Author response:
The following is the authors’ response to the original reviews.
Public Reviews:
Reviewer #1 (Public review):
Summary:
This manuscript asks how the uterine lumen is remodeled across the peri-implantation window and whether this remodeling is functionally linked to embryo attachment and subsequent pregnancy establishment. The authors combine whole-organ three dimensional imaging of optically cleared mouse uteri with single-cell and spatial transcriptomic profiling, conditional deletion of p38α at the uterine-wide versus epithelial-restricted level, and rescue experiments using progesterone and leukemia inhibitory factor. Based on these datasets, the authors propose that the luminal epithelium undergoes a previously underappreciated phase of organ-scale architectural reorganization before attachment, and that a p38α-dependent stress responsive program coordinates epithelial remodeling together with epithelial-stromal communication required for implantation competence.
Strengths:
By moving beyond local attachment-site morphology to a horn-level representation of luminal topology, the work provides anatomical context that is difficult to reconstruct from conventional section-based approaches and should be broadly useful to the implantation community. The integration of organ-scale morphology with single-cell and spatial transcriptomic datasets, together with genetic perturbation and rescue experiments, adds breadth and increases the potential long-term utility of the dataset for investigators interested in uterine receptivity and embryo-uterine interactions.
We thank the reviewer for the positive comments and constructive suggestions.
Weaknesses:
(1) The whole-uterus analysis of luminal folds and creases requires stronger methodological support. Given the mechanical compliance of the uterine lumen, it is difficult to evaluate from the current description whether dissection, fixation, clearing, and/or mounting could influence the observed luminal topography. This issue is particularly important because several key conclusions depend on the spatial distribution of folds across the uterine horn. A fuller account of tissue handling and reconstruction, together with validation that the preparation preserves native morphology, would substantially strengthen confidence in the organ-scale conclusions.
We thank the reviewer for raising this important methodological concern.
To address this concern, we have expanded the Methods section to provide a fuller description of tissue preparation, image acquisition, three-dimensional reconstruction, and quantitative analysis. Three-dimensional visualization of the uteri collected on days 1–4 and of implantation sites collected on days 5 and 6 was performed as previously described (PMID: 42022843). We have also clarified how representative cross-sectional images were selected from each reconstructed dataset and how identical analysis parameters were applied across samples.
Specifically, three representative two-dimensional cross-sectional images were generated from each three-dimensional reconstructed dataset. From day 4 onward, sections were obtained at the position of the embryo and at positions ±500 μm from the embryo. For samples collected at earlier stages, sections were obtained at the midpoint of the image stack and at positions ±500 μm along the longitudinal axis. Epithelial staining area and branching features were then quantified from these cross-sectional images using the same analysis criteria for all samples. These methodological details have been added to the revised Methods section (Page 27, Line 619).
In addition, to assess whether the whole-mount preparation preserved the overall luminal topography, we added representative transverse and longitudinal sections obtained using conventional two-dimensional histological analysis as Supplementary Fig. 1. The characteristic luminal morphology observed in these sections was comparable to that observed in the corresponding three-dimensional reconstructions. Although this comparison cannot completely exclude all possible effects of tissue processing, it supports the conclusion that the major morphological features identified by whole-organ imaging were not artifacts introduced by clearing, mounting, or reconstruction. The corresponding explanation has been added to the Results section (Page 8, Line 166).
(2) Several of the central morphological claims are supported primarily only by representative reconstructions. This includes the proposed flattening/creasing dynamics, alternating stretched and shrunken regions, persistence of abnormal folding in the mutant uterus, and the extent of structural rescue following progesterone supplementation. The authors could extract objective measures from the reconstructed luminal surface and provide more statistical analysis to demonstrate the reproducibility of the results.
We appreciate this constructive suggestion and agree that objective morphometric analyses are necessary to support the conclusions derived from the representative three-dimensional reconstructions.
Quantitative analyses of luminal epithelial flattening during days 1–4, differences between p38α uterine control and knockout mice, and morphological changes following P4/LIF administration to p38α uterine knockout mice were already included in Figures 1d and 7c. In the revised manuscript, we have expanded the descriptions of these analyses and clarified the experimental units and sample numbers in the figure legends.
We also performed additional quantitative analyses to characterize the alternating shrunken and stretched regions. We measured the orientation of luminal epithelial folds relative to the mesometrial–antimesometrial axis. In wild-type mice on day 4 at 20:00, folds in the regions surrounding embryos were predominantly vertically oriented, whereas folds at inter-implantation regions were predominantly horizontally oriented. These results provide an objective criterion distinguishing the areas that we refer to as “shrunken” and “stretched” and are now presented in Figure 2e.
