The autophagy protein, ATG14 safeguards against unscheduled pyroptosis activation to enable embryo transport during early pregnancy

Reviewer #1 (Public review):

This study by Popli et al. evaluated the function of Atg14, an autophagy protein, in reproductive function using a conditional knockout mouse model. The authors showed that female mice lacking Atg14 were infertile partly due to defective embryo transport function of the oviduct and faulty uterine receptivity and decidualization using PgrCre/+;Atg14f/f mice. The findings from this work are exciting and novel. The authors demonstrated that a loss of Atg14 led to an excessive pyroptosis in the oviductal epithelial cells that compromises cellular integrity and structure, impeding the transport function of the oviduct. In addition, the authors use both genetic and pharmacological approaches to test the hypothesis. Therefore, the findings from this study are high-impact and likely reproducible. However, there are multiple major concerns that need to be addressed to improve the quality of the work.

Reviewer #2 (Public review):

In this manuscript, Popli et al investigated the roles of autophagy related gene, Atg14, in the female reproductive tract (FRT) using conditional knockout mouse models. By ablation of Atg14 in both oviduct and uterus with PR-Cre (Atg14 cKO), authors discovered that such females are completely infertile. They went on to show that Atg14 cKO females have impaired embryo implantation as well as embryo transport from oviduct to uterus. Further analysis showed that Atg14 cKO leads to increased pyroptosis in oviduct, which disrupts oviduct epithelial integrity and leads to obstructive oviduct lumen and impaired embryo transport. The authors concluded that Atg14 is critical for maintaining the oviduct homeostasis and keeping the inflammation under check to enable proper embryo transport.

The authors have barely addressed most of my concerns in this revised version with a few minor issues remaining to be addressed:
(1) The authors tried to address my first concern regarding the statement that "autophagy is critical for maintaining the oviduct homeostasis". The revised statement in Line 53-54 "we report that Atg14-dependent autophagy plays a crucial role in maintaining..." is still not correct. It should be corrected as " we report that autophagy-related protein Atg14 plays a crucial role in maintaining...".
(2) Line 349-351 described 80-90% of blastocysts retrieved from oviducts of cKO mice, which is in consistent with Figure 3B (showing more than 98%).
(3) Line 447, "Fig. 5E" should be Fig. 6A. In addition, grammar error in the next sentence.
(4) In Figure 6D, why the composition of blastocysts in chemical treated group do not add up to 100%.

Reviewer #3 (Public review):

Summary:

The manuscript by Pooja Popli and co-authors tested the importance of Atg14 in the female reproductive tract by conditionally deleting Atg14 use PrCre and also Foxj1cre. The authors showed that loss of Atg14 leads to infertility due to the retention of embryos within the oviduct. The authors further concluded that the retention of embryos within the oviduct is due to pyroptosis in oviduct cells leading to defective cellular integrity. The revised manuscript has included new experimental data (Figs. S2B, 5B, 5C, and S3) that satisfied the concerns of this reviewer. The manuscript should provide important advancement to the field.

Author response:

The following is the authors’ response to the original reviews.

We greatly appreciate the opportunity to submit a revision of our manuscript entitled: "The Autophagy Protein, ATG14 Safeguards Against Unscheduled Pyroptosis Activation to Enable Embryo Transport During Early Pregnancy" by Popli et al. We thank all three Referees for underscoring the importance of our findings as well as the constructive critiques that we used to improve our paper. Most notably, we added the following new data:

· To provide more insight into whether pyroptosis activation occurs distinctly in the oviduct, we looked for GSDMD, (primary executioner of the pyroptosis pathway) expression in the uterus and ovary too. We observed no signs of pyroptosis activation in response to ATG14 loss in either the uterus or ovary of Atg14 cKO mice compared to control ones suggesting that ATG14 plays a distinct role in regulating pyroptosis specifically in the oviduct (Revised Figure 5F).

· To better understand the molecular mechanisms of pyroptosis activation in the oviducts, we examined various key markers of mitochondrial integrity, architecture, and function in control and Atg14 cKO oviducts. Our findings indicate a significant loss of mitochondrial structural and functional integrity, possibly contributing to the embryo retention phenotype via activating the pyroptosis pathway in the oviduct. (Revised Figure 5B & C).

