A stress-responsive morphogenetic program of the uterine epithelium safeguards the establishment of early pregnancy

19 Successful embryo implantation requires coordinated interactions between the 20 endometrial epithelium, stroma, and the embryo, yet underlying mechanisms have 21 not been fully understood. Using three-dimensional histological reconstruction 22 combined with single-cell and spatial transcriptomics, we identify a previously 23 unrecognized phase of luminal architectural reorganization preceding embryo 24 attachment. Within a narrow peri-implantation window, the luminal epithelium rapidly 25 remodels from a highly folded structure into a flattened, organized architecture that 26 provides a scaffold for embryo positioning. This morphogenetic transition is 27 accompanied by activation of stress-responsive signaling across epithelial and 28 stromal compartments. Functional analyses show that uterine-specific deletion of the 29 stress-responsive MAP kinase p38α disrupts luminal remodeling, leading to 30 persistent epithelial folding, failed embryo attachment, and infertility despite normal 31 hormone levels and embryo development. Although combined progesterone and 32 leukemia inhibitory factor supplementation rescues embryo attachment in 33 p38α -deficient uteri, luminal disorganization, abnormal stromal responses, and 34 impaired pregnancy progression persist. These findings identify a p38α -dependent, 35 stress-responsive morphogenetic program that coordinates epithelial dynamics and 36 epithelial–stromal communication to establish implantation-competent luminal 37 architecture. 38 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 3

39 Infertility is a social concern that affects 17.5% of the adult population worldwide1. 40 Some patients who undergo in vitro fertilization and embryo transfer (IVF‐ ET) 41 repeatedly fail to become pregnant, even after ET using high-quality embryos (2). 42 Embryo implantation, which is the starting point of pregnancy, can be divided into 43 several processes, including blastocyst spacing, apposition, attachment to the uterine 44 luminal epithelium, and invasion into the endometrial stroma2, 3. Establishing 45 implantation requires an exquisite interaction between the endometrium and embryo; 46 however, the detailed underlying mechanism remains unclear. 47 Based on the similarities in the influence of female sex hormones on the 48 endometrium during pregnancy, rodents have been utilized as an in vivo model of 49 human pregnancy. In mice, day 1 of pregnancy is defined by the observation of a 50 vaginal plug. After coitus, high serum levels of estrogen (E₂ ) induce epithelial 51 proliferation and P₄ production from the ovary increases on day 3. Embryos in the 52 oviduct reach the uterus on day 4 in a P₄ -dominant hormonal environment. In the 53 uterus, proliferation-differentiation switching (PDS), indicating the inhibition of 54 epithelial proliferation with induction of stromal proliferation the endometrium, is 55 evident on day 4 under the continuous influence of P₄ , resulting in the endometrium 56 acquiring implantation potential4, 5. With this dynamic change, the morphology of the 57 endometrial luminal epithelium reveals a slit-like narrowing, known as the formation of 58 a slit-like luminal structure4, 6, 7 (Fig. 1a). Late on the morning of day 4, blastocysts are 59 activated by a small estrogen surge that occurs as the starting signal for implantation2, 60 8. The blastocyst finally attaches to the luminal epithelium on day 4 midnight. The 61 luminal epithelium initiates the formation of a uterine crypt after attachment and the 62 embryo can be observed at the bottom of the crypt9. Stimulation from the embryo 63 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 4 attachment seems to be transmitted from the endometrial epithelium to the stroma, 64 where vascular permeability is increased (attachment reaction), and the surrounding 65 stromal cells initiate differentiation into decidual cells (decidualization). By the 66 evening of day 5, the endometrial luminal epithelium facing the attached blastocysts 67 disappears, and trophoblast invasion is initiated. 68 In recent studies, our group and others have demonstrated that a series of luminal 69 changes, that is PDS, formation of a slit-like luminal structure, and crypt formation, 70 are crucial for embryo implantation4, 5, 6, 7, 9, 10, 11. For PDS and formation of the slit-like 71 luminal structure, the P₄ -progesterone receptor (PGR) pathway is involved4, 5, 6. We 72 previously reported that mice with epithelial deletion of Pgr (Pgrfl/fl;LtfCre/+ mice; Pgr 73 eKO) showed sustained epithelial cell differentiation, resulting in defective uterine 74 receptivity6. As for embryo attachment and subsequent crypt formation, the Lif-Stat3 75 pathway is crucial, in addition to P4-Pgr signaling. Leukemia inhibitory factor (Lif) 76 activates uterine Lif receptors (Lifrs) to evoke Stat3-mediated gene transcription, thus 77 initiating the formation of implantation chambers (crypts)9, 11, 12, 13, 14. We have also 78 reported that uterine-specific retinoblastoma knockout mice (Rb uKO) and enhancer 79 zeste homolog 2 knockout mice (Ezh2 uKO) show impaired PDS because of cell 80 cycle abnormalities in the endometrium15, 16. Notably, in these uKO models, although 81 epithelial proliferation was continuously observed even on day 4, embryo attachment 82 occurred but subsequent embryo invasion was flawed. These differential phenotypes 83 of uterine-specific gene deletions suggest that essential mechanisms other than PDS 84 are involved in P4- and Lif-induced embryo attachment and subsequent pregnancy 85 maintenance. 86 Accumulating evidence has demonstrated that PDS, slit-like formation, and crypt 87 formation are important changes in the luminal epithelium for embryo attachment. 88 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 5 However, how the overall changes in the luminal epithelia are dynamically regulated 89 during the short period of embryo implantation remains unclear. It appears that PDS, 90 slit formation, and crypt shaping occur sequentially; however, how each step 91 influences the others remains uncertain. In particular, crypt formation occurs only in 92 the presence of embryos10, indicating that epithelial changes require reciprocal 93 interaction with embryos. To understand the molecular mechanism underlying these 94 epithelial changes, we especially examined p38α , a MAP kinase activated by 95 phosphorylation in response to various environmental stresses and inflammatory 96 cytokines, which regulates fundamental cellular processes such as proliferation, 97 apoptosis, cell differentiation17, 18, 19. Notably, p38α plays an important role in 98 development and tissue differentiation by modulating the localization of E-cadherin, 99 which is expressed in tissue epithelial cells, thus altering epithelial morphology20, 21. 100 Embryo implantation requires cell differentiation and death as well as glandular 101 epithelial development and secretion, which may be influenced by p38α because it 102 regulates various cell differentiation processes including mammary gland duct 103 formation22, 23, 24. Recently, deletion of p38α was reported to result in complete 104 pregnancy failure in mice25, supporting our notion of a p38α -regulated mechanism 105 shaping the endometrial epithelium during the peri-implantation period. 106 In this study, we first investigated the spatiotemporal changes of luminal shapes in 107 3D spanning the period from just after the coitus to the peri-attachment stage using a 108 time course analysis. Our analyses revealed that the surface of the endometrial 109 lumen not only became flattened but also showed creases, resulting in the even 110 zoning of embryos before attachment. The embryos were then attached to the 111 flattened area and crypts were formed. This epithelial shaping was impaired in the 112 mice with uterine-specific knockout (uKO) of p38α , giving rise to failed embryo 113 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 6 attachment. While treatment with P4 and Lif restored embryo attachment in p38α uKO 114 mice, embryo invasion and subsequent pregnancy maintenance remained disturbed 115 because of abnormal epithelial shaping, which was not rescued. In summary, we 116 discovered that dynamic morphological changes in the endometrial lumen prior to 117 implantation may influence the process of pregnancy establishment and maintenance, 118 which is regulated by previously unknown mechanisms independent of P4-Pgr and 119 Lif-Stat3. 120 121 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 7

