arg_1
stringlengths 4
5.08k
| round_1
float64 2
8
⌀ | ann_1
float64 1
2
⌀ | arg_2
stringlengths 8
2.19k
| round_2
float64 1
7
⌀ | ann_2
float64 1
2
⌀ | annotation_name
stringclasses 131
values | is_attacks
int64 0
1
|
|---|---|---|---|---|---|---|---|
When it comes to testing efficacy, we would like to use a treatment duration as long as possible to maximize the effect. However, this duration varies between in vitro and in vivo work. For cell cultures, the control samples continue to grow throughout the treatment duration, and so after 72 h the plates are very much confluent, and may not be left to continue and grow any further (without drastic changes in the environmental conditions, making them inadequate control cells). It is also noteworthy that such treatment periods for in vitro work are well accepted in the field of TTFields research. In vivo, prior studies also used similar time frames of 2 to 3 weeks from inoculation when working with the N1S1 model (Buijs et al., 2012 -; Ju et al., 2009; Thompson et al., 2012). Extending treatment duration was also limited by the physical status of the animals, as was explained in response 3.
| 2 | 1 |
Why the in vitro experiment was performed for 72 hours and in in vivo for 120 hours?
| 1 | 2 |
cancers14122959_makarova
| 1 |
Chloroquine was added to the cell cultures only at the final hours of the treatment for answering questions related to the mechanism of action, and not for boosting the efficacy of the other treatments. When using chloroquine in animal studies, it is for efficacy purposes, and so it is used throughout the treatment period. Since our animal experiment aimed to examine the efficacy of concomitant TTFields and sorafenib, with no additional agents, chloroquine was not employed.
| 2 | 1 |
Any explanation for why not using cloroquine in vivo to integrate better the in vitro data.
| 1 | 2 |
cancers14122959_makarova
| 1 |
For the cytotoxicity assay we remove the supernatant, washed the cells, and then collect the adherent cells following trypsinization and visual inspection to verify all cells were removed. Indeed, as the reviewer mentioned correctly, we “counted the live cells and plotted the final data as percentage of untreated controls”. We agree that this is not cytotoxicity per se, and that is why we also perform 7-AAD/annexin-V staining of the cells. For this apoptosis assay we do collect the supernatant together with the adherent cells. Increased apoptosis and/or necrosis indicates that reduction in the cell number observed in the cell count emanates at least in part from cytotoxicity. We agree with the reviewer that we should not confuse the cell count measurements with the term “cytotoxicity” prior to showing the effect on apoptosis, and therefore we have changed this terminology throughout the paper.
| 2 | 1 |
The authors claim that cytotoxicity was measured “by cell counting using iCyt EC800 (Sony Biotechnology) 123 flow cytometer, and expressed as a percentage relative to the control.” Does this imply that they counted the live cells and plotted the final data as percentage of untreated controls (as figure 1 suggests). However, is this a real cytotoxicity or a cell growth inhibition? Did they measure the adherent cells after trypsinization. Were the cells from supernatant counted (where are probably the majority of dead cells)? There is a big difference between a therapy which kills vs a therapy which induces a cellular arrest.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thank the reviewer for these questions. Preliminary tests using CD31 staining revealed no differences between the groups regarding blood vessel density and therefore we did not pursue the research in that direction. We agree that measuring the effect of TTFields on the anti-angiogenic effect of sorafenib and on resistance to sorafenib are interesting, these topics were however not within the scope of the current study, and remain for future investigations. Regarding possible discrepancy between the volume fold change and tumor weight, please see below a graph showing tumor volume (as measured by MRI) versus tumor weight, both measured at the end of the study. The graph shows very good correlation between the two parameters, indicating the reliability of the measurements. If the reviewer feels there is discrepancy, it may be due to the volume shown as fold change relative to the initial tumor volume and not as the end value.
| 2 | 1 |
Since sorafenib acts also on angiogenesis, did the authors investigate if TTFields may interfere with anti-angiogenic effect sorafenib-mediated? Also, is there any evidence that TTF may prevent the pretty common resistance to sorafenib observed in clinic? Moreover, there is a discrepancy between the fold changes in tumor weight vs. volume in the combination group vs. untreated group. Did the authors check changes in blood vessels density. Were the mice perfused before collecting the tumors?
| 1 | 2 |
cancers14122959_makarova
| 1 |
We routinely examine our cells for Mycoplasma. Regarding authenticity, the cells were used shortly after purchase, and so there was no need to examine this.
| 2 | 1 |
Was Mycoplasma testing done routinely? Were the cells checked also for authenticity?
| 1 | 2 |
cancers14122959_makarova
| 1 |
We appreciate this question. Autophagy is a process that is elevated in order to cope with cellular stress, but as stress level elevate, autophagy can no longer provide the protection the cell needs, and the cell will undergo apoptosis. This kind of kinetics means that the levels of autophagy markers will depend on the time point the cells are examined. This may be appreciated from figure 2 panel c and d, with the different kinetics displayed by the two cell lines. While for Huh-7D12 autophagy levels increase from 24 to 48 hours of treatment, in the HepG2 cells autophagy levels at 48 hours are lower than at 24 hours, indicating these cells are already after the autophagy peak and on the way to apoptosis. Indeed, figure 1d shows higher levels of apoptosis for HepG2 cells. To clarify we have added a few sentences to the text. In results sub-section 3.2: “However, autophagy kinetics seems to be faster in the HepG2 cells, in which LC3 markers are lower at 48 versus 24 hours, whereas elevation is seen from 24 to 48 hours for the Huh-7D12 cells.” In the discussion “While autophagy serves as a survival strategy of cells, when stress levels continue raising it may be over-activated and mediate cell death [9]. The faster autophagy kinetics seen for the HepG2 relative to Huh-7D12 cells following application of TTFields is in agreement with the higher apoptosis levels displayed by this cell line, and may serve as an additional rational for the higher efficacy of TTFields against it. Examination of the reasons for faster autophagy in HepG2 relative to Huh-7D12 cells is out of the scope of this work.” For more clarity we have also added a more in-depth kinetic study, including additional relevant markers and additional time points, examinations that were performed for the combined treatment as compared to TTFields and sorafenib alone. We thank the reviewer for this comment, as these additions add much clarity to the mechanism of action of TTFields in combination with sorafenib and provide a more coherent explanation for the in vivo results. These additions may be seen in Figure 3 and are described in results sub-section 3.4, Autophagy-apoptosis Interplay For Treatment with Concomitant TTFields and Sorafenib: “In order to investigate the mechanism of action of TTFields-sorafenib co-application, HepG2 and Huh-7D12 cells were treated for 6, 24, or 48 hours with TTFields, sorafenib (3µM), or the two modalities together, and then examined for expression levels of various proteins. For HepG2 cells, the autophagy marker beclin-1 demonstrated elevation after 6 hours of treatment, which was later replaced with diminished expression levels (Figure 3d). This type of behavior was seen in all treatment groups, but was most pronounced for TTFields-sorafenib co-application. The autophagy marker LC3 also displayed such bi-phasic characteristics, but with a somewhat slower kinetics, showing some elevation at 6 hours of treatment, but higher elevation at the 24 hours time point (Figure 3d). As in the case of beclin-1, the magnitude of the effect was higher for co-treatment of TTFields and sorafenib relative to the monotherapies. GRP78, a marker of ER stress, remained low in all treatment groups for 6 and 24 hours of treatment, but demonstrated elevated levels at the later, 48-hours time point (Figure 3e). The apoptosis marker cleaved PARP displayed increased expression in the combined group already after 24 hours, elevating even further after 48 hours of treatment. For the monotherapies, cleaved PARP increase was only evident at 48 hours of treatment, and to a lower extent than that in the co-treatment group (Figure 3f). The slower kinetics of the autophagy-apoptosis path in the Huh-7D12 cells, as seen from the elevation of LC3 after as much as 48 hours (Figure 2c and d), prevented from detecting such changes in the levels of these markers in this cell line (Figure S1).” And in the discussion part “Kinetic examination in the HepG2 cells revealed elevation in autophagy levels as early as 6 hours of TTFields or sorafenib treatment, which diminished and were replaced with ER stress and apoptosis for 48 hours of treatment. These results are in line with a previous study that focused on the effects of sorafenib on such markers in HepG2 cells [32]. The higher changes in expression levels and faster kinetics when TTFields and sorafenib were applied together rather than alone indicate higher stress levels imposed on the cells in the former case” Figure 3.
| 2 | 1 |
Is there any explanation why the combination of TTFields and sorafenib did not induce a significant level of autophagy as compared to untreated animals which invalid the initial hypothesis that “concomitant application of sorafenib and TTFields may increase stress levels enough to tilt autophagy towards the cell death pathway”.
| 1 | 2 |
cancers14122959_makarova
| 1 |
The sentence referred to was meant to describe only the in vivo outcomes. We have rephrased it to be more accurate and clear: “While each treatment alone elevated levels of autophagy relative to control, TTFields concomitant with sorafenib induced a significant increase versus control in tumor ER stress and apoptosis levels, demonstrating increased stress under the multimodal treatment.” Point 12: Finally, adding to all the above questions, I found a very weak Discussion section which must be extended.
| 2 | 1 |
The statement “TTFields concomitant with sorafenib induced a significant increase in apoptosis’ in the abstract section is overstated. When compared with sorafenib alone there is practically no difference. Moreover, TTFields failed to increases apoptosis when added to sorafenib and compared to sorafenib alone in one out of two human cells line investigated
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thanks the reviewer for this comment. We have now elaborated on many issues throughout the discussion, as was described thorough this response letter. Regarding the conclusion, we have rephrased it for better accuracy: “TTFields were identified to be most efficient for treatment of HCC cells at 150 kHz, and this frequency further demonstrated in vivo efficacy.” Why only one frequency was used in vivo, and the difference between frequency and dose, were explained in response 4.
| 2 | 1 |
Finally, adding to all the above questions, I found a very weak Discussion section which must be extended. Moreover, in the Discussion section, the authors concluded that “TTFields display efficacy for treatment of HCC in vitro and in vivo, with an optimal frequency of 150 kHz”. This is not a correct statement. While for in vitro data, the authors have data, for in vivo they used only one frequency of 150 Hz. At least one different dose should have been studied for comparison since this is a completely different tumor environment than the in vitro one.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thank the reviewer for this question. Indeed, the sorafenib dose used in this study proved to be slightly more efficacious than TTFields in controlling tumor fold increase. Nevertheless, these differences between the monotherapies, did not reach statistical significance. In accordance, there was no statistical difference between the monotherapies in the expression levels of the LC3 marker and the levels of cleaved PARP. In order to better understand the mechanism of action, we have added experiments to better characterize the autophagy-apoptosis interplay for treatment with concomitant TTFields and sorafenib, which are described in response 3. Of note, the clinical development of TTFields does not aim to replace sorafenib with TTFields, but rather to add TTFields on top of sorafenib and therefore this work focused mainly on the potential added value in combining these 2 modalities.
| 2 | 1 |
1) The authors demonstrate the efficacy of TTFields in vivo even when used as monotherapy. As shown in the Figure 4 C and D, TTFields were found less effective in terms of reducing the tumors volume and weight when compared with sorafenib. However, no differences in expression of LC3 marker were observed between these groups (treated with TTFields or with sorafenib)(as shown in Figure 4D). Similarly, low evidence of apoptosis (expression of cleaved PARP) was found in these groups, as shown in Figure 4F. What is the mechanism illustrating higher efficiency of sorafenib against HCC?