· To address the spatiotemporal and region-specific expression of ATG14 in the oviduct, we performed immunofluorescence analysis and observed the consistent expression of ATG14 in all the cellular compartments of oviducts including ciliary epithelial cells, secretory epithelial cells, and smooth muscle cells. Moreover, the region-specific expression analysis revealed that distinct expression of ATG14 in the ampullary region of cKO mice oviduct helps to preserve its structural integrity. Conversely, its loss in the isthmus region of the oviduct in concordance with active PR-cre activity causes completely distorted epithelial structures with luminal obliteration or narrowing resulting in an unorganized and obstructed lumen leading to embryo retention, suggesting that ATG14 is essential for maintaining the structural integrity of the oviduct (Revised Figure 3F & S2A).

· Considering the expression of PR-cre in the pituitary, which could potentially influence hormonal secretion and ovulation, we evaluated the levels of E2 and P4 during pregnancy. Our findings show that these hormone levels remained unchanged in Atg14 cKO mice, indicating that the absence of ATG14 does not negatively affect the HPG axis or pituitary function (Revised Figure 2F).

· ATG14 is an essential factor for the initiation of autophagy, and its loss can lead to reduced or inhibited autophagic activity. Consistently, we observed elevated levels of LC3b and p62 proteins, two well-known markers of autophagic flux in the oviducts of Atg14-deficient mice implying that loss of ATG14 leads to defective autophagy potentially disturbing the structural integrity of oviductal epithelial cells and impairing embryo transport. (New Supplementary Figure S2B).

Reviewer #1 (Public Review):

This study by Popli et al. evaluated the function of Atg14, an autophagy protein, in reproductive function using a conditional knockout mouse model. The authors showed that female mice lacking Atg14 were infertile partly due to defective embryo transport function of the oviduct and faulty uterine receptivity and decidualization using PgrCre/+; Atg14f/f mice. The findings from this work are exciting and novel. The authors demonstrated that a loss of Atg14 led to an excessive pyroptosis in the oviductal epithelial cells that compromises cellular integrity and structure, impeding the transport function of the oviduct. In addition, the authors use both genetic and pharmacological approaches to test the hypothesis. Therefore, the findings from this study are high-impact and likely reproducible. However, there are multiple major concerns that need to be addressed to improve the quality of the work.

Major comments:

(1) It is interesting that deletion of Atg14 using PgrCre results in pyroptosis only in the oviduct; the authors should speculate/evaluate why the oviduct, but not the uterus or follicles. Is there any cellular specificity that is sensitive to autophagy/pyroptosis in the oviduct but not in other cell types? This has not been evaluated or discussed in the manuscript. Is it possible to include GSDMD IHC for the uterine section to ensure that there was no pyroptosis event in the cKO uteri?

We performed GSDMD IHC and found that, unlike in the oviduct, the cKO uteri and ovaries do not exhibit detectable pyroptosis (Revised Figure 5F). Additionally, we have added text to the discussion section addressing possible reasons for the differential impact of Atg14 loss on pyroptosis along the reproductive tract continuum (Line number: 532-538)

(2) Please include an explanation of how a loss of Atg14, important for the initiation process of autophagy (as indicated in line 88), can lead to pyroptosis. There was some discussion about inflammation. But the connection is still missing.

We thank the reviewer for noting on this. We have now included a possible explanation of how autophagy could impact pyroptosis in the discussion section (Line number: 532-538)

(3) No expression data of ATG14 using IHC/IF analysis were included in the manuscript - this is missing. This is needed and important as the authors found that Foxj1Cre/+; Atg14f/f cKO mice had no fertility defect. Is it possible that ATG14 is not present in the ciliated epithelial cells of the oviduct? In addition, the data in Figure 5B also points to this speculation. This is because the GSDMD (the pyroptosis marker) is only observed in the isthmus region but not the ampulla.

We thank the reviewer for this nice suggestion. We performed the immunofluorescence analysis for ATG14 expression in control and Atg14 cKO oviducts and observed the consistent expression of ATG14 in all the cellular compartments of oviducts including ciliary epithelial cells, secretory epithelial cells, and smooth muscle cells (New Supplementary Figure S2A). We also looked for α-tubulin expressions in the oviduct of Foxj1Cre/+; Atg14 f/f mice and control mice and observed that ciliated epithelial cells that were positive for acetylated α-tubulin staining did not appear to be different in Foxj1Cre/+; Atg14 f/f mice oviduct compared to controls (Revised Figure 4C). However, due to the unavailability of reliable fluorescent-labeled antibodies for both Foxj1 and Atg14, we were unable to conduct the co-localization study as intended. This limitation hindered our ability to precisely determine the spatial overlap of these proteins within the tissue.

