text
stringlengths 4
179k
| predicted_class
stringclasses 5
values | confidence
float64 0
100
|
|---|---|---|
Histones were acid-extracted according to a published protocol . One microgram of each sample was used for Western blot analysis with 15% SDS-PAGE gels and PVDF membranes (Merck Millipore) as described above using antibodies listed in Additional file 1. Histone H3 was used as a histone loading control. | other | 38.44 |
ChIP-IT™ Express Kit (#53008, Active Motif, La Hulpe, Belgium) was used according to the manufacturer’s instructions. Rabbit Gamma Globulins (#31887, Invitrogen) served as a background control. Quantitative real-time PCR was used to determine enrichment of indicated gene regions at their transcriptional start site (TSS) as well as 2 kb upstream and downstream of each TSS. For a list of antibodies and primers, see Additional file 1. | other | 37.53 |
Cells were transfected by X-tremeGENE 9 DNA transfection reagent (Roche, Penzberg, Germany) with p57 Double Nickase Plasmid (sc-400444-NIC-2, Santa Cruz Biotechnology, Heidelberg, Germany) encoding a GFP marker, puromycin resistance, and two different sgRNAs targeting CDKN1C exon 1 and D10A mutant Cas9 (Nickase). Double Nickase control Plasmid (sc-437281) with non-targeting sgRNAs was used as a control. GFP expression allowed monitoring of transfection efficiency. Transfected cells were selected with 0.5 μg/ml puromycin for 5 days before single-cell seeding into 96-well plates. Genomic DNA was extracted from single-cell clones using QIAamp DNA Mini Kit (Qiagen). An amplicon spanning the sgRNA binding sites was amplified using HotStarTaq polymerase (Qiagen); PCR products were cloned into PCR4-TOPO TA Vector (450030, Invitrogen) and Sanger-sequenced. Mutant sequences were compared to the NCBI CDKN1C reference sequence (NG_008022.1). Successful knockout of p57KIP2 was verified by Western blot analysis. | other | 34.94 |
Since BRD4 is considered the most important target of the BET inhibitor JQ1 in various cancers, we evaluated BRD4 protein expression in a series of UCCs compared to the benign urothelial control cell lines HBLAK, TERT-NHUC, and NHUC (Fig. 1a). BRD4 was expressed in all tested UCCs at variable levels, in some UCCs more strongly than in the normal controls.Fig. 1Effects of BRD4 knockdown on urothelial carcinoma cell lines. a Western blot analysis of BRD4 expression in 10 UCCs compared to the benign urothelial control cell lines HBLAK, NHUC, and NHUC-TERT. α-tubulin served as a loading control. b Western blot analysis of BRD4 and c-MYC expression after siRNA-mediated knockdown of BRD4 in VM-Cub1 and UM-UC-3 cells. Cells transfected with BRD4 expression plasmid served as a positive control, with 3 μg instead of 20 μg protein loaded. c Relative viability of VM-Cub1 and UM-UC-3 cells after BRD4 knockdown for 48, 72, and 120 h compared to treatment with control siRNA. Relative viability is displayed on the ordinate in percent of the control cells treated with non-targeting siRNA. Differences between control and targeting siRNA were analyzed using Student’s t test (***p ≤ 0.001). d Clonogenicity assays of VM-Cub1 and UM-UC-3 cells after BRD4 knockdown for 120 h and results of quantification (**p ≤ 0.01). e Flow cytometric cell cycle analysis of VM-Cub-1 and UM-UC-3 cells after BRD4 knockdown for 48, 72, or 120 h. Percentages of cells in the respective cell cycle phase are given | other | 30.5 |
Effects of BRD4 knockdown on urothelial carcinoma cell lines. a Western blot analysis of BRD4 expression in 10 UCCs compared to the benign urothelial control cell lines HBLAK, NHUC, and NHUC-TERT. α-tubulin served as a loading control. b Western blot analysis of BRD4 and c-MYC expression after siRNA-mediated knockdown of BRD4 in VM-Cub1 and UM-UC-3 cells. Cells transfected with BRD4 expression plasmid served as a positive control, with 3 μg instead of 20 μg protein loaded. c Relative viability of VM-Cub1 and UM-UC-3 cells after BRD4 knockdown for 48, 72, and 120 h compared to treatment with control siRNA. Relative viability is displayed on the ordinate in percent of the control cells treated with non-targeting siRNA. Differences between control and targeting siRNA were analyzed using Student’s t test (***p ≤ 0.001). d Clonogenicity assays of VM-Cub1 and UM-UC-3 cells after BRD4 knockdown for 120 h and results of quantification (**p ≤ 0.01). e Flow cytometric cell cycle analysis of VM-Cub-1 and UM-UC-3 cells after BRD4 knockdown for 48, 72, or 120 h. Percentages of cells in the respective cell cycle phase are given | other | 30.34 |
Next, we investigated by siRNA-mediated knockdown how dependent different UCCs are on BRD4 function and whether effects of siRNA knockdown differed from that by pharmacological inhibition in the same cell line. VM-Cub1 and UM-UC-3 cells were chosen for their very different phenotypes and have had been extensively characterized for their response to HDACi previously [3, 4]. As shown below, they were also differentially sensitive to JQ1 (Table 1 upper part). Efficient knockdown of BRD4 protein was confirmed at 48, 72, and 120 h post-transfection. Expression of c-MYC, a common target of BRD4, was strongly decreased in VM-Cub1 cells 120 h post-transfection, but not in UM-UC-3 (Fig. 1b). | review | 27.73 |
Cell viability decreased following BRD4 knockdown in a time-dependent manner in both cell lines (Fig. 1c). Concordantly, clonogenic growth was significantly suppressed (Fig. 1d; p ≤ 0.01). Changes in cell cycle distribution intensified over time; VM-Cub1 cells became arrested in G0/G1, whereas UM-UC-3 cells accumulated in G2/M (Fig. 1e). Only minor increases in subG1 fractions were observed. | review | 29.38 |
Next, we investigated the dose response of eight different UCCs, including VM-Cub1 and UM-UC-3, to JQ1. JQ1 diminished viability in all cell lines, but they clearly fell into two groups, with low and high sensitivity. Calculated IC50 values were below 0.5 μM in the first group, but well above 1 μM in the second group (Table 1). Notably, at the highest concentration of JQ1 (25 μM) tested, a sizeable fraction of cells survived the treatment. This fraction correlated only partially with the respective IC50, indicating that JQ1 alone acted rather in a cytostatic fashion. No correlation was seen between sensitivity to JQ1 and BRD4 protein expression or any other previously investigated characteristics of the cell lines. Benign control cells generally responded to low concentrations of JQ1 (IC50 0.26–0.4 μM). | other | 26.47 |
Cell viability results from dose-response curves for JQ1, and Romidepsin were used to generate combination index plots (Fa/CI-Plots; fraction affected/combination index; Fig. 2a) by the Chou-Talalay method . Strong synergisms (CI < 1) were detected for all four UCCs at effect rates (Fa; percentage of dead cells) from 0.2 (T24) to 0.4 (UM-UC-3). Synergies intensified with increasing effect rates. Intriguingly, the strongest synergies were seen in the less JQ1-sensitive 639-V and UM-UC-3 cells. Similarly, primary cancer cell cultures established from patient tissues responded strongly towards combined treatment resulting in strong synergies (Additional file 2). In contrast, only a limited synergy of the combined treatment was seen with benign HBLAK cells (Fig. 2a).Fig. 2Effects of combined treatment with JQ1 and Romidepsin on proliferation and clonogenic growth of UCCs. a Combination index (CI/Fa) plots for the combination of JQ1 and Romidepsin. Cell viability was measured at five constant dose ratio experimental data points by ATP assay after 72-h treatment. CI plots were then generated using CompuSyn software. CI < 1 indicates synergism. Benign HBLAK cells were compared to four UCCs. b Clonogenicity assay following treatment for 48 h with JQ1, Romidepsin, or both compounds compared to DMSO as indicated in Table 1 (lower part). c Quantification of colony counts from clonogenicity assays. Differences between control and treated cells were analyzed using Student’s t test (**p ≤ 0.01, *p ≤ 0.05). d Flow cytometric cell cycle analyses following the indicated treatment for 48 h in four different UCCs and in HBLAK cells. See Additional file 3 for cell numbers in the respective cell cycle phases | other | 27.31 |
Effects of combined treatment with JQ1 and Romidepsin on proliferation and clonogenic growth of UCCs. a Combination index (CI/Fa) plots for the combination of JQ1 and Romidepsin. Cell viability was measured at five constant dose ratio experimental data points by ATP assay after 72-h treatment. CI plots were then generated using CompuSyn software. CI < 1 indicates synergism. Benign HBLAK cells were compared to four UCCs. b Clonogenicity assay following treatment for 48 h with JQ1, Romidepsin, or both compounds compared to DMSO as indicated in Table 1 (lower part). c Quantification of colony counts from clonogenicity assays. Differences between control and treated cells were analyzed using Student’s t test (**p ≤ 0.01, *p ≤ 0.05). d Flow cytometric cell cycle analyses following the indicated treatment for 48 h in four different UCCs and in HBLAK cells. See Additional file 3 for cell numbers in the respective cell cycle phases | other | 27.48 |
Next, we compared the effects of single and combined treatment on cell proliferation and apoptosis. Dosages for the combined treatment were chosen from synergistic combinations and mostly below individual agent IC50s (Table 1). Single treatment with JQ1 alone diminished clonogenic growth of the more sensitive VM-Cub1 cells (Fig. 2b), whereas the less sensitive UM-UC-3 cells readily formed colonies. Clone formation by benign HBLAK cells was rather unaffected. Next, we evaluated the combination of JQ1 and Romidepsin in two more (VM-Cub1 and T24) and two less (UM-UC-3 and 639-V) JQ1-sensitive UCCs and in HBLAK as a benign urothelial control. The combination treatment synergistically suppressed long-term proliferation of all four UCCs (at least 4-fold compared to DMSO; Fig. 2b, c), much more strongly than each inhibitor alone at the same doses (at least 3.5-fold compared to single treatments; Fig. 2b, c). Remarkably, no significant inhibition of clonogenic growth was observed in HBLAK cells. | review | 27.88 |
To characterize the cellular effects of the drug combination in more detail, we performed cell cycle analysis by flow cytometry. Treatment with JQ1 alone led to accumulation of cells in the G0/G1-phase and a decreased S-phase fraction in most UCCs (Fig. 2d, Additional file 3), except in UM-UC-3, where an increased G2/M fraction was observed. The subG1 fraction increased slightly in several, albeit not in all UCCs. HBLAK cells were likewise arrested in G0/G1, with decreased S-phase fraction, but no increased subG1 fraction. Cell cycle distribution changes after treatment with JQ1 were very similar to those after BRD4 knockdown (see Fig. 1e). Likewise, UCCs responded towards single treatment with Romidepsin by accumulating in G0/G1 or G2/M (Fig. 2d) in accord with our previous results . | other | 27.52 |
In contrast, combination treatment resulted consistently in a strong increase of the subG1 fraction across all four UCCs (Fig. 3a), whereas the number of S-phase cells was reduced dramatically, indicating cell cycle inhibition and cell death (Fig. 2d, Additional file 3). Although the cell lines reacted rather uniformly in so far, they differed in the cell cycle phase, in which they became arrested. Whereas VM-Cub1 and UM-UC-3 cells showed a strong G2/M increase with decreased G0/G1 cells, 639-V and T24 cells displayed an increased G0/G1 fraction (Fig. 2d, Additional file 3). HBLAK cells arrested in G0/G1 with a remarkable decrease of the S-phase and G2/M fractions. Notably, the combination induced a subG1 fraction much more prominently than either agent on its own in UCC, but not in HBLAK cells (Fig. 3a).Fig. 3Induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. a Increase of subG1 fraction by combination treatment as determined by flow cytometry. b Flow cytometric analysis of UCCs with indicated treatment (Table 1 lower part) after combined staining with PI and Annexin V. Percentages of viable (lower left), early (lower right), or late (upper right) apoptotic and necrotic (upper left) VM-Cub1 cells subsequent to indicated treatments. c Percentage of early apoptotic cells as measured by Annexin V staining for all UCCs and HBLAK control cells are displayed in bar graphs for the respective treatment. d PARP and Caspase-3 cleavage 48 h after treatment assessed by Western blot analysis. e Four UCCs received the combination treatment with or without the Pan-Caspase inhibitor Q-VD-Oph at 30 μmol/l. All cells received the same concentration of DMSO. Cell viability displayed on the ordinate was measured by ATP assay after 48 h. Differences between J + R treatment and J + R treated plus Caspase inhibitor were analyzed using Student’s t test (***p ≤ 0.001, **p ≤ 0.01) | other | 30.39 |
