BRG1 Is Dispensable for Sertoli Cell Development and Functions in Mice
Round 1
Reviewer 1 Report
Major comment
1) In Figure 1, authour need to co-staining BRG1 with germ cell marker (GCNA or VASA etc..)
Generally, male germ cell enter meiosis around 8 dpp. BRG1-positive cells were which stage of germ cell
spermatocyte? unddiferentiated germ cell or differentiated germ cell? entire germ cell?
please clarify with this issue.
2) Line 167, authour said " did not observed obvious changes in the number of Sertoli cells ~"
authour did not count the numbers of sertoli cells Brg1 conditional KO and WT.
3) In Figure 4, add immunoblotting data (SYCP3, TRA98, TNP1 etc) can improve quality of paper
Transcirption levels of germ cell marker such as VASA, PLZF, GFRa1 and Protamin experiment can improve quality of paper as well.
4) Figure 4E, SYCP3-positive cells seems more in the controls than cKO in image
Count of SYCP3-positive cells in tubules experiemnt will clarify this.
In addition, author need to cont of WT1 or Sox9-postiive sertoli cells to confrim change of sertoli cell number or proliferating
in KO mice.
5) It has been reproted that roles of Brg1 and Brm in spermatogenesis from Jianguan Wang et al. (Biol Reprod 2012, PMID: 22495890)
The author should clearly state the reason for why this study need in Sertoli population (introduction section)
Minor comment
1) why don`t you combine figure 5 and Figure S4. please combine as Figure 5
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this manuscript the authors Wang et al. examine effects of conditional knockouts of the epigenetics gene Brg1 in mouse Sertoli cells. Perhaps surprisingly, the authors determine that there was little or no effect on fertility or germ cell function from loss of Brg1 in Sertoli cells. The authors show that BRG1 protein is expressed in both Sertoli and germ cells in mouse testis. While using Amh-Cre mice from Jackson Laboratory, they noted that they did not see complete deletion of Brg1 from Sertoli cells. However when using Amh-Cre mice from a second source that was not named (EMMA?), they noted more complete deletion of Brg1. This is a useful observation that I think the authors should highlight in the manuscript. However, the authors did not make clear until the last paragraph of the Discussion what they meant by "two different Amh-Cre lines." Therefore, I was very confused as to what the authors were doing until I had re-read the manuscript. I recommend clarifying this issue (see Specific Comments).
Overall, the manuscript was well-written and clear. But the confusion with the two different Amh-Cre mouse lines made that part unclear. I highly recommend the authors explain the differences earlier in the manuscript and clarify why they essentially did the conditional knockout experiments twice.
Specific Comments:
1. Page 1, lines 14-24 (Abstract) and page 2, lines 72-75 (Introduction) -- The potential problems with Amh-Cre mice from JAX Laboratory and your solution using a different mouse line (EMMA) will be of great interest to workers in the field so that they can avoid this problem in the future. Please mention these findings in the Abstract and in the last paragraph of the Introduction.
2. Page 3, lines 91-92 (Results) -- As above, please explain that two Amh-Cre mouse lines were used in different parts of this study and distinguish the two lines clearly. For example, “Amh-Cre (Jackson)” and “Amh-Cre (EMMA).”
3. Page 4 (Figure 2A, B) and page 5 (figure legend) -- In the legend to this figure, please explain how this was done. Indicate that genomic PCR was used on genomic DNA from tail snips. Indicate on 2A where PCR primers were placed for the PCR.
4. Page 5 (figure legend 2B) -- The Authors state here that this gel indicates “Homozygous deletion of floxed region….” That cannot be true. The deletion would occur only in the cell types in which Amh-Cre is expressed, presumable Sertoli cells. If I understand correctly, this gel indicates that heterozygous Brg1(F/+) mice were mated to produce Brg1(F/F) offspring, and these offspring were used in subsequent experiments. Please clarify.
5. Page 5, lines 129-130 (Results 2.3) -- The Authors state, “At 42 dpp, the first wave spermatogenesis is completed in mice and Sertoli cells cease proliferation.” While it is correct that the first wave of spermatogenesis is probably complete at 42 dpp, it is not correct that Sertoli cells cease proliferation at that stage. More likely, Sertoli cell proliferation has concluded much earlier in mice. Please clarify.
6. Page 7, line 162-163 (Results 2.4) -- The Authors state “we crossed Brg1 conditional knockout mice (Brg1f/f) with another Amh-Cre mice, in which CRE expression starts at 14 dpc.” This statement is confusing because it was not made clear earlier that two different Amh-Cre mouse strains were used in this study and the results compared. Please clarify here (A) which Amh-Cre line was used previously (presumably the Amh-Cre Jackson line), (B) which Amh-Cre line was used here (presumably the Amh-Cre EMMA line), and (C) why the two different lines were used (presumably because the Jackson line did not efficiently delete Brg1 expression in your mice).