(4) In line with the previous comment, is ATG14 present in the human Fallopian tube? If so, which cell type? This needs to be addressed.

Author’s Response: We appreciate the reviewer's valuable suggestion. While we currently lack access to human fallopian tube biopsies, the Human Protein Atlas (https://www.proteinatlas.org/ENSG00000126775-ATG14) demonstrates distinct ATG14 expression in various fallopian tube cell types, with localization in the cytoplasm, membrane, and nucleus.

(5) As PgrCre is also expressed in the pituitary, is it possible that the deletion of Atg14 using PgrCre would affect pituitary function – hence a change in the FSH/LH secretion that subsequently affects ovulation? Although the uterine and ovarian histology in the Atg14 cKO looks similar to the controls, is it possible that cyclicity is also affected? The authors should evaluate whether the estrous cycle takes place regularly.

Author’s Response: Thank you for the insightful comment. However, evaluating the estrous cycle requires significant time and effort and is beyond the scope of the current manuscript. Nonetheless, we have now shown that both P4 and E2 levels were not altered in Atg14 cKO mice, indicating that the loss of Atg14 did not adversely impact the HPG axis, and by extension, pituitary function (Revised Figure 2F).

(6) The number of total embryos/oocytes in the cKO compared to the control has not been evaluated - this data must be included. Do the changes in autophagy in Atg14 cKO affect preimplantation embryo development? Please categorize the embryos found in the oviduct/uterus in both genotypes. i.e., % blastocyst, % morula, % developmentally delayed, % non-viable etc. It would be interesting to evaluate if the oviduct with heavy pyroptosis can support preimplantation embryo development.

Author’s Response: We thank the reviewer for this nice suggestion. We categorized the embryos into different categories as suggested and included the data (Revised Figure 3C and Figure 6D).

(7) It is unclear why the superovulation+mating experiment (Figure 3C) was performed. Please provide justification. Why was the data from natural mating (Figure 3A) insufficient?

Author’s Response: In Figure 3C, superovulation was employed to complement the natural mating studies and to provide stronger evidence for the embryo retention phenotype observed in the oviduct.

(8) In lines 297-298, the conclusion that "ATG14 is required for P4-mediated but not for E2-mediated actions during uterine receptivity" is not entirely correct. This is because the authors also observed that the downregulation of MUC1 (E2-target protein) is absent in the PgrCre/+;Atg14f/f cKO female uteri.

We thank the reviewer for noting this. We detected more E2-induced targets in D-4 pregnant uterine samples and found no change in their expression in response to Atg14 depletion in cKO females (Revised Figure 2E).

(9) Figure 3D: Please include an image that also represents the ampulla region. All images are from the isthmus region. It would be informative to see if the loss of cell boundaries also takes place at the ampulla region in the cKO oviduct.

We thank the reviewer for this nice suggestion. We included the ampulla section from the cKO and control female oviducts (Revised Figure 3F). As PR-cre activity is limited to isthmus only [1, 2], we did not see any structural abnormality in ampulla sections of cKO oviducts.

(10) Figure 3E: Please indicate which region the TEM was performed. Isthmus? Ampulla? Were the changes in mitochondrial phenotype observed across all oviductal regions?

The TEM imaging was performed by the WashU Core services. Although we clearly mentioned the core person to look into the isthmus region only, we are not sure if they accurately follow the instructions.

(11) Figure 4B; the evaluation of FOXJ1 IHC. The authors need to include sections that also have an ampulla region-especially in the cKO. In addition, it is misleading to state that there were fewer FOXJ1+ cells (line 361) in the cKO if the region being evaluated is the isthmus (which has a lot fewer ciliated epithelial cells in general) while the control image showed an ampulla where the abundancy of ciliated epithelial cells (FOXJ1+) is higher than that of the isthmus. The authors also need to include a higher resolution image (a zoom-in at the ciliated epithelial cells with FOXJ1+ signal) as well as the quantification of FOXJ1+ cells.