We performed the same analysis in control and p38α-deficient uteri and found that the spatially organized transition between vertically and horizontally oriented folds was disrupted in the mutant uterus. These results are now presented in Figure 7e and quantitatively support the persistence of abnormal luminal folding in p38α-deficient mice.
In addition, we measured the angle of the luminal epithelium relative to the longitudinal uterine axis at three positions: the embryo center and positions 30 μm on either side of the center. The average of the three measurements was calculated for each biological sample. This analysis revealed a significant difference between day 4 at 20:00, before stable attachment, and day 4 at 24:00, after attachment, and is now presented in Figure 2f.
Together with the existing measurements of epithelial area and branching features, these additional analyses provide quantitative support for the reproducibility of the luminal flattening, folding, and regional remodeling phenotypes. Nevertheless, we acknowledge that the reconstructed surface contains additional features, such as local curvature and fold depth, that could be examined in greater detail in future studies.
(3) The manuscript appears to over-reach in concluding that luminal remodeling zones embryos before attachment from day 4 to 5. As presented, the data support a correlation between luminal architecture and embryo position, but do not discriminate between (i) luminal remodeling directing embryo positioning, (ii) embryos locally shaping the lumen, or (iii) parallel regulation of both. The evidence is based on observations of the uterus and the inside blastocysts at certain time points around implantation. Without the time-lapse analysis within the uterus, the dynamic interactions between embryos and the uterus couldn't be determined.
We agree with the reviewer that our data demonstrate a strong spatial and temporal association between luminal architecture and embryo position but do not establish the direction of causality.
Our analyses were performed at defined time points during the peri-implantation period and therefore cannot distinguish whether uterine luminal remodeling directs embryo positioning, whether embryos locally induce changes in luminal morphology, or whether both processes are coordinately regulated by common uterine and embryonic signals. As the reviewer correctly notes, direct determination of these dynamic interactions would require longitudinal or time-lapse imaging of embryos and the uterine lumen within the same animal.
We have therefore revised the manuscript to moderate the causal language concerning embryo “zoning.” We now describe the findings as showing that regional luminal remodeling is associated with, and may contribute to, the spatial organization of embryos before attachment, rather than concluding that luminal remodeling directly determines embryo position. We have also added an explicit statement that the directionality of the interaction between embryo position and luminal remodeling remains unresolved in the present study. The relevant text in the Discussion has been revised accordingly (Page 19, Line 444).
(4) The key conclusion of the manuscript is that uterine p38α regulates luminal epithelial remodeling required for embryo attachment, as shown in the title. Against this background, the finding that epithelial-restricted loss of p38α does not overtly impair fertility is notable, as it suggests that the major function of p38α may not be epithelial cell-autonomous but instead may arise through other uterine compartments that secondarily influence the epithelium. At present, however, this conclusion remains insufficiently supported: the epithelial-specific model is not characterized in sufficient depth during the peri-implantation period, and the transcriptomic evidence for altered epithelial-stromal communication does not by itself explain the phenotypic difference between uterine-wide and epithelial-specific deletion. If stromal p38α is proposed as the critical upstream regulator, more direct testing, such as stromal-specific deletion, would be needed.
We thank the reviewer for highlighting this important point. We agree that the mild phenotype observed following epithelial-specific deletion of p38α argues against a solely epithelial cell-autonomous mechanism and that our current data do not directly demonstrate that stromal p38α is the exclusive upstream regulator of luminal epithelial remodeling.
The uterine p38α knockout mice generated using Pgr-Cre lack p38α in multiple uterine compartments, including epithelial and stromal cells. To assess the contribution of epithelial p38α, we generated epithelial-specific p38α knockout mice using the Ltf-Cre driver. These mice did not exhibit the complete infertility or severe luminal remodeling abnormalities observed in the Pgr-Cre model and showed only mild uterine abnormalities. These results indicate that epithelial deletion alone is insufficient to reproduce the severe phenotype caused by broader uterine deletion of p38α.
We have revised the Results and Discussion sections to describe the epithelial-specific knockout analysis more clearly and to distinguish between conclusions directly supported by the genetic models and the proposed mechanism of epithelial–stromal communication (Page 15, Line 344; Page 20, Line 459). We also clarify that stromal- or decidual-cell-specific deletion of p38α will be required to directly determine whether stromal p38α is necessary for luminal remodeling, embryo attachment, and subsequent embryo invasion. These experiments represent an important direction for future investigation.
