Cytochalasin B-Induced Membrane Vesicles from TRAIL-Overexpressing Mesenchymal Stem Cells Induce Extrinsic Pathway of Apoptosis in Breast Cancer Mouse Model

Round 1

Reviewer 1 Report

Technically sound manuscript and with interesting results

Author Response

Thank you for your comments.

Reviewer 2 Report

1-  Please mention to the “methods” that you performed in the study at “Abstract” section completely.

2-  Please mention to “mean ± SD” at “Statistical analysis” section.

3-  You should explain the “limitations” of the study at “Discussion” section.

4-  You should present some suggestions to clarify further investigations regarding the issue in the future at the end of “Discussion” section.

Author Response

Thank you for your valuable comments. We have corrected them point by point within the manuscript accordingly (your comments are in bold text and our responses are in ordinary type):

1) Please mention to the “methods” that you performed in the study at “Abstract” section completely.

We have added the information about all the methods that were used to analyze anti-tumor properties of CIMVs-TRAIL in the in vivo model (Page 1, Lines 18-19).

2) Please mention to “mean ± SD” at “Statistical analysis” section.

It was added to the text (Page 8, Lines 356-357).

3) You should explain the “limitations” of the study at “Discussion” section.

We have noted that our study was carried out on a model of one type of cancer, and the antitumor effect of the obtained vesicles relative to other tumors has yet to be investigated. Moreover, evaluation of the effectiveness of administering various CIMV-TRAIL doses and therapy duration would allow the development of more optimal vesicle-based treatment protocol. Another limitation of our study lies in the fact that vesicles isolated from MSCs contain a large number of biologically active molecules that can also contribute to the tumor progression. This may also be the reason why the introduction of CIMVs-TRAIL had no effect on tumor growth, although it induced apoptosis in the tumor cells. Subsequent investigation aimed at studying the protein and RNA profile of the isolated vesicles will significantly expand the understanding of the biological effects of CIMVs.

We have added appropriate information in the Discussion (Page 19, Lines 776-786).

4) You should present some suggestions to clarify further investigations regarding the issue in the future at the end of “Discussion” section.

We believe that other biologically active molecules contained in the vesicles isolated from MSCs require equally comprehensive investigation than the protein of interest TRAIL. Therefore, further studies will focus on the molecular composition of CIMVs and its contribution to their antitumor activity. The appropriate information was added in the Discussion (Page 19, Lines 782-786).

Reviewer 3 Report

Major Comments:

In this study, the authors examine TRAIL-induced apoptosis in TRAIL-resistant breast cancer cell lines using cytochalasin B-induced membrane vesicles of TRAIL-overexpressing mesenchymal stem cells. However, reviewers have many questions about the results presented by the authors. First of all, the reason for using mesenchymal stem cells for TRAIL-overexpressing cells is unclear. For example, what kind of results can be expected if a TRAIL-overexpressing cell line is established using other cancer cell lines and similar studies are conducted? Next, the authors investigated the differentiation ability of the established TRAIL-overexpressing cell lines, and examined the amount of cytokines produced by the established lines and their parent lines in the culture medium. The authors found that TRAIL-overexpressing cell lines showed reduced production of inflammatory cytokines. The authors see that the expression pattern of marker proteins on secretory vesicles does not differ significantly from that of producing cells, but do these results affect the outcome of TRAIL-induced apoptosis? If so, the authors need to examine and substantiate the causes and mechanisms that affect the results more specifically. However, it is considered unlikely that cytokines secreted in the culture supernatant will contaminate directly into the experimental system. In addition to induction of apoptosis by stimulation of receptors such as DR4 and DR5 by TRAIL ligands on membrane vesicles, some researchers reported various substances encapsulated in TRAIL-expressing membrane vesicles taken up by TRAIL-resistant breast cancer cell lines also induced cell death. This reviewer thinks the authors should also consider the possibility of doing so. This should be clarified by conducting experiments to examine each target.

Minor Comments:

1.     Please also overlay data for native MSCs in Figure 1D. 

2.     Is "MSCs-TRAIL by 3, 2 and 3.2 times" on line 428 not "MSCs-TRAIL by 2.3, 2 and 3.2"?

3.     Lines 458 to 470 show that the cell surface markers of CIMVs secreted from MSCs were particularly reduced in CD29 and CD105 compared to parental MSCs. What does this mean?

4.     Figure 3H on line 472 is incorrect. Please correct to Figure3I.

5.     From line 475 to line 476, it says "the protein was detected on the surface of 7.4 +/- 0.9% CIMVs-TRAIL". Where did this data come from?