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thank the reviewer for pointing out this issue. Quantification of the IHC images was done automatically. The whole slide was scanned, and the CaseViewer software was used to exclude non-tumor areas. The signals of the stained protein and the nuclei were resolved by color deconvolution and quantified separately using the FIJI software (ImageJ) software. Average signal per cell or percent of positive cells was calculated. As the reviewer pointed out, the high magnification images we chose to show do not correctly reflect the quantification performed by the software, and we have now replaced them with better representative fields of the slides.
| 2 | 1 |
2) Despite the expression of cleaved PARP was very low in the tumors treated with TTFields or sorafenib alone ( as shown in IHC-images in Figure 4F), the authors declare about ~ 20% of positive cells, as show in the graphs below IHC-staining. Similar, the graphs illustrating the LC3 expression are not in a proper fit with the images shown in Figure 4D.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We sincerely appreciate this well-taken comment. To better explain the correlation between autophagy and apoptosis we have performed additional examinations to include more markers and at additional time points. These examinations were performed for the combined treatment as compared to TTFields and sorafenib alone. We thank the reviewer for this comment, as these additions add much clarity to the mechanism of action of TTFields in combination with sorafenib and provide a more coherent explanation for the in vivo results. These additions may be seen in Figure 3 and are described in results sub-section 3.4, Autophagy-apoptosis Interplay For Treatment with Concomitant TTFields and Sorafenib: “In order to investigate the mechanism of action of TTFields-sorafenib co-application, HepG2 and Huh-7D12 cells were treated for 6, 24, or 48 hours with TTFields, sorafenib (3µM), or the two modalities together, and then examined for expression levels of various proteins. For HepG2 cells, the autophagy marker beclin-1 demonstrated elevation after 6 hours of treatment, which was later replaced with diminished expression levels (Figure 3d). This type of behavior was seen in all treatment groups, but was most pronounced for TTFields-sorafenib co-application. The autophagy marker LC3 also displayed such bi-phasic characteristics, but with a somewhat slower kinetics, showing some elevation at 6 hours of treatment, but higher elevation at the 24 hours time point (Figure 3d). As in the case of beclin-1, the magnitude of the effect was higher for co-treatment of TTFields and sorafenib relative to the monotherapies. GRP78, a marker of ER stress, remained low in all treatment groups for 6 and 24 hours of treatment, but demonstrated elevated levels at the later, 48-hours time point (Figure 3e). The apoptosis marker cleaved PARP displayed increased expression in the combined group already after 24 hours, elevating even further after 48 hours of treatment. For the monotherapies, cleaved PARP increase was only evident at 48 hours of treatment, and to a lower extent than that in the co-treatment group (Figure 3f). The slower kinetics of the autophagy-apoptosis path in the Huh-7D12 cells, as seen from the elevation of LC3 after as much as 48 hours (Figure 2c and d), prevented from detecting such changes in the levels of these markers in this cell line (Figure S1).” And in the discussion part “Kinetic examination in the HepG2 cells revealed elevation in autophagy levels as early as 6 hours of TTFields or sorafenib treatment, which diminished and were replaced with ER stress and apoptosis for 48 hours of treatment. These results are in line with a previous study that focused on the effects of sorafenib on such markers in HepG2 cells [32]. The higher changes in expression levels and faster kinetics when TTFields and sorafenib were applied together rather than alone indicate higher stress levels imposed on the cells in the former case.” Figure 3.
| 2 | 1 |
3) It will be much better to provide the data to explain the mechanisms illustrating why the monotherapy of TTFields or sorafenin induced autophagy, whereas the tumors treated with combination developed the substantial apoptotic death of tumor cells.
| 1 | 2 |
cancers14122959_makarova
| 1 |
Both cell lines experience elevation of apoptosis following application of sorafenib in a dose dependent manner, as evident from the AnnV/7AAD results. However, while TTFields greatly elevate apoptosis in HepG2 cells, they have a low effect on apoptosis levels in the Huh-7D12 cells, seen both in Figure 1d and in Figure 3c. As was explained in the discussion, this difference between the cell lines may be attributed to the different p53 status, wild type in HepG2 and mutated in Huh-7D12, as there are previous indications of lower TTFields-induced apoptosis in cell lines with mutated p53. As suggested by the reviewer, in order to back up the AnnV/7AAD results we added WB for cleaved PARP, as described in response 3.
| 2 | 1 |
4) Since Annexin V/7-ADD data was not convincing and the authors observed the minor increase of apoptotic cells after HCC cells were treated with combination of TTFields and sorafenib ( when compared to the cells treated with TTFiealds and sorafenib alone), I suggest to run the WBs to examine the expression of the cleaved forms of PARP and caspase-3 ( for both HCC cell lines). This might be helpful and make the in vitro data more relavant with the data shown in vivo.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We apologize for accidentally leaving out this figure, and have now added it.
| 2 | 1 |
1) Figure 4A is missing.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thank the reviewer for this comment. In the in vitro experiments we used cell lines derived from humans. However, these cell lines cannot be implanted to rats, and so for the in vivo experiments we had to use a cell line from rats.
| 2 | 1 |
2) the different HCC cell lines were used for in vitro and in vivo experiments, therfore making difficult to compare these data.
| 1 | 2 |
cancers14122959_makarova
| 1 |
We thank the reviewer for this important question and would be happy to clarify. TTFields are alternating electric fields in the range of 100 to 500 kHz, with maximal efficacy seen at a different frequency for different tumor types. Since TTFields cannot be applied at several different frequencies simultaneously, one specific frequency needs to be selected, and so the first step in examination of a new tumor type is a frequency scan. The effect of TTFields is not dichotomic, such that only one frequency is effective while the other are not at all. From the frequencies showing efficacy the purpose is to select the most effective one. It should be clarified that higher frequencies of the alternating electric fields do not mean higher energy, and hence there is no disadvantage in working with higher frequencies and no added toxicity or increased side effects for higher frequencies within the TTFields frequency range. Of note, TTFields at 200 kHz is already approved and has been applied to more than 18,000 patients with glioblastoma, with skin irritation being the main treatment related adverse event. TTFields at 150 kHz is approved for treatment of unresectable malignant pleural mesothelioma. To address these issues, we have rephrased the discussion: “It has been previously shown that maximal effectivity of TTFields occurs at a different frequency for different cancer types, owing to the specific electrical properties of the cells [15, 28]. Hence, the first step in applying TTFields to a new tumor type includes frequency scans.” and “The lower TTFields intensity required for treatment of HepG2 relative to Huh-7D12 cells for achieving the same level of efficacy suggest higher sensitivity of the former to TTFields.” Point 2: What is the ‘n’ number of the samples and experiments in each group?
| 2 | 1 |
What is the rationale of using 150 kHz TTFields? TTFields at 100 kHz has also shown significant differences as cytotoxicity at p<0.01 and p<0.001 in HepG2 and Huh-7D12 cells which have around 60% and 50% of cell survival. Why did the authors use high frequency of TTFields though the cytotoxicity was also observed in lower dose (Fig 1A).
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for pointing this was not clear. As mentioned in the statistical analysis sub-section within the methods part, all in vitro experiments were repeated at least 3 times, and so depicted We have now added this also to the legends of figures 1, 2, and 3. Per the in vivo study, as was mentioned in the methods (sub-section 2.8), 52 animals were included in the study. The specific numbers in each treatment group are now mentioned in the results (sub-section 3.4): “During the treatment period, average tumor volumes of control animals (n = 11) increased 5.9-fold (Figure 4b and 4c; and Figure S3 for tumor images). For animals treated with TTFields (n = 15) or sorafenib (n = 10), tumor growth was significantly lower, 3.3- and 2.3-fold, respectively. In the TTFields-sorafenib combination group (n = 16), a 1.6-fold increase in tumor volume within the treatment period was observed, a growth significantly lower than that for control or for each treatment alone.” It is also mentioned in the legend of figure 4: “Rats (n = 52) were inoculated orthotopically with rat HCC N1S1 cells, and treated with sham-vehicle (n = 11), TTFields alone (n = 15), sorafenib alone (n = 10), or TTFields + sorafenib (n = 16), according to the depicted timeline (a).” Point 3: LC3 is increased in all the groups except in control tissue, however, cell death was increased only in combined group.
| 2 | 1 |
What is the ‘n’ number of the samples and experiments in each group? Also mention the ‘n’ number in all the experiments involved for invitro and invivo.
| 1 | 2 |
cancers14122959_perova
| 1 |
First, we would like to emphasize that indications of autophagy were diverse and determined from increased cellular granularity, amplified lipidation of LC3 (from LC3-I to LC3-II) detected by Western blot, and elevated levels of LC3 foci observed by fluorescent staining. Nevertheless, to better explain the correlation between autophagy and apoptosis we have performed additional examinations to include more markers and additional time points. As per the reviewer’s request, Beclin-1 levels were increased more than 4-fold relative to control in all treatment groups, while intensity of LC3 staining was increased about 3-fold relative to control in the individual TTFields and sorafenib groups, but elevated only 2-fold in the combination group (Figure 4e).
| 2 | 1 |
LC3 is increased in all the groups except in control tissue, however, cell death was increased only in combined group. How can the authors correlate autophagy with apoptosis? It is not so trustworthy that the only expression of LC3B indicate the treated condition have increased autophagy in the tumor tissue. Other important autophagy and degradation markers like Beclin1 and P62 need to be shown to reflect the regulatory mechanism of TTFields, as well as for the combined treatment with Sorafenib.
| 1 | 2 |
cancers14122959_perova
| 1 |
Per the reviewer’s request we show below images of the tumors within the liver, which we have also provided now in the supporting material of the paper. As may be seen, the liver itself is much larger than the tumor, and so we did not see value in measuring total liver weight, and hence cannot provide these values. While we tried to reduce variability between the animals by excluding those that at treatment initiation had tumor volumes not within the range 30–100 mm3, there was still divergence between the animals that resulted in the standard deviation bars depicted in the figure. It is however important to mention that there was a very good correlation between tumor volume (as measured by MRI) versus tumor weight at the end of the study, as may be seen in the graph below. Figure S3. Images of tumors within the livers of the rats treated with sham heat (control rats), TTFields, sorafenib, or TTFields plus sorafenib.
| 2 | 1 |
What was the total weight of livers in all groups? Please update the images of whole liver showing tumors on it, or at least pictures of tumors which were removed after sacrifice, if applicable.
