IFITM proteins inhibit placental syncytiotrophoblast formation and promote fetal demise

See allHide authors and affiliations

Science  12 Jul 2019:
Vol. 365, Issue 6449, pp. 176-180
DOI: 10.1126/science.aaw7733

Placenta formation and fetal demise

A critical step of placental development is the fusion of trophoblast cells into a multi-nucleated syncytiotrophoblast layer. Trophoblast fusion is mediated by syncytins, encoded by endogenous retrovirus–derived envelope glycoproteins. Buchrieser et al. report that interferon-induced transmembrane (IFITM) proteins inhibit syncytin-mediated syncytiotrophoblast formation, restricting placental development and triggering fetal demise (see the Perspective by Kellam and Weiss). The results provide a molecular explanation for the placental dysfunctions observed in interferon-mediated disorders such as intrauterine growth retardation, TORCH (toxoplasmosis, other, rubella, cytomegalovirus, and herpes) infections, and some forms of preeclampsia.

Science, this issue p. 176; see also p. 118


Elevated levels of type I interferon (IFN) during pregnancy are associated with intrauterine growth retardation, preterm birth, and fetal demise through mechanisms that are not well understood. A critical step of placental development is the fusion of trophoblast cells into a multinucleated syncytiotrophoblast (ST) layer. Fusion is mediated by syncytins, proteins deriving from ancestral endogenous retroviral envelopes. Using cultures of human trophoblasts or mouse cells, we show that IFN-induced transmembrane proteins (IFITMs), a family of restriction factors blocking the entry step of many viruses, impair ST formation and inhibit syncytin-mediated fusion. Moreover, the IFN inducer polyinosinic:polycytidylic acid promotes fetal resorption and placental abnormalities in wild-type but not in Ifitm-deleted mice. Thus, excessive levels of IFITMs may mediate the pregnancy complications observed during congenital infections and other IFN-induced pathologies.

The placenta is the primary maternal–fetal barrier, achieving metabolic exchanges, hormone production, and protection from pathogens and the maternal immune system. The placental tissue stems from embryonic cytotrophoblasts. The structural and functional unit of human placenta is the chorionic villous, which includes proliferative mononuclear villous cytotrophoblasts (VCTs) at the basement and the syncytiotrophoblast (ST) layer at the surface. The multinuclear ST arises from the differentiation and fusion of VCTs. In humans, an abnormal ST is observed in pregnancy pathologies including intrauterine growth retardation (IUGR), preeclampsia, lupus, and Down syndrome (13). In contrast to the human placenta, the mouse placenta has a labyrinthic structure and exhibits two ST layers that separate fetal capillaries from maternal blood (4). In gestating mice, interferon-beta (IFN-β) triggered by Zika or other viruses or by administration of polyinosinic:polycytidylic acid (Poly-IC), promotes IUGR and fetal demise (2, 57). IFN-β signaling leads to abnormal labyrinth and ST structures, with the presence of many unfused cells and reduced fetal blood vessels (7).

Cytotrophoblast fusion is mediated by envelope glycoprotein (env)–derived genes of endogenous retroviruses (ERVs) that have been captured by mammals (3). These genes were termed “syncytins” on the basis of their fusogenic capacity. The capture of syncytin genes, sometimes as pairs (syncytin-1 and -2 in humans and syncytin-A and -B in mice) (3, 8, 9), occurred independently from different ERVs in diverse mammalian lineages 10 to 85 million years ago. Knocking out the syncytin-A gene or both syncytin-A and syncytin-B genes leads to fetal growth restriction and mid-gestational lethality (8, 9).