122 The morphology of the endometrial luminal surface dynamically changes 123 before embryo attachment. 124 To identify the spatiotemporal changes in the luminal epithelium prior to embryo 125 attachment, we adopted 3D visualization, which recently revealed the detailed 126 topography of the endometrial epithelium and the mechanism of embryo 127 implantation10. Blood vessels enter the uterus from the mesometrium, situating the 128 uterus along the mesometrial–anti-mesometrial (M–AM) axis. Once blastocysts 129 attach to the surrounding luminal cells, an implantation chamber (crypt) is formed by 130 luminal epithelial (LE) evaginations toward the AM pole10, 26, 27 (Fig. 1a, b). Although 131 the morphological changes in the endometrial lumen after embryo attachment have 132 been reported10, 28, little is known regarding the luminal morphology prior to 133 attachment. How the epithelial dynamics prior to embryo attachment influence 134 subsequent pregnancy processes also remains unclear. 135 Therefore, we analyzed the luminal morphology in wild-type mice on the mornings 136 of days 1–4 of pregnancy. On day 1, the endometrial lumen extended in a disorderly 137 manner against the M-AM plane. During days 2–4, the lumen became flattened with 138 some folding in the M-AM axis that remained even (Fig. 1c). A cross-sectional view of 139 the tissues revealed that the lumen became more slit-like with epithelial mass 140 decreasing daily prior to embryo attachment (Fig. 1c, d), which was consistent with 141 the results of luminal changes previously depicted using conventional histology4, 29. 142 Next, we investigated the relationship between luminal folding and blastocyst 143 movement by observing day 4 endometria from the morning to midnight, when 144 blastocysts arrived and attached to the uterine lumen, respectively. We observed the 145 3D morphology of the uterine lumen and positions of the blastocysts on day 4 146 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 8 morning (day 4 10:00), day 4 evening (day 4 16:00), day 4 evening (day 4 20:00), day 147 4 midnight (day 5 0:00), and day 5 morning (day 5 10:00). Once embryo attachment 148 occurred, vascular permeability increased in the surrounding endometrium, which 149 could be observed by the injection of blue dye into mice (blue dye reaction)30. The 150 blue dye reactions were 0% (0/5) at 16:00 on day 4, 18.8% (3/16) at 20:00 on day 4, 151 66.7% (4/6) at 0:00 on day 5, and 100% (7/7) at 10:00 on day 5 (Fig. 2a-c). Embryo 152 attachment was completed in more than half of the individuals at midnight on day 4; 153 therefore, we defined the evaluation time immediately before embryo attachment as 154 day 4 20:00. As previously described, the surface of the endometrial lumen was 155 flattened at day 4 10:00. On Day 4, from 16:00 to 20:00, the flat lumen exhibited 156 alternating shrunken and stretched areas. Shrunken areas with multiple folds were 157 observed at regular intervals. No regularity in the position of the embryos was 158 observed until day 4 16:00; however, at day 4 20:00, the embryos entered the 159 shrunken areas and gradually moved to the stretched areas over the time course. At 160 day 4 midnight, the embryos were attached to the AM pole of the lumen in the 161 stretched areas, and crypts were then formed on day 5 morning (day 5 10:00) (Fig. 162 2d). These results demonstrate that the uterus just prior to embryo attachment 163 showed dynamic changes in luminal morphology once the embryos arrived. 164 165 Uterine p38α activation is crucial for luminal morphology and the subsequent 166 embryo attachment 167 We then examined how luminal dynamics before attachment were regulated. 168 We performed single-cell RNA-seq (scRNA-seq) analysis for days 4 and 5 uteri (Fig. 169 3). Uterine tissues contain multiple cell types, including epithelial, stromal, vascular 170 endothelial (VE), and immune cells (Fig. 3a, Supplementary Fig. 1a, Table S1 and S2). 171 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 9 Among them, luminal cell clusters (LE) can be divided into two types: conventional 172 luminal cells (LE) which are highly observed on day 4, and activated LE 173 (LE_activated), whose population increases on day 5 (Fig. 3a). To determine the 174 characteristics of the LE_activated cells, we performed enrichment analyses using 175 Enrichr31 (Fig. 3b, Table S2). Transcription factor protein-protein interactions (PPIs) 176 revealed enrichment of ATF2 and JUNB, which are related to stress signalling32. In 177 agreement with this result, a pathway analysis using MSigDB HallMark also revealed 178 TNFα -related and hypoxia signaling. Furthermore, gene ontology analysis using 179 Metascape33 identified pathways related to oxidative stress and cell motility (Fig. 3c, 180 Table S2). We then investigated stromal cell types in the same manner (Fig. 3d, 181 Supplementary Fig. 1b, and Table S3). These were clustered into five types: 182 non-proliferative (Non-pro), proliferative (Pro), sub-epithelial (Sub-epi), 183 sub-endothelial (Sub-endo), and attached (Attached). Because the day 5 stroma 184 exclusively contained Attached cluster, we analyzed the enriched pathways in this cell 185 type. Similar to LE_ activated, this cluster was enriched in stress-related 186 transcriptional factors, signals and Gene Ontology (GO) terms (Fig. 3e, f, and Table 187 S4). 188 As a possible regulator of day 5-specific LE and stromal cell types, we 189 focused on p38α , a Map kinase protein. p38α is phosphorylated for activation in 190 response to various kinds of cellular stimuli18. Notably, phosphorylated p38α (pp38α ) 191 translocates into nuclei to activate ATF2-JUN-induced transcription of cytokines, 192 including TNFα 32. Our immunostaining revealed that p38α was highly expressed and 193 activated by phosphorylation in the pre-attachment luminal epithelia (Fig. 3g), 194 indicating the role of p38α in the luminal epithelium before embryo attachment. After 195 attachment, pp38α was expressed both in the endometrial epithelium and stroma, 196 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 10 especially around the attached embryos on days 6 and 8. 197 To investigate the roles of uterine p38α , we established mice with 198 uterine-specific deletion of p38α (p38α uKO) by crossing p38α -floxed mice with those 199 carrying a Pgr-Cre driver. Efficient deletion and inactivation of p38α protein in p38α 200 uKO was confirmed by immunostaining for p38α and pp38α (Fig. 4a). To examine the 201 reproductive phenotypes of p38α uKO and littermate p38α -floxed females (p38α Ctrl), 202 we mated them with fertile wild-type (WT) male mice. p38α uKO mice showed 203 complete infertility (Fig. 4b). As we detected corresponding numbers of hatched 204 blastocysts by flushing the uterine cavity with saline solution on day 4 morning in 205 each genotype (Fig. 4c), uterine p38α did not influence embryo development before 206 attachment. We then intravenously injected Chicago blue dye solution to examine the 207 implantation sites on day 5 of pregnancy. However, p38α uKO uteri showed no 208 attachment sites on day 5 of pregnancy, and blastocysts were recovered by saline 209 flushing of the uterine cavities (Fig. 4d), which indicated flawed embryo attachment. 210 On day 6 of pregnancy, the number of embryo attachment sites was significantly 211 reduced in p38α uKO compared with those in the p38α uCtrl (Fig. 4e). Further, our 3D 212 imaging on day 5 morning revealed failed crypt formation in p38α uKO, suggesting 213 that p38α uKO mice show infertility because of embryo attachment failure (Fig. 4f). 214 Our observations indicating luminal activation of p38α in the pre-attachment 215 period (Fig. 3g) as well as infertility with flawed embryo attachment in p38α uKO 216 females (Fig. 4) motivated us to investigate the involvement of this molecule in 217 luminal dynamics. We then compared the morphological changes in the lumen of 218 p38α uCtrl and p38α uKO from days 1 to 4 using 3D imaging (Fig. 4g). Similar to the 219 observation in the wild type mice, p38α uCtrl showed that the folding of the 220 endometrial lumen in the M-AM axis gradually disappeared but some folding 221 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 11 remained evenly. Accordingly, the endometrial lumen flattened as the pregnancy 222 progressed from days 1 to 4. In contrast, in p38α uKO, folding in the both 223 cross-sectional longitudinal axes remained even on day 4, which appeared as 224 serrated luminal shapes in the 2D view. Considering that embryo attachment failed in 225 p38α uKO, the luminal changes in the pre-attachment phase could influence 226 successful embryo attachment. 