Hypoxia is not required for human endometrial breakdown or repair in a xenograft model of menstruation

Abstract

Menstrual endometrial breakdown induced by estradiol and progesterone withdrawal is regularly attributed to vasospasm of spiral arteries causing ischemia and hypoxia. We investigated whether hypoxia actually occurred in an in vivo model of menstruation. Three complementary approaches were used to look for signs of hypoxia in fragments of human functionalis xenografted to ovariectomized immunodeficient mice bearing pellets-releasing estradiol and progesterone, and then deprived of ovarian steroids. Hormone withdrawal 21 d after grafting induced menstrual breakdown and MMP expression within 4 d. Local partial oxygen pressure (pO2) was measured by electron paramagnetic resonance using implanted lithium phtalocyanine crystals. In mice with hormone maintenance until sacrifice, pO2 was low one week after grafting (14.8±3.4 mmHg) but increased twofold from the second week when tissue was largely revascularized. After 3 wk, pO2 was not modified by hormone withdrawal but was slightly increased on hormone reimpregnation 4 d after removal (34.7±6.1 mmHg) by comparison with hormone maintenance (27.1±8.6 mmHg). These results were confirmed using fluorescence quenching-based OxyLite measurements. In a further search for signs of hypoxia, we did not find significant HIF1-α immunostaining, nor pimonidazole adducts after hormone withdrawal. We conclude that hypoxia is not needed to trigger menstrual-like tissue breakdown or repair in human endometrial xenograft.