this level of governmental involvement override the procreational right of any party wishing to avoid parenthood [35,36]? This brings forward the important issues of 'abandoned' embryos [37] currently in storage at Irish fertility clinics, and the equally vexing challenge of mandatory transfer of embryos known to carry a genetic disease [38]. By extension, Irish Courts might determine that all IVF embryos come within the protection of the Eighth Amendment/Article 40.3.3. If this were to be the case, it would directly impact IVF providers and patients since it would require that all embryos produced would have to be transferred to the |
woman's uterus. Presumably, this would have to include the transfer of embryos even if they were found to be defective via pre-implantation genetic diagnosis. However, if the Eighth Amendment/Article 40.3.3 were determined not to apply to non-transferred IVF embryos, then there would be no Constitutional impediment to pre-implantation genetic diagnosis where embryos are "de-selected" for transfer based on a genetic condition. In 2005, the CAHR report recommended that the embryo formed by IVF should not attract legal protection until it was transferred in utero, at which stage it should have the same legal protection as the embryo created in vivo [18]. But is procreative liberty (the personal freedom either to have children or to avoid having them) Constitutionally protected? If it exists at all, any legal 'right' to reproduce appears to be a negative right, meaning that other persons have a duty not to interfere with the exercise of procreative choice. It does not imply any obligation for others to provide resources necessary to exercise that choice [39]. [40], the applicants wished to have cryopreserved sperm destroyed 'to prevent the birth of a fatherless child, disruption of their existing family, and additional emotional, psychological and financial stress'. In a dispute over the fate of frozen IVF embryos, a similar argument may be advanced by the party seeking to avoid procreation. In the Hecht case, the Court emphasised the argument that the embryo constituted property by ruling that Hecht was entitled to use the sperm bequeathed to her. In Hecht v. Superior Court of Los Angeles County Where a conflict of procreative autonomy has arisen, thus far judges tend to favour the party seeking to avoid procreation, as the alternate decision would force unwanted parenthood on that party [35,41,42]. The Irish judiciary could adopt a similar approach should any conflict of procreative rights arise here. Undoubtedly, the relevance of the Constitution to such rights will also have to be considered by the court. In a dispute over frozen embryos, the party wishing to use the embryos for transfer may argue that they have a Constitutional right to procreate, and any legal manoeuvre to prevent embryo transfer would be an infringement of that right. This argument may be forwarded if the court decides that the embryo is not protected by virtue of the 'right to life' guarantee contained in the Eighth Amendment/Article 40.3.3 of the Irish Constitution. Indeed, this may also be argued if the embryo is regarded as having an intrinsic 'right to life', as procreational rights will then be assessed without the impingement of constitutional provisions. However, while there is a Constitutional right to procreate, that right is not absolute in its guarantee. Article 40.3.1 protects a host of unenumerated rights, one of which is the right to procreate. This issue first came before the Court in Murray v. Ireland [43], where the plaintiffs were husband and wife both serving sentences of penal servitude for life. They had no children and conjugal visits were denied during incarceration. The plaintiffs claimed that as a married couple, they had a right to beget children, that this right was protected by the Irish Constitution, and that it was being infringed as a function of their imprisonment. The plaintiffs argued that as lawfully married persons, they had a basic human right to procreate and that this right was Constitutionally protected. Although they argued that there was a hierarchy of protected rights, and that the right to procreate children was very high on that scale of values, the Court rejected this claim [43]. Such a position may be taken by a party seeking to avoid procreation where the non-transferred embryo has been accorded Constitutional protection. A legislative approach for Ireland Considering the absence of a clear Constitutional role in the regulation of assisted reproduction in Ireland, political leaders may be influenced by legislation in the United Kingdom. The issues of greatest concern were addressed by the HFE Act 1990. In response to an amendment to this Bill seeking to clarify the legal status of the human embryo, a former Lord Chancellor stated, 'An embryo is not a chattelA human entity which is living is not a chattel and neither is it a person in any ordinary senseIt is wrong to try and define a human embryo in terms of established legal definitions which are plainly inapplicable to human embryos. Why must an embryo be one or the other? Why cannot it just be an embryo?' [44]. These remarks from the UK suggest that the non-transferred embryo is sui generis, occupying a category of interim property. In any event, the amendment was withdrawn and consequently the legal status of the embryo remained undefined. The Warnock Report [8] which preceded the HFEA, expressed that 'Until now the law has never had to consider the existence of embryos outside the mother's uterus. The existence of such embryos raises potentially difficult problems as to ownership. The concept of ownership of human embryos seems to us to be undesirable. We recommend that legislation be enacted to ensure that there is no right of ownership in a human embryo' [45]. But creating a new, third category of 'interim property' may not fully remedy the challenges brought by non-transferred IVF embryos. It has been claimed that 'the court would have no choice but to treat an extra-corporeal embryo as either a person or a chattel. The likely outcome would be that it would be held to be a chattel. Such law that exists points in this direction and the pragmatism of the common law would see that to treat an extra-corporeal embryo as a chattel is more consistent with common sense than for it to be given the rights of a person' [46]. While the HFE Act 1990 does not explicitly define the legal status of the embryo, it does vest control of such embryos in the providers of the genetic material [11]. In the UK, this is accomplished by a mechanism whereby the consents of gamete providers determine the disposition of the embryos and define the power to control the use of the embryos derived therefrom. Specifically, the gamete provider must, at the time that the gametes are procured, indicate in writing for what purpose(s) those gametes may be used. The gametes, and/or any resulting embryos may only be used in accordance with the provisions obtained from written informed consent [36,47,48]. In the UK, this documentation confers legal control to the gamete providers, thereby minimising the type of disputes which have taken place in the United States regarding the disposition of gametes and embryos. The IVF provider must provide thorough counselling and obtain an effective written informed consent from the couple seeking treatment (the gamete providers) to delineate the scope of use to which the gametes and/or embryos may be put [49]. While the topics covered in these medical consents are unpleasant and may be awkward for couples about to undergo IVF, it is imperative that the treating clinic makes an effort to document the patients' wishes across a wide range of contingencies before commencing the IVF sequence. These scenarios should include what is to be done with gametes and/or non-transferred embryos if the couple separate or divorce, if the gamete donor dies or becomes incompetent, the use to which the material can be put, and the maximum period for which the material can be cryopreserved. The task of discussing these various interand post-treatment eventualities is assigned to the physician who collects the informed consent, because the IVF doctor is considered to be the individual best suited to respond to medical questions arising from the advanced reproductive technologies. A similar regulatory process could be introduced in Ireland to combat the inconsistencies that emanate from the absence of explicit regulation. Indeed, it has been suggested that an IVF informed consent defining the disposition of non-transferred embryos would do much to avoid future unspecified situtations that may come before Irish courts [46]. The legislative model in the UK gives dispositional control to the gametic contributors of the embryo via a detailed written informed consent process that must be obtained before commencing IVF treatment. While Irish legislators could adopt a similar approach, they will have to be mindful of any determinations the Courts assert in the current Roche case regarding the constitutional status of such embryos. A property model for non-transferred embryos? Judges and legislators must now regard embryos and gametes in ways never before necessary or possible. If Irish courts do not ascertain that the frozen IVF embryo deserves Constitutional protection, then they could rely on the property model as a determining factor in resolving a dispute over the fate of frozen embryos. Such 'a property analysis may allow more flexibility in defining the legal status of the embryo and properly serve the needs of public policy' [36,47,50]. The lack of legal authority in Ireland has significant implications for the landmark Roche case now before the Irish judiciary. In looking to other jurisdictions, the Court may explore the issue of whether or not the embryo can be perceived as property for the purposes of ascertaining who has dispositional control over that embryo. However, this approach is not without difficulty: 'In law it is usual to categorise things into property, which can be owned and controlled, and persons, which cannot. It is difficult to decide which of these categories is most appropriate for the embryo' [39]. The classification of a non-transferred IVF embryo as property means that its ownership will be vested in the parents of those gametes that gave rise to the embryo. This would permit the couple to determine and control the fate of their IVF embryos including disposal of any such embryo(s) in any way they wished. Yet, there is the proposition that the language of property is inappropriate for discussions of the human embryo. Indeed, it may be argued that it is unrealistic to suggest that one person has 'property' in another whole living body. A host of complex legal issues also would emerge from this application of the property model in assisted reproduction including which of the parents can claim custody, the possibility of sharing embryos equally between the disputing couple, and the consequent moral difficulties that follow. Because this issue has come before senior courts in other jurisdictions, Irish judges may consider the settled case law elsewhere to ascertain how best to resolve a dispute over the fate of stored embryos here. In the United States, a property model was applied in Davis v. Davis [51], a case regarding non-transferred embryos. In this dispute, the central question was who should have dispositional control over stored IVF embryos following divorce. The ex-wife wished to use the embryos in an attempt to become pregnant. The ex-husband did not wish to become a father, and further contended that to proceed with embryo transfer against his wishes was a direct infringement of his right to avoid procreation. The Tennessee Supreme Court considered the 'person v. property' dichotomy, and held that 'Preembryos are not, strictly speaking, either 'persons' or 'property' but occupy an interim category that entitles them to special respect because of their potential for human life'. Judgment was affirmed in the ex-husband's favour. Therefore, although the Tennessee Supreme Court rejected the application of an absolute property model in Davis, it did recognise the decision-making control over the embryo by the couple. That decision was influenced by a report by The American Fertility Society, which recommended that 'Within the limits set by institutional policies, decision-making authority regarding pre-embryos should reside with the persons who have provided the gametes...in the absence of specific legislation on the subject' [52]. However, the matter of decisional authority has been criticised as just another way to ask the basic question of who 'owns' the IVF embryo. As one author noted 'Ownership does not signify that embryos may be treated in all respects like other property. Rather the term merely designates who decides which legally available options will occur, such as creation, freezing, discard, donation, use in research and placement in a uterus. Although the bundle of property rights attached to one's ownership of an embryo may be more circumscribed than for other things, it is an ownership or property interest nonetheless' Philosophy, Ethics, and Humanities in Medicine 2009, 4:8 http://www.peh-med.com/content/4/1/8 [53]. With respect to the |
property model, the decision in Davis could be regarded as an instance where 'recognition of the existence of a proprietary or perhaps pseudo-proprietary or quasi-ownership interest' [54] prevailed. Accordingly, if the IVF embryo constitutes property and belongs to the couple who contributed the genetic material to create that embryo, then it follows that the couple should have decisional authority over its fate. The property model was also applied in York v. Jones [55], where an IVF clinic refused to transport embryos to California despite a request from the couple involved. The Court considered the bailor-bailee relationship that existed between the couple whose non-transferred embryo from IVF was stored in a fertility clinic [54]. The Virginia clinic argued that their contract with the couple only allowed for three possible dispositions in the event the couple did not proceed with IVF at the clinic of origin: 1) embryo donation to another infertile couple, 2) donation for human embryo research, or 3) destruction. The York decision is significant because the language used is an explicit recognition of the couple's proprietary interests in the embryo. The case was decided in favour of the couple, as the Court noted that the medical treatment agreement did not state that the attempt to achieve a pregnancy was limited to procedures employed at the clinic of origin. Paraplaix v. CECOS [56] also framed the issue of reproductive autonomy in a property context. This matter concerned the use of sperm for posthumous reproduction, where a widow requested the release of sperm from her deceased husband in order to achieve pregnancy posthumously. However, he had not made any specific disposition of the sperm his will. His widow nevertheless contended that the sperm formed part of the property of the deceased's estate, and was thus capable of being inherited. In contrast, CECOS argued that sperm was an indivisible part of the decedent's body (similar to a limb) and was therefore not inheritable as property. The widow was ultimately successful on grounds of her procreational autonomy argument, but not on the basis of the property argument. The Court held that 'sperm does not constitute a thing in commerce but secretion containing the seed of life destined for human procreation'. While Paraplaix involved gametes rather than IVF embryos, it remains relevant because it offers one route Irish courts might take in construing the embryo as property. Since this decision held that 'sperm does not constitute a thing in commerce', if the embryo is aligned with sperm for analysis purposes, then Irish Courts might also reject the property model. The parallel is imperfect, however, since it could be argued that someone has a right to decide what is to be done with stored sperm. The person who made the deposit should 'own' or have a property interest in the stored sperm as it 'is his semen in a biological and property sense, and thus he has the right to decide what happens to it' [53]. Furthermore, sperm is significantly different from the non-transferred embryo as the latter is created by sperm and ova derived from a participating couple, unlike sperm which is the simple deposit of one individual -a specimen by itself lacking the potential to develop into a human being. For frozen sperm, a property analysis can therefore be more easily argued. Although these precedents in other jurisdictions could influence Irish Courts in determining which model to adopt where the fate of surplus embryos created by IVF is disputed, this influence hinges on the applicability of the Constitution to the IVF embryo. Should Irish courts determine that Constitutional protections are not applicable to non-transferred embryos, then a modified property model may be used. This does not accept the property approach absolutely, but rather would recognise the respect owed to the human embryo while acknowledging that the participating couple have an interest in what happens to the subsequent embryo. But is the non-transferred embryo ever 'owned'? To ascertain if the embryo can be regarded as 'property' so as to allow one party in a dispute to exercise control over the destiny of that embryo, it is useful to analyse the debate surrounding the ownership of other body materials or components. Intuitively, it is quite natural for an individual to think in terms of 'my body' and to infer that because it is 'my' body, that the person can determine precisely what is done to it, or its parts [57]. In Anarchy, State and Utopia [58], the concept of 'self-ownership' was introduced and has since figured prominently in philosophical debates on issues ranging from taxation and exploitation to suicide, abortion and surrogacy [59]. 'Since there is an open-ended set of use-privileges and control-powers over one's own body, it seems natural enough to speak of 'owning' one's bodyif I am not a slave, nobody else owns my body. Therefore I must own myself. Therefore I must own all my actions, including those which create or improve resources. Therefore I own the resources, or the improvements, I produce' [60]. The matter of body ownership was explored in Doodeward v. Spence [61], which highlighted a dispute where the plaintiff sought the return of the preserved corpse of a two-headed stillborn baby from the police. The plaintiff claimed that he had a 'property' claim over the corpse. The Australian High Court agreed, stating 'when a person has by the lawful exercise of work or skill so dealt with a human body or part of a human body in his lawful possession that it has acquired some attributes differentiating it from a mere corpse awaiting burial, he acquires a right to retain possession of it, at least as against any person not entitled to have it delivered to him for the purpose of burial, but subject of course, to any positive law which forbids its retention under the particular circumstances.' The Court of Appeal in R v. Kelly & Lindsay [62] accepted the methodology of Doodeward, and concluded that in certain circumstances parts of corpses are 'property' and are therefore capable of being stolen. In Kelly, a sculptor was alleged to have stolen human body parts from the Royal College of Surgeons. The question that arose for the Court was: Could these body parts be considered as property in order to convict the accused of theft? The Court held that parts of a dead body may be property (and may therefore be capable of being stolen) when 'they have acquired different attributes by virtue of the application of skill, such as dissection or preservation techniques, for exhibition or teaching purposes'. This exception, coupled with the decision in Doodeward, may be useful to Irish Courts in drawing conclusions that the embryo may be considered to constitute property. Application of the principles from Doodeward may yield a useful analogy to IVF embryos, as their creation certainly requires an element of special skill. It should also be noted that the storing and thawing of embryos presents some technical challenge, and could be regarded as constituting 'preservation' for the purposes of affinity with Doodeward and ascribing a property status to the IVF embryo. Interesting and complicated questions will arise as ownership rights over non-transferred IVF embryos are explored in Ireland. Perhaps the State could recognise property rights for IVF embryos, but also devise a framework within which those rights must be exercised. Otherwise, there could be some judicial hesitancy to resolve a dispute over stored embryos in the context of a pure property model. Furthermore, in the absence of a legally defined framework, application of an absolute property model to nontransferred embryos could provide a couple with dispositional control the power to excericse unlimited rights over their embryos -potentially doing whatever they wished with that 'property'. Informed consent and non-transferred embryos: limitations and legal challenges While the legal status of non-transferred embryos awaits legal and/or judicial clarification, the process of informed consent still must be transacted between IVF patients and their doctors. Medical informed consent is a crucial factor in any dispute arising over the disposition of stored embryos. In a medical context, informed consent is a formal process by which a patient can be said to have given permission for a proposed treatment based on a full and clear appreciation and understanding of facts, the implica-tions and future consequences of the proposed treatment, the expected outcome(s) following treatment, awareness of alternative treatment(s), and the possible consequences of refusing treatment. To give informed consent, the individual concerned must have adequate reasoning faculties and be in possession of all relevant facts when consent is given. Unless a valid informed consent is obtained, any subsequent physical interaction between doctor and patient constitutes assault, unless the encounter is covered by the concept of necessity. Since IVF is always an elective medical procedure, therapy cannot begin unless the treating physician obtains written informed consent first [36,47,63]. Although there have only been a small number of cases involving the status of IVF embryos thus far, some Courts have regarded the informed consent (when properly executed) as a binding contract, and decide who shall have dispositional control over the embryos on that basis [20,47]. In an Irish context, if Courts should adopt this contractual model in order to decide 'custody' of any disputed non-transferred embryos, then this would allow the judiciary to avoid the difficult questions raised by the Constitutional status of the 'unborn'. Unfortunately, the language of informed consent for IVF in Ireland is not uniform and has never been standardised, so variation among clinics could be problematic. As background, surplus or non-transferred embryos from IVF may be safely cryopreserved for extended intervals [64,65], although a specific informed consent is also required for this procedure. In Ireland, there is no limit to the duration of such storage, in contrast to the UK, where embryos are frozen for a maximum period of ten years. After ten years, surplus embyos in the UK are thawed without transfer, and thus are 'allowed to perish'. It should be noted that the 'thaw without transfer' provision for nontransferred embryos permitted in the UK cannot be easily adopted in Ireland, due Medical Council guidelines and the unresolved question of Constitutional protection of the unborn as contained in Article 40.3.3. The potential conflict between 'thaw without transfer' and the Eighth Amendment/Article 40.3.3 of the Irish Constitution notwithstanding, some IVF providers in Ireland include language in the written informed consent stipulating certain circumstances where the clinic retains the right to dispose of non-transferred embryos by thawing them without transfer. Patients undergoing an IVF sequence at such a facility must agree in writing that when the wife's age exceeds 45 years, or when the cryopreserved embryo(s) have been in frozen storage in excess of 5 years -whichever occurs first -any non-transferred embryos will be removed from storage, thawed, but not transferred (i.e. destroyed). Any informed consent executed in Ireland containing such a 'thaw without transfer' clause invites the possibility of a judicial test of contract enforcability. This is because institutions using this type of informed consent in Ireland make access to IVF conditional upon a couple's agreement that their non-transferred embryos could be 'thawed without transfer', a procedure itself inconsistent with current Medical Council guidelines. While this may have been influenced by standards lawfully set in the UK, a written informed consent in Ireland enumerating situations where intentional destruction of human embryos by 'thaw without transfer' treads on terrain that has not yet been Constitutionally tested. In the event of circumstances changing in the relationship of the couple who underwent IVF treatment, such as the death of either partner, separation or divorce [63,66], what would be the fate of any non-transferred cryopreserved embryos? What should be the scope of informed consent for IVF in such cases? From a public policy perspective, is an IVF clinic's registration status jepordised (with the Medical Council) if it incorporates a standard 'thaw without transfer' clause into its treatment agreement? Would such an informed consent be considered legally enforcable? How might Irish Courts analyse this issue? The Irish Constitution (particularly Article 40, regarding the 'unborn' and personal rights) is the only substantive law in Ireland which can be applied to these matters at present. Courts in Ireland may consider decisions in related cases from other jurisdictions to resolve such issues. The UK case, R v. Human Fertilisation and Embryology Authority (ex |
parte Blood) [67] involved a dispute over the fate of surplus embryos and specifically addressed informed consent. Under existing UK law the HFEA was allowed to prohibit a woman from using her husband's sperm in the UK, because it was obtained without consent from her husband (who was incapacitated due to a comatose state). In 1997, the HFEA permitted the wife to export the sperm to Brussels for use in an IVF treatment there. The Court of Appeal's judgement confirmed that the principle of informed written consent is an essential part of English law, and that the posthumous storage or use of sperm or eggs without such consent is unlawful. The European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR) [68], offers a useful appeal mechanism for any individual from the E.U. wishing to revisit a Supreme Court decision regarding the status of non-transferred embryos. In Article 12, this convention provides that 'Men and women of marriageable age have the right to marry and to found a family, according to the national laws governing the exercise of this right.' And, since Article 8 states that 'Everyone has the right to respect for his private and family life' [68], it could be argued that the right to procreate is inherent in either ECHR article. Accordingly, should the Supreme Court of Ireland make any pronouncement on the Constitutional right to procreate, then the Roche case could be taken before the European Courts with the assertion that the decision infringed upon the individual right to procreate (or to avoid procreation) under the ECHR. That Court's judgment is binding on each EU member state that has ratified the Convention, however there is no mechanism by which the member state can be forced to implement that judgment. Unlike EU law, the ECHR cannot overrule the Constitution of Ireland, although many of the rights in the ECHR are already contained in the Constitution of Ireland. Thus, although individuals have the opportunity to seek appeal in the European Courts, it is evident that the State may avoid any obligation to implement a judgment from European Courts if such judgment is inconsistent with the Irish Constitution. An example of this chain of appeal of a member state's Supreme Court to the European Court is Evans v. Johnston HFEA et al [69]. This case presented the first opportunity the UK had to address the question of who should decide the fate of a frozen embryo when disputed by its two progenitors [70]. Following passage through the highest levels of the English judicial system, the Evans case was deliberated by the European Court of Human Rights. Due to the lack of precedent regarding the fate and status of non-transferred embryos in Ireland, the substance and trajectory of the Evans case are likely to affect the Roche case, or any similar case, brought before Irish Courts. In this dispute, Ms. Evans and Mr. Johnston were engaged to be married and the couple had undergone IVF treatment as chemotherapy treatment left Ms. Evans infertile. They had pre-implantation embryos in storage intended for future use. For Ms. Evans, the frozen non-transferred embryos therefore represented her only chance of having a child to which she was biologically related. When the relationship ended and Mr. Johnston decided not to proceed with IVF, it was contended that his withdrawal of consent would interfere and permanently frustrate Ms. Evans' overwhelming ambition to have children. The central issue here was whether the interference with her private life was necessary and proportionate given the circumstances. Ms. Evans advanced an number of arguments that the English law's interference with her private life was unlawful. Mr. Johnston wanted the six non-transferred embryos destroyed, while Ms. Evans wanted to use them to achieve pregnancy. Ms. Evans sought an injunction from the English courts compelling Mr. Johnston to restore his consent. She claimed that: (i) Mr. Johnston may not vary or withdraw his consent to the use of the embryos, (ii) The embryos may be stored throughout the remainder of the ten year period, (iii) Ms. Evans may lawfully be treated with embryos during the storage period, (iv) A declaration of incompatibility to the effect that Section 12 and Schedule 3 of the 1990 HFE Act 1990 breached her Article 8,12 and 14 rights of the ECHR, and (v) That the embryos are entitled to protection under Articles 2 and 8 of the ECHR [70]. In UK Courts, Ms. Evans unsuccessfully argued that her expartner should not be allowed to withdraw his earlier consent to the creation, storage, and use of the embryos. The Court took notice that under the HFE Act 1990, consent from both partners is required at every stage of the IVF process, including transfer. The Court of Appeal dismissed her appeal and held that Ms. Evans and Mr. Johnston only gave consent to the use of the embryos for 'treatment together'. As the couple were no longer together, the informed consent provided by Mr. Johnston's at the beginning of the IVF sequence was considered no longer valid. The Court highlighted that under Schedule 3 of the HFE Act 1990, Mr Johnston had an unconditional statutory right to withdraw or vary consent to the continued frozen storage of the embryos as well as their transfer. The Court of Appeal also held that the statutory requirement of continuing consent did not breach the European Convention on Human Rights, as it was proportionate to the legislative aim of protecting the rights and freedoms of both parties. The Court stated that a non-transferred embryo does not have a qualified right to life under Article 2 of the ECHR. While the Court of Appeal determined that the refusal of treatment (due to Mr Johnston's withdrawal of consent) was an interference with Ms. Evans' right to respect for private life under Article 8 of the ECHR, the Court held this interference was proportionate to the need that made it legitimate. They did not recognise that this was Ms Evans's last chance to have a genetic child [70]. After exhausting all avenues of appeal in the UK, Ms. Evans next turned to the European Court of Human Rights in Strasbourg, arguing that by making the exercise of her reproductive rights dependant on Mr Johnston's consent, her rights governed by Article 8 had been infringed. The European Court found that such a position would diminish the respect owed to Mr. Johnston's private life in proportion, as it enhanced the respect accorded to hers. The Court also noted that a balance must be struck between Ms. Evans and Mr. Johnston's privacy rights, yet the legislature (acting through its designee, the HFEA) had done this by requiring consent from both parties. Ms. Evans also argued discrimination under ECHR Article 14, on the grounds that in natural reproduction, men cannot withdraw consent to reproduction once the sperm has been voluntarily given whereas in IVF, men could modify or withdraw their consent at various stages. Nevertheless, the European Court decided that this difference in IVF was justifiable because even though in natural conception the male does not have the right to withdraw consent, the HFE Act 1990 aimed to treat both parties equally even though this does not happen in nature. The UK scheme of consents inherent in the Human Fertilisation and Embryology Act was therefore upheld, with Mr. Johnston prevailing [69]. It should be noted that Ms. Evans' argument that the nontransferred embryos had a 'right to life' under ECHR Article 2 was denied. The Court denied this argument because English law does not protect the rights and interests of the foetus prior to birth thus, so it would be anomalous to give such rights to non-transferred embryos. However, application of this component of the Evans decision in Irish law will be difficult given the differences between Ireland and the UK with respect to Constitutional issues. Evans was the first case on fertility treatment to be considered by the Human Rights Court, and the judges acknowledged that there was no international consensus about its regulation or to the status of non-transferred embryos. The Court was sensitive to the fact that their verdict deprived Ms. Evans of the ability to give birth to her own child, but they did not seem to consider that this may be the determining factor to allow her use of the stored embryos. The Court also insisted that Ms. Evans's inability to override the withdrawal of consent by the other 'genetic parent' did not breach her human rights. While some commentators on the Evans case recognised that women's investment in pregnancy, child-birth and the IVF process is greater than men's [42,71], no jurisdiction thus far has regarded this as a sufficient reason to elevate female rights above those of the male, with respect to ECHR Article 8. Given the lack of precedent in Ireland, the Evans case is likely to be a substantial persuasive authority on the Roche case, or any similar case brought before Irish Courts. AZ v. BZ [72] furnishes another example where the validity of consent forms signed by a couple before beginning IVF treatment and embryo storage was considered. The Court held that a consent paper signed by a man stipulating that in the event of separation his wife could have the use of stored embryos, could not be enforced against him years later, as it would require him to become a parent against his wishes. The Court further held that agreements entered into by the couple when they initiated IVF should be enforceable, but determined that either party could later change his or her mind about the disposition of the embryos, and if, after a divorce the parties disagreed as to disposition of any non-transferred embryos derived from IVF, then the party wishing to avoid procreation should prevail. The Appellate Division of the New Jersey Superior Court held that 'a contract to procreate is contrary to New Jersey public policy and is unenforceable', relying on the 1988 decision In the Matter of Baby M [73], which held that a maternal surrogacy contract was unenforceable as a matter of public policy. In New Jersey, the Court ultimately decided that the right not to procreate outweighed the right to procreate. While the Court suggested that it was willing to enforce agreements dealing with non-transferred IVF embryos, its rule ultimately rendered all such contracts unenforceable. This is because mere disagreement by either party vitiates the contract in favour of the non-procreative rights of the party wishing to disallow embryo transfer and subsequent potential pregnancy. A thoroughly prepared and properly obtained written informed consent prior to IVF can still be helpful at law [48,63,66]. The verdict in Roman v. Roman [74] determined that a written informed consent signed by a couple specifying that their frozen embryos be discarded in the event of divorce was valid. The ruling reversed a trial Court's award of frozen embryos to the wife, who had wanted to use them for in utero transfer and potential procreation. The Court found that written agreements between embryo donors and fertility clinics to which all parties have consented are valid and enforceable, provided that the parties have an opportunity to withdraw their consent to the terms of the agreement. The Court determined that the public policy of Texas would permit a husband and wife to enter voluntarily into an agreement before IVF treatment commences and make provisions for the disposition of any non-transferred embryos in the event of contingencies such as divorce, death or changed circumstances. The decision in Roman suggests that carefully drafted IVF consent forms can be enforced, consistent with public policy. Indeed, the preliminary ruling in the Roche case directly addressed the issue of informed consent, and hinged on whether Mr. Roche had given consent for the future implantation of any non-transferred embryos [75]. Mrs. Roche asserted that her former husband was the legal father of the frozen embryos, having signed a contract agreeeing to take full responsibility for the outcome of IVF. But Mr. Roche told the Court that circumstances had changed, he no longer wanted to have more children with his former wife, and claimed that they never made an agreement about the future use of frozen embryos [76]. The Court ruled that Mr. Roche had not given his express or implied consent for |
future utilisation of the non-transferred embryos. Regarding this case, an officer of the Irish Fertility Society stated that, 'Each episode of treatment requires the con-sent of all parties involved and this consent may be revoked by any party should their circumstances changeWe cannot accept that either partner should be coerced into any fertility treatment, even if he or she has already had treatment which has lead to the creation of embryos' [21]. Yet is there merit in an alternate view that assumes the couple, by agreeing to undergo an elective medcial procedure such as IVF, create an implied contract to use any embryos derived from such treatment to attempt to have a child? Any subsequent disputes should therefore be resolved in favour of the party seeking to procreate, as that was the original intention of the treatment [77][78][79]. Voluntary participation in the IVF process could be regarded as conduct reasonably leading to the assumption that both parties have committed to reproduction. In the event of changed circumstances, the doctrine of promissory estoppel becomes essential, as the party who subsequently seeks to use any non-transferred embryos relies, to his or her detriment, on the other party's commitment to reproduce jointly: 'The partner who opposes implantation of the embryos should be estopped from asserting his or her right not to reproduce' [80]. But the Court determined that when Mrs. Roche initially gave informed consent when she agreed that the embryos would be transferred in utero, that the three embryos mentioned in the informed consent document were those which were transferred 'fresh', not frozen. The purpose of cryopreservation of any surplus embryos was to use them if the first 'fresh' transfer failed. Mr Justice McGovern noted that, 'In light of the evidence in this case and the documents which have been produced, it cannot be said that it was the presumed intention of the parties that the three frozen embryos would be implanted in the woman's uterus in the circumstances which have arisen, namely following the success of the first implantation procedure and the legal separation of the couple' [76]. In Litowitz v. Litowitz [81] the written informed consent stipulated that any non-transferred embryos be allowed to perish after five years in storage. A dispute developed after the couple separated, as the husband wanted to put the embryos up for adoption and the wife wanted to have them transferred to a surrogate mother. In this case, there was also an egg donor involved who wanted the embryos back if the wife did not prevail. The Court considered the husband's interest to avoid procreation as stronger than that of the wife, and ruled in his favour. While parties can trust courts to 'do the right thing' by balancing the conflicting interests should disputes over nontransferred embryos arise, IVF clinics have a responsibility to craft their informed consent in a way that is as compre-hensive as possible [47,63]. This is because medical practitioners have the best expertise in the field and are keenly aware of the potential difficulties that could develop [82]. It was just such an unfortunate outcome that led to the dispute in the Roche case, where 'neither the wife or the husband adverted to the issue [of embryo disposition] until their marriage broke downin the absence of the consent forms indicating agreement either expressly or by implication, there was no agreement as to what was to happen to the three frozen embryos in the circumstances which have arisen' [75]. One of the curious features about the Roche case was that the consent forms provided at the time treatment commenced did not specifically address the disposition of any frozen embryos [83], either in the event that pregnancy was achieved from the first embryo transfer, or if their circumstances changed such as upon the death of either partner or a separation or divorce. The informed consent process in the Roche case was therefore found unsatisfactory from a legal sense [76], because it outlined the fresh transfer sequence but did not anticipate contingencies arising from any cryopreserved embryos. Informed consents for IVF in Ireland mandating 'thaw without transfer' are at risk of similar judicial criticism, but for different reasons. While the Court's decision in the Roche case provided some minimal conditions for how informed consent should be structured for IVF in Ireland, the fertility clinic that treated the Roche's had already modified their consent process before the case became public. Nevertheless, clarification about what should and should not be included in Irish IVF consent forms is needed. Conclusion The two conflicting interests in cases dealing with the disposition of non-transferred embryos represent the right to procreate, and the right to avoid procreation [2,20]. The obvious problem for Irish Courts, in common with all other jurisdictions, is that enforcement of one person's rights denies the rights of the opposite party [84,85]. It is therefore necessary to determine whether these rights are equal, or whether one right should outrank the other [86]. The 'right to procreate' has been variously regarded as either superior to the 'right to avoid procreation', or that the 'right to avoid procreation' does not exist at all in the context of IVF [63,85]. Similarly, it has been argued that there is only a negative moral right to have children, therefore no one should be allowed to interfere with procreative efforts. Couples have no positive moral right to have children, which implies that childless couples have no positive right to assisted procreation and that the State has no obligation to provide such services. While various arguments posit that the rights of the party wishing to avoid procreation from use of non-transferred embryos should prevail [87], an exception has been suggested for the partner who has no other possibility of having a genetic child except by transfer of such embryos [79]. It has even been suggested that 'if the non-consenting party simply wants to avoid having custody or financial reponsibility, a court could convert the party's status from being the parent of a frozen embryo to being an "egg donor" or "sperm donor" without the custody or financial obligations of parenthood' [88]. But whether or not a parent has the right to obtain such a waiver or release from the other parent (i.e., a disengagement of a gamete provider for the express purpose of avoidance of support obligations) on behalf of the child remains uncertain [89]. However, the 'right to procreate' has not yet gained the same legal traction that the right to avoid procreation has received [35,90], and procreational autonomy is only one model the Irish Courts could adopt in resolving a dispute over the fate of non-transferred embryos. This would require a difficult assessment of the rights of both parties involved (and potentiallly the rights of the embryo) in deciding who will control embryo disposition. Our judiciary is likely to balance the parties' rights in determining disposition of non-transferred embryos with consideration to precedents in the UK, U.S., and EHCR; such influences are likely to favour the party wishing to avoid procreation. Yet the need for statutory guidance is highlighted by the Roche case, as inconsistencies or omissions in treatment consent forms signed by Irish couples undergoing IVF are now left for medical providers to remedy at the clinic level. If this does not happen, then Irish Courts will expend considerable resources to resolve the resulting disputes. Indeed, 'the Irish Government has not provided a confidence-inspiring track record of legislatively dealing with complex and controversial problems raised either by reproduction or reproductive technologies. Policy and regulation is needed in the area of [assisted reproduction] and many observers of democracy assume that the public should have a positive role in debating desirable priorities for application' [91]. The vital public policy question therefore must be asked: Why has comprehensive regulation of IVF remained elusive in Ireland, despite our costly and impressive array of competent professional and academic agencies? If it were established that there is a Constitutional right to procreate in Ireland, then prior decisions rendered elsewhere suggest that there is a superior correlative right not to procreate. In a dispute over the fate of non-transferred embryos, would the Constitutional protection of the right to life of the embryo (if it were defined as 'unborn' for the purposes of the Eighth Amendment/Article 40.3.3) then override the desire of a party seeking to avoid procreation, if that right 'inalienable and imprescriptable' was protected by Article 41 of the Constitution as 'superior to all positive law'? In the absence of any precedent here, the direction taken by Irish courts awaits the Supreme Court's judgment in the Roche case. As the incidence of infertility rises, the number of IVF cycles will continue to grow. This trend, when added to the regrettable upsurge in marriage separation and divorce, cannot help but increase the potential for the involvement of Irish Courts in disputes involving nontransferred embryos. Although deciding the fate of nontransferred embryos in the wake of a relationship breakdown is an arduous task, the public policy consequences of ignoring the problem are undoubtedly more ominous. All branches of government in Ireland must act in a coordinated fashion and engage urgently on this far-reaching and profound issue with multi-generational ethical, political and legal ramifications. Predictive Factors for the Need of Tracheostomy in Patients With Large Vessel Occlusion Stroke Being Treated With Mechanical Thrombectomy Background: Patients with large vessel occlusion stroke (LVOS) eligible for mechanical thrombectomy (MT) are at risk for stroke- and non-stroke-related complications resulting in the need for tracheostomy (TS). Risk factors for TS have not yet been systematically investigated in this subgroup of stroke patients. Methods: Prospectively derived data from patients with LVOS and MT being treated in a large, academic neurological ICU (neuro-ICU) between 2014 and 2019 were analyzed in this single-center study. Predictive value of peri- and post-interventional factors, stroke imaging, and pre-stroke medical history were investigated for their potential to predict tracheostomy during ICU stay using logistic regression models. Results: From 635 LVOS-patients treated with MT, 40 (6.3%) underwent tracheostomy during their neuro-ICU stay. Patients receiving tracheostomy were younger [71 (62–75) vs. 77 (66–83), p < 0.001], had a higher National Institute of Health Stroke Scale (NIHSS) at baseline [18 (15–20) vs. 15 (10–19), p = 0.009] as well as higher rates of hospital acquired pneumonia (HAP) [39 (97.5%) vs. 224 (37.6%), p < 0.001], failed extubation [15 (37.5%) vs. 19 (3.2%), p < 0.001], sepsis [11 (27.5%) vs. 16 (2.7%), p < 0.001], symptomatic intracerebral hemorrhage [5 (12.5%) vs. 22 (3.9%), p = 0.026] and decompressive hemicraniectomy (DH) [19 (51.4%) vs. 21 (3.8%), p < 0.001]. In multivariate logistic regression analysis, HAP (OR 21.26 (CI 2.76–163.56), p = 0.003], Sepsis [OR 5.39 (1.71–16.91), p = 0.004], failed extubation [OR 8.41 (3.09–22.93), p < 0.001] and DH [OR 9.94 (3.92–25.21), p < 0.001] remained as strongest predictors for TS. Patients with longer periods from admission to TS had longer ICU length of stay (r = 0.384, p = 0.03). There was no association between the time from admission to TS and clinical outcome (NIHSS at discharge: r = 0.125, p = 0.461; mRS at 90 days: r = −0.179, p = 0.403). Conclusions: Patients with LVOS undergoing MT are at high risk to require TS if extubation after the intervention fails, DH is needed, and severe infectious complications occur in the acute phase after ischemic stroke. These factors are likely to be useful for the indication and timing of TS to reduce overall sedation and shorten ICU length of stay. INTRODUCTION Mechanical thrombectomy has been shown to be highly effective and is the standard of care for large vessel occlusion stroke (LVOS) (1). A large meta-analysis of randomized controlled trials, however, showed that >20% of patients with LVOS and mechanical thrombectomy (MT) had unfavorable outcomes with modified Rankin scores (mRS) of 4 or 5, likely requiring pronged in-patient stay with treatment on intensive care units (ICU) and increased risks for stroke and non-stroke related complications (1). Taking all ischemic stroke patients into account, it has been estimated that around 24% of this population needs ICU treatment (2). Various complications in LVOS-patients are associated with ICU treatment and lead to unfavorable functional outcomes (3). These complications include neurological causes of decreased consciousness by large infarctions, cerebral edema, or symptomatic intracerebral hemorrhage (sICH), which also |