7. Page 7, line 162-163 (Results 2.4) -- The Authors state “we crossed Brg1 conditional knockout mice (Brg1f/f) with another Amh-Cre mice, in which CRE expression starts at 14 dpc.” Are the Authors claiming that in the Amh-Cre (Jackson) line, CRE expression does not begin at 14 dpc? I think that expression of Amh-Cre should be similar in both Amh-Cre (Jackson) and Amh-Cre (EMMA). Do the Authors have a citation to indicate differently? Please clarify.
8. Page 7, lines 174-188 -- I think that this result is more interesting than the Authors are claiming, not because there is no effect of Brg1 deletion on spermatogenesis, but because the two different Amh-Cre lines gave different results. Further, the Authors are not explaining well why they did the conditional knockouts twice. These results will be of use to other workers using Amh-Cre cKO mice. Please clarify why the two experiments were done and how they tested the efficiency of Brg1 deletion in the two strains.
9. Page 7 and Figure S3A -- Figure S3A shows quantitative RT-PCR for expression of several mRNAs, comparing germ cell and Sertoli cell RNA. What was the source of the germ cell RNA?
10. Page 10, lines 225-227 -- As discussed above, please clarify what was different about the two Amh-Cre mouse lines, indicate their names and sources, and state how this tested the efficiency of deletion of Brg1 in Sertoli cells.
11. Page 11, line 259 -- This is the only place in the text where the Authors “Amh-Cre (EMMA)” indicate that there was a second Amh-Cre line from a different source (“EMMA,” which is not explained). Please clarify.
12. Page 11, lines 277-279 -- Are the Amh-Cre (EMMA) mice available to researchers outside China? These might be useful reagents for researchers in the Sertoli cell field.
13. Page 11, lines 281-283 (Materials and Methods) -- The Authors state “male conditional Brg1 knockout mice (Brg1f/f) were bred with female Amh-Cre; Brg1f/f, and the progeny mice were genotyped with PCR analyses.” Did the Authors expect to determine whether the deletions had occurred using tail snip genotyping? Please explain.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Wang et al. have studied the importance of the transcription activator BRG1 for the development of Sertoli cells and their functioning during spermatogenesis in mice. Although somatic Sertoli cells are essential for testis formation and spermatogenesis, peculiarities of gene expression regulation in these cells underlying their unique properties are not well understood. On the other hand, the germline-specific expression of BRG1 ATPase was previously shown to be indispensable for spermatogenesis and male fertility. In the present study, previously generated transgenic mouse lines were utilized to conditionally knockout (cKO) Brg1 in Sertoli cells using two slightly different Cre drivers. Based on the obtained results, the authors came to the conclusion that BRG1 is dispensable for Sertoli cell development and functioning. This is an interesting and novel finding. However, I think that these conclusions are not well justified, as not all necessary control experiments were performed. In addition, the content of the manuscript seems to be redundant and requires some restructuring.
MAJOR comments:
1) The efficiency of the Brg1 cKO in Sertoli cells was estimated primarily by the immunohistofluorescence (IHF) analysis, which may not be sensitive enough especially when studying enzymes. Indeed, the RT-PCR data shown in Figure S3 indicate that even in the case of “good” Amh-Cre driver (EMMA), which ensures “efficient”/“complete” Brg1 deletion in Sertoli cells, the Brg1 gene expression was reduced only to about 30% compared to the control. This suggest that substantial amount of BRG1 protein is still present in Sertoli cells, although IHF analysis does not detect it. The important question is whether the reduction to 30% of the BRG1 protein level is equal to the complete loss of this protein or not. I doubt it. In addition, the mRNA levels do not directly correspond to the protein levels, which also depend on the protein half time. Therefore, the amount of BRG1 protein in sorted cKO Sertoli cells should be somehow analyzed, for example, by Western blot analysis. Next, depending on the results, the conclusion about complete Brg1 cKO should be most probably corrected.
2) The authors describe in detail very similar results they obtained using (i) Amh-Cre driver from Jackson Laboratory that provided “partial” Brg1 deletion and (ii) Amh-Cre driver from Shanghai Model Organisms Center (EMMA) that provided “efficient”/“complete” Brg1 deletion. Undoubtedly, it is a good practice to verify the results by using different drivers for the same cell type. However, due to the partial activity of the first driver, the corresponding results obtained are not really conclusive. Therefore, it would be much more logical to show only the data obtained with the good EMMA driver and just mention that very similar results were obtained with another less efficient driver (all the appropriate data should be moved to Supplementary materials).