We appreciate the reviewer for the suggestion. In Figure 4A, we have already shown the ampulla region from both control and cKO oviducts, wherein alpha-tubulin staining was evident in both oviducts.

We agree with the reviewer that the isthmus usually has fewer ciliary epithelial cells than the ampulla, however, as illustrated in Figures 4A and 4B, Atg14 depletion causes a marked disruption of structural integrity with loss of cell boundaries specifically in the isthmus, which is far more pronounced than in the ampulla. One reason for this is the reported Pgr Cre activity, which is much more robust in the isthmus than in the ampulla [1, 2] . This disruption leads to the substantial loss of both ciliated and secretory cells, compromising the epithelial architecture to such an extent that it is impossible to accurately quantify the Foxj1 signal as can be seen in higher resolution images in New Supplementary Figure S3.

For more clarity, we modified the statement in the revised file (Line Number: 393-396)

(12) All IHC/IF and embryo images need to include the scale bars.

We thank the reviewer for this suggestion. We now included the scale bar in all the images.

(13) Figure 5H: although IL1B is being discussed, there was no data in this study to support the figure.

In Figure 5H, IL1B is presented as part of the pyroptosis signaling pathway. As we have already shown other key executioners of this pathway: Caspase 1 and GSDMD, we believe that additional IL1B data would not provide new insights beyond what has already been shown.

Minor comments:

(1) Please include n (sample size) for all data, including the histology image in the figure legends for all studies.

We now included the sample size in figure legends for all data shown in the manuscript.

(2) Line 32, did the authors mean to say, "Self-digestion of..." instead of "Self-digestion for..."?

In Line 32, we meant, “Cellular self-digestion for female reproductive tract functions”. We have now corrected the statement.

Fig. 1A - please include negative control.

We included the negative control (Revised Figure 1)

(3) Figure 1E left panel and Figure 4C - please label "Average no. of pups/female/litter" as each female has more than one litter over her reproductive lifespan. If the authors represent pups/females, then the number should be accumulative in the range of 35-40pups/females in the control group.

We thank the reviewer for noting this. We now corrected the label in both Revised Figure 1E and Revised Figure 4E.

(4) Line 273: please remove "& F" as there is no Figure F in the image.

We removed “&F” from the Line 273.

(5) The presence of CL is not always indicative of normal hormonal levels; therefore, the authors should include the measurement of progesterone levels at 3.5 dpc in the cKO compared to the control group. Hormonal regulation is also crucial for embryo transport.

We thank the reviewer for this suggestion. We measured not only P4 but also E2 levels in D4 pregnant females and found no significant difference in their levels compared to corresponding controls (Revised Figure 2F).

(6) Figure 2A shows that KRT expression is not present in the control uteri. Although the KRT8 levels may have decreased at 4 dpc, they should be present (see Figure S2A).

We observed no decrease in KRT expression in control uteri on 5 dpc. We included better-resolution images for KRT expression (Revised Figure 2A).

(7) The dotted white lines in Figure 2A are too thick. It's difficult to see the Ki67 positive signal in the luminal epithelial cells. Please also add a quantitative analysis of Ki67+ cells in the luminal epithelium vs. stromal cells.

We now corrected the dotted lines in Revised Figure 2B. However, as the Ki-67 proliferation is evident in the representative images, we believe quantification analysis will not add anything new to the existing conclusion.

(8) Figure 2D - the y-axis mentions the weight ratio. However, the figure legend describes the transcript levels of Atg14 - please correct this.

We corrected the label in the revised manuscript.

(9) Line 294 - Please correct Figure 2C to Figure 2B.

We corrected it.

(10) Line 308 - Please correct Figure 2E to Figure 2F.

We corrected it.

(11) Line 310 - Please correct Figure 2F to Figure 2G.

We corrected it.

(12) Line 311 - Please correct Figure 2F to Figure 2G.

We corrected it.

(13) Information in Figure S2A and S2B should be included in the main figure.

We thank the reviewer for this nice suggestion. We now included the figures S2A and S2B in the main figure (Revised Figure 2C & D).