Since the cell cycle regulator protein p57KIP2 (encoded by the CDKN1C gene, hereafter p57) had been reported to mediate synergistic induction of apoptosis by combined HDAC and BET inhibition in pancreatic ductal adenocarcinoma , we evaluated its expression in four UCCs after treatment. p57 mRNA was induced between 17-fold (T24) and up to 1000-fold (VM-Cub1) by combination treatment. It was also induced regularly, but to a smaller extent, by Romidepsin, but not consistently by JQ1 (Additional file 5). Western blot analysis revealed a double band around the predicted molecular weight, of which especially the lower band was strongly induced by the combination treatment (Fig. 4e). The upper band likely reflects a phosphorylated isoform generated by active AKT protein kinase which is destined to degradation via a SKP2-containing E3 ubiquitin ligase complex. Accordingly, a (T310) phospho-specific p57 antibody demonstrated loss of phosphorylated p57 following combination treatment, indicating accumulation of the more stable unphosphorylated p57 protein (Fig. 4e). | other | 26.53 |
Accordingly, active pAKT was diminished in three of four cell lines by JQ1 single treatment, but particularly strongly by combined treatment (Fig. 4f). SKP2 expression, too, was strongly downregulated in all four cell lines by the combination treatment. VM-Cub1 cells displayed the strongest induction of p57 protein and mRNA and the most pronounced decrease in its negative regulators c-MYC, pAKT, and SKP2. Conversely, T24 cells displayed the weakest increase of p57 protein and pAKT levels remaining unchanged. These findings suggest that, in addition to increased transcription of CDKN1C, stabilization of p57 contributes to its accumulation. Of note, reduced activity of AKT by combined treatment is likely to diminish pro-survival signaling in UCCs by other pathways as well. | study | 27.88 |
Induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. a Increase of subG1 fraction by combination treatment as determined by flow cytometry. b Flow cytometric analysis of UCCs with indicated treatment (Table 1 lower part) after combined staining with PI and Annexin V. Percentages of viable (lower left), early (lower right), or late (upper right) apoptotic and necrotic (upper left) VM-Cub1 cells subsequent to indicated treatments. c Percentage of early apoptotic cells as measured by Annexin V staining for all UCCs and HBLAK control cells are displayed in bar graphs for the respective treatment. d PARP and Caspase-3 cleavage 48 h after treatment assessed by Western blot analysis. e Four UCCs received the combination treatment with or without the Pan-Caspase inhibitor Q-VD-Oph at 30 μmol/l. All cells received the same concentration of DMSO. Cell viability displayed on the ordinate was measured by ATP assay after 48 h. Differences between J + R treatment and J + R treated plus Caspase inhibitor were analyzed using Student’s t test (***p ≤ 0.001, **p ≤ 0.01) | other | 30.3 |
The increased subG1 fractions and according morphological changes in cell morphology indicated that combination treatment elicited pronounced apoptosis. Concordantly, the number of early apoptotic cells determined by Annexin V staining (Fig. 3b, c and Additional file 4) as well as cleaved Caspase 3 and cleaved PARP were significantly enhanced (Fig. 3d). These markers of apoptosis were not or only weakly induced by either a single drug treatment. Of note, induction of apoptosis was less pronounced in UM-UC-3 compared to the other UCCs. Moreover, cytotoxicity of the combination treatment was partly prevented by the Pan-Caspase inhibitor Q-VD-Oph in all four UCCs (Fig. 3e) indicating that the combination treatment efficiently elicited caspase-dependent apoptosis. | other | 28.16 |
Concurringly, mRNA expression of the antiapoptotic regulators BCL-XL (Fig. 4a) and BCL-2 (Fig. 4b) was diminished by combination treatment in UCCs, again with the exception of UM-UC-3. Expression of BCL-XS was either increased or remained unchanged (Fig. 4c). Similarly, the antiapoptotic BCL-XL and Survivin proteins were generally downregulated by the combination treatment (Fig. 4d). These anti-apoptotic proteins were not consistently downregulated by either single agent and in some cases actually induced. Thus, in contrast to single drug treatments, combined treatment led to stronger and more consistent downregulation of antiapoptotic regulators. This downregulation was less pronounced in UM-UC-3, presumably accounting for the weaker apoptotic response in this cell line.Fig. 4Changes in expression of apoptotic regulators and cellular signaling after combined treatment with Romidepsin and JQ1. Relative mRNA expression levels of antiapoptotic BCL-XL (a) and BCL-2 (b) and proapoptotic BCL-XS (c). mRNA levels were measured by qRT-PCR and normalized to the expression of TBP. Fold change compared to DMSO control 48 h after treatment is displayed on the ordinate. Differences between combination treatment samples and untreated controls were analyzed using Student’s t test (**p ≤ 0.01, *p ≤ 0.05). d Protein expression of antiapoptotic BCL-XL and Survivin after combination treatment for 48 h assessed by Western blot. e Protein expression of c-MYC, EZH2, p57KIP2, and phospho-p57KIP2 (Thr310) assessed by Western blot analysis. f Phosphorylated AKT (S473) and expression of AKT and SKP2 was detected by Western blot analysis | study | 27.73 |
Changes in expression of apoptotic regulators and cellular signaling after combined treatment with Romidepsin and JQ1. Relative mRNA expression levels of antiapoptotic BCL-XL (a) and BCL-2 (b) and proapoptotic BCL-XS (c). mRNA levels were measured by qRT-PCR and normalized to the expression of TBP. Fold change compared to DMSO control 48 h after treatment is displayed on the ordinate. Differences between combination treatment samples and untreated controls were analyzed using Student’s t test (**p ≤ 0.01, *p ≤ 0.05). d Protein expression of antiapoptotic BCL-XL and Survivin after combination treatment for 48 h assessed by Western blot. e Protein expression of c-MYC, EZH2, p57KIP2, and phospho-p57KIP2 (Thr310) assessed by Western blot analysis. f Phosphorylated AKT (S473) and expression of AKT and SKP2 was detected by Western blot analysis | study | 30.69 |
STAT3 activation may also be involved in regulation of cell cycle progression and anti-apoptotic response and had been reported to be inactivated by combined HDAC and BET inhibition in pancreatic ductal adenocarcinoma . However, we neither observed significant changes in STAT3 phosphorylation nor expression after combined treatment with JQ1 and Romidepsin (Additional file 6). | other | 32.1 |
To investigate to which extent p57 contributes to apoptosis induction by combined treatment in UCCs, we generated VM-Cub1 p57 knockout clones by CRISPR/Cas9-mediated gene editing. In contrast to the parental cells, p57 protein remained undetectable in cell clones with successful knockout even after combined treatment (Fig. 5a). Surprisingly, however, p57 knockout did not rescue the cells from induction of cell death; viability was instead further decreased (Fig. 5b) and cleaved PARP was enhanced (Fig. 5a). Colony formation assays displayed concurring results (Fig. 5c), as did cell cycle analysis. Double treatment of p57 knockout cells disrupted the cell cycle completely, and the majority of cells died rapidly (Fig. 5d, Additional file 3).Fig. 5Effect of p57KIP2 gene knockout on induction of apoptosis by combination treatment. a Protein expression of p57 and PARP1 in parental and knockout VM-Cub1 cells with or without combination treatment. b Relative viability of VM-Cub1 parental and p57 knockout cells after combination treatment. Differences between samples with combination treatment of parental and knockout cells were analyzed using Student’s t test (***p ≤ 0.001). c Examples of clonogenicity assay, with quantified colony counts (***p ≤ 0.001) and d cell cycle analysis for the indicated cells following combined inhibitor treatment. See Additional file 3 for cell numbers in the respective cell cycle phases | other | 29.17 |
Effect of p57KIP2 gene knockout on induction of apoptosis by combination treatment. a Protein expression of p57 and PARP1 in parental and knockout VM-Cub1 cells with or without combination treatment. b Relative viability of VM-Cub1 parental and p57 knockout cells after combination treatment. Differences between samples with combination treatment of parental and knockout cells were analyzed using Student’s t test (***p ≤ 0.001). c Examples of clonogenicity assay, with quantified colony counts (***p ≤ 0.001) and d cell cycle analysis for the indicated cells following combined inhibitor treatment. See Additional file 3 for cell numbers in the respective cell cycle phases | other | 29.56 |
To follow the effects of the drug combination on the chromatin level, we analyzed changes in histone acetylation. Overall, acetylation of histones H3 and H4 increased after treatment with Romidepsin alone, but was further enhanced by combination treatment (Fig. 6a). We further analyzed H3K27 acetylation and H3K4 trimethylation, which are established markers for active transcription, around the transcriptional start sites of six genes affected by the drug combination using chromatin immunoprecipitation (ChIP). ChIP results for acetylated H3K27 revealed a broad and strongly enhanced enrichment of acetylation after combined treatment encompassing at least 2 kb upstream of the transcriptional start site (TSS) to 2 kb downstream of the TSS of all six investigated genes (Fig. 6b). By comparison, treatment with Romidepsin alone did not significantly increase H3K27 acetylation compared to DMSO controls. Strikingly, H3K4 trimethylation was increased at the TSS of BCL2, CDKN1C/p57, and SKP2 genes after treatment with JQ1 alone, but decreased, mostly significantly, at the TSS of genes downregulated by the combination treatment.Fig. 6Changes in histone modifications elicited by combination treatment. a Changes in H3 and H4 acetylation were detected after single and combined treatment in VM-Cub1 and UM-UC-3 cells by acid histone extraction and Western blot analysis. Total histone 3 served as a loading control. b Chromatin immunoprecipitation analysis of H3K27 acetylation at transcriptional start sites (TSS) as well as 2 kb upstream (− 2 kb) and 2 kb downstream (+2 kb) of BCL-2, BCL2L1/BCL-X, BIRC5/Survivin, c-MYC, CDKN1C/p57 and SKP2 genes after single and combined treatment of VM-Cub1 cells. Immunoprecipitated DNA amounts measured by qPCR for each condition are expressed as percentage of input DNA. Additionally, H3K4 trimethylation was analyzed at the TSS of the same genes. Rabbit IgG served as a background control. Differences between samples with combination treatment and DMSO treated cells were analyzed using Student’s t test (***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05) | other | 31.47 |
Changes in histone modifications elicited by combination treatment. a Changes in H3 and H4 acetylation were detected after single and combined treatment in VM-Cub1 and UM-UC-3 cells by acid histone extraction and Western blot analysis. Total histone 3 served as a loading control. b Chromatin immunoprecipitation analysis of H3K27 acetylation at transcriptional start sites (TSS) as well as 2 kb upstream (− 2 kb) and 2 kb downstream (+2 kb) of BCL-2, BCL2L1/BCL-X, BIRC5/Survivin, c-MYC, CDKN1C/p57 and SKP2 genes after single and combined treatment of VM-Cub1 cells. Immunoprecipitated DNA amounts measured by qPCR for each condition are expressed as percentage of input DNA. Additionally, H3K4 trimethylation was analyzed at the TSS of the same genes. Rabbit IgG served as a background control. Differences between samples with combination treatment and DMSO treated cells were analyzed using Student’s t test (***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05) | other | 32.12 |
