In this study, we found a significant upregulation of CD46 protein expression in the EOSPE and LOSPE group. No difference in CD46 protein expression was observed between the EOSPE and E-control groups or between the EOSPE and LOSPE groups. Additionally, no significant difference in CD55 protein expression was discovered between the preeclampsia group and the control group.
In recent years, many studies have revealed that circulating C4d, C3a, C5a, and sC5b-9 levels were higher in patients with preeclampsia than in healthy pregnant women[3–7, 21]. Meanwhile, complement-activated fragments, including C4d, C5a and MAC, were expressed at much higher levels in placentas of patients with preeclampsia than in normal pregnant women[8–11]. Considering these findings, we estimated that the upregulation of CD46 protein expression in severe preeclampsia was a maternal-fetal interface feedback mechanism to attempt to limit local complement activation.
Our recent study found that excessive activation of the complement system through the alternative and classical pathways was observed as early as the first trimester in patients with preeclampsia later in pregnancy. Then, the concentrations of complement components became similar to those in normal pregnancy thereafter, even in late pregnancy. Complement activation was not core factor in the pathogenesis of preeclampsia. So, upregulation of CD46 protein expression could not prevent the progression of preeclampsia.
In this study, we did not discover significant differences in CD46 and CD55 expression between the EOSPE group and the E-control group. In fact, it was difficult to enroll a standardized E-control group without any potential factor of complement activation. In this study, we enrolled patients with preterm delivery, premature rupture of membranes, placental implantation and fetal hemolysis in the E-control group. No placenta infection occurred in this group. Complement activation was previously reported to be involved preterm labor[23, 24], and CD55 expressed on white blood cells was elevated in patients with preterm labor[25, 26]. Overexpression of complement regulatory protein may have also occurred in placentas of patients with preterm delivery and premature rupture of membranes in the E-control group. Therefore, the expression of complement regulatory proteins in the EOSPE group requires further study.
To date, several studies have examined the expression of complement regulatory protein in placentas of preeclampsia patients. Lokki et al. found that membrane-bound regulatory proteins (CD46, CD55, and CD59) were widely expressed on the syncytiotrophoblast layer by immunofluorescence staining and histochemistry. There was no difference observed in the expression pattern of any regulatory protein between the preeclampsia and control groups or between the early-onset and late-onset preeclampsia groups. Another study by Buurma et al. discovered that the mRNA expression of CD55 and CD59, but not CD46, was much higher in early-onset preeclampsia patients than in normal pregnant women with term delivery. However, this study did not enroll late-onset preeclampsia and recruited the early-onset preeclampsia group without strict standards, including 8 cases of chronic hypertension and 2 cases of inherited thrombophilia. Perhaps due to the complex regulatory pathways, our results were inconsistent.
In addition to mRNA and protein expression, some researchers have focused on the mutations of CD46 that predispose patients to preeclampsia. Salmon et al. confirmed the relationship between hypomorphic variants of CD46 and non-autoimmune preeclampsia. However, Lokki et al. did not identify an association between genetic polymorphisms of CD46 and preeclampsia. Further study should focus on the role of functional CD46 isoforms in the pathogenesis of severe preeclampsia.
In cellular experiments, we found no significant difference in expression of CD46 and CD55 on the HTR-8/SVneo cell surface after VEGF treatment at different concentrations. Mason et al. found that human umbilical vein endothelial cells and dermal endothelial cells expressed more CD55 after VEGF treatment, but not CD46 and CD59. However, another study has found that VEGF upregulated the expression of factor H, CD46 and CD55 in glomerular. Extravillous trophoblasts invade and replace endothelial cells of spiral artery to remodel blood vessels. Perhaps limited concentration gradient may result in negative results. Furthermore, we speculated that complement regulatory proteins’ response to VEGF in heterogeneous endothelial cells may vary between different vascular beds.
Our study had the following limitations: (i) As previously mentioned, E-control group enrollment was not strict, and perhaps CD46 and CD55 were upregulated in some patients. Additional time will be needed to enroll a standardized group of patients. (ii) This study had a relatively small sample size, and further studies with larger sample sizes need to be implemented. (iii) In cellular experiments, HTR-8/SVneo cells had a tendency to express more CD46 with the increase of VEGF concentration, but without statistical difference. We should set higher concentration of VEGF and examine circulating regulatory proteins in further experiment.