(14) Figure 3C - due to a lot of cellular debris after flushing, it's difficult to see. But it seems like there are secondary follicles in the flushing of control oviducts - this is highly unlikely. This could be due to an artifact of an accidental poking of the ovaries during collection.

We agree with the reviewer. It might be due to the unintentional poking of the ovaries. We will take extra care in future experiments to avoid this and ensure clean flushing to prevent any confusion from debris or artifacts.

(15) Figure 2B and Figure 3D signals from DAPI are missing - it's black with no blue signal. This could be the data loss during file compression for manuscript submission.

We included better-resolution pictures for the DAPI signal in Revised Figure 2B & Figure 3F.

(16) Explain why some embryos in the cKO make it to the uterus when the females are superovulated.

It might be due to the heightened hormonal stimulation provided by the superovulation which could facilitate the movement of some embryos through the oviduct despite any defects or abnormalities caused by the loss of ATG14 in the oviduct.

Reviewer #2 (Public Review):

Summary:

In this manuscript, Popli et al investigated the roles of the autophagy-related gene, Atg14, in the female reproductive tract (FRT) using conditional knockout mouse models. By ablation of Atg14 in both oviduct and uterus with PR-Cre (Atg14 cKO), the authors discovered that such females are completely infertile. They went on to show that Atg14 cKO females have impaired embryo implantation and uterus receptivity due to impaired response to P4 stimulation and stromal decidualization. In addition to the uterus defect, the authors also discovered that early embryos are trapped inside the oviduct and cannot be efficiently transported to the uterus in these females. They went on to show that oviduct epithelium in Atg14 cKO females showed increased pyroptosis, which disrupts oviduct epithelial integrity and leads to obstructive oviduct lumen and impaired embryo transport. Therefore, the authors concluded that autophagy is critical for maintaining the oviduct homeostasis and keeping the inflammation under check to enable proper embryo transport.

Strengths:

This study revealed an important and unexpected role of the autophagy-related gene Atg14 in preventing pyroptosis and maintaining oviduct epithelial integrity, which is poorly studied in the field of reproductive biology. The study is well designed to test the roles ofATG14 in mouse oviduct and uterus. The experimental data in general support the conclusion and the interpretations are mostly accurate. This work should be of interest to reproductive biologists and scientists in the field of autophagy and pyroptosis.

Weaknesses:

Despite the strengths, there are several major weaknesses raising concerns. In addition, the mismatched figure panels, the undefined acronyms, and the poor description/presentation of some of the data significantly hinder the readability of the manuscript.

(1) In the abstract, the authors stated that "autophagy is critical for maintaining the oviduct homeostasis and keeping the inflammation under check to enable embryo transport". This statement is not substantiated. Although Atg14 is an autophagy-related gene and plays a critical role in oviduct homeostasis, the authors did not show a direct link between autophagy and pyroptosis/oviduct integrity. In addition, the authors pointed out in the last paragraph of the introduction that none of the other autophagy-related genes (ATG16L, FIP200, BECN1) exhibited any discernable impact on oviduct function. Therefore, the oviduct defect is caused by Atg14 specifically, not necessarily by autophagy.

We thank the reviewer for noting this. We corrected the statement in the revised manuscript (Line number: 53-54).

(2) In lines 412-414, the authors stated that "Atg14 ablation in the oviduct causes activation of pyroptosis", which is also not supported by the experimental data. The authors did not show that Atg14 is expressed in oviduct cells. PR-Cre is also not specific in oviduct cells. It is possible that Atg14 knockout in other PR-expressing tissues (such as the uterus) indirectly activates pyroptosis in the oviduct. More experiments will be required to support this claim. In line with the no defect when Atg14 has knocked out in oviduct ciliary cells, it will be good to use the secretory cells Cre, such as Pax8-Cre, to demonstrate that Atg14 functions in the secretory cells of the oviduct thus supporting this conclusion.

We now included the ATG14 expression data in the oviduct (New Supplementary Figure S2A). Consistent with previous studies reporting PR-cre activity in the isthmus [1, 2] , we observed that Atg14 depletion was more pronounced in the isthmus compared to the ampulla. However, generating a secretory Pax-8 cell Cre mice model will require a substantial amount of time and effort, and we respectfully note that this is beyond the scope of the current manuscript.