In this study, we examined the efficacy of a combination treatment with small-molecule inhibitors of chromatin regulators in urothelial carcinoma cell lines. In previous studies, we had identified HDAC class I-specific inhibitors like Romidepsin to be the most efficient HDAC inhibitors in UC. However, as this treatment did not induce apoptosis straightforwardly and did not appear particularly selective for tumor cells, we searched for a second drug for an improved combination therapy [2, 4]. Here, we combined Romidepsin with the BET inhibitor JQ1, which had been proposed to synergize with HDACi in several cancer types and to induce a canonical apoptotic response . To our knowledge, the particular combination of Romidepsin and JQ1 has only been investigated by Jostes et al. in testicular cancer cell lines. These authors reported about reduced tumor burden even at lower and less frequent doses in xenograft experiments . Other studies to date applied mainly pan-HDAC inhibitors like panobinostat or SAHA [23–25]. | study | 27.81 |
The results of our current study clearly demonstrate a significant synergistic effect on cell proliferation and prominent apoptotic cell death for the combination of the BET protein inhibitor JQ1 and the HDAC class I-specific inhibitor Romidepsin. As UC cells are quite resistant to apoptosis induction under many conditions, we were especially interested in characterizing these cellular effects. | other | 32.3 |
A pioneer study on the efficacy of single treatment with JQ1 alone in two related UC cell lines (T24 and EJ) provided first direct evidence that UC cells may be dependent on BRD4 . shRNA mediated knockdown in the respective cells, decreased cell viability in a time-dependent manner (72 h), and induced cell cycle arrest in G0/G1 phase. In addition, in vivo tumor growth was reduced by BRD4 inhibition in a xenograft mouse model. We also found BRD4 overexpressed in some of the 12 investigated UC cell lines compared to benign urothelial cells, but overexpression was not uniformly observed, indicating that T24 may not be representative for all UC cell lines. Indeed, in our study, siRNA-mediated knockdown of BRD4 in both VM-Cub1 and UM-UC-3 cells resulted in a time-dependent reduction of cell viability and long-term proliferation, but the cells responded differentially with respect to cell cycle arrest. VM-Cub1 cells displayed the expected arrest in G0/G1, whereas UM-UC-3 arrested in G2/M. In many further functional assays, we likewise observed that UM-UC-3 cells responded differently or less efficiently, e.g., with regard to induction of apoptosis. Similar heterogeneity of BRD4 expression and response to BRD4 inhibition may occur in cancer patients due to the pronounced heterogeneity of urothelial carcinoma. This prompted us to characterize the response of UC towards single and combined treatment more comprehensively across different cell lines. | other | 29.6 |
Indeed, our investigations on the cellular effects of pharmacological BRD4 inhibition in eight UC cell lines chosen to represent the heterogeneity of urothelial carcinoma demonstrated that the previously reported results on T24 cannot be simply extended to all UC cell lines. Instead, dose-response curves for JQ1 in eight UC cell lines demonstrated that they could be clearly classified into two groups with low and high sensitivity and that not all UCCs are as sensitive as T24. The differential sensitivity was not related to BRD4 protein expression or any other obvious characteristics of the cell lines and remains to be further investigated. | other | 33.44 |
Like T24 in the study by Wu et al. , most cell lines accumulated in G0/G1 following treatment with IC50 doses of JQ1 (0.18–6.8 μM), but UM-UC-3 accumulated in G2/M. At these doses, we detected only a slight increase of the subG1-fraction. Comparable results were reported for colon cancer and medulloblastoma cells, which arrested in G0/G1 after treatment with 0.1–0.5 μM JQ1 [14, 26]. However, in contrast to various other cancer cell types, JQ1 treatment alone neither significantly inhibited long-term proliferation of UCCs nor induced significant apoptosis suggesting that UC cells are generally less sensitive towards JQ1 than other tumor types [8, 14, 24, 27] and recover over time. Benign urothelial cells also arrested at rather low concentrations, but no apoptosis was observed and long-term proliferation ability was retained. | study | 28.86 |
Using the Chou-Talalay method , we determined doses for the combined treatment with JQ1 and Romidepsin with synergistic effects on cell viability at doses mostly lower than the IC50 concentrations of either single treatment. Importantly, all UC cell lines, including those with low sensitivity to JQ1 alone like UM-UC-3 and 639-V, as well as primary tumor cultures, responded strongly to the combination treatment. The combination of JQ1 with Romidepsin was clearly more efficacious in inhibiting proliferation of UCCs than JQ1 alone. Moreover, the combination elicited apoptosis much more efficiently than either JQ1 or Romidepsin. Benign control cells were strikingly less affected suggesting that the combined treatment might achieve a higher tumor specificity with lower toxic side effects for normal cells than single treatment with Romidepsin . Accordingly, long-term proliferation was not affected in HBLAK cells, but was very strongly diminished in all cancer cell lines, again irrespective of their sensitivity towards JQ1 alone. The cell cycle was heavily disturbed, and we observed very strongly increased subG1 fractions as well as high numbers of early apoptotic cells subsequent to combination treatment in UCCs, but not in benign HBLAK cells. Similar synergistic effects on proliferation and apoptosis have been reported in pancreatic cancer, neuroblastoma, and AML cells by combination treatment using JQ1 with the pan-HDAC inhibitors SAHA (vorinostat) or panobinostat [8, 23, 24] and in melanoma cells by panobinostat and BET inhibitor I-BET151 . Borbeley et al. also combined the class I-specific inhibitor mocetinostat with JQ1 for treatment of breast cancer cells . Interestingly, normal melanocytes, embryonic fibroblasts, and normal hematopoietic progenitor cells were spared from apoptosis similar to HBLAK cells, again emphasizing the increased tumor specificity of the combination [7, 23, 24]. | other | 28.78 |
Molecular biomarkers and application of the caspase-inhibitor Q-VD-OPh confirmed that the combination treatment, other than treatment with either inhibitor alone, efficiently induced canonical caspase-dependent apoptosis in UC cells. Similar changes in apoptosis-related proteins were found in pancreatic cancer , neuroblastoma , melanoma , and AML cells . Anti-apoptotic proteins like BCL-2 may be induced by activated STAT3 [25, 28], and in pancreatic adenocarcinoma cells, the combination of JQ1 and SAHA diminished STAT3 phosphorylation to downregulate BCL-2 . However, in UC cells, we observed neither diminished phosphorylation nor decreased expression of STAT3, again indicating that the mechanisms mediating cellular effects may differ between cancer entities. Accordingly, gene expression profiles following treatment with BETi and HDACi drug combination in studies on different cancer cell types each describe hundreds of differentially expressed genes, with limited overlap, but converging on a cellular response resulting in cell cycle arrest and apoptosis [23, 27]. These relatively uniform ultimate cellular effects thus appear to be driven by cell type-dependent transcriptional programs. | other | 30.33 |
In addition, suppression of AKT-mediated survival signaling is likely to contribute to the observed cell death. Diminished AKT kinase activity may especially contribute to the dramatic overexpression of p57KIP2 via stabilization of the protein following decreased phosphorylation priming for SKP2-mediated degradation [29, 30], as especially non-phosphorylated p57KIP2 accumulated. SKP2 itself was as well strongly diminished by the treatment, presumably further contributing to p57 accumulation. In accordance with other studies reporting p57KIP2 as a prominent factor induced in treated cells [6, 8], we found both its mRNA and protein strongly elevated. | other | 28.78 |
The function of p57 with regard to regulation of apoptosis is controversial. Obviously, p57 can prevent or promote apoptosis dependent on the cellular context . Promotion of apoptosis may occur by p57 translocation into mitochondria to trigger the intrinsic apoptotic pathway . In pancreatic cancer cells, depletion of p57 by shRNA decreased apoptotic markers induced by JQ1 and SAHA combination treatment; Cas9-mediated knockout of Cdkn1c in mice significantly diminished apoptosis of combination-treated animals . In an analogous approach, we generated p57 knockout UC cells by gene editing. Intriguingly, this knockout did not rescue the cells from apoptosis induced by combination treatment. This finding underlines the context-dependent function of p57 on the one hand. On the other hand, it supports again the idea that compounds like JQ1 targeting general transcriptional regulators do rarely act through their effect on a single gene but elicit broad effects on the transcriptome which are cell type-dependent due to the respective transcriptional programs . For instance, in other studies, suppression of c-MYC was proposed to be the central event mediating effects in the treated cells, but forced overexpression of this supposedly crucial target gene failed to rescue all phenotypes . In this respect, we note that downregulation of c-MYC was observed in all UCCs treated with the drug combination, but that its extent varied. | study | 29.52 |
Recent publications have suggested a mechanism for the downregulation of oncogenic and anti-apoptotic factors like c-MYC or BCL-2 after combined treatment with HDAC and BET inhibitors [23, 33]. HDAC inhibitor treatment induces histone hyperacetylation, as also observed in UC cells. Genes undergoing hyperacetylation of H3K27 within a 5 kb region around their transcriptional start site by this treatment appear to become particularly dependent on binding of BET proteins like BRD4 to their regulatory regions, including enhancers, to promote gene transcription. Accordingly, genes activated by HDAC inhibitor treatment were reported to be enriched for BRD4 . Abolishment of BRD4 binding by JQ1 then leads to their transcriptional inactivation. This mechanism has been proposed for combined treatment with 4SC-202, Entinostat, or mocetinostat [27, 33], all of which predominantly target class I HDAC isoenzymes, similar to Romidepsin. Indeed, Romidepsin alone resulted also in activation of anti-apoptotic factors like BCL-2 and BCL-X in UC cells, suggesting that this mechanism also applies in our experimental setting. Accordingly, ChIP analysis revealed broad H3K27 hyperacetylation around the transcriptional start sites of all investigated genes after combined treatment, which would render them sensitive to transcriptional repression by JQ1. Consequently, H3K4 trimethylation, a marker for transcriptionally active genes, was significantly reduced at the TSS of four genes downregulated by combination treatment in UC cells. Notably, Mishra et al. reported globally increased H3K4me3 at the TSS of genes upon treatment with the HDAC inhibitor 4SC-202, which—different from Romidepsin—additionally inhibits the LSD1 histone demethylase. | study | 29.22 |
In conclusion, we have demonstrated that the combination of JQ1 and Romidepsin exerts strong synergistic antineoplastic effects on urothelial carcinoma cells. Due to the strong synergy, the applied doses were well below the IC50 doses of the single treatments and affected benign control cells only mildly. UC cells are often particularly resistant towards induction of caspase-dependent apoptosis . Therefore, this novel combination treatment constitutes a promising approach for a small-molecule therapy of UC with reduced toxic side effects to normal cells. | study | 27.64 |
Detailed flow cytometry results from cell cycle analysis. Displayed are percentages of cells in the indicated cell cycle phase as measured by flow cytometric cell cycle analyses following the indicated treatment for 48 h in four different UCCs and in HBLAK cells (Fig. 2d) or in VM-Cub1 knockout cells (Fig. 5d). (PDF 177 kb) | study | 34.38 |