| 1 | 2 |
cancers14122959_perova
| 1 |
IHC tests the presence of a protein already expressed in the tissue, and 30 min are sufficient to allow recognition and binding of the antibody to the target protein. The secondary antibody is an HRP conjugated goat anti rabbit, that is a part of the Leica HRP-refine detection kit (Cat # DS9800). We have added the missing antibody information to section 2.10. We unfortunately do not have any tissue left to allow for performing WB analysis from the in vivo study, but we have conducted additional cell line studies and added WB analysis of cleaved PARP for the in vitro examinations and additional IHC examinations for autophagy and ER stress markers (as shown in response 3).
| 2 | 1 |
In the IHC experiment for PARP, the authors have incubated with primary antibody only for 30 min? Is that timing enough to get protein expression? What source of secondary antibody was used? In addition to IHC, I suggest performing western blot using PARP antibody where the full length and cleaved bands are observed in the same blot.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for identifying this mistake, and have corrected this, changing the label of the LC3 data to e.
| 2 | 1 |
The figure number is mislabeled in Fig 4D-LC3 Immunofluorescence.
| 1 | 2 |
cancers14122959_perova
| 1 |
Error bars are not missing for the control groups, rather they are null, as the values displayed in these figures are relative to control, and so per definition all control experiments have the exact same value of 100 or 1 (for percentage or fold change, respectively).
| 2 | 1 |
Why there is no error bar in the control group of all bar graphs? Please include error bars and re-calculate the statistical analysis for all the data wherever missing.
| 1 | 2 |
cancers14122959_perova
| 1 |
We apologize for this mistake, and have now corrected the figure legend. The quantification of immunofluorescence of LC3 foci formation is c, and immunoblotting showing LC3-II to LC3-I ratio is d. Point 9: There is no data in Fig 4A.
| 2 | 1 |
Figure legends of Fig2D is missing. Fig C is repeated in the legend. Please label the figures appropriately. It is so frustrating to understand.
| 1 | 2 |
cancers14122959_perova
| 1 |
We apologize for accidently omitting Fig 4A. This figure depicts the timeline of the in vivo experiment for easier understanding of the study designed. We have now corrected this.
| 2 | 1 |
There is no data in Fig 4A. Whole data is missing but have explained in the result section and in figure legends.??
| 1 | 2 |
cancers14122959_perova
| 1 |
We understand from the reviewer’s comments that the sentence “The higher effects seen in the presence of CQ reveal that the observed phenomenon is due to increased autophagic flux, rather than decreased autolysosome degradation” is not clear enough. We have hence rephrased the sentence to be more accurate and better deliver the message: “CQ is an inhibitor of lysosome degradation, commonly used to decipher whether the elevation of LC3 is due to upregulation of the autophagy process or reduced autophagosome turnover [30]. The higher TTFields-induced elevation of LC3 seen in the presence of CQ suggest that the observed phenomenon is due to increased autophagic flux, rather than decreased autolysosome degradation.” Point 11: Typos: Line 111, ‘invitro’ spelling is not correct.
| 2 | 1 |
Line 337 seems over statement since the data are not shown in the manuscript.
| 1 | 2 |
cancers14122959_perova
| 1 |
The “inovitro” in line 111 is not a typo, it is the name of the system used for applying TTFields in vitro. In line 342 we removed the duplicate “and”, and thank the reviewer for catching this typo.
| 2 | 1 |
Typos: Line 111, ‘invitro’ spelling is not correct. Double ‘and’ in Line 342. Need language and grammar check.
| 1 | 2 |
cancers14122959_perova
| 1 |
For the in vitro experiments with TTFields (sub-section 2.3) we tried not to elaborate too much and referenced previous work describing all details. We understand that we have cut off too many details, and are happy to add them “HepG2 and Huh-7D12 cell suspensions (500 µl, 25 x 103 cells/plate) were placed as a drop in the center of 35-mm inovitro™ dishes composed of high dielectric constant ceramic (lead magnesium niobate–lead titanate [PMN-PT]), with two perpendicularly pairs of transducer arrays printed on their outer walls. Cells were incubated overnight at 37 °C to allow attachment to the dish, and then 2 ml of fresh media were added.” Regarding the number of rats included in the final analysis, we apologize this information was not clear. The 52 rats mentioned in the text were in fact the actual final number of animals in the analysis, since “all rats reached the required usage limit of ≥18 h/day”, as is now explicitly stated in the results sub-section 3.4.” Point 2: What is puzzling in this investigation is showing functional data using two human cell lines and in vivo data using rat cells.
| 2 | 1 |
In vitro experiments sub-section in Methods sections lacks many experimental details (e.g., type of plate/flask, plating overnight or not before experiment, number of plated cells). A very important missing is not showing the actual number of rats included in the final analyses (the ones who successfully received therapy for more than 18 hours/day). This must be added.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for raising this important issue. Using a xerograph model is indeed a good suggestion, however it would require working in immunosuppressed mice. It is currently not feasible for us to perform studies in which we apply TTFields to immunosuppressed animals, as these animals are of smaller weight thus the burden of wearing the electrodes may induce too much stress. Furthermore, these animals are more prone to contract infections during the process of electrodes placement and replacement on the animals. On the other hand, using the rat cells for in vitro experiments in the inovitro dishes turned out to be very technically challenging, as the cells were non-adherent, and so we could not pursue this important avenue. Nevertheless, we agree with the reviewer that there is a possibility that the optimal frequency may differ in the N1S1 relative to the human cell lines. Therefore, we have addressed this issue by including results of an additional frequency scan experiments done with N1S1 cells. We now mention this in the text, results sub-section 3.4: “The efficacy of combining TTFields with sorafenib relative to each modality alone was examined in the N1S1 HCC rat orthotopic model (timeline in Figure 4a). In vitro experiments confirmed that the 150 kHz TTFields frequency found optimal for treatment of the human cell lines was also optimal for treatment of the murine N1S1 cells used for the in vivo study (Figure S2).” We provide the relevant frequency scan figure in the supplementary material. Per the good question by the reviewer regarding the p53 status of the N1S1 cells we added: “It is also worth mentioning that the N1S1 murine cells used for this study, like the HepG2 cells, are p53 wild type.” Figure S2. TTFields frequency scan in rat N1S1 HCC cells. N1S1 cells were treated with TTFields (1.7 V/cm RMS) across a frequency range of 100–400 kHz, and cell count were determined following 24 hours of treatment. *p < 0.05 relative to control; one-way ANOVA. ANOVA = analysis of variance; HCC = hepatocellular carcinoma; RMS = root mean square; SEM = standard error of the mean; TTFields = Tumor Treating Fields.
| 2 | 1 |
What is puzzling in this investigation is showing functional data using two human cell lines and in vivo data using rat cells. While I do not know if the authors have the technology to perform TTFileds in mice, where for sure they should have done xenograft models with the two human cell lines, why the rat cell line was not studied in vitro using the same experimental strategies as for human cells. This must be done and included for a better understanding of TTFieldds activity from in vitro to in vivo data. What are the p53 status and the apoptosis signaling pathway function in N1S1 cells? Maybe the cytotoxicity data will reveal a different better frequency.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for this comment, and apologize that the illustration of the timeline was accidentally missing from the submission (we have now added it). The limiting factor for study duration was the well being of the animals. The tumors in the control group were very large, causing stress and weight loss of the animals. The arrays placed on the animals, together with the individual housing needed to prevent wire entanglement, adds even more stress and increases animal weight loss. Overall, it was non-ethical to continue the study further. We have now added this explanation to the discussion section to clarify this limitation of the study: “In the HCC animal model, the acute effects of TTFields and sorafenib were examined. Due to the large tumors developed in the control group and the stress experienced by the animals as a result of the individual housing and motility limitations imposed by the sham and TTFields arrays, longer treatment durations were not feasible.” Point 4: There is no explanation why the most effective dose was 150 kHz and higher does actually decreases the killing.
| 2 | 1 |
Another puzzling experiment is the schedule for the in vivo work. Since the authors missed to add Figure 4A for timeline, based on Methods section the rats were treated for 5 days with TTFields and or sorafenib and a day later the rays were sacrificed. While a short “acute” follow-up is welcome the most important experiment should allow the follow-up for much more days to indeed observe the effect of TTFields added to sorafenib. I could not find an explanation for not including a long term follow-up. In my opinion, this is a key therapeutically experiment which must be done and included in the study.
| 1 | 2 |
cancers14122959_perova
| 1 |
It is important to understand that the TTFields frequency is not the treatment dose, and hence increased frequency does not mean increased efficacy. In the case of TTFields, intensity, duration, and usage are the factors that contribute to the “treatment dose”. Nevertheless, we have added to the discussion an explanation on this issue of optimal frequency: “It has been previously shown that maximal effectivity of TTFields occurs at a different frequency for different cancer types, owing to the specific electrical properties of the cells [15, 28]. Hence, the first step in applying TTFields to a new tumor type includes frequency scans.” Regarding the option that the optimal frequency will differ between in vitro and in vivo setting, this is indeed a concern. However, as shown by the in vitro frequency scans, the effect of TTFields is not dichotomic, and effectiveness can be seen at more than one frequency. Therefore, the relatively low expected differences between animals treated with different TTFields frequencies together with the variability inherent to animal studies, makes it technically problematic to perform frequency scans in vivo. As of today, in vitro frequency scans are the common practice for determining the optimal TTFields frequency for delivery to animals and to humans. In fact, in vitro frequency scans were the basis for the clinical studies leading to FDA approval of TTFields at 200 kHz for the treatment of GBM and at 150 kHz for the treatment of MPM. We have added this issue to the discussion as a limitation of the study: “Since the optimal TTFields frequency has been shown to be dependent on the electrical properties of the cells and it is not clear how much effect the tumor microenvironment has on these properties, and because it is technically problematic to perform TTFields frequency scans in vivo, the frequency detected in the cell cultures was also employed for the animal studies.” Point 5: Why the in vitro experiment was performed for 72 hours and in in vivo for 120 hours?
| 2 | 1 |
There is no explanation why the most effective dose was 150 kHz and higher does actually decreases the killing. Can this dose observed in vitro on only tumor cells be translated to in vivo work where the tumor microenvironment is total different?
| 1 | 2 |
cancers14122959_perova
| 1 |
When it comes to testing efficacy, we would like to use a treatment duration as long as possible to maximize the effect. However, this duration varies between in vitro and in vivo work. For cell cultures, the control samples continue to grow throughout the treatment duration, and so after 72 h the plates are very much confluent, and may not be left to continue and grow any further (without drastic changes in the environmental conditions, making them inadequate control cells). It is also noteworthy that such treatment periods for in vitro work are well accepted in the field of TTFields research. In vivo, prior studies also used similar time frames of 2 to 3 weeks from inoculation when working with the N1S1 model (Buijs et al., 2012 -; Ju et al., 2009; Thompson et al., 2012). Extending treatment duration was also limited by the physical status of the animals, as was explained in response 3.
| 2 | 1 |
Why the in vitro experiment was performed for 72 hours and in in vivo for 120 hours?