3) I have also to note that the difference in the behavior of two Amh-Cre drivers is an interesting finding. However, the authors did not investigate the reasons of this difference and just provided some speculations about it. Actually, these two drivers have not so much in common besides their name Amh-Cre. First, the different Amh gene promoter fragments as well as other different regulatory elements were used to construct them. Second, these drivers were integrated at different genomic locations being thus subjected to chromatin position effects. As far as I understand, the genomic location of the less efficient driver obtained from Jackson Laboratory is not even known. Altogether, it might be quite possible that this inefficient driver is not transcriptionally active in a subset of Sertoli cells. Such possibility should be at least mentioned in the manuscript.
Also, if the authors decide to keep the story about both drivers unchanged (see the previous point), then they have to characterize further the Amh-Cre driver from Jackson Laboratory. Specifically, the localization of CRE recombinase in Sertoli cells can be studied by IHF (using commercially available antibodies against this enzyme). If CRE will be found in all Sertoli (WT1+) cells, this would support the hypothesis about poor accessibility of loxP sites in some cells.
4) I missed the logical link between BRG1 and Sertoli cells in the Introduction. Why this particular protein was chosen for the analysis among hundreds of other chromatin proteins? Is it known to be involved (in some tissue(s)) in regulation of genes that are highly expressed in Sertoli cells?
MINOR comments:
1) The title could be slightly extended to indicate that not only the development, but also functioning of Sertoli cells is not affected upon (complete or partial?) loss of BRG1.
2) I did not understand why authors sometimes say “BRG1 proteins” or “EYFP proteins”. See, for example, lines 65, 104 and 248.
3) The gene and transgene names should be always written in italic. See, for example, “Brg1” in lines 90 (there should be also “Brg1” instead “BRG1”), 109, 149 and 190; “EYFP” in line 106 and in the legend of Figure S1B; “Amh-Cre” in Figure 1A and line 109 (there should be also “Amh” instead “AMH”); “Cre” in line 268 (there should be also “Cre” instead “CRE”).
4) The extra space should be deleted before “f/f” in line 96.
5) There is a mistake in the description of the Rosa26-EYFP reporter transgene (lines 101 and 106): there is a strong transcriptional stop sequence (triple SV40 polyadenylation sequence) and not “the stop codon” before EYFP coding sequence (see [42]). The same thing should be also corrected in the legend of Figure S1B.
In addition, I did not find mentioning of this reporter construct in 3 out of 4 papers cited in line 281 [39-41].
6) “sequences” should be changed to “sequence” in line 107.
7) The Brg1 gene name should be added to Figure 1A right panel.
8) There is no consistency in writing “f/f” and “f/+”– see, for example, Figure 1B (“F/F” and “F/+”).
9) “Proteins” should be changed to “proteins” in line 114.
10) The speculations in lines 118-122 should be deleted.
11) “of” should be added after “wave” in line 129.
12) Statistical analysis should be added to Figures 1B and S2A.
13) Name of statistical test used should be always indicated in Figure legends.
14) “functions” should be changed to “function” in line 173.
15) “from” should be changed to “in” in lines 174 and 185.
16) References to papers describing SYCP3, TRA98 and TNP1 localization patterns should be added to the sentences in lines 178-183.
17) “Brg1 disruption” should be changed to “Amh-Cre-mediated cKO of Brg1” or “cKO of Brg1 in Sertoli cells”. Also, it is worthwhile to specify the Brm mRNA level reduction (down to about 50%) in this sentence.
18) “BTB formation” should be changed to “BTB functioning” in line 229.
19) “is” should be added before “expressed” in line 247. Also, “that” should be added at the end of this line.
20) “Immunohistofluorensce” should be corrected to “Immunohistofluorescence” in line 296.
21) What is “H&E” in line 300?
22) There is no consistence in writing pH values: for example, “pH=5” in lines 291-292 and “pH 6.0” in line 302.
23) It is not clear what is the genotype of samples shown at the left in all subpanels (except the one at the bottom right) of Figure S1A.
24) “(Amh1 and Wt1)” should be added after “and Sertoli cell genes” in the legend of Figure S3A.
25) The reference gene should be indicated in legend of Figure S3B as it is done in the legend of Figure S4.
26) What are “Amh-Cre1#” and “Amh-Cre2#” in Figure S4?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Revised manuscripts was much improved by author. I am impressed wtih the detailed responsed by the author to each my commnet.
Manuscripts is well written and refrence, and i have no futher major comments.
Reviewer 2 Report
It appears that the Authors have addressed all my comments and have made the manuscript more readable. They have also improved its utility to the germ cell/Sertoli cell community by better explaining their experiments using the two Amh-Cre mouse lines. Well done.
Reviewer 3 Report
The authors have addressed all my concerns and therefore in my opinion the manuscript is suitable for publication.