Data on induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. Induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. Flow cytometric analysis of UCCs with indicated treatment after combined staining with PI and Annexin V. Percentages of viable (lower left), early (lower right), or late (upper right) apoptotic and necrotic (upper left) VM-Cub1 cells subsequent to indicated treatments. (PDF 1052 kb) | study | 35.44 |
Additional file 1:Information on primer sequences and antibodies. Sequence information and amplicon sizes for qRT-PCR and ChIP qPCR primers as well as product information and dilution of applied antibodies are given. (PDF 271 kb) Additional file 2:Data on synergistic effects on cell viability of primary cancer cells by combined treatment with Romidepsin and JQ1. (a), (b) Relative viability of primary cultures 1 and 2 established from primary tumor tissue of two different patients after single or combined treatment for 48 h. Relative viability is displayed on the ordinate in percent of the control cells treated with DMSO. (c) Primary culture 2 was used for an extended dose response curve analysis after 24 h to perform Chou-Talalay calculations (***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05). (d) Combination index (CI/Fa) plot for the combination of JQ1 and Romidepsin. Cell viability was measured at five constant dose ratio experimental data points by ATP assay for primary culture 2. CI plots were then generated using CompuSyn software. CI < 1 indicates synergism. (PDF 293 kb) Additional file 3:Detailed flow cytometry results from cell cycle analysis. Displayed are percentages of cells in the indicated cell cycle phase as measured by flow cytometric cell cycle analyses following the indicated treatment for 48 h in four different UCCs and in HBLAK cells (Fig. 2d) or in VM-Cub1 knockout cells (Fig. 5d). (PDF 177 kb) Additional file 4:Data on induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. Induction of apoptotic cell death by combined treatment with Romidepsin and JQ1. Flow cytometric analysis of UCCs with indicated treatment after combined staining with PI and Annexin V. Percentages of viable (lower left), early (lower right), or late (upper right) apoptotic and necrotic (upper left) VM-Cub1 cells subsequent to indicated treatments. (PDF 1052 kb) Additional file 5:Data on expression changes of of p57KIP2 mRNA by combination treatment in UC cells. Relative mRNA expression levels of p57KIP2 mRNA after single and combined treatment with Romidepsin and JQ1. mRNA levels were measured by qRT-PCR and normalized to the expression of TBP. Fold change compared to DMSO control 48 h after treatment is displayed on the ordinate. (PDF 100 kb) Additional file 6:Data on STAT3 activation and expression after combination treatment in UC cells. Phosphorylated and total STAT3 protein was detected by Western blot analysis in four UC cell lines cells after indicated treatment. α-tubulin served as an additional loading control. (PDF 216 kb) | study | 28.56 |
Data on synergistic effects on cell viability of primary cancer cells by combined treatment with Romidepsin and JQ1. (a), (b) Relative viability of primary cultures 1 and 2 established from primary tumor tissue of two different patients after single or combined treatment for 48 h. Relative viability is displayed on the ordinate in percent of the control cells treated with DMSO. (c) Primary culture 2 was used for an extended dose response curve analysis after 24 h to perform Chou-Talalay calculations (***p ≤ 0.001, **p ≤ 0.01, *p ≤ 0.05). (d) Combination index (CI/Fa) plot for the combination of JQ1 and Romidepsin. Cell viability was measured at five constant dose ratio experimental data points by ATP assay for primary culture 2. CI plots were then generated using CompuSyn software. CI < 1 indicates synergism. (PDF 293 kb) | study | 35.3 |
Data on expression changes of of p57KIP2 mRNA by combination treatment in UC cells. Relative mRNA expression levels of p57KIP2 mRNA after single and combined treatment with Romidepsin and JQ1. mRNA levels were measured by qRT-PCR and normalized to the expression of TBP. Fold change compared to DMSO control 48 h after treatment is displayed on the ordinate. (PDF 100 kb) | study | 33.9 |
Data on STAT3 activation and expression after combination treatment in UC cells. Phosphorylated and total STAT3 protein was detected by Western blot analysis in four UC cell lines cells after indicated treatment. α-tubulin served as an additional loading control. (PDF 216 kb) | study | 35.56 |
Patients with return of spontaneous circulation (ROSC) after out-of-hospital cardiac arrest (OHCA) suffer from prolonged, complete, whole-body ischemia and reperfusion, and have a grim prognosis [1, 2]. Post-cardiac arrest ischemia-reperfusion injury to the brain may reportedly be attenuated by induced hypothermia [3–5]. Thus, the 2015 international guidelines for cardiac arrest recommend a targeted temperature-management (TTM) strategy for adults sustaining an OHCA with an initial shockable rhythm who remain unresponsive after the ROSC [6, 7]. | other | 28.55 |
Patients resuscitated after cardiac arrest are reportedly at high risk of developing infectious diseases, especially pneumonia, several days after ROSC [2, 8–13]. Although recent guidelines recommend a TTM strategy [6, 7], hypothermia can impair the immune system and increase the incidence of infection [14, 15]. In addition, invasive medical techniques, such as extracorporeal membrane oxygenation (ECMO), intra-aortic balloon pumping (IABP) and continuous renal replacement therapy (CRRT), are used during TTM and organ support after OHCA . These invasive devices (especially ECMO) may themselves increase the risk of infection [17–19]. Consequently, patients undergoing TTM after cardiac arrest are likely to be at greater risk of infection and death. | other | 30.38 |
Although antibiotic administration is the mainstream treatment for infectious diseases, only a few retrospective studies have suggested a beneficial influence on the survival of OHCA patients who received early antibiotic administration while undergoing TTM [20, 21]. Moreover, there are no current reports of a relationship between early antibiotic administration and mortality among patients with OHCA requiring invasive life-supporting interventions. | other | 32.1 |
We hypothesized that prophylactic systemic antibiotic therapy during TTM may reduce mortality in OHCA patients, especially in those requiring invasive life-supporting interventions. The purpose of this study was to evaluate our hypothesis using a large nationwide inpatient database in Japan. | other | 34.88 |
We retrospectively evaluated the Japanese Diagnosis Procedure Combination (DPC) database, which was described previously [16, 22, 23]. Briefly, the DPC database includes administrative claims and discharge abstract data for all inpatients discharged from more than 1000 participating hospitals, covering all 82 academic hospitals and more than 90 % of all tertiary-care emergency hospitals in Japan [16, 22, 23]. The database includes the following information for each patient recorded using a uniform data-submission form: age; sex; primary diagnosis, comorbidities on admission, and post-admission complications coded using the International Classification of Diseases, 10th Revision (ICD-10) codes; medical procedures; daily records of all drugs administered and devices used; and discharge status. Patient follow-up began on the day of admission and ended on the date of discharge, either to home, to another hospital, or because of death. Patients with cardiac arrest upon arrival at the hospital were defined as experiencing OHCA, with chest compressions performed on/after arrival. We assumed that cardiac arrest was of cardiac aetiology unless there was an obvious non-cardiac cause (i.e. cerebrovascular disease, respiratory disease, severe trauma, drowning, asphyxiation or drug overdose) , in accordance with the International Liaison Committee on Resuscitation consensus statement [24, 25]. We defined patients with non-cardiogenic cardiopulmonary arrest as follows: ICD-10 codes at primary diagnosis and comorbidities upon admission of J60 to J61.9 [subarachnoid and cerebral haemorrhage], I71.0 to I71.9 [aorta dissection and aneurysm], I26.0 to I26.9 [pulmonary embolism], J45.9 and J46 [severe asthma], S00 to T98 [trauma, burns, hanging, accidental hypothermia, drowning, electrocution, anaphylaxis, drug overdose], J69 and T71 [asphyxiation], or K25.0, K25.2, K26.0, K26.2, K27.0, K27.2, K28.0, K28.2, K92.0 to 92.2, and I85.0 to I85.9 [acute gastrointestinal bleeding or oesophageal/gastric varices]). | other | 34.4 |
We designated hospital volume as the number of eligible patients treated for the current study and categorized hospitals into tertiles (i.e. low, medium, and high). The hospital type was categorized as academic or non-academic. The Japanese Ministry of Health, Labour and Welfare officially approved TTM for patients with cardiac arrest only (i.e. not in other disorders such as traumatic brain injury) in April 2006. | other | 32.78 |
Adults with cardiogenic OHCA were identified in the DPC database from July 2007 to March 2013. We did not include in-hospital cardiac arrest cases in this study. The inclusion criteria were as follows: (i) age ≥18 years; (ii) diagnosis of OHCA (ICD-10 codes at primary diagnosis and comorbidities on admission of I46.0 [cardiac arrest with successful resuscitation], I46.1 [sudden cardiac death], or I46.9 [cardiac arrest, unspecified]); (iii) and administration of therapeutic hypothermia on days zero and one. Exclusion criteria were as follows: (i) patients with non-cardiogenic cardiopulmonary arrest; (ii) patients with malignancy, caducity, and pneumonia upon admission (to maintain comparability between the groups; C00 to C97 [malignancy], R54 [caducity], J13 to J18, and J690 [bacterial pneumonia and aspiration pneumonia]); (iii) patients discharged within 2 days of admission (to avoid immortal time bias ). Thus, we compared patients who received antibiotics within 2 days of admission (antibiotics group) and those who did not receive any antibiotics within that time (control group) . | other | 37.28 |
We performed propensity scoring and instrumental variable analyses. To estimate the propensity score, we fit a logistic regression model designed for antibiotic treatment as a function of background-patient characteristics and in-hospital treatments or interventions performed on day 0. We included the following: age; sex; hospital type; hospital volume; fiscal year; cardiac arrest on hospital arrival; administration of epinephrine; ventricular fibrillation; defibrillation performed; requirement for PCI, ECMO, IABP and/or CRRT; requirement for catecholamines or vasopressin; and requirement for antiarrhythmic drugs, sivelestat sodium use, and blood transfusion. We performed a one-to-one matched analysis using nearest-neighbour matching based on the estimated propensity scores of the patients. A match occurred when a patient in the antibiotics group had an estimated score within 0.2 standard deviations of a patient in the control group . We examined the balance in baseline variables using standardized differences, where >10 % was regarded as imbalanced . Descriptive statistics are presented for all patients and propensity score-matched groups. Continuous variables were compared using a t-test. Categorical variables were compared using the chi-square test or Fisher’s exact test. We performed logistic regression analysis fitted with generalized estimating equations to examine the association between antibiotics use and survival and accounted for the paired nature of the propensity score-matched patients. We used Cox regression analysis to assess differences in in-hospital survival rates between patients with and without antibiotic treatment in the propensity score-matched groups. Odds ratios (ORs) and hazard ratios (HRs) with 95 % confidence intervals (CIs) were calculated. | other | 34.78 |
Survival plots for all patients treated with or without early antibiotics administration in propensity score-matched groups. There was no significant difference in 30-day mortality between the antibiotics and control groups. (Cox regression analysis; hazard ratio, 0.88; confidence interval 95 %, 0.75 to 1.04) | other | 28.61 |
In the instrumental variable model, the null hypothesis of ‘no association between patterns of hospital antibiotics use and actual antibiotics use’ was rejected (P < 0.001). Our instrument of pattern of hospital antibiotics use was not weak (F statistic = 702). The estimated reduction in 30-day mortality was not significantly associated with administration of antibiotics (6.6 %, CI 95 %, −0.5 to 13.7 %, p = 0.28). | other | 28.78 |