The immune-related IFN-induced transmembrane proteins IFITM1, IFITM2, and IFITM3 are restriction factors that protect uninfected cells from viral infection. They block the entry into host cells of many enveloped viruses by inhibiting virus–cell fusion at the hemifusion stage (1012) and act by altering the biophysical properties or cholesterol content of the cellular membranes in which they are found (1113). IFITMs modify the rigidity and/or curvature of the membranes without evidence of a direct interaction with the fusogenic viral envelopes (11, 12). Because of different sorting motifs, IFITM1 mostly resides at the plasma membrane, whereas IFITM2 and IFITM3 accumulate in the endolysosomal compartment after transiting at the cell surface. IFITM proteins are expressed at various basal levels in different cell types and are up-regulated by IFNs and other cytokines (11). In addition to their antiviral activity, the physiological cellular functions of IFITM proteins remain poorly characterized. Transgenic mice in which the cluster of Ifitm genes is knocked out (hereafter referred to as IfitmDel mice) are more sensitive to various viral infections (12) but exhibit no overt abnormalities apart from a slight overweight (14). Little is known about the impact of IFITMs on fusion events mediated by cellular proteins or by ERV-derived env proteins. Here, we investigated whether IFITMs impair syncytin-mediated fusion and thus affect fetal development.

We first investigated whether IFITMs inhibit cell fusion mediated by exogenously expressed human syncytins. We generated 293T cells carrying a green fluorescent protein (GFP)–Split complementation system, in which two cells separately produce half of the reporter protein, generating a signal only upon fusion (Fig. 1A). The extent of fusion was quantified by measuring the GFP+ area (Fig. 1A). Transfection of syncytin-1 induced the appearance of multinucleated GFP+ cells. Fusion was significantly decreased when syncytin-1 was cotransfected with FLAG-tagged IFITM1, 2 or 3, but not with a control plasmid (Fig. 1A). The extent of inhibition of fusion varied between individual IFITMs. The IFITM3Δ1-21 and IFITM3∆ub mutants, which lack endocytic and ubiquitination motifs, respectively, and accumulate in the cell (15, 16), were more active at inhibiting fusion than was wild-type (WT) IFITM3. By contrast, an IFITM3∆palm mutant, whose levels are reduced (15, 16), was inactive. The coexpression of three IFITMs strongly inhibited fusion (Fig. 1A). IFITMs similarly inhibited fusion mediated by syncytin-2 (fig. S1A). IFITMs did not decrease the levels of syncytin-1 in transfected cells (fig. S1B). To distinguish the effect of IFITMs in syncytin-1+ (donor) cells from that in target cells (expressing SLC1A5, the syncytin-1 receptor), we used an HIV Tat and long terminal repeat luciferase transactivation coculture system, in which cell fusion generates luciferase activity (13). IFITMs inhibited fusion when present in either donor or target cells (fig. S1C).

Fig. 1 IFITMs inhibit syncytin-mediated cell fusion.

(A) Left: 293T-GFP1-10 and -GFP11 were cocultured at a 1:1 ratio and cotransfected with syncytin-1 and IFITM or control plasmids. Cell fusion was quantified by measuring the GFP+ area at 18 hours. Middle: Representative images of GFP-Split 293T cells cotransfected with syncytin-1 and control or IFITM2 expression plasmids. GFP areas are outlined in white. Scale bar, 200 μm. Right: Quantification of GFP areas. Results are shown as mean ± SD from four independent experiments. (B) Left: BeWo β-Gal-α and β-Gal-ω were transduced with IFITM or control vectors. Cells were cocultured at a 1:1 ratio and fusion was induced by forskolin for 48 hours. Right: Fusion index (β-Gal activity) measured with a colorimetric (chlorophenol red-β-D-galactopyranoside, CPRG) assay. Results are shown as mean ± SD from four independent experiments. (C) Left: WT or IfitmDel MEFs were cotransfected with GFP and syncytin-A plasmids and syncytia were quantified after 24 hours. Right: Quantification of syncytia (cells with more than three nuclei) per well. Results are shown as mean ± SD from three to six independent experiments. Statistical analysis: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, one-way analysis of variance (ANOVA); ns, not significant.