can lead to respiratory complications like pneumonia caused by stroke-associated dysphagia and again can result in hemodynamic instabilities caused by systemic inflammatory responses (4). In these scenarios, multiple factors influence the decision if-, and at which time point to perform a tracheostomy (TS) to achieve long term ventilatory support, drastically decrease sedatives and shorten prolonged orotracheal intubation, both in combination associated with increased rates of pneumonia and length of ICU stay (5). Failed extubation rates of around 17% in mixed neurological ICU (neuro-ICU) populations with acute brain injury have been reported, not only describing an association with quantitative and qualitative measurements of consciousness (e.g., a Glascow Coma Scale (GCS) of ≤8 points and the inability to obey commands), but also identifying multiple other factors like chronic obstructive pulmonary disease or congestive heart failure being associated with increased risk of reintubation, reflecting that conventional extubating criteria are not applicable in this patient group (6,7). Therefore, the decision and timepoint of TS in major stroke patients should most likely be guided by clinical and functional status, emerging complications, and past medical history rather than only focusing on the state of consciousness and ventilatory function (5). If, in this context, an early TS translates into better clinical outcomes in major stroke patients is being investigated in a still unpublished randomized control trial (8), after Bösel et al. (9) showed feasibility and safety of early tracheostomy in major stroke patients in a pilot study. Patients with LVOS, in contrast to a general population of major stroke patients, represent a well-characterized subgroup both at risk for intravenous thrombolysis (IVT)-/MT associatedas well as cardio-respiratory associated complications increasing the risk for ICU-treatment, as most patients receive general anesthesia with orotracheal intubation prior MT. In this subgroup of major stroke patients, a combination of factors increasing the risk for TS, to date, has not been systematically investigated. The aim of this study was to describe the proportion and characteristics of LVOS-patients receiving a TS after MT and to determine factors predicting the need for TS during their ICU stay. Patient Population and Clinical Characteristics We used a prospectively derived, single-center database including all patients receiving MT for LVOS treatment between 2014 and 2019 in a large, academic stroke center and being treated on a neuro-ICU. This database includes information on time metrics, imaging, intervention, patient history, and stroke-as well as non-stroke associated complications after MT. Information on the timepoint of tracheostomy, blood gases, failed extubation and pulmonary diseases were obtained retrospectively through a chart review [IntelliSpace Critical Care and Anesthesia (ICCA) information system (Koninklijke Philips N.V., 2004-2017)]. Ethics approval was sought from the Ethics Committee of the University Medicine Göttingen (13/7/15) and all patients or next of kin gave informed written consent for the anonymized use of disease-related data on hospitalization. Patients were included in the analysis if a predefined dataset on pre-stroke history (comorbidities), periinterventional data (imaging, time metrics, scores), and clinical data on the post-stroke course (e.g., infectious complications) was complete and available, which was the case in 635 (88.7%) from 716 patients included in the databank in the mentioned time period. In the case of complete datasets, we included all patients with LVOS receiving MT with or without prior IVT. As the success of the reperfusion therapy is a major predictor for clinical outcome, we only included patients with no-or minimal perfusion (complete or near-complete vessel occlusion) of the occluded vessel on the first angiogram prior to MT [defined as modified Thrombolysis in cerebral infarction (mTICI) ≤ 1]. Therefore, patients with spontaneous reperfusion or reperfusion through IVT were excluded. All patients were treated on the Neuro-ICU of the University Medical Center Göttingen, Germany. We included all types of LVO being treated at the discretion of the neurointerventionalist performing the MT. The indication for IVT was according to the current German guidelines and all patients were treated from stroke experienced intensive care trained neurologists only. The decision for the need for a TS and its timepoint was made by ICU-trained senior consultant neurologists without the use of standard criteria or SOPs. All TSs were performed as surgical TSs by experienced Ears, Nose, and Throat (ENT) specialists. The indication for a decompressive hemicraniectomy was made in consensus with stroke experienced neurosurgeons. Stroke and Non-stroke Associated Complications After MT Following complications after MT were documented and investigated: failed extubation was defined as ≥ 1 complete removal of the orotracheal tube with the need of reintubation within a 48 h period. Hospital-acquired pneumonia (HAP) was diagnosed if an infiltrate/suspicion of an infiltrate was visualized on chest X-ray in presence of at least two clinical signs such as pyogenic secretion, fever (>38 • C), leukocytosis, or leukopenia (>10,000 or <4,000/l) or the detection of a pneumonia-typical pathogen in the bronchial secretion or blood (10). For the diagnosis of sepsis, the sequential (sepsis-related) organ failure assessment (SOFA) score has been used and was ≥2 in combination with positive blood cultures (11). Statistical Analysis Statistical analysis was performed using SPSS 21 (IBM SPSS Statistics, Armonk, NY, USA). Characteristics of all patients are shown as M ± SD if normally distributed, and as median with interquartile range (IQR), if not. Category variables were given as absolute frequencies and percentages and examined by the Pearson Chi-Square test for statistically significant differences between the compared groups. The different groups were examined for significant differences by using independent samples T-test or Mann-Whitney U test, as appropriate. Uniand multi-variate logistic regression analysis were performed for clinical and imaging factors as well as complications being unequally distributed between the TS and non-tracheostomy (nTS) group with p < 0.1 in a univariate pre-test. In multivariate logistic regression, the pre-identified factors for TS were included and the backward selection (Wald) was used. The clinical scores National Institute of Health Stroke Scale at discharge, mRS at discharge, and after 90 days were not included in the models and were only used to investigate an association between time from admission to TS and clinical outcome. In a final step, the model's regression coefficients of the identified predictors for TS were used to create a score, which again was analyzed using the Area Under the Receiver Operating Characteristic (AUROC) method. The cut-off score was defined as a score with maximal Youden-Index (Youdens J = sensitivity + specificity -1). Pearson correlations have been used to investigate the strength and direction of associations between treatment times and functional outcome scores. In all procedures, a p < 0.05 was considered statistically significant. With exception for diabetes mellitus and symptom-onset-torecanalization time, all factors described above were predictive for the need of TS in the univariate analysis. As shown in Table 2 Table 3). The regression coefficients of the multivariate logistic regression model given in Table 3 were used to create a score for the prediction of TS using the equation given in Supplementary Figure 1. This score, ranging from −7 to 4 points, showed an excellent predictive value for the need for TS (AUROC, 0.929, 95%CI, 0.884-0.974, p < 0.001). Patients with TS had a median score of −1 (IQR, −1 to 1), and patients without TS had a median score of −6 (IQR, −6 to −3; p < 0.001) points. A cut-off score of −2 points has been identified with a sensitivity of 81% and a specificity of 94%. As it can be assumed that in a high proportion of LVOSpatients with severe neurological deficits and complications the therapy was limited and therefore no TS has been performed, we conducted a secondary analysis combining patients who died during their neuro-ICU stay (discharge mRS = 6) with TS patients in one group (TS plus severe cases group) and compared these patients to the nTS group. Baseline characteristics are given in Supplementary Table 1, which showed multiple differences, especially concerning clinical scores, imaging characteristics, and complication rates. All these different factors were included in a multivariate logistic regression model with stepwise backward selection, which again revealed HAP, failed extubation, sepsis, and DH as strongest predictors for a combined endpoint TS and death at discharge (Supplementary Table 2). Table 3, there was no difference in admission pH, paO 2 , paCO 2 , oxygenation index, or oxygen saturation. Patients with TS showed a trend towards higher lactate on admission (1.5 ± 1.2 vs. 1.3 ± 0.7, p = 0.062). DISCUSSION In the present study, we identified HAP, failed extubation, the need for DH, and Sepsis as strong predictors for patients to undergo TS during their neuro-ICU stay after mechanical thrombectomy. TS is believed to have distinct advantages to ease weaning in patients with severe dysphagia, reduced level of consciousness, and post-stroke complications. These advantages include a lower death space due to a shorter cannula with weaning facilitation, reduced sedatives, and therefore better mobilization and reduced complications like pneumonia and bedsores, and increases patient comfort (12,13). However, reliable indications if-and when to perform a TS for patients with acute brain injury are not yet established (13). While studies on TS in mixed ICU populations showed conflicting results (14-16) a first prospective, randomized trial by Bösel et al. (9) (SETPOINT study) showed reduced mortality and sedatives associated with early TS in a population with ischemic and hemorrhagic strokes. A followup study (SETPOINT 2) in this respect is highly anticipated (8). In contrast to all possible advantages of TS in patients with acute brain injury, the procedural risks of TS (13) in its two forms as percutaneous tracheostomy and surgical tracheostomy in everyday clinical practice must be balanced against its benefits. Therefore, patient selection is key to being able to determine in which patients TS should be performed early and which patients are more likely to benefit from prolonged orotracheal intubation and later extubation trials. All previously published studies on the role of TS included either mixed ICU populations or mixed patient groups with acute brain injury or subtypes of stroke. Our study aimed to specifically address risk profiles for TS in patients with LVOS after mechanical thrombectomy, who became an important and intensively studied subgroup of stroke patients during the past years. Patients with LVOS with a short symptom to recanalization time, lower NIHSS, higher ASPECTS as well as no hemorrhagic transformation after MT are very likely to be early extubated in contrast to patients with MT/IVT associated early ICHs or large ischemic cerebral areas. Interestingly, we did not observe ischemic stroke-specific parameters or scores as strongest predictors for TS, but factors commonly complicating the clinical course of major ischemic stroke patients like HAP and (possible associated) sepsis. This reflects the finding, thatbesides the reduced level of consciousness-in around 35% of cases cardio-respiratory complications lead to an ICU admission of stroke patients (17). In contrast, stroke severity is likely to be reflected by the high predictive value of DH and failed extubation. TS has been reported to be required in up to a third of patients requiring a DH (18) and is usually performed in patients with a reduced level of consciousness with high ICP associated spaceoccupying cerebral edema and it is highly variable if a patient requiring this surgery can be readily extubated in a short time period after the surgery (19). In addition, failed extubations are highly likely in patients with stroke-induced dysphagia and reduced level of consciousness, being both highly prevalent in LVOS patients (20,21). Besides dysphagia and reduced level of consciousness also pneumonia contributes to the failure of extubations, prolonged ventilatory support, and TS (4). The risk for pneumonia itself is associated with stroke severity and stroke outcome and has been reported to be as high as 60% in multiple studies (4). In our study, almost all patients in the TS group had pneumonia compared with 37.6% in the nTS group, which is within the reported range of 4% to 56% of all stroke patients being treated in stroke units (4). This high prevalence of pneumonia increases the risk for sepsis again, which in most cases is caused by pneumonia (22). Taking all thoughts on these contributing factors to indicate a TS into consideration, and overall patient type emerges known to neurointensivists, as all these factors are influencing each other. Patients with severe strokes are at high risk to develop HAP and are likely to require DH, both again leading to failed extubation and to severe systemic inflammatory responses caused by multiple infectious pathways and |
DH as major surgery itself. These major predicting factors for TS also were directly or indirectly represented in a previously published score to predict the need for TS in a mixed stroke population [Stroke-Related Early Tracheostomy Score (23)]. Our data suggest that especially these kinds of patients are likely to undergo TS and therefore can be scheduled early for this procedure if these risk factors are present. These patients also might represent the subgroup of ischemic stroke patients most likely to benefit from early TS, given the possible benefits mentioned above. In this respect, however, clinical outcome in our study was not correlated with time from admission to TS. This might be explained by the current practice in our hospital that TSs are performed late to give patients the chance to functionally improve and to perform weaning and extubation trials. Therefore, differences in timepoints of TS were small and no possible differences in outcome could have been detected. Another reason could be, that the TS group is too small, and therefore not enough statistical power is present to detect influences of earlier TS on clinical outcome. What we observed, however, was a significant positive correlation between the duration of mechanical ventilation and clinical outcome. Rather of being an epiphenomenon likely to be explained by the fact that patients with complications and more comorbidities are also more likely to require longer ventilatory support, detrimental effects of mechanical ventilation (e.g., barotrauma, immunotrauma triggering systemic inflammatory responses) and prolonged exposure to sedatives itself can contribute to unfavorable outcomes (24,25). In this respect, as a correlation between mechanical ventilation periods and clinical outcome-, but not between the period from admission to TS and outcome was detected, from our data it can be speculated that not the modality of ventilatory support (TS vs. orotracheal tube) influences functional outcome, but the total time the patient is ventilator dependent. This again supports the notion and current recommendation that weaning should be initiated early and prolonged intubation should be avoided in stroke patients (5). The strength of our study is the inclusion of a large number of LVOS patients with prospectively derived data, all undergoing MT and being solely treated in a specialized neuro-ICU. However, using this prospectively derived data does not change the retrospective design of this study, which represents one of its major limitations. Other limitations include a single-center study design, only reflecting local practices and procedures associated with the use of TS, the lack of information concerning extend and type of dysphagia as a major contributing factor for the decision to perform TS, and a relatively low number of TS patients in this ischemic stroke subgroup. Another major limitation of this study is represented by a selection bias applicable to age and limitation of medical and surgical therapy according to the presumed and the actual patient will. Therefore, our data must be interpreted with caution and the results only applied in patients wishing full treatment. In addition, the score created with our single-center data must be validated in other datasets and can only be used considering the aforementioned limitations. In conclusion, the combination of HAP, failed extubation, DH, and sepsis after MT may be useful for patient selection to indicate TS. If, however, earlier TS-and not solely limitation of mechanical ventilation time-translate into better functional outcomes must be determined by prospective trials like the SETPPOINT 2 study. DATA AVAILABILITY STATEMENT The datasets presented in this article are not readily available because data can only be made accessible upon reasonable request. Requests to access the datasets should be directed to [email protected]. ETHICS STATEMENT The studies involving human participants were reviewed and approved by Ethics Commission of the University Medicine Göttingen, Germany. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. AUTHOR CONTRIBUTIONS IM designed the study and was involved in the acquisition and statistical analysis of the data, drafted, finalized the manuscript, and approved the manuscript before submission. KS was involved in the acquisition and statistical analysis of the data and approved the manuscript before submission. MB and JL contributed to the manuscript and approved the manuscript before submission. DB and M-NP contributed to the manuscript, was involved in the acquisition of the data, and approved the manuscript before submission. All authors contributed to the article and approved the submitted version. The Bayesian Design of Adaptive Clinical Trials This paper presents a brief overview of the recent literature on adaptive design of clinical trials from a Bayesian perspective for statistically not so sophisticated readers. Adaptive designs are attracting a keen interest in several disciplines, from a theoretical viewpoint and also—potentially—from a practical one, and Bayesian adaptive designs, in particular, have raised high expectations in clinical trials. The main conceptual tools are highlighted here, with a mention of several trial designs proposed in the literature that use these methods, including some of the registered Bayesian adaptive trials to this date. This review aims at complementing the existing ones on this topic, pointing at further interesting reading material. Introduction This paper is a bird's eye view of the recent literature on adaptive designs of clinical trials from a Bayesian perspective. Statistics plays a prominent role in the design as well as the analysis of the results of a clinical study and Bayesian ideas are well received by clinicians. In their book, Spiegelhalter and his coauthors [1] make a strong case in favour of Bayesian methods in health care, and in the last two decades Bayesian statistics has had a large impact in the medical field (see the superb review by Ashby [2]), the more so as its implementation gets easier thanks to better computational facilities. "Bayesian clinical trials: no more excuses" is the title of an editorial in Vol 6(3) of Clinical Trials [3]. The Bayesian approach has a good reputation at producing scientific openness and honesty. The Bayesian paradigm is especially appropriate at the planning stage of a clinical trial, when external information, such as historical data, findings from previous studies, and expert opinions, is often available and awaiting to be made the most of. As Donald Berry and his colleagues state in [4], we are all Bayesian at the design stage! Health authorities have issued important statements on the statistical, clinical and regulatory aspects of Bayesian clinical trials ( [5,6]), recently allowing and even advocating the use of innovative methods, in particular adaptive design; as the editors of this Special Issue point out, most statistical and biomedical journals have recently hosted proposals of trial designs with a Bayesian slant, in some cases virtual re-executions of published trials. A search carried out in PubMed in August 2020 has returned nearly 300 publications (half of them published in the last decade) which either propose or use Bayesian adaptive methods in the design of a clinical trial. This may be also thanks to the popularization by Donald Berry [7][8][9][10][11] and the efforts made by statisticians working in the pharmaceutical industry, one of the main players in the design of clinical trials, to incorporate Bayesian methods. This is shown in leading journals in clinical trial methodology, like Pharmaceutical Statistics, The Journal of Biopharmaceutical Statistics or Biometrical Journal. Some confusion occasionally arises between the concepts of "Bayesian" and of "adaptive" design, because of similarities in the outlook: in the Bayesian paradigm, accrued data are used to update the prior distribution on the parameters, via Bayes' Theorem, and in response-adaptive experiments the accrued data are used at each step, namely after each observation or predefined group of observations, to update the next planning decision. Either approach (Bayesian or adaptive) can stand on its own, and has developed independently of the other: we clarify this point later. We are interested in trial designs that are both Bayesian and adaptive. The data are recursively evaluated during the experiment: the posterior parameter distribution is recursively updated and used to modify the execution of the trial according to a previously established rule. The textbook by Berry, Carlin, Lee, and Muller [12] successfully illustrates Bayesian adaptive methods in clinical research and deals with design issues too. It goes almost without saying that randomization is a must in a clinical trial (for Bayesians too), to counteract several types of bias, for instance selection bias. The subject is advancing fast in many directions and it is impossible to keep up with all the different threads. Dealing with this topic properly would require several books. In this paper it has been chosen to focus on the methodology, followed by several examples of instances in which the methods are applied. The subject matter is organized as follows: Section 2 is a general discussion of Bayesian designs. In Section 3 we summarize the theory of response-adaptive designs from the frequentist viewpoint, with emphasis on their importance in clinical studies. Moving on to Bayesian adaptive trial designs, Section 4 deals with the different methodological approaches (utility-based, probabilityonly, predictive) and their use for randomization, sample size determination and early stopping. Section 5 reports examples from the literature. Section 6 lists some well-known real life trials performed according to a Bayesian adaptive design. In Section 7 we mention the on-going debate on the relevance of response-adaptive randomization designs in clinical trials, although at present the controversy is not directly addressed to Bayesian methods. Conclusions are drawn in Section 8. Bayesian Designs To start with, the very meaning of the term "Bayesian design" has to be contextualized. It is important to specify whether the Bayesian approach relates just to the pre-experimental plan, or to the analysis of the data as well. Because "Bayesian" is sometimes taken to mean "subjective", for clinical trials an "objective" data analysis is often preferred. In clinical trials in particular the operating characteristics of the design often include the power of some statistical test, strictly speaking a non-Bayesian feature. So some studies are hybrid: the inference is frequentist, but a prior probability on the parameters is used at the design stage. Several authors, for example Ventz et al. [13] have presented ways of combining Bayesian models and frequentist analyses in clinical trials. At the start, design issues were neglected by Bayesian statisticians, but Bayesian design has latterly become a well-established subdiscipline, as shown by the outstanding number of quotations of Chaloner and Verdinelli's [14] paper throughout the years. Unfortunately, that review has never been updated, and does not mention the topic of adaptive design, which at the time was only beginning to appear in the literature. As is well known, in the Bayesian model-based paradigm, the parameters of the model are treated as random quantities and the a priori knowledge about unknown effects of the interventions is expressed by means of a prior probability distribution on the model parameters. In the non-adaptive case, the experiment is designed entirely before the new data are acquired, and its performance may depend on the data still to be observed The fully Bayesian way to designing the trial would be a utility-based one, in accordance with Lindley [15]. In the model-based approach, the utility U will be a function of the experimental plan, the data y, and the model parameter values θ. A design action (deterministic or randomized) is chosen so as to maximize the expectation of the utility function with respect to the joint distribution of the parameters and the future data. In the eighties Smith and Verdinelli [16], Giovagnoli and Verdinelli [17] and Chaloner [18] proposed Bayesian designs that maximize the precision of the posterior distribution of the treatment effects under a linear statistical model. For non-linear models, Chaloner and Larntz [19] suggested choosing the design that optimizes the expectation, with respect to the parameters' prior, of one of the well-known design criteria: the best known ones-the so-called D-optimality and trace-optimality for instance ("Optimal design" [20])-have also Bayesian interpretations. Since optimality criteria are functions of the Fisher information matrix, which is already an average over the possible responses, this approach too is not conditional on the data. This suggestion has become very popular and keeps being used in the literature, with many variants, not all of which are entirely convincing. It should be stressed that the choice of a suitable utility function is related to the objectives of the trial. The specification of the function U is certainly demanding, as |
various aspects of the decision context have to be formally quantified. The main purpose of any experiment is obviously inference. Even if the subjects are patients, the main goal of a trial is knowledge, not healthcare, so the expected gain in information is a relevant component of U . In the past, designs for Bayesian analysis of the data would mostly choose utility functions based on Shannon information or on a quadratic loss. Nowadays, it is not uncommon to find Bayesian designs motivated by the precision of the estimates and power of the tests, in view of the fact that the experimental data will be analysed by frequentist tools. On the other hand, in a clinical trial the utility function may also need to reflect care for the subjects involved in the experiment. An opposite, rather extreme, example in the binary response case is the 0-1 utility: the gain is 1 for success, 0 for failure, which takes no account of inference. Since a clinical trial is a multi-objective experiment, the utility will more likely be a trade-off between several purposes: ethical, inferential and possibly economic. Verdinelli and Kadane [21] suggested a utility function which is a linear combination of the total expected value of future observations and the Shannon's information of the experiment. Clyde and Chaloner [22] too have proposed weighted combinations of utility functions, which are still mathematically valid utilities. In this way optimal designs for multiple objectives can be handled in the decision-theoretic framework, so long as the utilities are standardized to a comparable scale. For instance with binary outcomes one of the usual optimality criteria-determinant or trace, standardized-can be combined with the percentage of expected successes. Clyde and Chaloner [22] show that this approach is equivalent to finding the optimal experiment under one (primary) criterion subject to constraints on the minimal efficiency of the experiment under criteria for other objectives. The weights will be chosen by the experimenter to reflect the relative importance given to each utility component: it is advisable that the ethical and/or economic impact should not completely outweigh the inferential goal. Bayesian non-adaptive design has had numerous applications in clinical trials up to the end of the nineties and beyond, as shown in the paper by Ashby [2]. Adaptive Designs: The Frequentist Approach Response-adaptive-or simply adaptive-experiments are those in which the accrued data and the past allocations are used at each step to choose how to perform the next observations, according to pre-established rules. In statistics, "response-adaptive" (sometimes also "flexible") is a technical term. It means a predefined step-wise sequential algorithm which prescribes the modalities of the next step as a (deterministic or probabilistic) function of the accrued data. The choice of the pre-established rule is an essential part of the designas indicated by the FDA [6]-and must be taken into account to draw correct inferences. Response-adaptive design does not mean simply that at any stage you may want to change your plans. It should be stressed that the updating process of an experiment cannot take place in a haphazard manner, which could undermine the ensuing statistical analysis of the data. Typically it could lead to miscalculating the variance of the estimates or to drawing wrong conclusions. Burman and Sonesson [23] give a very clear-minded discussion. Thus the correct design of adaptive experiments poses challenging statistical problems: because of the dependence on the past data, the design itself will be a stochastic process, whose realization is the sequence of actual assignments of the experiment. A good introduction to adaptive designs is Chapter 1 of the collective volume edited by Pong and Chow [24]. It is useful to distinguish among the different types of adaptation useable in clinical trials: (1) Adaptive allocation rule-change in the randomization procedure to modify the allocation proportion. (2) Adaptive sampling rule-change in the number of study subjects (sample size) or change in study population: entry criteria for the patients change. (3) Adaptive stopping rule-during the course of the trial, a data-dependent rule dictates whether and when to stop for harm/futility/efficacy. (4) Adaptive enrichment: during the trial, treatments are added or dropped. A more detailed classification is the one by Huskins et al. [25]. All types of adaptation may also take into account covariates, i.e., the patient's prognostic factors/biomarkers. Adaptation may aim to ensure approximate balance of categorical covariates between the treatment arms in randomization (minimization and stratified randomization methods) or to best suit the characteristics of the subjects in the trial, when the covariates are of the "predictive" type, namely they interact with the treatments. Despite the intricacies of the theory, including whether inference should be conditional on the design or unconditional as Baldi-Antognini and Giovagnoli point out [26], adaptive methods are attractive to experimenters: they look and sometimes are more flexible, efficient, ethical and/or economical, and proceeding sequentially seems to reflect real life practice. Among the possible applications, at present the clinical ones are playing a prominent role, to the point that response-adaptive designs are almost identified with clinical trials, notwithstanding their relevance in other disciplines: experimental economics, computer science, aerodynamics, combustion engineering, psychology, social sciences, and many others, as a simple search in Google Scholar easily shows. Altogether, adaptive design seems to be a more congenial perspective for practicing clinicians. Adaptive procedures are often feasible in clinical trials, in which typically the recruitment of subjects takes place sequentially, especially when they are healthy volunteers. In particular, they are fairly frequent in early phase studies, because of the inherently adaptive nature of such trials. The "steps" may consist of just one observation at-a-time, or more than one, as explained earlier. They may also represent natural stages of the experiment. Even a twostage trial may be regarded as adaptive, if it is ruled in advance what decisions will be made conditional on the first stage data. Clearly such step-by-step procedures must be described and justified in the study protocol. A review of the literature unfortunately shows that some trials described as adaptive report not-so-rigorous ad-hoc adjustments of the experimental design along the way. A quick overview of non-Bayesian adaptive designs methods may be helpful, to present the main ideas. An authoritative starting point is Rosenberger's 1996 review [27] but important methodological advances have been made since. Dose Finding Trials for new drug development involve dose-finding. In Phase I we are usually interested in finding which value of the dosage x will produce a prescribed mean response, typically a small tolerated amount of toxicity. The experimental problem is essentially adaptive by its very nature: after each observation a decision has to be taken as to whether to leave the dosage unaltered for the next cohort or change it. There are the so-called rule-based designs without assumptions on the dose-toxicity response curve, like the notorious 3 + 3 rule. I have chosen not to dwell on these non-Bayesian methods, since a description can be easily found in several books about statistical methods in clinical trials. Another non parametric method is the Up-and-Down design, in which the assignment rule of the next patient involves a random component: see for instance Baldi-Antognini and Giovagnoli [26] for the theoretical properties. Surprisingly, only few developments of the Up-and-Down incorporating Bayesian ideas have taken place. In the parametric set-up, the mean response is modelled as a smooth increasing function Q(x, θ) of the dose, depending on an unknown vector parameter θ. Frequent choices for Q(x, θ) are the logistic quantile function and the probit function. The parameter θ can be recursively estimated through Maximum Likelihood. The design problem is how to modify the doses efficiently each time. Adaptive ideas show their potential also in "seamless" designs for Phase I and Phase II simultaneously ( [28]), or Phase II together with Phase III ( [29]). Response-Adaptive Randomization When equal allocation to the treatment arms was regarded as optimum, the only accepted randomization rule with two treatments was "tossing a fair coin". Since Efron ([30]), randomization schemes that move away from the traditional equal allocation have gained consent. Efron's Biased Coin is a randomization rule skewed in favour of the underrepresented treatment(s), regardless of the data; response-adaptive randomization, on the other hand, is a "biased coin" that skews the allocations using the accrued data, usually to favour the best performing treatment. An important introduction to the mathematics of response adaptive randomization in clinical trials is the book by Hu and Rosenberger [31]. For two treatments A and B with binary responses and p A and p B success probabilities, basic methods are the well-known Play-the-Winner and Randomized Play-the-Winner: see the book by Rosenberger and Lachin [32] for details and also for recent theoretical developments in adaptive randomization. The modern approach consist in first choosing an ideal allocation proportion of the treatments-a "target"-obtained by a trade-off between several purposes: ethical, inferential and possibly economic, as suggested by Baldi-Antognini and Giovagnoli [33]: see also [34]. Then an adaptive, possibly randomized, procedure is devised with the property of converging to this target allocation for all values of the unknown model parameters. Suggested adaptive rules that achieve this purpose are the Sequential Maximum Likelihood design, the Doubly-adaptive Biased Coin Design and the Efficient Randomized Adaptive Design (ERADE), which is a response-adaptive version of Efron's Biased Coin rule. These are explained, for instance, in Chapter 1 of the collective book edited by Sverdlov [35] that contains a description of the state-of-the art in adaptive randomization in clinical trials. The other contributions in the same volume dwell on further developments. Sequential Monitoring Selecting the sample size is the number one design concern in almost all experiments. In frequentist statistics the sample size is usually calculated so as to achieve a prescribed power for the statistical test of interest, under reasonable assumptions for the true state of nature. An adaptive approach to this problem is typically applied in clinical trials, when the trial design is in two or more stages. At the end of each stage the appropriate sample size for the next stage gets re-estimated, making use of the accrued data. The sample size issue is also related to early stopping, since as is well-known in clinical trials another approach is to fix a (maximum) number for the sample and include the possibility to stop earlier, if certain conditions are met. Interim analyses of the data at predetermined time periods are conducted before the data collection has been completed, in order to stop the trial early if some of the treatments tested show to be clearly harmful or clearly useless, or obviously superior. This is the oldest practice in adaptive design, usually referred to as group-sequential. Decision rules for stopping consist in setting boundaries for a predetermined test statistic so that some error probability requirements are satisfied. There is a rich literature on this topic [36][37][38][39]. Adaptive early termination has been applied, for instance, in the recent 2020 World Health Organization Solidarity Trial for COVID-19 [40]. In bio-pharmaceutical experiments the attention has focussed in particular on stochastic curtailment ( [41]). The stochastic curtailment rule computes the conditional power, i.e., the conditional probability that the summary statistics at the end of the trial is in the rejection region, given the current data available, under the null hypothesis of no effect or the alternative hypothesis (a clinically significant difference). This approach is based on a "prediction", thus it leans towards the Bayesian philosophy. Group-sequential designs in general do not consider treatment allocations other than equal size randomization. For some models, however, Jennison and Turnbull [42] have shown that response-adaptive randomization can be incorporated into a general family of group sequential tests without affecting the error probabilities. Zhu and Hu [43] have studied both theoretical and finite sample properties of designs which combine sequential monitoring with response-adaptive randomization. The Bayesian Viewpoint in Response-Adaptive Designs Adaptive designs are (deterministic or randomized) rules that at each stage of the trial, conditionally on the accrued data, prescribe how to choose the treatment assignments and/or include new or drop old treatments and/or choose the sample size for the next experimental stage and/or choose the next patients, and/or decide whether to stop the trial. In the Bayesian set-up, in order to choose the next step's or stage's settings in an optimal way the design rule makes use of the posterior distribution of the parameters, updated each time. If the approach |