We also performed instrumental variable analysis as confirmatory analysis for the propensity score analyses . We used the hospitals’ antibiotics prescribing preference as an instrumental variable, and computed the differences in the risk of 30-day mortality between the groups with and those without early antibiotic administration. This approach was implemented using a 2-stage least-squares method that also adjusted for patient demographic characteristics. We classified hospitals that administered early antibiotics to the 50th percentile or more of eligible patients as hospitals with a preference for early antibiotics administration, and those that administered to less than the 50th percentile of eligible patients as hospitals without a preference for early antibiotics administration. We estimated the risk difference with its 95 % CI. To confirm that the percentage of hospital use of antibiotics was not a weak instrument, we used a partial F test. The null hypothesis was that there was no association between patterns of hospital antibiotics use and actual antibiotics use. An F statistic >10 suggests that the instrument is not weak . | other | 37.22 |
We also performed subgroup analyses to identify the patients who may have benefited the most from antibiotics therapy. We selected relevant subgroups among the propensity score-matched patients and estimated their risk ratios and 95 % CIs for 30-day mortality. For robustness, we further evaluated the subgroup of patients who had the most benefit from early antibiotics administration (i.e. the lowest risk ratio group of patients for 30-day mortality), and performed the same analyses as all patients, including Cox regression and instrumental variable analyses. | other | 34.53 |
Among unmatched patients, there was a significantly lower incidence of pneumonia after admission in the antibiotic group than in the control group (12.6 % vs. 15.3 %, difference −3.7 %; CI 95 %, −0.2 to −5.3 %, p = 0.04). However, this difference was not significant in the matched group (13.0 % vs. 15.8 %, difference −6.3 %; CI 95 %, −0.2 to 0.6 %, p = 0.10). | other | 31.42 |
The risk of 30-day mortality associated with antibiotics use in the propensity score-matched patients for subgroups is presented in Fig. 4. There was a significant association between the use of antibiotics and lower mortality in the subgroup of patients who required ECMO, IABP, or CRRT. The lowest risk of 30-day mortality associated with the administration of antibiotics was observed in the subgroup who required ECMO (risk ratio 0.67; CI 95 %, 0.50 to 0.89, p = 0.006). Among propensity score-matched patients who did not require ECMO, there was no significant difference in 30-day mortality between the two groups (28.6 % in the antibiotics group versus 27.8 % in the control group, difference 0.7 %; CI 95 %, −3.9 to 5.5 %, p = 0.76). In contrast, among patients who required ECMO, there was a significant difference in 30-day mortality between the 2 groups (62.9 % in the control group versus 43.5 % in the antibiotics group, difference 19.3 %; CI 95 %, 5.9 to 32.7 %; p = 0.005).Fig. 4Risk ratios of in-hospital mortality associated with antibiotic use in propensity score-matched patients. CRRT, continuous renal replacement therapy; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pumping; PCI, percutaneous coronary angiography/intervention; vf, ventricular fibrillation | study | 31 |
During the study period, 95,960 OHCA patients were identified in the database. Of the patients initially identified, 2803 eligible cardiogenic patients with OHCA were treated with TTM in 371 hospitals (Fig. 1). These patients were categorized into the antibiotic-treated (n = 1272) and control groups (n = 1531), and 802 propensity score-matched pairs were generated (Fig. 2). Table 1 shows the demographic and clinical characteristics of the patients in the two groups. In the antibiotics group (n = 1272), the most common antibiotics administered were as follows: first-generation cephalosporin for 58.6 % (746 patients), third-generation cephalosporin for 10.4 % (145 patients), and broad-spectrum penicillin for 6.3 % (80 patients).Fig. 1Patient selection Fig. 2Propensity score matching process Table 1Patient characteristics, and initial treatments and interventions Unmatched groups Propensity score-matched groups VariableControl (n = 1531)Antibiotics (n = 1272)Standardized differences, %Control (n = 802)Antibiotics (n = 802)Standardized differences, %Age, mean (SD)60.1(15.3)60.7(14.3)−3.560.7(14.8)60.6(14.6)1.0Sex (male)a 1,151(75.2)984(77.4)−5.1613(76.4)609(75.9)1.2Academic hospital620(40.5)392(30.8)20.3262(32.7)279(34.8)−4.5Hospital volume, cases Low, <10484(31.6)505(39.7)−16.9317(39.5)305(38.0)3.1 Medium 11–22470(30.7)419(32.9)−4.8271(33.8)264(32.9)1.9 High, >23577(37.7)348(27.4)22.2214(26.7)233(29.1)−5.3Ventricular fibrillation746(48.7)645(50.7)−4.0397(49.5)413(51.5)−4.0Required defibrillation on admission328(21.4)409(32.2)−24.4186(23.2)204(25.4)−5.2Cardiac arrest on admission540(35.3)586(46.1)−22.1320(39.9)319(39.8)0.3Epinephrine provided on admission597(39.0)617(48.5)−19.3342(42.6)338(42.1)1.0Percutaneous coronary intervention872(57.0)902(70.9)−29.4507(63.2)536(66.8)−7.6Intra-aortic balloon pumping308(20.1)483(38.0)−40.1218(27.2)204(25.4)4.0Continuous renal replacement therapy136(8.9)248(19.5)−30.899(12.3)104(13.0)−1.9Extracorporeal membrane oxygenation system Pharmacologic intervention147(9.6)298(23.4)−37.9105(13.1)108(13.5)−1.1 Dopamine741(48.4)703(55.3)−13.8408(50.9)425(53.0)−4.2 Dobutamine286(18.7)403(31.7)−30.3189(23.6)176(21.9)3.9 Norepinephrine487(31.8)584(45.9)−29.2300(37.4)300(37.4)0.0 Vasopressin51(3.3)53(4.2)−4.428(3.5)26(3.2)1.4 Amiodarone371(24.2)463(36.4)−26.7234(29.2)244(30.4)−2.7 Nifekalant76(5.0)103(8.1)−12.745(5.6)57(7.1)−6.1 Lidocaine370(24.2)455(35.8)−25.5233(29.1)241(30.0)−2.2 Sivelestat sodium44(2.9)84(6.6)−17.632(4.0)33(4.1)−0.6Blood transfusion Red blood cells133(8.7)268(21.1)−35.392(11.5)99(12.3)−2.7 Fresh frozen plasma86(5.6)199(15.6)−33.059(7.4)66(8.2)−3.3 Platelets21(1.4)61(4.8)−19.916(2.0)15(1.9)0.9 anumbers in parentheses are proportions (%) unless otherwise stated | review | 26.73 |
The overall 30-day mortality was 32.5 % (910 of 2803 patients), and the incidence of pneumonia after admission was 14.1 % (395 of 2803 patients). Among unmatched patients, significant differences in 30-day mortality were observed between the two groups (30.3 % in the control group compared with 35.1 % in the antibiotics-treated group, difference 4.8 %; CI 95 %, 1.2 to 8.2 %, p = 0.007), but not among matched patients (33.0 % versus 29.9 %, respectively, difference 3.1 %; CI 95 %, −1.4 to 7.7 %, p = 0.18). Logistic regression analyses showed that there was no significant association between the use of antibiotics and lower 30-day mortality in the propensity score-matched groups (OR 0.89; CI 95 %, 0.73 to 1.1; p = 0.24). The Cox regression analysis did not indicate significant differences in-hospital mortality between the control and antibiotics groups for the propensity-matched groups (HR 0.88, CI 95 % 0.75 to 1.04, p = 0.12; Fig. 3).Fig. 3Survival plots for all patients treated with or without early antibiotics administration in propensity score-matched groups. There was no significant difference in 30-day mortality between the antibiotics and control groups. (Cox regression analysis; hazard ratio, 0.88; confidence interval 95 %, 0.75 to 1.04) | review | 28.94 |
Risk ratios of in-hospital mortality associated with antibiotic use in propensity score-matched patients. CRRT, continuous renal replacement therapy; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pumping; PCI, percutaneous coronary angiography/intervention; vf, ventricular fibrillation | study | 34.62 |
Among patients who required ECMO, a logistic regression analysis indicated that there was a significant association between the the early use of antibiotics and lower 30-day mortality in the propensity score-matched groups (OR 0.47, CI 95 %, 0.27 to 0.83, p = 0.009). Cox regression analysis showed significant mortality differences between the control and antibiotics groups for the propensity-matched groups (HR 0.61, CI 95 %, 0.43 to 0.87, p = 0.005; Fig. 5). In the instrumental variable model, the estimated reduction in 30-day mortality associated with receipt of antibiotics was 18.2 % (CI 95 %, 2.1 to 34.4 %, p = 0.03). The lower incidence of pneumonia after admission in the early antibiotics-treated group than in the control group was not significant among patients who required ECMO (9.3 % vs. 13.3 %, difference −4.1 %; CI 95 %, −12.6 to 4.4 %, p = 0.35).Fig. 5Survival plots for a subgroup of patients who required extracorporeal membrane oxygenation in the propensity score-matched groups. There was a significant difference in 30-day mortality between the antibiotics and control groups (Cox regression analysis; hazard ratio, 0.61; 95 % confidence interval, 0.43 to 0.87) | other | 29.66 |
Survival plots for a subgroup of patients who required extracorporeal membrane oxygenation in the propensity score-matched groups. There was a significant difference in 30-day mortality between the antibiotics and control groups (Cox regression analysis; hazard ratio, 0.61; 95 % confidence interval, 0.43 to 0.87) | review | 28.31 |
The results of this study using a nationwide database suggest that there was no significant association between the use of early antibiotics administration and survival in a cohort of patients with OHCA receiving TTM. However, we found a significant association between the use of antibiotics and lower mortality in the subgroup of patients who required ECMO, IABP, or CRRT. | study | 29 |
The use of ECMO in the intensive care unit has substantially increased over the past ~10 years [17, 29, 30], including during cardiopulmonary resuscitation (CPR) and afterwards [16, 31, 32]. The results of a recent meta-analysis by Xie et al. suggested that ECMO has the potential to improve outcomes after cardiac arrest; however, they also suggested that ECMO is associated with significant complication rates, including infection, which must be incorporated into the risk-benefit analysis when considering treatment . The latest resuscitation guidelines state that extracorporeal CPR is a reasonable rescue therapy for selected patients with cardiac arrest when initial conventional CPR is failing; however, they provide no guidance regarding the role of antibiotic administration [6, 7]. | study | 35.03 |
A low cardiac output state combined with high systemic vascular resistance results in poor systemic perfusion in OHCA patients who successfully achieve ROSC. Post-cardiac arrest acute kidney injury occurs in more than 50 % of cardiac arrest patients . Potential treatments to counteract this severe hemodynamic status and acute renal failure include IABP and CRRT. Although these devices themselves may increase the risk of infection, the use of prophylactic antibiotics is controversial . No studies have evaluated the relationship between early antibiotic administration and mortality among OHCA patients requiring these invasive life-supporting interventions. | study | 31.47 |
We found that patients with cardiogenic OHCA who required ECMO, IABP, or CRRT might benefit from early antibiotic treatment in our propensity score and instrumental variable analyses. Early antibiotic administration may have the greatest therapeutic benefits in patients with the most severe OHCA patients with TTM. Grimaldi et al. recently reported that nearly half of the patients with post-cardiac arrest shock had high blood endotoxin concentration in the first 12 h following ROSC. They also reported that endotoxemia was associated with shock duration, mean daily dose of vasopressors and organ failure . Thus, among those patients with OHCA and severe hemodynamic disturbance to require ECMO, IABP, and/or CRRT, early antibiotics administration may have the potential to improve outcomes. | study | 31.47 |