| 1 | 2 |
cancers14122959_perova
| 1 |
Chloroquine was added to the cell cultures only at the final hours of the treatment for answering questions related to the mechanism of action, and not for boosting the efficacy of the other treatments. When using chloroquine in animal studies, it is for efficacy purposes, and so it is used throughout the treatment period. Since our animal experiment aimed to examine the efficacy of concomitant TTFields and sorafenib, with no additional agents, chloroquine was not employed.
| 2 | 1 |
Any explanation for why not using cloroquine in vivo to integrate better the in vitro data.
| 1 | 2 |
cancers14122959_perova
| 1 |
For the cytotoxicity assay we remove the supernatant, washed the cells, and then collect the adherent cells following trypsinization and visual inspection to verify all cells were removed. Indeed, as the reviewer mentioned correctly, we “counted the live cells and plotted the final data as percentage of untreated controls”. We agree that this is not cytotoxicity per se, and that is why we also perform 7-AAD/annexin-V staining of the cells. For this apoptosis assay we do collect the supernatant together with the adherent cells. Increased apoptosis and/or necrosis indicates that reduction in the cell number observed in the cell count emanates at least in part from cytotoxicity. We agree with the reviewer that we should not confuse the cell count measurements with the term “cytotoxicity” prior to showing the effect on apoptosis, and therefore we have changed this terminology throughout the paper.
| 2 | 1 |
The authors claim that cytotoxicity was measured “by cell counting using iCyt EC800 (Sony Biotechnology) 123 flow cytometer, and expressed as a percentage relative to the control.” Does this imply that they counted the live cells and plotted the final data as percentage of untreated controls (as figure 1 suggests). However, is this a real cytotoxicity or a cell growth inhibition? Did they measure the adherent cells after trypsinization. Were the cells from supernatant counted (where are probably the majority of dead cells)? There is a big difference between a therapy which kills vs a therapy which induces a cellular arrest.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for these questions. Preliminary tests using CD31 staining revealed no differences between the groups regarding blood vessel density and therefore we did not pursue the research in that direction. We agree that measuring the effect of TTFields on the anti-angiogenic effect of sorafenib and on resistance to sorafenib are interesting, these topics were however not within the scope of the current study, and remain for future investigations. Regarding possible discrepancy between the volume fold change and tumor weight, please see below a graph showing tumor volume (as measured by MRI) versus tumor weight, both measured at the end of the study. The graph shows very good correlation between the two parameters, indicating the reliability of the measurements. If the reviewer feels there is discrepancy, it may be due to the volume shown as fold change relative to the initial tumor volume and not as the end value.
| 2 | 1 |
Since sorafenib acts also on angiogenesis, did the authors investigate if TTFields may interfere with anti-angiogenic effect sorafenib-mediated? Also, is there any evidence that TTF may prevent the pretty common resistance to sorafenib observed in clinic? Moreover, there is a discrepancy between the fold changes in tumor weight vs. volume in the combination group vs. untreated group. Did the authors check changes in blood vessels density. Were the mice perfused before collecting the tumors?
| 1 | 2 |
cancers14122959_perova
| 1 |
We routinely examine our cells for Mycoplasma. Regarding authenticity, the cells were used shortly after purchase, and so there was no need to examine this.
| 2 | 1 |
Was Mycoplasma testing done routinely? Were the cells checked also for authenticity?
| 1 | 2 |
cancers14122959_perova
| 1 |
We appreciate this question. Autophagy is a process that is elevated in order to cope with cellular stress, but as stress level elevate, autophagy can no longer provide the protection the cell needs, and the cell will undergo apoptosis. This kind of kinetics means that the levels of autophagy markers will depend on the time point the cells are examined. This may be appreciated from figure 2 panel c and d, with the different kinetics displayed by the two cell lines. While for Huh-7D12 autophagy levels increase from 24 to 48 hours of treatment, in the HepG2 cells autophagy levels at 48 hours are lower than at 24 hours, indicating these cells are already after the autophagy peak and on the way to apoptosis. Indeed, figure 1d shows higher levels of apoptosis for HepG2 cells. To clarify we have added a few sentences to the text. In results sub-section 3.2: “However, autophagy kinetics seems to be faster in the HepG2 cells, in which LC3 markers are lower at 48 versus 24 hours, whereas elevation is seen from 24 to 48 hours for the Huh-7D12 cells.” In the discussion “While autophagy serves as a survival strategy of cells, when stress levels continue raising it may be over-activated and mediate cell death [9]. The faster autophagy kinetics seen for the HepG2 relative to Huh-7D12 cells following application of TTFields is in agreement with the higher apoptosis levels displayed by this cell line, and may serve as an additional rational for the higher efficacy of TTFields against it. Examination of the reasons for faster autophagy in HepG2 relative to Huh-7D12 cells is out of the scope of this work.” For more clarity we have also added a more in-depth kinetic study, including additional relevant markers and additional time points, examinations that were performed for the combined treatment as compared to TTFields and sorafenib alone. The higher changes in expression levels and faster kinetics when TTFields and sorafenib were applied together rather than alone indicate higher stress levels imposed on the cells in the former case” Figure 3. In the animal study we may only measure one time point. The lower autophagy seen at this time for the combined treatment together with the higher apoptosis indicate that we are further along the kinetic timeline of autophagy relative to the monotherapies, suggesting higher stress in animals receiving the combined treatment. We have now added IHC examination of beclin-1, an additional autophagy marker, and of GRP78, a marker for ER stress, as described in results sub-section 3.4.
| 2 | 1 |
Is there any explanation why the combination of TTFields and sorafenib did not induce a significant level of autophagy as compared to untreated animals which invalid the initial hypothesis that “concomitant application of sorafenib and TTFields may increase stress levels enough to tilt autophagy towards the cell death pathway”.
| 1 | 2 |
cancers14122959_perova
| 1 |
The sentence referred to was meant to describe only the in vivo outcomes. We have rephrased it to be more accurate and clear: “While each treatment alone elevated levels of autophagy relative to control, TTFields concomitant with sorafenib induced a significant increase versus control in tumor ER stress and apoptosis levels, demonstrating increased stress under the multimodal treatment.” Point 12: Finally, adding to all the above questions, I found a very weak Discussion section which must be extended.
| 2 | 1 |
The statement “TTFields concomitant with sorafenib induced a significant increase in apoptosis’ in the abstract section is overstated. When compared with sorafenib alone there is practically no difference. Moreover, TTFields failed to increases apoptosis when added to sorafenib and compared to sorafenib alone in one out of two human cells line investigated
| 1 | 2 |
cancers14122959_perova
| 1 |
We thanks the reviewer for this comment. We have now elaborated on many issues throughout the discussion, as was described thorough this response letter. Regarding the conclusion, we have rephrased it for better accuracy: “TTFields were identified to be most efficient for treatment of HCC cells at 150 kHz, and this frequency further demonstrated in vivo efficacy.” Why only one frequency was used in vivo, and the difference between frequency and dose, were explained in response 4.
| 2 | 1 |
Finally, adding to all the above questions, I found a very weak Discussion section which must be extended. Moreover, in the Discussion section, the authors concluded that “TTFields display efficacy for treatment of HCC in vitro and in vivo, with an optimal frequency of 150 kHz”. This is not a correct statement. While for in vitro data, the authors have data, for in vivo they used only one frequency of 150 Hz. At least one different dose should have been studied for comparison since this is a completely different tumor environment than the in vitro one.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for this question. Indeed, the sorafenib dose used in this study proved to be slightly more efficacious than TTFields in controlling tumor fold increase. Nevertheless, these differences between the monotherapies, did not reach statistical significance. In accordance, there was no statistical difference between the monotherapies in the expression levels of the LC3 marker and the levels of cleaved PARP. In order to better understand the mechanism of action, we have added experiments to better characterize the autophagy-apoptosis interplay for treatment with concomitant TTFields and sorafenib, which are described in response 3. Of note, the clinical development of TTFields does not aim to replace sorafenib with TTFields, but rather to add TTFields on top of sorafenib and therefore this work focused mainly on the potential added value in combining these 2 modalities.
| 2 | 1 |
1) The authors demonstrate the efficacy of TTFields in vivo even when used as monotherapy. As shown in the Figure 4 C and D, TTFields were found less effective in terms of reducing the tumors volume and weight when compared with sorafenib. However, no differences in expression of LC3 marker were observed between these groups (treated with TTFields or with sorafenib)(as shown in Figure 4D). Similarly, low evidence of apoptosis (expression of cleaved PARP) was found in these groups, as shown in Figure 4F. What is the mechanism illustrating higher efficiency of sorafenib against HCC?
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for pointing out this issue. Quantification of the IHC images was done automatically. The whole slide was scanned, and the CaseViewer software was used to exclude non-tumor areas. The signals of the stained protein and the nuclei were resolved by color deconvolution and quantified separately using the FIJI software (ImageJ) software. Average signal per cell or percent of positive cells was calculated. As the reviewer pointed out, the high magnification images we chose to show do not correctly reflect the quantification performed by the software, and we have now replaced them with better representative fields of the slides.
| 2 | 1 |
2) Despite the expression of cleaved PARP was very low in the tumors treated with TTFields or sorafenib alone ( as shown in IHC-images in Figure 4F), the authors declare about ~ 20% of positive cells, as show in the graphs below IHC-staining. Similar, the graphs illustrating the LC3 expression are not in a proper fit with the images shown in Figure 4D.