6.     Shouldn't the 72 hours listed on line 488 be written as 24 hours or 72 hours?

7.     The contents described in lines 490 to 500 have slightly different numbers, but I think they are duplicates. Please check it.

8.     "The expression of the anti-apoptotic BCL-2 gene mRNA in the same group of animals remained unchanged" on line 570, and "Interestingly, the mRNA level of the BCL-2 gene in the tumor sample injected with native CIMVs was also increased by 1.4" is contradictory. It should be rewritten as a whole.

9.     The mAB on line 596 should be written as mAb.

10.  Is it really necessary to consider what is stated in lines 613-651 of the discussion part of this paper? Cytokines and chemokines produced by MSCs are humoral factors and are therefore unlikely to contaminate the CIMVs used in this experiment. In this paper, which should have examined only the effect of membrane vesicles, I do not think it is necessary to consider the findings of other researchers described here in the results of this experiment.

11.  According to "A number of studies have shown the ability of vesicles from MSCs to suppress tumor progression due to the transfer of microRNA molecules" described in line 732, the contribution of miRNA in CIMVs-TRAIL used in this study should be clarified. It should also be examined whether miR-20a-3p described in Ref.53 was included in the CIMVs-TRAIL used in this study.

Author Response

Thank you for your comments which have helped us to improve our manuscript. We have corrected your comments point by point within the manuscript accordingly (your comments are in bold text and our responses are in ordinary type):

1) Please also overlay data for native MSCs in Figure 1D. 

The histogram of native MSCs was overlaid (Page 9, Figure 1D).

2)  Is "MSCs-TRAIL by 3, 2 and 3.2 times" on line 428 not "MSCs-TRAIL by 2.3, 2 and 3.2"?

Thank you for this comment. It was corrected.

3) Lines 458 to 470 show that the cell surface markers of CIMVs secreted from MSCs were particularly reduced in CD29 and CD105 compared to parental MSCs. What does this mean?

We suggest that such a significant decrease in CD29 and CD105 level on the surface of the vesicles is due to the fact that the density of these markers on the parental cells is comparatively low and some of the vesicles are built from that part of the cell membrane where these receptors are not presented. We have added this information into the Discussion (Page 17, Lines 690-693). 

4) Figure 3H on line 472 is incorrect. Please correct to Figure3I.

It was corrected.

5) From line 475 to line 476, it says "the protein was detected on the surface of 7.4 +/- 0.9% CIMVs-TRAIL". Where did this data come from?

The data were obtained using flow cytometry staining with anti-TRAIL antibody. This information was added in the text (Page 12, Lines 476-477).

6) Shouldn't the 72 hours listed on line 488 be written as 24 hours or 72 hours?

It was corrected.

7) The contents described in lines 490 to 500 have slightly different numbers, but I think they are duplicates. Please check it.

Indeed, the data are almost identical as they are obtained using two different apoptosis detection assays, the APC Annexin V Apoptosis Detection Kit and the Vybrant FAM Caspase-8 Assay Kit. Similar data once again confirm the reliability of the results of the antitumor activity analysis of CIMVs-TRAIL.

8) "The expression of the anti-apoptotic BCL-2 gene mRNA in the same group of animals remained unchanged" on line 570, and "Interestingly, the mRNA level of the BCL-2 gene in the tumor sample injected with native CIMVs was also increased by 1.4" is contradictory. It should be rewritten as a whole.

The expression of the anti-apoptotic BCL-2 gene mRNA remained unchanged in the group of animals, received CIMVs-TRAIL. It was corrected to clarify the meaning (Page 15, Line 572-574).

9) The mAB on line 596 should be written as mAb.

It was corrected.

10) Is it really necessary to consider what is stated in lines 613-651 of the discussion part of this paper? Cytokines and chemokines produced by MSCs are humoral factors and are therefore unlikely to contaminate the CIMVs used in this experiment. In this paper, which should have examined only the effect of membrane vesicles, I do not think it is necessary to consider the findings of other researchers described here in the results of this experiment.

The method of vesicle isolation we use assumes that during the formation of the vesicles, proteins secreted by parental cells are also incapsulated in the CIMVs. This was demonstrated in our previous study of CIMVs containing interleukin 2 (PMID: 33579033). For this reason, we consider it important to mention in the text the pro-tumor effects of cytokines/chemokines the secretion of which was significantly changed in the CIMVs-TRAIL.

11) According to "A number of studies have shown the ability of vesicles from MSCs to suppress tumor progression due to the transfer of microRNA molecules" described in line 732, the contribution of miRNA in CIMVs-TRAIL used in this study should be clarified. It should also be examined whether miR-20a-3p described in Ref.53 was included in the CIMVs-TRAIL used in this study.

