Endometrial Hemostasis / Thrombosis and Angiogenesis Unit

Charles Lockwood, M.D.

Our unit consists of a team of researchers and clinicians that are devoted to understanding the basic mechanisms underlying the maintenance of hemostasis during blastocyst invasion and the establishment of the uteroplacental circulation as well as the paradoxically controlled hemorrhage of menstruation. Progesterone stimulates the estrogen-primed human endometrium to undergo decidualization, the process of growth and differentiation that transforms precursor stromal cells into decidual cells. Decidualization begins around blood vessels of the midluteal-phase endometrium. Under the continued influence of estradiol (E2) and progesterone, the decidualization reaction spreads throughout the late luteal phase and gestation. Implantation is initiated by attachment of the blastocyst to the uterine luminal epithelium. Subsequently, syncytiotrophoblasts invade the underlying endometrial stromal compartment. They breach capillaries and venules that are enmeshed in stromal cells at various stages of decidualization to establish the primordial utero-placental circulation. Extravillous cytotrophoblasts then penetrate the uterine spiral arteries and initiate morphological changes that increase intervillous blood flow. These processes provide the developing embryo with essential oxygen and nutrients before placentation. They risk decidual hemorrhage, however, which can lead to spontaneous abortion, placental abruption, and preterm birth. In species with a hemochorial placenta, the degree of trophoblast invasiveness is positively corrrelated with the extent of decidualization with human trophoblasts and human endometrium, respectively, displaying the most invasiviness and most extensive decidualization. Moreover, the association of ectopic pregnancy with hemorrhage is highest in species lacking a true decidua. Their localiziation at perivascular sites postions decidual cells to promote local hemostasis, thus counteracting the threat of hemorrhage during endovascular trophoblast invasion and subsequent remodeling of endometrial blood vessels.

We showed that decidualized stromal cells of luteal phase and pregnant human endometrium express tissue factor (TF), the primary initiator of hemostasis, and plasminogen activator inhibitor (PAI-1) the major inhibitor of fibrinolysis, thereby suggesting a mechanism by which perivascular decidual cells can mitigate the risk of hemorrhage during endovascular trophoblast invasion. Progestins enhanced TF and PAI-1 mRNA and protein levels in monolayers of human endometrial stromal cells (HESCs), with estradiol (E2) + progestin, further enhancing TF and PAI-1 levels despite a lack of response to E2 alone. This differential ovarian steroid response has been found for several decidualization markers. Further studies with cultured HESCs established that elevated TF and PAI-1 levels are mediated by the progesterone receptor and are maintained for weeks in response to E2 plus progestin, thus simulating the chronic upregulation of TF and PAI-1 levels observed in decidualized HESCs in vivo. Recent studies revealed that elevated TF and PAI-1 expression during in vitro decidualization of HESCs involved both the EGFR and progesterone receptor. Thus, enhancement of TF and PAI-1 mRNA and protein levels in the HESCs required co-incubation with a progestin (MPA) and an EGFR agonist such as EGF or TGF-alpha.

Further studies in our lab established that progestin-enhanced TF expression in HESCs was trancriptionally regulated. The absence of canonical estrogen and progesterone response elements from either the TF or PAI-1 gene promoters suggested that the progestational induction was mediated by another intermediate. Indeed we found that the transcriptional factor Sp1 was central to this induction.

We are also investigating the pathophysiology of uterine thrombosis and of abnormal uterine bleeding (AUB) associated with the perimenarchal and perimenopausal periods, anovulation, myomas, polyps, carcinomas and long-term progestin-only contraceptive therapy. Of key interest to these studies are the identification of pathways which may lead to aberrant angiogenesis.

While aberrant angiogenesis may lead to AUB, uteroplacental thrombosis promotes adverse pregnancy outcomes including fetal death, growth restriction and, paradoxically, abruption at sites of ischemic spiral arteries. Inherited thrombophilias are both a leading cause of maternal thromboembolism and are associated with an increased risk of certain adverse pregnancy outcomes including second and third trimester fetal loss, abruptions, severe intrauterine growth restriction (IUGR) and early-onset, severe preeclampsia. Current information suggests that all patients with a history of prior venous thrombotic events and those with these characteristic adverse pregnancy events should be evaluated for thrombophilias. The most common, clinically significant, inherited thrombophilias are heterozygosity for the factor V Leiden and prothrombin G20210A mutations. The autosomal dominant deficiencies of protein C and protein S are of comparable thrombogenic potential but are far less common. Homozygosity for the 4G/4G mutation in the type-1 plasminogen activator inhibitor gene and the thermolabile variant of the methylenetetrahydrofolate reductase gene, the leading cause of hyperhomocysteinemia, while relatively common, confer a low risk of thrombosis. In contrast, autosomal dominant antithrombin deficiency and homozygosity or compound heterozygosity (i.e., carriers of one copy of each mutant allele) for the factor V and prothrombin mutations are very rare but highly thrombogenic states. Regardless of their antecedent histories, pregnant patients with these highly thrombogenic conditions are at very high risk for both thromboembolism and characteristic adverse pregnancy outcomes and require full therapeutic heparin therapy throughout pregnancy and need at least 6 weeks of postpartum oral anticoagulation. There is also compelling evidence that patients with the less thrombogenic thrombophilias and a history of venous thrombotic events or characteristic adverse pregnancy outcomes require prophylactic anticoagulant therapy during pregnancy and, in the case of prior thromboembolism, during the puerperium. Antepartum anticoagulation does not appear warranted among patients with less thrombogenic thrombophilias who are without a history of venous thromboembolism, characteristic adverse pregnancy outcomes or other high risk factors for venous thrombosis.

While inherited thrombophilias are associated with adverse pregnancy outcomes most patients with thrombophilias have normal pregnancies and most patients with adverse outcomes do not have an identifiable thrombophilia. We are investigating the precise risk of adverse pregnancy outcome and the presence of inherited thrombophilias. In particular, we are establishing markers of enhanced uteroplacental thrombosis that will facilitate intervention trials.

Lastly, we are funded by the March of Dimes to study the pathogenic processes leading to preterm delivery. This study seeks to combine the most useful biophysical and biochemical markers of such processes with optimal clinical and epidemiologic predictors into a composite, easily applied risk tool. This integrated approach will identify at-risk asymptomatic patients with high sensitivity, specificity, and positive and negative predictive values, and also ascertain underlying pathogenic processes to facilitate targeted therapy.