| 1 | 2 |
cancers14122959_perova
| 1 |
values are mean (N ≥ 3) ± SEM. We sincerely appreciate this well-taken comment. these examinations were performed for the combined treatment as compared to TTFields and sorafenib alone. We thank the reviewer for this comment, as these additions add much clarity to the mechanism of action of TTFields in combination with sorafenib and provide a more coherent explanation for the in vivo results. These additions may be seen in Figure 3 and are described in results sub-section 3.4, Autophagy-apoptosis Interplay For Treatment with Concomitant TTFields and Sorafenib: “In order to investigate the mechanism of action of TTFields-sorafenib co-application, HepG2 and Huh-7D12 cells were treated for 6, 24, or 48 hours with TTFields, sorafenib (3µM), or the two modalities together, and then examined for expression levels of various proteins. For HepG2 cells, the autophagy marker beclin-1 demonstrated elevation after 6 hours of treatment, which was later replaced with diminished expression levels (Figure 3d). This type of behavior was seen in all treatment groups, but was most pronounced for TTFields-sorafenib co-application. The autophagy marker LC3 also displayed such bi-phasic characteristics, but with a somewhat slower kinetics, showing some elevation at 6 hours of treatment, but higher elevation at the 24 hours time point (Figure 3d). As in the case of beclin-1, the magnitude of the effect was higher for co-treatment of TTFields and sorafenib relative to the monotherapies. GRP78, a marker of ER stress, remained low in all treatment groups for 6 and 24 hours of treatment, but demonstrated elevated levels at the later, 48-hours time point (Figure 3e). The apoptosis marker cleaved PARP displayed increased expression in the combined group already after 24 hours, elevating even further after 48 hours of treatment. For the monotherapies, cleaved PARP increase was only evident at 48 hours of treatment, and to a lower extent than that in the co-treatment group (Figure 3f). The slower kinetics of the autophagy-apoptosis path in the Huh-7D12 cells, as seen from the elevation of LC3 after as much as 48 hours (Figure 2c and d), prevented from detecting such changes in the levels of these markers in this cell line (Figure S1).” And in the discussion part “Kinetic examination in the HepG2 cells revealed elevation in autophagy levels as early as 6 hours of TTFields or sorafenib treatment, which diminished and were replaced with ER stress and apoptosis for 48 hours of treatment. These results are in line with a previous study that focused on the effects of sorafenib on such markers in HepG2 cells [32]. The higher changes in expression levels and faster kinetics when TTFields and sorafenib were applied together rather than alone indicate higher stress levels imposed on the cells in the former case.” Figure 3. HepG2 cells were treated for 6, 24, or 48 hours with 150 kHz TTFields, 3 µM sorafenib, or the two treatments combined, followed by Western blot examination of the autophagy markers beclin-1 and LC3 (d), the ER stress marker GRP78 (e), and the apoptosis marker cleaved PARP (f). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001 relative to time-respective control; two-way ANOVA. In the animal study we have now added IHC examination of beclin-1 and of GRP78, a marker for ER stress, as described in results sub-section 3.4. Concomitant TTFields with Sorafenib Enhances Treatment Efficacy in Vivo: “Tumor histology and immunostaining for beclin-1 and LC3, GRP78, and cleaved PARP were performed to examine autophagy, ER stress, and apoptosis levels, respectively. GRP78 levels in the groups treated with TTFields or sorafenib alone remained unchanged from the control, but were elevated 2-fold in the TTFields plus sorafenib group (Figure 4f). Additionally, the percentage of cells positive for cleaved PARP was significantly higher relative to control only in the combination group (Figure 4g).” and also in the discussion: “The lower autophagy accompanied by the higher ER stress and apoptosis displayed in the conjunction group relative to the monotherapies groups following 6 days of treatment suggest that these animals were pushed further along the autophagy-apoptosis kinetic timeline due to the higher levels of stress experienced by these animals, in accordance with the results described for the cell cultures.” Figure 4. tumor slices were subjected to immunohistochemical analysis for beclin-1 and LC3 (e), GRP78 (f), and cleaved PARP (g). Values are mean ± SD. *p < 0.05, **p < 0.01, and ****p < 0.0001 relative to control for labels above bars, or between indicated groups; Student’s T-test. Beclin-1 levels were increased more than 4-fold relative to control in all treatment groups, while intensity of LC3 staining was increased about 3-fold relative to control in the individual TTFields and sorafenib groups, but only 2-fold in the combination group (Figure 4e). To better explain the correlation between autophagy and apoptosis we have performed additional examinations to include more markers and at additional time points.
| 2 | 1 |
3) It will be much better to provide the data to explain the mechanisms illustrating why the monotherapy of TTFields or sorafenin induced autophagy, whereas the tumors treated with combination developed the substantial apoptotic death of tumor cells.
| 1 | 2 |
cancers14122959_perova
| 1 |
Both cell lines experience elevation of apoptosis following application of sorafenib in a dose dependent manner, as evident from the AnnV/7AAD results. However, while TTFields greatly elevate apoptosis in HepG2 cells, they have a low effect on apoptosis levels in the Huh-7D12 cells, seen both in Figure 1d and in Figure 3c. As was explained in the discussion, this difference between the cell lines may be attributed to the different p53 status, wild type in HepG2 and mutated in Huh-7D12, as there are previous indications of lower TTFields-induced apoptosis in cell lines with mutated p53. As suggested by the reviewer, in order to back up the AnnV/7AAD results we added WB for cleaved PARP, as described in response 3.
| 2 | 1 |
4) Since Annexin V/7-ADD data was not convincing and the authors observed the minor increase of apoptotic cells after HCC cells were treated with combination of TTFields and sorafenib ( when compared to the cells treated with TTFiealds and sorafenib alone), I suggest to run the WBs to examine the expression of the cleaved forms of PARP and caspase-3 ( for both HCC cell lines). This might be helpful and make the in vitro data more relavant with the data shown in vivo.
| 1 | 2 |
cancers14122959_perova
| 1 |
We apologize for accidentally leaving out this figure, and have now added it.
| 2 | 1 |
1) Figure 4A is missing.
| 1 | 2 |
cancers14122959_perova
| 1 |
We thank the reviewer for this comment. In the in vitro experiments we used cell lines derived from humans. However, these cell lines cannot be implanted to rats, and so for the in vivo experiments we had to use a cell line from rats.
| 2 | 1 |
2) the different HCC cell lines were used for in vitro and in vivo experiments, therfore making difficult to compare these data.
| 1 | 2 |
cancers14122959_perova
| 1 |
We have introduced new several sentences in this paragraph of Materials and Methods, which define these ad-hoc rules. Lines 154-155: “mgR51C read-outs have been introduced into the classification system as PVS1_O or BP7_O codes of variable evidence strength depending on the splicing outcome [P, Sup-porting (±1 point); M, Moderate (±2); Strong (±4); Very Strong (±8)].” Lines 163-179: “we have developed some ad-hoc rules that take into consideration the coding potential of each individual transcript and its relative contribution to the overall expression to reach the appropriate PVS1_O or BP/_O evidence strength. In brief, for each complex read-out we have applied the following algorithm: (i) De-convolute mgR51C read-outs in-to individual transcripts; (ii) apply ACMG/AMP evidences to each individual transcript; (iii) produce an overall PVS1_O (or BP7_O) code strength based on the relative contribu-tion of individual transcripts/evidences to the overall expression. Thus, if pathogenic supporting transcripts contribute ≥90% to the overall expression, PVS1_O_ code is applied (if different transcripts support different pathogenic evidence strengths, the lowest strength contributing >10% to the overall expression is selected for overall evidence strength). Similarly, BP7_O_ code is applied if benign supporting transcripts contribute ≥90% to the overall expression (if different transcripts support different pathogenic evi-dence strengths, the lowest strength contributing >10% to the overall expression is selected for overall evidence strength). If neither pathogenic nor benign supporting transcripts contribute ≥90% to the overall expression, the splicing assay is considered not providing any evidence in favor, or against, pathogenicity. Recently, we have used a similar ap-proach to deal with complex PALB2/ATM minigene read-outs [20,30].” 2 -
| 2 | 1 |
Line 156 – the authors refer to “ad-hoc rules” that they have developed for consideration of the different coding transcripts associated with the same spliceogenic variant in variant interpretation and classification. Although a reference is provided so that the reader can look up what these ad-hoc rules are, it would also be helpful to briefly describe these in the current manuscript.
| 1 | 2 |
cancers14122960_makarova
| 1 |
The five RAD51 paralogs are known to be required for homologous recombination and maintenance of genomic stability. Indeed, RAD51C interacts with RAD51B, RAD51D, XRCC2 and XRCC3 in two different complexes that play a role in homologous recombination. Miller et al studied the interaction between RAD51B and D (and also XRCC3) with deletion mutants. These authors found that Rad51C1-285 (includes β-strands 1-5) or Rad51C285-376 (includes β-strands 6-9) did not bind RAD51B. So, a complete beta-sheet is important in maintaining the overall fold of the protein. Moreover, the missense variant p.Arg312Trp (ß-strand 6) has been shown to impair RAD51C function (Gayarre et al 2017). Both studies indicate that this protein region is essential for RAD51C function so that transcripts lacking any of the β-strands, such as ▼(E6q4)-a, ▼(E6q4)-b, Δ(E7), Δ(E8) or the in-frame isoform Δ(E5), is probably deleterious. - We have modified this part, adding several sentences to clarify it. Lines 270-277: “ The integrity of the β-sheet is important in maintaining the overall fold of the RAD51C protein and the interaction with RAD51B, so that alterations of any ß-strand of RAD51C should be considered deleterious [33]. Further, structural features (the order of the ß--strands in space is not the same as their order in sequence) predict that proteins lacking any single b-strand would fail to form the ß--sheet resulting in a collapse of the protein core and misfolding of the protein [33]. Moreover, the missense variant p.Arg312Trp (ß-strand 6) has been shown to impair RAD51C function [34]. Altogether these data,…” 3-
| 2 | 1 |
Lines 250-251 and 263-265– In these lines the authors discuss transcripts in which the encoded proteins lack some beta strands. The lack of these protein structures are used as evidence to support pathogenicity. It is not clear from the text how lack of these beta strands is predicted to impact protein function. Is there evidence from another source that these beta strands are critical to protein function and that their loss is deleterious (rather than resulting in normal or slightly reduced protein activity)?
| 1 | 2 |
cancers14122960_makarova
| 1 |
We have added all the transcript names and their contribution in Table 2 (Column PVS1_O/BP7_O mgR51C_ex2-8).
| 2 | 1 |
Lines 259-274 – This paragraph refers to evidence used in the variant interpretation of three aberrant transcripts that kept the open reading-frame (Δ(E2p3), Δ(E5) and ▼(E8p3)). However, the corresponding tables that summarizes variant classification according to the ACMG/AMP-based criteria (Table 2), does not include these transcript isoform names. As such, in order to correlate the discussion in this paragraph with the information in table 2, the reader also needs to cross reference Table 1 or 3. Incorporation of the transcript isoform names in table 2 would assist the reader in correlating this discussion of transcript isoforms with the corresponding evidence used to classify each of the variants.
| 1 | 2 |
cancers14122960_makarova
| 1 |
We have included cross-references to Table 1 to facilitate understanding of the manuscript.
| 2 | 1 |
Lines 259-274 – This paragraph refers to evidence used in the variant interpretation of three aberrant transcripts that kept the open reading-frame (Δ(E2p3), Δ(E5) and ▼(E8p3)). However, the corresponding tables that summarizes variant classification according to the ACMG/AMP-based criteria (Table 2), does not include these transcript isoform names. As such, in order to correlate the discussion in this paragraph with the information in table 2, the reader also needs to cross reference Table 1 or 3. Incorporation of the transcript isoform names in table 2 would assist the reader in correlating this discussion of transcript isoforms with the corresponding evidence used to classify each of the variants.
| 1 | 2 |
cancers14122960_makarova
| 1 |
Very important comment. As the reviewer indicates, it would be essential to define the threshold of RAD51C expression from which it keeps its tumor suppressor activity. Unfortunately, it is not known by now but this finding would provide critical information to determine the pathogenicity of leaky spliceogenic variants.
| 2 | 1 |
Lines 342-359 – This paragraph discusses two RAD51C variants, c.404+3A>G and c.705+3A>G, for which mg-FL transcripts were detected in 26.3% and 21.3% or transcripts, respectively. The discussion of these variants in this section (and in other sections) does not address the significance of canonical transcripts in these cases. Leaky splice variants have been reported in various genes and sometimes can be associated with milder phenotypes (or no phenotypes), presumably because the canonical isoforms contribute to a “phenotypic rescue”. Is it known whether there is a threshold of RAD51C deficiency that is tolerated before associated cancer risks become increased?