We agree that other biologically active molecules contained in the vesicles isolated from MSCs require equally comprehensive investigation than the protein of interest TRAIL. The lack of investigation on the contribution of other proteins and RNAs to the antitumor effect of CIMVs-TRAIL is a limitation of our study, we discussed it in Page 19, Lines 776-786. The molecular composition of CIMVs and their contribution to antitumor activity will be investigated in our further studies.

Round 2

Reviewer 3 Report

First of all, the authors have not responded to the reviewers' major comments. It only addresses minor comments. What should be clarified in this paper is to approach the mechanism of why MSC-derived EVs overexpressing TRAIL can enhance the apoptosis-inducing effect on cancer cells. A similar experiment using lung cancer cell lines has already been reported, so at present, even though the target cancer is different, and there is no novelty in terms of content. As the authors are also aware, it is easy to imagine that miRNAs, cytokines, chemokines, etc. contained in CIMVs act additionally. If so, additional experiments should be conducted to clarify the mechanism even a little. There is no progress in science if we stay in the realm of speculation. This paper cannot be accepted in its current state.

Author Response

Thank you for your review. We find your comments reasonable, but would like to clarify a few key points. First of all, mesenchymal stem cells were chosen for this cell therapy approach, since they have a number of significant properties, in particular, low immunogenicity and the ability to migrate toward tumor foci. However, we have decided to switch to a cell-free system, since the use of MSCs in therapy is associated with the possibility of their malignant transformation.

Secondly, in order to get closer to understanding the mechanism of apoptosis induction in tumor cells, we have analyzed changes in the level of the mRNA of CASP8, BAX and BCL-2 genes and caspase 8 protein. In our opinion, an increase in the expression level of the CASP8 gene and level of activated caspase 8 protein indicates that the induction of apoptosis in tumor cells occurred via the extrinsic pathway of apoptosis. The novelty of our study lies in the use of Cytochalasin B for the isolation of artificial vesicles. Since cytochalasin B-induced membrane vesicles differ from the naturally derived ones, we are still continuing to investigate and compare their properties. The fact that we observed effect of CIMVs on the viability of tumor cells may be mediated not only by the protein profile, which has been changed after genetic modification, and requires more comprehensive study, which cannot fit into the scope of this investigation. However, the aim of the study was to analyze the effect of CIMVs-TRAIL on tumor cells and evaluate whether membrane TRAIL induces apoptosis via an extrinsic pathway. The influence of the properties of cytochalasin B-induced vesicles on tumor cells, including the protein and RNA profile, will be analyzed in more detail in further studies.

Round 3

Reviewer 3 Report

If the authors find novelty and superiority in cytochalasin B induced MVs, they should compare and analyze non-induced data as a control. In this paper, the positive effect of cytochalasin B induced is not understood by the readers because the comparative analysis is limited to TRAIL expression and non-expression. The outline of the paper should be reconsidered.

Author Response

Thank you again for your review and attempt to improve our manuscript. First of all, we would like to point out once again that we are working with “artificial” membrane vesicles, completely different from naturally occurring EVs. As such, it is impossible to use un-induced by Cytochalasin B control because such control does not produce membrane vesicles (besides natural at very low level). Main idea of our approach is to use normal, not transformed/cancer cells as source of cell-free therapeutics. This is a safer approach. And using induction by Cytochalasin B we dramatically, several logs, increase the yield of membrane vesicles production, making the technology industrially attractive. We agree with the reviewer, that additional mechanism studies are needed, however, we consider this manuscript as sort of short-communication, a proof-of-principle study, demonstrating feasibility of our approach. We thank the reviewer one more time for thoughtful comments and will use them in our new grant applications and future studies. However, we don’t have resources at this point to conduct additional in-depth mechanism studies. 

Round 4

Reviewer 3 Report

Although the authors' comments are not entirely incomprehensible, Yuan et al., cited by the authors, collected EVs from MSCTRAIL and control MSC culture supernatants and compared their anti-tumor effects. According to their method, the authors should be able to compare CIMVs and non-induced MVs. With the addition of these data, the superiority of CIMVs should be well understood by the readers, and there is room for consideration in the authors' desired comment that "the analysis of the mechanism will be clarified in the next paper." This reviewer is asking for a Major Revision with additional experiments, not a Minor Revision with comments and minor corrections. As noted above, the authors' inability to recover EVs that other research groups have had success with is far from acceptable. The authors should ask the above group for guidance and overcome it.