227 228 Supplementation with P₄ and Lif, two major factors supporting embryo 229 implantation, rescues the flawed embryo attachment, but not the subsequent 230 pregnancy maintenance in p38α uKO 231 Our group and others have previously revealed that PDS, wherein epithelial cell 232 proliferation is terminated before implantation, is an indicator of endometrial embryo 233 receptivity5. P₄ is an inducer of PDS as well as slit formation in the endometrial 234 lumen4, 6. Immunostaining for Ki67, a cell proliferation marker, showed an increased 235 number of proliferating cells in the luminal epithelium in p38α uKO on day 4 morning, 236 suggesting that PDS was impaired in this milieu (Fig. 5a, b). This result motivated us 237 to examine whether P₄ supplementation could rescue the phenotype of p38α uKO. 238 We thus treated p38α uKO with P₄ in the preimplantation period and examined the 239 resulting morphological changes in the endometrial lumen using 3D analysis. On 240 days 3 and 4, P4 treatment suppressed folding in both the cross-sectional and 241 longitudinal axes, thus flattening the lumen (Fig. 5c-e). Further, PDS was also 242 rescued by P4 injection to p38α uKO (Fig. 5f, g). Notably, p38α uKO showed a normal 243 hormone-producing capacity of the ovary, as serum estradiol-17β (E₂ ) and 244 progesterone (P₄ ) levels were comparable (Supplementary Fig. 2a). We also 245 confirmed that the expression of estrogen receptor (ERα ) and progesterone receptor 246 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 12 (PGR) in the uterus were not influenced by p38α deletion (Supplementary Fig. 2b). 247 Although luminal morphology was improved by P4 treatment, embryo attachment was 248 still failed in the mutant (Fig. 5h), suggesting that some other factor is required for 249 p38α -dependent embryo attachment. 250 Besides the P ₄ -PGR pathway, leukemia inhibitory factor (Lif) is a critical factor 251 for embryo attachment9, 11, 34. Lif is an interleukin-6 family member secreted by the 252 endometrial gland on the day 4 morning34, 35, 36. In situ hybridization showed 253 significant decreases of Lif in the p38α uKO uterus on day 4 morning even after P4 254 treatment (Fig. 6a). Further, activation of Stat3, a transcription factor downstream of 255 Lif, was also downregulated as determined by immunostaining for phosphorylated 256 Stat3 in p38α uKO uterus on day 4 morning (Fig. 6b). Based on these results, we 257 examined whether activation of Lif-Stat3 can restore the flawed embryo attachment in 258 p38α uKO. 259 These results prompted us to determine whether supplementation of Lif along with 260 P₄ can ameliorate abnormal implantation in p38α uKO. Five groups were established: 261 vehicle administration to p38α uCtrl and p38α uKO, P₄ alone to p38α uKO (at 10:00 262 on day 2–day 4), Lif alone to p38α uKO (at 9:00 and 18:00 on day 4), and both P₄ and 263 Lif to p38α uKO (Fig. 6c). The number of embryo attachment sites in the single 264 treatment of either P₄ or Lif did not differ compared with that in the vehicle group. In 265 contrast, simultaneous supplementation with P₄ and Lif increased the number of 266 embryo attachment sites in p38α uKO, which was comparable to that in the p38α 267 uCtrl group as shown by the clear blue reactions on day 5 morning, indicating that 268 both P₄ and Lif are required for p38α -dependent embryo attachment (Fig. 6d, e). 269 COX2 (Fig. 6f) and phosphorylated Stat3 (Supplementary Fig. 3), which are induced 270 at embryo attachment sites11, were found to be expressed in p38α uKO upon P₄ and 271 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 13 Lif treatment, also supporting our notion. However, despite embryo attachment, P4 272 and Lif treatment could not recover full term pregnancy with litters in p38α uKO (Fig. 273 6g). On day 8, although the number of implantation sites was comparable between 274 the two groups, none of the implantation sites underwent normal embryogenesis and 275 formed hematomas in p38α uKO treated with P₄ and Lif (Supplementary Fig. 4), 276 indicating that pregnancy maintenance did not occur normally. Eventually, P4 and 277 rLif-treated p38α uKO females failed to give birth, accompanied by severe embryo 278 resorption (Fig. 6g). This suggests that p38α is required for P4- and Lif- induced 279 embryo attachment as well as healthy pregnancy maintenance. 280 281 p38α plays an important role in establishing the luminal epithelial morphology 282 for embryo attachment, and changes by day 4 morning are an important 283 scaffold for embryo attachment site formation 284 Flawed pregnancy maintenance in the mutant even after treatment led us to 285 examine how pregnancy events after the embryo attachment were disturbed in 286 p38α -deleted uteri. We thus investigated the luminal morphologies on days 5 and 6 287 when embryo attachment and invasion were evident (Fig. 7a). We found that folding 288 was evident in the longitudinal axis in p38α uKO regardless of any treatment. 289 Simultaneous treatment with P4 and Lif created a crypt on day 6, but still longitudinal 290 folding was observed around the crypt. Further, crypt formation was poorly initiated 291 on day 5 morning in P4 and Lif-treated p38α uKO showing an obvious longitudinal 292 folding. Considering that this folding should be eliminated by day 4 night in normal 293 pregnancy (Fig. 2), we investigated the luminal morphology at day 4 20:00, just 294 before embryo attachment (Fig. 7b). Similar to the preparatory phase of embryo 295 attachment on day 4 morning (10:00 am), the surface of the stretched lumen, i.e., the 296 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 14 site of embryo attachment, was flat in p38α uCtrl. In contrast, p38α uKO and p38α 297 uKO+Lif (P₄ non-treated group) showed remarkable persistence of luminal folding in 298 the longitudinal axis with a poorly established stretched area. In p38α uKO+P₄ and 299 p38α uKO+Lif+P₄ (P₄ treated group,) the flattening and stretching of the lumen was 300 partially rescued but longitudinal folding remained (Fig. 7b, left). We also observed 301 the luminal shapes in cross-sectional views to examine the relationship between the 302 shape of the slit-like lumen and embryo location (Fig. 7b, right, c, and d). In the p38α 303 uCtrl, the stretched luminal areas serving as embryo attachment sites, were flat and 304 slit-like, predicting smooth guidance of the embryo to the AM pole (Fig. 7b, right). In 305 contrast, regardless of the treatment, flattening of the lumen failed in the p38α uKO, 306 accompanied by strayed positioning of embryos probably because of failure of M-AM 307 axis formation. These abnormal morphologies of the luminal epithelia appeared as an 308 increased epithelial mass (Fig. 7c) and increased epithelial branching in the uKO (Fig. 309 7d). Overall, these results suggest that p38α is responsible for morphological 310 changes in the lumen before embryo attachment, which P₄ , but not Lif, partially 311 assists in. Considering that P4 could not solely restore embryo attachment in p38α 312 uKO, Lif seems to act as an inducer of attachment under the influence of P4. 313 314 p38α is an important signal transducer between luminal epithelia and stroma 315 for embryo attachment 316 We then examined how p38α influences feto-maternal interactions to accomplish 317 healthy embryo implantation and luminal dynamics. Notably, we found that 318 epithelial-specific deletion of p38α did not critically alter female fertility 319 (Supplementary Fig. S5), indicating that p38α regulates epithelial-stromal crosstalk. 320 To obtain this information, we performed scRNA-seq in Ctrl uteri on day 4 2000 h 321 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 15 (just before attachment) and 2400 h (at attachment), as well as in uKO mice with or 322 without P4 + rLif treatment on day 4 2400 h (Fig. 8a, Table S5). Similar to Fig. 3a, we 323 found multiple cell types in the uteri, including epithelial, stromal, myometrial, and 324 immune cells. Among these, we focused on stromal cells as their population 325 increased during the embryo attachment process in the uCtrl (Fig. 8a). The stromal 326 cells were further clustered into non-proliferative (Non-prolif), proliferative (Pro), 327 decidualizing (Dec), sub-epithelial (Sub-epi), sub-endothelial (Sub-endo), and 328 uKO-specific clusters (Fig. 8b, Table S6). The uKO-specific cluster was highly 329 enriched in uKO uteri. We then traced stromal differentiation using pseudo-time 330 analysis (Fig. 8c and d). Intriguingly, the uKO-specific cluster was poorly differentiated 331 compared with other stromal cell types, except for the Non-prolif cluster (Fig. 8d). 332 We then examined the signal transduction between the epithelial (LE and GE) and 333 stromal clusters, focusing on secretory molecules that can bridge the epithelial and 334 stromal compartments (Fig. 8e). In uCtrl, non-canonical Wnt (ncWnt) from the LE and 335 stroma and IGF from the stroma affected the decidualizing and luminal cells, which 336 were enhanced upon embryo attachment (Fig. 8e, upper). In contrast, in 337 p38α -deficient uteri, only epithelial cells strongly sent and received signals (Fig. 8e, 338 lower left), which remained even with P4 and rLif treatment (Fig. 8e, lower right). 339 These results demonstrate flawed epithelial-stromal communication in the p38α uKO 340 milieu. 