Reviewer #2 (Public review):
Summary:
In this study, the authors aimed to characterize the architectural reorganization of the uterine luminal epithelium during the implantation period. Using 3D histological reconstruction, single-cell RNA sequencing, and spatial transcriptomics, the authors characterize luminal remodeling during the peri-implantation period and employ a mouse model to explore the role of p38α in regulating luminal flattening.
Strengths:
This study clearly described the changes in luminal architecture during implantation. Moreover, they also used integration of multiple advanced techniques, including 3D tissue reconstruction, single-cell transcriptomics, and spatial transcriptomics, which together provide a detailed description of the molecular characteristics of the uterine architecture during implantation.
We thank the reviewer for the positive comments and constructive suggestions.
Weaknesses:
The authors used PR-Cre to generate uterine p38α knockout mice. This Cre driver deletes p38α not only in epithelial cells but also in stromal compartments. Therefore, it remains unclear whether the observed phenotype arises from epithelial cells, stromal cells, or a combination of both. Previous studies have shown that p38α regulates epithelial polarity, cytoskeletal organization, and E-cadherin localization. However, the current study does not examine changes in cell adhesion or epithelial junction integrity. Previous studies have reported that uterine fluid absorption during implantation is closely associated with luminal closure and remodeling. It would be important to determine whether epithelial transport-related genes are altered in the mutant uterus. Could dysregulated fluid homeostasis contribute to the implantation defects observed in the p38α-deficient mice?
We thank the reviewer for identifying these important mechanistic issues. We agree that the Pgr-Cre driver deletes p38α in multiple uterine compartments, including both epithelial and stromal cells, and therefore does not by itself establish which compartment is primarily responsible for the observed phenotype.
To assess the contribution of epithelial p38α, we generated epithelial-specific p38α knockout mice using the Ltf-Cre driver. These results are presented in Supplementary Fig. S6. In contrast to the complete infertility and severe luminal remodeling defects observed in Pgr-Cre-mediated uterine p38α knockout mice, epithelial-specific deletion of p38α did not critically impair female fertility and caused only mild uterine abnormalities. These findings indicate that epithelial loss of p38α alone is insufficient to reproduce the severe phenotype observed in the Pgr-Cre model.
We have revised the Results and Discussion sections to describe the Ltf-Cre analysis more clearly and to emphasize that the phenotype of the Pgr-Cre model is unlikely to result solely from epithelial-autonomous loss of p38α (Page 15, Line 344; Page 20, Line 459). Nevertheless, we agree that these findings do not directly establish stromal p38α as the exclusive upstream regulator. The phenotype may reflect the combined effects of p38α deletion in multiple uterine compartments and disrupted communication between epithelial and stromal cells. Direct testing using stromal- or decidual-cell-specific deletion will therefore be required to determine the precise contribution of stromal p38α.
We also agree that epithelial adhesion, polarity, and cytoskeletal organization are relevant to the luminal remodeling phenotype. In the original study, we examined E-cadherin and β-catenin, which are associated with adherens junctions and epithelial polarity. In the luminal epithelium of day 4 p38α uterine knockout mice, staining for E-cadherin and β-catenin was stronger and more persistent than in control mice. These findings do not suggest a simple loss of epithelial junction integrity. Instead, they are consistent with impaired remodeling of adhesion- and polarity-associated epithelial structures during the peri-implantation period.
We have revised the Results and Discussion sections to interpret these findings more explicitly in the context of epithelial adhesion and polarity. We now discuss the possibility that defective remodeling of epithelial junctional or cytoskeletal structures contributes to the persistence of abnormal luminal folding. However, because epithelial-specific p38α deletion did not reproduce the severe phenotype of the Pgr-Cre model, these epithelial abnormalities may arise, at least in part, from altered signaling between epithelial and non-epithelial uterine compartments rather than from an exclusively epithelial-autonomous defect.
Finally, in response to the reviewer’s suggestion concerning uterine fluid homeostasis, we re-examined our RNA-seq dataset with a focus on genes associated with epithelial transport, barrier function, and fluid regulation. Several transport- and barrier-associated genes were altered in the p38α-deficient uterus, including Aqp3, Slc26a6, Cldn10, Gjb2, Trpv6, Trpm6, and Atp6v0a4. These genes are involved in processes such as water transport, ion transport, paracellular permeability, intercellular communication, and pH or ion homeostasis.
