Pregnancy is associated with unique hormonal milieu that facilitates the accommodation of the intricate needs of the embryo and the placenta. Successful angiogenesis and the creation of a low-resistance uteroplacental vascular system is a critical feature for normal pregnancy. Factors that affect angiogenesis are likely to affect pregnancy-specific vascular adaptations and maintenance. 2-ME2, an endogenous estrogen metabolite, is of a particular interest in this respect. 2-ME2 is one of the most potent endogenous antiangiogenic and proapoptotic steroids. 2-ME2 is a known inhibitor of endothelial cell proliferation, differentiation, and migration.10,11,19
Recently, it was shown that 2-ME2 is essential for regulation of uteroplacental vascular homeostasis in normal pregnancy and that its deficiency in a mouse model was associated with preeclampsia-like symptoms and fetal wastage.18
This raises the question as to how 2-ME2 would promote apoptosis and inhibit angiogenesis in some tissues and yet be essential for normal placentation and pregnancy on the other hand. In particular, we were interested in investigating whether the unique environment of pregnancy modulates the effects the 2-ME2 on uterine vascular endothelial cell proliferation and differentiation. Therefore, in the current study, we have investigated the effects of 2-ME2 on angiogenesis and proliferation of primary UAECs lines from nonpregnant and pregnant ewes.
We assessed angiogenesis by utilizing 3-dimensional basement membrane assays (capillary tube formation assay on Matrigel) that examines both vascular endothelial cell proliferation and differentiation. We showed that 2-ME2 inhibited capillary tube formation in F-UAECs only and not in L-UAECs or P-UAECs as evident by the preserved density capillary tubular structures and the branching points in L- and P-UAECs in the presence of 2-ME2. Furthermore, we showed that 2-ME2 equally induced apoptosis and microtubular architecture disruption in pregnant and nonpregnant UAECs. Treatment of UAECs with shorter durations of lower doses of 2-ME2 did not cause cell apoptosis while higher concentrations of 2-ME2 increased apoptosis equally in P-UAECs and NP-UAECs. In addition, higher concentrations of 2-ME2 treatment resulted in microtubule clumping, condensation, and polymerization around the nucleus when compared to 17βE2 treatment or control. The effect of 2-ME2 on microtubular structure was equally seen in P-UAECs and NP-UAECs. We concluded that the milieu of pregnancy may create a unique pregnancy-specific environment that modulate the action of 2-ME2 on capillary tube formation from UAECs.
We have previously shown that in vitro culture of primary sheep UAECs is a viable model for studying angiogenesis as they closely recapitulates the in vivo signaling events and maintain their memory in regard to their cell of origin.20
We have used this system to study the effects of 17βE2 on angiogenesis and showed that 17βE2 plays an important role in endothelial cell proliferation and differentiation. Here, we use this system to investigate the effects of 2-ME2 on UAECs in order to identify possible differential effects of 2-ME2 on capillary tube formation in pregnancy. We show that the inhibitory effect of 2-ME2 on tube formation was less pronounced in UAECs from pregnant and luteal phase sheep when compared to follicular phase.
The precise underlying etiology for the differential action and attenuation of the antiangiogenic effects of 2-ME2 in pregnancy is not apparent. The cellular effects of 2-ME2 are likely dependant on the cell type and the unique cellular milieu in different tissues. Potential factors that may explain the differential action of 2-ME2 in nonpregnant and pregnant endothelial cells include differences in the concentration and metabolism of 2-ME2 in different cell types. Varying concentrations of 2-ME2 were shown to have different effects on cell proliferation and the process of angiogenesis overall. Low doses of 2-ME2 were stimulatory to the proliferation of primary endometrial stroma and Ishikawa cells while higher doses were inhibitory.24
In this study, we show no effects of lower concentrations of 2-ME2 on UAECs, while higher concentrations resulted in cell apoptosis. Another potential factor that could contribute to the differential effects of 2-ME2 on angiogenesis in pregnancy include the distinctive hormonal milieu of pregnancy and the dissimilar expression of hormone receptors, receptor isoforms, and receptors coactivators and corepressors between nonpregnant and pregnant uterine artery endothelium.