We next examined the effect of IFITMs on endogenous syncytins. In trophoblast-like BeWo human choriocarcinoma cells, addition of the adenylate cyclase activator forskolin triggers syncytin-1 production and syncytium formation (17). To quantify fusion, we generated BeWo cells carrying two complementation systems based on either β-galactosidase-α/ω (β-Gal-α/ω) (Fig. 1B) or GFP-Split (fig. S2C). These cells were then transduced with lentiviral vectors to express a control protein or FLAG-tagged IFITM1-3 (fig. S2, A and D). As expected, forskolin triggered fusion and production of β-Gal, which was detected in fixed cells and quantified in cell lysates (Fig. 1 B and fig. S2B). The fusion efficacy was significantly reduced by IFITMs. The presence of IFITMs did not modify BeWo growth nor alter their ability to produce syncytia-independent β-Gal (fig. S3 A to D). Similar results were observed with the GFP-based system. Videomicroscopy analysis showed delayed and reduced fusion with each IFITM (fig. S2, C to E, and movie S1).

We next tested the capacity of endogenous IFITMs to block fusion in mouse embryonic fibroblasts (MEFs) derived from WT or IfitmDel mice (14). To this aim, MEFs were cotransfected with syncytin-A and GFP plasmids (Fig. 1C and fig. S4A). Syncytin-A induced numerous and large GFP+ syncytia (with up to 20 nuclei) in IfitmDel MEFs, and fewer and smaller syncytia in WT MEFs. Transduction of IfitmDel MEFs with a murine Ifitm3 vector (fig. S4B) restored resistance to fusion (Fig. 1C and fig. S4A). WT MEFs were poorly sensitive to fusion, likely due to high basal IFITM3 levels (fig. S4B).

We then evaluated the effect of type I IFN and IFITMs in primary human trophoblasts. We first asked whether IFITMs were up-regulated by type I IFN in human placental explants. It has been reported that in human midterm placental chorionic villi explants, addition of IFN-β, but not IFN-λ, leads to defects including cellular damage, decreased production of human chorionic gonadotropin (hCG), and the appearance of syncytial knots (7). In such explants, IFN-β up-regulated hundreds of IFN-stimulated genes (ISGs), including IFITM1 (7). We took advantage of this large set of published RNA data to investigate the expression of IFITMs, syncytins, and their receptors. IFITMs were up-regulated by IFN-β compared with mock- and IFN-λ–treated explants, whereas syncytins and their receptors were not modulated (fig. S5).

These observations were made in whole placental explants, in which nontrophoblastic cells are also present. We thus studied the impact of IFN-β on primary VCTs isolated from eight term human placentas. VCTs can be cultured for up to 3 days and spontaneously differentiate in multinucleated ST (18). Treating VCTs with IFN-β (100 or 1000 IU/mL for 48 hours) strongly up-regulated IFITM1 and IFITM2-3, as shown by immunofluorescence and Western blot (fig. S6, A and C). IFN-β slightly up-regulated RNA levels of syncytin-2 but did not affect those of syncytin-1 and syncytin receptors (fig. S6D). We next quantified VCT fusion, using a method based on the detection of the transcription factor GATA3 by immunofluorescence and its down-regulation upon ST differentiation (18). Cells were also stained for E-cadherin to visualize plasma membranes. IFN-β inhibited VCT fusion to the same extent as GW9662, a peroxisome proliferator–activated receptor gamma (PPARγ) antagonist known to block this process (18) (Fig. 2, A and B). Because it acts independently of IFN signaling, GW9662 did not induce IFITMs (fig. S6, A and C). The effect of IFN-β on ST differentiation was supported by a decreased release of hCG, which is secreted by ST but not by VCTs (Fig. 2C). The reduction of fusion was not due to a cytotoxic effect of IFN-β, because levels of an apoptosis marker (cCK18) were not augmented (fig. S6, B and C).

Fig. 2 IFN-β inhibits fusion of primary human VCTs.

