Controlling Trophoblast Cell Fusion in the Human Placenta-Transcriptional Regulation of Suppressyn, an Endogenous Inhibitor of Syncytin-1

Abstract

Cell fusion in the placenta is tightly regulated. Suppressyn is a human placental endogenous retroviral protein that inhibits the profusogenic activities of another well-described endogenous retroviral protein, syncytin-1. In this study, we aimed to elucidate the mechanisms underlying suppressyn's placenta-specific expression. We identified the promoter region and a novel enhancer region for the gene encoding suppressyn, ERVH48-1, and examined their regulation via DNA methylation and their responses to changes in the oxygen concentration. Like other endogenous retroviral genes, the ERVH48-1 promoter sequence is found within a characteristic retroviral 5' LTR sequence. The novel enhancer sequence we describe here is downstream of this LTR sequence (designated EIEs: ERV internal enhancer sequence) and governs placental expression. The placenta-specific expression of ERVH48-1 is tightly controlled by DNA methylation and further regulated by oxygen concentration-dependent, hypoxia-induced transcription factors (HIF1α and HIF2α). Our findings highlight the involvement of (1) tissue specificity through DNA methylation, (2) expression specificity through placenta-specific enhancer regions, and (3) the regulation of suppressyn expression in differing oxygen conditions by HIF1α and HIF2α. We suggest that these regulatory mechanisms are central to normal and abnormal placental development, including the development of disorders of pregnancy involving altered oxygenation, such as preeclampsia, pregnancy-induced hypertension, and fetal growth restriction.

Keywords: DNA methylation; HERV; cell fusion; hypoxia inducible factor (HIF); placenta; promoter; suppressyn; syncytin.

Figure 1

Schematic diagram of the suppressyn promoter region. The top panel shows the genome structure encoding suppressyn (ERVH48-1), and the middle panel shows an enlarged schematic of the 5′ LTR (long terminal repeat) and 5′ flanking regions. Lollipops indicate predicted CpG methylation sites. The lower panel shows the 5′ LTR sequence, with white outlining the U3, grey outlining the R, and black outlining the U5 regions. Arrows indicate the transcription start point of the NM001308491.2 transcript. Red lettering indicates TATA sequences, blue indicates predicted binding sequences for the HIF/ARNT complex, and green indicates predicted binding sequences for GATA transcription family members.

Figure 2

Promoter activity of the suppressyn 5′ LTR and its flanking region. On the left is a schematic diagram of the suppressyn 5′ LTR and 5′ flanking sequence, as cloned into the pGL3 vector. The enhancer region predicted by the cCRE is shown schematically (EH38E2145129). The right side of the diagram depicts the results of luciferase activities measured after the transfection of these constructs into various cell lines (BeWo, JEG3, HTR8, Ishikawa, and HeLa cells). pGL3-basic was used as reference for comparison to each of the other constructs. Statistical analyses were made using Mann–Whitney U tests compared to pGL3-LTR in each cell line (* p < 0.05, ** p < 0.01). Experiments were performed in duplicate, and the means and ± SDs were calculated from three independent experiments.

Figure 3

Bisulfite sequencing analysis of ERVH48-1 promoter regions. Bisulfite sequencing of the ERVH48-1 5′ LTR and 5′ flanking region. White circles represent non-methylated CpG sites and black circles represent methylated CpG sites. The top columns show the location of CpG dinucleotides from 1 to 28, with numbers 19–28 depicting the methylation status of CpG sites within the 5′ LTR region. The bottom columns list the methylation rates (%) for each 5′ LTR region.

Figure 4

Luciferase activities of the ERVH48-1 promoter and novel enhancer sequences (EIEs). (a) The schematic diagram on the left depicts the suppressyn 5′ LTR and its flanking sequences, including the 5′ flanking region of ERVH48-1 and putative enhancer sequences (ERV Internal Enhancers; EIEs). Below, the separate constructs cloned into the pGL3 vector and the regions they cover, including the 5Enh, LTR, and EIE sequences, are shown. On the right, relative luciferase activity was plotted for these constructs for each cell line used (BeWo, JEG3, HTR8, Ishikawa, and HeLa cells). (b) The left side of this diagram depicts a schematic of the ERVH48-1 5′ LTR and EIE sequences. Details of the whole and the restriction enzyme-truncated (SmaI, EcoRV, and BglII) pGL3 constructs are shown below this. Three enhancer regions in the EIEs were predicted via cCRE. Predicted binding sites for GATA transcription factors in this region are presented (…GATA…). Luciferase activity associated with the transfection of these constructs into BeWo cells is shown on the right. Luciferase activity associated with the presence of the pGL3-basic construct was used as a comparator for statistical analysis. Mann–Whitney U tests were performed, and statistical significance was set at ** p < 0.01. Experimental replicates were performed in duplicate, and the means ± SDs were calculated from three independent experiments.

Figure 5

ERVH48-1 promoter activity in low and high oxygen conditions. (a) Comparison of ERVH48-1 promoter activity under low and high oxygen conditions (2% or 20%). Luciferase activity of ERVH48-1 promoter sequences in BeWo, JEG3, HTR8, Ishikawa, and HeLa cells when cultured at 20% O₂ (left column) or 2% O₂ (right column). (Bars from left to right: pGL3-basic, -LTR, -5Enh-LTR, -LTR-EIE, and -5Enh-LTR-EIE.) Luciferase activities for each construct were normalized to pGL3-basic and compared to values for each reporter construct cultured under 20% O₂ results to assess statistical differences. (Mann–Whitney U test: * p < 0.05; ** p < 0.01.) Experiments were performed in duplicate, and the means ± SDs were calculated from three independent experiments. (b) The effects of transient expression of hypoxia-inducible proteins (HIF1α, HIF2α, and ARNT) on the transcriptional activity of each ERVH48-1 promoter construct in BeWo cells and (c) in HTR8 cells. pGL3-basic was included as a reference. Statistical analysis was performed, comparing the luciferase activities in cells exposed to the vector-only transfected control with those in which each gene or genes were transiently transfected (Mann–Whitney U test: ** p < 0.01). Experiments were performed in duplicate, and the means ± SDs were calculated from three independent experiments.