These transcriptomic changes raise the possibility that dysregulated epithelial transport and fluid homeostasis contribute to defective luminal closure and remodeling in p38α-deficient mice. We have added this interpretation to the revised manuscript. However, because gene-expression analysis does not directly measure transporter activity, uterine fluid volume, or fluid absorption, our current data do not establish a causal relationship between altered fluid homeostasis and implantation failure. We have therefore presented this mechanism as a possibility and have added direct functional analysis of uterine fluid absorption and epithelial transport as an important direction for future study.
Recommendations for the authors:
Reviewer #1 (Recommendations for the authors):
The main need for the current manuscript is to apply greater analytical rigor and sharper framing.
We thank the reviewer for the positive comments and constructive suggestions.
(1) Provide a fuller description of dissection, orientation, fixation, clearing, mounting, acquisition settings, and reconstruction/segmentation criteria, including quality-control steps and exclusion criteria.
Thank you for your advice. We added these sentences in the method sections (Page27, Line619). 3D visualization of the days 1–4 uteri or day 5 and 6 implantation sites was performed as previously reported (PMID 42022843).
To quantitatively evaluate epithelial flattening morphology, three representative 2D cross-sectional images were generated from each 3D reconstructed dataset. Sections were collected at the site of embryo located and its ±500 μm positions on day 4 onwards, otherwise midpoint of the image stack and its ±500 μm positions along the longitudinal (x) axis. Epithelial staining area and epithelial branching features were subsequently quantified from these cross-sectional images using identical analysis parameters across samples.
These detailed methods are now added into the revised manuscript.
Add validation that the whole-uterus preparation preserves the luminal original physiological topography.
We appreciate this advice. Representative transverse and longitudinal sections were added as Supplementary Fig. 1 to confirm that luminal morphology was comparable between conventional two-dimensional section analysis and three-dimensional evaluation. The main text was revised accordingly (Page8, Line 166).
(2) Strengthen the support for the whole-uterus morphology data by adding quantitative morphometric analysis across biological replicates. Metrics related to fold density, epithelial branching, curvature, surface roughness, or related parameters would make the central morphological conclusions more rigorous.
Thank you for this advice.
Quantification of luminal flattening from days 1 to day 4 in wild-type mice, comparisons between p38α uCtrl and uKO mice, and the morphological changes induced by LIF/P4 administration in p38α uKO mice have already been presented in Figures 1d and 7c. We now added detailed explanation about these data.
To further quantify the transition between shrunken and stretched areas in wild-type mice before and after embryo attachment, we quantified the angle of luminal foldings against M-AM axis, showing shrunken areas have vertical folding (around 0 degree) while stretched areas have horizontal folding (around 90 degree) on day 4 20:00 (Figure 2e), which were dysregulated in p38α-deleted uteri (Figure 7e). We additionally measured the angle of the luminal epithelia against the longitudinal axis at three positions (the embryo center and ±30 μm from the center) using ImageJ. The average of these three measurements was calculated for each sample (n = 3). This analysis revealed a significant difference between day 4 20:00 (pre-attachment) and day 4 24:00 (post-attachment) (Figure 2f), indicating that embryo location is moved in alignment with the time course.
(3) Define the experimental unit clearly in each quantitative analysis. In several panels, it is not fully clear whether n refers to animals, uterine horns, implantation sites, sections, or embryos.
Thank you for pointing this out. Now the definition of n has been clarified in the figure legend.
(4) The causal language around embryo zoning should be moderated unless additional evidence can be provided. As written, the data support a strong association between luminal architecture and embryo position, but not yet a directional mechanism.
Thank you for this opinion. According to this suggestion, we have revised the Discussion to avoid overstating this conclusion (Page19, Line 444).
(5) Figure presentation should be revised for consistency and clarity. For example:- Scale bars and their labeling should be standardized throughout.
- In Figure 4e, the representative uterine photographs do not appear to be shown at a fully matched scale between each other.
- In Figure 4f, the control and mutant three-dimensional views appear to be displayed from different angles, which makes comparison of crypt formation and depth more difficult. As these are three-dimensional datasets, matched viewpoints should be feasible.
- Figure 9c appears to be mislabeled. The panel is labeled as β-actin, but based on the main context, this should be β-catenin.
Thank you for pointing this out. We have corrected Figures 4e, 4f, and 9c as suggested.
(6) The apparent reuse of the uKO image between Figures 4e and 5h should be clarified. If the same image is intentionally reused, this should be stated explicitly; otherwise, a different representative image would be preferable.