Another explanation could be that ECMO, IABP, and/or CRRT themselves increase the risk of nosocomial infection. Patients on ECMO for extracorporeal CPR reportedly developed 24.7 infections per 1000 ECMO days according to data from the Extracorporeal Life Support Organization Registry , and these infections were associated with increased mortality. Nevertheless, the Task Force does not currently recommend routine antibiotic administration during ECMO, due to lack of evidence of its efficacy . Furthermore, previous studies showed that the incidence of catheter-related bacteremia in patients who required hemodialysis ranged between 0.6 and 6.5 episodes per 1000 catheter days, and increased linearly with the duration of catheter use [18, 36]. Although we cannot infer a robust cause-and-effect relationship from this study, we speculate that there may be an association between early antibiotic administration and improved survival in OHCA patients requiring ECMO, IABP, and/or CRRT during TTM. We believe that our nationwide retrospective data will be useful when planning for future international trials. | other | 31.98 |
We analysed real-world clinical data of more than 2,800 adults with cardiogenic OHCA treated at 371 hospitals. In the present study, the decision to provide antibiotics was left to the discretion of physician of each hospital. To date, there have been no randomised trials or nation-wide studies regarding this issue. Accordingly, there are no guidelines whether to provide antibiotics for patients after cardiac arrest due to the lack of data. Approximately, 45 % of the patients received antibiotics in the present study. Although analysis of the baseline patient characteristics in the unmatched group showed that antibiotics were used in a greater proportion of patients with more severe status after cardiac arrest (i.e. those at greater risk of death and more likely to require defibrillation upon admission, PCI, IABP, ECMO, catecholamines, antiarrhythmic medications, or blood products), one-to-one propensity score-matching successfully balanced the characteristics of the antibiotic and control groups. Although patients receiving antibiotics appeared less likely to die than did similar patients who did not receive antibiotics (mortality rates 33.0 and 29.9 %, respectively), this difference was not statistically significant. | other | 31.22 |
Similar results were observed in the instrumental variable analysis, which also found that there was no significant reduction in 30-day mortality associated with the receipt of antibiotics. When hospitals have highly consistent antibiotic-use policies for post-cardiac arrest patients treated with a TTM strategy, decisions regarding antibiotic use may be made independently of individual patient characteristics, including unmeasured variables, as OHCA occurs independent of the location of the admitting hospital. In this situation, the hospital’s antibiotics use policy may act as an instrumental variable, thereby setting the conditions for a ‘natural experiment’ that allows for an unbiased estimate of risk in patients with OHCA undergoing TTM. Considering the results of the propensity score and instrumental variable analyses, early antibiotic-administration appears to have little influence on survival during TTM after OHCA. Our negative results of the present study may have resulted from a lack of statistical power. However, the absolute difference in 30-day mortality in the propensity-matched analysis was 3.1 %, thereby suggesting that at least 7272 patients (3636 patients per group) would be required to show a significant difference in future trials (α error = 0.05, 1-β [power] = 0.8). | other | 31.11 |
Although the incidence of pneumonia after OHCA is inconsistent across previous studies, the incidence of pneumonia in the present report may be lower than that of previous reports outside Japan [8, 10, 15, 20, 21]. This may be due to several reasons. Firstly, because of its retrospective design, the diagnosis of pneumonia was assigned to the doctors in charge of each 371 hospitals. Thus, there might be some under-reporting (or over-reporting) in the current results. Secondly, in the present study, we included only cardiogenic OHCA patients that underwent TTM. In Japan, during the study period, most TTMs were managed with a milder target temperature for a longer period (e.g. duration of cooling for >2 days in 34 °C and >2 days for rewarming [32, 37]), than the recommendations in international guidelines published during the study period (i.e. cooling to 32 °C to 34 °C for 12 or 24 h ). Due to these differences, both the incidence and recognition of pneumonia by the attending doctors may have decreased. Further studies are required to clarify our speculations. | other | 32.03 |
Our study has several limitations. Firstly, although we evaluated nationwide data using a propensity score and instrumental variable analyses, the study was conducted retrospectively. Therefore, a cause-and-effect relationship cannot be established. Large randomized trials are necessary to confirm our findings; however, such studies are challenging in this vulnerable population. Indeed, the mortality rate exceeds 50 % in those who underwent ECMO. Secondly, a limited number of covariates could be analysed as the database did not include several important factors, particularly pre-hospital data including whether the OHCA was witnessed by a bystander, the quality of resuscitation given, whether resuscitation was dispatcher-assisted, and how the emergency services treated the patients . Although we endeavoured to compensate for these potential unmeasured confounders using instrumental variable analyses, further studies are required. Thirdly, we evaluated cardiogenic OHCA patients who underwent therapeutic hypothermia without pneumonia on hospital arrival in the present study. However, since pneumonia often presents gradually after admission, there might be some undiagnosed or underreported pneumonia. In addition, we could not identify catheter-related infections from the current database. Fourthly, the DPC is an administrative database with information input in relation to patient discharge. We could not follow up patients after discharge from the hospital because this information was not available in the database. Finally, we evaluated only adults with cardiogenic OHCA treated by TTM. Our results cannot be generalized to the use of prophylactic or early administration of antibiotics for children, adults with non-cardiogenic OHCA (especially who already had pneumonia on admission), in-hospital cardiac arrest patients, or those not treated with TTM. | study | 29.88 |
Although there was no significant association between the use of early antibiotics administration and mortality in a large cohort of patients with cardiogenic OHCA undergoing TTM, antibiotics appear to improve survival in a subgroup of patients that require ECMO. Further studies are required to confirm our results. | study | 27.66 |
Cyanobacteria are a group of primitive microorganisms present in several geographical areas worldwide that are able to grow and proliferate in all aquatic and terrestrial ecosystems, from tropical forests to deserts, oceans, and lakes . Eutrophication and climate change may enhance the bloom of cyanobacteria, increasing the amount of bioactive secondary metabolites in freshwaters, compounds that might pose a risk to humans, animals, and plants . Among these secondary metabolites, the alkaloid Cylindrospermopsin (CYN) is an emerging toxin, originally produced by the cyanobacteria Cylindrospermopsis raciborskii, but now exhibiting a cosmopolitan distribution pattern . CYN producers belong to several genera, such as Aphanizomenon, Anabaena, Raphidiopsis, Lyngbya, and Umezakia , recorded from a great variety of habitats. CYN is a cytotoxin with the structure of a tricyclic guanidine along with a uracil ring . | other | 27.17 |
The mechanisms of the toxic action of CYN in several organs are not totally clarified , although evidence suggests that, both in vitro and in vivo, CYN induced a concentration-dependent protein and glutathione synthesis inhibition as well as genotoxicity due to DNA fragmentation . Besides, CYN activation by cytochrome P450 seems to be related to its increased toxicity , suggesting that the initial compound and its possible metabolites could act by different mechanisms . Oxidative stress is also involved as a mechanism of CYN toxicity in both in vitro and in vivo assays . For consumers protection and to prevent CYN adverse effects, 0.03 µg kg−1 of body weight (b.w.) has been settled as the provisional Tolerable Daily Intake (TDI) for this toxin . | study | 31.47 |
In comparison to mammals, the amount of studies that investigated the physiological and biochemical effects of CYN on diverse plants is scarcer. These studies have been reviewed by Corbel et al. (2014) and, more recently, by Machado et al. (2017) . These effects include alterations in growth , germination, and development , oxidative stress , decreases in chlorophyll , chromosomal aberrations , changes in mineral content and in the proteome . | other | 28.72 |
Generally, humans can be exposed to cyanotoxins by the oral route through several pathways: when drinking water contaminated with cyanotoxins, when consuming fish, vegetables, crops and even food supplements susceptible to contain cyanotoxins, or accidentally by ingestion of water while performing recreational activities . The ability of cyanotoxins such as microcystins (MCs) to accumulate in the tissues of a wide range of agricultural crops has been described and reviewed , although only a few studies have been carried out on the leaves of edible plant species, such as lettuce (Lactuca sativa) . | study | 27.53 |
However, in spite of the human health risks associated with CYN, less attention has been given to its bioaccumulation in edible agricultural plants compared with MCs , which is considered a public health concern, given CYN's high water-solubility and its ability to transfer to higher trophic levels . This bioaccumulation may be possible because of the absorption of toxins by plants if surface water contaminated with cyanotoxins is used in agriculture, thus posing risks for food safety. In this sense, field studies have shown a broad range for concentrations of CYN in waters worldwide, although ecologically relevant CYN levels potentially used for the irrigation of vegetables ranged between 5 and 100 µg L−1 . At these concentrations, the ability of CYN to enter the food web by this last route has been minimally studied. To fill this gap in knowledge, advances in this field could provide helpful information for public health, concerning the bioaccumulation of CYN in plants and how water resources should be used to minimize crop contamination. White et al. (2005) suggested that the aquatic macrophyte Hydrilla verticillata might adsorb CYN in the plant cell walls instead of taking up the toxin through the cells. CYN transdermal absorption could represent an important route of plant uptake because of CYN high presence in dissolved form in the environment . Silva & Vasconcelos (2010) observed that the roots of L. sativa, Phaseolus vulgaris, and Pisum sativum accumulated higher concentrations of CYN in comparison to the stems. Additionally, Prieto et al. (2011) found accumulation of CYN in plants of Oryza sativa exposed to 2.5 μg CYN L−1 from an extract, showing a higher amount of the toxin in the roots, compared to the leaves. CYN was also detected in various Brassica vegetables (roots and leaves) after exposure to a CYN-containing extract, and this accumulation seemed to depend on the concentrations applied to the roots (18–35 μg L−1), with the levels of CYN ranging from 10 to 21% in the leaves . Recently, a study showed that the bioaccumulation of CYN in lettuce (L. sativa) and arugula (Eruca sativa) depends indirectly on the exposure concentration, and also on time and on the species . For example, after 7 days of exposure, the CYN concentrations measured in lettuce leaves were 8.29, 4.19, and 3.78 µg CYN kg−1 for exposure concentrations of 3, 5, and 10 µg CYN L−1, respectively. In the case of arugula, these values were 11.49, 10.41, and 9.47 µg CYN kg−1 for the three levels of exposure. | study | 28.73 |