| 1 | 2 |
cancers14122960_makarova
| 1 |
So, we have modified this paragraph to introduce this information: “Only two variants (c.404+3A>G and c.705+3A>G) displayed 26% and 21% of the mgFL-transcript, respectively. Unfortunately, it is not known the minimal amount of RAD51C expression to confer tumor suppressor haplosufficiency so, these splicing assays were not considered informative (PVS1_O_N/A).” Note that leaky variants generate complex minigene read-outs (two or more transcripts), and are therefore classified accordingly (see methods).
| 2 | 1 |
Lines 342-359 – This paragraph discusses two RAD51C variants, c.404+3A>G and c.705+3A>G, for which mg-FL transcripts were detected in 26.3% and 21.3% or transcripts, respectively. The discussion of these variants in this section (and in other sections) does not address the significance of canonical transcripts in these cases. Leaky splice variants have been reported in various genes and sometimes can be associated with milder phenotypes (or no phenotypes), presumably because the canonical isoforms contribute to a “phenotypic rescue”. Is it known whether there is a threshold of RAD51C deficiency that is tolerated before associated cancer risks become increased?
| 1 | 2 |
cancers14122960_makarova
| 1 |
Acknowledge this suggestion. We have added two references of BRCA1 and BRCA2 studies.
| 2 | 1 |
in the Introduction in the sentence 61-69 the authors do not mention anything about BRCA1 (they mention MLH1 though); I would expect that as BRCA1 splice variants are deeply studied.
| 1 | 2 |
cancers14122960_makarova
| 1 |
In this section we have tried to describe the classification approach and the rules we have followed to classify the variants. Certainly, we agree with the referee as sometimes the method is mixed with some results. So, we have moved some sentences of the last paragraph of Materials and Methods to Results. “The PM3 evidence (in trans with a pathogenic variant in a recessive disorder) did not con-tribute to the final classification. Not surprisingly (FANCO is an extremely rare FA complementation group) [37], none of the tested variants has been identified in Fanconi Anemia patients (ClinVar and Global Variome shared LOVD databases and literature search-es). Similarly, the BS2 evidence (in trans with a pathogenic variant in a healthy individual) does not contribute to the final classification of our tested variants. Finally, we have considered that some pathogenic (PS2, PM1, PM6, PP2, PP4) and benign (BP1, BP3, BP5) codes are not applicable to the classification of RAD51C variants.” 3-
| 2 | 1 |
In the Materials and Methods, the section 2.6 is not actually a section that described some methodology. It appears to be more appropriate as results and maybe some information could be in the introduction.
| 1 | 2 |
cancers14122960_makarova
| 1 |
We have included explanations of these acronyms in text and Table 1 and have modified the following sentence: “Of the 19 characterized transcripts, 14 introduced premature termination codons (PTC; PTC transcripts), and of these, 10 are predicted to be degraded by the Nonsense-Mediated Decay pathway (NMD; PTC-NMD transcripts) that is considered convincing evidence of deleteriousness (Supplementary Table S3).
| 2 | 1 |
The acronyms need to be explained. Not all readers are familiar with PTC (premature termination codon) and NMD (nonsense mediated decay); as some variants are found PTC-NMD, this needs to be explained better. I would also stress a little more that FL is almost undetectable.
| 1 | 2 |
cancers14122960_makarova
| 1 |
We have modified the sentence of the Fl-transcript: “All variants altered splicing, 18 of which produced no traces of the mgFL-transcript or almost undetectable levels (<2.4%, c.904G>A)…”
| 2 | 1 |
The acronyms need to be explained. Not all readers are familiar with PTC (premature termination codon) and NMD (nonsense mediated decay); as some variants are found PTC-NMD, this needs to be explained better. I would also stress a little more that FL is almost undetectable.
| 1 | 2 |
cancers14122960_makarova
| 1 |
Acknowledge this comment. Splicing variants are called solely based on the size difference. It is good to look into Sanger sequence of splice variants after gel elution of the band and sequencing. We have sequenced the RT-PCR products of all variant assays (indicated in Materials and Methods, section 2.5. Minigene Splicing Assays). In fact, all the *.ab1 sequence and*.fsa fragment analysis files of RT-PCR products will be freely available at http://hdl.handle.net/10261/270934; https://doi.org/10.20350/digitalCSIC/14662 upon manuscript acceptance (links indicated in the manuscript section “Data Availability Statement”). Unfortunately, Sanger sequencing only allowed us to characterize the main transcripts, while the minor ones (<10% of the overall expression) are really difficult to characterize since gel band extraction does not work properly with these small amounts, or other methods (e.g. subcloning of RT-PCR products into a PCR-vector, or RNAseq of minigene outcomes) are laborious and not cost-effective. Anyway, we have also been using Fluorescent fragment analysis for transcript characterization in our previous studies (since Acedo et al, 2012). We have shown that this technique is highly sensitive, accurate and shows high resolution. For example, in Figure 1c, transcripts with minimal size differences (1-3 nt) are well-discriminated. So, for minor-rare transcripts is a good option (not perfect, we agree with the reviewer) to annotate them. - To clarify it, we have modified this part: Lines 132-134: “RT-PCR products were sequenced by Macrogen (Madrid, Spain), which allowed the characterization of the main variant-induced transcripts. Minor transcripts were annotated according to fluorescent fragment electrophoresis size data (see below).”
| 2 | 1 |
Splicing variants are called solely based on the size difference. It is good to look into Sanger sequence of splice variants after gel elution of the band and sequencing.
| 1 | 2 |
cancers14122960_makarova
| 1 |
We would like to thank reviewer #1 for his/her time and effort to review our manuscript. Please find our point-by-point responses below.
| 2 | 1 |
This manuscript adds evidence that NR2F1 is a dormancy marker. In addition, it shows that NR2F1 is primarily expressed in CAFs, particularly in inflammatory CAFs.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We totally agree with the reviewer that it is puzzling how NR2F1 expressed in CAFs of the primary breast cancer contribute to the dormancy of DTCs, thus the addition of a discussion on this point in more detail will strengthen this manuscript. Based on our results, NR2F1 expression in primary bulk tumor is associated with several pathways related to dormancy, and NR2F1 is most predominantly expressed in CAFs in the tumor microenvironment. However, we did not prove the underlying mechanism through which CAF-expressed NR2F1 regulates dormancy, and we do not intend to claim a causal relationship. Single-cell sequence data of metastatic tumor cohorts will allow us to investigate whether the expression of NR2F1 in CAFs in the metastatic TME is related to the dormancy of DTCs. We added the following sentences to the discussion section.
| 2 | 1 |
Given that dormant disseminated tumor cells (DTCs) highly express NR2F1 (Fluegel et al. Nat Cell Biol 2017), it is puzzling how NR2F1 in CAFs of the primary tumor would contribute to dormancy of DTCs. It would be nice, if the authors would discuss this important point in more detail.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We agree with the reviewer that it would be interesting to see how NR2F1 levels would change with endocrine therapy. However, we do not have access to cohorts that include tumor samples before and after neoadjuvant endocrine therapy at this point. What we do have access to regarding endocrine therapy is a cohort comparing responders and non- responders to neoadjuvant endocrine therapy (GSE145325). We found that NR2F1 expression between responders and non-responders to endocrine therapy was not different. We added this to the results section as follows.
| 2 | 1 |
The authors choose chemotherapy as a treatment option to compare it with NR2F1 levels. It would be interesting to see, how NR2F1 levels would change with endocrine treatment and/or endocrine resistance.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We agree with the reviewer that breast cancer is a heterogenous disease and we should show subtype-specific data. We analyzed survival outcomes and the cell fraction in TME by each immunohistological subtype. We did not observe any validated difference in survival outcomes. Cell fractionation of immune cells and stromal cells showed almost similar trends for the scores such as intratumor heterogeneity, HRD, mutation rate, and neoantigens across all subtypes. The results for each immunohistological subtype of single-cell Cohort 2 are shown in Supplementary Figure 7, and each subtype showed the same trend. We have revised the results section and added supplementary data as follows.
| 2 | 1 |
Breast cancer is a heterogenous disease. Different subtypes behave differently in many aspects. It would be great to see some subtype-specific data.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We thank the reviewer for reading our manuscript closely and pointing out our oversight. As you indicated, we did not find any significance difference in 2D in Figure 1A and have corrected the results section as follows.
| 2 | 1 |
The authors state “We demonstrated that the expression of NR2F1, RARB, and TGFB1 genes are higher in previously established dormant cells (D2OR murine breast cancer cells [44] compared to the proliferative cells (D2A1 cells) in both 2D and 3D cultures (Figure 1A, all p < 0.02).” This is not true for NR2F1 in 2D. Please correct.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We are grateful to the reviewer for her/his time and effort in reviewing our paper, as well as for pointing out issues to improve the paper.
| 2 | 1 |
The authors present interesting findings that a tumor dormancy marker, NR2F1, is predominantly expressed in the inflammatory CAFs, and high expression of NR2F1 is associated with suppressed immune response and increased density of stromal cells. However, this reviewer has a few concerns that need to be addressed before accepting this article for publication.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We completely agree with the reviewer that it will be informative to demonstrate the NR2F1 expression in both primary and metastatic breast cancer and the possible correlation of NR2F1 expression with metastasis. In Figure 3C, we present NR2F1 expression in primary breast cancer with and without distant metastases. NR2F1 expression was not increased in the group with later recurrence in four cohorts in this analysis. We also present NR2F1 expression between primary and metastatic breast cancer in Figure 3E with no significant difference. We did not find a clear association of NR2F1 with distant metastasis in this study. On the other hand, NR2F1 expression was higher in primary breast tumors with lymph node metastasis in all four cohorts, as shown in Figure 3B, suggesting an association between NR2F1 and lymph node metastasis. Given these results, we revised the results section as follows.
| 2 | 1 |
This reviewer noticed that all the analysis was performed on the public data sets of bulk RNA-seq or single cell sequencing of primary breast tumors. Can the authors do some analyses using data generated from both primary and metastatic tumors to check if there is any difference of NR2F1 expression, and how NR2F1 expression is correlated with metastasis?
| 1 | 2 |
cancers14122962_makarova
| 1 |
We completely agree with the reviewer that NR2F1 expression in cancer cells and CAFs should be investigated not only in primary but also in metastatic breast cancer using single-cell sequence in order to prove that NR2F1 expression in CAFs affects late recurrence. The main finding of this study is that NR2F1 is predominantly expressed in CAFs rather than in all other cell types in the TME, and we do not intend to claim that CAF-expressed NR2F1 regulates dormancy. Further, we do not have access to single-cell sequence cohorts of metastatic breast cancer tumors, but it is of our interest, and this will be our future direction. In response to the reviewer, we added the following sentences in the discussion section.
| 2 | 1 |
Moreover, the authors need to provide some, at least minimum, evidence or clues that CAF-expressed NR2F1 is responsible for tumor dormancy regulation.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We thank the reviewer for closely reading our manuscript and pointing out our oversight. We have corrected the results section as answered in Responce1.