341 Among ncWNTs, Wnt5a plays a critical role in early pregnancy27, 28. Wnt5a activates 342 receptor tyrosine kinases Ror1 and Ror2 in the uterus Both overexpression and 343 deletion of Wnt5a compromises pregnancy outcomes owing to sustained apicobasal 344 polarity in the luminal epithelia27. Similar to Wnt5a, Igf1 is also involved in epithelial 345 depolarity and is highly expressed in day 4 uterine stroma, activating Igf1r and 346 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 16 downstream Stat3 in the luminal epithelia37. Uterine-specific deletion of Igf1r results in 347 flawed embryo attachment because of sustained epithelial polarity37. These contexts 348 prompted us to examine these two molecules in the p38α uKO milieu. We first 349 compared Wnt5a and Igf1 expression using scRNA-seq data and found significantly 350 upregulated Wnt5a and downregulated Igf1 in both the luminal epithelia (LE) and 351 decidualizing stroma (Dec) (Fig. 9a and b). In agreement with these results, we found 352 stronger staining of epithelial markers, β -catenin and E-cadherin, in day 4 uKO uteri 353 (Fig. 9c). We then asked whether embryo invasion was influenced by p38α deletion. 354 Our group has previously shown that epithelial removal after the loss of epithelial 355 polarity facilitates embryo invasion and the subsequent pregnancy processes15, 38. To 356 assess embryo invasion, the implantation sites were stained for cytokeratin 8 (CK8), 357 a marker of epithelial and trophoblastic cells. In p38α uCtrl, the endometrial luminal 358 epithelium disappeared and trophoblasts invaded the uterine stroma on the morning 359 of day 6, whereas the endometrial luminal epithelium around the embryo remained 360 and embryo invasion was incomplete in p38α uKO even after treatment with P4 and 361 Lif (Fig. 9d). 362 We also examined the spatial transcriptome of the embryo attachment site on the 363 morning of day 5 from Ctrl- and P4 + rLif-treated uKO mice to observe 364 epithelial-stromal interactions (Fig. 9e-h). Cross-sectional tissues of each 365 implantation site were clustered into six types – LE, embryo attached stroma 366 (Str_attached), proliferative stroma (Str_prolif), uKO-specific Str (Str_uKO_specific), 367 GE, and myometria (Myo) (Fig. 9e, f, Table S7). Notably, uKO tissues solely 368 contained Str_uKO_specific as the stromal cluster. To understand the characteristics 369 of this cell type, pathway analyses were performed using Metascape. We found that 370 the upregulated genes in Str_uKO_specific were enriched in pathways such as 371 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 17 “regulation of inflammatory response,” “negative regulation of cell differentiation,” and 372 “negative regulation of cell population proliferation” (Fig. 9g, Table S8) 373 Down-regulated genes were related to “Extracellular matrix organization,” “negative 374 regulation of canonical Wnt signaling pathway,”, and “regulation of cellular response 375 to growth factor stimuli” (Fig. 9h, Table S8), which are known to be associated with 376 healthy pregnancy outcomes36, 39. These results indicate that embryo attachment to 377 the uKO luminal epithelia induces inflammatory signals rather than physiological 378 reactions in the stroma, compromising subsequent pregnancy maintenance. In 379 summary, our data demonstrate the previously unappreciated role of uterine p38α as 380 crucial for appropriate embryo attachment independently of the P4 and Lif pathways. 381 382 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 18

383 Communication between embryos and endometria is crucial for the successful 384 establishment of pregnancy40. In preparation for blastocyst attachment, the 385 endometrium undergoes massive morphological changes, especially in the epithelial 386 layers: (1) PDS before embryo attachment, (2) slit formation of the endometrial lumen 387 before embryo attachment, and (3) crypt formation of the endometrial luminal 388 epithelium after embryo attachment. In this study, we conducted 3D histological 389 analysis as well as single cell and spatial transcriptome analyses to elucidate the 390 details of dynamic morphological changes in the endometrial lumen. Two molecular 391 pathways, the P₄ -PGR and Lif-Stat3 pathways, have been considered important for 392 embryo attachment so far. Further, we demonstrated a previously unappreciated role 393 of p38α in addition to the above two major pathways (Fig. 10). 394 395 The major finding of this study was that luminal shapes were altered throughout the 396 uterine horns during early pregnancy. In particular, changes were evident from the 397 morning of day 4 to midnight. This may explain the mechanism of embryo attachment 398 as well as embryo spacing: a previous study demonstrated that embryo spacing 399 occurs during day 4 noon to evening, which was compromised in mice with systemic 400 KO of Lpar3, a lipid receptor expressed in the luminal epithelia41. Indeed, we 401 observed that epithelial folding in the M-AM axis gathered evenly, corresponding to 402 the position of the embryos from morning to evening on day 4, indicating that luminal 403 layer movement contributes to embryo spacing in addition to myometrial, as 404 previously reported42. This 3D imaging experiment had some limitations. Because the 405 data were obtained from euthanized mice, following changes over time in the same 406 individual was impossible, and the influence of phenotypic time differences among 407 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 19 individuals could not be completely ruled out. To examine the epithelial morphology 408 and phenotype over time in the same individual, experimental system using 409 techniques such as live imaging in vivo needs to be established. 410 Considering the functional feature of p38α as a Map kinase, the outer- or 411 inner-cellular stimuli evoking p38α -dependent epithelial shaping remain unclear. Map 412 kinases can be activated downstream of various receptors, including tyrosine 413 receptor kinases and G protein-coupled receptors18. One candidate is the Ror1/Ror2 414 tyrosine receptor kinase, which can be activated by Wnt5a27. Uterine-specific deletion 415 of Wnt5a-Ror1/Ror2 axis resulted in defective embryo implantation because of 416 abnormal luminal morphology, similar to our observation in p38α uKO. Similarly, 417 deletion of Igf1-Igfr signaling causes poor embryo attachment, accompanied by 418 abnormal luminal integrity37. These contexts may explain why Wnt5a and Igf1 were 419 upregulated in p38α uKO uteri. 420 New insights into the morphological changes in the endometrial lumen may provide 421 an innovative approach for treating embryo attachment failure, focusing on the 422 endometrial lumen morphology. In humans, healthy implantation occurs at the fundus 423 of the uterus43, which is different from that in rodents with turbinal uterine structures. 424 However, epithelial integrity is a common feature that regulates appropriate embryo 425 implantation across species; for instance, after ovulation in humans, 426 epithelial-mesenchymal transition with reduced epithelial polarity occurs during 427 menstruation and implantation, thus influencing implantation outcomes44, 45, 46. A 428 previous study using human endometrial epithelial cell lines demonstrated that 429 deletion of p38α influenced the cellular transcriptome and metabolome, contributing 430 to cancer cell-like characters47, suggesting the role of p38α in maintaining luminal 431 epithelial integrity in humans. In addition to specific molecular mechanisms, 432 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 20 implantation failure may also be overcome from the mechanical aspect of tissue 433 mobility for pathological conditions such as uterine myoma and uterine adenomyosis, 434 wherein abnormal morphological changes in the endometrial lumen are presumed48. 435 These findings are expected to have broad applications in diagnosing and treating 436 human implantation failure, including the search for biomarkers of implantation ability 437 and supplementation with relevant molecules. As reported previously, p38α also plays 438 critical roles in mammary gland lumen formation22; therefore, the mechanism 439 discovered in this study could be applicable in other epithelial systems as well. 440 441 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 21