(3) With FOXJ1-Cre, the authors attempted to specifically knockout Atg14 in ciliary cells, but there are no clear fertility and embryo implantation defects in Foxj1/Atg14 cKO mice. The author should provide verification data to show that Atg14 had been effectively depleted in ciliary cells if Atg14 is normally expressed.

We understand the reviewer’s concern. We included new data for ATG14 expression in control and Atg14 cKO mice oviducts (New Supplementary Figure S2A). However, due to the unavailability of reliable fluorescent-labeled antibodies for both Foxj1 and Atg14, we could not conduct the co-localization studies as intended, and this limitation hindered our ability to precisely determine the spatial overlap of these proteins within the oviduct. Nonetheless, Foxj1-cre is a widely used mice model with reported cre-activity in ciliary epithelial cells including oviduct tissues [3]. Given the widespread expression of ATG14 in all the ciliary and secretory cells (New Supplementary Figure S2A) and distinct FOXJ1 expression in the oviduct (New Supplementary Figure S3), we are confident that Atg14 is deleted in the ciliary epithelial cells of Foxj1/Atg14 cKO mice oviducts.

(4) In lines 307-313, the author tested whether ATG14 is required for the decidualization of HESCs. The author stated that "Control siRNA transfected cells when treated with EPC seemed to change their morphological transformation from fibroblastic to epithelioid (Fig. 2E) and had increased expression of the decidualization markers IGFBP1 and PRL by day three only (Fig. 2F)". First, the labels in Figure 2 are not corresponding to the description in the text. Second, the morphology of the HESCs in the control and Atg14 siRNA group showed no obvious difference even at day 3 and day 6. The author should point out the difference in each panel and explain in the text or figure legend.

Decidualization is a post-implantation event, whereas our study primarily focuses on pre-implantation events in the oviduct. Therefore, we have removed all data related to human and mouse decidualization to enhance the clarity and precision of our study.

(5) In lines 332-336, the authors pointed out that the cKO mice oviduct lining shows marked eosinophilic cytoplasmic change, but there's no data to support the claim. In addition, the authors further described that "some of the cells showed degenerative changes with cytoplasmic vacuolization and nuclear pyknosis, loss of nuclear polarity, and loss of distinct cell borders giving an appearance of fusion of cells (Fig. 3D)". First, Figure 3D did not show all these phenotypes, and it is likely a mismatch to Figure 3E. Even in Figure 3E, it is not obvious to notice all the phenotypes described here. The figure legend is overly simple, and there's no explanation of the arrowheads in the panel. More data/images are required to support the claim here and provide a clear indication and explanation in the figure legend.

Dr. Ramya Masand, Chief pathologist in the Pathology Department at the Baylor College of Medicine, and a contributing author, assessed the H&E-stained oviduct sections from control and cKO mice. We have now included a new Supplementary Figure S3 with previous representative H&E images that depict the cellular alterations described in lines 332–336.

(6) In lines 317-325, it is rather confusing about the description of the portion of embryos from the oviduct and uterus. In addition, the total number of embryos was not provided. I would recommend presenting the numerical data to show the average embryos from the oviduct and uterus instead of using the percentage data in Figures 3A and 5G.

We thank the reviewer for this nice suggestion. We calculated the average number of embryos and found no difference in the number of embryos recovered from cKO or polyphyllin-treated pregnant mice at 4 dpc compared to their controls. (New Supplementary Figure S4A & B).

(7) In lines 389-391, authors tested whether Polyphyllin VI treatment led to activated pyroptosis and blocked embryo transport. Although Figures 5F-G showed the expected embryo transport defect, the authors did not show the pyroptosis and oviduct morphology. It will be important to show that the Polyphyllin VI treatment indeed led to oviduct pyroptosis and lumen disruption.

We performed the GSDMD staining IHC in Polyphyllin VI or vehicle-treated mice oviducts and observed elevated GSDMD expression with Polyphyllin V (New Figure 6E). However, no significant lumen disruption was detected, which may be attributed to the short-term exposure of the oviducts to pyroptosis induction, in contrast to the more pleiotropic effects observed in genetically induced models. Nonetheless, this observation clearly indicates that unscheduled or unwarranted activation of pyroptosis impedes embryo transport.

(8) In line 378, it would be better to include a description of pyroptosis and its molecular mechanisms to help readers better understand your experiments. Alternatively, you can add it in the introduction.

