Although it has been reported that hypoxia inducible factor 2 α (Hif2a), a major transcriptional factor inducible by low oxygen tension, is expressed in the mouse uterus during embryo implantation, its role in pregnancy outcomes remains unclear. This study aimed to clarify functions of uterine HIF using transgenic mouse models. Mice with deletion of Hif2a in the whole uterus (Hif2a-uKO mice) showed infertility due to implantation failure. Supplementation with progesterone (P4) and leukemia inhibitory factor (LIF) restored decidual growth arrest and aberrant position of implantation sites in Hif2a-uKO mice, respectively, but did not rescue pregnancy failure. Histological analyses in Hif2a-uKO mice revealed persistence of the intact luminal epithelium, which blocked direct contact between stroma and embryo, inactivation of PI3K-AKT pathway (embryonic survival signal), and failed embryo invasion. Mice with stromal deletion of Hif2a (Hif2a-sKO mice) showed infertility with impaired embryo invasion and those with epithelial deletion of Hif2a (Hif2a-eKO mice) showed normal fertility, suggesting the importance of stromal HIF2α in embryo invasion. This was reflected in reduced expression of membrane type 2 metalloproteinase (MT2-MMP), lysyl oxidase (LOX), VEGF, and adrenomedullin (ADM) in Hif2a-uKO stroma at the attachment site, suggesting that stromal HIF2α regulates these mediators to support blastocyst invasion. These findings provide new insight that stromal HIF2α allows trophoblast invasion through detachment of the luminal epithelium and activation of an embryonic survival signal.

Although cervical pregnancy and placenta previa, in which the embryo and placenta embed in or adjacent to the cervix, are life-threatening complications that result in massive bleeding and poor pregnancy outcomes in women, the incidence of these aberrant conditions is uncommon. We hypothesized that a local molecular mechanism is normally in place to prevent embryo implantation in the cervix. The ovarian hormones progesterone (P(4)) and estrogen differentially direct differentiation and proliferation of endometrial cells, which confers the receptive state for implantation: P(4) dominance causes differentiation of the luminal epithelium but increases stromal cell proliferation in preparation of the uterus for implantation. In search for the cause of cervical nonresponsiveness to implantation, we found that the statuses of cell proliferation and differentiation between the uterus and cervix during early pregnancy are remarkably disparate under identical endocrine milieu in both mice and humans. We also found that cervical levels of progesterone receptor (PR) protein are low compared with uterine levels during this period, and the low PR protein levels are attributed to elevated levels of microRNA(miR)-200a in the cervix. These changes were associated with up-regulation of the P(4)-metabolizing enzyme 20α-hydroxysteroid dehydrogenase (200α-HSD) and down-regulation of its transcriptional repressor signal transducer and activator of transcription 5 in the cervix. The results provide evidence that elevated levels of miR-200a lead to down-regulation of P(4)-PR signaling and up-regulation of (200α-HSD) in the cervix, rendering it nonresponsive to implantation. These findings may point toward not only the physiological but also the pathological basis of the cervical milieu in embryo implantation.

Cellular senescence, defined as an irreversible cell cycle arrest, exacerbates the tissue microenvironment. Our previous study demonstrated that mouse uterine senescent cells were physiologically increased according to gestational days and that their abnormal accumulation was linked to the onset of preterm delivery. We hypothesized that there is a mechanism for removal of senescent cells after parturition to maintain uterine function. In the current study, we noted abundant uterine senescent cells and their gradual disappearance in wild-type postpartum mice. F4/80+ macrophages were present specifically around the area rich in senescent cells. Depletion of macrophages in the postpartum mice using anti-F4/80 antibody enlarged the area of senescent cells in the uterus. We also found excessive uterine senescent cells and decreased second pregnancy success rate in a preterm birth model using uterine p53-deleted mice. Furthermore, a decrease in F4/80+ cells and an increase in CD11b+ cells with a senescence-associated inflammatory microenvironment were observed in the p53-deleted uterus, suggesting that uterine p53 deficiency affects distribution of the macrophage subpopulation, interferes with senescence clearance, and promotes senescence-induced inflammation. These findings indicate that the macrophage is a key player in the clearance of uterine senescent cells to maintain postpartum uterine function.

Although a close connection between uterine regeneration and successful pregnancy in both humans and mice has been consistently observed, its molecular basis remains unclear. We here established a mouse model of decellularized uterine matrix (DUM) transplantation. Resected mouse uteri were processed with SDS to make DUMs without any intact cells. DUMs were transplanted into the mouse uteri with artificially induced defects, and all the uterine layers were recovered at the DUM transplantation sites within a month. In the regenerated uteri, normal hormone responsiveness in early pregnancy was observed, suggesting the regeneration of functional uteri. Uterine epithelial cells rapidly migrated and formed a normal uterine epithelial layer within a week, indicating a robust epithelial-regenerating capacity. Stromal and myometrial regeneration occurred following epithelial regeneration. In ovariectomized mice, uterine regeneration of the DUM transplantation was similarly observed, suggesting that ovarian hormones are not essential for this regeneration process. Importantly, the regenerating epithelium around the DUM demonstrated heightened STAT3 phosphorylation and cell proliferation, which was suppressed in uteri of Stat3 conditional knockout mice. These data suggest a key role of STAT3 in the initial step of the uterine regeneration process. The DUM transplantation model is a powerful tool for uterine regeneration research.

Functions of tumor suppressor p53 and its negative regulator mouse double minute 2 homolog (Mdm2) in ovarian granulosa cells remain to be elucidated, and the current study aims at clarifying this issue. Mice with Mdm2 deficiency in ovarian granulosa cells [Mdm2‐loxP/progesterone receptor (Pgr)‐Cre mice] were infertile as a result of impairment of oocyte maturation, ovulation, and fertilization, and those with Mdm2/p53 double deletion in granulosa cells (Mdm2‐loxP/p53‐loxP/Pgr‐Cre mice) showed normal fertility, suggesting that p53 induction in the ovarian granulosa cells is detrimental to ovarian function by disturbing oocyte quality. Another model of Mdm2 deletion in ovarian granulosa cells (Mdm2‐loxP/anti‐Mullerian hormone type 2 receptor‐Cre mice) also showed subfertility as a result of the failure of ovulation and fertilization, indicating critical roles of ovarian Mdm2 in ovulation and fertilization. Mdm2‐p53 pathway in cumulus granulosa cells transcriptionally controlled an orphan nuclear receptor steroidogenic factor 1 (SF1), a key regulator of ovarian function. Importantly, MDM2 and SF1 levels in human cumulus granulosa cells were positively associated with the outcome of oocyte maturation and fertilization in patients undergoing infertility treatment. These findings suggest that the Mdm2‐p53‐SF1 axis in ovarian cumulus granulosa cells directs ovarian function by affecting their neighboring oocyte quality.—Haraguchi, H., Hirota, Y., Saito‐Fujita, T., Tanaka, T., Shimizu‐Hirota, R., Harada, M., Akaeda, S., Hiraoka, T., Matsuo, M., Matsumoto, L., Hirata, T., Koga, K., Wada‐Hiraike, O., Fujii, T., Osuga, Y. Mdm2‐p53‐SF1 pathway in ovarian granulosa cells directs ovulation and fertilization by conditioning oocyte quality. FASEB J. 33, 2610–2620 (2019). http://www.fasebj.org