442 Mice 443 WT (C57BL/6N, SLC), p38α -floxed (Mapk14-floxed; kindly provided by Dr. Kinya 444 Otsu, University of Osaka)19, Pgr-Cre49, were used in this study. Pgr-Cre is expressed 445 throughout the uterine layers49. Mice with p38α deletion in all uterine layers 446 (Mapk14flox/flox PgrCre/+; p38α uKO) were generated by crossing Pgr-Cre with 447 p38α -floxed mice. Cre-negative littermates (Mapk14flox/flox) served as controls. All 448 mice used in this study were housed at the University of Tokyo Animal Care Facility, 449 following the institutional guidelines for using laboratory animals. 450 451 Evaluation of pregnancy outcomes 452 To examine the pregnancy outcomes, p38α -uKO, or p38α -floxed (control) female 453 mice were mated with C57BL/6N fertile male mice, as reported in a previous study15, 454 38, 50. The day of vaginal plug detection was considered day 1 of pregnancy. Pregnant 455 mice were euthanized by cervical dislocation on the designated day of pregnancy to 456 evaluate pregnancy phenotypes and for sample collection. On days 2 and 3, both 457 sides of the oviducts were flushed with saline to confirm the presence of 2-cell 458 embryos on day 2 and 8-cell embryos or morula embryos on day 3. On day four, one 459 uterine horn was flushed with saline to confirm the presence of blastocysts. Embryo 460 attachment sites were observed as blue bands soon after intravenous injection of a 461 1% solution of Chicago blue dye (Sigma-Aldrich) in saline on days 5 and 6. When no 462 embryo attachment sites were observed as of day 5, both uterine horns were cut and 463 flushed with saline to collect the embryos. 464 To analyze implantation failure, pregnant mice were sacrificed on day 8 at 1000 h, 465 and implantation sites were histologically assessed. If obvious hematopoietic cell 466 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 22 infiltration was observed, implantation failure was diagnosed. Parturition events were 467 monitored daily from days 19 to 22; all mice were dissected, and the abdominal cavity 468 was observed for findings of miscarriage. 469 To evaluate the phenotype of embryo attachment in wild-type mice over time, we 470 observed the uteri on day 4 morning (day 4 10:00), day 4 evening (day 4 16:00), day 471 4 night (day 4 20:00), day 4 midnight (day 5 0:00), and day 5 morning (day 5 10:00). 472 For day 5 night, the observation time was defined as 18:00–22:00. In preliminary 473 experiments, we confirmed that the phenotypes were equivalent, with no significant 474 changes in the morphology of tissues collected during these 4 h. As described above, 475 samples with visible embryo attachment sites were evaluated for their numbers, and 476 those with no visible embryo attachment sites were evaluated as pregnant specimens 477 by flushing the contralateral uterus with saline and observing the blastocysts. 478 Daily subcutaneous injections of P₄ (2 mg/mouse/day) to p38α uKO mice were 479 performed from day 2 of pregnancy or from the criterion day of pregnancy at 10:00 as 480 previously described15. 481 rLif injections were performed as previously reported11; female mice received rLif 482 (20/i3 µg/head, i.p.) at 9:00 and 18:00 on day 4 of pregnancy. The rLif expression 483 vector was a kind gift from Prof. Eichi Hondo13. 484 485 Transmission electron microscopy (TEM) 486 TEM was performed on mouse uterine specimens collected on day 4 at 10:00 h. The 487 fixation solution was 2% glutaraldehyde-2% paraformaldehyde dissolved in 0.1 M 488 phosphate buffer (pH 7.4). The mice underwent the following perfusion procedures: 489 deep anesthesia was administered, the mouse was fixed in place, and the abdomen 490 to chest was incised; the diaphragm was quickly incised with tweezers and the heart 491 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 23 was exposed; next, a 26G needle was inserted into left ventricle, and saline solution 492 was injected; then, the saline was switched with the aforementioned fixative solution, 493 which was injected to ensure fixative solution spreading to the tissues. Finally, the 494 specimens were kept refrigerated in fixative solution. Specimen processing and 495 imaging were performed at the Hanaichi Institute of Chemical Microscopy after 496 embedding. 497 498 RNA extraction and real time-quantitative PCR (RT-qPCR) 499 RNA was prepared from homogenized frozen tissues, as previously described16. 500 qRT-PCR was performed using THUNDERBIRD SYBR qPCR Mix (TOYOBO). The 501 housekeeping gene Actb was used for internal standardization of mRNA expression. 502 Relative expression levels were determined using the ΔΔ Ct method51. The following 503 primers were used. 504 Gene Strand Sequence Mouse Actb Forward TGTTACCAACTGGGACGACA Reverse GGGGTGTTGAAGGTCTCAAA Mouse Lif Forward GCTATGTGCGCCTAACATGA Reverse AGTGGGGTTCAGGACCTTCT Reverse CCTGATTAAACACAGCCCAGCA Mouse Cdh1 Forward TGATGTTGCTGTCCCCAAGT Reverse CATCAACCGGCTTAATGGTG 505 H&E staining and immunostaining 506 H&E staining and immunostaining of uterine tissues were performed using 507 paraffin-embedded sections (6 μ m) or frozen sections (12 μ m) as previously 508 described15. For immunohistochemistry, the sections were incubated overnight with 509 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 24 primary antibodies including p38 (8690, Cell Signaling Technology, 1:800), pp38 510 (4511, Cell Signaling Technology, 1:800), Esr1 (ab32063, Abcam, 1:200), Pgr 511 (ab63605, Abcam, 1:100), pStat3 (ab76315, Abcam, 1:100), FOXA2 (8186, Cell 512 Signaling Technology, 1:200), and COX2 (AA570-598, Cayman,1:200). For 513 immunohistochemistry, signals were detected using a DAB substrate kit (#425011, 514 Nichirei) after incubation with horseradish peroxidase-conjugated secondary 515 antibodies (K4003, Dako). The images were captured using the Leica DM5000 B light 516 microscope. 517 For immunofluorescence analysis of the paraffin sections, the sections were 518 incubated overnight with primary antibodies, including CK8 (DSHB, 1:500), and 519 signals were detected using Alexa Fluor 488-conjugated anti-rat immunoglobulin G 520 (Thermo Fisher Scientific, A11006,1:500); nuclei were stained with 521 6-diamidino-2-phenylindole (DAPI) (Dojindo, 1:500). For immunofluorescence 522 analysis of frozen sections, sections were incubated overnight with primary 523 antibodies, including Ki67 (20701, Cell Signaling Technology, 1:200, Alexa Fluor® 524 555 Conjugate), Ecad (3199, Cell Signaling Technology,1:200, Alexa Fluor® 488 525 Conjugate), and β catenin (83539, Cell Signaling Technology, 1:200, Alexa Fluor® 526 555 Conjugate). Nuclei were detected using 6-diamidino-2-phenylindole DAPI (1:500). 527 Images were captured using an AXR microscope (Nikon). 528 529 Automated western blots with simple western (WES) 530 Proteins were extracted from cryopreserved and homogenized day 4 uterine tissues 531 using RIPA buffer (Sigma) supplemented with a proteinase inhibitor cocktail (Sigma) 532 and phosphatase inhibitor cocktail (Sigma). 533 Equal amounts of protein (2 µg/µl) were loaded into 12–230 kDa separation module 534 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 25 kit and analyzed using the Protein Simple Wes® System (Protein Simple, San Jose, 535 CA, USA) following the manufacturer's instructions. The antibodies used included 536 Actin (C-11, sc-1615, Santa Cruz, 1:2000), Stat3 (4904, Cell Signaling Technology, 537 1:2000), and pStat3 (9145, Cell Signaling Technology, 1:2000). Anti-goat IgG and 538 anti-rabbit IgG antibodies were used as secondary antibodies. Actin served as the 539 loading control. 540 541 3D visualization of uterine endometrial luminal epithelium 542 3D visualization of the day 1–4 uteri or day 5 and 6 implantation sites was performed 543 as previously reported10. To stain luminal and glandular epithelial cells, day 1–6 544 tissues were incubated with anti-E-cadherin antibodies (Cell Signaling Technology, 545 24E10, 1:500), followed by incubation with an anti-rabbit antibody conjugated with 546 Alexa 555 (A21428, Thermo Fisher Scientific, 1:500). 3D images were acquired using 547 the LSM 880 (Zeiss) and AXR (Nikon) microscopes. The surface tool in Imaris 548 (version 9.8; Oxford Instruments) was used to construct a 3D structure from the 549 images. 550 551 Measurement of serum E₂ and P4 levels 552 Blood samples were collected from mice on the indicated day of pregnancy. Serum P4 553 levels were measured as described previously38, using a progesterone 554 enzyme-linked immunosorbent assay (ELISA) kit (582601, Cayman). Serum E₂ levels 555 were measured using an estradiol ELISA kit (501890, Cayman). 556 557 Spatial transcriptomics 558 Spatial transcriptomes were analyzed using 10x Visium (10x Genomics) following the 559 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 26 manufacturer’s protocol. Day 6 uteri from control and Taz-uKO females were 560 collected. Frozen sections (10 μ m) were mounted on gene expression slides and sent 561 to KOTAI Bio Inc. (Osaka, Japan) for processing. Following a 30-minute Proteinase K 562 reaction, the sections were hybridized with spatial tags on the slides and 563 reverse-transcribed in situ. The cDNAs were analyzed by RNA sequencing using 564 DNBseq (MGI) with 300 million reads per sample. Raw FASTQ files and microscope 565 slide images for each sample were processed with Space Ranger software (version 566 1.1, 10× Genomics) using the “spaceranger count” pipeline, involving STAR with the 567 default parameters for aligning reads against the mouse reference genome mm10 568 “refdata-gex-mm10-2020-A.” This pipeline uses Visium spatial barcodes to generate 569 a feature spot matrix with unique molecular identifier counts. Clustering analysis was 570 performed using Seurat (version 5.0.0)52 and clusters were visualized using UMAP. 571 Differentially expressed genes between genotypes were identified using an adjusted 572 p-value 1.5. Metascape33 and Enrichr53 were used to 573 analyze the GO terms and upstream transcription factors within each cluster, 574 respectively. The Mouse Visium data were deposited to the GEO database 575 (Accession No. GSE305995). 