We thank the reviewer for this nice suggestion. We included literature on the pyroptosis pathway in the introduction section (Line Number: 105-118).

(9) Please make sure to provide definitions for the acronyms such as FRT, HESCs, GSDMD, etc.

We added definitions for the acronyms such as FRT, HESCs, and GSDMD used in the study.

(10) It is rather confusing to use oviducal cell plasticity in this manuscript. The work illustrated the oviducal epithelial integrity, not the plasticity.

We thank the reviewer for the suggestion. We have revised the manuscript accordingly to ensure clarity and precision in describing the oviductal epithelial structural changes observed in the absence of ATG14.

A few of the additional comments for authors to consider improving the manuscript are listed below.

(1) Some of the figures are missing scale bars, while others have inconsistent scale bars. It would be better to be consistent.

We now included the scale bars in all images.

(2) On a couple of occasions, the DAPI signal cannot be seen, such as in Figure 2B and Figure 3D.

We now included better-resolution images for the DAPI signal in all fluorescent images shown in the revised manuscript.

(3) Overall, the figure legends can be improved to provide more detailed information to help the reader to interpret the data.

We included additional details in all the figure legends in the revised manuscript.

(4) In Figure 2D, the Y-axis showed the stimulated/unstimulated uterine weight ratio, why did the author put "Atg14" at the top of the graph? At the same time, the X-axis title is missing in Figure 2D.

We apologize for the typo error. We removed “Atg14” from the top of the graph and included the X-axis title in the revised manuscript.

(5) In the left panel of Figure 2G, "ATG14" at the top should be "Atg14" to be consistent.

In Figure 2G, we are representing “ATG14” according to human gene annotation.

(6) In line 559, there miss "(A)" in front of Immunofluorescence analysis of GSDMD.

We thank the reviewer for noting this. We corrected it in the revised manuscript.

Reviewer #3 (Public Review):

Summary:

The manuscript by Pooja Popli and co-authors tested the importance of Atg14 in the female reproductive tract by conditionally deleting Atg14 using Pr Cre and also Foxj1cre. The authors showed that loss of Atg14 leads to infertility due to the retention of embryos within the oviduct. The authors further concluded that the retention of embryos within the oviduct is due to pyroptosis in oviduct cells leading to defective cellular integrity. The manuscript has some interesting findings, however there are also areas that could be improved.

Strengths:

The importance of Atg14 and autophagy in the female reproductive tract is incompletely understood. The manuscript also provide spatial evidence about a new mechanism linking Atg14 to pyroptosis.

We thank the reviewer for the positive statements and constructive comments on our manuscript.

Weaknesses:

(1) It is not clear why the loss of Atg14 selectively induces Pyroptosis within oviduct cells but not in other cellular compartments. The authors should demonstrate that these events are not happening in uterine cells.

We thank the reviewer for this nice suggestion. We performed GSDMD IHC and found that, unlike in the oviduct, the cKO uteri and ovaries do not exhibit detectable pyroptosis (Revised Figure 5F). Additionally, we have added text to the discussion section addressing possible reasons for the differential impact of Atg14 loss on pyroptosis along the reproductive tract continuum (Line number: 532-538)

(2) The manuscript never showed any effect on the autophagy upon loss of Atg14. Is there any effect on autophagy upon Atg14 loss? If so, does that contribute to the observation?

We thank the reviewer for the nice suggestion. We found LC3b and p62 protein levels, two well-known markers of autophagic flux are elevated due to Atg14 loss in the oviduct (New Supplementary Figure S2B). Since, p62 accumulation is an indicative of the reduced autophagic flux [4], we posit loss of Atg14 results in defective autophagy in the oviduct. Importantly, this defective autophagy adversely impacted the structural integrity of oviductal epithelial cells, causing impairment in embryo transport.

(3) It is not clear what the authors meant by cellular plasticity and integrity. There is no evidence provided in that aspect that the plasticity of oviduct cells is lost. Similarly, more experimental evidence is necessary for the conclusion about cellular integrity.

We thank the reviewer for the suggestion. We have revised the text for clarity and precision in describing the oviductal epithelial structural changes observed in the absence of ATG14. To avoid ambiguity, we have removed the term "cellular plasticity." We have already provided extensive evidence, including multiple H&E stains and immunofluorescence analyses for KRT8 and smooth muscle actin to illustrate cellular integrity in both control and cKO oviducts. However, we respectfully believe that performing additional experiments on cellular integrity would not contribute further to the conclusions already drawn.