Fresh products such as lettuce, spinach, cabbage, and sprouts are processed minimally. Their consumption has increased over the last decade, maybe because of their high nutritional value, changes in social eating habits, and wider accessibility . Some recommendations by international organizations and limit values have been set worldwide in order to prevent or manage the potential effects on human health induced by the exposure to CYN under different scenarios . However, no recommendations or limits have been established so far in the case of vegetables, despite the results demonstrating the risk of ingesting CYN through contaminated vegetables . Following the recommendations of the European Food Safety Authority (EFSA), there is a need to develop analytical methods for sample preparation and detection of cyanotoxins in complex matrices such as food items . In this sense, some methods for cyanotoxins analysis in different matrices such as natural blooms, cyanobacterial strain cultures, and biological samples, are reported in the literature and use Liquid Chromatography (LC)–Mass Spectrometry techniques , whose criteria and applications have been recently reviewed by Caixach et al. (2017) . Concerning CYN, no validation procedures with robustness tests have been developed in vegetables compared to other matrices such as waters, cyanobacterial cultures, or fish tissues . In fact, and specifically in relation to the matrices assayed in the present study, CYN has been detected in Brassica vegetables, lettuce, and arugula by ELISA and LC–MS/MS , but these methods have not been validated. Ultra-performance liquid chromatography–tandem mass spectrometry (UPLC–MS/MS) allows excellent specificity and sensitivity for cyanotoxins detection and quantification in waters and also in more complicated matrices, becoming the technique of choice for these purposes . | other | 29.17 |
For these reasons, the present work aimed for the first time to develop and validate an analytical method based on UPLC–MS/MS for the extraction and quantification of CYN in edible vegetables samples (Lactuca sativa L.). The method has been optimized and validated according to international guides . The present procedure has been designed for the routine determination of CYN in leaves of edible lettuces intended for human consumption, for prevention and risk assessment purposes. | other | 31.38 |
Before testing the efficiency of the CYN extraction method, the UPLC–MS/MS method was setup for this purpose. For this, commercially purchased CYN standard solutions were tested, acquiring mass spectra and adjusting the mobile phase strength, as previously performed in our laboratory . The spectrum was obtained on collision of m/z 416, corresponding to the pseudomolecular ion [M + H]+. The selected transitions for the quantitation and confirmation of the analyte CYN were 416.2/194.0 and 416.2/176.0, respectively. Specifically, the signals at m/z 194 and 176 correspond to the loss of SO3 and H2O from the fragment ion at m/z 274, corresponding to the loss of the [6-(2-hydroxy-4-oxo-3-hydropyrimidyl)] hydroxymethinyl moiety from the CYN structure . | other | 34.72 |
A prerequisite for carrying out a quantification is to establish a calibration function for the final measuring instrument . The response as a function of concentration was calculated from CYN standards prepared in fresh lettuce leaves extracts and was measured by a calibration curve including six points within the linear range 5–500 µg L−1 (equivalent to 5–500 ng CYN g−1 f.w. lettuce). The regression equation obtained was y = 378.55x + 13.424 (r2 = 0.9999) (Figure 1). | other | 31.86 |
Six different concentrations of CYN were spiked in blank extracts of lettuce leaves (5–500 ng CYN g−1 f.w.), and the samples were submitted to the proposed method. The response linearity was set according to Huber (1998) by plotting each concentration assayed against its respective response factors (signal response/analyte concentration). The target line represents the median of the response factors obtained and has zero slope, while the two parallel horizontal lines represent 0.95 and 1.05 times the median value (Figure 2). The linear range of the method applies to the full range studied because no intersections with the lines were found. Moreover, with the same signal responses, the corresponding ANOVA of the regression line was performed, indicating a lack-of-fit F ratio of 5.39 compared to a tabulated F value of 19.4, and, consequently, the calibration function can be considered as linear. | other | 33.72 |
The limits of detection (LOD) and quantitation (LOQ) are based upon the variability of the blank and are calculated according to the equation YLOD or LOQ = Yblank + nSblank, where Yblank is the mean value of 10 blank signals, Sblank its corresponding standard deviation, and n is a constant (3 for LOD and 10 for LOQ). The LOD and LOQ obtained were 0.22 ng CYN g−1 f.w. and 0.42 ng CYN g−1 f.w (equivalent to 0.22 and 0.42 µg L−1), respectively. Both limits are below the proposed guideline value of 1 µg L−1 for CYN in waters and are lower than the LOD and LOQ values of 0.5 and 80 µg L−1, respectively, reported in the same matrix, i.e., lettuce leaves, by Cordeiro-Araújo et al. (2017) . Moreover, the LOD and LOQ found in the present study would permit the quantification of CYN at the concentrations detected in other studies carried out with Brassica vegetables (2.71 ng g−1) , as well as with lettuce and arugula (3.76 and 5.5 ng CYN g−1 f.w, respectively) . | other | 35.44 |
Precision is a measure of the closeness of agreement between mutually independent measurement results obtained under specified conditions. It is generally dependent on the analyte concentration , and its measure includes three concepts: repeatability, intermediate precision, and reproducibility . Repeatability is a measure of variability in the results in the case that the measurements are performed on the same material by a single analyst employing the same method and equipment over a short timescale, while intermediate precision estimates the variation in the results in the case that the measurements are made on the same material in a single laboratory applying the same method over an extended timescale, and therefore refers to more variable conditions compared to repeatability. On the other hand, reproducibility is a measure of the variability in the results when the measurements are performed in different laboratories and, therefore, of what is expected to represent the greatest variation in the results . | other | 33.6 |
The values of repeatability (within-day and between-day, Sw and SB), intermediate precision (intralaboratory reproducibility, SIP), and SIP relative standard deviations (%RSDIP) were calculated analyzing three replicates of lettuce leaves extracts spiked with standard CYN solutions at different concentrations (20, 200 and 500 µg L−1) on the same day, following the ICH guidelines, and over a period of three consecutive days. Considering three different days as the main source of variation, the estimations of the precision parameters were obtained by performing an analysis of variance (ANOVA) for each validation standard according to González and Herrador (2007) and González et al. (2010) (Table 1). According to the Eurachem/CITAC Guide our results show that the values of repeatability (SW) represent the smallest variation. Moreover, the corresponding %RSDIP were compared with the acceptable RSD percentages obtained from the AOAC guidelines (2016) . Our results, at the three concentration levels considered, were lower than or of the same order of the tabulated %RSDAOAC (15–21% for 20 µg L−1, and 11–15% for 200–500 µg L−1). | review | 29.88 |
Trueness is the closeness of agreement between a test result and the accepted reference value of the property being measured. Trueness is stated quantitatively in terms of “bias”, with smaller bias indicating greater trueness , and can be expressed as the recovery obtained for each validation standard assayed. The total recovery for each validation standard (20, 200, or 500 µg L−1) was obtained as the ratio between the observed estimation of the standard concentration and the true value, expressed as fraction or as percentage. The recoveries (%) obtained were 104 ± 2 for 20 µg CYN L−1, 90 ± 5 for 200 µg CYN L−1, and 85 ± 3 for 500 µg CYN L−1. The suitability of these values was checked by comparison with the published acceptable recovery percentages depending on the analyte concentration, according to AOAC (2016) . Specifically, taking into account the CYN concentration of the three validation standards, the acceptable recovery range (%) according to AOAC could vary within 80% and 110%. Consequently, in terms of recoveries, the proposed method can be considered as acceptable. | other | 33.7 |
The ruggedness (sometimes also called robustness) is a parameter that has often been required in the validation studies of analytical methods and to assess the vulnerability of factors in the analytical methods optimized by using the chemometric tools . These studies evaluate a method’s capacity to remain unaffected by small variations in some method parameters . The strategy when performing the present ruggedness study was based on the method suggested by Youden (1967) , according to González and Herrador (2007) . Three deliberate variations in the extraction procedure of the method were chosen: F1, i.e., time for the sample to pass through the cartridge, F2, i.e., the water volume for washing the cartridge, and F3, i.e., the elution volume of DCM/MeOH. The levels were coded as specified in Table 2. By combining these parameters, the resulting eight different possibilities were tested. To determine whether the variations in an experimental setting have a significant effect on the results, a significance t test is applied by González and Herrador (2007) , and the t values (Fx) are compared with the 95% confidence level two-tailed tabulated value (ttab) with eight degrees of freedom (d.f.) derived from the precision study for each concentration tested. The effect of every factor is estimated as the difference of the average result obtained at the level +1 and that obtained at the level −1. The two-level full factorial design is the most efficient chemometric tool for ruggedness evaluation . In the present work, the experiments were performed using lettuce leaves spiked with an intermediate validation standard of 200 µg CYN L−1 (equivalent to 200 ng g−1 f.w.), and each factor was analyzed by triplicate on three different days obtaining eight d.f. The t values obtained were F1 = 1.36, F2 = 0.297, and F3 = 0.208, being in all cases lower than the tabulated value (ttab = 2.306). These results showed that the present validated method can be considered as a robust method against the three factors considered at the levels fixed in the study. | other | 33.2 |
The validated method was finally applied to quantify the unconjugated CYN concentrations in the leaves of the control lettuce (n = 5) and of lettuces grown in a soil-free hydroponic cultivation system (n = 5), following the experimental conditions indicated (‘Experimental exposure of vegetables and application of the validated method’ section). CYN was not detected in the control lettuces, and 2.37 ± 0.80 ng CYN g−1 f.w. was detected in the experimental lettuces exposed to 10 µg CYN L−1 (Figure 3). The CYN uptake obtained (23.7%) for lettuce agrees with the previous values presented by Kittler et al. (2012) for kale (Brassica oleracea var. sabellica), corresponding to 10–15%, and for vegetable mustard (Brassica juncea) (12–21%) exposed to 18.2–35.5 µg CYN L−1, using a hydroponic cultivation system. Similarly to these authors, in this work we have demonstrated that CYN is absorbed by the roots of lettuces and transported to the leaves and that the adsorptive effects of the soil particles have not been taken into account. Previously, the transfer of CYN to the upper part of a plant (stem) was also observed in lettuce (L. sativa), bean (Phaseolus vulgaris), and peas (Pisum sativum) exposed to CYN-contaminated water obtained by exposure to a C. raciborskii strain (0.57–57 µg CYN L−1) . Prieto et al. (2011) detected CYN in the leaves of O. sativa plants (measured by ELISA test) suggesting the uptake of CYN by the roots and its further translocation to other organs of the plant. Moreover, the only data found in the scientific literature reporting CYN levels in lettuce leaves, obtained by a non-validated method in this matrix by Cordeiro-Araújo et al. (2017) , were also similar to those found in the present work. These authors found a concentration-dependent CYN bioaccumulation in lettuce, and the mean value reported after exposure to 10 µg CYN L−1 for 7 days was 3.78 ± 0.25 ng g−1. Our results confirm the uptake of CYN in edible vegetables exposed to low and ecologically relevant concentrations of the toxin, posing potential public health consequences. | other | 28.61 |