is decision-based, the design rule recursively optimizes the posterior expected utility. Posterior distributions may also be used in more direct ways in the design of the experiment, without the decision theoretic framework, as we shall see in Section 4.1 about adaptive randomization. In the binary case, posterior probabilities correspond to the choice of the simplistic utility, where the gain is 1 for success, 0 for failure. An essential Bayesian tool is also the predictive probability of yet unobserved events, conditional on past data. Predictive distributions are useful in many adaptive design contexts, like trial monitoring and also deciding whether to conduct a future trial. Their use is shown in many clinical contexts in the book by Berry et al. [12]. Recently, Bayesian methods are more and more to be found naturally embedded in most of the emergent adaptive design ideas: the Continuous Reassessment Method for dose-finding in Phase I (see Section 5) is a typical example. Important review articles are by Chevret [44], who searched adaptive Bayesian design in the medical literature up to 2010, and by Rosner [45], who deals with Bayesian adaptive design in drug development. Bayesian adaptive designs are sometimes called BAD, not a very exciting acronym! When the randomization rule is adaptive, they are called Bayesian Adaptive Randomization (BAR), Bayesian RAR or Bayesian Response-Adaptive Randomization (BRAR); we prefer Bayesian Adaptively Randomized Designs (BARDs), a much more inspiring acronym. In the literature there is occasionally a temptation to describe as Bayesian some adaptive designs that make no use of priors or posteriors, like the Bayesian Biased Coin Design by Atkinson and Biswas [46]. Bayesian Adaptive Randomization The need to randomize treatments to patients is mandatory in all clinical trials, and is used in each Phase whenever possible, although it tends to be studied with particular reference to Phase III: these are multicenter case-control studies on large patient groups (300-3000 or more), aimed at assessing how effective the proposed new intervention(s) is in comparison with the current "gold standard". In spite of the popularity of Bayesian adaptive methods, Bayesian adaptive randomization for clinical trials does not seem to have been investigated extensively, as pointed out in the book by Atkinson and Biswas [47]. There is no straightforward Bayesian equivalent of Play-the-Winner for the case of binary data and two treatments, which is not surprising since Play-the-Winner is a myopic strategy, based on the most recent result, whereas the Bayesian paradigm is characterized by its use of all the past data. The very early paper by Thompson [48] is worth a special mention, because its Bayesian way to randomization in the binary model-reminiscent of the Randomized Play-the-Winner, and called "probability only"-is still very popular to this day. "If P is the probability estimate (meaning the posterior probability) that one treatment is better than a second, as judged by data at present available, then we might take some monotone increasing function of P, say f(P), to fix the fraction of such individuals to be treated in the first manner until more evidence may be utilised, where 0 ≤ f(P) ≤ 1; the remaining fraction of such individuals (1 − f(P)) to be treated in the second manner; or we may establish a probability of treatment by the two methods of f(P) and 1 − f(P), respectively". (Thompson, [48]) When f (P) = P this method is referred to as randomized probability matching: it is ethical from the patients' viewpoint but pays no attention to inference. However the function f may be chosen so as to act as a stabilizing transformation, to avoid excess variability which has a negative effect on the inferential efficiency. It has become customary to take f (P) and 1 − f (P) proportional to [Pr(p A > p B | y)] υ and [Pr(p B > p A | y)] υ , respectively, where υ is a positive quantity that modulates the tradeoff between the exploration (gaining information) and the exploitation (benefit for the patients) aims of the experiment. The value of υ recommended by Thall and Wathen [49], based on empirical experience, is υ = 1 2 , or υ = n/2N, where n is the present sample size and N the total proposed one. Extensions of the theory allow the progressive reassessment of υ based on interim analysis data. In principle, the idea of adaptive randomized designs converging to a prescribed target can be used in a Bayesian context as well. A utility function is chosen and at each step, a "temporary" target is found by optimizing the posterior expected utility. This pseudotarget-which is conditional on the data and gets updated each time-will be used, possibly after a suitable transformation, as the allocation probability of the randomization scheme. The intuitive meaning is to try to allocate the treatments-at each step-in the way that is optimal according to the present information. It may be looked at as a Bayesian equivalent of the frequentist Sequential Maximum Likelihood design mentioned in Section 3.2. Sample Size Determination and Early Stopping Although for a Bayesian analysis of the data in principle there is no need for preplanned sample sizes, nevertheless Bayesian design does consider sample size selection, both for practical reasons and for a potential classical inferential analysis. The prior distribution of the unknown quantities may be incorporated into finding the appropriate sample size in more ways than one (utility-based, pre-posterior, ecc.); see for instance [50]. Instead of the conditional power, the predictive power is often used, namely the predictive probability of rejecting the null hypothesis of no effect, or of no difference among effects: this approach indicates how the sample size of a clinical trial is to be adjusted so as to claim a success at the conclusion of the trial with an expected probability. The same ideas can be used when the trial is planned to be performed adaptively; at the end of each step the sample size of the next step will be selected. As to sequential monitoring, Donald Berry [51] and Spiegelhalter and Freedman [52] were perhaps the very first to suggest the application of Bayesian tools for the decision to stop the trial before the planned sample size is reached. Monitoring can be based on the posterior probabilities of the hypotheses of interest, like the posterior probability that the treatment benefit lies above or below some boundary, or based on the predictive probabilities of the consequences of continuing the study; the predictive power, namely the expectation of the power function with respect to the distribution of the true underlying effect size, is often relevant when deciding on whether to stop a clinical trial for futility. Useful references are [53,54]. Alternatively, the decision of whether to interrupt the trial may derive from a proper utility function that quantifies the gain associated with the consequences of stopping or continuing. A good discussion of the appropriateness of the different Bayesian decision rules is found in the books by Spiegelhalter et al. [1] and Berry et al. [12], and in [55]. Suggestions of Bayesian Adaptive Design from the Literature The best known Bayesian adaptive rule is the Continual Reassessment Method (CRM) by O'Quigley et al. [56,57] for Phase I: Cheung [58] devotes an entire book to it. It is aimed at finding a given quantile of the dose-response function Q(x, θ), with θ the unknown vector parameter. Often the response is toxicity and we are looking for the dose x p * corresponding to a given maximum probability p* of toxicity, called Maximum Tolerated Dose (MTD). Given a prior on θ, the expected value of x p * is calculated and the next set of observations is taken at the dose level nearest to it. The process is then iterated recusively and it can be proved to converge to the MTD. The main advantage of this design is that the majority of observations are centered around the dosage of interest. Many variants of the CRM have appeared over the years to handle different clinical scenarios, such as separate groups or late-onset toxicity. In particular: • The TITE-CRM by Cheung and Chappell [59] incorporates the time-to-event of each patient allowing patients to be entered in a staggered fashion. • Escalation with Overdose Control (EWOC) by Babb and Rogatko [60]: it is the same as CRM, except for the use of the αth-quantile of the MTD's posterior, instead of its mean, when selecting the next dose. This allows rapid dose escalation while controlling the probability of exceeding MDT. The extension of EWOC to covariate utilization permits personalization of the dose level for each specific patient. • The STARPAC design [61] uses a traditional rule-based design until the first patient has a dose limiting toxicity and then switches to a modified CRM. • Yin and Yuan [62] use the rather controversial idea of averaging the statistical model with respect to the parameter prior in conjunction with the Continuous Reassessment Method. Still about dose-finding trials of Phase I with a binary toxicity endpoint, examples of Bayes design rules obtained in a decision theoretic framework are: • The modified Toxicity Probability Interval (mTPI) design [63]. The decision to escalate or de-escalate the dose is made by partitioning the probability interval into three subintervals. The posterior probability that p* is in each subinterval is calculated, divided by the width of the subinterval. The interval with the highest posterior probability mass dictates the dose decision for the next patient. The mTPI possesses desirable large-and small-sample properties. These designs are compared in a numerical study in [64]. • The Adaptive Bayesian Compound Design by McGree et al. [65]: the authors use a compound utility functions to account for the dual experimental goals of estimating the MTD and addressing the safety of subjects. Bayesian optimal design theory is used adaptively in a two-stage Phase I design by Haines et al. [66]. There has been a widespread use of Bayesian response-adaptive randomization methods. Thompson's idea for adaptive randomization, extended from the case of two treatment arms to several arms, has been applied by Thall, Inoue and Martin [67] to the design of a lymphocyte infusion trial. Under a beta-binomial model, Yuan, Huang and Liu [68] design a trial for leukemia. The randomization assigns an incoming patient to the treatment arm such that the imbalance of a prognostic score across the treatments is minimized. This score depends on an unknown parameter whose posterior mean is continuously updated during the ongoing trial. Still for the Beta Binomial model, in Giovagnoli [69] the trace criterion is used as the utility function and a recursive "biased coin" is found that maximizes the posterior utility. The sequential randomized treatment allocation is shown to converge to Neyman's classical target, namely the optimal one according to the trace criterion. Under the same model, Xiao et al. [70] have defined a Bayesian Doubly-adaptive Biased Coin Design, using the posterior probabilities of p A > p B and of p B > p A , for the target and an assignment rule similar to the ERADE mentioned in Section 3.2. They derive some asymptotic properties of their Bayesian design, namely convergence and asymptotic normality of the allocation proportion. Giovagnoli and Verdinelli [71] choose a recursive target that optimizes the posterior expectation of a compound utility function and the ERADE algorithm for convergence. Turning to sample size determination and early stopping, sequential stopping is mainly associated with Phase II trials, but as early as 1994, Thall and Simon [72] developed a design with continuous monitoring until a high posterior probability is achieved that a drug is promising or that it is not promising, or until reaching the maximum sample size. This idea has been further refined and modified by a multitude of authors. • Wang [73] predicts how the sample size of a clinical trial needs to be adjusted so as to claim a success at the conclusion of the trial with an expected probability. • An interesting evaluation paper is by Uemura et al. [74]. • Continuous monitoring by means of predictive probabilities is given by Lee and Liu [75]: for the binary case, under a beta-binomial model, and a given maximum sample size, they recursively calculate the predictive probability of concluding the study rejecting the hypothesis of no efficacy of the new treatment. They search for the design parameters within the given constraints such |
that both the size and power of the test can be guaranteed. • Yin, Chen and Lee [76] have coupled Thompson's adaptive randomization design with predictive probability approaches for Phase II. • Zhong et al. [77] introduce a two-stage design with sample size re-estimation at the interim stage which uses a fully Bayesian predictive approach to reduce an overly large initial sample size when necessary. Decision-theoretic methods have been applied in this context too, for example by Cheng and Shen [78] for a comparative two-armed clinical trial. They specify a loss function, based on the cost for each patient and the costs of making incorrect decisions at the end of the trial. At each interim analysis, the decision to terminate or to continue the trial is based on the expected loss function while concurrently incorporating efficacy, futility and cost. The maximum number of interim analyses is determined adaptively by the observed data. Bayesian Adaptive Designs in Registered Trials Adaptive designs are mathematically sophisticated instruments. Their development is fairly recent, and the split that can be observed between theory and practice is not at all surprising. There are several obstacles-both technical and practical-to launching an adaptive trial, beyond the significant time and effort required by any clinical trial. Among other things, adaptive design requires updating information on accrued data, the speed of acquisition may be highly variable so there is the need to identify short-term endpoints that can be used to accurately predict treatment responses such as long-term mortality in terms of a gold-standard endpoint. The steps required to establish this type of design in a novel context are indeed fairly complex, as some case studies show (see for instance Mason et al. [79]. As to the Bayesian approach, this may include specialized software programs to run the study design, only made possible by recent advancements in computational algorithms and computer hardware ( [80]). Nevertheless, it is worth remarking that the philosophy of Bayesian adaptive designs has already made its way into the clinic. They are now fairly well established in cancer research ( [10]), and to a lesser extent, in other clinical areas. As well as single study designs, Bayesian adaptive methods are being employed to build "platform" designs (Adaptive Platform Trials). These are trials for simultaneous testing of multiple treatment strategies in separate groups, with plans to discontinue any group that is definitively inferior at planned interim analyses. Trial patients are enrolled in a continuous manner via a common master protocol, with interventions entering and leaving the platform on the basis of a predefined decision algorithm. Several Adaptive Platform Trials are now funded in various disease areas (see Angus et al. [81], Brown et al. [82] and Talisa et al. [83] for a discussion). The following is a non-exhaustive list of recent or still on-going clinical trials that incorporate Bayesian adaptive design features: Controversies There is an on-going debate on adaptive designs in clinical trials. The criticisms are not addressed specifically to Bayesian methods, nor to the general class of adaptive designs, but circle around the value of response-adaptive randomization in clinical trials. The much criticized Michigan ECMO trial (see [105,106]), which took place 35 years ago, is often advocated to discourage the use of response-adaptive randomization in practice. It is not an aim of this paper to get involved in these controversies: among the rest, there are still too many open methodological questions surrounding the use of adaptive designs and their inference, whether Bayesian or not. On the other hand, it would be unfair not to mention the existence of this debate. First of all the usefulness of adaptive randomized design is questioned: Korn and Freidlin [107] and Yuan and Yin [108] suggest that in the binary case outcome-adaptive randomization might not have substantial advantages over fixed-ratio when the response rate of the experimental treatment is not substantially higher than that of the standard treatment. Berry [109], however, disproves their conclusion. Korn and Freidlin [110] examine further the negative effects of response-adaptive randomization in some wellknown trials. Lee, Chen and Yin [111] give a balanced view, as the result of extended simulations. Another issue is whether adaptive randomization designs are ethical: Hey and Kimmelman [112] and Wathen and Thall [113] argue that the chance that adaptive random allocation will assign more patients to an inferior arm is too high. Other authors (Wason et al. [114]) worry that these designs may be too complex for use and that standard analysis methods of analyzing results from adaptive trials are valid only asymptotically. Another main concern is bias from temporal trends ( [115]). In conclusion, Thall, Fox and Wathen [116] state that adaptive randomization produces inferential problems that decrease potential benefit to future patients, and may decrease benefit to patients enrolled in the trial. These problems should be weighed against its putative ethical benefit. For randomized comparative trials to obtain confirmatory comparisons, designs with fixed randomization probabilities and group sequential decision rules appear to be preferable to adaptive randomization, scientifically and ethically. Whereas this controversy has been useful to learn how response-adaptive randomization might be used appropriately, for instance by introducing a burn-in period of equal randomization in some adaptive trials, the class of adaptive designs is really vast, so any discussion should focus around the value of a specific type of adaptive design in the specific context for which it is being proposed. Unfortunately, the arguments are often generic and not applied to the special features of adaptive procedures, as also Villar et al. [117] underline. In particular, as mentioned in Section 2 about the choice of the utility function, priorities about the very purpose or purposes of the experiment should be made clear in advance. Conclusions "Bayesian adaptive clinical trials: a dream for statisticians only?" asks Chevret [44]. Clearly, Bayesian adaptive experiments are not easy to design, let alone to implement. For a start, elicitation of a prior is not a simple matter. In clinical trials it is generally assumed to be based on historical data. In their book ([1]) Spiegelhalter, Abrams and Myles recommend attempting both an "enthusiastic" and a "skeptical" prior. On the other hand, Bayesian statistics exercises greater appeal than frequentist on most applied researchers, and the same can be said of adaptive design rules. This explains why the presence of Bayesian and adaptive design methods combined together has become massive in the biostatistical literature, notwithstanding the fact that adaptive algorithms are more complex than non-adaptive. It is this author's opinion that although there is a widespread consensus that the Bayesian and the adaptive approaches to design go very well together, the field is still rather fragmented. The development has taken place in a relatively short time and Bayesian adaptive designs are still awaiting in-depth investigation. It is a sad state of affairs that in general there is no sounder way to evaluate the performance of Bayesian (and non-Bayesian) designs other than by computer simulations. Often the simulation scenarios are chosen on the basis of the researchers' personal preferences, so the conclusions may be debatable. The book by Yin [118] is a thorough presentation of both Bayesian and frequentist adaptive methods in clinical trial design, but the two approaches are based on fundamentally different paradigms and a comparison of Bayesian and non-Bayesian designs is possible only in restricted cases. As an example, when several experimental treatments are available for testing, Wason and Trippa [119] compare Bayesian adaptive randomization, which allocates a greater proportion of future patients to treatments that have performed well, to multi-arm multi-stage designs, which use pre-specified stopping boundaries to determine whether experimental treatments should be dropped. The authors show that in this case both are efficient, but neither is superior: it depends on the true state of nature. In conclusion, it is worth quoting the words of Stallard et al. [120]: "Bayesian adaptive methods are often more bespoke than frequentist approaches . . . They require more design work than the use of a more standard frequentist method but can be advantageous in that design choices and their consequences are considered carefully". Obstructive sleep apnea syndrome (OSAS) in children and the contribution of orthodontist in the treatment Obstructive sleep apnea syndrome (OSAS) in children is a pathology that can lead to serious complications.It is therefore imperative to detect this disease as early as possible in order to be able to treat it.The narrowing of the airways responsible for OSAS is frequently the result of enlarged lymphoid tissue in children. Surgical removal of the adenoids and tonsils is the first line of treatment. However, this surgical treatment is not sufficient in young patients with maxillary hypodevelopment with narrow, arched palate, and mandibular retrusion or hyperdivergence.With two types of devices (mandibular propulsion orthosis, distractor), the orthodontist can effectively contribute to the treatment of pediatric OSAS. Introduction Obstructive Sleep Apnea Syndrome (OSAS) is a frequently occurring childhood malady, whose is incidence is largely underestimated. OSAS is thought to affect between 2 to 3.5% of children, with a peak incidence between three and five years [2] and a very slight male preponderance near puberty [15] . This syndrome is caused by intermittent, partial, or complete obstruction of the upper airways (pharynx and larynx) during sleep. This results in a decrease (hypopnea) or stop (apnea) of breathing for at least 10 seconds. When it is complete and prolonged, the obstruction of the airways causes breathing efforts which can lead to brief awakenings called "micro awakenings" which disrupt the quality of sleep [9] . OSAS causes serious complications without treatment: pulmonary dysfunction, heart failure, neurocognitive deficit with delayed learning, even disturbances in mood and attention and a break in the growth curve. It is therefore essential that patients with OSAS are identified and treated. Risk factors for children in the obstructive sleep apnea syndrome OSAS is associated with certain conditions such as asthma, ear, nose, and throat infections (recurring ear infections, rhinitis, sinusitis), obesity, drepanocytic anemia, and certain congenital afflictions such as trisomy 21, neuromuscular problems and maxillo-facial malformations [2,15] . In the case of major skeletal malformations, responsible for a three-dimensional narrowing of the middle or lower facial mass massive facial, severe OSAS is found, in particular for orofacial clefts. Much more discrete maxillofacial anomalies, far from these syndromic dysmorphoses, have been associated with OSAS. They can simply involve a decrease in space such as retrognathia (jaw retracted), brachygnathia (short jaw), endognathia (narrow jaw in the transverse direction) and mandibular hyper divergence characterized by a vertical lengthening of the mandible. They generate a descent and / or a retreat of the lingual mass, inserted at the level of the chin symphysis, which reduces the oropharynx. The dysmorphosis of these young patients is frequently accompanied by dental malocclusions, because on the one hand the insufficient development can cause a lack of space for the teeth or be accompanied by dental lags, and on the other hand the child adopts a lingual and mandibular posture of compensation which is at the origin of secondary anomalies. Symptomatology of OSAS Usually, OSAS is first discerned through assessment of clinical symptoms, a medical history, a morphological and functional evaluation of the upper airways, and finally by the indispensable polysomnographic examination, which, alone, can provide an irrefutable objective confirmation of the diagnosis. Nocturnal symptoms 3.1.1 Snoring Symptom most commonly associated with OSAS. It takes place when the airways are partially obstructed. When air passes through this limited space, it vibrates the soft tissues of the throat, uvula and soft palate. These vibrations create a sound that we call snoring. It is the first call sign, both for parents and for the medical team. In 96% of OSAS cases, this snoring is found, but it should be noted that simply noisy breathing, or a complete absence of snoring, may accompany certain confirmed OSAS [10] . Pauses in respiration (apnea) An apnea is defined as a total cessation of air flow for at least 10 seconds. In children, the pathological threshold is determined from one apnea per hour of sleep. OSAS is accompanied by frequent apneas, unknown to the patient, which end up worrying those around him. These apneas can cause sudden awakenings with sometimes a feeling of suffocation. Night sweats They are a good indication of increase of the blood level of carbon dioxide and bear witness to the hypoventilation that accompanies apnea. |
They can occur daily and be abundant. Nocturnal mouth breathing (which is sometimes occurs in the daytime) It is caused by an obstruction in the nasal air passageways. In healthy subjects, 70% of the inspired air passes through the nasal passages; in many children with apnea, mouth breathing will provide rescue breathing [3,8,15] . Abnormal sleeping positions By setting their heads in a hyper-extended position thus limiting lowering of the base of the tongue children can improve air flow [15,17] . Children also adopt this position when they nod off during long automobile trips; when they continue this behavior during waking moments it may become a chronic statural anomaly. We can also observe in children with apnea: a sleep is agitated with frequent position changes, shaking of the lower limbs, night terrors, recurrent nightmares, confusion in waking periods, bruxism, episodes of somnambulism or somniloquy (emission of sounds during sleep), repeated nocturnal arousals, nocturnal micturation or enuresis (involuntary and unconscious urination during sleep) , and, with nursing infants, nocturnal crying. Diurnal symptoms During the day, compared to children of the same age, we can observe: Diurnal behavioral disorders with pseudo-hyperactivity, an increase in psychiatric disorders [4] ; Difficulty concentrating and learning, disturbances in executive functions with reduced performance on intelligence tests, difficulties in school [15] . We thus note a risk of OSAS multiplied by 6 or 9 among the last 3 pupils of the class [1] ; Drowsiness, quite rare, affecting only about 20% of children with OSAS. Diagnostic Since none of the symptoms already seen are specific, the diagnosis of OSAS is made by recording the sleep. The polysomnography (PSG) The polysomnography makes registrations of different stages of sleep derived from electro-encephalographic graphs and an electro-oculogram that discerns specific phases of arousal and sleep, notably REM, or Rapid Eye Movement, and also records changes in leg placement from a position sensor with electromyograms along with a night-long electrocardiogram. We can also associate a microphone (snoring) or other sensors. This examination makes it possible to detect with great precision sleep breathing disorders and to define their type (central apnea, obstructive, mixed, hypopnea). It also makes it possible to determine the structure of sleep and to quantify the time spent in the different stages of sleep. The ventilation polygraphy (VP) The VP measures cardio-respiratory parameters including ventilation deficit, oxymetrics, and cardiac rate. It does not register brain activity, leg and eye movements. This examination, the sensitivity of which is slightly lower than polysomnography, however, makes it possible to confirm the diagnosis of OSAS when the clinical suspicion is strong. The oximetry It allows the oxygen level in the blood to be evaluated using a digital sensor. Alone is used less and less for the diagnosis of OSAS, given its significantly lower sensitivity and the inability to differentiate between obstructive and central apnea. Treatment of obstructive sleep apnea syndrome in children 5.1 Surgical treatment The treatment of choice for childhood OSAS is adenotonsillectomy, which effectively treats between 53 and 100% of cases. It will have the effects of lowering respiratory resistance, harmonizing facial growth by promoting nasal breathing, and improving behavioral and cognitive disorders if they are present. There is the problem of medium and long-term follow-up on the possibility of recurrence of OSAS. Indeed, some patients may have persistent ventilatory disorders after the operation (from 47% to 75% for the study by Tauman [14] on a population of obese children) or recurrence of symptoms a few years after being cured (for 14.5% of them according to Guilleminault et al [7] ). These results underline, on the one hand, the importance of an exploration of postoperative sleep and, on the other hand, the importance of cofactors such as obesity and craniofacial abnormalities. Orthodontic treatment Orthodontic devices would allow an expansion of the maxillary skeleton and an interesting mandibular anteriorization in young apneic patients with dysmorphosis [2] . They can be offered as second-line after failure of treatment with adeno-tonsillectomy, or as first-line for cases of moderate OSAS in the absence of obvious lymphoid organ hypertrophy. Beyond this mechanical approach, the restoration of spontaneous nasal ventilation is absolutely sought after treatment, whether surgical or orthopedic. Care should be taken, especially for young patients who may have developed the habit of ventilating through the mouth, even in the absence of any obstacles. To avoid recurrence, it may be worthwhile to offer real rehabilitation through ventilatory exercises. These exercises can be supported by small devices called "functional educator" [Fig. 1] which guide the tongue in a high position, keeping lips joined. Rapid maxillary disjunction The association between palatal morphology (arched palate) and oral ventilation is well known to orthodontists. Its explanation is based on the morphogenetic role of the tongue which, faced with chronic nasal obstruction, has adopted a low position over the long term to allow an oral ventilation supplement. The low position of the tongue no longer stimulates the maxillary sutures or counteracts the functional centripetal pressures exerted by the cheeks, which contributes to the narrowness and depth of the palate. When faced with a developmental insufficiency in the nasal stage, it is not uncommon to find associated narrowness of the dental arch, or even a lack of support for the cheekbones. These forms of narrow dental arches are often associated with dental crowding and an inversion of dental relations in the transverse direction (or linguoclusion); the maxillary arch should, under normal conditions, circumscribe the mandibular arch like the lid of a box. Rapid maxillary disjunction is an orthopedic treatment that seeks to separate the median intermaxillary and interpalatal sutures, which are still fibrous in children, to "catch up" to a deficit in sutural growth. The repercussions of rapid maxillary disjunction on ventilation have been described even before the identification of OSAS in children. It would widen the nasal passages and improve nasal ventilation. Secondly, creating space for the tongue would allow it to free up the oropharynx. Treatment can be started from the age of four to five, when all of the temporary teeth have erupted, and when the child can be cooperative. The MSE breaker (Maxillary Skeletal Expander) is an individual device, made from dental impressions processed in the laboratory, which is then sealed by the orthodontist on the patient's upper molars. The device consists of anchoring systems on the teeth [ Fig. 2] and a median cylinder which are connected by rigid arms. Once fixed in the mouth, the MSE breaker is activated daily, using a small wrench, from a quarter to half a millimeter, for 15 days to three weeks. The patient initially feels discomfort due to the bulkiness of the device, then tension with each turn of the key, but activation is not painful. An expansion of 5 to 8mm is thus obtained, depending on the initial transverse deficit. It is manifested by the opening of a large space (diastema) between the upper incisors. After the desired expansion has been achieved, the cylinder is blocked, allowing the spontaneous ossification of the disjointed suture, which stabilizes after three to six months. The median diastema usually closes spontaneously during this phase of contention. Mandibular advancement devices (activators) In adult OSAS cases, mandibular advancement devices (orthoses) advance the mandible and tongue during the night and thus reduce the risk of pharyngeal collapse during sleep. These same orthoses constitute a treatment route for pediatric OSAS [16] , provided the child has insufficient development of the mandible, adopting a position set back to the maxilla. They then seek to correct the rétrusion mandibular, anteriorizing the insertions of the tongue and normalizing dental reports. The mandibular advancement devices (or the activator, or Herbst rods [ Fig. 3]) is an individualized device, made by an orthodontist from dental impressions treated in the laboratory in resin or thermoplastic material. It is a removable device, worn at home and during sleep, and removed at mealtimes. Other therapeutic alternatives Therapeutic alternatives are proposed for patients who are not candidates for ODF, or in whom an obstruction persists: drug treatment (local anti-inflammatory drugs, or antihistamines), revision surgery (intervention on the basis of the tongue, etc…), management of obesity or replacement therapy, by setting up continuous positive airway pressure CPAP [12] . It is a small, portable "compressor" that fits next to the bed and delivers positive airway pressure through a hose and a nasal or face mask. This device induces an increase in the pressure inside the pharynx and thus prevents its obstruction during inspiration. This treatment makes it possible to normalize nocturnal breathing and suppress micro arousals in almost all patients suffering from obstructive apnea. Its undesirable effects remain significant and include among others: sensations of suffocation, skin lesions on the support areas of the mask, rhinitis, nasal and oral dryness and conjunctivitis in the event of an air leak in the eyes. The alteration in body image due to wearing a mask on the face is also not to be overlooked. Conclusion Collecting the symptoms of OSAS in children, which are sometimes very discreet, neglected or unrecognized by parents, is essential to lead to polysomnography, emphasizing the role of the clinician, and the quality of his anamnesis. The orthodontist is included in the development of the therapeutic plan after the diagnosis of OSA, and in the control of the restoration of physiological nasal ventilation. Platelet function, coagulation and fibrinolysis in patients with previous coronary and cerebrovascular ischemic events OBJECTIVES: Ischemic stroke (IS) or transient ischemic attack (TIA) history is present in 4-17% of patients with coronary artery disease (CAD). This subgroup of patients is at high risk for both ischemic and bleeding events. The aim of this study was to determine the role of platelet aggregability, coagulation and endogenous fibrinolysis in patients with CAD and previous IS or TIA. METHODS: A prospective case-control study that included 140 stable CAD patients divided into two groups: the CASE group (those with a previous IS/TIA, n=70) and the CONTROL group (those without a previous IS/TIA, n=70). Platelet aggregability (VerifyNow Aspirin® and VerifyNow P2Y12®), coagulation (fibrinogen and thromboelastography by Reorox®) and endogenous fibrinolysis (D dimer and plasminogen activator inhibitor-1) were evaluated. RESULTS: Patients in the CASE group presented significantly higher systolic blood pressure levels (135.84±16.09 vs 123.68±16.11, p<0.01), significantly more previous CABG (25.71% vs 10%, p=0.015) and significantly higher calcium channel blocker usage (42.86% vs 24.29%, p=0.02) than those in the control group. In the adjusted models, low triglyceride values, low hemoglobin values and higher systolic blood pressure were significantly associated with previous IS/TIA (CASE group). Most importantly, platelet aggregability, coagulation and fibrinolysis tests were not independently associated with previous cerebrovascular ischemic events (CASE group). CONCLUSION: Platelet aggregability, coagulation and endogenous fibrinolysis showed similar results among CAD patients with and without previous IS/TIA. Therefore, it remains necessary to identify other targets to explain the higher bleeding risk presented by these patients. This patient population has a higher prevalence of cardiovascular risk factors and a higher prevalence of established cardiovascular disease compared to patients without previous ischemic cerebrovascular events (IS/TIA). These characteristics imply a high risk for major cardiovascular events, including cardiovascular mortality (4)(5)(6)(7). On the other hand, patients with previous IS/TIA also have an increased incidence of major bleeding events, especially when subjected to the more potent modern antithrombotic regimens, which include some of the new antiplatelet drugs that are contraindicated in this population (8)(9)(10). Platelet aggregability, the coagulation system and endogenous fibrinolysis play key roles in the "ischemic-hemorrhagic" balance. However, little is known about the roles of these factors specifically in patients with prior cerebrovascular events. The purpose of this study was to evaluate whether patients with CAD and previous IS/TIA show any differences in terms of platelet aggregation, coagulation tests or endogenous fibrinolysis in comparison with patients without these complications, which could explain, at least in part, the increased bleeding risk in these individuals. ' METHODS Study design: This study was designed as a case-control study (with a 1:1 allocation ratio). Individuals in the case (those with a previous IS/TIA) and control groups were retrospectively selected and matched for sex, age, type of previous ACS (STEMI, NSTEMI and UA), and time between the ACS and inclusion in the study. Ethics approval was obtained from the local institutional review board. Inclusion criteria: The inclusion criteria included a prior ACS (over 12 months prior to inclusion), |
a history of IS/TIA prior to the diagnosis of ACS, chronic use of aspirin since the diagnosis of ACS, and signing of the inform consent form. Participants: The medical records of all patients who were diagnosed with ACS and had a history of IS/TIA, which were included in the coronary care unit and cardiac surgery service databanks, were analyzed. From January 2013 to April 2015, 918 records were evaluated; 412 did not meet the inclusion criteria (mainly because the IS/TIA occurred after the ACS diagnosis), 122 presented at least one of the exclusion criteria (mainly severe kidney dysfunction), and 314 were excluded for other reasons (mainly for the difficulty in finding a suitable matched control subject). As shown in Figure 1, 140 individuals were included in the present study, of whom 70 were in the case group and 70 were in the control group. Interventions: After the careful review of medical records and confirmation of inclusion and exclusion criteria, eligible patients were invited to participate in the study. After signing the informed consent form, patients underwent a clinical evaluation to confirm the data records and adherence to AAS. Laboratory tests: In addition to hematological (CBC) and metabolic (renal function, lipid and glucose profile) evaluations, laboratory tests directly related to the main purpose of this study were performed as follows: Platelet aggregability -platelet aggregation mediated by thromboxane A2 (TX A2) was assessed by VerifyNow Aspirin s , while platelet aggregation mediated by ADP was evaluated by VerifyNow P2Y12 s . The high residual platelet reactivity (HRPR) to AAS was defined by an ARU 4550 (11). Coagulationfibrinogen and thromboelastography (Reorox s ) were evaluated. Endogenous fibrinolysis -D dimer and plasminogen activator inhibitor-1 (PAI-1) were evaluated. Statistical analysis Data normality was tested by the Kolmogorov-Smirnov test. Continuous variables are expressed as the mean (+/-SD). Comparisons between the two groups were performed using Student's t-test for Gaussian variables and the Mann-Whitney U test for non-Gaussian variables. Categorical variables are expressed as the absolute number and relative frequency, and comparisons between the two groups were performed using the Chi-square test or Fisher's exact test, as indicated. Multivariable models In stepwise logistic regression analysis, previous IS/TIA was the dependent variable, and the baseline variables depicted in Table 1 were the independent variables. Each of the platelet, coagulation and endogenous fibrinolysis tests was separately included as an independent variable, thus producing a total of 6 adjusted models. SPSS version 20.0 software (IBM, USA) was utilized for the analyses, and a p-value o0.05 (2-tailed) was used to indicate statistically significant differences. With respect to missing data, we use a listwise deletion approach. Baseline characteristics As shown in Table 1, patients in the case and control groups were adequately matched for the prespecified variables. However, patients in the case group presented significantly higher systolic blood pressure levels (135.84 ± 16.09 vs 123.68 ± 16.11, po0.001), even though the use of antihypertensive medications was more common in this population (2.37 ± 1.09 vs 3.0 ± 1.23, p=0.006). Cardiologic risk factors in both groups did not show statistically significant differences. In the analysis of preexisting cardiovascular diseases, there were no significant differences between the two groups with regard to AF, PAD and previous PCI. However, the case group more frequently had a past history of CABG (25.71% vs 10%, p=0.015). All patients were taking aspirin, as this was mandatory for participation in the study. Few individuals were using a dosage higher than 100 mg/day. There were no statistically significant differences regarding previous medication usage, except for calcium channel blockers, which were more frequently used in the case group (42.86% vs 24.29%, p=0.02). The values for hematological and inflammatory variables were similar between patients in both groups. In relation to the metabolic profile, the lipid profile did not present any differences between the two groups. However, the case group presented higher creatinine levels (but similar values for the glomerular filtration rate) and higher fasting glucose levels (although similar glycosylated hemoglobin levels) compared to the control group. Platelet aggregability Platelet aggregation mediated by thromboxane A2, which was evaluated by VerifyNow s Aspirin, was not statistically significantly different between the groups. Furthermore, this same pattern was observed for the platelet aggregation mediated by ADP, which was evaluated by VerifyNow P2Y12 s . Upon analyzing the high residual platelet reactivity to AAS, no significant differences were again observed between the two groups ( Table 2). Coagulation The fibrinogen values of the case and control groups were not statistically significantly different, and a similar pattern was observed for the maximum clot firmness, which was evaluated by thromboelastography (Table 2). Endogenous fibrinolysis No statistically significant differences were observed in either the D dimer or PAI-1 levels ( Table 2). Table 3 shows the variables that were observed as being significantly associated with a previous IS/TIA in each of the adjusted models. The triglyceride values were associated with a previous IS/TIA in model 1 (including VerifyNow Aspirin), model 2 (including VerifyNow P2Y12) and model 5 (including PAI-1). Hemoglobin values were also significantly associated with a previous IS/TIA in 2 of the models (models 2 and 5). In model 4 (including maximum clot firmness by thromboelastography), systolic blood pressure was associated with a previous IS/TIA. As shown in Table 3, none of the variables in models 3 and 6 (which included fibrinogen and D-dimer, respectively) were associated with a previous IS/TIA. The most important finding was that none of the platelet aggregability, coagulation or fibrinolysis tests were independently associated with a previous cerebrovascular ischemic event. ' DISCUSSION The main finding of our study is that the higher bleeding risk related to patients with CAD and previous cerebrovascular ischemic events could not be explained by differences in platelet aggregability, coagulation or endogenous fibrinolysis in these patients. The ADAPT-DES study showed that HRPR to AAS (ARU4550) is a protective factor of major bleeding after drug-eluted stent implantation. In our patient population, HRPR to AAS was similar among patients with previous IS/ TIA and those without previous IS/TIA, which does not justify the difference in bleeding risk between these groups of individuals (12). An exacerbated response to clopidogrel (PRUo95) was associated with higher major bleeding rates in patients included in the ADAPT-DES study. In our study, there were no significant differences observed in the VerifyNow P2Y12 s results between patients in the IS/TIA and non-IS/TIA groups who did not utilize clopidogrel. Despite the fact that we did not analyze the eventual antiplatelet response to clopidogrel, the subanalysis of the POPULAR Trial did this, analyzing patients with and without previous IS/TIA, and the results indicated that there were no significant differences in the PRU after clopidogrel use between the groups (13). It is important to remember that many medications used in ACS treatment influence coagulation factors, which can lead to different bleeding outcomes. Moreover, there seems to be a stronger correlation between serum coagulation factor values and ischemic events than between serum coagulation factor values and bleeding events. At this moment, the only coagulation factor that is clearly associated with a higher incidence of bleeding outcomes is Factor XIII, which was not measured in our study (14). Although endogenous fibrinolysis is an opposite process to that of atherothrombosis, the correlation between serum values of fibrinolysis markers and ischemic or bleeding events is not clear. In this study, we observed no significant differences in the analyzed groups regarding PAI-1 and D dimer values. This finding suggests that fibrinolysis (at least when measured by the PAI-1 and D dimer values) does not justify the higher bleeding outcomes observed in patients with a previous IS/TIA (15). Three of the adjusted models used in our study suggested a relationship between low triglyceride levels and a previous IS/TIA. After corroborating our findings with those of the Rotterdam study, a strong correlation between low triglyceride levels and intracerebral hemorrhage was observed (16). In another study, Bonaventura and Cols observed a risk for hemorrhagic stroke in patients with low triglyceride levels (o0.94 mmol/L) that was twice as high as the risk for hemorrhagic stroke in those without low triglyceride levels (17). There are some possible explanations for these findings. First, some studies have suggested a positive relationship between high triglyceride levels, coagulation factors and blood viscosity, resulting in a possible prohemorrhagic state in patients with low triglyceride values (16,18). Second, cholesterol and fatty acids are essential elements of cell membranes. Experimental studies have suggested a relationship between low cholesterol levels, higher cellular membrane permeability (19) and endothelial weakness in small intracerebral arteries (20). However, the influence of triglyceride levels in this scenario has not been well studied. Hypertension is an important risk factor for ischemic events, including IS/TIA (21). However, hypertension is correlated with bleeding events, especially during antithrombotic treatment (22). In our study, patients with previous cerebrovascular ischemic events presented higher systolic arterial blood pressure values at baseline despite using more anti-hypertensive medications. These findings suggest a difficulty in controlling systolic hypertension among patients with a previous IS/TIA, which could be related to the higher risk for adverse events in this population. A low hemoglobin value is a strong predictor of major bleeding events in patients with ACS in the intrahospital phase and long-term follow-up phase (23)(24)(25)(26). In our study, low hemoglobin values were independently associated with a previous IS/TIA (individuals at a high risk for bleeding). In addition to being a possible marker for patients suffering from a previous hemorrhage (e.g., gastrointestinal bleedings), hemoglobin is directly associated with primary hemostasis via platelet adhesion, activation and aggregability (27)(28)(29)(30). In summary, the higher risk for bleeding demonstrated among patients with IS/TIA when subjected to more potent anti-thrombotic therapies could not be explained by the variables analyzed in this present study. Therefore, other pathophysiologic mechanisms should be responsible for these findings, as the presence of a dysfunctional hematoencephalic barrier has been demonstrated in individuals with previous cerebrovascular events (31)(32)(33). Our study has some limitations. First, it has a retrospective design. Therefore, despite having case-control matched for 4 different variables, other nonmatched variables could have influenced the results. However, because of the low incidence of a previous IS/TIA in patients with ACS, a prospective design would be unpractical. Second, the case group presented with mild sequelae (mean modified Rankin scale score=2), which may be justified by the worse outcome of patients with higher Rankin scores. Therefore, no conclusion can be drawn for patients with more severe sequelae post-IS/TIA. ' CONCLUSIONS Platelet aggregability, coagulation and endogenous fibrinolysis showed similar results among CAD patients with and without previous IS/TIA. Therefore, it remains necessary to identify other targets to explain the higher bleeding risk presented by these patients. ' ACKNOWLEDGMENTS This work was supported by the São Paulo Research Foundation, FAPESP. ' AUTHOR CONTRIBUTIONS Barbosa CJDG, Nicolau JC and Barreiros RS designed the study. Franci A, Arantes FBB, Furtado RHM, Strunz CMC, Rocha TRF and Baracioli LM acquired data, drafted the manuscript and approved its final version. Ramires JAF, Kalil-Filho R and Nicolau JC provided substantial contributions to the study conception/design, revised the intellectual content of the manuscript and approved its final version. Electrochemical Platform Based on Modified Ti Electrodes to Test a Food Allergen Presence Celiac disease (CD) is an autoimmune disorder. For these patients, the only treatment is a gluten free diet. In this work we showed that there can be an easy and cheap method to test gluten presence. We used several modified Ti electrodes. All modified electrodes were prepared using electrochemical methods. Modified electrodes were tested using differential pulse voltammetry in solutions with or without gluten. Selected modified Ti electrode was tested using different known gluten concentrations and extracts from different aliments. electrode (18). DPV was used with reduced graphene oxide modified electrodes with antigliadin antibody. At least 4 hours are needed to prepare the modified electrode (19). A research study was using both impedance and DPV. They were using modified electrodes with antibody based on the use of dithiols (20). A simpler method was using a graphite mine and rapid differential pulse voltammetry (5). In the present research, we used modified Titanium (Ti) electrodes and DPV. We used different surface modifications: nanotubes obtained by anodization, nanofibers and graphene oxide. We tested different food products with the modified electrodes. First, Ti round samples, with 8 mm diameter (99.99% purity, Alpha Aesar) were polished and cleaned with ultrasounds. Distiled water, |