(4) The mitochondrial phenotype shown in Figure 3 didn't appear as severe as it is described in the results section. The analyses should be more thorough. They should include multiple frames (in supplemental information) showing mitochondrial morphology in multiple cells. The authors should also test that aspect in uterine cells. The authors should measure Feret's diagram. Diff erence in membrane potential etc. for a definitive conclusion.

We appreciate the reviewer’s suggestion. We carried out the TOM20 (mitochondrial structural marker) and cytochrome C (mitochondrial damage and cell death marker) immune-colocalization study and found loss of TOM20 signal with concomitant cytochrome c leakage into the peri-nuclear space (Revised Figure 5B). Additionally, we also observed reduced expression of mitochondrial structural and functional markers by qPCR analysis (Revised Figure 5C). However, we respectfully argue that conducting membrane potential studies on murine oviducts is extremely complex and is beyond the scope of this study.

(5) The comment that the loss of Atg14 and pyroptosis leads to the narrowing of the lumen in the oviduct should be experimentally shown.

We have now included a New Supplementary Figure S3 with representative previous immunofluorescence images that clearly show the narrowing of the lumen with Atg14 loss in the oviduct.

(6) The manuscript never showed the proper mechanism through which Atg14 loss induces pyroptosis. The authors should link the mechanism.

We respectfully disagree with the reviewer on this point. We have provided substantial evidence regarding the cellular mechanisms through which the loss of Atg14 may lead to the activation of pyroptosis as outlined below:

(1) Cellular Changes: Loss of ATG14 in the oviduct results in cellular swelling and the formation of fused membranous structures, which are characteristic features of pyroptosis activation.

(2) Expression of Key Pyroptosis Proteins: We observed an induced expression of GSDMD and Caspase-1, primary executioners of the pyroptotic pathway, in response to Atg14 loss.

(3) Inflammatory Markers: Elevated levels of inflammatory markers such as TNF-α and CXCR3 were detected, both of which are known to promote pyroptosis [5, 6].

(4) Mitochondrial Damage: We have added new data demonstrating disrupted colocalization of TOM20 (a mitochondrial structural marker) and Cytochrome c (a cell death marker), resulting in Cytochrome c leakage into the perinuclear space (Revised Figure 5B). Additionally, qPCR analysis revealed reduced expression of mitochondrial structural and functional markers in cKO oviduct tissues (Revised Figure 5C).

Based on these evidences, we can clearly say that Atg14 has some direct or indirect link to inflammasome activation. However, understanding the complex rheostat between the Atg14-mediated autophagy and inflammation regulatory axis will necessitate future studies employing sophisticated models, such as combined knockout mice where ATG14 is deleted alongside key inflammatory regulators (e.g., NLRP3, GSDMD, or CASPASE-1). These dual knockout models could provide crucial insights into how ATG14 modulates inflammatory pathways.

References:

(1) Herrera, G.G.B., et al., Oviductal Retention of Embryos in Female Mice Lacking Estrogen Receptor alpha in the Isthmus and the Uterus. Endocrinology, 2020. 161(2).

(2) Soyal, S.M., et al., Cre-mediated recombination in cell lineages that express the progesterone receptor. Genesis, 2005. 41(2): p. 58-66.

(3) Zhang, Y., et al., A transgenic FOXJ1-Cre system for gene inactivation in ciliated epithelial cells. Am J Respir Cell Mol Biol, 2007. 36(5): p. 515-9.

(4) Mizushima, N., T. Yoshimori, and B. Levine, Methods in mammalian autophagy research. Cell, 2010. 140(3): p. 313-26.

(5) Vaher, H., Expanding the knowledge of tumour necrosis factor-alpha-induced gasdermin E-mediated pyroptosis in psoriasis. Br J Dermatol, 2024. 191(3): p. 319-320.

(6) Liu, C., et al., CXCR4-BTK axis mediate pyroptosis and lipid peroxidation in early brain injury after subarachnoid hemorrhage via NLRP3 inflammasome and NF-kappaB pathway. Redox Biol, 2023. 68: p. 102960.