Here, we show that the effects of 2ME2 were more pronounced in F-UAECs when compared to L-UAECs and P-UAECs. These effects may be due to progesterone which may play a role in modulating the effects of 2-ME2 on the vascular uterine endothelium in pregnancy. Both pregnant and luteal periods are high progesterone states whereas folicular period is high estrogen state. This difference is further augmented in the sheep model as, unlike the human ovary, the sheep corpus leuteum does not produce any estrogen. The role of progesterone receptors in the regulation of capillary tube formation by 2-ME2 in pregnancy warrants further studies especially in lieu of the current observation as the 2 physiologic states with high progesterone were protected from 2-ME2. Estrogen receptors (ER) on the other hand are unlikely to play a major role in modulating 2-ME2 action in pregnancy since 2-ME2 has weak affinity for ER. Recently, we have shown that ER-α antagonist has no effects on the proliferation of P-UAECs that were treated with or without 2-ME2 and that neither ER-α or ER-β contributed to the proliferative responses of pregnant UAECs to estrogen metabolites.8
The exact signaling cascade for the action of 2-ME2 in uterine artery endothelial cells remains unclear. We have previously shown in our lab that ERK1/2 and AKT phosphorylation signaling pathways are involved in mediating the acute nongenomic action of estradiol in UAEC; so here we wanted to assess if 2-ME2 work through a similar mechanism.21
We could not show conclusive evidence of 2-ME2-mediated increase in the acute phosphorylation of ERK1/2, Akt or eNOS in the current model. Our lab is currently reinvestigating the acute nonestrogen-receptor-mediated actions of 2-ME2 via phosphorylation of ERK1/2 and p38 MAPK under extensive time courses (Jobe and Magness, unpublished observation). Many other genes were shown to be regulated by 2-ME2 and could potentially modulate the function of 2-ME2 on UAEC. 2-ME2 inhibited ERK mitogen-activated protein kinase in human adrenal carcinoma cell line.12
2-ME2 also inhibited the α-subunit of hypoxia-inducible factor (HIF-1 α) in the murine placenta.18,13,25
Other genes that are regulated by 2-ME2 include SMAD that mediates signaling of transforming growth factor β (TGFβ), cyclin B1 protein and its associated kinase Cdk1/Cdc2, glucocorticoid receptor, nuclear factor-κB (NF-κB), and IRS-1 that promote mitogenesis through activation of the insulin receptor and the insulin-like growth factor I.17
We have previously shown that higher concentrations of 2-ME2 downregulated Bcl2 and VEGF protein expression in endometrial stroma cells, thus accounting for some of the proapoptotic and antiangiogenic effects of 2-ME2.24
It is possible that the differential regulation of some of these genes by 2-ME2 in uterine vasculature and placental endothelium may alter the effects of 2-ME2 in pregnant and nonpregnant state.
Technical limitations of this study include the fact that 2 different incubation perioids were used to assess the effects of 2ME2 on angiogenesis and apoptosis. This was due to the fact that angiogenesis and apoptosis became detected at different time intervals in our cell line. Despite this limitation, our data further reinforce the role of 2ME2 in reproduction. Other data supporting a physiological role of 2-ME2 in reproduction is rather strong. 2-ME2 is totally and exclusively produced by COMT enzyme. COMT-/-
mice have shed the light on many of the reproductive roles of 2-ME2.26
Using this model, Zacharia and colleagues have reported that lack of 2-ME2 in COMT -/-
mice caused failure to dilate aortic endothelial cells after treatment with 17βE2 when compared to wild-type littermates indicating that 2-ME2 is the main mediator of the vasodilitotory effect of 17βE2 in aortic endothelial cells.6,26,27
Kanasaki and colleagues have shown that 2-ME2 prevented placental clotting, fetal wastage, and preeclampsia in a mice model and that COMT-/-
mice that were deficient in 2-ME2 developed preeclampsia-like symptoms.18
Furthermore, COMT expression is downregulated in mid-secretory phase that corresponds to the implantation window in human endometrium and is overexpressed in the placenta during the first trimester of pregnancy.24,28
Herein, we show that pregnancy protected P-UAECs from the antiangiogenic effect of 2-ME2, a process that would promote neovascularization and thus elevations in uteroplacental blood flow when the fetal demand for oxygen and nutrients is greatest. Moreover, this has the potential of ensuring that 2-ME2 is not deleterious to angiogenesis of the delicate endothelial cells of the uterine vasculature during pregnancy. Despite the protective effects of pregnancy from the antiangiogenic action of 2-ME2, there were no differences in apoptosis or tubular polymerization between when higher doses of 2-ME2-were used. This may indicate that the processes of angiogenesis and apoptosis in UAEC are independently regulated. If proven, this may further facilitate vascular remodeling process in early pregnancy.