ethanol and acetone were used (15 minutes each). Electrode was named E1. The nanotubes (TiNT) were prepared at room temperature in a viscous electrolyte: EG: 0.5 wt% NH4F + 2 vol.% H2O at 50 V, 2 h. E1 was used as working electrode and a Pt wire as counter electrode. A Matrix MPS-7163 was used as power supply. After preparation, they were rinsed with distilled water, and subjected to ultrasonic treatment 30 seconds in distilled water. Electrode was named E2. TiO2 nanofibers (TiNF) were prepared at high voltages using a previous described method (21) with some modifications. Briefly, electrolyte was: 4 wt.% TBut + 8 wt.% PVP in DMF/isopropanol (mass ratio 1/1) + 2 wt.% acetic acid. Prior to electrodeposition, the samples were coated with 100 μl solution using a spin-coater Laurell WS-650-23 (Laurell Technologies). The dynamic mode was used, with 3000 rpm, 30 seconds. For nanofibers formation, solution was fed at a rate of 0.5 ml/h using a Legato180 pump (Kd Scientific) from a plastic syringe with a blunt tip needle. 20 kV were applied between the needle tip and Cu collector plate (at 10 cm distance) for 30 min. A high-power supply was used -PS/EJ30P20 (Glasmann). Samples were dried overnight and treated 4 h at 450°C. E1 was used as working electrode. Resulted electrode was named E3. Gluten presence evaluation was made using DPV between 0 V and 1.2 V, with 50 mV amplitude at 20 mV scan rate. Electrolyte used was DMSO:acetate buffer (pH 4.8) 1:4. Acetate buffer was prepared from natrium acetate and glacial acetic acid; with or without gluten. Gluten extraction: flour samples (1g) were mixed with distilled water (10 mL) for 1 h under ultrasound stirring. After centrifugation at 8000 rpm for 10 min a pellet was formed. This pellet was immersed into ethanol (60%, 50 mL) for 1 h, centrifuged (30 min, 8000 rpm). 5 mL of the supernatant was used for analysis dissolved in 25 mL buffer pH 4.8 Equipment: scanning electron microscopy (SEM) images for all samples were recorded with Thermo Scientific FEI Quanta 650 FEG (Hillsboro, OR, USA) high-performance scanning electron microscope. All electrochemical data were recorded with a potentiostat/galvanostat from Metrohm Autolab (PGSTAT 302N). Electrochemical cell was of three electrodes in a single compartment. Results and discussions Recorded curves during sample preparation and their corresponding SEM images for modified electrodes are presented in Figure 1. For GO deposition on Ti using CV (30 cycles) the recorded curve is seen in Figure 1 a. It is a typical CV curve for a redox process. In the SEM image (Figure 1b) it can be seen the graphene agglomerations. For sample Ti/NT, the curve recorded during anodization is presented in Figure 1 c. It is typical anodization curve with revealing the mechanism of nanotubes formation previously described (22). In the SEM image (Figure 1d), very ordered nanotubes are visible, having diameters around 100 nm. Using high voltage process, random aligned TiO2 nanofibers were prepared (Figure 1 e). The CV curve for GO deposition on top of nanotubes (TiNTGO), the currents recorded during CV (Figure 1 f) were higher compared with the ones recorded in TiGO case (Figure 1a). It is visible that the current was increasing with each cycle. In the SEM image (Figure 1 g), it is visible that the nanotubes were covered with a film. Their diameters are smaller compared with Ti/NT sample. Most of the nanotubes remained open after GO deposition. The Ti surface modification influence is visible once again in the curve recorded during GO electrodeposition on top of nanofibers ( Figure 1h). This curve is very different compared with the ones recorded for TiGO30 sample and TiNTGO sample. GO electrodeposition on top of nanofibers is clearly seen in Figure 1 i. Nanofibers diameters were changed after GO electodeposition ( Figure 1 e and i). To see which of the proposed electrodes it is desirable to evaluate the presence of gluten, we tested six electrodes using DPV. One electrode was E1 (polished and cleaned Ti) and the others were modified Ti electrodes. From the Figure 2, Figure 3. Only five concentrations were presented. At sixth concentration, probably Ti was oxidized and the signal was not proportional with the gluten concentration anymore. Gluten concentration was plotted against peak height. We calculated the limit of detection (LOD) and the limit of quantification (LOQ) in each case, using relations described in literature (23), with Excel program. In first case we obtained LOD 36 ppm and LOQ 119 ppm. In the second case, results were: LOD 66 ppm and LOQ 220 ppm. It can be seen that increasing the cycles number during GO electrodeposition did not increased the GO film conductivity, so the results were not improved. To see if the E4 30 TiGO electrode can be used to analyze real samples, we prepared three kind of flour samples extracts. From Figure 4, it can be seen that with this modified Ti electrode, we were able to distinguish between flour samples. The higher signal was obtained for wheat flour. This is known to have gluten. A smaller signal compared with the one obtained in DMSO:acetate buffer with 66 ppm gluten was obtained for corn flour , labeled by manufacturer "may contain traces of gluten". A signal compared with the one obtained in DMSO:acetate buffer without gluten was obtained for flour labeled by manufacturer "gluten free". Figure 4. Flour samples testing with E4 30 TiGO To verify the reproducibility and repeatability, we freshly prepared E4 30 TiGO at different times. We recorded the signal in DMSO:acetate buffer with 100 ppm gluten. From results presented in Figure 5 a. The result is similar. We also performed five consecutive scans in DMSO:acetate buffer with 100 ppm gluten. We obtained a good reproducibility (Figure 5b). Conclusions Different nanostructures (nanotubes, nanochannels, nanofibers) were prepared for a potential gluten detection application. GO was also used as titanium or nanostructured titanium surface modification method. Their structure was analyzed using SEM. Modified electrodes were tested for gluten presence evaluation using an electrochemical method (DPV). Ti modified with graphene oxide and nanofibers gave a signal in DPV in solutions containing https://doi.org/10. 37358/Rev. Chim.1949 Rev. Chim., 71 (5) known gluten concentrations. Increasing graphene oxide layer thickness (by increasing cycles number during electrodeposition), does not increased the conductivity. We tested different kinds of flour with Ti modified with graphene oxide: wheat flour (known to have gluten), corn flour (labeled "may contain traces of gluten") and a mixed fluor labeled "gluten free". We were able to distinguish between them. Future experiments need to be performed to try to lowed LOD and LOQ, but up to this point, the proposed electrode (Ti with graphene oxide) and proposed method (DPV) can be used to detect gluten from aliments which were contaminated with gluten during manufacturing process. Amoxycillin and clavulanic acid induced Stevens-Johnson syndrome: A case report Stevens-Johnson syndrome (SJS) is an immune mediated hypersensitivity reaction. Significant involvement of oral, nasal, eye, vaginal, urethral, GI and lower respiratory tract mucous membrane may develop. It is usually a reaction due to a medication or due to an infection. In 95 % of case reports, drugs were found to be an important cause for the development of SJS. In this case report, a 32 year old female reported chief complaint of itch skin eruptions all over the body along with erosive lesions on tongue, lips, buccal mucosa and genital mucosa. The reaction occurred after administration of augmentin (containing amoxycillin and clavulanic acid). She was treated with antimicrobials, antiallergics and conservative management. The patient improved and was discharged from the hospital. Causality assessment using Naranjo Adverse Drug Reaction Probability Scale revealed that amoxycillin and clavulanic acid combination was a possible cause for the adverse reaction with a score of 4. INTRODUCTION Stevens-Johnson syndrome (SJS) is a rare but very serious disorder of skin and mucous membranes. It is usually a reaction due to a medication or due to an infection. It is considered as an acute life-threatening condition and a medical emergency and requires hospitalization. In SJS recovery takes weeks to months, depending on the severity of the patient's condition (Mayo Clinic, 2014). The proportion of females has been estimated to be 33-62 %. The largest series reports 39.9 % of females in a group of 315 patients with Stevens-Johnson syndrome. In a large cohort, the mean age of patients with Stevens-Johnson syndrome was 25 years. In a smaller series, the mean age of patients with Stevens-Johnson syndrome was reported as 47 years. However, cases have been reported in children as young as 3 months and adults as old as 78 years (Foster, 2016). SJS may present as a nonspecific febrile illness leading to malaise, headache, cough, rhinorrhea with polymorphic lesions of the skin and mucous membrane characterized by acute blisters and erosions (French, 2006). Various etiologic factors have been implicated as causes of Stevens-Johnson syndrome. Drugs most commonly are blamed. The four etiologic categories are infectious, drug induced, malignancy-related and idiopathic. Stevens-Johnson syndrome is idiopathic in 25-50 % of cases. Drugs and malignancies are most often implicated as the etiology in adults and elderly persons. Pediatric cases are related more often to infections (Foster, 2016). CASE REPORT A 32 year old female came to our hospital in emergency department with itchy skin eruptions all over the body. She gave a history of sore throat for which she was administered augmentin (containing amoxycillin and clavulanic acid). The eruptions were seen after taking first dose which she took at night. The eruptions were erythematous, hyper pigmented, target-like, round lesions. The eruptions were 2-3 cm in diameter and present throughout the body, more concentrated on the upper and lower limbs, upper chest, abdomen, face, palms and soles. On examination, erosive lesions were present on the lips, buccal mucosa and the tongue. Painful erosions were also seen on the genital mucosa. Redness of eyes and blurring of vision was also noted. A diagnosis of Stevens-Johnson syndrome was made. Subsequently, multiple vesicles and bullae with antral necrosis developed in the area of the lesions. Blistering of the lesions was noted. Swelling of the face and lips were noted. A slough and whitish plaques were observed over the tongue. The lesions later crusted on the skin and the oral cavity. Bleeding time and clotting time were in the normal limits. Absolute eosinophil count was normal. Blood urea and serum creatinine levels were normal. Liver function tests were normal. Neutrophils were increased. The following drugs were administered: -Injection piperacillin/tazobactam 4.5 g i.v. three times a day -Injection linezolid 600 mg i.v. two times a day -Injection albumin 50 ml i.v. once a day for three days -Injection pheniramine 8 mg i.v. twice a day -Tablet fluconazole 150 mg once weekly -Tablet loratidine 10 mg at bed time daily -Tablet folic acid 5 mg once daily -Syrup paracetamol two teaspoons full three times a day -Syrup K-lyte (potassium bicarbonate potassium citrate) two teaspoon full three times a day -Potassium permanganate mouth wash -Polymyxin B ointment -Fusidic cream -Nystatin drops -Saline wash. During her stay in ward swelling was noted in her both legs and diagnosis of DVT was made through color Doppler. Following drugs were given for DVT: -Tablet Rivaroxaban 15 mg twice a day used for seven days, then shifted to Tablet warfarin 10 mg once a day -Injection Enoxaparin sodium 60 mg subcutaneously twice a day. The patient improved with the above treatment and was subsequently discharged from the hospital with the advice not to be administered beta-lactam antimicrobials in the future. DISCUSSION In 1922, Stevens and Johnson described 2 male patients of 7 and 8 years old, who developed extraordinary generalized eruption with fever and inflamed buccal mucosa (Barvaliya et al., 2011;Stevens and Johnson, 1922). SJS can be differentiated from other skin conditions on three clinical criteria, (i) the pattern of individual skin lesions, (ii) distribution of lesions, and (iii) extent of epidermal detachment. The characteristic findings in SJS are widespread erythematous or purpuric macules which form flat atypical target lesions as the disease progresses to cause full thickness epithelial necrosis (French, 2006). In the oral cavity, SJS causes widespread ulcerative lesions. A prodrome occurs in about 30 % of cases and may begin within 1 to 3 weeks of starting a new drug and lasts 1 to 2 weeks before the |
onset of mucocutaneous manifestations, presenting with flu-like symptoms, sore throat, headache, arthralgias, myalgias, fever, bullous and other rashes, pneumonia, nephritis or myocarditis (Farthing et al., 2005). Balanitis, urethritis and vulval ulcers may occur. Our patient did not report any prodrome, but skin, mouth and genital ulcerations were present. Drug-induced SJS is characterized by mucosal erosions plus widespread distribution of atypical targets or purpuric macules and epithelial detachment involving less than 10 % BSA on the trunk, face and extremities (Ayangco and Rogers, 2003). SJS has to be clinically differentiated from viral stomatitides, pemphigus, EM, TEN and the sub-epithelial immune blistering disorders like pemphigoid. There are no specific diagnostic tests for SJS (Farthing et al., 2005). Our case showed ulceration of oral cavity, involvement of eye with redness, ulceration of genital region along with numerous healed lesions on chest, abdomen and limbs which showed typical appearance of "target lesions". The lesions were widespread as compared to EM, which is localized. Adverse drug reactions (ADRs) are one of the leading causes of death among hospitalized patients and occur in 0.3 to 7 per cent of all hospital admissions. These may vary from mild rashes to severe reactions such as Stevens-Johnson syndrome (Doshi et al., 2012). Hällgren et al. (2003) stated that antibiotics are the most common cause of Stevens-Johnson syndrome, followed by analgesics, cough and cold medication, NSAIDs, psychoepileptics and antigout drugs. Of antibiotics, penicillins and sulfa drugs are prominent; ciprofloxacin has also been reported (Hällgren et al., 2003). Patel et al. (2012) reported in their study that antimicrobials were the most commonly suspected drugs (45 %) causing SJS as has also been reported in Australia (Sanmarkan et al., 2011). In a study conducted on 225 references in India, 10 references were included as per selection criteria. The major causative drugs were antimicrobials (37.27 %), anti-epileptics (35.73 %) and nonsteroidal anti-inflammatory drugs (15.93 %), Carbamazepine (18.25 %), phenytoin (13.37 %), fluoroquinolones (8.48 %) and paracetamol (6.17 %) (Patel et al., 2013). Coamoxiclav is a generally well tolerated antimicrobial. Its most frequently reported adverse effects are gastrointestinal adverse reactions and hepatotoxicity (Salvo et al., 2007). It has been reported in few publications as etiology of SJS in adults (Abou-Elhamd, 2009). Our patient did not report any of these. Amoxycillin and clavulanic acid combination therapy was identified as the causative agent because of the temporal relationship between the administration of the combination and the beginning of the eruptions. There have also been several other previous reports linking amoxycillin and clavulanic acid to Stevens-Johnson syndrome. According to Naranjo Adverse Drug Reaction Probability Scale, amoxycillin and clavulanic acid induced SJS was possible in our patient (a score of 4). The first step in the management was an immediate withdrawal of the offending agent followed by supportive care. Garcia-Doval et al. (2000) report that earlier the drug is withdrawn, better the prognosis while exposure to drugs with longer half-lives increases the risk of death. Supportive care must include the management of fluid and electrolyte requirements (Garcia-Doval et al., 2000). Adjuvant treatments such as corticosteroids and immunosuppressants may also be used in severe cases of SJS (Gerull et al., 2011). Financial interests None declared. Conflicts of interest The authors declare that they have no conflict of interest. CONCLUSION This case report reports the fact that severe hypersensitivity reactions can occur with amoxycillin and clavulanic acid, which can be possibly dangerous and life-threatening. Therefore, clinicians must be more cautious while prescribing this drug. Affected patients and their first-degree relatives should be educated regarding the adverse effects of beta lactam antimicrobials and instructed to avoid them in future. Bacterial Profile and Antibiotic Resistance in Patients with Diabetic Foot Ulcer in Guangzhou, Southern China: Focus on the Differences among Different Wagner's Grades, IDSA/IWGDF Grades, and Ulcer Types Objective To understand the bacterial profile and antibiotic resistance patterns in diabetic foot infection (DFI) in different Wagner's grades, IDSA/IWGDF grades, and different ulcer types in Guangzhou, in order to provide more detailed suggestion to the clinician about the empirical antibiotic choice. Methods 207 bacteria were collected from 117 DFIs in Sun Yat-sen Memorial Hospital from Jan.1, 2010, to Dec.31, 2015. The clinical data and microbial information were analyzed. Results The proportion of Gram-negative bacteria (GNB) was higher than Gram-positive bacteria (GPB) (54.1% versus 45.9%), in which Enterobacteriaceae (73.2%) and Staphylococcus (65.2%) were predominant, respectively. With an increasing of Wagner's grades and IDSA/IWGDF grades, the proportion of GNB bacterial infection, especially Pseudomonas, was increased. Neuro-ischemic ulcer (N-IFU) was more susceptible to GNB infection. Furthermore, with the aggravation of the wound and infection, the antibiotic resistance rates were obviously increased. GPB isolated in ischemic foot ulcer (IFU) showed more resistance than the N-IFU, while GNB isolates were on the opposite. Conclusions Different bacterial profiles and antibiotic sensitivity were found in different DFU grades and types. Clinician should try to stay updated in antibiotic resistance pattern of common pathogens in their area. This paper provided them the detailed information in this region. Introduction Diabetes is a metabolic syndrome characterized by hyperglycemia, which has become a heavy burden to China [1]. Deregulated metabolism in diabetics is link to many complications, including neuropathy, retinopathy, nephropathy, atherosclerosis, and foot ulcers [2]. Diabetic foot ulcer (DFU) is an outcome of complicated amalgam of several risk factors such as peripheral vascular disease, peripheral neuropathy, trauma, and impaired resistance to infection [3], and continues to be a major reason for lower extremity amputation worldwide [4]. Diabetic foot infection (DFI) was considered as one of the most frequent and disastrous complications of diabetes. As reported, 60% of DFU are infected on presentation [4], which can increase the risk of a lower extremity amputation by 50% compared to the DFUs without infection [5,6]. Because the diabetics' infection can worsen quickly, clinician must pursue the diagnosis aggressively [7] to select an initial antibiotic regimen for the likely pathogens, which need more microbiological information about the DFUs before the wound cultures and antibiotic sensitivity test. Thus, there is an urgent need for the bacterial profile and antibiotic resistance suggestion in more details to give their empirical antibiotic selection "a best guess." There were several researches reported that acute DFI is usually caused by aerobic Gram-positive cocci, but deep or chronic wounds often harbor aerobic Gram-negative and obligate anaerobic bacteria, often polymicrobial flora [8][9][10][11], while few studies investigated the differences of bacterial profiles in different DFIs in more detail. According to the patients' clinical features acquired at the "first sight" of clinician, including the patient as a whole (e.g., cognitive, metabolic, and fluid status), the affected foot or limb (e.g., the presence of neuropathy and vascular insufficiency) and the infected wound [12], different classification systems are used to assess the severity of DF, the most often used of which were the Wagner-Meggit classification system that takes into consideration the depth of ulcer, presence of gangrene, and level of tissue necrosis [13] and IDSA/IWGDF classification system for defining the presence and severity of an infection of DF [7]. Besides, DF can be classified into three types according to whether with or without peripheral arterial or nerve diseases [9], named ischemic foot ulcer (IFU), neuropathic foot ulcer (NFU), and neuro-ischemic foot ulcer (N-IFU), respectively. More detailed information about pathogens and antibiotic resistance according to different DFU grades and types presents further practical significance for suggesting a more specific antibiotic choice. On the other hand, to better provide optimal antimicrobial therapy, clinician should be familiar with the common microbial isolates and antibiotic resistance patterns in their own region of practice. Many studies from different regions showed different bacterial profiles in DFIs, especially in warm climate in Asia and Africa [3]. As the main metropolises with a large population and a typical subtropical climate in Southern China, Guangzhou may have a unique bacterial profile and antibiotic resistance in patients with diabetic foot ulcer, while rarely studied. With the aim of understanding the bacterial profile and antibiotic resistance patterns in DFUs in Guangzhou, furthermore in different Wagner's grades, IDSA/IWGDF grades, and different ulcer types, 117 DFI patients and 207 bacterial isolates were collected from Sun Yat-sen Memorial Hospital from Jan. 1, 2010, to Dec. 31, 2015. The clinical data and microbial information were compared among the different DFUs' grades and types. This knowledge will provide more practical advice about antibiotic agent choice to the clinicians. [7,14]. Briefly, clinical severity of ulcer was assessed by Wagner-Meggit classification system [13] and severity of infection was quantified according to the IDSA/IWGDF classification system [7] as previous description. The patients were classified into four Wagner's grades and three IDSA/IWGDF grades based on these systems (Table 1). The diagnosis of peripheral sensory neuropathy was based on failure to appreciate a 10 g Semmes-Weinstein monofilament test and on vibration detection test performed with a 128 Hz tuning fork, and the peripheral arterial disease was diagnosed by limb arterial color Doppler investigation. According to whether with peripheral arterial disease or peripheral sensory neuropathy, the diabetic foot ulcer (DFU) can be classified into ischemic foot ulcer (IFU), neuropathic foot ulcer (NFU), and neuro-ischemic foot ulcer (N-IFU). The DFUs only with peripheral arterial disease were defined as IFU, the IFUs together with peripheral sensory neuropathy were defined as N-IFU, and the DFUs with peripheral sensory neuropathy but without peripheral arterial disease were defined as NFU. Specimen Collection and Microbiological Culturing. All the specimens were sampled to the microbiology laboratory within 48 h after hospital admission. Swabbing were collected from each wound after the wound had been cleansed (using 0.9% sterile saline and gauze) and debrided (removal of necrotic tissue, foreign material, calluses, and undermined wound edges) [15]. No antimicrobial agent or antiseptic was introduced into the wound before specimen collection. Each wound was swabbed by rotation of a wound swab over a 1 cm 2 area of the wound for 5 seconds, using sufficient pressure to extract fluid from the inner part of the wound [16]. The specimens were placed into sterile transport containers and sent to the microbiology laboratory for aerobic culturing within 30 minutes. Anaerobic culturing was not performed in this study. Totally, 207 isolates were collected from the 117 patients. To avoid sample duplication, isolates that were consecutively isolated from the same individual were excluded. All isolates underwent phenotypic identification using the VITEK® 2 microbial identification system (bioMérieux, Marcy l'Etoile, France) according to the manufacturer's instructions. Susceptibilities were determined using the disk diffusion method in accordance with the performance standards for antimicrobial susceptibility testing, recommended by the Clinical and Laboratory Standards Institute. Enterobacter cloacae ATCC 700323 and Staphylococcus saprophyticus ATCC BAA-750 were used as the quality control strain for phenotypic identification. Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, and Staphylococcus aureus ATCC 29213 were used as the quality control strain for antibiotic sensitivity test. Statistical Analysis. In descriptive statistics, the frequency of isolate distribution and antibiotic resistance was treated as categorical variables. The chi-square or two-sided Fisher's exact test was used to discriminate whether the distributions were significantly different between different groups. The distributed variables were expressed as the mean ± standard deviation and compared by one-way ANOVA. Variables without normal distribution were expressed as the media (interquartile range) and compared by Kruskal-Wallis H test. It was considered statistically significant if the two-side P value < 0.05. All statistical analyses were carried out using SPSS 19.0 for Windows (IBM). All susceptibility data and molecular test results were analyzed using WHONET software, version 5.6. Characteristics of Patients and Wounds. Totally, 95.8% DF patients suffered from DFU (388/405), 30.2% (117/388) of which were clinically infected. Additionally, the DFIs in this study were mainly classified in the moderate or severe grades (Wagner's 2~4 grades and IDSA/IWGDF 2~3 grades), rarely in the mild stage, and only 5 patients were NFU (4.3%). All the patients enrolled were type 2 diabetes ones. The percentage of newly diagnosed DFUs was 26.4%, mainly in Wagner's grades 2 and 4 (32.4% and 37.5%), IDSA/IWGDF grade 2 (38.8%), and N-IFUs (33.8%). With an increasing Wagner's grades and IDSA/IWGDF grades, the serum Creaction protein (CRP) and procalcitonin (PCT) level had an increased trend (P < 0 05). There were no significant differences in the majority of the clinical characteristics examined (Table 1). With an increasing of Wagner's grades and IDSA/ IWGDF grades, the proportion of Gram-negative bacterial infection was obviously increased (Figure 2). Staphylococcus aureus and |
Enterococcus were the main Gram-positive bacteria isolated in every Wagner's grades and IWGDF/IDSA grade DFIs, while there was some differences about the Gram-negative isolates in different grade DFIs. Enterobacteriaceae, mainly including Escherichia coli, Enterobacter cloacae, and Klebsiella pneumonia, were the main Gramnegative bacteria isolates in the mild DFIs (Wagner's grade 1 and IWGDF grade 2), and Proteus appeared in the moderate wounds (Wagner's grade 2~3 and IWGDF 2~3). Furthermore, Pseudomonas and Acinetobacter raised to another two Table 2 and Figure 2. More than a half of the DFIs in this study were polymicrobial (59.8%, 70/117), with aerobic Gram-positive cocci (GPC), and especially staphylococci, the most common causative organisms. Especially in the IWGDF grade 2, Wagner's grade 2/4 DFUs, and N-IFU patients ( Table 1). All the fungal infections (n = 23) were polymicrobial with bacteria. standard definition for acquired resistance in 2012 [17]. Totally, 22 MRSA were detected in the 207 isolates, distributed in different grades and types. One CRE (carbapenemresistant Enterobacteriaceae) were isolated from a 49-yearold male patient who was diagnosed DFU four years ago, given systemic and local antibiotic therapy for several times during his three hospitalization periods and outside hospital, whose wound was classified to Wagner grade 3, IWGDF grade 4, and N-IFU when the carbapenem-resistant Escherichia coli was isolated. No VRE (vancomycin resistant Enterococcus), PDRAB (pandrug-resistant Acinetobacter baumanii), and PDRPA (pandrug-resistant Pseudomonas aeruginosa) were detected. As the main pathogens of DFI, the antibiotic sensitivity information of Staphylococcus aureus and Enterobacteriaceae was analyzed. Different antibiotic resistance patterns were shown in different wound grades and types. According to the resistance rates of 33 antibiotic agents of the two major pathogens above, we defined the regimens whose resistance rate was <30% as "potential empirical regimens" and the ones whose resistance rate > 70% as "alarming empirical regimens" in every grades and types. Details showed in Table 3. Discussion To our knowledge, this is the first prospective study on microbiological profile and antibiotic resistance pattern of the diabetic foot infection based on the different classification systems, in order to give the clinicians more suggestions in details for initial empirical antibiotic selection according to the comprehensive assessment of the patients. DFU continues to be a major reason for lower extremity amputation worldwide [4], about half of which are clinically infected at presentation [18]. In our study, 95.8% DF patients suffered from DFU, 30.2% of which were clinically infected, and mainly the chronic ulcer with infection. The polymicrobial infection, including polybacterial infection and bacteriafungus infection, accounted for 59.8% of the DFIs in this study, which coincide with the previous reports [12,19]. As the other studies, Staphylococcus is the predominant Grampositive bacteria, including Staphylococcus aureus and CN-S (Coagulase-negative staphylococcus). Compared with the Gram-positive bacteria, there were more species of Gramnegative bacteria infected by DFIs. From the general and species of the bacteria, Proteus and Pseudomonas aeruginosa were the predominant pathogens in Gram-negative bacteria, followed by Klebsiella pneumonia, different from some reports in which the dominating Gram-negative flora was Escherichia coli in other areas [20], may due to the warm climates in Guangzhou. However, the predominant flora was Enterobacteriaceae. Coinciding with some studies which showed that the Gram-negative organisms were the most frequent isolates in DFIs in warm climates, especially in Southeast Asia and Africa [21,22], the prevalence of Gramnegative was some higher than the positive aerobes in this study, as Guangzhou has a warm and humid climates. To the DFIs, selection of an initial antibiotic regimen is usually empirical, so the likely pathogens and their antibiotic sensitivity often are "guessed" by the clinician before the microorganism cultivation and sensitivity tests. Therefore, a detailed bacterial profile and antibiotic resistance pattern associated with the different severity and types of DFIs is urgently needed for the clinicians. Actually, the severity of the DFU and infection is first determined by the clinical classification scheme. Various classification systems have been proposed to assess the severity of diabetic foot lesion that attempt to encompass different characteristics of ulcer including ulcer size, depth, ischemia, infection, and neuropathy [3]. Wagner-Meggit classification system is the most widely used classification system [23] but cannot help to take into consideration about ischemia and infection. Another classification system given by the Infectious Disease Society of America (IDSA) and International Working Group on the Diabetic Foot (IWGDF) can define the presence and severity of an infection of the diabetic foot, named IWGDF/ IDSA classification [7]. Besides, clinician used to classify the DF to IFU (ischemic foot ulcer), NFU (neuropathic foot ulcer), and N-IFU (neuro-ischemic foot ulcer) according to more detailed vessel and nerve check. In this study, different bacterial profiles and antibiotic sensitivity were found in different Wagner's grade, IWGDF grade, and DFU types. With the aggravation of the wound and infection, the Gram-negative bacterial species harbored and increased, especially the proportion of Pseudomonas, a common nosocomial infection bacteria, resistant to many kinds of antibiotics, which coincided with the previous study [24]. The polymicrobial infection distributed mainly in moderate wound (IWGDF grade 2 and Wagner's grade 2 DFUs) and N-IFU patients, which was beyond our expectation that the severe wound and infection may tend more polymicrobial. Combined with the clinical characteristics, the patients in moderate wound and N-IFUs in our study had more newly diagnosed rate compared with the other groups, who had not received systemic antibiotic treatments, which may cause the results above. the likely pathogens. In details, this study gave the clinician suggestions about the most possible regimens as the "potential antibiotics," and the regimens should not be used for their high resistance as the "alarming antibiotics" according to different wounds. When the patients were evaluated by Wagner-Meggit classification system and IWGDF/IDSA classification system, and the ulcers were typed as IFU, NFU, or N-IFU, clinicians can choose the overlapping of different systems according to Table 3. For example, if the wound of a DFI patient was graded as Wagner-Meggit grade 2 and IWGDF/IDSA grade 3 and was diagnosed as an ischemic foot ulcer (IFU), combined with the bacterial profile and antibiotic resistance, the clinician can try cefotetan, β-L-ase, carbapenem, fluoroquinolone, or aminoglycosides as the empirical antibiotics to cover the main possible pathogens and avoid penicillin, ampicillin, the first to third generation cephalosporin and tetracycline in order to prevent the infective treatment and MDR bacteria due to antibiotic abuse. If the DFU patient was classified in more severe IWGDF grades, less potential empirical regimens could be chosen and more should be avoided, then modified the regimens according to the available clinical and microbiological information. However, this paper only provided the empirical regimens selected suggestion about the predominant GNB and GPB, while did not cover all the pathogens. Actually, some other pathogens showed higher resistance rates to more antibiotics due to their natural resistance, for example, the Pseudomonas aeruginosa, Enterococcus faecium, and Stenotrophomonas maltophilia. Therefore, more attention should be paid to the DFIs with high risk of the natural resistance pathogens above, like the Wagner's grade 4 and IWGDF/IDSA grade 4 wound. The major limitation of this study is the lack of anaerobic culturing. Further study is required to evaluate the anaerobic distribution and drug sensitivity in the different grades of DFUs. Another limitation is the small number of included patients, especially those with Wagner-Meggit grade 1 or IWGDF/IDSA grade 1 wound, and rarely neuropathic ulcerations. Tissue biopsy is known as the most standard method, and swab cultures are considered as not reliable since it generally includes the colonizers and not the causative pathogen [15]. But in this study, the swabs were obtained after a complete debridement in order to avoid the colonizers, and the CNS, as the main colonized organisms in the skin, were detected lower than 10% in this study, which showed that the swabs were reliable. Conclusions Different bacterial profiles and antibiotic sensitivity were found in different Wagner's grades, IWGDF grades, and DFU types. Clinician should try to stay updated in antibiotic resistance pattern of common pathogens in their area, especially when practice on the empirical antibiotic use. This paper provided the detailed practical information (potential empirical regimens and alarming empirical regimens) to the clinician based on the assessments to the DFIs from the different aspects in this region. Disclosure The funders played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. A case of two foci of primary melanoma with metastasis to the skin in giant congenital melanocytic nevi Congenital melanocytic nevi are moles that are present at birth or arise within the first weeks of life [1]. They vary in size, from small, medium to very large or giant ones (more than 20 cm in diameter) [1, 2]. Giant congenital melanocytic nevi occur in approximately 1 in 20 000–50 000 newborns [3]. It is established in the literature that patients with giant melanocytic nevus have an elevated risk of developing melanoma, which ranges from 2% to 42% in the literature [1–3]. To date, no universal guidelines to treat giant melanocytic nevi have been given [3]. The aim of this paper is to present a case of a 25-yearold female patient who developed two sites of primary melanoma and metastasis to the skin, all in the area of giant melanocytic nevi. In March 2017, a 25-year-old female came to our OutPatient Plastic Surgery Clinic with giant congenital nevi in order to remove two distressing lesions (Figure 1). Additionally the patient is diagnosed with diabetes mellitus type 1. Before qualification to surgery the patient was asked to go through ultrasonography procedure. The obtained results show that those two lesions were very well vascularized and their picture was similar to angioma. On the day of surgery done in local anaesthesia, the lesion on the back was removed with excessive pigmented skin. The histopathological result demonstrated melanoma Breslow > 2 mm, Clark V diagnosed. The patient was referred to the Department of Oncologic Surgery in order to perform sentinel lymph node biopsy and excision of the second tumour. Histopathological examination did not show any melanoma metastasis in the sentinel node and no melanoma at the site of the primary tumour. The histopathology of the second excised tumour revealed metastatic melanoma (Figure 2). The genetic tests did not show a BRAF mutation. No metabolically active tumours were revealed in the positron emission tomography-computed tomography (PET-CT) study. In December 2017, the patient was admitted again to the Plastic Surgery Clinic to have another skin tumour excised from the back. Histopathological examination showed another primary cutaneous melanoma, Breslow > 2 mm, Clark V. The patient also underwent a procedure to widen the tumour Congenital melanocytic nevi are moles that are present at birth or arise within the first weeks of life [1]. They vary in size, from small, medium to very large or giant ones (more than 20 cm in diameter) [1,2]. Giant congenital melanocytic nevi occur in approximately 1 in 20 000-50 000 newborns [3]. It is established in the literature that patients with giant melanocytic nevus have an elevated risk of developing melanoma, which ranges from 2% to 42% in the literature [1][2][3]. To date, no universal guidelines to treat giant melanocytic nevi have been given [3]. The aim of this paper is to present a case of a 25-yearold female patient who developed two sites of primary melanoma and metastasis to the skin, all in the area of giant melanocytic nevi. In March 2017, a 25-year-old female came to our Out-Patient Plastic Surgery Clinic with giant congenital nevi in order to remove two distressing lesions ( Figure 1). Additionally the patient is diagnosed with diabetes mellitus type 1. Before qualification to surgery the patient was asked to go through ultrasonography procedure. The obtained results show that those two lesions were very well vascularized and their picture was similar to angioma. On the day of surgery done in local anaesthesia, the lesion on the back was removed with excessive pigmented skin. The histopathological result demonstrated melanoma Breslow > 2 mm, Clark V diagnosed. The patient was referred to the Department of Oncologic Surgery in order to perform sentinel lymph node biopsy and excision of the second tumour. Histopathological examination did not show any melanoma metastasis in the sentinel node and no melanoma at the site of the primary tumour. The histopathology of the second excised tumour revealed metastatic melanoma (Figure 2). The genetic tests did |
not show a BRAF mutation. No metabolically active tumours were revealed in the positron emission tomography-computed tomography (PET-CT) study. In December 2017, the patient was admitted again to the Plastic Surgery Clinic to have another skin tumour excised from the back. Histopathological examination showed another primary cutaneous melanoma, Breslow > 2 mm, Clark V. The patient also underwent a procedure to widen the tumour excision with a 2 cm margin, histopathological examination did not show melanoma at the site of this second primary tumour. The patient is scheduled for subsequent excision of smaller moles and till July 2019 her follow-up is preceded, the patient is being well. In the literature we can find data suggesting that patients with giant melanocytic nevi who develop melanoma tend to have poorer prognosis than patients with melanoma de novo. It results from deeper Breslow or Clark depth, increased rate of metastasis and increased rate of positive lymph nodes on presentation [3]. Turkeltaub et al. paid attention to the fact that melanoma arising in patients with giant melanocytic nevi may be more difficult to discern and Coughlin et al. additionally underlined that melanoma in such cases may even appear beneath the skin after the giant nevi were removed [3,4]. Similar observations result from our paper. In the described patient, the Clark scale was the highest (V), metastasis was present in the skin, however the sentinel lymph node was negative. Diagnosed melanoma was present as a subcutaneous tumour, soft, painless on palpation, and it was more like neurofibroma or angioma. The patients with giant congenital melanocytic nevi must be qualified for frequent physical and dermoscopic examinations. The suspected lesions should be removed and histopathological/immunohistochemical examination should be performed. If melanoma is detected, the treatment should be adequate to the degree of disease. The patients must have very frequent follow-up examinations (at least every 2 months). Nevertheless, prognosis is uncertain in such patients. Evolution of microRNA in primates MicroRNA play an important role in post-transcriptional regulation of most transcripts in the human genome, but their evolution across the primate lineage is largely uncharacterized. A particular miRNA can have one to thousands of messenger RNA targets, establishing the potential for a small change in sequence or overall miRNA structure to have profound phenotypic effects. However, the majority of non-human primate miRNA is predicted solely by homology to the human genome and lacks experimental validation. In the present study, we sequenced thirteen species representing a wide range of the primate phylogeny. Hundreds of miRNA were validated, and the number of species with experimentally validated miRNA was tripled. These species include a sister taxon to humans (bonobo) and basal primates (aye-aye, mouse lemur, galago). Consistent with previous studies, we found the seed region and mature miRNA to be highly conserved across primates, with overall structural conservation of the pre-miRNA hairpin. However, there were a number of interesting exceptions, including a seed shift due to structural changes in miR-501. We also identified an increase in the number of miR-320 paralogs throughout primate evolution. Many of these non-conserved miRNA appear to regulate neuronal processes, illustrating the importance of investigating miRNA to learn more about human evolution. Introduction Comparative genomics is an indispensable tool for studying the evolutionary history of any organism. Humans are no exception: people are perpetually fascinated with the molecular phenotypes that differentiate us from other primates. Studies comparing protein coding sequence data has uncovered rapid evolution between primates in many key areas, including immunity, sensory perception, reproduction, and keratinization [1,2,3]. However, results from genomic scale analyses continue to reinforce that much, if not most, phenotypic differences between species result from changes in gene expression and not amino acid divergence [3,4,5,6,7]. While there are many layers of gene regulation that exist between DNA sequence data and expressed proteins, emphasis is often placed on the mechanisms that regulate transcription. There has been much work characterizing the coevolution of transcription factors and their DNA binding elements [8]. However, less is known concerning the evolution of post-transcriptional regulatory elements. In addition to RNA binding proteins, microRNAs (miRNA) are an important class of post-transcriptional trans-acting factors that regulate mRNA stability a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 and rates of translation [9]. Despite their critical importance in seemingly every biological process (cell proliferation, differentiation, metabolism, apoptosis) [10], the role of miRNA in primate evolution has yet to be thoroughly examined. In this manuscript we provide an in depth characterization of miRNA identification and evolution across multiple primate lineages. MicroRNA biochemistry MiRNAs are short, noncoding, single-stranded RNAs important for post-transcriptional regulation in eukaryotes. MiRNAs are a relatively new addition to our understanding of genetics: the first miRNA was discovered in Caenorhabditis elegans in 1993, but the widespread effects of miRNAs were not fully recognized until the early 2000s [11]. Since then, discoveries in the miRNA field have expanded our understanding of genetics, illustrating the complexity of regulatory networks and the interplay between sequence and structure. Phylogenetic studies have shown that miRNAs have been present throughout the evolution of metazoans and that increased number and expression of miRNAs are positively associated with structural and organismal complexity [6,11]. Non-conserved miRNAs can be an indicator of adaptation in the genome of an organism, leading to novel phenotypes and a number of diseases, including heart disease, schizophrenia, and numerous types of cancers [6,10,12]. The intricate process of miRNA biogenesis (Fig 1) plays a crucial role in generating diverse phenotypes in organisms, as an alteration to any step may have profound downstream effects. MiRNA genes are transcribed from the genome, resulting in a primary miRNA transcript that may include a single miRNA or a cluster of miRNAs [11]. Regions of a primary miRNA form hairpin structures that are recognized by the endonuclease drosha, which cleaves the doublestranded stem region of the hairpin to produce an approximately 83 nucleotide (nt) precursor Regions of the primary miRNA form a hairpin structure that is recognized by the endonuclease drosha, which cleaves the double-stranded stem region of the hairpin to create a pre-miRNA of 83 nt in length. The pre-miRNA is exported to the cytoplasm where it is further processed by dicer, which cleaves off the loop region of the hairpin. This results in an approximately 22 to 23 nt double-stranded RNA called the miRNA-miRNA* duplex. The mature miRNA strand is loaded into the RNA-induced silencing complex (RISC), where its 8 nt seed region complementarily base pairs with messenger RNA targets, leading to their downregulation. miRNA (pre-miRNA) [13]. After being exported to the cytoplasm, pre-miRNA are further processed by a second endonuclease, dicer, which cleaves off the loop region of the hairpin to produce an approximately 22 nt double stranded RNA duplex that contains the mature miRNA and its complement (termed the star strand, or miRNA Ã ). The strand with the less thermodynamically stable 5' end becomes the mature sequence and is loaded into the RNAinduced silencing complex (RISC), while the star sequence is degraded. Occasionally a pre-miRNA has both of its miRNA and miRNA Ã strands lead to mature sequences. The mature miRNA base-pairs with complementary sequence within the 3' untranslated region (UTR) of messenger RNA (mRNA). This process guides RISC to specific transcripts, resulting in downregulation of the targets through degradation of the transcripts or inhibition of translation [14]. While there are varying degrees of complementarity between a miRNA and its mRNA target, binding is most highly dependent on positions 1 through 8 of the 5' end of the mature miRNA, known as the seed region [11]. While 75% of downregulated mRNA have canonical seed sites in their 3' UTR, the seed region is not always sufficient for causing downregulation [15]. The 3' end of the mature miRNA can also have an effect: positions 13-16 are highly conserved, and their proper complementary base pairing to a mRNA target is associated with downregulation [15]. About 60% of all human transcripts contain known or predicted miRNA target recognition sites [16]. A single type of miRNA can have one to thousands of targets, which establishes the potential for small changes in miRNA sequence to have profound phenotypic effects: each miRNA may result in varying degrees of phenotypic plasticity for different cell types, which have different target mRNAs to act upon. Additionally, many miRNA have multiple paralogs throughout the genome. Gene duplication followed by mutation in one copy is a common avenue for the evolution of novel functions: by maintaining more than one gene copy for a given miRNA, purifying selection to preserve function is often relaxed for one of the paralogs, allowing for mutational acquisition, differentiation, neofunctionalization, and subfunctionalization [17]. Most miRNA are highly conserved across species and show higher rates of purifying selection than protein-coding regions of the genome, suggesting that variation found in miRNA sequence may play a vital role in the evolution of metazoans [18,19,20,21]. MiRNA structure While many studies focus on changes in miRNA expression or variants in the seed region, changes in pre-miRNA secondary structure can also dramatically affect downstream function through several different mechanisms. In general, variants in the stems of pre-miRNAs that decrease overall structural stability of the hairpin reduce the production of mature miRNA [22]. If the pre-miRNA has a secondary structure that is very divergent from the standard hairpin, the ability of drosha to recognize and process the pre-miRNA may be reduced or completely eliminated. Small changes in sequence may have drastic effects: a variant in the mature sequence of miRNA-125a blocks the processing of primary miRNA to pre-miRNA, resulting in complete loss of function [23]. However, sequence divergence does not always imply structural divergence, as compensatory mutations often help conserve a pre-miRNA's hairpin structure. A mutation in the primary miRNA sequence could also result in a different but stable hairpin structure. This could alter the location of drosha cleavage sites during pre-miRNA biogenesis [24], and in turn shift the cleavage sites of dicer, resulting in a different mature miRNA and thus different seed region. Sun et al. identified such a variant with an altered cleavage site and seed region shift [25]. Previous studies have only investigated a limited number of pre-miRNA variants, and the effects of most are still unknown. Primate miRNA diversity A major roadblock to studying miRNA across primates is the lack of experimentally verified miRNA in non-human primates. Only 12 of the~300 known primate species [26] have any entries in miRBase (Fig 2, S1 Table) [27]. The number of characterized human pre-miRNAs (n = 1881) is still more than twice as large as that of chimpanzee (n = 655), the most well studied non-human primate. The majority of these miRNA are predicted based only on homology to the human genome; only four species (chimpanzee, gorilla, orangutan, and rhesus macaque) have sequences that are experimentally validated through RNAseq or other expression analysis. While homology is a useful tool for identifying orthologs, it may result in an overrepresentation of conserved sequences with respect to humans, missing sequence diversity in more distantly related primate species. Homology alone cannot identify if a sequence is actually expressed or forms a stable hairpin that successfully completes processing by drosha and dicer. Even in cases where a mature miRNA is produced, one cannot determine the exact boundaries of the mature sequence (and thus seed region) without expression data [28]. The current best mature predictive software (Mature Bayes) only comes within 1 nt of the true mature sequence 49% of the time, which can result in an incorrect seed region and thus target repertoire [29]. Despite this lack of validated miRNAs, some research has compared differences in miRNA across primates. Berezikov et al. used high throughput sequencing technology to discover miRNAs in the brains of human fetuses and chimpanzee adults [30]. While they discovered hundreds of miRNAs specific to primates with dozens not conserved between humans and chimpanzees, they did not perform any functional validation of the targets of these miRNAs. Hu et al. recently discovered several miRNA that were differentially expressed in human and chimpanzee brains, and that this differential expression resulted in downregulation of several neuronal genes [31]. Zhang et al. discovered an X-linked miRNA cluster that was rapidly evolving in primates, and these miRNA had increased expression during male sexual maturation [32]. However, techniques investigating only the expression level of miRNA would miss any phenotypic differences caused by changes in miRNA target specificity. Target specificity could |