576 577 scRNA-seq and data analysis 578 The 10x Genomics Chronium FRP protocol was followed for scRNA-seq analysis. On 579 days 4 and 5 for WT , or day 4 evening and midnight for p38α floxed and uKO, uterine 580 horns were excised, snap-frozen, and sent to Takara Bio Co.(Osaka, Japan). After 581 fixing the cells with formaldehyde, a single-cell suspension was prepared using a 582 GentleMACS (Miletenyi Biotec). The cells were used for RNA sequencing library 583 preparation using the Chromium Next GEM Single Cell Fixed RNA Sample 584 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 27 Preparation Kit, Chromium Fixed RNA Kit, Mouse Transcriptome, Chromium Mouse 585 Transcriptome Probe Set v1.0.1, Chromium Next GEM Chip Q Single Cell Kit, Dual 586 Index Kit TS Set A, and Chromium X (10x genomics, USA). Paired-end sequencing 587 was performed on an Illumina next-generation sequencer (NovaSeq 6000; Illumina). 588 Raw FASTQ files were processed using Cell Ranger software (10x Genomics, USA), 589 and Seurat (http://www.satijalab.org/seurat) v5.0.052 was used to process read counts. 590 Cell trajectory was determined using Monocle354. CellChat was used for the cell-cell 591 interaction assay. The mouse scRNA-seq data were deposited in the GEO database 592 (Accession No. GSE296581 and GSE305994). 593 594 Statistical analyses 595 Statistical analyses were performed using a two-tailed Student’s t-test or one-way 596 analysis of variance (ANOVA), followed by Bonferroni post-hoc tests, in GraphPad 597 Prism10. Statistical significance was set at P < 0.05. 598 599 Study approval 600 All animal experiments were approved by the Institutional Animal Experiment 601 Committee of the University of Tokyo Graduate School of Medicine (approval 602 numbers P20-076 and A2023M165). 603 604 Data and material availability 605 The RNA-seq experimental data will be made publicly available upon publication 606 (GSE305994 and GSE305995). This study did not involve the development of custom 607 code or algorithms. All the software used in this study is publicly available and is cited 608 in the main text and Methods sections. 609 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 28

610 1. 1 in 6 people globally affected by infertility: WHO. Saudi Med J 44, 524–525 611 (2023). 612 2. Dey SK , et al. Molecular cues to implantation. Endocrine reviews 25, 341–373 613 (2004). 614 615 3. Fukui Y, et al. Uterine receptivity, embryo attachment, and embryo invasion: 616 Multistep processes in embryo implantation. Reprod Med Biol 18, 234–240 617 (2019). 618 619 4. Haraguchi H , et al. MicroRNA-200a locally attenuates progesterone signaling 620 in the cervix, preventing embryo implantation. Mol Endocrinol 28, 1108–1117 621 (2014). 622 623 5. Hirota Y . Progesterone governs endometrial proliferation-differentiation 624 switching and blastocyst implantation. Endocr J 66, 199–206 (2019). 625 626 6. Gebril M , et al. Uterine Epithelial Progesterone Receptor Governs Uterine 627 Receptivity Through Epithelial Cell Differentiation. Endocrinology 161, 628 (2020). 629 630 7. Zhang S , et al. Uterine Rbpj is required for embryonic-uterine orientation and 631 decidual remodeling via Notch pathway-independent and -dependent 632 mechanisms. Cell Res 24, 925–942 (2014). 633 634 8. Egashira M, Hirota Y . Uterine receptivity and embryo-uterine interactions in 635 embryo implantation: lessons from mice. Reprod Med Biol 12, 127–132 636 (2013). 637 638 9. Fukui Y, et al. Uterine Epithelial LIF Receptors Contribute to Implantation 639 Chamber Formation in Blastocyst Attachment. Endocrinology 162, (2021). 640 641 10. Yuan J, Deng W, Cha J, Sun X, Borg JP , Dey SK. Tridimensional visualization 642 reveals direct communication between the embryo and glands critical for 643 implantation. Nature communications 9, 603 (2018). 644 645 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 29 11. Aikawa S , et al. Spatiotemporal functions of leukemia inhibitory factor in 646 embryo attachment and implantation chamber formation. Cell Death Discov 10, 647 481 (2024). 648 649 12. Hiraoka T , et al. Differential roles of uterine epithelial and stromal STAT3 650 coordinate uterine receptivity and embryo attachment. Sci Rep 10, 15523 651 (2020). 652 653 13. Kobayashi R , et al. The contribution of leukemia inhibitory factor (LIF) for 654 embryo implantation differs among strains of mice. Immunobiology 219, 655 512–521 (2014). 656 657 14. Sun X, Bartos A, Whitsett JA, Dey SK. Uterine deletion of Gp130 or Stat3 658 shows implantation failure with increased estrogenic responses. Mol 659 Endocrinol 27, 1492–1501 (2013). 660 661 15. Akaeda S, et al. Retinoblastoma protein promotes uterine epithelial cell cycle 662 arrest and necroptosis for embryo invasion. EMBO Rep 22, e50927 (2021). 663 664 16. Fukui Y, et al. The EZH2-PRC2-H3K27me3 axis governs the endometrial cell 665 cycle and differentiation for blastocyst invasion. Cell Death Dis 14, 320 (2023). 666 667 17. Coulthard LR, White DE, Jones DL, McDermott MF, Burchill SA. p38(MAPK): 668 stress responses from molecular mechanisms to therapeutics. Trends Mol 669 Med 15, 369–379 (2009). 670 671 18. Morrison DK. MAP kinase pathways. Cold Spring Harb Perspect Biol 4, 672 (2012). 673 674 19. Nishida K, et al. p38alpha mitogen-activated protein kinase plays a critical role 675 in cardiomyocyte survival but not in cardiac hypertrophic growth in response to 676 pressure overload. Mol Cell Biol 24, 10611–10620 (2004). 677 678 20. Zohn IE, Li Y, Skolnik EY, Anderson KV, Han J, Niswander L. p38 and a 679 p38-interacting protein are critical for downregulation of E-cadherin during 680 mouse gastrulation. Cell 125, 957–969 (2006). 681 682 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 30 21. Liu Y, Martinez L, Ebine K, Abe MK. Role for mitogen-activated protein kinase 683 p38 alpha in lung epithelial branching morphogenesis. Dev Biol 314, 224–235 684 (2008). 685 686 22. Wen HC , et al. p38 α Signaling Induces Anoikis and Lumen Formation During 687 Mammary Morphogenesis. Sci Signal 4, ra34 (2011). 688 689 23. Canovas B, Nebreda AR. Diversity and versatility of p38 kinase signalling in 690 health and disease. Nat Rev Mol Cell Biol 22, 346–366 (2021). 691 692 24. Chen Y, Ramakrishnan DP , Ren B. Regulation of angiogenesis by 693 phospholipid lysophosphatidic acid. Front Biosci (Landmark Ed) 18, 852–861 694 (2013). 695 696 25. Tang Y, et al. P38alpha MAPK is a gatekeeper of uterine progesterone 697 responsiveness at peri-implantation via Ube3c-mediated PGR degradation. 698 Proc Natl Acad Sci U S A 119, e2206000119 (2022). 699 700 26. Cha J, Sun X, Dey SK. Mechanisms of implantation: strategies for successful 701 pregnancy. Nat Med 18, 1754–1767 (2012). 702 703 27. Cha J , et al. Appropriate crypt formation in the uterus for embryo homing and 704 implantation requires Wnt5a-ROR signaling. Cell Rep 8, 382–392 (2014). 705 706 28. Arora R , et al. Insights from imaging the implanting embryo and the uterine 707 environment in three dimensions. Development 143, 4749–4754 (2016). 708 709 29. Ye X. Uterine Luminal Epithelium as the Transient Gateway for Embryo 710 Implantation. Trends Endocrinol Metab 31, 165–180 (2020). 711 712 30. Psychoyos A. Endocrine control of egg implantation. American Physiology 713 Society (1973). 714 715 31. Xie Z , et al. Gene Set Knowledge Discovery with Enrichr. Curr Protoc 1, e90 716 (2021). 717 718 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 31 32. Lau E, Ronai ZA. ATF2 - at the crossroad of nuclear and cytosolic functions. J 719 Cell Sci 125, 2815–2824 (2012). 720 721 33. Zhou Y , et al. Metascape provides a biologist-oriented resource for the 722 analysis of systems-level datasets. Nature communications 10, 1523 (2019). 723 724 34. Stewart CL, et al. Blastocyst implantation depends on maternal expression of 725 leukaemia inhibitory factor. Nature 359, 76–79 (1992). 726 727 35. Chen JR, Cheng JG, Shatzer T, Sewell L, Hernandez L, Stewart CL. Leukemia 728 inhibitory factor can substitute for nidatory estrogen and is essential to 729 inducing a receptive uterus for implantation but is not essential for subsequent 730 embryogenesis. Endocrinology 141, 4365–4372 (2000). 731 732 36. Song H, Lim H, Das SK, Paria BC, Dey SK. Dysregulation of EGF family of 733 growth factors and COX-2 in the uterus during the preattachment and 734 attachment reactions of the blastocyst with the luminal epithelium correlates 735 with implantation failure in LIF-deficient mice. Mol Endocrinol 14, 1147–1161 736 (2000). 