In comparison to other cyanotoxins, MC-LR showed a similar pattern of accumulation in agricultural plants . Thus, in lettuce sprayed with MCs (ranged between 0.62 and 12.5 µg L−1 for 15 days), the concentrations found in the foliar tissues were 8.31–177.8 µg MCs kg−1 f.w. , higher than the values obtained by Crush et al. (2008) . The accumulation of MCs in lettuce had a significant positive correlation with the different exposure concentrations of MCs (MC-LR and MC-RR) and could imply serious human health risks when consuming contaminated vegetables , although the possibility of MC-LR decontamination of lettuce leaves has been demonstrated . | other | 30 |
Assuming a human consumption of 40 g of vegetable per day for an adult of 60 kg (equivalent to 0.65 g kg−1 b.w. day) , the ingestion is 1.58 ng CYN kg−1 b.w. per day (2.37 ng CYN/g f.w. × 40 g vegetable/60 kg b.w.). Taking into account the recommended TDI for CYN of 0.03 µg kg−1 per day, the consumption of lettuce leaves in this case (1.58 ng CYN kg−1 b.w. per day) would account for 5.3% of the TDI of this toxin. For a more detailed exposure assessment in relation to different age groups, we took into account the data provided by the EFSA, Comprehensive European Food Consumption Database (Table 3) for normal and high consumers (95th percentile). Thus, it was evidenced that the CYN uptake in adults (mean ranged between 0.07–1.54 ng CYN kg−1 b.w.day) was similar to that reported by Cordeiro-Aráujo et al. (2017) . Moreover, the age group with a higher CYN exposure would by the elderly (65–75 years) with a potential CYN exposure of 0.17–12.77 ng CYN kg−1 b.w.day (mean values) or of 0.57–70.53 for high consumers. These values could contribute to the TDI of CYN in a range of 0.55–42.58% and 1.90–235.1%, respectively. Therefore, in this group age health risks after consumption of CYN-contaminated lettuce cannot be excluded. These findings highlight the need to develop adequate validated methods for an accurate determination of CYN in order to know and to prevent the health risks for the consumers. | study | 30.17 |
A UPLC–MS/MS method has been developed and validated for the first time in order to detect unconjugated CYN in the leaves of lettuce (L. sativa), showing acceptable sensitivity, reproducibility, accuracy, and robustness. Its linearity, the LOD and LOQ, recoveries (85–104%), and intermediate precisions obtained (12.72–14.70%) confirm its validation. In addition, this method has been successfully applied to detect CYN at ecologically relevant concentrations in lettuce leaves exposed to CYN, showing that CYN can accumulate in edible vegetables intended for human consumption. Although the TDI was exceeded in most age groups considered, in elderly high consumers health risks after consumption of CYN-contaminated lettuce cannot be excluded. These results demonstrate the need and importance of developing accurate analytical methods to monitor CYN in leaf vegetables with the intention of reducing or preventing the associated potential health risks. | study | 27.1 |
Cylindrospermopsin standard (95% purity) was purchased from Alexis Corporation (Lausen, Switzerland). CYN standard solutions were prepared in Milli-Q water (100 µg mL−1) and further diluted for their use as working solutions (5–500 µg L−1). All reagents and chemicals employed in this work were of analytical grade. HPLC-grade methanol, acetonitrile, acetic acid, dichloromethane, and trifluoroacetic acid (TFA) were supplied by Merck (Darmstadt, Germany). Deionized water (418 MΩ cm−1 resistivity) was acquired from a Milli-Q water purification system (Millipore, Bedford, MA, USA). BOND ELUT® Carbon cartridges (PGC, 500 mg, 6 mL) were obtained from Agilent Technologies (Amstelveen, The Netherlands, Europe). Reagents for UHPLC–MS/MS were of LC–MS grade; formic acid was supplied by Fluka (Steinheim, Germany) and water and acetonitrile by VWR International (Fontenay-sous-Bois, France). The lettuce samples without CYN were purchased from a local supermarket. In this way, we ensured that they had passed the quality controls required for their human consumption. | other | 35.12 |
Preliminary tests were performed to evaluate the influence of the lyophilization process on CYN concentration. For this, 1 mL of a CYN solution (200 μg L−1) was added to 1 g of lettuce (fresh weight, f.w.) and the sample was lyophilized (−80 °C). In parallel, 1 mL of the same CYN solution (200 μg L−1) was added to 0.05 g of lyophilized lettuce (dry weight, d.w.), corresponding to 1 g of f.w. These procedures were performed in triplicate. Subsequently, CYN was extracted from all the samples to evaluate the influence of these processes. Since no significant differences were observed between the conditions (data not shown), we decided to work with fresh-weight lettuce, once the plants had been exposed to the toxin, as they reach the final consumer. | other | 35.03 |
To develop the extraction and purification procedures, a modified version of the method from Kittler et al. (2012) was performed, adding a purification stage with graphitized carbon cartridges. Briefly, control fresh lettuce leaves (1.05 ± 0.14 g f.w.) were fortified with CYN standard solutions to obtain three concentration levels: 20, 200, and 500 ng CYN g−1 f.w. Afterwards, the toxin was extracted with 6 mL of 10% acetic acid; after ultraturrax homogenization for 30 s, the sample was sonicated (15 min) and stirred (15 min). The resulting mixture was centrifuged at 12,000 rpm for 15 min. Once the extract was obtained, a purification step was applied. For this purpose, BOND ELUT® Carbon cartridges were activated with 10 mL of a solvent mixture of DCM/MeOH (10/90) acidified with 5% formic acid and rinsed with 10 mL of Milli-Q water. Then, the sample was loaded, the column was washed with 10 mL of Milli-Q water, and the analyte was eluted with 10 mL DCM/MeOH (10/90) acidified with 5% formic acid. To concentrate the sample, the extract was evaporated in a rotary evaporator and resuspended in 1 mL Milli-Q water, prior to its UPLC–MS/MS analysis. | other | 34.78 |
The chromatographic separation was carried out with a UPLC Acquity (Waters, Milford, MA, USA) coupled to Xevo TQS-micro (Waters, Milford, MA, USA) consisting of a triple quadrupole mass spectrometer equipped with an electrospray ion source operated in positive mode. UPLC analyses were performed on a 50 × 2.1 mm Acquity BEH C18 1.7 µm column, at a flow rate of 0.45 mL min−1. For chromatographic separation, a binary gradient was used with (A) water and (B) methanol as mobile phases, containing, both of them, 0.1% formic acid (v/v). The injection volume was 5 µL. The profile for elution was: 0% B (0.8 min), linear gradient to 90% B (2.2 min), 90% B (1 min), and finally 100% B (1 min). Multiple Reaction Monitoring (MRM) was applied, by which the parent ions and fragments ions were monitored at Q1 and Q3, respectively. The transitions for the analyte CYN are 416.2/194.0 and 416.2/176.0. The transition 416.2/194.0 (quan) was chosen for quantitation of CYN, and the transition 416.2/176.0 (target) as confirmatory. The ion ratio were measured in the CYN standard solution and in the lettuce samples as target area/quan area × 100, giving a mean ion ratio of 38.8% with SD of 1.2% for the CYN standard solution (n = 10), and a mean ion ratio of 39.6% with SD of 5.1% for CYN in the lettuce samples (n = 5) exposed to the toxin. For UPLC–ESI–MS/MS analyses, the mass spectrometer was set to the following optimised tune parameters: Capillary voltage: 3.0 kV, Source Temperature: 500 °C, and source Gas flow: 1000 L h−1. | other | 30.9 |
The lettuce plants (Lactuca sativa) were purchased from a local market (Porto, Portugal) as sprouts. The roots were washed with deionized water to remove all remaining soil and prepare the plants for hydroponic cultivation. The plants were introduced into opaque glass jars, and the roots were completely immersed in culture medium at pH 6.5, as described by Freitas et al. (2015) . The plants were acclimated for one week with white fluorescent light (light–dark period of 14–10 h), at a controlled temperature of 21 ± 1 °C until the start of the experiment. | other | 38.47 |
The Chrysosporum ovalisporum (LEGE X-001) cyanobacterial CYN-producing strain (CYN+) was isolated from Lake Kinneret, Israel and was grown for biomass production in the Interdisciplinary Centre of Marine and Environmental Research, CIIMAR (Porto, Portugal) as described in Guzmán-Guillén et al. (2014) . The extraction of CYN from the culture was performed according to Guzmán-Guillén et al. (2012) with modifications described in Guzmán-Guillén et al. (2017) . CYN retention time was 9.55 min, and the concentration obtained was 2.14 μg CYN mg−1 of lyophilized cells. | other | 35.97 |
After acclimation, five specimens of lettuce were exposed to a solution of CYN extracted from the culture at a concentration of 10 µg L−1. In order to do this, the culture medium was changed three times a week for 21 days, adding to the medium the volume of C. ovalisporum culture necessary to reach the exposure level (10 μg L−1). Moreover, five lettuce plants were included in the experiment as a control group without exposure to the toxin. When the experiment was over, the plants were washed with distilled water, frozen, and lyophilized (Telstar Lyoquest) for CYN extraction, following the validated method proposed in the present study. | other | 37.78 |
The proposed method was validated taking into account the Eurachem (2016), the AOAC (2016), and González and Herrador (2007) guides for linearity, sensitivity, precision, recovery, and robustness. Three CYN validation standards were employed, performing the measures in triplicate each day for three consecutive days, and these concentrations tried to cover the optimal working range. | other | 34.28 |
One mL of solutions at three different CYN concentrations (20, 200, and 500 µg CYN L−1) was added to the plant matrix to obtain 20, 200, and 500 ng g−1 f.w. Precision and recovery were obtained by applying a one-factor analysis of variance (ANOVA), as explained in the Results and Discussion section, and then they were compared with tabulated reference values. Moreover, a robustness (ruggedness) study was performed by spiking the matrices with an intermediate validation standard of 200 µg CYN L−1 (equivalent to 200 ng g−1 f.w.), according to Youden's procedure (1967) . Small, deliberate variations in the following parameters were tested with Student’s t test in order to evaluate the ruggedness of the method: (1) time for the sample to pass through the cartridge; (2) water volume for washing the cartridge; and (3) volume of DCM/MeOH employed for CYN elution. | other | 35.16 |
Intracranial tumors (ICTs) are either signs of advanced distant malignant processes or local processes that can be either benign, malignant, or reactive in nature since non-tumor volume-occupying lesions (NVOL) can be viewed as distinct types of specific intracranial processes . These lesions present an immediate threat to patients' lives and need to be diagnosed and treated as soon as possible, regardless of their origin . Irrespective of the differences in the distinct tumor groups and the involvement of many medical specialists, these tumors are hard to diagnose and treat due to their state of advancement, biological properties, and the specifics of the blood-brain barrier . | other | 30.16 |
Despite variances in the pathogenesis, evolution of ICT groups, and even within a clinical ICT entity, the primary symptom and reason for the diagnosis of an ICT is almost always the onset of neurological deficits . The first sign of a distant malignant process can rarely be a neurological deficit caused by its intracranial spread, while, in most other cases, these metastases are the first sign of advancement of a condition and are difficult or resistant to treatment . In the case of local tumors, the neurological deficit is virtually always the first sign of the disease, and its rate, measured via the Karnofski neurological status score, can be used as a predictive factor . Primary ICTs seldom have a prior generalized manifestation, such as the case with some astrocytomas that can cause epileptic seizures and pituitary adenomas, which can cause a variety of conditions, such as gigantism, acromegaly, Cushing’s disease, or generalized panhyperpituitarism . | study | 31.47 |
Subsets and Splits