change due to sequence differences in the seed region, or sequence differences that change the secondary structure of pre-miRNA and thus alter its downstream processing. Non-conserved miRNA provide insights into primate evolutionary history, including what differentiates humans from other primates. In order to investigate how conserved or divergent miRNA are within primates, we sequenced miRNA from thirteen species, greatly expanding the number of experimentally validated non-human primate miRNA and our knowledge of their evolution. MiRNA discovery In order to better characterize patterns of miRNA evolution across primates, small RNAseq was performed on fibroblast cells cultured from 13 divergent primate species (Fig 3, see Methods). Our study drastically expanded our knowledge of primate miRNA, tripling the number of primate species with experimentally validated miRNA in miRBase (from 4 to 13), and adding dozens to hundreds of miRNA per species (Fig 4, S2 Table). This includes the only experimentally validated miRNA sequences available for bonobo, one of human's closest evolutionary relatives; and for the first time, experimentally validated miRNA sequences are available for New World monkeys, lemurs, and a galago. For non-human primate miRNAs that to date were computationally predicted by homology alone, 27% (211/766) were sequenced in at least one of our primate species, with the majority of these sequences (86%) being represented by at least two species (Fig 5). Primate miRNA evolution Sequenced miRNAs were computationally clustered into groups of homologs that had at least 70% identity within the mature region (see Methods). Homology groups containing paralogs were further subdivided into their individual miRNA orthologs, resulting in 188 particular miRNA ortholog groups with representation in at least three primate species. As expected based on previous studies [19], primate miRNA appears to be highly conserved, with 173 of 188 miRNA ortholog groups (~92%) showing no variation within the mature region across primates. Of the 15 miRNA ortholog groups that contained variation within the mature region, none of these variants occurred within the seed region. This is consistent with previous studies that show the seed region to be the most highly conserved region of miRNA and the most important determinant of target recognition (Fig 6) [15]. Only one variant was found within positions 13-16, the second most conserved region of miRNA that is sometimes involved in 3' complementary base pairing during target recognition. Most variation was observed in basal primate species: 14/21 variant sequences were from the basal Strepsirrhini suborder, and 5/21 were from New World monkeys (Table 1). This is concordant with the hominoid slowdown hypothesis, which shows that rates of nucleotide substitution in primates decrease as generation time increases [33]. Structural analysis We analyzed the thermodynamic stability and structural conservation of any miRNA with at least 5 species in its alignment (n = 152, see Methods). Our pre-miRNA structures are thermodynamically stable as measured by z-score (where more negative values indicate stability), with most of the analyzed miRNAs (120/152) having a z-score that indicates very conserved structures (z < -3.0) (Fig 7). Structural stability is further evidence that a miRNA sequence is genuine [34]. Structural conservation as measured by the Structural Conservation Index (SCI), where an SCI of 1 indicates complete structural conservation, generally decreases as sequence divergence increases (Fig 8), but this correlation is weak (R 2 = .1719). This is to be expected, as SCI only approximately captures true structure conservation, but is also concordant with the properties of miRNA: a miRNA with low sequence identity may still be structurally conserved due to compensatory mutations, or a miRNA with high sequence identity may have few variants that result in drastic (and perhaps functionally significant) structural difference. miR 2355-5p MiR-2355-5p was the only alignment to contain a variant within the conserved 3' binding location (positions 13-16) of the mature miRNA. Homologous sequences from additional primate species and a mouse (Mus musculus) outgroup extracted from the UCSC Genome Browser reveal an interesting evolutionary event: the closest sister taxa to humans (chimpanzee, bonobo, and gorilla) have variant T15C in the mature sequence, while humans have seemingly reverted to the ancestral T (Fig 9) [35]. This reversion is conserved across humans in dbSNP [36]. This specific transition, as well as other mutations throughout the pre-miRNA, do not seem to alter the secondary structure of the hairpin. This is the first time RNAseq has confirmed the expression of this miRNA in a non-human primate (chimpanzee, bonobo, gorilla, and orangutan); the only other species confirmed to express this miRNA is cow (Bos taurus) [37], suggesting it is likely to be expressed in other primates as well. Human miR-2355-5p has MiRDeep2 scores range from -10 to 10, with a higher number corresponding to increased likelihood that the miRNA is genuine. A cut-off of 0 was used to be included in this study. MiRNA already annotated in miRBase are represented in black and gray: black represents miRNA with experimental validation, and gray represents miRNA previously predicted solely by homology to the human genome that have now been validated in this study. Novel miRNA are shown in a color corresponding to their miRDeep2 score; this score is partially determined by the availability of any previously annotated miRNA, which would inherently result in lower scores for our primates with no information in miRBase. been shown to be expressed in embryonic stem cells [38], neural stem cells [39], and throughout the female reproductive tract [40,41]. The specific targets and function of miR-2355-5p are currently unknown. miR-299-3p miR-299-3p was completely identical across all primates except for a single change (C10T) in humans. The secondary structure of miR-299 was entirely conserved (SCI = 1) and was the most thermodynamically stable hairpin of all the miRNA in this study (z = -10.33). Previous research identified this change based on sequences from human, chimpanzee, and macaque, and we show that the ancestral sequence is shared across all primates except human. Initial research found that miR-299-3p has human-specific expression, with preferential expression in neurons; while targets of this miRNA were enriched for neuronal function and axon guidance, there was no difference in target specificity between the human and chimpanzee versions [42]. A more recent study confirmed that miR-299-3p is expressed in cerebellum and has targets enriched for neuronal function; however, changes in target repertoire and expression levels between humans and non-human primates were identified, illustrating how a change outside of the seed region can still have profound effects [43]. miR-501-3p Of all of the alignments, miR-501 had the lowest mean pairwise identity (91%) across primates, similar to the average pairwise identity between most human and mouse pre-miRNA (>90%) [19]. However, miR-501 still appeared to be structurally conserved due to a number of compensatory mutations (SCI = 0.95, covariance contribution = -0.24). The mature sequence miR-501-3p also contains a variant just outside the seed region at position 9 in the basal Strepsirrhini suborder that introduces an additional bulge in the hairpin. This variant, as well as variants outside of the mature sequence, likely alters the overall secondary structure of the hairpin, resulting in the mature sequence being shifted downstream by 1 nt in apes relative to Strepsirrhines (Fig 10). This would alter the dicer cleavage position and shift the seed region by one nt, likely changing the target repertoire of this particular miRNA. Previous research has shown that miR-501-3p localizes to dendrites and plays a key role in NMDA-induced dendritic spine remodeling, which is thought to be the "structural basis of information storage in the brain for cognitive functions such as learning and memory" [42]. Suppression of GluA1 expression is necessary for long term maintenance of NMDA-induced spine modifications: experimental assays showed that miR-501-3p targets the transcript of GluA1, and their expression are inversely correlated during postnatal brain development. NDMA stimulation increased the expression of the primary miR-501 transcript, and still increased mature miR-501-3p levels even when a transcription inhibitor was present, suggesting that miR-501-3p undergoes post-transcriptional regulation [42]. This regulation may be controlled by the structure of miR-501, as hairpin structural stability increases the production of the mature miRNA [22]. The sampled apes in our study have increased complementary base pairing throughout the hairpin, lacking the mid-mature bulge found in Strepsirrhines. This structural difference and its resulting seed shift may indicate an important moment in primate brain evolution. Evolution of microRNA in primates miR-320 One of our largest homology groups was composed of the miRNA-320 paralogs (Fig 11). In humans, the miR-320 family consists of one copy of miR-320a (chr8), two copies of miR-320b (chr1), two copies of miR-320c (chr18), two copies of miR-320d (chr13 & chrX), and a single copy of 320e (chr19). Gene duplication allows novel functions to evolve, as one paralog is maintained by purifying selection for its previous function while other copies are allowed to acquire mutations and neofunctionalize and/or subfunctionalize. While our data showed miR-320a to be present across the entire primate lineage, we only identified RNAseq reads for miR-320b and miR-320c in apes and Old World monkeys, matching the copy number found in humans (we did not sequence any copies of miR-320d or miR-320e from any species, likely because they are not expressed in this particular cell type). This pattern of only being found in the most derived species is unlikely to occur by random chance, and may indicate that these additional paralogs do not exist in these genomes. miR-320a is mammalian-specific, and its paralogs have only been identified in previous studies (experimentally or computationally) in primates: 320b in gorilla, 320b and 320c in macaque, and all paralogs in chimpanzee and orangutan [27,44,45,46]. We confirmed the presence or absence of each miR-320 paralog using blastn (see Methods, S3 Table). 320a was present in all primates, concordant with our RNAseq data. 320b1, 320c1, 320c2 and 320d2 were present in all primates except for the basal Stepsirrhini suborder. 320b2, 320d1, and 320e were present only in apes and Old World monkeys, with 320e having an additional duplication in orangutan. Alignments clearly indicated whether or not a sequence was present; for example, the pre-miRNA of miR-320b1 is absent in Strepsirrhines despite the conservation of Red indicates the mature miR-501-3p sequence supported by reads, yellow the predicted loop, and blue the predicted star sequence. In mouse lemur and galago, the mature sequence contains a variant (arrow) immediately following the seed region (underlined); this as well as variants outside of the mature sequence appear to alter the overall secondary structure of the hairpin, resulting in the mature sequence and thus the seed region being shifted downstream by 1 nt (circled in black). https://doi.org/10.1371/journal.pone.0176596.g010 Evolution of microRNA in primates flanking sequence, illustrating the insertion event that took place sometime after the Strepsirrhini suborder split from the rest of the primate lineage (Fig 12). Paralogs found in a genome but missing in our sequencing data may be identified in future RNAseq efforts; however, it is also possible that despite being in the genome, they are not expressed, are expressed only in different cell types, or are not successfully processed into mature miRNA. Our data indicates that the miR-320 family has undergone multiple gene duplications throughout primate evolution, and suggests that only more derived species successfully express mature forms of miR-320b and miR-320c, but more extensive investigation of expression in other species is needed to expand on these conclusions. Numerous studies have illustrated the role of the miR-320 family as a regulator of neural development. The family is highly expressed in rat neurons, with enrichment specifically in axons [47], and is also targeted by REST, a transcription factor that silences neuronal genes in non-neuronal tissue and is essential in neuronal differentiation as well as the maintenance of neural stem cells [48,49]. Transfection experiments confirm that miR-320b inhibits expression of neuron-related mRNA targets, and in situ hybridizations to investigate histological expression patterns show miR-320b co-localized with neurons in both human and macaques [50]. Increased levels of miR-320b have also been shown to increase neurite length, further suggesting that increased copies of this miRNA may play a role in neuronal development [51]. The miR-320 family is also frequently dysregulated in neurological disorders: miR-320 was downregulated in blood of schizophrenia patients [52] and the striatum of forebrains of Huntington Disease patients [53], and upregulated in the cortex of patients with sporadic Alzheimer's disease [54] and mouse brains undergoing prion-induced neurodegeneration [55]. In both control and Huntington Disease forebrains, miR-320b and 320c have such high rates of posttranscriptional |
nucleotide substitutions compared to their primary transcripts that researchers Evolution of microRNA in primates have suggested these edited sequences be considered the reference miRNA in brain tissue; additionally, these two paralogs were the only miRNA with substitutions at several positions across the mature sequence [53]. In addition to its role in neuronal development, miR-320c appears to have a wide range of functions, including regulation of chondrocytes in cartilage [56], differentiation of skeletal stem cells [57], inhibition of proliferation, migration, and invasion in bladder cancer [58], and induction of resistance to the chemotherapeutic agent gemcitabine in pancreatic cancer [59]. Given the number of duplications this miRNA family has undergone throughout primate evolution and the enrichment of neuronal functions among its paralogs, it is likely that the miR-320 family plays a role in primate brain evolution. Discussion In recent decades, our understanding of functional genomics has been facilitated by technological advances in high throughput transcriptomic and proteomic methods. These tools are also rapidly expanding our understanding of miRNA, a relatively new class of trans-acting regulators of gene expression that are quickly being recognized as potential sources of phenotypic variation or as biomarkers for disease. Reliable sequence information across a diverse range of species is required for researchers to comprehensively study miRNA evolution. In this study, we have greatly expanded the number of experimentally validated nonhuman primate miRNAs, especially in more divergent sister taxa. Inclusion of New World monkeys, lemurs, and a galago in this study made it possible to identify more ancient evolutionary events that have shaped primate evolution, such as the mature sequence shift we identified in miR-501-3p and the duplications of the miR-320 family. We have also demonstrated that more than a fourth of all computationally predicted primate miRNAs were found within our data (despite having only sequenced one cell type), lending confidence to prediction by homology as a method of miRNA discovery. However, our results also illustrate the Fig 12. Alignment of miR-320b1 homologous sequences. The yellow box denotes the pre-miRNA sequence, red outlines the mature sequence, and variants with respect to humans are marked in grey. miR-320b1 is found in all apes, Old World monkeys, and New World monkeys (only human and marmoset are shown for simplicity). The entire pre-miRNA sequence is clearly absent in Strepsirrhines (aye-aye and galago), despite conservation of flanking sequence, demonstrating an insertion event that took place after the Strepsirrhini suborder split from the rest of the primate lineage. importance of validating mature miRNA through RNAseq: mature miRNA sequence shifts caused by changes in secondary structure cannot be reliably determined by homology alone, and require sequencing reads to determine their boundaries. Because this study only sequenced miRNA from a single cell type (cultured fibroblasts), we likely captured a subset of the miRNAs expressed within each species. However, it is notable that fibroblasts and neurons are both derived from the ectoderm [60], and conversion of fibroblasts to neurons is relatively easy [61]; additionally, fibroblasts are frequently used as a model when studying brain disorders because of their neuron-like signal transduction pathways [62,63]. This relationship between fibroblasts and neurons possibly explains the abundance of neuronally-expressed miRNA identified in this study. Nonetheless, even a single cell type was sufficient for finding a number of interesting evolutionary events. More sequencing efforts across a diverse range of cell types and stages of development are likely to reveal additional insights into primate evolution, especially in cell types already known to have undergone significant phenotypic changes (i.e., neuronal tissue). More research is needed to confirm the actual biological targets of these miRNA of interest, as target prediction software is not accurate given the complicated binding interactions of miRNA: TargetScan and miranda, two of the best predictive software currently available, both have false positive rates of~25% [64]. Even target identification is starting to benefit from high throughput technology, such as expression profiling after miRNA knockdown or overexpression [65], as well as UV cross-linking miRNA-mRNA duplexes to RISC to be pulled out via immunoprecipitation [18]. We hope that this study serves as a foundation for future research into the evolution of miRNA and gene regulation in primates. Primate samples In order to represent a broad span of the primate phylogeny, we selected thirteen primate species that had sequenced genomes and fibroblast cell cultures available through Coriell Cell Repositories (Fig 3, S4 Table): Marmoset is the only species where the genome of a closely related species (C. jacchus) was used as the reference for the cell line available (C. geoffroyi). RNA sequencing Mature miRNA from fibroblast cells was extracted using the Qiagen miRNeasy Mini Kit following the manufacturer's protocols (Catalog #217004, Qiagen, Valencia, CA). We prepared and barcoded samples using Illumina's TruSeq Small RNA Library Preparation Kit (Catalog #RS-200-0012, Illumina). Barcoded samples were multiplexed for 25bp paired-end sequencing on a single lane of an Illumina MiSeq. The Picard tool ExtractIlluminaBarcodes separated raw Illumina reads by barcode, and IlluminaBasecallsToFastq output the results in fastq format. Fastqc was run on the original fastq, with FastqTOSam importing the fastqs into bam format. Bam files were processed with Picard MarkIlluminaAdapters tool. The adapter-masked fastqs were put into PEAR with no minimum overlap size in order to merge the reads. Fastqc was run on merged, trimmed fastq and compared to the original fastqc to confirm that adapters were trimmed. MiRNA identification miRDeep2 was used to predict novel and identify previously annotated miRNA [66]. Merged, trimmed reads were mapped to their respective genomes using the miRDeep2 mapper.pl module with the following parameters: -c -j -l 18 -m -p -s -t-v. MiRDeep2 was executed with default parameters. When making novel miRNA predictions, miRDeep2's algorithm accounts for already known miRNAs of the species being analyzed and of any related species. We retrieved a list of known miRNAs from miRBase (release 21) for any of our primates that were in the database (H. sapiens, P. troglodytes, P. paniscus, G. gorilla, P. pygmaeus, M. mulatta), and used all known metazoan miRNAs as our "related species" reference. MiRDeep2 assigned a score from -10 to 10 to each miRNA, with a higher number corresponding to increased likelihood that the putative miRNA is functional. This score is partially determined by the availability of any known miRNA, which would inherently result in lower scores for our primates with no information in miRBase. Because of this, we chose a relaxed score cut-off of 0 and minimum read depth of 3 to include a miRNA in our analyses, with the expectation that false positives would be removed during paralog clustering and alignment. Confirmation of predicted miRNAs We compiled a list of 776 different miRNAs from miRBase that lacked experimental validation in non-human primates prior to this study. Blastn (NCBI blast 2.2.21) was then used to find a conservative match (100% identity over at least 18 nt) of these sequences within our experimentally validated miRNA. If a match was found in at least one of our primate species, the miRBase name and genome coordinates were extracted from the header of the match using a custom perl script, and that particular miRNA was counted as now having experimental validation in a non-human primate. When determining whether a given primate genome contained a particular predicted miRNA, paralogs were collapsed into a single group and the highest score was taken, as it is difficult to distinguish between paralogs that have identical or nearly identical mature sequences. Homolog clustering To determine which miRNA in our data set were homologous (either as orthologs or paralogs), our mature miRNA sequences were clustered by an all-versus-all search using blastn (NCBI blast 2.2.21) at a permissive e-value (1E-02). Additional e-value cut-offs were evaluated, but given the short nature of the search queries (~22 nt mature miRNAs), it was assessed that a more permissive value was necessary to find correct matches for sequences of this length. The results were filtered for matches between sequences that had 70% identity over at least 18 nt. The sequences that remained were then clustered into groups by a custom Python script, with a sequence being added to a group if it shared at least 70% identity to any other sequence in that group. In this way, every sequence found in the search was placed into a group or was identified as not having known homologs within our dataset. Phylogenetic analysis The 100 largest homology groups were selected for phylogenetic analysis, with two trees generated per group: one based on our experimentally validated mature sequence, and another based on the excised sequences as predicted by miRDeep2. These excised sequences contain the actual pre-miRNA plus~20 nt of flanking sequence on either side. Specifically, miRDeep2 searches for the highest local stack of mature reads and excises it twice, once with 20 nt upstream and 70 nt downstream flanking sequence, and once with 70 nt upstream and 20 nt downstream flanking sequence. This is in order to determine if the mature reads occur on the 5' or 3' part of the hairpin, with miRDeep2 attempting to fold both excised sequences into stable hairpins. Thus, any excised sequence that is confirmed to include a pre-miRNA will have exactly 20 nt flanking on one side, and~20 nt flanking the other (exact length of this flank can vary depending on the length of the mature, star, and loop sequences). This flanking sequence adds robustness to our alignments of already very short sequences. Sequences were aligned using the Fast Statistical Alignment Algorithm (FSA) [67]. Trees were then generated using RAxML with the following parameters: -f a -m GTRGAMMA -p 12345 -x 12345 -# 1000 -s -n [68]. Mature and excised miRNA alignments were visualized with the Max Plank Institute's Bionformatics Toolkit [69] and trees were visualized with Phylodendron (http://iubio.bio. indiana.edu/treeapp/). Trees were visually examined for evidence of potential adaptation, such as an excess of paralogs found only in a particular subgroup of primates, or basal primates whose sequences represent an intermediate step between paralogs. These large paralog trees were then subdivided into groups that contained only one particular miRNA (in at least three species), based on visual assessment of both mature and excised miRNA alignments and trees. In nearly all cases, miRNAs were clustered into obvious ortholog groups. In rare cases where it was difficult to determine where a particular sequence belonged, miRNA sequences in the trees were searched within the miRBase database for the closest match. These searches clearly labeled known paralogs, confirming that our self-blast clustering worked as intended. After subdivision, each particular miRNA was realigned using FSA and trees reconstructed with RAxML. Each alignment was then searched for sequence variants within the mature region of a particular miRNA, as these changes are likely to have phenotypic consequences. Structural analysis The exact sequence of the pre-miRNA hairpin was retrieved from the excised sequences based on the folding predictions of miRDeep2. Pre-miRNA alignments were analyzed by the Vienna RNA Package's RNAz program [70], which predicts the secondary structure of noncoding RNA and calculates different measures of structural conservation. Thermodynamic stability of a particular secondary structure is indicated by the z-score, which is the number of standard deviations between the minimum free energy (MFE) of a sequence compared to the MFE of random sequences of the same length and base composition; RNAz circumvents this computationally intensive step by using support vector regression to estimate mean MFE and standard deviation [71]. Lower z-scores imply greater thermodynamic stability, and scores below -3 generally indicate very stable structures that are unlikely to arise by random chance. The Structure Conservation Index (SCI) is the most accurate measure of structural conservation currently available [72]. SCI compares the average minimum free energy (MFE) of individual sequences in an alignment to a consensus MFE of that alignment. This consensus MFE is weighted by a "covariance contribution," which gives a bonus to compensatory and consistent mutations that conserve structure, and a penalty to inconsistent mutations; a negative covariance contribution indicates more compensatory mutations. A SCI close to 1 indicates structural conservation, but SCIs cannot necessarily be compared since they depend on the number of sequences in an alignment and its mean pairwise identity. In general, SCIs near or above the mean pairwise identity of the alignment indicate good candidates for conservation [73]. Paralog confirmation Genomic coordinates of the human pre-miRNA for each member of the miR-320 family were obtained from miRBase, and were then used to extract pre-miRNA |
sequences plus 1000 nt of flanking sequence on either side from the UCSC Genome Browser. These~2080 nt sequences were then searched against our thirteen primate genomes using blastn at an e-value of 1E-10, and were permissively filtered for matches with a minimum of 70% identity over at least 300nt. Repetitive elements in the flanking regions that were found in thousands of locations within an individual genome were removed. For each species, the match with the highest blast score that overlapped with the pre-miRNA was identified as the paralog. Some species had no overlapping matches, but did have unique matches in the flanking regions; for these, the sequence containing the hypothetical location of the pre-miRNA (100 nt upstream and 200 nt downstream from the pre-miRNA start) was extracted with a custom perl script, aligned with FSA [67], and visualized with the Max Plank Institute's Bionformatics Toolkit [69] in order to investigate conservation of the pre-miRNA. Data access miRBase policy requires acceptance of publication prior to submission. Accession numbers will appear here once processed by miRBase. Supporting information S1 Breast diseases in women over the age of 65 in Monastir, Tunisia As life expectancy is on the rise, it is predicted that a growing number of people will live beyond the age of 65 and therefore a higher number of elderly women will have breast diseases requiring significant health care and services. This study is aimed at investigating the characteristics, the treatment and outcomes of women older than 65 years old treated for breast diseases at our institution. This was a retrospective study covering the period from January 2003 to December 2011. It involved 92 patients aged over 65 and treated for breast disease in the Maternity Center of Monastir, Tunisia. The data included characteristics of patients and tumors, treatment and outcomes that were obtained through data extraction sheets. We reported a study of 92 women over the age of 65 of whom 77 women had malignant breast disease (83.6%) and 15 benign breast diseases (16.4%). Breast cancer was discovered at a mean age of 72.5 ± 6.6 years. Distant metastases were found in 5.3% of cases and infiltrative ductal carcinoma was detected in 85.7% of patients. Hormonal receptors were positive for estrogens in 64.7% of cases. Surgical treatment was performed in 73 patients and adjuvant treatment was prescribed for 67 women (86%). The complication rate was 16.6% among the 73 patients who underwent surgery. Benign breast diseases represented 16.3% of the mammary pathologies. Abscesses and fibrocystic mastopathy were the most frequent histological diagnoses. Despite great interest in geriatric gynecological pathology worldwide, many questions related to how optimally treat this patient population remain unanswered. In this study, a surgical treatment was performed in 94.8% of breast cancer patients and the complication rate was 16.6%. Introduction Nowadays, the health care of elderly people is of great interest. Population trends will result in a growing number of people over the age of 65. In addition, this age group is increasingly demanding and requires significant health care and services. In Tunisia, the prevalence of elderly people, which was 9% in 2003, reached 9.5% in 2009 and will reach 17% in 2029 [1]. Greater interest in gynecological pathology has been centered on sexually active woman than on elderly woman. This is underlined by health programs aimed primarily at promoting the health of mothers and children. To our knowledge, only a few aspects of pathologies of the breast in Tunisian elderly women have been dealt with. Methods This was a retrospective study from January 2003 to December 2011. It involved 92 patients treated for breast diseases in the Maternity Center of Monastir and fulfilling the following inclusion criteria: aged over 65 and having malignant or benign breast disease. The data were obtained through data extraction sheets. The variables studied were: age at diagnosis, family and personal history, age of puberty, age of menopause, parity, time and reason for consultation, TNM stage, histological type and treatment. We included 92 patients: 77 breast cancer cases and 15 benign conditions. The statistical analyses were performed using the SPSS 18.0 statistical software package. Data analysis was descriptive in nature. Conventional formulae were used to calculate the means and standard deviations. Malignant breast diseases The average age of patients with breast cancer was 72.5 ± 6.6 years [65 -89 years]. This cancer affected mainly the 65-74 age group. Four women (5.2%) had a family history of breast cancer. Of the 77 women in this group, 4 women had precocious puberty (< 12 years). The mean age at puberty was 13.1 ± 1.4 years [10 -16 years]. Parity for the breast cancer group was very close to that of all the series. Indeed, the average parity was 5.8 ± 3.5 [0 -17]. In addition, nulliparity was found in 12 women (15.6%). The absence of breastfeeding was noted for 14 women (18.2%) and was not specified for 9 women. The mean age at menopause was 49.4 ± 4.4 years [38 -57]. Late menopause onset (> 55 years) was noted in 5 women (6.5%) and 2 women had used hormonal contraception. A total of 61 women (79.2%) with breast cancer had a history of high blood pressure and diabetes. In addition, in these women, a surgical history was reported in 31 women (40.2%). We particularly noted 3 cancers, 5 breast abscesses and a breast adenofibroma. For breast cancer, the palpable mass was the chief complaint. It was found in 68 women followed by mastitis and nipple discharge in 8 and 5 women respectively. The average consultation delay was 5.2 ± 5 months [0 -16 months]. This time was calculated for 76 women. The most represented T stage for breast cancer was T2 according to the TNM classification followed by T4 with respective rates of 54% and 38%. It should be noted that one woman had bilateral involvement: T2 on one side and T1 on the other (Table 1). For breast cancer, N1 stage of nodal invasion was reported in 40 women (51.9%) followed by N0 and N2 in 35(45.4%) and 2 (2.7%) women respectively. The woman with bilateral involvement had N0 on one side and N1 on the other (Table 1). At the time of diagnosis, 4 patients (5.3%) had distant metastases. The most common histological type was invasive ductal carcinoma (IDC) detected in 68 women (87.2%). The histoprognostic grade was specified in 75 women. It was grade I for 38.7% of women and II for 38.7%. The hormone receptor study was performed in 74 cases: it was positive estrogen receptors (ER+) in 47 cases (63.5%) and positive progesterone receptors (PR+) in 36 cases (48.6%). A surgical treatment was performed on 73 patients: surgery was first in 65 women (83.3%) and it followed neo-adjuvant chemotherapy for 8 patients whose tumor was diagnosed at an advanced stage. Patey's mastectomy with nodal dissection was performed in 54 cases (70.1%). A conservative treatment was performed in 6 patients including a woman with a bilateral cancer treated on the left with Patey's mastectomy and on the right by a simple lumpectomy. A palliative mastectomy was performed in 13 cases (16.6%). Adjuvant therapy was prescribed for 67 women (87%). This included radiotherapy for 41 (61.2%) women, chemotherapy for 48 (71.6%) women and hormone therapy for 24 (35.8%) women according to different protocols. Among the 73 patients who underwent surgery, the follow-up was simple for 60 cases. The sequels were mainly dominated by infectious complications. One death was reported in a 76 year old female patient (hypertensive and hemiplegic) on day 4 after cardiac failure. After an average follow-up of 4.5 years, 5 patients had locoregional recurrence with or without axillary involvement. A case of distant metastasis was also observed. Benign breast diseases As for benign breast diseases, they represented 16.3% of the mammary pathologies. Fibrocystic mastopathy and abscesses were the most frequent histological diagnoses. The mean age of patients in this group was 68.2 ± 3.1 years [65 -75 years]. The main surgical procedures varied according to the indications. These consisted of three tumorectomies, three pyramidectomies, two zonectomies and three biopsies. A flattening under antibiotic cover was performed for the 4 breast abscesses of our series. The operative follow-ups were simple in the majority of cases. A single hematoma case of the scar was observed. Discussion Due to the epidemiological characteristics including population aging, the rise in life expectancy as well as the increasing incidence of cancer with age, elderly to very old women represent a large proportion of breast cancer patients. The U.N. population division estimates one in five persons are expected to be 65 or older by 2035 in the United States. In addition, data from the Surveillance Epidemiology and End Results Program indicate that the incidence of breast cancer increases up to 80 years of age and plateaus between 80 and 85 years of age [2]. For elderly women in Europe, the incidence of breast cancer is 320/100,000 women between 65 and 75 years old and 300/100,000 women after the age of 80 [3]. Like all cancers, the frequency of breast cancer increases with age [4]. The authors reported that 45% of breast cancers were diagnosed in women over 65 [5]. The nodule or palpable mass is the most frequent reason for consultation. In the Chatzidaki et al. series [6], palpable nodule and nipple discharge were the reasons for consultation in 58.5% and 7.5% of cases. In the Jedidi series [7], the most common reasons for consultation were palpable nodule (77.9%), skin changes (22%) and nipple discharge (10.3%). In our series, the most common reasons for consultation were palpable nodule (89.5%), carcinomatous mastitis (10.5%) and nipple discharge (6.6%). There is a consultation delay of more than 3 months or more for elderly women with breast cancer [8]. In this age group, this delay is mainly due to the absence of similar cases in the family, the absence of a partner or the fear of the consequences of cancer [8]. In the Jedidi series [7], the average consultation delay was 11.2 months in Sfax Hospital in the south of Tunisia. In our series, the average consultation delay was shorter (5.2 months) and this can be explained by the fact that Sfax Hospitals admitted more women from the inner regions. In fact, in the inner regions, there are poor information campaigns and low awareness of health issues among women. To reduce this delay and improve the prognosis of breast cancer, Forbes et al. [9] offer elderly women an education program that involves the following objectives: To recognize the symptoms of breast cancer, identify the risk factors for breast cancer, encourage early consultation, discuss the benefits of early consultation and reduce the fear of the consequences of breast cancer. The size of the tumor is a prognostic factor. The literature reports that 45-80% of elderly patients present with T2 or greater or palpable tumors [10]. In our series and those of Jedidi [7], tumors classified T2 are more frequent followed by those classified T4 (Table 1). One would hypothesize that the lack of standard screening guidelines for elderly women may lead to advanced disease at diagnosis. Some studies have reported that elderly women over the age of 80 have a higher likelihood of presenting with advanced disease [11], with up to 44% axillary nodal involvement [12], or with breast cancers that have poor prognostic features [12]. However, other studies suggest that older women have less aggressive disease [13,14] and are less likely to have nodal involvement [15]. Like those of Jedidi [7], our series revealed that node invasion was mainly of classes N0 and N1 (Table 1). Infiltrating ductal carcinoma remains the most common histological subtype of breast cancer diagnosed in older and younger patients. Older patients are reported to have a greater frequency of tumors with more indolent histologies and an overall more favorable biological tumor profile [16]. This profile is characterized by a higher percentage of estrogen receptor-positive (ER) tumors (83% ER in patients under 65, 87% to 91% ER tumors in patients 65 and older) [12]. The proportion of ER tumors continues to rise even within the over-65 cohort, with 87% of patients aged 65 to 74 years having ER tumors, compared with 91% of patients aged 85 and older. Reduced proliferation markers (such as S-phase fraction) and HER2/neu negativity are also features of breast |