(A) Left: Quantification of fusion. Mononucleated VCTs express the nuclear GATA3 protein, which is down-regulated after fusion. Right: After 48 hours, nuclei were visualized by 4′,6-diamidino-2-phenylindole (DAPI) and plasma membranes with E-cadherin. Representative images are shown. Syncytia, defined as large cells containing multiple GATA3 nuclei, are outlined in white. (B) The fusion index was quantified as the percentage of GATA3-negative nuclei. (C) hCG levels in culture supernatants at 72 hours. Results are shown as mean ± SD from eight independent experiments. Statistical analysis: ****P < 0.0001, one-way ANOVA.

To further assess the effect of IFITMs in this system, VCTs were transfected with a FLAG-tagged IFITM3 or a control GFP plasmid. The poor efficiency of transfection (10% of the cells expressing IFITM3 or GFP) precluded precise quantification of syncytia. However, a visual examination indicated that IFITM3+ cells remained mononucleated, even in the vicinity of large syncytia, whereas GFP-transfected cells were able to fuse (fig. S7). These results demonstrate that IFN-β induces IFITMs in VCTs and that both IFN-β and IFITMs inhibit ST formation.

We then evaluated the role of IFITMs in pregnant mice. Administration of Poly-IC has been extensively used as a model of type I IFN induction, triggering fetal growth retardation and resorption (2, 6, 7). We thus tested the effect of Poly-IC in gestating WT, Ifnar−/− or IfitmDel dams that had been mated to males of the same genotypes (Fig. 3A). Poly-IC was injected at embryonic day 7.5 (E7.5), a time point that shortly precedes ST formation (~E8.5) (4). A dose of 200 μg, which results in resorption of almost all fetuses within 48 hours (7), was injected. With Poly-IC, WT fetuses, but neither Ifnar−/− nor IfitmDel fetuses, were resorbed at E9.5 (Fig. 3B). The fetus size was similar in Ifnar−/− and IfitmDel mice (Fig. 3C and fig. S8) and slightly reduced compared with untreated animals (Fig. 3C and fig. S8). We confirmed that IfitmDel and WT mice similarly responded to Poly-IC, as shown by the induction of various ISGs (Irf7, Oas1a, Stat1) measured in the liver 14 hours after injection, indicative of a systemic inflammation (fig. S9B). A local response was also observed in placental extracts of WT mice, which up-regulated Ifitm1, Ifitm3, and Irf7 RNAs. As expected, IfitmDel placentas did not express Ifitm genes but up-regulated Irf7 (fig. S9 C). Levels of syncytin-A and -B RNAs were not significantly affected by Poly-IC (fig. S9C).

Fig. 3 IFITMs are key mediators of IFN-induced fetal demise in mice.

(A) Gestating dams were injected intraperitoneally with Poly-IC or phosphate-buffered saline (PBS) at E7.5. The number, size, and resorption of the embryos were assessed at E9.5. Numbers of litters were as follows: WT PBS, n = 3; WT Poly-IC, n = 6; Ifnar−/− PBS, n = 3; Ifnar−/− Poly-IC, n = 3; IfitmDel PBS, n = 3; and IfitmDel Poly-IC, n = 7. (B) Percentage of resorption for the indicated conditions. Numbers in brackets indicate the number of resorptions/total number of fetoplacental units. Statistical analysis: ****P < 0.0001, Mann-Whitney test; ns, not significant. (C) Representative images of E9.5 embryos. Scale bars, 500 μm. (D) WT and IfitmDel placentas (E9.5) stained for E-cadherin (red) and Hoechst (blue). White arrows indicate ST. Bottom panels are magnified areas indicated by white boxes in the top panels. Representative images from three independent experiments are shown.

We next performed histological analysis of the placentas to examine the consequences of Poly-IC treatment. Placentas from WT and IfitmDel fetuses were stained for E-cadherin to label trophoblasts (Fig. 3D). As expected, ST formation was detectable in untreated placenta from both WT and IfitmDel fetoplacental units (arrows). By contrast, with Poly-IC, less ST was detected in WT placentas, whereas it was still present in IfitmDel placentas. We next triple stained placenta 14 hours after Poly-IC injection for IFITM3, E-cadherin, and CD31 (a marker of endothelial cells) (fig. S9D). As expected, IFITM3 was up-regulated in placentas of Poly-IC-treated WT mice, but not in those of IfitmDel mice. IFITM3 overexpression upon Poly-IC injection was observed in CD31+ E-cadherin cells (endothelium) and CD31 E-cadherin+ cells (trophoblasts), demonstrating that IFITM3 is induced in trophoblasts in response to Poly-IC, consistent with in vitro results.