Thank you for pointing this out. As you indicated, the same image had been inadvertently reused. We sincerely apologize this mistake and have replaced it with an alternative representative image in figure 5h.
(7) Please check the figure legends and Methods carefully for internal consistency, including panel labels, marker names, genotype names, sampling time points, and nomenclature.
Thank you for your advice. We have standardized the notation of gene names, protein names, and other terminology throughout the manuscript.
(8) The Methods should include a clear description of the epithelial-specific p38α deletion model.
We apologize for the lack of explanation. We have added a description of how the epithelial-specific p38α knockout mice were generated in the Methods section.
(9) The manuscript requires thorough editing for language and presentation. For example:
- From line 53 to 5: "In the uterus, proliferation-differentiation switching (PDS), indicating the inhibition of epithelial proliferation with induction of stromal proliferation in the endometrium, is evident on day 4 under the continuous influence of P4, resulting in the endometrium acquiring implantation potential."
- From lines 257 to 261, the content of two consecutive sentences is repeated: "Based on these results, we examined whether activation of Lif-Stat3 can restore the flawed embryo attachment in p38α uKO. These results prompted us to determine whether supplementation of Lif along with P4 can ameliorate abnormal implantation in p38α uKO.
Thank you for your valuable advice. We have revised the sentences accordingly and further refined the manuscript through professional English editing.
Reviewer #2 (Recommendations for the authors):
(1) The authors used PR-Cre to generate uterine p38α knockout mice. This Cre driver deletes p38α not only in epithelial cells but also in stromal compartments. Therefore, it remains unclear whether the observed phenotype arises from epithelial cells, stromal cells, or a combination of both. An epithelial-specific deletion model would help clarify the cell-type-specific role of p38α in luminal remodeling.
We thank the reviewer for raising this important point. We agree that Pgr-Cre deletes p38α in multiple uterine compartments, including epithelial and stromal cells, and that this point should be clearly addressed when interpreting the phenotype.
To clarify the contribution of the epithelial compartment, we have already generated and analyzed epithelial-specific p38α knockout mice using the Ltf-Cre driver. These data are shown in Supplementary Fig. S6. In contrast to the complete infertility and severe luminal remodeling defects observed in Pgr-Cre-mediated uterine p38α knockout mice, epithelial-specific deletion of p38α did not critically impair female fertility and caused only mild uterine abnormalities. Thus, epithelial deletion alone is not sufficient to recapitulate the severe phenotype observed in the Pgr-Cre model.
We agree that this point was not sufficiently emphasized in the original manuscript. We have therefore revised the Results and Discussion sections to explicitly describe the Ltf-Cre analysis and to clarify that the severe phenotype in Pgr-Cre mutants is unlikely to arise solely from an epithelial-autonomous loss of p38α. Instead, our data support a model in which p38α in non-epithelial uterine compartments, particularly the stroma, contributes to luminal epithelial remodeling through epithelial–stromal communication.
We have also softened the wording to avoid implying that the Pgr-Cre model alone definitively identifies the stromal compartment as the exclusive source of the phenotype. Rather, the combination of the Pgr-Cre phenotype, the mild Ltf-Cre phenotype, and the single-cell/spatial transcriptomic analyses supports a critical role for p38α-dependent epithelial–stromal crosstalk in establishing implantation-competent luminal architecture.
We now added explanations regarding eKO in the result sections (Page15, Line344) as well as Discussion sections (Page20, Line459)
(2) Previous studies have shown that p38α regulates epithelial polarity, cytoskeletal organization, and E-cadherin localization. However, the current study does not examine changes in cell adhesion or epithelial junction integrity. The authors should evaluate markers of cell adhesion and polarity and discuss whether the observed phenotype could result from impaired epithelial junctions or altered cytoskeletal organization.
We agree with the reviewer that epithelial adhesion, polarity, and cytoskeletal organization are important potential mechanisms underlying the luminal remodeling phenotype. In the original manuscript, we examined E-cadherin and β-catenin, both of which are associated with adherens junctions and epithelial polarity. We observed stronger and more persistent staining of these markers in day 4 p38α uKO luminal epithelium compared with controls, suggesting impaired remodeling of epithelial adhesion/polarity-associated structures rather than a simple loss of epithelial junctions.
We have revised the Results and Discussion sections to explicitly describe these data in the context of epithelial adhesion and polarity. We further discuss that altered junctional or cytoskeletal remodeling may contribute to the persistent luminal folding phenotype. However, because epithelial-specific p38α deletion using Ltf-Cre did not reproduce the severe phenotype of Pgr-Cre-mediated uterine knockout mice, we interpret these epithelial abnormalities as likely arising, at least in part, from disrupted epithelial–stromal communication rather than from an exclusively epithelial-autonomous junctional defect.