tumors in the elderly [12]. In our series we found RE + in 64.7% of cases and RP+ in 48.6% of cases. The gerontological community has long recognized the heterogeneity within the elder patient group. The concept of chronological versus physiological age is difficult to quantify, yet is of inherent importance in clinical decision-making. Aging affects multiple body systems and remains an individualized process that correlates poorly with chronological age. The concept of "functional age" has thus emerged. Two surrogate markers of functional age, comorbidities and the gerontological assessment, are currently used in cancer care-related decisions. The present recommendations by the National Comprehensive Cancer Network (NCCN) include the use of a geriatric assessment tool in developing care plans for all cancer patients aged 70 and older [17]. No current assessment tool has emerged as the preferred choice. Today, the majority of breast cancer treatment recommendations appear to be modeled on younger patients, as most clinical trials focus on healthy, young women [18]. Several authors agree that older women should have the same treatment options as young women [19]. Survival without recurrence and/or without metastasis and the relative risk of mortality depend on a good indication for adjuvant therapy, the presence of prognostic factors represented mainly by SBR grade, lymph node involvement, tumor size and the presence of hormonal receptors. The management of the mammary pathology in the elderly patient is often suboptimal with a general tendency to under-treatment. It is characterized not only by a decrease of the adjuvant therapeutic indications but also by an increase in abstentions and a reduction in palliative care, analgesic and comfort management [20]. In the elderly, it is difficult to evaluate the indication of adjuvant chemotherapy whose toxicities seem to be accentuated with age. Surgery is still the cornerstone in the treatment of early-stage breast cancer. Several studies have shown that breast conservation surgery followed by radiation therapy is as effective as more extensive surgery options such as mastectomy [21]. Nevertheless, some studies suggest that elderly women are less likely to be offered breast conservation surgery than their younger counterparts [22,23]. Partial surgery would be better tolerated in the short and medium terms on the aesthetic, functional and psycho-cognitive aspects [24]. For those who have a lot of co-morbidities, surgery under local anesthesia may be better tolerated than general anesthesia especially in cases of severe psycho-cognitive disorders [19]. Perioperative morbidity is low and mortality ranges from 0 to 2%, which is due to co-morbidities and not to chronological age [19]. In patients with hormone receptor-negative breast cancer, chemotherapy has been proven to have a survival benefit. A growing body of data suggests that chemotherapy also leads to improved survival in elderly patients [25]. Clinical trials show that radiotherapy after breast conserving surgery reduces breast cancer recurrence among older women with early-stage disease [26]. However, survival as well as local control is likely to be diminished for older women with life expectancies of 10 years or more with large or node-positive tumors who do not receive radiation therapy [27]. In contrast to the trends in breast surgical procedures, elderly women are less likely to undergo axillary staging procedures compared to younger women [12,13]. This is despite previous studies demonstrating that a nodal involvement retains prognostic significance in older patients with breast cancer [28] and that adjuvant treatment decisions are altered in many elderly patients based on axillary staging results [29]. Whether the absence of axillary staging necessarily results in less favorable outcomes is not very clear. Martelli et al. [30] reported a 5-year follow-up on a cohort of 219 women aged 65-80 with clinical T1N0 breast cancer who were randomized to axillary lymph node dissection versus no axillary surgery. Two women who had no axillary surgery developed an axillary recurrence, but there was no difference in overall or disease-specific mortality between the two groups [30]. Mandelblatt et al. [31] found that women aged 67 and older who underwent axillary staging, either by sentinel lymph node dissection or axillary lymph node dissection, developed three times more upper extremity complications than younger women. Those sequelae also had a larger impact on physical and mental function in the older patients [31]. The risks of axillary procedures in elderly women must be weighed against the potential prognostic information and benefits obtained. This is true for women of all ages, but may be particularly relevant in the elderly population. In the conclusions of the AMAROS [32] study, the authors considered that radiotherapy in elderly women with sentinel nodules is a good alternative in order to avoid morbidity associated with ganglion dissection. Conclusion In this study, the average age of breast cancer patients was 72.5 years, the palpable mass was the main complaint and the most common histological type was invasive ductal carcinoma. Even if the interest in geriatric gynecological pathology started early in Western countries, many questions remain unanswered on how to optimally treat this patient population. In this study, a surgical treatment was performed in 94.8% of breast cancer patients and adjuvant treatment was prescribed for 87% of women, although there is a lack of good quality evidence regarding the role of adjuvant chemotherapy. What is known about this topic 45% of breast cancers are diagnosed in women over 65; Only few aspects of breast pathologies of the elderly women in Tunisia have been treated. What this study adds Breast cancer affected mainly the 65-74 years old age group; The palpable mass was the main chief complaint; Even if data suggest that older women could not have aggressive treatment and that standard treatments cause more complications in elderly patients, surgical treatment was performed in 94.8% of breast cancer patients. Tumor is 2 cm or less across. T2: Tumor is more than 2 cm but not more than 5 cm across. T3: Tumor is more than 5 cm across. T4: Tumor of any size growing into the chest wall or skin. This includes inflammatory breast cancer. N categories for breast cancer NX: Nearby lymph nodes cannot be assessed (for example, if they were removed previously). N0: Cancer has not spread to nearby lymph nodes. N1: Metastases to movable ipsilateral axillary lymph node(s) Health Risks of Sarcopenic Obesity in Overweight Children and Adolescents: Data from the CHILT III Programme (Cologne) Sarcopenic obesity is increasingly found in youth, but its health consequences remain unclear. Therefore, we studied the prevalence of sarcopenia and its association with cardiometabolic risk factors as well as muscular and cardiorespiratory fitness using data from the German Children’s Health InterventionaL Trial (CHILT III) programme. In addition to anthropometric data and blood pressure, muscle and fat mass were determined with bioelectrical impedance analysis. Sarcopenia was classified via muscle-to-fat ratio. A fasting blood sample was taken, muscular fitness was determined using the standing long jump, and cardiorespiratory fitness was determined using bicycle ergometry. Of the 119 obese participants included in the analysis (47.1% female, mean age 12.2 years), 83 (69.7%) had sarcopenia. Affected individuals had higher gamma-glutamyl transferase, higher glutamate pyruvate transaminase, higher high-sensitivity C-reactive protein, higher diastolic blood pressure, and lower muscular and cardiorespiratory fitness (each p < 0.05) compared to participants who were ‘only’ obese. No differences were found in other parameters. In our study, sarcopenic obesity was associated with various disorders in children and adolescents. However, the clinical value must be tested with larger samples and reference populations to develop a unique definition and appropriate methods in terms of identification but also related preventive or therapeutic approaches. Introduction Obesity in children and adolescents is a growing health problem [1]. Between 1975 and 2016, the global prevalence of overweight in children and adolescents worldwide increased from 0.7% to 5.6% in girls and 0.9% to 7.8% in boys [2]. In Germany, the prevalence of overweight including obesity was 15.4% among children aged 3-17 years based on the Child and Adolescent Health Survey (KiGGS, wave 2 2014-2017) [1]. The current COVID-19 pandemic is expected to lead to a further increase [3]. In addition to the possible persistence of overweight and obesity into adulthood, along with its corresponding health consequences [4], the significantly increased risk of cardiometabolic, orthopaedic, and psychological comorbidities is problematic even in this young age group [5,6]. Children and adolescents with overweight and obesity are more likely to have cardiovascular risk factors such as high blood pressure, lipid metabolism disorders, and glucose metabolism disorders. This may lead to the development of noncommunicable diseases such as type 2 diabetes mellitus [1,7,8], endothelial dysfunction, non-alcoholic fatty liver disease (NAFLD), and musculoskeletal dysfunction [1,6,9]. The complete picture of the metabolic syndrome (MetS) is found in 6% to 39% of overweight children, depending on the underlying definition [10]. Visceral fat content, the associated secretion of so-called adipocytokines [11], and the presence of low-threshold systemic inflammation play a central role in the development of the above-mentioned comorbidities [12]. In addition, an inverse relationship between cardiometabolic risk factors and low or disproportionate (with respect to body fat) muscle mass is increasingly being described in adulthood as expression of the so-called sarcopenic obesity [13][14][15]. Classically, sarcopenia is associated with underweight due to the loss of muscle mass as well as reduced muscle strength and function. The European Working Group on Sarcopenia in Older People (EWGSOP) expanded the definition of sarcopenia to include both primary sarcopeniacharacterised by reduced muscle mass, limited muscle function, and strength at an older age and secondary sarcopenia in the context of chronic diseases, including obesity [16]. However, people with sarcopenic obesity can be of normal weight or 'only' overweight, but their relatively low muscle mass may be masked by a higher fat mass [14,15,17]. Thus, in addition to measuring handgrip strength, the muscle-to-fat ratio (MFR) is used to determine the severity of sarcopenic obesity [17][18][19]. MFR is an indicator for cardiometabolic risk factors and metabolic syndrome in adults, while it correlates negatively with waist circumference, systolic blood pressure, and blood lipid levels [19]. Additionally, MFR can also be used to assess cardiometabolic health in children [20]. Despite these initial indications, few studies to date have examined the presence of sarcopenic obesity in childhood and adolescence or possible concomitant diseases [17,18]. In order to avoid the negative health consequences of sarcopenic obesity, adequate knowledge and evidence-informed countermeasures are urgently needed. Therefore, we investigated the correlations between the presence of cardiometabolic risk factors and the occurrence of sarcopenia/sarcopenic obesity using the Children's Health Interventional Trial (CHILT III) programme, which is an outpatient weight management programme for obese children and their families. Sample Description In this study, the input data of the CHILT III programme of the German Sport University Cologne from the years 2003-2021 were used, which is a family-based, multimodal, outpatient programme for obese children and adolescents aged between 8 and 16 years [21]. Of 538 subjects, 119 (47.1% female) could be included in the analysis, as it was possible to classify sarcopenia by MFR followed by Kim et al. [22]. The MFR cut-off values were defined according to McCarthy et al. [20] (cut-off = mean value − 2SD of the MFR of the middle fifth of the BMI range). Based on this, the MFR cut-off value for sarcopenia is at 1.25 for boys of all ages, 1.1 for girls between 5-10 years, and 0.8 for girls between 10-18 years [20]. Accordingly, the diagnosis of sarcopenia was made when the MFR was below these cut-off values. Therefore, in 83 children (69.7%), sarcopenia was present according to the above criteria (see Figure 1). None of them suffered from an overt diabetes mellitus type I or II. Anthropometric Data The height of the children and adolescents was measured barefoot in cm; weight was measured in kg. A calibrated scale and a stadiometer were used for this purpose [23]. BMI was divided into percentiles according to Kromeyer-Hauschild et al. [24]. Following the guidelines of the 'Arbeitsgemeinschaft für Adipositas (AGA)', a BMI above the 90th percentile was classified as overweight and a BMI above the 97th percentile was classified as obese [25]. In addition, the BMI standard deviation score (SDS) was calculated using the least mean squares (LMS) method for non-normally distributed characteristics [25]: , and S[t] are parameters for the participants' age and sex. Waist circumference was measured in cm using a standard tape measure. The measurement was taken midway between the anterior superior iliac spine and the lowest |
rib with the participant standing upright. Using a body fat caliper (Harpender Skinfold Caliper HSK-BI, British Indicators, West Sussex, UK), skinfold thickness was measured in triplicate to the nearest 0.2 mm in triceps and subscapular according to a standardised protocol [26], and the mean of the three results was reported. Sex-and age-specific equations were used to calculate body fat percentage, as in similar studies [27][28][29]. Bioelectrical Impedance Analysis Fat mass and muscle mass were determined by bioelectrical impedance analysis (BIA; Nutriguard-MS, Data Input GmbH, Pöcking, Germany). This was determined to 0.1 kg and 0.1% with the four-point measurement. The frequency of the measurement was 50 kHz [30]. The MFR was calculated according to McCarthy et al. [20]: For the calculation of the MFR, in this study, skeletal muscle mass and fat mass were given and used by the NutriPlus software of BIA measurements [31]. Blood Pressure Blood pressure was measured oscillometrically three times using an automatic blood pressure monitor after approximately 5-10 min of rest [23]. The cuff size was chosen so that two-thirds of the upper arm length was covered. The mean value from all three calculations was calculated and documented. We classified hypertension according to the S2k guideline of the German Society for Paediatric Cardiology using age-and height-specific reference values [34]. Laboratory Parameters Blood values were taken after fasting (12 h food and drink abstinence, including plain water, no teeth brushing) and analysed in the laboratory of the German Sport University. Fasting blood glucose, total cholesterol, high-density lipoprotein (HDL), and triglycerides were measured directly. Low-density lipoprotein (LDL) cholesterol was determined indirectly from total cholesterol, HDL, and triglycerides using the Friedewald equation [35]. Insulin was determined using human insulin standards (Elecsys Insulin) from Roche Diagnostics, Mannheim [36]. The homeostatic model assessment (HOMA index) was used as a parameter of insulin sensitivity and was calculated by the following formula [23,37]: In addition, GGT (in U/L) was determined by a kinetic photometric assay using reagents from ABX Pentra (HORIBA ABX 2007). GPT and GOT (each in U/L) were determined via an optimised UV assay without pyridoxal phosphate using reagents from ABX Pentra (HORIBA ABX 2005/2007) [36]. High-sensitivity (hs) CRP (in mg/L) was determined using the cobas c system by Roche/Hitachi. Leptin was measured by a direct sandwich enzyme-linked immunosorbent assay (ELISA, kit from MERCK/Millipore KgaA, Darmstadt, Germany). Definition of Metabolic Syndrome (MetS) The metabolic syndrome was defined according to the International Diabetes Federation (IDF) classification modified for children and adolescents up to 16 years [38]. For adolescents over 16 years, we applied the IDF criteria for adults [39] (see Appendix A Table A1). Cardiorespiratory Fitness/Ergometry Maximum cardiorespiratory performance capacity (in watts) was determined by bicycle ergometry (Ergometrics er900, Ergoline, Bitz, Germany). The children started at 25 watts, and workload was increased by 25 every 2 min until the maximum load was reached. The children were encouraged to continue until they had reached their maximum physical capacity. Relative watts (watts/kg) was defined as the maximum watts in relation to body weight [21]. Children with acute illnesses, such as febrile infections, asthma attacks, or metabolic diseases were excluded from ergometry. Other contraindications include cardiomyopathies, certain vascular anomalies, and heart failure [40]. Muscular Fitness/Standing Long Jump The standing long jump was used to measure muscular fitness, resp. jumping strength, and associated leg muscle strength. This test was based on the Dordel-Koch test with defined standard values. In the test, the children and adolescents had to jump as far as possible using both legs without a run-up. The children made two attempts, and the better one was scored [41]. Statistical Analysis The data were analysed using IBM SPSS Statistics, version 28.0; descriptive statistics were presented as means and standard deviations (SD). Normal distribution of the parameters was checked using the Kolmogorov-Smirnov test. Parameters with normal distribution were examined with parametric tests. Parameters that were not normally distributed were tested using the Mann-Whitney U-test. A t-test was used to compare the means of continuous parameters with variance homogeneity. In the case of variance heterogeneity, the T-test with Welch correction was used. We tested categorical parameters using a Chi-squared test. Statistical significance was defined as a p-value < 0.05. Table 1 shows the anthropometric data of the total sample, as well as the possible differences between participants with and without sarcopenia based on the above-mentioned cut-offs. Of the children with sarcopenia (n = 83), 69.9% were male and 30.1% were female; in the group without sarcopenia, (n = 36), 13.9% were male and 86.1% were female (p < 0.001). On average, children and adolescents with and without sarcopenia were the same age, height, and weight (see Table 1). However, there were significant differences in BMI-SDS (p = 0.018) and waist circumference (p = 0.024). In children with sarcopenia, boys had a higher waist circumference (p = 0.042), higher GPT (p = 0.006), higher skeletal muscle mass, and higher SMI (p < 0.001), as well as a higher MFR (p = 0.026) than girls (see Appendix A Tables A2-A4). Laboratory Parameters and Blood Pressure Children with sarcopenia showed a significantly higher GGT (p = 0.028), higher GPT (p = 0.003), a significantly higher hs-CRP (p = 0.009), and significantly higher diastolic blood pressure (p = 0.046). The other parameters did not differ significantly among participants with and without sarcopenia (see Table 2). Muscle Mass and Cardiorespiratory/Muscular Fitness Children with sarcopenia had higher BIA fat mass (p = 0.032) and lower MFR (p < 0.001), lower physical performance (p = 0.001), and lower jumping distance (p = 0.041). Muscle mass, SMM, and SMI did not differ significantly (see Table 3). Sarcopenia and MetS Regarding the classification of the metabolic syndrome according to the modified IDF classification [38] (see Appendix A Table A1) and the presence of sarcopenia, the Chi-squared test showed no significant difference (p = 0.747). Discussion To our knowledge, this is one of the first studies to examine sarcopenia in overweight and obese children and adolescents and its association with selected cardiometabolic risk factors and exercise capacity. Of the participants, 69.7% were characterised as sarcopenic, which was associated with higher values for waist circumference, BMI-SDS, GGT, GPT, hs-CRP, and diastolic blood pressure, and lower cardiorespiratory and muscular fitness. There was no correlation between sarcopenia and systolic blood pressure, lipids, or fasting blood glucose, insulin levels, and HOMA index, or components of the MetS. However, already in this age group, indications of systemic inflammation, NAFLD, and blood pressure are shown. It may be possible that the disease value would have been clearer with a larger and less homogeneous sample (all the children and adolescents were obese). So far, the occurrence of sarcopenia or sarcopenic obesity has been analysed mainly in the context of older and/or chronically ill subjects [42]. Orkin et al. showed that BIA measures of muscle and fat mass correlate strongly with magnetic resonance imaging (MRI) measures of total psoas muscle surface area (tPMSA) and fat areas in children with obesity and NAFLD. However, they did not explicitly take into account the occurrence of sarcopenia [43]. As with older persons, there is a lack of gold standard for the definition of sarcopenic obesity in children and adolescents [22,42]. We followed the cut-off values of the MFR (cut off = mean value − 2SD of the MFR of the middle fifth of BMI range) defined by McCarthy et al. using BIA muscle and fat mass [20]. Further studies using dual energy X-ray (DEXA) for body composition showed that with a lower cut-off value (cut-off value of the mean value minus 1 SD of the MFR for the third BMI quintile), the proportion of children below this value is higher [22]. Therefore, it must be critically questioned whether sarcopenic obesity is adequately represented by the ratio of muscle to fat mass. However, in addition to the MFR, other methods for determining muscle strength, mass, and power are recommended for the diagnosis of sarcopenia [16]. While DEXA serves as the gold standard for determining muscle mass, it is significantly more time consuming and expensive than BIA. Segmental BIA has proven to be an effective and practical alternative [16,20,44] especially when using multi-frequency devices as in our analysis [30]. Chen et al. compared the results of BIA and DEXA in 1476 children and adolescents aged 7-17 years and showed high comparability in the determination of body fat [45]. Similar studies used the appendicular skeletal muscle mass (the sum of the skeletal muscle mass in all four extremities) to calculate muscle mass by Tanita BC-418MA single frequency (50 Hz) Segmental Body Composition Analyser [20]. Additionally, muscle strength in general is measured by handgrip strength [16]. Due to a lack of data on handgrip strength, we analysed the results of the standing long jump to determine muscular fitness in the lower limbs, as other studies have shown that this is a valid parameter [46,47]. In older persons, muscle performance is often determined by the short physical performance battery, the 6-min walk, or the timed get-up-and-go test [16]. For our younger population, we added cardiorespiratory fitness measured in watts/kg. However, taking into account the methodological approach, this cross-sectional analysis shows that sarcopenic obesity in children and adolescents increases the risk of systemic inflammation or NAFLD, diastolic blood pressure, and poorer cardiorespiratory and muscular fitness. In association with other inflammatory cytokines, adipokines and/or myokines such as IL-6 and TNF-alpha should be examined in larger collectives. However, first of all, a unique definition and assessment in this age group has to be developed. Additionally, future prospective studies should consider measuring the clinical significance of sarcopenia and sarcopenic obesity in children and adolescence. In addition to general primary prevention measures in kindergartens and schools to promote a healthy lifestyle (including the preservation of muscle mass), we recommend including parameters such as MFR in paediatric health examinations. This study has strengths and limitations in addition to those already mentioned. One strength is the presence of factors relevant to the assessment of sarcopenic obesity and possible associated disorders. However, in this rather small, selected group, the pubertal status was not taken into account, because the Tanner stage was not recorded. Studies have shown that puberty can reduce the risk of elevated total cholesterol and LDL cholesterol [48]. In addition, puberty involves a physiological insulin resistance of the body, which is a central element in the development of a MetS [11]. Another limitation is the use of BIA to determine body compositions as mentioned above. A determination of the body compositions by DEXA and on this basis determined could possibly lead to more precise results [17]. For further studies, especially in the development of a uniform definition of sarcopenic obesity in this age group, the use of a DEXA is recommended. As a main limitation, we have already pointed out the lack of a clear definition and methodical recording of sarcopenia in children and adolescents. The laboratory determination of GGT and GPT does not necessarily mean that NAFLD is present. Abdominal ultrasound examinations or even liver biopsies were not possible in our study. In addition to the determination of the MFR, simpler and more accessible methods for diagnosing sarcopenia should be implemented. One suggestion is the handgrip-to-BMI ratio defined by Steffl et al. [17]. Lower relative handgrip strength in children was associated with higher BMI and waist circumference [18]. Therefore, the authors recommended using the handgrip to BMI ratio to identify children at a risk of sarcopenic obesity [17]. The extent to which this would lead to different results remains speculative at present. Conclusions In summary, sarcopenia according to the used definition is present in more than two-thirds of our population of children and adolescents with overweight and obesity. In this group, sarcopenic obesity is associated with poorer cardiorespiratory and muscular fitness, elevated GGT, GPT, and hs-CRP levels, and elevated diastolic blood pressure. However, to identify the clinical value, a unique definition and methods not only based on the ration between muscle and fat mass to identify children at risk are preconditions. Subsequently, appropriate preventive and therapeutic countermeasures at an early stage have to be developed. Informed Consent Statement: Informed consent was obtained from the participants' parents involved in the study. Data Availability Statement: The data used and analysed during the current study involve sensitive patient information and indirect identifiers. As a result, the datasets are not |
available. cardiorespiratory fitness, and standing long jump of the group of sarcopenia present in girls and boys. Minimally Invasive Anterior Longitudinal Ligament Release for Anterior Column Realignment Study Design: Review of the literature. Objectives: Anterior column realignment (ACR) is a powerful but relatively new minimally invasive technique for deformity correction. The purpose of this study is to provide a literature review of the ACR surgical technique, reported outcomes, and future directions. Methods: A review of the literature was performed regarding the ACR technique. A review of patients at our single center who underwent ACR was performed, with illustrative cases selected to demonstrate basic and nuanced aspects of the technique. Results: Clinical and cadaveric studies report increases in segmental lordosis in the lumbar spine by 73%, approximately 10° to 33°, depending on the degree of posterior osteotomy and lordosis of the hyperlordosis interbody spacer. These corrections have been found to be associated with a similar risk profile compared with traditional surgical options, including a 30% to 43% risk of proximal junctional kyphosis in early studies. Conclusions: ACR represents a powerful technique in the minimally invasive spinal surgeon’s toolbox for treatment of complex adult spinal deformity. The technique is capable of significant sagittal plane correction; however, future research is necessary to ascertain the safety profile and long-term durability of ACR. Introduction Adult spinal deformity (ASD) has a high prevalence in the elderly population, such that surgical management of ASD has increased 3.4-fold among patients 60 years of age or older from 2004 to 2011. 1 ASD prevalence in a volunteer adult population with an age greater than 60 years was found to be 68% in a recent epidemiologic study. 2 Surgical correction of deformity has been found to dramatically improve quality of life in patients with ASD; in particular, sagittal balance had been found to demonstrate a linear relationship with the severity of symptoms and impairment of the quality of life. Sagittal realignment is therefore critical to maximally improve surgical and quality-of-life outcomes. [3][4][5] Traditionally, surgical correction of ASD has been performed with use of a posterior open approach. However, in recent decades, several advances have been achieved in surgical treatment of ASD through minimally invasive surgery (MIS) approaches. In the past, MIS techniques have been useful in the restoration of coronal balance but have been more limited in achieving sagittal correction. Anterior column realignment (ACR) represents a relatively new MIS technique for treatment of sagittal imbalance that provides significant correction of segmental lordosis comparable to that achievable via pedicle subtraction osteotomy (PSO). ACR consists of anterior longitudinal ligament (ALL) release with partial anterior annulotomy and anterolateral discectomy via a lateral transpsoas surgical corridor ( Figure 1). 6,7 Combining the benefits of indirect ALL release and hyperlordotic lateral cage placement, ACR can significantly restore sagittal balance with use of minimally invasive techniques or minimally invasive techniques that serve as an adjunct to traditional open approaches. Outcomes ACR can effectively restore lumbar lordosis (LL) and thereby significantly improve spinopelvic mismatch. 8,9 Complimentary posterior column osteotomies (PCOs) may be performed to increase the amount of segmental lordosis possible during ACR. 10 The amount of lordosis gained varies according to the angle of the cage used and the extent of adjunct PCO. According to the Schwab osteotomy classification system, grade I osteotomy consists in resection of the inferior facet and entire joint capsule at a given spinal level. 11 Grade II consists of removal of both inferior and superior facets of an articulation at the index level as well as the flavum ligament; other posterior elements of the vertebra, including the lamina and the spinous processes, can also be removed. 12 With the posterior elements intact, ACR can result in an increment of 10 in segmental lordosis. 13 A multicenter analysis reported that the addition of PCO to an ACR increased the segmental lordosis gain by 72.7%. 10 Grade I osteotomy can lead to 21 to 27 of angle change in segmental lordosis, whereas grade II osteotomy can afford segmental LL to match the cage lordosis, in this case 32 to 33 when a 30 cage is used. 13 In different studies, the increase in overall LL with a single-level ACR ranges from 25 to 27 . 14,15 The traditional 3-column osteotomy, or PSO, has been used as a gold standard tool for sagittal realignment and yields approximately 25 to 30 of lordosis restoration. Several studies have demonstrated similar radiographic outcomes in matched cohorts of patients with ACR versus PSO. 16 Notably, there is significant variability across procedures for ACR in terms of location, interbody choice, and technique for posterior fixation and arthrodesis. Authors have described both percutaneous and open approaches, and the influence of this on lordosis and complication rates is not well delineated in the literature. Biomechanics From a biomechanical perspective, the ACR is highly destabilizing due to ALL release but also facilitates the placement of a large-footprint lordotic invertebral device. It is critical to emphasize that the ALL provides significant stability in extension and has contributions during lateral bending and axial rotation. 17 As a result, the requisite ALL transection significantly reduces the stability of the spine. 8 This instability may, to a certain extent, be counteracted by the addition of a hyperlordotic interbody device (anterior column lengthening), anterolateral fixation via plating, and posterior segmental fixation. 18 Nonetheless, it is likely that the ACR holds a theoretical advantage in terms of biomechanical stability and rod protection compared with open techniques due to stress reduction and anterior load-sharing capacity. The combination of the widefootprint intervertebral support, anterolateral plate fixation, and the anterior lengthening effect of ACR may ultimately grant improved stability and facilitate load-bearing capacity through the anterior column. 18 Earlier finite element modeling studies have suggested an inherent biomechanical advantage of the ACR construct over a PSO deformity construct. In their study, Januszewski et al 19 found that the anterior lengthening approach of ACR results in less acute rod bending and may contribute to decreased posterior rod strain. This can be compared with a lack of anterior lengthening effect and the posterior shortening feature of the PSO, which leads to hyperacute sharp-contoured posterior rods, as opposed to ACR, and makes it more vulnerable to rod fracture because of the strain concentration. A recent study investigating the risk of rod fracture across ACR constructs demonstrated a 4% rate of rod fracture with 1-year follow-up, potentially representing a lower risk of rod fracture. 20 However, to date no cadaveric biomechanical studies of ACR have been performed to corroborate these findings. Target/Patient Selection Although ACR is an MIS technique that is growing in popularity with the capacity for significant sagittal correction, it is important to keep in mind the limitations of this procedure. Appropriate patient selection remains paramount in regard to patient safety and health outcomes. Several publications have suggested treatment algorithms for approaching the management of ASD with MIS techniques. 21 The ACR technique, while capable of significant sagittal plane correction, is limited to free segments and is constrained by lumbosacral plexus anatomy and risk of neurological deficit from direct trauma or retraction injury. Those patients with severe deformity are less likely to achieve restoration of sagittal alignment and clinical improvement. 22 Patients with mild to moderate sagittal deformity are candidates for ACR, while those with fixed or severe deformity are typically not considered surgical candidates for ACR ( Figure 2). The minimally invasive spinal deformity surgery algorithm has suggested that treatment with MIS benefits patients who meet the following parameters: LL-PP mismatch less than 30 , pelvic tilt less than 25 , thoracic kyphosis less than 60 , sagittal vertical axis less than 6 cm, and Cobb angle less than 20 . 21 Patients who have sagittal vertical axis of more than 6 cm but flexible curves are also candidates. Restoring LL and sagittal balance also takes precedence over scoliosis correction during ASD surgery. 23 Evaluation for appropriateness, including thorough preoperative radiographic assessment, is necessary to ascertain the working surgical corridor at the level of interest in terms of vascular, neurological, and bony anatomy. Imaging modalities, such as computed tomography (CT) and magnetic resonance imaging (MRI), can provide information regarding the anatomy of the great vessels, segmental vessels (that may preclude safe docking of retractor or annular release contralaterally), or potentially scarred vessels to annulus. Notably, a preoperative MRI provides reliable information about the location of the vessels anterior and the lumbar plexus ( Figure 3). Patients with ASD may have displacement of the psoas muscle and the surrounding neurovascular structures. Additionally, we recommend evaluation of segmental vasculature on sagittal MRI to identify large segmental arteries or veins across the disc space that may represent a surgical bleeding risk. Contraindications for ACR are shared with any lateral transpsoas approach. These include a history of retroperitoneal surgery, presence of scar tissue on this region, and evidence of fusion across the disc space ( Figure 4). Patients should undergo bone-quality evaluation with dual-energy X-ray absorptiometry before ACR. Severe osteopenia and osteoporosis, with a T-score less than À2.0 at the femoral neck, should also contraindicate lateral ACR, because intervertebral subsidence represents an important concern and can cause failure of the treatment. 12 Surgical Technique The ACR surgical technique consists in using the traditional lateral transpsoas approach described by Akbarnia et al. 6 Access to the lateral spine is obtained via the traditional lateral transpsoas corridor. Using radiography or neuronavigation, the location of the spinal level of interest is identified and marked on the surface of the skin. Lateral incision is performed overlaying the disc space, followed by subcutaneous dissection, facial opening with coagulation, blunt abdominal wall muscle dissection sparing the traversing neurovascular structures, retroperitoneal dissection, and safe identification of the lateral aspect of the lumbar spine. After confirmation of appropriate location and verification of safe dilation, a tubular dilator is positioned and rigidly fixed to the surgical bed to facilitate access to the lumbar spine. Next, a blunt dissection tool is placed anterior to the ALL. The discectomy is then performed in the standard fashion: (1) annulotomy, (2) contralateral annulotomy, and (3) discectomy. Next, blunt dissection is performed by progressive blunt dissector advancement, carefully developing the plane between the adventitia tunica of the anterior vessels and the ALL, which has been previously studied by means of MRI to discover a safe plane. Once a safe plane has been developed, a sharp blade is advanced over the grooved anterior retractor to perform sharp dissection of the ALL with radiographic verification. Once sharp release of the ALL has been performed on approximately 75% of the ALL, the retractor is removed, and a manual indirect distracting tool is placed ( Figure 5). Under fluoroscopic guidance, the distracting tool is advanced and progressively opened to complete the ALL release indirectly. Oftentimes, an audible pop will be heard, and a sudden loss of resistance will be felt. It is critical at this point to visually inspect for bleeding. Following discectomy, a hyperlordotic interbody device implant is placed and fixated anterolaterally with either 1 or 2 screws ( Figure 6). Although debate exists regarding the use of a buttress plate (which requires 1 screw) or plate fixation (which requires 2 screws), in our experience, plate fixation does not significantly limit achievable lordosis. The fascia overlying the abdominal musculature and skin incision are reapproximated in standard fashion to avoid abdominal wall herniation. Posterior construct supplementation with pedicle screws and rods should follow ACR, and posterior osteotomy can be used to increase lordosis gain if needed (Table 1). 12 Technology Minimally invasive surgery, particularly lateral corridor surgery, is heavily reliant on imaging and monitoring technology. Reliable and accurate neuromonitoring is a prerequisite for safe and efficient lateral access surgery. Due to the likely prolonged retraction on the psoas muscle, identification and protection of the lumbosacral plexus is of paramount importance in reducing the incidence of neural injury. Use of imaging technology is crucial, and fluoroscopy, low-dose image-enhanced fluoroscopy, [24][25][26] or potentially combined CT and radiographic imaging can be of assistance. 27 Neuromonitoring. Electrical neurophysiological monitoring is a necessary part of minimally invasive lateral transpsoas approaches to the lumbar spine. The nerve roots that contribute to the lumbar plexus and the genitofemoral nerve are at risk of injury when the surgeon crosses the psoas muscle with dilators and also during the retractor positioning. |
The nerves forming the plexus run obliquely outward, enclosed by the psoas muscle, from a dorsal situation at L1 to a ventral situation at L5; at the lower levels, they are heavier and denser, the roots conjoin, and the main motor branches originate. 28,29 The use of directional dynamically triggered electromyography (t-EMG) with discrete threshold responses is a highsensitivity method that has been associated with a decrease in the neural complication rate from 30% to less than 1%. 30,31 Use of t-EMG provides threshold results of lower limb nerve root function on bilateral vastus medialis, tibialis anterior, biceps femoris, and medial gastrocnemius myotomes while manipulating the psoas muscle, spanning responses from the L2-S2 spinal nerves. t-EMG is integrated with dilator apparatuses and can provide rotational 360 information on the position and distance of the motor nerves surrounding the instrumentation, where a lower response threshold indicates closer nerve proximity compared with a higher response threshold. Once docking has safely proceeded, plexus injury may still occur, induced by compression or distention by the retractor blade, which causes microvascular and electrophysiological changes that may result in neurological deficit or pain. 32,33 The magnitude and duration of the distortion are decisive in the occurrence and severity of iatrogenic nerve injury. 34 The retractor blade can contain a t-EMG test to quantify changes in threshold throughout the surgery and can indicate whether an ongoing compression and retractor should be released. 35 Spontaneous unremarkable (free-run) EMG monitoring has relatively low sensitivity to sustained retraction of nerve roots. t-EMG provides real-time electrophysiological feedback of the situation of the nerves. In addition to allowing the anatomy of the lumbar plexus and safety zones to be perceived, t-EMG increases the elementary basement for a safe navigation and is an integral and necessary part of ACR. 29 Imaging. Similarly, imaging is critical for performing the ACR safely and effectively. Fluoroscopy is considered the gold standard for ACR during positioning, marking the skin, confirming the level, and implanting. However, increased radiation exposure to the patient and surgeon is a disadvantage associated with MIS. Even with frequent use of fluoroscopy, there is a risk for inappropriate cage positioning. Spinal navigation with 3-dimensional CT is an option to decrease the incidence of implant misplacement and amount of radiation exposure. 27 Once discectomy and endplate prep have been achieved, a navigated trial is inserted under stereotactic guidance for sizing. The cage is then attached to a navigated holder that allows insertion with image guidance. Navigation can also be used during the psoas muscle phase for positioning the retractor. In multilevel surgery, a significant concern is shifting the anatomy after the first cage insertion, which may potentially cause decreased accuracy. In such cases, use of fluoroscopy is helpful to verify accuracy. Complications A recent study from the International Spine Study Group that reviewed 953 patients treated surgically for ASD reported a major complication rate of 7.6%. 36 Major complications were 106S Global Spine Journal 10(2S) defined according to Carreon et al, 37 with the most common complications being excessive blood loss (defined as the loss of >4 L of blood), deep wound infection, and pulmonary embolism. Although the ACR is an MIS technique that may potentially decrease complications related to large exposures and extensive soft-tissue disruption, overall complications can occur. Notably, ACR is associated with significantly less blood loss than traditional 3-column osteotomy. 16,38 Most important, neurological deficit associated with lumbosacral plexus neuropraxia caused by psoas muscle retraction represents an important concern with use of the lateral transpsoas approach. In the largest series of cases in the literature, among 600 patients who underwent surgery with the lateral transpsoas approach, only 0.7% had neurological deficits, and these deficits were transient in all cases. 39 However, in the same study, 62.7% of patients experienced some postoperative thigh-related symptoms. 39 Prolonged retraction time is a predictor of decreasing nerve function and decreasing retraction time, and use of t-EMG throughout retraction may reduce the incidence of this complication. 40 The majority of postoperative transpsoas thigh symptoms are temporary; half of those patients with such symptoms will recover in 3 months, and 90% will recover within 1 year. 39,40 Surgeons must be experienced and familiar with the retroperitoneal and abdominal anatomy when performing lateral approaches to safely identify and separate the ALL from the anterior structures before releasing it. The direct and indirect ALL transection is an extremely advanced technique that exposes spine surgeons to unfamiliar regional anatomy, risking injury to the autonomic plexus, visceral organs, or the great vessels (femoral artery and femoral vein). 41 Visceral or major retroperitoneal vessel injuries are rare but are some of the most feared complications. 42 Bowel injuries have been previously reported in the literature and require prompt repair via laparotomy and potentially colostomy to avoid peritonitis, which can be fatal. 43 Injury to the great vessels can also occur, which demands immediate repair with assistance from a vascular surgeon. The most common mechanical complications in patients with ASD are proximal junctional kyphosis (PJK) and proximal junctional failure (PJF). 44 Others causes of revision surgery are pseudarthrosis, rod fracture, and loss of correction. 45 PJK is a concern in ASD surgery regardless of the approach used, and its prevalence is 20% to 40%. 46 PJK results from failure of the proximal junction caused by mechanical stress concentration at the upper instrumented level and its adjacent level. Consequently, the posterior ligament complex, bone elements, and instrumentation are affected at this site. Age greater than 65 years is a major risk factor for PJK, and other factors that may also be associated with PJK include disruption of the posterior tension band, bone quality, severity of preoperative deformity, extent and rigidity of construct, and incomplete realignment. [46][47][48][49] Gandhi et al 47 reported a series of cases in which, among patients with ACR only, the incidence of PJK was 30% and that of PJF was 11%. Among patients with ACR plus PCO, the incidence of PJK was 42.9% and that of PJF was 40%. The addition of PCO is associated with an increased incidence of PJK and PJF; nevertheless, the patients who require ACR plus PCO are usually those who have the most severe sagittal imbalance and require greater reduction. Recent investigation into combining ACR with adjacent PSO did not report any events of PJK or PJF in an average 1.9-year followup; importantly, however, this was not the primary objective of the study. 50 Additional complications have also been reported. These include rod fracture and subsidence. A retrospective review from several institutions found rod fracture in 4.4% of cases treated with ACR at a mean of 8.6 months after surgery across 98 patients (follow-up range, 1-7.4 years), with the length of the surgical construct an independent predictor of rod failure. 20,51 Subsidence of the interbody graft into the adjacent vertebral bodies is another risk associated with use of lateral approaches and can result from high mechanical stress at the endplate. Subsidence of the interbody graft can cause severe pain, impaired fusion, and even fracture of the vertebral body itself. Over-distraction, insufficient cage width, and endplate violation have been described in the literature as potential risk factors. 52,53 An increased incidence of subsidence has been reported among patients with impaired bone mineral density, but ACR is feasible in this population. 54 All patients who will undergo ACR should be screened with dual-energy X-ray absorptiometry. Pearls and Pitfalls of ACR The ACR technique represents a powerful, recently introduced MIS option to address ASD and has a strong potential for correction of sagittal imbalance and restoration of alignment. It is, however, an advanced technique and should be performed by surgeons experienced with lateral approaches to the spine. The surgeons who perform ACR would ideally be familiar with the lateral approach and retroperitoneal anatomy, and they also must be prepared to control, or at least slow, bleeding in case of vascular tearing (affecting major or minor vessels). Once the retractor is docked, the surgeon should be able to operate efficiently and promptly to decrease the opened retractor time to avoid or limit the extent of neurological deficits. Preoperative evaluation of patient-specific anatomy with MRI to ensure a safe surgical corridor is paramount, particularly in assessing the potential for adherence of vascular structures to the ALL. In addition, a requisite for ACR is a nonfused disc space; in this regard, we recommend evaluation of the intended disc space with both standard radiography and CT. Although performing a discectomy for a standard lateral approach is possible with angled instruments, we advise against use of angled instruments for ALL release. Therefore, evaluation of the crest line in relation to the L4-5 level on standing anteroposterior radiographs is crucial to assess feasibility of an L4-5 ACR. Conclusions ACR represents a powerful technique in the minimally invasive spinal surgery armamentarium for sagittal plane correction in ASD or as a less invasive adjunct to traditional open techniques. This technique is heavily dependent on technology but represents a fascinating and exciting new chapter in spine surgery. Although studies demonstrate good radiographic outcomes with a complication profile comparable to that associated with traditional 3-column techniques, further investigation is necessary to better characterize and develop this technique. There’s Plenty of Room at Bottom of Biostatistics and Biometrics Renowned physicist Richard Feynman delivered a talk in 1959 at Caltech with title “There’s plenty of room at bottom”. This talk was criticized for almost thirty years and it was hard to believe until early 1990. At present, this unusual talk of Feynman is considered as the foundational talk on “nanotechonology” [1]. Being unfamiliar of biological terminologies, it is customary for me to use mathematical terminologies. Readers of this article may find opinions of this article absurd for the time being, but we shall try to develop some sketches of ideas for future in these directions. Opinion Renowned physicist Richard Feynman delivered a talk in 1959 at Caltech with title "There's plenty of room at bottom". This talk was criticized for almost thirty years and it was hard to believe until early 1990. At present, this unusual talk of Feynman is considered as the foundational talk on "nanotechonology" [1]. Being unfamiliar of biological terminologies, it is customary for me to use mathematical terminologies. Readers of this article may find opinions of this article absurd for the time being, but we shall try to develop some sketches of ideas for future in these directions. Most of statistical ideas deal with quantitative analyses. Qualitative analyses are often difficult for existing statistical methodologies to deal with. For example; if we compare quantities relative to attributes, then it is difficult to do statistical analyses. Thus, it must be assumed that presence of uncertainties in data with respect to attributes make tasks of statistical analyses most difficult. Biological data are related to attributes, which are present in terms of hormones, symptoms, molecules etc. So, there are scopes of development of statistical ideas and their applications in biology and associated areas. Prior to 1999, there were four mathematical tools to deal with uncertainties viz. probability theory, fuzzy set theory, rough set theory and the theory of interval mathematics [2]. Fuzzy set deals with membership values. Membership values are often difficult to consider and varies from person to person, which yields problems in decision making, data analyses etc. Thus, the faults which are present in Fuzzy set, led Molodtsov [3] to introduce soft set theory in 1999. Definition 1 [3] A soft set F A on the universe U is defined by the set of ordered pair Here, A f is called an approximate function of the soft set x may be arbitrary. Some of them may be empty, some may have non-empty intersection. We will denote the set of all soft sets over U as ( ). Above definition clearly shows presence of attributes, words, adjectives, [0,1] etc. One may refer to Acharjee [4] for extensive review on soft set theory and its hybrid structures for researchers other than mathematicians. If we consider Pearson's correlation coefficient, then difficulties arise related to data in terms of attributes. Either two attributes may be similar, "semantically" similar, or dissimilar. Statistical methodologies fail to distinguish between these three characteristics of attributes. Hence, can we say that biostatistics has scopes to adopt new statistical methodologies? Of course "yes". Prior to Acharjee and |