737 738 37. Zhou C , et al. Sequential activation of uterine epithelial IGF1R by stromal IGF1 739 and embryonic IGF2 directs normal uterine preparation for embryo 740 implantation. J Mol Cell Biol 13, 646–661 (2021). 741 742 38. Matsumoto L , et al. HIF2α in the uterine stroma permits embryo invasion and 743 luminal epithelium detachment. J Clin Invest 128, 3186–3197 (2018). 744 745 39. Bondarenko V, et al. Embryo-uterine interaction coordinates mouse 746 embryogenesis during implantation. EMBO J 42, e113280 (2023). 747 748 40. Cha J, Sun X, Dey SK. Mechanisms of implantation: strategies for successful 749 pregnancy. Nat Med 18, 1754–1767 (2012). 750 751 41. Ye X , et al. LPA3-mediated lysophosphatidic acid signalling in embryo 752 implantation and spacing. Nature 435, 104–108 (2005). 753 754 42. Hama K, et al. Embryo spacing and implantation timing are differentially 755 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 32 regulated by LPA3-mediated lysophosphatidic acid signaling in mice. Biol 756 Reprod 77, 954–959 (2007). 757 758 43. Jinno M, Ozaki T, Iwashita M, Nakamura Y, Kudo A, Hirano H. Measurement of 759 endometrial tissue blood flow: a novel way to assess uterine receptivity for 760 implantation. Fertil Steril 76, 1168–1174 (2001). 761 762 44. Uchida H. Epithelial mesenchymal transition in human menstruation and 763 implantation. Endocr J 71, 745–751 (2024). 764 765 45. Owusu-Akyaw A, Krishnamoorthy K, Goldsmith LT, Morelli SS. The role of 766 mesenchymal-epithelial transition in endometrial function. Hum Reprod 767 Update 25, 114–133 (2019). 768 769 46. Whitby S, Zhou W, Dimitriadis E. Alterations in Epithelial Cell Polarity During 770 Endometrial Receptivity: A Systematic Review. Front Endocrinol (Lausanne) 771 11, 596324 (2020). 772 773 47. Joseph S , et al. MAPK14/p38α shapes the molecular landscape of endometrial 774 cancer and promotes tumorigenic characteristics. Cell Rep 44, 115104 (2025). 775 776 48. Hiraoka T, Osuga Y , Hirota Y. Current perspectives on endometrial receptivity: 777 A comprehensive overview of etiology and treatment. J Obstet Gynaecol Res 778 49, 2397–2409 (2023). 779 780 49. Soyal SM , et al. Cre-mediated recombination in cell lineages that express the 781 progesterone receptor. Genesis 41, 58–66 (2005). 782 783 50. Daikoku T , et al. Conditional deletion of Msx homeobox genes in the uterus 784 inhibits blastocyst implantation by altering uterine receptivity. Dev Cell 21, 785 1014–1025 (2011). 786 787 51. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using 788 real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 789 402–408 (2001). 790 791 52. Hao Y, et al. Dictionary learning for integrative, multimodal and scalable 792 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 33 single-cell analysis. Nat Biotechnol 42, 293–304 (2024). 793 794 53. Chen EY , et al. Enrichr: interactive and collaborative HTML5 gene list 795 enrichment analysis tool. BMC Bioinformatics 14, 128 (2013). 796 797 54. Cao J , et al. The single-cell transcriptional landscape of mammalian 798 organogenesis. Nature 566, 496–502 (2019). 799 800 801 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 34 Acknowledgments 802 We thank Ms. Atsumi Miura for providing technical assistance. We are grateful to 803 Francesco J. DeMayo (National Institute of Environmental Health Sciences) for 804 providing Pgr-Cre mice, and to Kinya Otsuki (Osaka University) for providing 805 Mapk14-floxed mice. 806 807 Funding 808 This work was supported by the Japan Society for the Promotion of Science (JSPS) 809 KAKENHI (grant nos. 23K08278, 23K27176, 24K22157, 24K21911, 25K02779, 810 25H01065), Japan Agency for Medical Research and Development (AMED) (grant no. 811 JP25gn0110085, JP24gn0110069, JP25gk0210039, JP24lk0310083, 812 JP25gn0110097, JP25gk0210042 and JP25gk0210045), Children and Families 813 Agency (Grant Number JPMH23DB0101), Japan Science and Technology Agency 814 (JST) Fusion Oriented Research for Disruptive Science and Technology (FOREST) 815 (grant no. JPMJFR210H), Mochida Memorial Foundation for Medical and 816 Pharmaceutical Research, Uehara Memorial Foundation, Inoue Foundation for 817 Science, Astellas Foundation for Research on Metabolic Disorders, The Naito 818 Foundation and the fund of joint research with NIPRO corporation. 819 820 Author contributions 821 Conceptualization: S.A. and Y .H.; Funding acquisition: S.A. and Y.H.; 822 Investigation:C.I., S.A., Y.F., X.H., R.S.H., D.H., T.H., M.M.; Data analysis: C.I., S.A.; 823 Data interpretation: C.I., S.A., Y .H.; Project administration and supervision: Y .H.; 824 Writing - original draft preparation: C.I., S.A.; Writing – review and editing: S.A. 825 826 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 35 Competing interests 827 All authors declare they have no competing interests. 828 829 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 36 Figure legends 830 Fig. 1 Three-dimensional observation of endometrial lumen morphology before 831 embryo attachment. a A schematic diagram of embryo implantation. b Schematic of 832 the longitudinal and cross-sectional views of the uterine horns. c Two-dimensional 833 (2D) and Tridimensional (3D) views of the uterine epithelium stained for E-cadherin. 834 The luminal epithelium was segmented and magenta colored using Imaris. Scale bar: 835 1 mm (left) and 200 µm (middle and right). Schematic diagrams of luminal shapes on 836 each pregnancy day are shown in the right panel. d The area of the luminal 837 epithelium per sectional view was quantified. n = 3 for each sample and n = 3 for each 838 day of pregnancy; in total, n = 9 sections were quantified. Data are presented as 839 means ± SEM, ***P < 0.001, ****P < 0.0001 by one-way ANOVA followed by 840 Bonferroni’s post-hoc test. 841 842 Fig. 2 Dynamic morphological changes in the endometrial luminal epithelium 843 occur on day 4 night, just before embryo attachment. a Representative 844 photographs of pregnant uteri from day 4 morning to day 5 morning, which were 845 injected with blue dye to depict embryo attachment sites. Scale bar: 1 cm. b 846 Percentage of implantation-positive females in (a). The number of replicates and 847 percentage of implantation-positive females are shown above each bar. c The 848 number of implantation sites in (a) and (b). Data are represented as means ± SEM. d 849 2D and 3D longitudinal views of uterine epithelia stained for E-cadherin. In the right 850 panels, the luminal epithelium is segmented and colored in magenta using Imaris. 851 Scale bar: 1 mm (left) and 200 µm (right). Asterisks indicate the locations of embryos. 852 853 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 37 Fig. 3 Stress-related signals are activated in the uteri during the 854 peri-attachment phase. a UMAP of scRNA-seq cell types on days 4 and 5 of 855 pregnancy. Dots indicate individual cells, and colors indicate different clusters. b 856 Enrichment analyses for upstream transcription factors (left) and gene ontology (GO) 857 (right) of highly expressed genes in the LE_activated vs. LE cells. c Network of GO 858 terms related to the upregulated genes in the LE_activated cells compared with those 859 in the LE cells. Each node represents an enriched term and is colored according to its 860 cluster ID. d UMAP of different stromal cell types on days 4 and 5 of pregnancy. Dots 861 indicate individual cells, and colors represent different clusters. e Enrichment 862 analyses for upstream transcription factors (left) and gene ontology (GO) (right) of 863 highly expressed genes in the Attached cluster compared with those in the other 864 stromal clusters. f Network of GO terms related to the upregulated genes in the 865 Activated cells compared with those in the other stromal clusters. Each node 866 represents an enriched term and is colored according to its cluster ID. g 867 Representative images of p38α (top) and phosphorylated p38α (pp38α ; bottom) 868 immunohistochemistry during days 1, 4, 6, and 8 of pregnancy. Scale bar: 100 µm. 869 LE: luminal epithelia, GE: glandular epithelia, Str: Stroma, Em; embryo, Le; Luminal 870 epithelium, St; Stroma. Arrowheads indicate embryos. At least three independent 871 samples were evaluated for each day of pregnancy. 872 873 Fig. 4 Morphological changes in the pre-attachment endometrial luminal 874 epithelium are impaired in uterus-specific p38α KO mice. a Efficient p38α 875 deletion was confirmed by immunostaining for p38α and phosphorylated p38α 876 (pp38α ) in the uteri on day 4 of pregnancy. Scale bar = 100 µm. b Average litter size 877 for each genotype. The numbers of replicates are shown in the graphs. Data 878 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 38 represent the mean/i3 ±/i3 SEM, and ****P < 0.0001 by Student’s t-test. c Comparable 879 numbers of flushed blastocysts (left) and their morphology (right). The graph shows 880 the number of dams tested. Data represent the mean/i3 ±/i3 SEM, n.s.: not significant 881 according to Student’s t-test. d Representative photographs of the uteri from each 882 genotype (left) and the average number of implantation sites on day 5 of pregnancy. 