| 2 | 1 |
Line 170, “Figure 1E” should be “Figure 1D”.
| 1 | 2 |
cancers14122962_makarova
| 1 |
We totally agree with the reviewer that it is puzzling how NR2F1 expressed in CAFs of the primary breast cancer contribute to the dormancy of DTCs, thus the addition of a discussion on this point in more detail will strengthen this manuscript. Based on our results, NR2F1 expression in primary bulk tumor is associated with several pathways related to dormancy, and NR2F1 is most predominantly expressed in CAFs in the tumor microenvironment. However, we did not prove the underlying mechanism through which CAF-expressed NR2F1 regulates dormancy, and we do not intend to claim a causal relationship. Single-cell sequence data of metastatic tumor cohorts will allow us to investigate whether the expression of NR2F1 in CAFs in the metastatic TME is related to the dormancy of DTCs. We added the following sentences to the discussion section.
| 2 | 1 |
Given that dormant disseminated tumor cells (DTCs) highly express NR2F1 (Fluegel et al. Nat Cell Biol 2017), it is puzzling how NR2F1 in CAFs of the primary tumor would contribute to dormancy of DTCs. It would be nice, if the authors would discuss this important point in more detail.
| 1 | 2 |
cancers14122962_perova
| 1 |
We agree with the reviewer that it would be interesting to see how NR2F1 levels would change with endocrine therapy. However, we do not have access to cohorts that include tumor samples before and after neoadjuvant endocrine therapy at this point. What we do have access to regarding endocrine therapy is a cohort comparing responders and non- responders to neoadjuvant endocrine therapy (GSE145325). We found that NR2F1 expression between responders and non-responders to endocrine therapy was not different. We added this to the results section as follows.
| 2 | 1 |
The authors choose chemotherapy as a treatment option to compare it with NR2F1 levels. It would be interesting to see, how NR2F1 levels would change with endocrine treatment and/or endocrine resistance.
| 1 | 2 |
cancers14122962_perova
| 1 |
We agree with the reviewer that breast cancer is a heterogenous disease and we should show subtype-specific data. We analyzed survival outcomes and the cell fraction in TME by each immunohistological subtype. We did not observe any validated difference in survival outcomes. Cell fractionation of immune cells and stromal cells showed almost similar trends for the scores such as intratumor heterogeneity, HRD, mutation rate, and neoantigens across all subtypes. The results for each immunohistological subtype of single-cell Cohort 2 are shown in Supplementary Figure 7, and each subtype showed the same trend. We have revised the results section and added supplementary data as follows.
| 2 | 1 |
Breast cancer is a heterogenous disease. Different subtypes behave differently in many aspects. It would be great to see some subtype-specific data.
| 1 | 2 |
cancers14122962_perova
| 1 |
We thank the reviewer for reading our manuscript closely and pointing out our oversight. As you indicated, we did not find any significance difference in 2D in Figure 1A and have corrected the results section as follows.
| 2 | 1 |
The authors state “We demonstrated that the expression of NR2F1, RARB, and TGFB1 genes are higher in previously established dormant cells (D2OR murine breast cancer cells [44] compared to the proliferative cells (D2A1 cells) in both 2D and 3D cultures (Figure 1A, all p < 0.02).” This is not true for NR2F1 in 2D. Please correct.
| 1 | 2 |
cancers14122962_perova
| 1 |
We completely agree with the reviewer that it will be informative to demonstrate the NR2F1 expression in both primary and metastatic breast cancer and the possible correlation of NR2F1 expression with metastasis. In Figure 3C, we present NR2F1 expression in primary breast cancer with and without distant metastases. NR2F1 expression was not increased in the group with later recurrence in four cohorts in this analysis. We also present NR2F1 expression between primary and metastatic breast cancer in Figure 3E with no significant difference. We did not find a clear association of NR2F1 with distant metastasis in this study. On the other hand, NR2F1 expression was higher in primary breast tumors with lymph node metastasis in all four cohorts, as shown in Figure 3B, suggesting an association between NR2F1 and lymph node metastasis. Given these results, we revised the results section as follows.
| 2 | 1 |
This reviewer noticed that all the analysis was performed on the public data sets of bulk RNA-seq or single cell sequencing of primary breast tumors. Can the authors do some analyses using data generated from both primary and metastatic tumors to check if there is any difference of NR2F1 expression, and how NR2F1 expression is correlated with metastasis?
| 1 | 2 |
cancers14122962_perova
| 1 |
We completely agree with the reviewer that NR2F1 expression in cancer cells and CAFs should be investigated not only in primary but also in metastatic breast cancer using single-cell sequence in order to prove that NR2F1 expression in CAFs affects late recurrence. The main finding of this study is that NR2F1 is predominantly expressed in CAFs rather than in all other cell types in the TME, and we do not intend to claim that CAF-expressed NR2F1 regulates dormancy. Further, we do not have access to single-cell sequence cohorts of metastatic breast cancer tumors, but it is of our interest, and this will be our future direction. In response to the reviewer, we added the following sentences in the discussion section.
| 2 | 1 |
Moreover, the authors need to provide some, at least minimum, evidence or clues that CAF-expressed NR2F1 is responsible for tumor dormancy regulation.
| 1 | 2 |
cancers14122962_perova
| 1 |
We thank the reviewer for closely reading our manuscript and pointing out our oversight. We have corrected the results section as answered in Responce1.
| 2 | 1 |
Line 170, “Figure 1E” should be “Figure 1D”.
| 1 | 2 |
cancers14122962_perova
| 1 |
According to the reviewer suggestion, FT-IR peaks shift related to chromium ion adsorption are discussed in the mechanism section and is incorporated in the revised manuscript.
| 2 | 1 |
Some FT-IR peaks shift with adsorption of Cr, and some not. Please add the discussion which peak shift is related to Cr, the authors can combine these with the discussion of the mechanism.
| 1 | 2 |
catal12030290_makarova
| 1 |
EDX analysis (In Figure 3) – is a conditional type of analysis of the chemical composition of the surface. In EDX, often, spectra with different atomic abundances of elements can be obtained even from the same sample.
| 2 | 1 |
In Figure 3, what is the origin of the increase concentration of Al and S?
| 1 | 2 |
catal12030290_makarova
| 1 |
The sulphuric acid disintegrated the leaves. Most of the substances in the leaves are reduced to carbon after two hours. Charring takes place by adding sulphuric acid and by the action of heat, charring removes hydrogen and oxygen from the solid, so that the remaining char is composed primarily of carbon. It also helps to remove the moist content in the leaves. The pollution problem may not be takes place.
| 2 | 1 |
In the experiment, H2SO4 is added in the first step, what is the purpose? Will this bring pollution problem?
| 1 | 2 |
catal12030290_makarova
| 1 |
Chromium is a potentially toxic metal occurring in water and groundwater as a result of natural and anthropogenic sources. The prepared adsorbent is well suitable in the real condition. Since the prepared adsorbent (Al-GNSC) is successfully reduces the chromium (VI) ion form the real groundwater samples.
| 2 | 1 |
How about the Cr concentration in the polluted ground water? Is the products suitable for this real condition?
| 1 | 2 |
catal12030290_makarova
| 1 |
Based on the reviewer suggestion, regeneration studies were performed briefly. Four different desorption agents such as tap water, 0.1M HCl, 0.1M H2SO4 and 0.1M NaOH were utilized to remove the adsorbed chromium ions from the Al-GNSC adsorbent. From this various desorption agents it was identified that 0.1M NaOH was more effective. Hence, the reuse of Al-GNSC from Cr(VI)-loaded material was studied for sorption and desorption cycles using sodium hydroxide as a regenerated agent. Relevant references are quoted in the revised manuscript. Reference: Sujitha Ravulapalli and Ravindhranath Kunta reported that sodium hydroxide was used as a regenerating agent for the sorption and desorption of Cr(VI) by activated carbon derived from Lantana camara plant. [Enhanced removal of chromium (VI) from wastewater using active carbon derived from Lantana camara plant as adsorbent. Water Sci Technol (2018) 78 (6): 1377–1389. https://doi.org/10.2166/wst.2018.413] Reference: M.A. Tandal and B.N.OZA. reported that Sodium hydroxide as a regenerating agent for the sorption and desorption of Cr(VI) by Granular activated carbon. [Adsorption and regeneration studies for the removal of Chromium (VI) from the waste water of electroplating industry using Granular activated carbon. Asian Journal of chemistry. Vol. 17, No.4 (2005), 2524-2530] Page 2 of 2 Comment 2:
| 2 | 1 |
The issue of regeneration has not been sufficiently worked out, but the authors themselves write about this, and it is not clear why NaOH was used for this.
| 1 | 2 |
catal12030290_makarova
| 1 |
We acknowledge the reviewer’s opinion. For the preparation of 1kg of Al-GNSC adsorbent, approximately 150 to 200 mL of hydrochloric acid was added in order to blend the aluminum in to the ground nut shell carbon. Thank you for your valuable comment.
| 2 | 1 |
It was also not clear to me how much hydrochloric acid was eventually added during the preparation of the adsorbent.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, the expression of separation factor is mentioned in the revised manuscript. 𝑅𝐿= 1 1+𝑏𝐶𝑖, where “Ci” is the initial concentration of Cr(VI) and “b” is the Langmuir constant.
| 2 | 1 |
The authors did not indicate how the “separation factor ‘RL’ is calculated and did not give a link to the equation for its calculation, and if the readers are not quite in the subject, then what kind of factor they do not understand.
| 1 | 2 |
catal12030290_makarova
| 1 |
However, we regret that we were not able to investigate the BET analysis due to pandemic situation, which could definitely give us additional information.
| 2 | 1 |
One could also estimate the specific surface area and porosity. So, the paper can be published after revision.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, the complete terms of all abbreviations are mentioned before the first use in the revised manuscript.
| 2 | 1 |
The authors should write the complete terms of all abbreviations (including the instruments) before the first use in the abstract and main manuscript i.e. FT-IR and SEM in abstract section et al.
| 1 | 2 |
catal12030290_makarova
| 1 |
As suggested, the novelty of the research work is explained in the introduction part of the revised manuscript, and the Page 2 of 4 obtained results are compared with the recent literature.
| 2 | 1 |
The authors should clearly explain the innovation and importance of their work on the introduction of the manuscript. They should justify the value of the work and compare their work with previously similar published papers.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, the SEM images with same scale are provided in the revised manuscript.
| 2 | 1 |
Fig. 1 - for a more effective visual comparison, authors recommended to provide SEM images of the same scale. In such form is rather difficult to make adequate comparison.
| 1 | 2 |
catal12030290_makarova
| 1 |
We hope the reviewer understand the experimental deficiencies at the stage of the present experiments. We deeply appreciate the comment raised by the reviewer. Thank you very much. However, we regret that we were not able to investigate the XPS analysis due to pandemic situation, which could definitely give us additional information about the elemental confirmation.
| 2 | 1 |
First of all, authors have to attach a EDX mapping images before/after sorption of chromium ions. And the authors are strongly recommended to add XPS spectra to this section of revised manuscript. The XPS method is much more sensitive and more accurately determines changes in the chemical composition of samples.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, XRD study has been performed and incorporated in the revised manuscript.
| 2 | 1 |
Why authors did not use XRD technique for sample characterization?