Mordeson [5], many statistical methodologies were developed in terms of uncertainties of mathematics viz. fuzzy set, intuitionistic fuzzy sets etc. These methodologies have focused mostly on degrees of membership, degree of non-memberships etc. with modified statistical formulations. Every soft set can be represented in tabular form in terms of 0 and 1. In [5], Acharjee and Mordeson connected attributes with statistical methodologies in the presence of binary digits 0 and 1. Though, significance of these methodologies has been shown in human trafficking, yet the application part shows that data were divided by attributes related to vulnerability and government responses in "The Global Slavery Index, 2016". In a similar manner, biological data are often related with symptoms, drugs, hormonal activities, molecules etc. So, extensive analysis for the development of soft statistics is seemed necessary for biostatisticians. In the meanwhile, Acharjee and Mordeson [5] have included of mathematical ideas from economics viz. utility theory, to connect human choice behaviour and its significance in statistical analyses are shown. Let us think about occurrence of a biological activity, and a researcher has preference among the attributes, which are responsible for the activity. Acharjee and Mordeson [5] developed methodologies by using preference, which reduce entire data to the specific data which signify preference and attributes simultaneously. Let us consider a case of biological activity, where some causes of this activity are unknown and a researcher will have to guess about the activity in terms of attributes and degree of membership. Thus, fuzzy soft set is the suitable tool to deal with Biostatistics and Biometrics Open Access Journal ISSN: 2573-2633 this biological activity. But, fuzzy statistical methodologies are not suitable tool to deal with this case. Moreover, there is a lack of methodology to analyze this case. So, Acharjee et al. [6] have used fuzzy -cut to develop fuzzy soft statistical ideas with utility. For the biologists, let us assume that α-cut is the restriction in the data in the sense of a layman, which they will have to reduce for their analyses or they will have to overcome in the course of time to have the necessary data from a huge data. Similarly, presence of degree of non-membership as well as degree of membership related to attributes together yields intuitionistic fuzzy soft statistics [7]. Thus, we can see that classical statistical methodologies may fail in many cases, where qualities are present. Biostatisticians should focus on these tools and think to develop these areas for the betterment of human race. In biometrics, ideas of similarity play crucial roles. During measures of similarities between two objects, various existing definitions focus on common domain of two objects relative to each of them independently. The value of similarity measure lies between 0 and 1. It is often observed that certain negligible changes in the objects yield difficulties during recognition of biometrics. These types of difficulties can be dealt with similarities of soft sets. In [8], Acharjee encountered some errors and redefined similarity measures of soft set based on Kharal's definitions [9]. Soft similarity and associated structures have scopes to find similarities between the face of a person after burn and his original face before the burn. There are many cases similar to this, where biometric processes fail to provide suitable methods of recognitions. These problems are to overcome for the purposes of defence security system of a country. Thus, if one develops biometric tools and methodologies based on attributes, then soft similarities and associated structures will be helpful to indentify even after some deformations of one face or thumb etc. Significance of our justifications can be understood from the prospective of wounded face of a defence officer and security password related to confidential top secret's defence account, which is nothing but his original face before the wound. Hence, improvisations of security systems are mandatory. At the end of this article, we want to focus on aspects of nanotechnology for biometrics. Uses of nanotechnology in biometrics can be found in [10]. Multisets deal with multiplicity of elements, thus theoretical studies on nanotechnology have scopes to use soft multiset topology. Without going for mathematical details, intuitively we can say that, if number of atoms matter in nanotechnology, then multiset theoretic topological study will gave the theoretical knowledge about behaviour of atoms in nano level of nanotechnology. In similar manner, we can say same for soft multiset topology. Thus, we want to conclude with above sketches of ideas which may be beneficial for biostatisticians and we hope that this short paper may have some though provoking impacts in future. Understanding the impact of lockdowns on short-term excess mortality in Australia During 2020 and 2021, Australia implemented relatively stringent government restrictions yet had few COVID-19 deaths. This provides an opportunity to understand the effects of lockdowns and quarantining restrictions on short-term mortality and to help provide evidence in understanding how such public health policies can impact on health. Our analysis is based on preliminary mortality data collected by the Australian Bureau of Statistics. Rates were estimated by disease and over time and compared with mortality statistics in the period 2015–2019. Comparing deaths in 2020-2021 with 2015–2019 show the annual mortality rate (per 100 000 people) fell by 5.9% from 528.4 in 2015–2019 to 497.0 in 2020–2021. Declines in mortality are across many disease categories including respiratory diseases (down 9.4 deaths per 100 000), cancer (down 7.5 deaths per 100 000) and heart disease (down 8.4 deaths per 100 000). During 2020 and 2021, Australian age-standardised mortality rates fell by 6%. This drop was similar for men and women, and was driven by a reduction in both communicable and non-communicable causes of death. Such evidence can help inform public health policies designed to both control COVID-19 and other infectious diseases. INTRODUCTION A recent commentary 1 in this journal has highlighted the ongoing debate about the overall impact of 'lockdowns' which they define as a restrictive set of non-pharmaceutical interventions aimed at preventing spread of COVID-19. An important element of this debate is the impact of these measures on overall mortality. While lockdowns have proved effective in reducing transmission 2 and deaths from COVID-19 3 some have argued it comes at a high cost. For example, signatories of the Great Barrington Declaration 4 argued that 'Current lockdown policies are producing devastating effects on short and long-term public health… [including]… worsening cardiovascular disease outcomes, fewer cancer screenings and deteriorating mental health'. The Declaration called on governments to end stay-at-home orders, and focus on building herd immunity in the broader population. One of the countries that is highlighted in the commentary 1 is Australia, which had lockdowns without experiencing large numbers of COVID-19 cases and deaths in the first 2 years of the pandemic (this changed in 2022, after lockdowns were lifted). By the end of 2021, Australia had fewer than 100 COVID-19 deaths per million people, less than one-sixth of the worldwide death rate. 5 Yet while Australia's COVID-19 death rate was relatively low during that period, its movement restrictions were relatively stringent. Across 34 advanced nations, Australia had the seventh-most stringent lockdown in 2020 and 2021 (see online supplemental appendix 1). In those years, Australia is unusual in having both a low death rate from COVID-19 and stringent lockdowns. While previous international comparisons noted Australia 'did not have large numbers of excess deaths', 6 the country has not been included in many previous international comparisons due to incomplete data. To address this gap, we explore excess deaths in Australia using recently released mortality data collected by the Australian Bureau of Statistics. This allows us to understand both disease and temporal patterns, which SUMMARY BOX ⇒ Analysis of short-term excess mortality suggests that lockdowns are not associated with large numbers of deaths in Australia. ⇒ This decline in short-term mortality was due to declines in both communicable and non-communicable causes of death. ⇒ Google mobility data indicates that the drop in deaths tracked reductions in movement outside the home. ⇒ Virtual working and other online activities may be a means to reduce mortality from both communicable and non-communicable causes of death. BMJ Global Health provide insights into short-term impacts of lockdowns on mortality. Our analysis is based on provisional mortality data collected by the Australian Bureau of Statistics. For allcause mortality, these data cover the full universe of deaths. For cause-specific mortality, the data cover only doctor-certified deaths (excluding the 11%-14% of deaths that are certified by a coroner). 7 Full details of our data are provided in online supplemental appendix 2. Our analysis compares deaths in 2020-2021 with deaths in the period 2015-2019 using age-standardised death rates. Following the Australian Bureau of Statistics, we refer to this as 'excess mortality', but it is important to acknowledge that it represents only a simple comparison of two time periods. In this sense, our measure of excess mortality differs from the WHO's measure of excess deaths, which also takes into account factors such as the average ambient temperature. As well as looking at the entire period, we also study mortality rates by week, enabling us to analyse the relationship between mortality and population mobility. To better understand the fall in mortality, it is instructive to look at how it tracks government lockdowns and population movements. For this purpose, we use one indicator of the impact of lockdowns, Google Community Mobility reports, 8 which show daily mobility trends in six categories of places: grocery and pharmacy, parks, transit stations, retail and recreation, residential, and workplaces. For each category, these figures are expressed as a deviation from the baseline in early-2020, and account for day-of-week effects. More detail about these data are provided in online supplemental appendix 2. Table 1 shows that the annual age-standardised mortality rate (per 100 000 people) fell from 528.4 in 2015-2019 to 497.0 in 2020-2021. This represents a 5.9% fall in the mortality rate. This figure is similar for males and females (for an explanation of why the percentage change for persons is slightly smaller than the figure for either males or females, see online supplemental appendix 2). Looking at causes of death, we find declines in mortality across the board. In absolute terms, the mortality drop is largest for respiratory diseases (down 9.4 deaths per 100 000), cancer (down 7.5 deaths per 100 000) and heart disease (down 8.4 deaths per 100 000). In proportional terms, the mortality drop is largest for respiratory diseases (down 20.5%) and the respiratory disease subcategory of influenza and pneumonia (down 41.8%). MORTALITY TRENDS Strikingly, while Australia experienced an average of around 600 influenza deaths each year in 2015-2019, influenza claimed fewer than 50 lives in 2020. In 2021, fewer than five Australian deaths were recorded from influenza (see online supplemental appendix 2 for details). In figure 1, we analyse the seasonal nature of these effects, plotting weekly deaths across the year. During 2015-2019, the weekly death rate shows a marked seasonal effect, rising from around 9.5 deaths per 100 000 in summer to around 11.5 deaths per 100 000 in winter. By contrast, the years 2020-2021 saw less seasonality in Australian deaths, with the weekly mortality rate averaging around 9 deaths per 100 000 in summer and 10 deaths per 100 000 in winter. Figure 2 shows the six mobility trends indicators, with a horizontal line depicting the baseline. As these data show, the COVID-19 pandemic significantly reduced visits to retail and recreation, parks, transit stations and workplaces, and substantially increased the amount of time spent in residences. The decreased time in public places and workplaces coincided with the two lockdown periods: April-May 2020 (when all Australians were BMJ Global Health under lockdown) and August-September 2021 (when two-thirds of Australians were under lockdown). Visits to grocery stores and pharmacies show a slightly different pattern: spiking as Australians stocked up at the start of lockdowns, and steadily rising over the 2 years (perhaps reflecting increased demand for medicines). Alongside these mobility indicators, we show the share of the Australian population under lockdown. We also show how age-adjusted mortality compared with previous years. To do this, we take the age-adjusted death rate for each week in 2020 and 2021, and estimate the ratio of that mortality rate to the mortality rate for the corresponding week in the period 2015 -2019. Thus, a ratio of 1.1 indicates that the |
death rate was 10% higher than in prior years, while a ratio of 0.9 indicates that the death rate was 10% lower than the historical average. This chart shows that the death rate fell sharply as social distancing increased. By mid-2020, the death rate was around 10% below its historical average. The death rate then returned close to its historical average in late-2020, before dropping (that is, improving) in late-2021, a period in which Australia's largest two states were under lockdown. How does mobility track mortality? In figure 3, we present scatterplots of the mobility levels (relative to baseline) against the death rate (relative to prior years). Each dot denotes the figures for a single week. These data show that higher rates of activity in retail and recreation, grocery and pharmacy, transit stations, and workplaces were associated with higher mortality rates, while greater levels of activity in parks and more time spent in residences were associated with lower levels of mortality. The simple bivariate regressions are significant at the 1% level for retail and recreation, grocery and pharmacy, transit stations, workplaces and residences. The association between time spent in parks and mortality is statistically insignificant. Finally, we show the relationship between mortality and the share of the population under official lockdown. Consistent with the mobility data, this relationship is negative and statistically significant at the 3% significance level. EXPLAINING THE MORTALITY DECLINE Australia was among a small number of countries that saw a decline in overall age-standardised mortality during 2020-2021. This decline in mortality was due to declines in both communicable and non-communicable causes of death. Using Google mobility data, we provide suggestive BMJ Global Health evidence that the drop in deaths tracked reductions in movement outside the home. In the short term, government lockdowns and social distancing appear to have significantly reduced overall mortality at least in the short term, which may help shape future public policy. While the reduction in mortality from infectious diseases can explained by social distancing measures, the reduction in deaths from non-communicable conditions such as cancer and heart disease are more surprising. In the case of cardiovascular disease (which contributes to around one third of the reduction in deaths), possible explanations could be people spending greater time indoors over winter months (which are traditionally known to have high excess deaths), 9 faster access to emergency services, 10 lower pollution 11 and a reduction in cardiovascular events following influenza. 12 Explanations for the reductions in cancer deaths are more difficult to identify and might be due to potential limitations in the coding of deaths for people with multimorbidities. 13 A recent analysis indicates that unlike other countries COVID-19 did not impact on provision of many healthcare services. For example, a recent analysis has indicated that COVID-19 pandemic did not disrupt cancer care in Australia. 14 It is possible that reductions in influenza may also have played a role, given evidence that influenza mortality rates are higher among cancer patients than the general population. 15 It will be important to continue track mortality in countries such as Australia to see if there are long-run mortality impacts of lockdowns, for example, from the disruption of breast cancer screening services 16 or declining levels of physical exercise 17 both of which may lead to an increase in mortality rates over the longer term. Although our analysis of overall deaths includes coroner-reported deaths, our cause-specific mortality figures are based only on doctor-certified deaths, excluding coroner-reported deaths and so exclude deaths from suicides, accidents and assaults (see online supplemental appendix 2). A recent international analysis of suicides in the early stages of the pandemic (which included data from Australia) has shown that rates 'remained largely unchanged or declined in the early months of the pandemic'. 18 Future research should look separately at suicides and homicides when full data become available. It should also be noted that excess deaths can be calculated using a variety of different methodologies which can produce some variation in results, although consistently 2020 is a year of below average mortality for many diseases. 19 Further, the Australian Bureau of Statistics is not the only source of data on deaths. The World Mortality Dataset 20 is a valuable resource used by numerous international organisations to compare excess deaths across countries that could be used in future analyses. Another caveat is that we are considering only the impact on mortality and do not capture the impact on morbidity from poor mental health or from family, domestic and sexual violence. 21 For example, a recent Australian study using a quasi-experimental design has shown that lockdowns were 'associated with a modest negative change in overall population mental health'. 22 Quality-adjusted life-years could be a useful broader metric, capturing both the negative and positive health impacts of home working and virtual interactions in a single measure. This could provide the basis of an evaluative framework BMJ Global Health of COVID-19 pandemic response policies. 23 Additional impacts of lockdowns such as disruption to economic activity and schooling also need to be considered. Beyond the pandemic, it is also worth considering whether a move to virtual working and other online activities may be a means to reduce mortality from a range of diseases such as respiratory infections during the winter months. One must also weigh against the potential benefits in mortality, the impact on other aspects of health such as general well-being from social interaction. While large-scale randomised studies 24 have shown that home working can increase productivity, the potential for it as a health intervention in randomised studies has received less attention. 25 Given the mortality benefits observed here at a population level, it would worth exploring if the health and social impacts of home working and virtual activities could be evaluated using large-scale randomised trials. CONCLUSION Our results do provide some evidence that, at least in terms of short-run mortality, some of the concerns that led to the Great Barrington Declaration have not eventuated. Comparisons of deaths in 2020-2021 with 2015-2019 shows a drop in the annual mortality rate across many disease categories including respiratory diseases, cancer and heart disease. Hence, social distancing measures in Australia may have averted thousands of non-COVID deaths during the first phase of the pandemic. Understanding how patterns of mortality are impacted by public health interventions such as lockdowns can help provide a better evidence base for dealing with infectious diseases such as COVID-19, which is likely to remain endemic for the foreseeable future. Contributors AL conceived the idea, acquired the data and undertook the analysis and contributed to the writing-up. PC contributed to the interpretation of the analysis and the writing up. Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Competing interests None declared. Patient consent for publication Not applicable. Ethics approval As this study is based on publicly available databases without any identifying individual information, ethical approval was not needed. Provenance and peer review Not commissioned; externally peer reviewed. Data availability statement The data in this study were obtained from publicly available databases. Replication files are available on request from the authors. Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise. Treatment of patients with aortic atherosclerotic disease with paclitaxel-associated lipid nanoparticles OBJECTIVE: The toxicity of anti-cancer chemotherapeutic agents can be reduced by associating these compounds, such as the anti-proliferative agent paclitaxel, with a cholesterol-rich nanoemulsion (LDE) that mimics the lipid composition of low-density lipoprotein (LDL). When injected into circulation, the LDE concentrates the carried drugs in neoplastic tissues and atherosclerotic lesions. In rabbits, atherosclerotic lesion size was reduced by 65% following LDE-paclitaxel treatment. The current study aimed to test the effectiveness of LDE-paclitaxel on inpatients with aortic atherosclerosis. METHODS: This study tested a 175 mg/m2 body surface area dose of LDE-paclitaxel (intravenous administration, 3/3 weeks for 6 cycles) in patients with aortic atherosclerosis who were aged between 69 and 86 yrs. A control group of 9 untreated patients with aortic atherosclerosis (72-83 yrs) was also observed. RESULTS: The LDE-paclitaxel treatment elicited no important clinical or laboratory toxicities. Images were acquired via multiple detector computer tomography angiography (64-slice scanner) before treatment and at 1-2 months after treatment. The images showed that the mean plaque volume in the aortic artery wall was reduced in 4 of the 8 patients, while in 3 patients it remained unchanged and in one patient it increased. In the control group, images were acquired twice with an interval of 6-8 months. None of the patients in this group exhibited a reduction in plaque volume; in contrast, the plaque volume increased in three patients and remained stable in four patients. During the study period, one death unrelated to the treatment occurred in the LDE-paclitaxel group and one death occurred in the control group. CONCLUSION: Treatment with LDE-paclitaxel was tolerated by patients with cardiovascular disease and showed the potential to reduce atherosclerotic lesion size. ' INTRODUCTION Severe atherosclerotic aortic disease is a life-threatening condition that markedly increases the risks of thromboembolic accidents such as stroke, acute myocardial infarction and obstruction of peripheral vessels, especially in elderly subjects with large, complex lesions (1,2). In this setting, the use of non-invasive energetic treatments to reduce lesion size and inflammation is essential for the prevention of subsequent cardiovascular events. The most potent anti-proliferative drugs currently available are chemotherapeutic agents used for cancer treatment. However, the systemic use of these drugs at high doses for the treatment of atherosclerotic cardiovascular diseases (3) is unlikely due to their significant, often life-threatening toxicity. Nonetheless, the toxicity of such agents can be strongly diminished by the use of optimized drug-delivery systems. In a pioneer study performed on patients with acute leukemia, Maranhão et al. (4) reported the potential of a cholesterol-rich non-protein nanoemulsion (LDE) as a drug targeting agent. LDE particles have lipid compositions and structures that resemble low-density lipoprotein (LDL) and can be injected directly into the bloodstream. When LDE particles come into contact with plasma, they acquire exchangeable apolipoproteins from native lipoproteins, such as apolipoprotein (apo) E, which binds the particles to LDL receptors (5). In neoplastic cells, lipoprotein receptors are overexpressed (6), such that uptake of native LDL and of LDE particles is increased relative to that in normal tissues. It was also shown that the uptake of LDE particles by the aortas of cholesterol-fed rabbits is increased in comparison to normal aortas (7) and that rabbit-grafted hearts take up the nanoemulsion at amounts fourfold greater than native hearts (8). When tested in mice by measuring classical pharmacological parameters, the toxicity of paclitaxel was markedly reduced upon association of the drug with LDE particles. The LD50 of LDE-paclitaxel was tenfold higher than that of a commercial preparation that uses Cremophor EL as a drug delivery vehicle (9). Subsequently, patients with advanced solid cancers treated with LDE-paclitaxel at the high doses used in clinical practice did not show significant clinical or laboratory toxicity (10). LDE-paclitaxel treatment of rabbits induced to exhibit atherosclerosis via high cholesterol intake resulted in a 65% reduction in lesion size (7). In rabbits that underwent heterotopic heart transplantation, LDE-paclitaxel treatment markedly reduced heart graft damage by preventing coronary vessel destruction and macrophage invasion into the myocardium (8). The low toxicity of LDE-paclitaxel shown in cancer patients (10) and the convincing anti-atherosclerotic effects of this preparation on the aortic lesions of cholesterol-fed rabbits (7) prompted us to design the current pilot clinical study. The aim of this study was to investigate whether patients with aortic atherosclerotic disease showed good tolerability to high-dose LDE-paclitaxel treatment and whether this formulation could achieve reductions in plaque area that could be documented using a non-invasive, angiotomography-based approach. To the best of our knowledge, |
this is the first study in which patients with cardiovascular disease were treated with a systemic chemotherapeutic scheme used for cancer treatment. Patients The study group included consecutive patients referred for computerized tomography of the aorta at the Cardiovascular Computed Tomography Section of Dante Pazzanese Cardiology Institute in the city of São Paulo and at the Diagnosis Center of the Paraná Hospital in the city of Maringá. The recruited patients were on clinical follow-up due to aortic disease. The inclusion was a diagnosis of severe aortic atherosclerosis (plaque thickness 44 mm) by multidetector computed tomography (MDCT) angiography. Ten volunteers were enrolled in the study for treatment with LDE-paclitaxel. Seven of the volunteers were male and three were female. They ranged in age from 69 to 86 years and all showed aortic dilations and ulcers on angiography. All had systemic arterial hypertension, 7 had type 2 diabetes mellitus, 3 had dyslipidemia and 3 were ex-smokers. All were using statins. A control group (without LDE-paclitaxel treatment) of nine consecutive volunteer patients composed of 8 men and one woman aged 72-84 years with severe aortic atherosclerosis diagnosed by MDCT angiography was also studied. All had systemic arterial hypertension and dyslipidemia, 4 had type 2 diabetes mellitus and 5 were ex-smokers. All were using statins. Patients with renal dysfunction defined as creatinine clearance o60 mL/min, or with aortic dissection were excluded from both groups. The study protocol was approved by the medical ethics committees of the Dante Pazzanese Cardiology Institute and Paraná Hospital and all participants provided informed written consent. Preparation of LDE particles containing paclitaxel oleate LDE particles containing paclitaxel oleate were prepared by high-pressure homogenization from a pre-emulsion obtained by ultrasonic irradiation until complete drug dissolution, according to previously described methods (11). After the homogenization cycles, the formed emulsion was centrifuged, sterilized by passage through a 0.22-mm pore polycarbonate filter (Millipore, Billerica, MA) and kept at 4 o C until use. The final concentration of paclitaxel oleate LDE particles was confirmed by high-performance liquid chromatography (Shimadzu, Columbia, ML), which indicated a yield of association of paclitaxel oleate to the LDE particles of 495%. Based on measurements made by a laser scattering technique (12), the average particle diameter was 52 nm. Treatment of patients with LDE-paclitaxel particles Administration of the treatment commenced at approximately 10:00 a.m. The patients were infused via the brachial vein over an approximate 90-min time course with 175 mg/m 2 LDE-paclitaxel dissolved in a 0.9% saline solution to a total volume of 250 ml. The patients were clinically monitored for adverse reactions during the infusion and for 1 hour after the procedure. The treatment was administered every 3 weeks for a total of 6 cycles. Clinical and laboratory evaluations were performed 2 weeks after each cycle and the patients could contact the researchers anytime to report symptoms. Toxicity ratings based on NCI Common Toxicity Criteria were assessed before each treatment cycle and treatment was delayed or suspended if a patient experienced serious adverse reactions or if significant laboratory changes were detected. The evaluated laboratory parameters included red and white blood cell counts, platelet counts, electrolytes, cholesterol, triglycerides, apolipoproteins A-I and B, C-reactive protein, liver and renal function and coagulation tests. Imaging studies In the group of patients treated with LDE-paclitaxel, MDCT angiography was performed before treatment initiation and again 1-2 months after the last treatment cycle. In the control group of patients, MDCT angiography was performed twice with an interval of 6-8 months. Images were acquired using a 64-slice MDCT scanner (Aquilion 64; Toshiba Medical Systems, Otawara, Japan), which is a 64x0.5 mm collimation scanner with a gantry rotation speed of 400 ms/rotation. Scanning was performed at 100 kV and 400-500. Depending on each patient's body mass index (BMI), the table feed was 6.4 mm/gantry rotation with a beam pitch of 0.2. For thoracic aorta scans, a retrospective electrocardiogram (ECG)-gated protocol with dose modulation was adopted; abdominal aorta scans were performed using a non-gated protocol. The patients received 100 ml intravenous iodinated contrast (iopamidol 370 mg/dl) or a maximum dose of 2 ml/kg. Atherosclerotic plaques were quantified along the aorta by planimetry performed on 1-mm-thick axial images from preand post-treatment scans, such that an exact comparison of the plaque measurements before and after treatment could be performed. Plaque volume was independently quantified by two trained imaging technicians who were blinded to whether the images were taken pre-or post-treatment. Variations in the measurements of plaque volume between pre-and post-treatment were considered accurate when o3%. Statistical analysis Bland-Altman analysis was performed using Prism 6 GraphPad software (San Diego, CA) to verify that agreement existed in the plaque volume measurements taken by the two observers. The differences between the pre-and posttreatment values of plaque volume were compared using a paired Student's t test. Results were considered significant at the o0.05 level. ' RESULTS A total of 59 treatment cycles were performed and analyzed with respect to toxicity parameters. Eight patients completed the 6 scheduled treatment cycles without showing appreciable clinical or laboratory toxicity. A 78-year-old male patient experienced sudden death after the fifth treatment cycle. The patient had high cardiovascular risk and was found in his home under cardiopulmonary arrest and transported to the nearest hospital where his death was registered. Necropsy was not performed due to family refusal. This patient had not shown any clinical or laboratory toxicities at any point during the treatment period, and his death was probably unrelated to the intervention. Another patient did not return for the post-treatment angiography within the pre-established period of 1-2 months after the last (6 th ) treatment cycle. None of the patients showed any common toxicities associated with taxane use, i.e., myelosuppression, nausea, vomiting, alopecia, arthralgia, myalgia or peripheral neuropathy, or any other toxicities (Table 1). Mild leukopenia (white blood cell o4300/ml) was observed after the second and sixth cycles in one patient who remained asymptomatic. Isolated elevation of alkaline phosphatase (AP) and gammaglutamyl transpeptidase (GGT) was found in one patient throughout the treatment. This patient's baseline levels of AP and GGT were already elevated at study entry and increased by 1.5-and 4-fold, respectively. Bilirubin, AST and ALT levels; coagulation test results; and liver ultrasound remained normal in all patients throughout the treatment period. Mild alterations in renal function that did not require intervention were observed in up to 3 patients after treatment cycles 1, 2, 3, 5 and 6 (Table 1). Total, LDL and HDL cholesterol, triglycerides and apolipoprotein A1 and B concentrations did not significantly change during the treatment cycles (Table 2). Table 3 displays the individual data on aortic wall thickness volume of the eight patients in whom pre-and post-treatment angiographic exams were performed, as measured by the two blinded observers. Bland-Altman analysis showed that there was a strong correlation between the plaque volume measurements of the two observers, with a o3% variation. After treatment with LDE-paclitaxel, plaque volume was reduced by 43% in four patients (patients 1, 4, 5 and 7), remained stable in two patients (patients 3 and 8) and increased by 43% in two patients (patients 2 and 6). Table 1 -Toxicity of LDE-paclitaxel administered as an intravenous infusion over 2 h every 3 weeks to 10 patients with aortic atheroma at a dose of 175 mg/m 2 . Toxicity was classified from grade 1 (mild; asymptomatic or mild symptoms; clinical or diagnostic observations only; intervention not indicated) to grade 5 (death related to adverse event). The data represent the number of patients who presented grade 1 toxicity only because no grade 2-5 toxicities were observed. However, the differences between the pre-and post-treatment values were not significant (p=0.348). In Table 4, the plaque volume data obtained from the nontreated control patients at 6-8 month intervals are shown. In this group of nine patients, no one showed a decrease in plaque volume over the period between the first and the last observation. In 6 patients, the plaque volumes were stable and in 3 there was a 43% increase in plaque volume. Comparing the data from the non-treated group, there was a statistically significant increase in plaque volume from the first to the second exam (po0.05). ' DISCUSSION Aortic atheroma is a manifestation of systemic atherosclerosis in which the presence of aortic plaques is associated with the presence of carotid artery or coronary artery plaques. Thus, the identification of aortic atheromas is a strong predictive risk factor for major cardiovascular events, such as stroke or acute myocardial infarction (13,14). In this context, the risk factors for aortic atheroma are similar to those for coronary or cerebrovascular artery atherosclerotic disease, namely, older age and male gender, smoking, hypercholesterolemia, hypertension and diabetes (15). Patients with aortic atherosclerotic disease also seem to benefit from statin use (16). A 175 mg/m 2 body surface area paclitaxel dose was tested in the patients with aortic atheroma, which is equal to that used in cancer chemotherapy schemes. Paclitaxel is a major drug in cancer chemotherapy and is also used in pharmacological stents to inhibit restenosis. The mechanism of action for the drug is based on the ability of taxanes to increase tubulin polymerization into stable microtubules. It also promotes microtubule stabilization, thus preventing depolymerization. The overall effect is inhibition of the mitotic apparatus formation and arrest of mitosis in the G2/M phase of the cell cycle (17,18). The toxicity of commercially available paclitaxel, in which Cremophor EL is used as a drug delivery vehicle, is cumulative. The main toxicities are hematological, with neutropenia being the most prominent toxicity of the drug. Neurotoxicity, arrhythmias and alopecia are other frequent toxicities elicited by paclitaxel, together with major hypersensitivity reactions related to Cremophor EL. The fact that the toxicity of commercial paclitaxel is cumulative and that paclitaxel-associated LDE particles were administered here at a dose roughly twice that used in the study reported by Margolis et al. (19) for six cycles without producing appreciable toxicities highlights the superior tolerability of LDE-paclitaxel. To document this, a total 59 chemotherapy cycles were analyzed and the results confirmed the absence of the significant toxicities previously shown in cancer patients. This reduced toxicity can be ascribed to several factors, such as the unique biodistribution pattern created by the carrier, the envelopment of the drug in circulation, which diminishes contact with normal tissues and organs and the increased half-life of the drug. In a previous study, treatment of rabbits harboring atherosclerotic lesions induced by high cholesterol consumption with LDE-paclitaxel markedly decreased macrophage and smooth muscle cell invasion of the intima, as well as the presence of apoptotic cells (7). This accounted for the great reduction in lesion area and intima width observed in the aortas of the treated rabbits. In rabbits with heart grafts, the expression levels of pro-inflammatory factors such as tumor necrosis factor-a, interleukin-1b, interleukin-18, monocyte chemotactic protein-1, vascular cell adhesion molecule-1, matrix metalloproteinase-9, and matrix metalloproteinase-12, as well as the anti-inflammatory factor interleukin-10, were strongly up-regulated in the heart grafts of non-treated animals compared with native hearts (8). Treatment with LDE-paclitaxel resulted in a several fold reduction of the gene expression of both pro-and anti-inflammatory factors in the grafted hearts (20). It is worthwhile to note that similar results were achieved following the treatment of cholesterolfed rabbits with other preparations of chemotherapeutic agents associated with LDE, such as LDE-etoposide (21) and LDE-methotrexate (22). Thus, it can be assumed that lesion size reduction, as was observed in patients 1, 4, 5 and 7 in the current study, is a consequence not only of the antiproliferative actions of the drug but also of the reduction of inflammation in the lesion area. It is also worth noting that the inflammatory process increases the actions of proteinases, which can result in plaque rupture and hemorrhage, thrombus formation and thromboembolic accidents. The large caliber and thickness of the aortic vessel facilitates the quantification of lesion volumes using MDCT, which is non-invasive. The reliability of the quantitative analyses that were independently made by the two blinded imaging technicians was confirmed by the o3% interobserver variability. Although the study was limited by the sample containing only eight patients, it is noteworthy that four of the patients showed a regression in aortic plaque volume, whereas none of the patients in the non-treated control group showed a regression in lesion size. Additionally, although there was no significant reduction in plaque volume in the LDE-paclitaxel group |
after treatment, in the control group, probably due to the small number of patients, there was an increase in plaque volume, suggesting disease progression in these patients. With respect to the data on plasma lipid and apolipoprotein concentrations collected during the study, the lack of observed effect following the LDE-paclitaxel treatment was expected because paclitaxel does not interfere with plasma lipid metabolism. Indeed, the natural history of aortic atheroma is not always progressive; rather, a slow progression in plaque thickness is the most common clinical course (1,2). This pilot study showed that treatment with high-dose LDE-paclitaxel had low enough toxicity to permit its use in patients with cardiovascular disease. Although not statistically significant, there was an average 18% reduction in plaque volume in four out of the eight participants, which is a promising finding. This result was especially noteworthy in view of the short 18-week treatment period and when considering that plaque reduction did not occur in any of the control group patients. In contrast, statistically significant disease progression was observed in the non-treated control patients. The results of the current study are supportive for future trials aiming to confirm the efficacy of this novel treatment in a large number of participants. ' AUTHOR CONTRIBUTIONS Shiozaki AA researched the study design, wrote the manuscript and executed the work. Senra T and Paladino-Filho AT executed the work, interpreted the data and performed case selection. Morikawa AT and Deus DF. Pinto IM performed data interpretation and case selection. Maranhão RC researched the study design and helped writing the manuscript. Oral Health Policy and Improvement Strategies in Oklahoma Oklahoma an ethnically, financially and geographically diverse population has unique oral health care challenges. These challenges include poor overall oral health, inadequate oral health coverage, significant physical access to care barriers and a shortage of oral health care workers. Just as the oral health care barriers are diverse, so are the potential solutions. Potential solutions include efforts at all levels of government, innovations of health care delivery and recognition of the unique needs of Oklahoma American Indian population. Potential strategies address each of these opportunities and recognize both the short and long term needs of Oklahoma oral health. INTRODUCTION Despite being a large contributor to overall health and wellness with linkages to chronic disease such as diabetes, metabolic syndrome and cardiovascular disease [1], oral health care is an often overlooked and underprovided service in Oklahoma. Oklahoma's oral health statistics reflect a population with high rates of childhood tooth decay, adult tooth loss and absence of regular oral health appointments primarily due to cost. Root causes include financial and physical barriers and widespread oral health care professional shortages. Only 72% of Oklahoman children have received one or more dental visits compared to 80% nationally, and 66% of third graders have experienced tooth decay compared to only 52% of the national population [2]. The statistics are no better for adults: 60% of Oklahoman adults visited the dentist in 2019 compared to 68% nationally, and only 35% of pregnant women had their teeth cleaned during pregnancy compared to 46% nationally [2]. Additionally, 42% of Oklahomans aged 65 or older have lost at least six teeth to decay or gum disease [3]. Inequities in oral health access and outcomes for Oklahomans show some populations experience greater barriers than others. Cost is cited as the main reason for avoiding dental care [4,5] with low income adults and children in Oklahoma being less likely to see a dentist and lowincome older adults are also more likely to experience tooth loss (Table 1) [6,7]. Much of the oral disease Oklahomans experience is preventable [8]; however, accessing the necessary oral health cares can be expensive and difficult. POLICY OPTIONS AND IMPLICATIONS Oral Health Are Access Barriers: Financial Approximately 23% of American Indian or Alaska Native (AI/AN) people live in poverty compared to 10.5% for the US general population, and in Oklahoma, poverty affects 19.3% of AI/AN peoples [9]. Currently, 29% of Oklahoma's population receive Medicaid benefits. AI/AN people living in Oklahoma make up 11% of the total Medicaid enrollment for the state [10]. In 2020, prior to the recent Medicaid expansion, 533,000 Oklahomans were uninsured, the second highest rate in the nation, leaving many Oklahomans without realistic options for dental care [11]. Of insured patients, 14.6% of adults avoid oral health care due to non-covered costs [12]. For the uninsured, 30% of adults do not receive the dental care they need due to cost [4]. A recent report outlining the challenges and potential solutions for oral health, show 68% of respondents stated cost as the biggest barrier for improved oral health in Oklahoma communities [13]. Oral Health Care Access Barriers: Physical Physical access to oral health care is also a burden as Oklahoma, a largely rural state, encompasses 70,000 square miles and 4 million residents, 45% of whom live in rural areas. According to data published in 2017, 66% of the AI/AN population live in rural Oklahoma [14]. Further, many rural areas in Oklahoma do not have adequate oral health providers. According to the Health Resources and Serviced Administration (HRSA), 1.3 million Oklahomans live in counties considered dental health professional shortage areas, further hindering access to care [15]. The lack of oral health providers in these areas leave residents with limited options other than to travel long distances for care [16]. Those with unreliable transportation are more likely to receive inadequate or no care. Oklahoma does not have the number of oral health providers needed to support the population, with 55.3 dental care providers per 100,000 people compared to 61.2 nationally [12]. Additionally, fewer than half of dentists in Oklahoma accept Medicaid [17]. Those who face health care expense obstacles are even more challenged when that provider is physically distant or inaccessible. Oral Health Care Access Barriers: Workforce In both private and Tribal health systems employers are challenged to fill dental provider positions. The national workforce is estimated to need over 10,000 additional dental health professionals to meet the current population's needs [18]. As a result, over 80% of dental practices report recruitment challenges [19]. A recent search of the Indian Health Service (IHS) open positions database shows an ongoing trend of dentist, hygienist and assistant vacancies, with multiple open positions at some practices. This indicates a true and unacceptable deficiency in the ability to provide adequate care to the American Indian population [20]. Oklahoma's Indian health care delivery systems include federally managed clinics, urban Indian organizations and Tribally operated health care systems, referred to as I/T/Us. Rural Oklahoma has 45 I/T/U systems, which serve over 380,000 patients statewide [21]. As mentioned, staffing shortages within the I/T/Us contribute to delayed care and inadequate funding often limit services resulting in high rates of underfunded referrals to outside providers [22]. As mentioned the lack of health care providers is not unique to the I/T/U system and is felt across service delivery models. Specific root causes include lack of sustainable funding to recruit and retain providers and the high administrative burden now placed on health care providers, which ultimately reduces patient care productivity. Oral Health Care Access Solutions: Financial After the Medicaid expansion on July 1, 2021, an additional 250,000 Oklahomans, most of which were adults, enrolled in Medicaid [10]. As mentioned, ∼11% of the Oklahoma Medicaid population is American Indian [10]. Also in 2021 the Medicaid oral health coverage benefit which was previously limited to children, was expanded to adults [23,24]. The cumulative effect of Medicaid expansion and addition of adult oral health should provide new and robust services to the adult population, which may partially solve the previous oral health care coverage issues. However, increased utilization and demand for preventative services are expected and may put further strain on the system. Fortunately, with the latest Medicaid budget changes, the I/T/U systems can expect a significant increase in revenue for adult services already provided. This change will directly improve the services available for Medicaid recipients. It may indirectly improve oral health services by increasing clinical capacity through increased funding to I/T/Us as well. Oral health insurance coverage options and equitable policies should continue to be explored, as I/T/U systems may still find it difficult to provide the services needed and meet optimum staffing levels. Oral Health Care Access Solutions: Physical In the absence of an increase in oral health care providers in Oklahoma, innovative care delivery options should be explored. The COVID-19 pandemic forced health care providers to transition some services to telehealth. This trend resulted in significant improvements in access to health care, including those with transportation issues and physical limitations. To support the benefits of telehealth, the Oklahoma State Legislature has defined telehealth modalities and coverage mechanisms [25], as well as basic teledentistry [26]; and the American Dental Association has issued a position statement supporting teledentistry to provide patient care and education [27]. To further maximize teledentistry in Oklahoma, legislation and regulation to solidify insurance coverage and reimbursement requirements are necessary. Oral Health Care Access Solutions: Workforce Addressing deficiencies in access to care and provider availability, especially in the overburdened I/T/U system, requires examination of the workforce and care delivery. Expanding the services that dental assistants and hygienists can perform and authorizing a new provider type, the dental therapist (DHAT), might provide necessary care to more rural and Tribal populations. DHATs are licensed mid-levels that provide some of the most common dental procedures, such as exams and fillings. DHATs work under the supervision of a dentist [28], and are able to bring care directly to people including in schools and nursing homes, Tribal communities and rural areas [28]. DHATs have been utilized since 2005 and are currently authorized to work, in varying capacity, in 12 states [28]. While DHATs have been working globally for a century, the dental therapy profession was brought to the US by Alaskan Tribal leaders seeking to improve oral health in their communities. Under federal authority, these leaders created a program that focused on educating Alaska Natives to deliver the care their communities needed most. By focusing the scope of DHATs on a small set of commonly needed procedures, they created an education program that was accessible and affordable. The result has been a sustainable, culturally appropriate dental therapy program that creates well-paying jobs in underserved communities while increasing access to dental care and improving oral health outcomes [29]. In 2015, the US Commission on Dental Accreditation (CODA), the organization responsible for accrediting education programs for dentists and dental hygienists, issued dental therapy standards [30]. Having CODA accredit dental therapy education programs ensures the providers are educated to the same high standards as other dental providers and takes the burden off individual states to set the education requirements for dental therapists [30]. Dental therapists, supervised directly by a licensed dentist, have been shown to improve access to care for underserved communities and overall oral health [29,31]. Transferring lower-level dental procedures to DHATs would allow dentists to focus on more complex care, making the dental team more efficient and accessible [32]. Employing dental therapists means lower employment cost and allows community health centers and I/T/U clinics to stretch limited budgets and makes private practice dentists more profitable [33]. Because dental therapy education is shorter and less expensive and can be offered at Tribal or community colleges, dental therapy has created accessible workforce pipelines from and to underserved communities. Access to Care Solutions: A Community Perspective Oral health care deficiencies are felt beyond the oral health care professions. A survey of AI/AN and the general population found that a lack of rural oral health care services and covered benefits were of highest concern. Approximately 61% of respondents felt expanding the scope of practice for Oklahoma dental hygienists would help address the provider shortage, and 58% want to expand the oral health care profession types in Oklahoma. The addition of a DHAT was supported by 96% of respondents; however, 31% of the respondents voiced a need for advocacy to improve understanding of the profession (Figure 1). These outcomes show a general awareness of the Oklahoma oral health care status and need for advocacy for oral health care funding, expanded oral health care professions and scopes of work [13]. Partnerships with Oklahoma tribes are essential to identify community-based data collection methods and community-informed approaches that encompass the breadth of the disparities. |