(3) Previous studies have reported that uterine fluid absorption during implantation is closely associated with luminal closure and remodeling. It would be important to determine whether epithelial transport-related genes are altered in the mutant uterus. Could dysregulated fluid homeostasis contribute to the implantation defects observed in the p38α-deficient mice?
We thank the reviewer for this thoughtful suggestion. As the reviewer noted, uterine fluid absorption and epithelial transport are important processes associated with luminal closure and remodeling during implantation.
In response to the reviewer’s suggestion, we re-examined our RNA-seq dataset with a focus on genes related to epithelial transport and fluid homeostasis. We found that several transport- and barrier-associated genes were altered in the P4/LIF-untreated p38a-deficient uterus, including Aqp3, Slc26a6, Cldn10, Gjb2, Trpv6, Trpm6, and Atp6v0a4. These genes are associated with water transport, ion transport, paracellular permeability, epithelial communication, and pH/ion homeostasis. These findings suggest that dysregulated epithelial transport and fluid homeostasis may contribute, at least in part, to impaired luminal closure/remodeling and implantation defects in the mutant uterus.
We have added this point to the revised manuscript. Nevertheless, because transcriptomic data do not directly measure transporter activity or uterine fluid dynamics, further functional studies will be required to determine whether altered fluid absorption directly contributes to the abnormal luminal architecture. We have revised the Discussion to acknowledge this limitation and to propose direct functional analysis of uterine fluid absorption and epithelial transport as an important future direction.
(4) It is intriguing that P4 plus LIF supplementation rescues embryo attachment, but pregnancy still fails at later stages. The authors should discuss whether this defect might result from p38α deletion in stromal cells or from persistent abnormalities in luminal remodeling.
Thank you for your advice. It is difficult to determine from the current study whether the observed phenotype is primarily caused by persistent luminal epithelial abnormalities or by stromal p38α deficiency. In the present study, P4 and LIF supplementation restored embryo attachment in p38α uKO mice; however, embryo invasion remained completely absent, resulting in subsequent pregnancy failure. Previous studies from our group have shown that uterine deficiency of Hif2α, Rb1, Ezh2, or Lox impairs embryo invasion and leads to pregnancy failure, highlighting the importance of coordinated epithelial–stromal remodeling during implantation (PMID: 41803091, 3719814, 2012472, 33399260). In the p38α uKO uterus, although P4 and LIF supplementation restored key molecular signals required for attachment, abnormalities in luminal morphology persisted. Therefore, persistent defects in luminal remodeling may contribute to the failure of embryo invasion and subsequent pregnancy progression. However, because embryo attachment is severely impaired in untreated p38α uKO mice, the post-attachment role of stromal p38α cannot be directly assessed in the current model. Further studies using decidual cell-specific genetic approaches will be required to clarify the contribution of stromal p38α to embryo invasion. We added these sentences in the discussion sections.
(5) The terminology describing luminal morphogenesis requires clarification, particularly the definitions of "shrunken areas" and "stretched areas." These terms should be clearly defined and, if possible, supported by quantitative criteria
Thank you for your suggestion. To quantify the transition between shrunken and stretched areas in wild-type mice, we measured the angle of luminal folding around the embryo (±500 µm) as well as inter-implantation sites on day 4 2000h using ImageJ. This analysis revealed the epithelia around the embryos are more shrunken with vertically oriented folding compared to inter-implantation sites (Figure 2e). We similarly analyzed uteri from control and p38α uKO, and found uKO poorly formed shrunken and stretched areas, which may compromise embryo implantation in this milieu (Figure 7e).
(6) The exact collection time points for day 4 and day 5 uteri should be clearly specified (Line 169), as implantation-related changes are highly time-dependent.
Thank you for your advice. We have added “10:00” to both day 4 and day 5.
(7) The manuscript relies heavily on qualitative visual comparisons of luminal morphology. The authors should provide quantitative morphometric analyses, such as measurements of lumen width, epithelial curvature, or fold depth, to support their conclusions.
Thank you for your suggestion. To quantify the transition between shrunken and stretched areas in wild-type mice before and after embryo attachment, we measured the angle between the LE folding adjacent to the embryo and inter-implantation sites against the M-AM axis. The corresponding results have been added to the manuscript (Figure 2e and 7e).