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, kinetic study has been performed and incorporated in the revised manuscript.
| 2 | 1 |
The adsorption capacities of Al-GNSC adsorbents at different contact times have been provided. Which kinetics are right? Please add missing information about appropriate kinetic model in revised manuscript.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion, the conditions for testing adsorbents of Cr(VI) is incorporated in the Table and is mentioned in the revised manuscript.
| 2 | 1 |
It is rather difficult to make an adequate comparison of certain properties (catalysts or sorbents) with the already available results, since the concentration of the pollutant and the mass of the loaded sorbent vary in each experiment. Therefore, the authors are recommended to add the missing information (i.e. conditions for testing sorbents of Cr(VI) ions) to Table 3.
| 1 | 2 |
catal12030290_makarova
| 1 |
According to the reviewer suggestion the adsorption capacity of groundnut shell activated carbon (Qe= 7.4 mg/g) is mentioned in the revised manuscript.
| 2 | 1 |
In order to confirm proposed mechanism of Cr(VI) adsorption (illustrated on the fig 6) Authors should provide data on adsorption capacity of pristine groundnut shell activated carbon (not modified with Al).
| 1 | 2 |
catal12030290_makarova
| 1 |
Thank you for your valuable suggestion. According to the reviewer suggestion the conclusion section is elaborated with specific conclusions in the revised manuscript.
| 2 | 1 |
The conclusion section should be elaborated and improved. The author should bring specific conclusions in accordance with obtained results.
| 1 | 2 |
catal12030290_makarova
| 1 |
We acknowledge the reviewer’s opinion. As suggested by the reviewer, we checked the manuscript carefully and enlisted a professional English language service to eliminate the spelling mistakes and grammatical errors. Thank you for your valuable suggestion for strengthening the quality of the manuscript.
| 2 | 1 |
Moderate English changes required
| 1 | 2 |
catal12030290_makarova
| 1 |
Indeed, the liposomes are prepared by a conventional method. However, here and in our previous work [1] we use the term liposomal nanotraps to reflect the functional aspect of the liposomal action. In this context, the liposomes act as traps for bacterial toxins.
| 2 | 1 |
The authors used the term liposome nanotraps. The authors have used conventional liposome preparations. It is not clear why they are calling the preparations as the nanotraps.
| 1 | 2 |
cells11010166_perova
| 1 |
We verified the total protein content of the supernatants by Coomassie blue staining, which was similar for all strains (Figure. 1, for referee inspection only). The experiments were performed using different bacterial supernatant batches and results were remarkably consistent. All the supernatants were collected at the exact same bacterial culture’s optical density to harvest the bacteria in a comparable state between batches and strains. In our previous work we have shown that cholesterol-dependent cytolysins (SLO, PLY) displayed different kinetics and dynamics of their hemolytic activity compared to SLS. As a result, the total hemolytic activities of individual streptococcal supernatants were not represented by a simple sum of activities (concentrations) of their individual toxins but displayed more complex time- and amount-dependent behavior: e.g. – the activity of PLY/SLO was prevalent at the initial times of incubation and at relatively high amounts (volume) of supernatants, whereas SLS activity fully developed only after initial lag period but was prevalent when relatively low amounts of supernatants were used in the assays. The relative quantifications of toxins between streptococcal species and strains were performed in a previous publication [1] and are referred to throughout the manuscript. We added a paragraph in the result section (line 215-222) that summarizes those points. Moreover, in our current experiments we aim not only at the neutralization of the whole hemolytic secretome of streptococcus (as described in [1]) but also include (putative) cytotoxic/cytostatic activities that might be carried out either by hemolysins (SLO/SLS) or other not yet identified toxins that display cytotoxic/cytostatic but no hemolytic activities. For these reasons, in our current experiments, we use specific cytotoxic/cytostatic activities of total supernatants derived from a toxicity assay displayed in (Figure. 1), instead of concentrations of individual (partly unknown) toxins. However, we agree with the reviewers that using volume units is confusing. We therefore, edited our manuscript to display the lethal dose (LD%) interpolated from results shown in Figure 1 instead. Using LD% units accounts for the batch variability.
| 2 | 1 |
The authors have used bacterial culture supernatants to examine the cytotoxic effect on the THP-1, Jurkat, and Raji cell lines. Quantitation fo the the culture medium is required (by measuring the total protein content, or by providing some quantitative indicator). Volume of the culture supernatant is mentioned in the results. It is not quantitative, and will vary from batch to bath of the bacterial culture.
| 1 | 2 |
cells11010166_perova
| 1 |
While LDH-release or MTT assay are frequently used techniques and might be preferred by some investigators, other approaches assessing cell viability might be more popular by others, dependent on particular experimental settings of a particular study. Figure 2 (for referee inspection only) demonstrates that Alamar blue cell viability assay and Trypan blue live/dead quantification provide results that are identical to those obtained in the cell proliferation protocol used in our study. We believe that the latter protocol is the most suitable experimental approach for our study since it allows distinguishing between cell lysis, cytotoxisity and cytostatic effects (i.e. between cytolysins, cytotoxins and cytostatic agents) in a single experiment. The techniques and algorithms used by the CellDrop relies on accurate and unbiased measurements, akin to previously published techniques[2,3].
| 2 | 1 |
LDH-release assay of cytotoxicity or MTT assay of cell viability should have been used for measuring the cell death.
| 1 | 2 |
cells11010166_perova
| 1 |
The focus of our study is to highlight that successful protection against the whole palette of streptococcal toxins can be achieved by using liposomal nanotraps and to show that the liposome requirements differs between bacterial species and between different types of immune cells. We agree that the SLO neutralization is not novel and simply confirm results from previous publications by us and others. In the current work we do not intend to reveal new neutralization mechanisms either for SLO or for SLS.
| 2 | 1 |
It is well-known that SLO binds to the cholesterol-containing lipid bilayer. Therefore, it is obvious that the cholesterol-containing liposomes would neutralise SLO present in the bacterial culture. Therefore, no new information is provided with these experiments.
| 1 | 2 |
cells11010166_perova
| 1 |
Streptolysin S (SLS) is a small, non-immunogenic, peptide. This means that no commercial antibody against it is available and the peptide is too small for reliable mass spectrometry detection. It is heavily post-translationally modified and is the product of a complex operon. Its exact mode-of-action is still not yet fully clarified. In a previous publication, we were able to show that SLS is neutralized by phosphatidylcholine as well as sphingomyelin liposomes [1]. However, given the poor characterization of SLS, its unavailability from commercial providers and its extremely tedious purification, the mechanistic details of the SLS neutralization will require an extensive a project of its own and are beyond the scope of the current study.
| 2 | 1 |
In my opinion, differential inhibition of the SLS activity in GGS supernatant by the liposome preparations is interesting, and may provide new insights, if explored in more detail.
| 1 | 2 |
cells11010166_perova
| 1 |
We used two controls in the protection experiments. In one control we challenged the cells with bacterial supernatant without adding liposomes, to determine that the baseline cytotoxicity is in line with the results displayed in the Figure 1. The second control consists of the immune cells without toxin or liposomes. This control represents 100% survival and allows us to normalize our survival data. We also tested the intrinsic toxicity of liposomes on their own to see if they did impede cell growth or were cytotoxic. At the concentrations used in current study the liposomes were not cytotoxic (data not shown). The following text was added to the material and method section: The data were normalized to a control incubated with PBS instead of bacterial supernatant (considered as 0% cell death). A control challenged by bacterial supernatant without liposomes was added for each assay to verify the expected cytotoxic activity 2.
| 2 | 1 |
Although the sensitivity of immune cells to GAS or GGS supernatants is shown in the first section of the results, were any controls used in the neutralization assays with liposomes?
| 1 | 2 |
cells11010166_perova
| 1 |
We selected the specific supernatant volumes used to reach a similar lethal dose to compare the protection capability and efficiency of the liposomal nanotraps. However, as the immune cell lines have different sensitivities depending on the toxin profile of the tested strain, a similar lethal dose corresponds to different supernatant volumes. We used the results displayed in Figure 1 of the manuscript to determine which volume to use. We agree that the way we presented it can lead to confusion and we replaced the supernatant volume values by LD% values. The following text was added to the material and method section: The added supernatant volume was determined based on the toxicity assay results and was used either at saturating dose (lethal dose >90, LD>90) or at non-saturating dose (LD60-90) to study minor toxin activities. We agree that the distinction between saturating (LD>90) and non-saturating (LD60-90) was confusing. To clarify our process we did add clarifications throughout the results part and we added a panel in figure 2 to include results obtained after challenge of LD>90 of GGS 5804 supernatant.
| 2 | 1 |
In the material and methods section it is stated that the cells survival assays start with the addition of a fixed volume of supernatant but it is not clear to me the criteria for reflecting different amounts of supernatant in Figures 2 and 3. It should be explained more clearly. What is the reason for using different amounts of supernatant with cytolysins depending on its source or the type of cell line it is tested against in figures 2 and 3? This should be explained. Furthermore, it should be justified why this difference does not influence the comparison of the results with the different cell lines (THP1, Jurkat, Raji).
| 1 | 2 |
cells11010166_perova
| 1 |
All the corrections are highlighted in yellow. Corrected the legend of Figure 4 (EM/HPHYT/BHT, EM/HPHYT, ET/HPHYT).
| 2 | 1 |
Some errors in the legend of Figure 4, such as EM/PHYT/BHT, EM/PHYT is not presented in Figure 4.
| 1 | 2 |
cosmetics2030248_makarova
| 1 |
Response: The errors were corrected in text on page 09. (EM/HPHYT/BHT, EM/HPHYT, ET/HPHYT).
| 2 | 1 |
In Figure 4, the results is represented in three times or signal one or other times. Therefore, you must rewritten the legend of Figure 4.
| 1 | 2 |
cosmetics2030248_makarova
| 1 |
All the corrections are highlighted in red. The “galenic” was mentioned in the article due to physician Claudio Galeno, which created the first creams formulations. Therefore we decided to change galenic to cosmetic.
| 2 | 1 |
We recommend “galenic” can be changed to herbaceous, because galenic is not usually to see for non-professional fields or general cosmetics reader.
| 1 | 2 |
cosmetics2030248_makarova
| 1 |
Inserted parenthesis at line 132 (Tpeak = 48.17 °C).
| 2 | 1 |
Less a parenthesis at line 132.
| 1 | 2 |
cosmetics2030248_makarova
| 1 |
Corrected at line 160 “Figureure” to “Figure”.
| 2 | 1 |
At line 160, “Figureure” should be “Figure.”
| 1 | 2 |
cosmetics2030248_makarova
| 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.