Federal Level Solutions To solve oral health care professional shortages, discussions around expanding the scope of work for dental hygienists and/or adding new levels of oral health care professionals are necessary. The Indian Health Care Improvement Act (IHCIA) [34], the authority for the health care provision to AI/AN people, provides guidance for increasing the number of oral health care professionals. The IHCIA specifically identifies DHATs as a profession that Tribes can utilize to expand access to care. However, the IHCIA requires that Tribes outside Alaska not employ DHATs under the Community Health Aide Program unless the Tribe is in a state with a dental therapy licensing law. Tribes and national organizations such as the National Indian Health Board (NIHB) are advocating for Congress to remove this restriction, as the requirement infringes Tribal sovereignty and because DHATs practicing under the Community Health Aide Program receive federal certification, not necessarily state licensure [34]. Urban Indian clinics are also specifically excluded from accessing this benefit within the IHCIA [34], thus limiting the benefit to the AI/AN patients accessing Tribal health services. To overcome this regulatory barrier state legislation is required. States have the flexibility to determine what dental benefits are provided to adult Medicaid enrollees. There are no minimum requirements for adult dental coverage. Making adult dental services mandatory in Medicaid would expand access to dental care for millions of low-income AI/AN peoples outside the I/T/U system while improving the ability of I/T/U systems to meet their Frontiers in Oral Health | www.frontiersin.org patients' oral health needs. The optional status of Medicaid adult dental coverage means that states can take away these benefits at any time. Medicaid adult dental benefits are often subject to state budget cuts during economic downturns, especially in states with more comprehensive coverage. This means that states may offer different oral health coverage to people in different eligibility categories, such as during pregnancy or people with disabilities. This narrow definition of benefits can be confusing for enrollees and oral health providers, especially when covered services change with state budget fluctuations [4]. Because AI/AN people covered by Medicaid are not required to pay premiums or enrollment fees, standardizing Medicaid adult dental benefits could significantly increase access to affordable dental care for the hundreds of thousands of AI/AN adults already covered by Medicaid. Tribal Level Solutions Tribal health systems, by virtue of sovereignty, have the option to expand and self-regulate health care professionals within their employment. Tribes, in conjunction with HRSA and following CODA standards [28], may exercise Tribal sovereignty and train, license and employ dental therapists. Precedent for this process has been set by the Swinomish Indian Tribal Community of Washington state [35]. However, this solution does not extend to non-Tribal Federally Qualified Health Centers, urban clinics, or private practices; therefore, it does not comprehensively address the oral health care needs in Oklahoma or the AI/AN population. Excluding Tribal sovereignty, the best route for expanding the types of oral health care professionals in Oklahoma is a change in the statute to authorize DHATs to practice statewide. State Level Solutions Within the state of Oklahoma, the oral health policy landscape is largely directed by the Oklahoma Dental Act [36], which is enforced by the Oklahoma Dental Board. Currently, dentists and dental hygienists are recognized as oral health care providers in Oklahoma. Changes to this statute, such as the addition of a mid-level oral health care provider or changes in scopes of practice, must be approved by the Oklahoma State Legislature and the Oklahoma Dental Board [36]. The Oklahoma Dental Board currently consists of nine dentists, one dental hygienist and two at-large members [37]. Sixty-four percent of the current Oklahoma Dental Board represents urban communities, and none represent Tribal or Urban Indian clinics. Due to this lack of diverse representation, the current board makeup presents challenges for AI/AN health care advocacy. Oral health care in Oklahoma can be improved through the unified efforts of Tribes, urban Indian organizations, dental associations and others and state-level legislation that targets access to care barriers and benefits all oral health providers. ACTIONABLE RECOMMENDATIONS: STRATEGIES FOR IMPROVING AI/AN ORAL HEALTH CARE IN OKLAHOMA To improve Oklahoma AI/AN oral health, solutions are required that will benefit the needs of all citizens regardless of where they live or seek care. As with any complex health issue, these solutions include creative and all-inclusive thinking, broad education, and strong advocacy. Strategies designed to address each of the deficiencies discussed above are outlined below and were prioritized for feasibility, benefit applicability and potential for uptake. To fully overcome state oral health barriers, effective and all-inclusive state legislation is required. Financial Access to Care 1) Advocate for standardized, mandatory Medicaid adult dental benefits. 2) Advocate for defined coverage and reimbursement parity for teledentistry services for all payers. a. Leverage recent Medicaid oral health inclusion language to support the need for expanded benefits. 3) Advocate for the broad uptake and utilization of available oral health Current Procedural Terminology (CPT) codes within all I/T/U health systems. Physical Access to Care 1) Advocate for technical support for the uptake and utilization of teledentistry services. 2) Advocate for broad coverage and reimbursement for transportation options to oral health care providers. 3) Advocate for the continued and expanded recruitment and retention of oral health care professionals within I/T/U health systems. Workforce Deficiencies 1) Advocate for addition/expansion of scopes of practice for oral health care providers. 2) Provide education about cost savings, revenue streams and services of mid-level oral health care providers to community/professional partners, medical and dental associations, and policy authorities. a. Provide education about the rigor and requirements of CODA educational standards for dental therapy programs. b. Provide education and support for state-led efforts for addition/expansion of scopes of practice for oral health care providers. 3) Support I/T/U representation on the Oklahoma Dental Board. 4) Advocate for Tribal-level dental therapy programs, encouraging the use of Tribal sovereignty to implement training programs, licensing and regulation. 5) Advocate for the continued and expanded recruitment and retention of oral health care professionals within I/T/U health systems. CONCLUSION Oral health is critical for overall health and wellness. Improving access to dental care by adopting policies that increase access can help decrease medical costs associated with chronic conditions that continue to put Oklahoma in the top rankings for various leading causes of death [38]. Timely advances in Oklahoma's oral health policies, including the adult Medicaid oral health benefit, increases patient-driven demands for oral health care services. For facilities that receive the Office of Management and Budget (OMB) rate, significant increases in revenue are expected. If reinvested back into oral health care services, this additional revenue will increase the depth and breadth of services provided resulting in reduction to some of the access-to-care barriers discussed. However, improved financial access to care may further strain an already understaffed workforce. Further discussion of solutions to workforce shortages is necessary now, as they are predicted to become more significant in the near future. Solutions for this problem require a unified voice and open-minded discussions between Tribal and non-Tribal entities, dental associations and legislators to support Oklahoma's unique needs. Successful Treatment of Bloodstream Infection due to a KPC-Producing Klebsiella Pneumoniae Resistant to Imipenem/Relebactam in a Hematological Patient Novel carbapenem-β-lactamase inhibitor combination, imipenem/relebactam (IMI-REL), has been recently approved for treatment of infections with limited or no alternative treatment options. In this study, we described the emergence of the IMI-REL-resistance in a KPC-producing Klebsiella pneumoniae (KPC-Kp) strain collected from a hematological patient with no evidence of prior colonization. Interestingly, IMI-REL-resistance was associated with meropenem/vaborbactam (MER-VAB) cross-resistance but was not associated with cross-resistance to ceftazidime/avibactam (CAZ-AVI). Although treatment with CAZ-AVI and gentamicin completely eradicated the infection due KPC-Kp cross-resistance to IMI-REL and MER-VAB, the patient became colonized subsequently by KPC-Kp strains susceptible to IMI-REL and MER-VAB. Whole-genome sequencing performed by hybrid approach using Illumina and Oxford Nanopore platforms demonstrated that all KPC-Kp strains isolated from hematological patient belonged to the ST512 and were clonally related. Analysis of antimicrobial and porins genes demonstrated that cross-resistance to IMI-REL and MER-VAB was associated with increased blaKPC-3 copy number and truncated OmpK35 and OmpK36 with GD134-135 insertion. Phylogenetic analysis demonstrated that KPC-Kp cross-resistance to IMI-REL and MER-VAB was clonally related to a KPC-Kp resistant to IMI-REL as previously described, demonstrating the spread of this multidrug resistant clone in the hematological unit. In conclusion, the results presented in this study reported the emergence of cross-resistance to MER-VAB and IMI-REL in a KPC-Kp strain isolated from a hematological patient and highlight the potential development and diffusion of new multidrug resistance traits. Introduction Carbapenem-resistant Enterobacteriales (CRE) represent a public health problem [1]. Resistance to carbapenems is commonly due to the production of plasmid-encoded carbapenemase, which are often associated with several antimicrobial resistance determinants [2,3]. Treatment of infections due to carbapenemase-producing Enterobacteriales (CPE) is a particular concern for clinicians, mainly due to the limited available antimicrobial molecules [4]. On the basis of limited options available, a common strategy to overcome MDR resistance is to combine antimicrobial molecules with minimal in vitro activity. Although different studies suggest the superior of the combination therapy rather than monotherapy, many questions remain regarding which combination to use and the effectiveness of combined Whole Genome Sequencing and Bioinformatic Analysis During the study period, we collected 1 KPC-Kp strain from blood and 10 KPC-Kp strains from a rectal swab. Whole genomic DNA, extracted from KPC-Kp strains isolated from the patient, was sequenced using Illumina and ONT platforms, as previously described [12]. Briefly, short reads were obtained by using the Illumina iSeq 100 platform (iSeq Reagent Kit v2, Illumina, San Diego, USA) with a 2 by 150 paired-end run after the Nextera DNA Flex paired-end library. Long reads were obtained by using the MinION Mk1C platform (Oxford Nanopore Technologies, Oxford, United Kingdom), with the onedirection (1D) library prepared with a rapid sequencing kit (SQK-RAD004). All read sets were evaluated by FastQC software prior to assembly. Hybrid assemblies were performed with Unicycler version 0.4.7, and contigs were polished with Illumina reads using Pilon v.1.23. Annotation was automatically performed with RAST. Assembled genomes were screened for antimicrobial resistance and Sequence type (ST) by the CGE server [16]. ß-lactamase content was manually investigated by BLAST analysis using the Beta-Lactamase-Database [17]. Porin genes and Tn4401 isoform were manually investigated by BLAST analysis. Phage regions within the KPC-Kp genome were assessed using PHASTER web tool (PHAge Search Tool Enhanced Release) using default parameters [18]. Comparative analysis based on SNPs and indels within the genome's populations of the KPC-Kp strains was performed by aligning Illumina reads against an annotated genome, as previously described [14]. Genetic relationships among isolates included in this study were performed as previously described [11]. Briefly, a phylogenetic tree of the three KPC-Kp genomes was included in this study and, from 58 Italian strains, was generated based on the core genome single nucleotide polymorphisms (SNPs) analysis using Parsnp software and NJST2581 strain as reference. Analysis of the bla KPC copy number was evaluated using the comparative Ct method by normalizing the C T value of the target gene to the endogenous 16S rDNA control gene [12]. The fold change (2 -∆∆CT ) was analyzed by using reference strain BAT15. Genome Accession Number The sequences of KPC-Kp genomes included in this study were deposited at EMBL/EBI under the project study (EBI project PRJNA809903) with the following accession numbers: BAT15 (SAMN26209579), BAT16 (SAMN26209580), and BAT17 (SAMN26209581). Ethical Statements The study was conducted in the context of a normal clinical routine. The study was conducted in accordance with the Declaration of Helsinki. Samples were coded and analysis was performed with an anonymized database. Case Description A 65-year-old man was admitted to the Emergency Department for fever, breathlessness, and fatigue (D0). The patient's blood cell count showed hyperleukocytosis, anemia, and thrombocytopenia and a peripheral blood smear confirmed the diagnosis of acute myeloid leukemia. Two days later, the patient was admitted to the Hematology Unit. On the same day, cytoreductive therapy and antibiotic treatment with meropenem and tigecycline were introduced. Antibiotic therapy was managed without Infectious Disease (ID) consultation. On D7, chemotherapy with cytarabine, daunorubicin, and midostaurin was started. During the following days, mucositis and febrile neutropenia occurred, and KPC-Kp susceptible to CAZ-AVI, named BAT16, was isolated from blood cultures (D15), while prior surveillance rectal swabs did not show CPE colonization. |
According to an antimicrobial susceptibility pattern, the patient received a combination therapy with CAZ-AVI and gentamicin. On D21, a rectal swab yielded for once KPC-Kp, named BAT15, though this time the strain was resistant to aminoglycosides while it retained residual susceptibility to colistin and tigecycline. In accordance with active rectal CPE screening in our hospital [19], the patient was monitored weekly for the presence of rectal CPE colonization. Subsequent rectal CPE screening revealed that the patient remained colonized for 4 months by the IMI-REL-susceptible KPC-Kp strain (BAT17). Meanwhile, clinical conditions rapidly deteriorated, and on D23, patient was admitted to the Intensive Care Unit (ICU) for multi-organ failure with septic shock and severe acute kidney injury requiring continuous veno-venous hemodiafiltration. New blood cultures were collected, and after ID consultation, savage antibiotic therapy with high-dose CAZ-AVI (2.5 g every 6h), colistin, and tigecycline was administered. Since blood cultures were negative and considering progressive clinical improvement, this antibiotic regimen was continued up to 14 days. The patient recovered and on D47 was discharged from ICU with residual renal failure needing temporary intermittent dialysis and permanent aminoglycoside-induced bilateral deafness. Phenotypic Characteristics of KPC-Kp Isolates Phenotypic characteristics of the KPC-Kp strains included in the study period are shown in Table 1. Antimicrobial susceptibility testing of KPC-Kp strains showed that all strains were resistant to carbapenem (meropenem MICs ≥32 mg/L, imipenem MICs >=32 mg/L) and cefiderocol (MICs range 4-8 mg/L), while all strains were susceptible to CAZ-AVI (MICs range 4-8 mg/L). At the same time, antimicrobial susceptibility analysis revealed that two KPC-Kp strains collected from rectal swabs (BAT15 and BAT17) were susceptible to novel BL-BLICs, while KCP-Kp isolated from blood (BAT16) exhibited an increased MIC of 6 and 2-folds, respectively, for MER-VAB and IMI-REL in comparison to susceptible strains, resulting in resistance to both antimicrobial molecules. Comparison of the core genome SNPs among isolates included in this study showed that BAT15 and BAT17 differed by 38 and 44 SNPs compared to the BAT16 strain. At the same time, alignment of Illumina reads of cross-resistant BAT16 to annotated genomes of IMI-REL susceptible strains (i.e., BAT15 and BAT17) showed that most SNPs occurred within intergenic regions (Table S1 in the Supplementary Material). Furthermore, to investigate the basis of cross-resistance to IMI-REL and MER-VAB, we examined the sequence rearrangement of occurred plasmids derived from different strains included in this study. Genome analysis revealed that Tn4401 was located within 228 Kb [IncFIB(K)]/ IncFII(K)] and 118 Kb [IncFII(K)/ IncFIB(pQil)] plasmids in all KPC-Kp strains, while the third copy present in the IMI-REL and MER-VAB cross-resistant KPC-Kp was found within an IncX3 plasmid of 57 Kb in size. Interestingly, the 57 Kb plasmid carrying the bla KPC-3 gene in BAT16 showed a backbone common with the IncX3 plasmid from BAT17 and BAT15 strains (Figure 1). The region of difference between IncX3 plasmids was flanked by IS3000 and IS26 elements and contained a region harboring the bla SHV-11 gene in the 43 kb plasmids of BAT15 and BAT17 strains, while it carried a region harboring bla KPC-3 , carrying Tn4401 transposon in the 57 Kb of IMI-REL-resistant strain (Figure 1). Deep sequence analysis demonstrated the region harboring Tn4401 transposon and, bracketed by IS3000-IS26, was also present in plasmids of 228 and 118 Kb shared among all KPC-Kp strains, thus resulting in three bla KPC-3 copies in the IMI-REL-resistant strain and two copies in the IMI-REL-susceptible strains (Figure 2). In order to confirm the relationship between the increased bla KPC copy number and IMI-REL resistance, qPCR was performed among all strains included in this study. Relative quantification of the bla KPC-3 gene demonstrated that IMI-REL-resistant KPC-Kp exhibited an increased copy number rather than IMI-REL-susceptible strains. KPC-Kp was found within an IncX3 plasmid of 57 Kb in size. Interestingly, the 57 Kb plasmid carrying the blaKPC-3 gene in BAT16 showed a backbone common with the IncX3 plasmid from BAT17 and BAT15 strains (Figure 1). The region of difference between IncX3 plasmids was flanked by IS3000 and IS26 elements and contained a region harboring the blaSHV-11 gene in the 43 kb plasmids of BAT15 and BAT17 strains, while it carried a region harboring blaKPC-3, carrying Tn4401 transposon in the 57 Kb of IMI-REL-resistant strain ( Figure 1). Deep sequence analysis demonstrated the region harboring Tn4401 transposon and, bracketed by IS3000-IS26, was also present in plasmids of 228 and 118 Kb shared among all KPC-Kp strains, thus resulting in three blaKPC-3 copies in the IMI-REL-resistant strain and two copies in the IMI-REL-susceptible strains ( Figure 2). In order to confirm the relationship between the increased blaKPC copy number and IMI-REL resistance, qPCR was performed among all strains included in this study. Relative quantification of the blaKPC-3 gene demonstrated that IMI-REL-resistant KPC-Kp exhibited an increased copy number rather than IMI-REL-susceptible strains. During the study period, we isolated a KPC-Kp strain resistant to IMI-REL, MER-VAB, and CAZ-AVI in the same ward from a different patient [20]. Our previous findings demonstrated that KPC-Kp strain cross-resistant to IMI-REL, MER-VAB, and CAZ-AVI, named CAZ47, harbored similar plasmids content and chromosome by exhibiting high sequence identity. Genome comparison demonstrated that BAT16 strain differed to CAZ47 by the presence of different blaKPC alleles (blaKPC40 and blaKPC53 in CAZ47 and blaKPC3 in BAT16) and by an additional plasmid of 118 Kb carrying the third copy of blaKPC-3. In order to confirm the closely genomic relationship of KPC-Kp strains collected from two neutropenic patients, we performed a phylogenetic analysis based on the core genomes SNPs of KPC-Kp genomes isolated in Italy. Our phylogenetic tree showed that BAT15, BAT16, and BAT17 clustered closely with KPC-Kp strains collected from the second patient (i.e., BO318, CAZ59 and CAZ47), thus confirming the clonal spread of KPC-Kp strains cross-resistant to IMI-REL and MER-VAB in the same hematology unit ( Figure 3). During the study period, we isolated a KPC-Kp strain resistant to IMI-REL, MER-VAB, and CAZ-AVI in the same ward from a different patient [20]. Our previous findings demonstrated that KPC-Kp strain cross-resistant to IMI-REL, MER-VAB, and CAZ-AVI, named CAZ47, harbored similar plasmids content and chromosome by exhibiting high sequence identity. Genome comparison demonstrated that BAT16 strain differed to CAZ47 by the presence of different bla KPC alleles (bla KPC40 and bla KPC53 in CAZ47 and bla KPC3 in BAT16) and by an additional plasmid of 118 Kb carrying the third copy of bla In order to confirm the closely genomic relationship of KPC-Kp strains collected from two neutropenic patients, we performed a phylogenetic analysis based on the core genomes SNPs of KPC-Kp genomes isolated in Italy. Our phylogenetic tree showed that BAT15, BAT16, and BAT17 clustered closely with KPC-Kp strains collected from the second patient (i.e., BO318, CAZ59 and CAZ47), thus confirming the clonal spread of KPC-Kp strains cross-resistant to IMI-REL and MER-VAB in the same hematology unit (Figure 3). of 9 Microorganisms 2022, 10, x FOR PEER REVIEW 7 of 9 Figure 3. Phylogenetic tree of KPC-producing K. pneumoniae (KPC-Kp) included in this study and genomes of Italian isolates. Strains described in this study are highlighted in green, while KPC-Kp strains described in a previous study [20] are highlighted in red. Discussion In this study, we described the emergence of the cross-resistance to IMI-REL and MER-VAB in a KPC-Kp isolated from a neutropenic patient. Clinical data showed that antimicrobial combination therapy based on CAZ-AVI in combination with gentamicin completely eradicated the infection due to cross-resistant KPC-Kp strain and that, following the BSI episode, the patient became colonized by IMI-REL-susceptible strains, which resulted clonally, related to KPC-Kp cross-resistance to IMI-REL and MER-VAB. Of note, our findings demonstrate that the emerging cross-resistant KPC-Kp was genetically related to a KPC-Kp strain resistance to IMI-REL, MER-VAB, and CAZ-AVI described in our previous study [20]. Based on these findings, it seems that the patient acquired an Figure 3. Phylogenetic tree of KPC-producing K. pneumoniae (KPC-Kp) included in this study and genomes of Italian isolates. Strains described in this study are highlighted in green, while KPC-Kp strains described in a previous study [20] are highlighted in red. Discussion In this study, we described the emergence of the cross-resistance to IMI-REL and MER-VAB in a KPC-Kp isolated from a neutropenic patient. Clinical data showed that antimicrobial combination therapy based on CAZ-AVI in combination with gentamicin completely eradicated the infection due to cross-resistant KPC-Kp strain and that, following the BSI episode, the patient became colonized by IMI-REL-susceptible strains, which resulted clonally, related to KPC-Kp cross-resistance to IMI-REL and MER-VAB. Of note, our findings demonstrate that the emerging cross-resistant KPC-Kp was genetically related to a KPC-Kp strain resistance to IMI-REL, MER-VAB, and CAZ-AVI described in our previous study [20]. Based on these findings, it seems that the patient acquired an infection due to cross-resistant KPC-Kp strain during hospitalization and that the combination therapy based on meropenem, in association with tigecycline, did not select strains cross-resistant to IMI-REL or MER-VAB. In accordance with previous findings [9,20,21], our results confirmed that the IMI-REL-resistance was associated with an increased blaKPC copy number in a OmpK35 porin deficiency strain harboring OmpK36 porin mutations. In particular, previous studies demonstrated that KPC-Kp strains are resistant to IMI-REL exhibited porins deficiency (i.e., truncated OmpK35 and OmpK36 truncated or with GD134-135 insertion) and increased blaKPC copy number (i.e., 2-4 folds) rather than IMI-REL-susceptible strains [20,21]. At the same time, we demonstrated that the mechanism at the basis of IMI-REL-resistance was associated with the cross-resistance to MER-VAB. Genome analysis of KPC-Kp strain cross-resistance to IMI-REL and MER-VAB showed that a region containing blaKPC-3 carrying Tn4401 transposon was shared among IncX3, IncFIB(pQIL)/IncFII(K), and In-cFIB(K)/IncFII(K) plasmids and based on the absence of flank-ing direct repeats. We hypothesized that this region was transposed via homologous recombination mediated by IS3000 and IS26 elements. At the same time, our results demonstrated that the KPCcarrying Tn4401 transposon was present in triple copies within the cross-resistant KPC-Kp and in double copies among IMI-REL-susceptible KPC-Kp strains included in this study. Similar findings have been reported in previous studies, demonstrating that these elements may be at the basis of the high transferability efficiency of antimicrobial resistance genes in MDR microorganisms [20,21]. In conclusion, here we reported the successful treatment of bacteremia due to KPC-Kp strain resistant both to IMI-REL and MER-VAB. Cross-resistance to IMI-REL and MER-VAB represents an emerging threat that could pose further limitations for clinicians to the treatment of infections due to difficult to treat (DTR) pathogens, especially in critical patients. In light of emergence diffusion of antimicrobial resistance and the recent COVID-19 pandemic, there is an urgent need to optimize antimicrobial therapy and integrate antimicrobial stewardship activities and infection control measures to limit the spread of novel resistance [22]. In this context, further studies should be necessary to identify the suboptimal drug exposure related to the cross-resistance to IMI-REL and MER-VAB, characterize specific-mutations related to cross-resistance emerging in vivo under different antimicrobial therapy, and define the optimal antimicrobial treatment for infections due to cross-resistant KPC-Kp. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Ethics Committee of IRCCS Azienda Ospedaliera-Universitaria di Bologna (protocol code 409/2019/Oss/AOUBo 17/02/2020). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: Not applicable. Conflicts of Interest: The authors declare no conflict of interest. Recently published papers: A review of novel strategies in the prevention of hospital-acquired infections, the ability of intensivists to perform echocardiography, and the benefit of polymyxin B haemoperfusion in abdominal sepsis Chlorhexidine bed baths seem to reduce the incidence of methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococcus detected by surveillance cultures. There is also some evidence on the benefit of chlorhexidine mouthwashes in the prevention of ventilator-associated pneumonias. Acid-suppressing drugs increase the incidence of hospital-acquired pneumonias in non-intensive care unit patients, although this association has not been shown in the intensive care setting. Intensivists can be trained to perform basic echocardiography in a short period of time, but their errors could lead to incorrect changes in management. Polymyxin B haemoperfusion was shown in interim analysis to improve patients with abdominal sepsis to such an extent that the EUPHAS randomised controlled trial was halted on ethical grounds, although other authors have criticised this decision. Hospital-acquired infections and ventilatoracquired pneumonias: avoidance strategies Prevention of hospital-acquired infections and ventilatorassociated pneumonias is a key concern in the reduction of mortality in |
both the critical care and general ward patient populations. A number of recent articles have evaluated differing approaches relating to this topic [1][2][3][4]. Climo and colleagues studied the effect of daily 4% chlorhexidine baths on colonisation and subsequent bloodstream infection with multidrug-resistant organisms, in a multicentred before-after study [1]. The acquisition of methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococcus, as detected by active surveillance cultures, was shown to be significantly reduced (methicillin-resistant S. aureus incidence density 5.04 vs. 3.44 cases per 1,000 eligible patient-days, P = 0.046; vancomycin-resistant enterococcus incidence density 4.35 vs. 2.19 cases per 1,000 eligible patient-days, P = 0.008). Furthermore, the authors showed that the incidence of vancomycin-resistant enterococcus bacteremias decreased significantly following the intervention (2.13 vs. 0.59 cases per 1,000 patient-days). Owing to the low rate of methicillin-resistant S. aureus bacteremia prior to the change in practice, Climo and colleagues were unable to show a significant reduction in methicillin-resistant S. aureus bloodstream infections, although they suggest that with sufficiently powered studies this may also be shown to be the case. Panchabhai and colleagues compared oropharyngeal cleansing using 0.2% chlorhexidine with the study intensive care unit (ICU) policy of twice-daily 0.01% potassium permanganate in a randomised controlled trial [2]. Within their study population of 471 patients in total there was no significant difference between groups. The authors do observe, however, that the incidence of nosocomial pneumonias on their ICU in the 3 months prior to the study and in the 3 months after the study was significantly higher than during the 6-month study period (21.7% vs. 7.4%; P <0.001). The authors conclude that during the study period the meticulous oral hygiene in both groups and the weak antiseptic properties of potassium permanganate may have contributed to this unanticipated but significant finding. Two other studies have also found that oral decontamination with either chlorhexidine or a chlorhexidine/ collistin mixture did significantly reduce the incidence of ventilator-associated pneumonias [3,4]. The National Institute for Health and Clinical Excellence and the National Patient Safety Agency have recommended this practice as a part of the ventilator care bundle since August 2008 [5]. In the management of severe sepsis or septic shock, the Surviving Sepsis Campaign [6] strongly recommends the use of a histamine-2 blocker or a proton-pump inhibitor in order to Recently published papers: A review of novel strategies in the prevention of hospital-acquired infections, the ability of intensivists to perform echocardiography, and the benefit of polymyxin B haemoperfusion in abdominal sepsis provide stress ulcer prophylaxis. Herzig and colleagues undertook a large single-centred cohort study of the effect of acid-suppressing medications on the incidence of hospitalacquired pneumonias [7]. They included all patients admitted to their hospital for longer than 3 days over a period of 3 years (n = 42,093), excluding patients admitted to the ICU. Of these patients, acid-suppressing medication was ordered in 32,933 (52%) admissions. There was a large baseline difference in the characteristics of the group receiving acidsuppressing medications versus those patients not receiving the medications. Once these groups had been propensity matched, the authors were able to compare 16,396 paired patients and showed that the use of an acid-suppressing drug showed an increased association with developing a hospital-acquired pneumonia with an odds ratio of 1.3 (95% confidence interval = 1.1 to 1.4). However Beaulieu and colleagues in their 2008 study did not show an increased risk of nosocomial pneumonia in medical intensive care patients receiving proton-pump inhibitors [8]. Given these findings, the prescription of acid-suppressing medications should be given careful consideration -particularly when used for the purpose of stress ulcer prophylaxis, where these drugs should be ceased as soon as enteral feeding has been reestablished. Schweickert and colleagues present the argument that early physical and occupational therapy can reduce the duration of time on a ventilator, and can thereby reduce the mortality associated with this intervention [9]. They performed a dualcentre randomised controlled trial comparing standard ICU practice of daily sedation holds and therapy as per the primary care team versus daily sedation holds with specific early exercise and mobilisation (physical and occupational therapy). In their group of 104 patients, early physical and occupational therapy significantly increased the number of patients who had returned to independent functional status by the point of discharge (59% vs. 35%, P = 0.02). Early therapy also was shown to reduce the average number of days with ICU delirium (2.0 vs. 4.0, P = 0.03). Early therapy did not, however, reduce the duration of ICU stay (5.9 vs. 7.9, P = 0.08) or of hospital stay (13.5 vs. 12.9, P = 0.93). Knight and colleagues looked at a less commonly described intervention, enteral synbiotic (a mixture of prebiotic and probiotic) therapy, on the incidence of ventilator-associated pneumonias [10]. In their blinded, randomised, placebocontrolled trial, the authors compared a total of 259 patients with a primary outcome variable as the development of ventilator-acquired pneumonia. The use of synbiotics had no significant effect on the development of ventilator-acquired pneumonia, nor the secondary endpoints of oropharyngeal flora, ventilator days, ventilator-associated pneumonia rate per 1,000 ventilated days, ICU length of stay and hospital length of stay. Intensivist assessment of left-ventricular function using transthoracic ultrasound To appropriately manage the critically ill patient, an accurate assessment of left ventricular function is often called for. Owing to equipment or personnel limitations these investigations can be significantly delayed, and a rapid assessment of left ventricular function is often not possible. Melamed and colleagues asked whether, after brief training, intensivists were able to make an accurate assessment of left ventricular function using transthoracic ultrasound [11]. They trained a group of intensivists with 2 hours of didactic teaching and 4 hours of practical teaching, and provided a range of pre-recorded examinations for the intensivists to study in their own time. The study group comprised 44 patients in whom the intensivists were asked to first identify normal or abnormal left ventricular function and then to categorise those with abnormal function into mild to moderate or severe. Ultimately, the intensivists correctly categorised 36 out of 44 cases (82%). It would be easy to conclude that this is a resounding endorsement for transthoracic echocardiography performed by the intensivist, but the cases they incorrectly identified were overestimations of actual function and this could potentially alter management incorrectly. If the study had also evaluated the degree of certainty that the intensivists held over their findings, it would help to illustrate whether inappropriate changes of management might have resulted from their transthoracic echocardiography findings. Effect of polymyxin B haemoperfusion on patients with abdominal sepsis The EUPHAS randomised controlled trial examined the effect of polymyxin B fibre haemoperfusion on the outcome of patients with abdominal septic shock [12]. Polymyxin B is an antibiotic with a high affinity for endotoxin. Haemoperfusion was performed using haemoperfusion polystyrene fibres with bound polymyxin B. The study was stopped early because there was a signal of decreased mortality in the treatment group, based on a Cox proportional hazards regression survival model. The crude mortality analysis, however, did not show a statistically significant difference between groups (11/34 polymyxin group vs. 16/30 conventional group, odds ratio = 2.39 (95% confidence interval = 0.87 to 6.60), P = 0.09). The difference between the crude mortality analysis and the Cox regression model can be explained by a larger occurrence of late deaths in the treated group. This is a concerning observation in an unblinded trial, as it could be caused by a brief prolongation of life support in the treatment group and a resultant increase in early survival data. Despite their early withdrawal from the study the authors call for larger multicentred studies to confirm their findings. The decision for early cessation, the study's insufficient power and the lack of a patient-centred endpoint has been criticised in other editorials [13]. Tetrastigma hemsleyanum Vine Flavone Ameliorates Glutamic Acid-Induced Neurotoxicity via MAPK Pathways Glutamic acid (Glu) is a worldwide flavor enhancer with various positive effects. However, Glu-induced neurotoxicity has been reported less. Tetrastigma hemsleyanum (TH), a rare herbal plant in China, possesses high medicinal value. More studies paid attention to tuber of TH whereas vine part (THV) attracts fewer focus. In this study, we extracted and purified flavones from THV (THVF), and UPLC-TOF/MS showed THVF was consisted of 3-caffeoylquinic acid, 5-caffeoylquinic acid, quercetin-3-O-rutinoside, and kaempferol-3-O-rutinoside. In vitro, Glu caused severe cytotoxicity, genotoxicity, mitochondrial dysfunction, and oxidative damage to rat phaeochromocytoma (PC12) cells. Conversely, THVF attenuated Glu-induced toxicity via MAPK pathways. In vivo, the neurotoxicity triggered by Glu restrained the athletic ability in Caenorhabditis elegans (C. elegans). The treatment of THVF reversed the situation induced by Glu. In a word, Glu could cause neurotoxicity and THVF owns potential neuroprotective effects both in vitro and in vivo via MAPK pathways. Introduction Glutamic acid (Glu), as one of the basic amino acids, is widely applied in industry. Glu is involved in many crucial chemical reactions in the body and plays an important role in protein metabolism in organisms [1]. Meanwhile, it is ubiquitous in the human diet; as a flavor enhancer and food additive in food field, Glu is wildly used to improve the taste of beverages and foods and preserves the freshness of animal food [2,3]. For example, sodium glutamate, commonly known as monosodium glutamate, is a typical flavor agent that can be used alone or with other amino acids [4]. In addition, studies have proven that Glu is an excellent hairgenerating agent. Ottersen et al. reported that Glu effectively promoted the proliferation of hair papilla cells; besides, it can also expand blood vessels and accelerated blood circulation, resulting in hair regeneration [5,6]. It has also been estimated that Glu poses the ability to reduce wrinkles [7]. As an auxiliary drug for liver diseases, Glu is taken and combined with blood ammonia to form glutamine, which can relieve the toxic effect of ammonia in the metabolic process, thus preventing and treating hepatic coma [8,9]. Besides, brain tissue cannot oxidize amino acids except glutamate, and therefore, glutamine can be used as an energy substance to improve the function of the brain [10,11]. As a supplement to the nerve center and cerebral cortex, Glu exhibits a certain effect on the treatment of concussion, nerve damage, epilepsy, and mental retardation [12,13]. Data show that Glu peaks the largest productive amino acid variety worldwide. Although Glu shows an irreplaceable role in various fields, it may also turn from a protective agent to a neurotoxin in many cases. A growing number of studies have found that the Glu is closely associated with the etiology and pathology of many neurological and psychiatric disorders, such as cerebral ischemia, epilepsy, Alzheimer's disease, Huntington's disease, schizophrenia, and Pico disease [14]. Oxidative stress caused by Glu is an important cause of neurodegenerative diseases [15,16], high concentration of Glu inhibits cysteine/glutamate antiporter or system x-CT in neuronal cells, causes calcium overload that interferes with mitochondrial respiratory chain function, thereby inhibiting cystine uptake and causing intracellular glutathione (GSH) deprivation and ROS accumulation, and ultimately leading to cell necrosis or apoptosis [17,18]. Moreover, studies have shown that the caspase-dependent apoptotic pathway is related with Glu-neurotoxicity [19]. Natural plant extracts are attracting researchers' attention because of their safety, nontoxicity, and various biological activities [20]. Many natural plants have been already proved to exhibit various bioactive capacities such as antioxidative activity, anti-inflammatory activity, hypolipidemic effects, and even antitumor capacity [20]. Tetrastigma hemsleyanum Diels et. Gilg (TH), initially used for folk treatment of cancer (Li et al., 2019), is now not only a traditional Chinese medicine but also a type of functional food. Previous studies have shown that TH has antioxidant, anti-inflammatory, anticancer, and immunomodulatory properties that can effectively treat high fever, infantile febrile seizures, pneumonia, asthma, hepatitis, rheumatism, menstrual disorders, sore throats, and sores [21,22]. It has reported that TH contains many phytochemicals, such as flavonoids, phenolic acids, polysaccharides, and phytosterols, resulting in its various biologically activities such as anti-inflammatory, antioxidant, antiproliferative, antitumor, and antiviral effects [22][23][24]. However, there are fewer studies that paid attention to vine of TH, which is usually regarded useless and often discarded as a by-product, resulting in a waste of source. In this study, we have extracted and purified the THVF, then identified and characterized the main compounds of THVF by UPLC-TOF/MS. In addition, we adopted the PC12 cell line and evaluated the protective effects of the THVF against damage induced by |