883 Flushed embryos from p38α uKO are shown next to the uterine photographs. Scale 884 bar = 1 mm (uteri) and 100 µm (embryos). The number of replicates is shown on the 885 graph. Data represent the mean/i3 ±/i3 SEM, and ****P < 0.0001 by Student’s t-test. e 886 Representative photographs of the uteri from each genotype (left) and the average 887 number of implantation sites on day 6 of pregnancy. Scale bar = 1 mm. The number 888 of replicates is shown on the graph. Data represent the mean/i3 ±/i3 SEM, and ****P < 889 0.0001 by Student’s t-test. f Representative images of day 5 pregnant uteri stained for 890 E-Cadherin in 3D. The luminal and glandular epithelia were segmented and colored 891 magenta and cyan, respectively. Scale bar = 100 µm. g Representative 3D views of 892 luminal epithelia during days 1–4 of pregnancy, segmented based on epithelial 893 staining for E-cadherin. Scale bar = 100 µm. 894 895 Fig. 5 P₄ supplementation to p38α uKO mice partially rescues structural 896 changes and proliferation-differentiation switching (PDS) in the endometrial 897 luminal epithelium before embryo attachment. a, b Representative images of Ki67 898 immunofluorescence in the luminal epithelium of the uteri on day 4 morning. The 899 percentages of Ki67-positive cells per total luminal cells are shown in (b). The number 900 of replicates is shown on the graph. Data represent the mean/i3 ±/i3 SEM, **P < 0.01 by 901 Student’s t-test. c The schedule of P4 treatment to p38α uKO females during days 1 902 to 5 of pregnancy. d Representative 3D longitudinal views of luminal epithelia from 903 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 39 the control and p38α uKO uteri with or without P4 treatment, segmented from 904 epithelial staining for E-cadherin. Scale bar = 100 µm. e Cross-sectional view of the 905 luminal epithelia in (d). Scale bar = 100 µm. f, g Representative images of Ki67 906 immunofluorescence in the luminal epithelium on day 4 morning in the uteri from 907 control and p38α uKO mice with or without P4 treatment. The percentages of 908 Ki67-positive cells per total luminal cells are shown in (g). The number of replicates is 909 shown on the graph. Data represent the mean/i3 ±/i3 SEM, **P < 0.01 and ***P < 0.001 910 by one-way ANOVA followed by Bonferroni’s post-hoc test. 911 912 Fig. 6 Impaired Lif-Stat3 pathway affects embryo attachment in p38α uKO mice. 913 a Representative images of Lif in situ hybridization (top and middle) and 914 phosphorylated Stat3 (pStat3) immunostaining (bottom) in day 4 uteri from control 915 and p38α uKO mice with or without P4 treatment. Epithelial cells immunostained for 916 CK-8 are shown in the top and middle panels. The area indicated by a dashed line in 917 the top panel is shown in the middle panel. Scale bar = 50 µm. b The schedule of P4 918 and rLif treatment in p38α uKO female mice during days 1 to 5 of pregnancy. c 919 Representative photographs of day 5 pregnant uteri from control and p38α uKO mice 920 with or without P4 and rLif treatment. Arrowheads indicate sites of embryo attachment. 921 Scale bar = 5 mm. d The number of implantation sites in (c) was calculated. The 922 number of replicates is shown on the graph. Data represent the mean/i3 ±/i3 SEM, ***P 923 < 0.001, ****P < 0.0001 and n.s.: not significant by one-way ANOVA followed by 924 Bonferroni’s post-hoc test. f e Representative images of COX2 immunostaining on 925 day 5 implantation sites from control and p38α uKO uteri treated with P4 and rLif. 926 Scale bar = 100 µm. M: mesometrial pole; AM: anti-mesometrial pole. Arrowheads 927 indicate embryos. f A representative image of day 20 pregnant uteri from p38α uKO 928 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 40 mice treated with P4 and rLif. The percentage of deliveries per total number of 929 pregnant females is shown on the right graph. The number of replicates is shown on 930 the graph. Data represent the mean/i3 ±/i3 SEM, ***P < 0.001 by Student’s t-test. 931 932 Fig. 7 Embryo attachment in p38α uKO mice is rescued by supplementation 933 with P₄ and Lif, but luminal positioning remains inappropriate. a Representative 934 3D views of the luminal epithelia collected from each genotype on days 5 and 6 of 935 pregnancy. Scale bar = 100 µm, asterisks indicate embryos. b Representative 3D 936 views of luminal epithelia on day 4 midnight, immediately before embryo attachment, 937 collected from each genotype. Scale bar = 100 µm, asterisks indicate embryos. The 938 graphs in the right panels show the luminal shapes for each genotype and condition. 939 c Average area per sectional view calculated from the images shown in (b). Data 940 represent the mean/i3 ±/i3 SEM, and P-values were determined by one-way ANOVA 941 followed by Bonferroni’s post hoc test. d Average number of luminal branches per 942 sectional view calculated from the images shown in (b). Data represent the 943 mean/i3 ±/i3 SEM, and *P < 0.05, ***P < 0.001, ****P < 0.0001, n.s.: not significant by 944 one-way ANOVA followed by Bonferroni’s post-hoc test. 945 946 Fig. 8 p38α plays an important role in stromal differentiation to ensure 947 appropriate epithelial-stromal interactions before embryo attachment. a UMAP 948 of scRNA-seq for various cell types from control uteri on day 4 evening and midnight, 949 and from p38α uKO with or without P4 and rLif treatment on day 4 midnight. Dots 950 indicate individual cells, and colors indicate different clusters. b UMAP of scRNA-seq 951 of stromal cell types from control uteri on day 4 evening and midnight, and from p38α 952 uKO with or without P4 and rLif treatment on day 4 midnight. Dots indicate individual 953 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 41 cells, and colors indicate different clusters. c UMAP plots show the diffusion 954 pseudotime of stromal cells. The colors of the spots indicate the pseudotime from 955 early (blue) to late (dark red). d Boxplot showing the distribution of pseudo-time within 956 different stromal cell clusters. Colors and labels indicate the cell types corresponding 957 to those shown in (b). e Heatmap showing the outgoing (left) or incoming (right) 958 communication strengths of key pathways between the major uterine endometrial cell 959 types in each genotype and condition. 960 961 Fig. 9 p38α uKO mice supplemented with P₄ and Lif show embryo invasion 962 failure owing to sustained epithelial polarity. a, b Luminal epithelial ( a) and 963 stromal (b) expression of Igf1 and Wnt5a determined using scRNA-seq in each 964 genotype and condition. ***P < 0.001, ****P < 0.0001, ns: not significant by one-way 965 ANOVA followed by Bonferroni’s post-hoc test. c Representative images of 966 immunostaining for β -actin (red) and E-cadherin (green) in day 4 uteri from each 967 genotype. M: mesometrial pole, AM: anti-mesometrial pole. Scale bar = 50 µm. d 968 Representative images of immunostaining for CK-8 (green) on day 6 implantation 969 sites from the control and p38α uKO mice treated with P4 and rLif. Areas demarcated 970 by dashed lines in the top panels are shown in the bottom panels. M: mesometrial 971 pole, AM: anti-mesometrial pole. Asterisks indicate embryos. Scale bar = 200 µm 972 (top) and 100 µm (bottom). e H&E staining (upper) and visualization of the spatial 973 transcriptome (lower) in day 5 implantation sites from the control and p38α uKO mice 974 treated with P4 and rLif. M: mesometrial pole; AM: anti-mesometrial pole, LE: luminal 975 epithelia, GE: glandular epithelia, Str: stroma. Arrowheads indicate embryos. f UMAP 976 analysis of the spatial transcriptome dataset colored according to cell type for day 5 977 implantation sites in control (left) and p38α uKO uteri treated with P4 and rLif (right). 978 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint 42 Each dot color is mapped according to the spatial transcriptome visualized in the 979 uterine sections, as shown in (e). g, h Network of GO terms related to the upregulated 980 (g) or downregulated genes (h) in Str_uKO_specific compared to other stromal 981 clusters. Each node represents an enriched term and is colored according to its 982 cluster ID. 983 984 Fig. 10 The role of uterine p38α in embryo attachment. p38α contributes to 985 flattening of the endometrial luminal surface and induction of luminal narrowing prior 986 to embryo attachment, through pathways activated by P₄ . It is also suggested that 987 uterine p38α induces glandular Lif, thus inducing embryo attachment. In addition to 988 these pathways, p38α contributes to a decrease in epithelial polarity by activating 989 epithelial-stromal interactions, thus resulting in successful embryo attachment and 990 invasion. 991 992 993 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted February 2, 2026. ; https://doi.org/10.64898/2026.01.29.702684doi: bioRxiv preprint