Altogether, these results strongly suggest that IFITM up-regulation in trophoblasts inhibits ST formation in vivo, which likely contributes to type I IFN-associated placental dysfunction and fetal demise in mice.

In this study, we have uncovered a previously unknown function for the IFITM family of antiviral restriction factors. We report that IFITMs impair the fusogenic activity of syncytins required for ST formation and maintenance. Our results provide a possible molecular explanation for placental dysfunctions associated with IFN-mediated disorders such as IUGR and “TORCH” infections (toxoplasmosis, other, rubella, cytomegalovirus, and herpes) (19). In addition to infection, genetic and autoimmune interferonopathies such as Aicardi–Goutières syndrome and systemic lupus erythematosus (SLE) are associated with pregnancy complications (2). High serum IFN level is a marker of poor pregnancy outcome in SLE (20). Down syndrome in trisomy 21 (T21) patients is also associated with serious birth defects (1). In vitro, VCTs from T21 patients poorly fuse into ST (1). This may be due to the location of the IFN receptor on chromosome 21, rendering T21 cells hyperresponsive to IFN (21) and thus potentially expressing high amounts of IFITMs.

As other restriction factors, Ifitm genes are polymorphic in humans and primates (11, 12). It will be worth determining whether placental pathologies of unknown etiology, such as certain preeclampsia or early spontaneous abortions, implicate IFITM proteins and variants. It is also tempting to suggest that blockade of IFITMs may represent a possible intervention strategy to prevent pregnancy complications linked to interferonopathies.

Supplementary Materials

Materials and Methods

Figs. S1 to S9

References (2237)

Movie S1

References and Notes

Acknowledgments: We thank members of the Virus and Immunity Unit for discussions and help, M. Maaran Rajah for critical reading of the manuscript. We thank patients who participated in the study. We also thank members of Embryology and Core Breeding teams (DTPS-C2RA-Central Animal Facility platform) for technical support with strain revivification, M. Cohen-Tannoudji for hosting mouse experiments. We thank the Cellular and Molecular Imaging facility of the Faculty of Pharmacy of Paris (UMS3612 CNRS/US25 INSERM), and the UtechS Photonic BioImaging (Imagopole). Funding: Work in the laboratory of O.S. is funded by Institut Pasteur, ANRS, Sidaction, the Vaccine Research Institute (ANR-10-LABX-77), Labex IBEID (ANR-10-LABX-62-IBEID), “TIMTAMDEN” ANR-14-CE14-0029, “CHIKV-Viro-Immuno” ANR-14-CE14-0015-01, CNRS and the Gilead HIV cure program. Work in the laboratory of M.L. is funded by Institut Pasteur, INSERM, ERC, Labex IBEID (ANR-10-LABX-62-IBEID), Institut Convergence, and Université Paris Descartes. C.M. was supported by a fellowship from grant no. ANR-10-LABX-62-IBEID. M.V.A. was supported by NSF GRFP grant no. 1644869. Author contributions: J.B., D.A.D., T.H., X.M., T.F., M.L., and O.S. designed the experimental strategy. J.B., Q.N., D.A.D., F.P., M.V.A., and K.R. designed and performed experiments with cell lines. S.A.D. and T.F. designed and performed experiments with primary trophoblasts. J.B., T.C., C.M., Q.N., O.D., and X.M. designed and performed mouse experiments. E.P. and K.-H.H. performed the bioinformatic analysis. A.D. and T.H. provided vital materials and expert advice. J.B. and O.S. wrote the manuscript and all authors agreed on the final version. Competing interests: The authors declare no competing interests. Data and materials availability: All data are available in the manuscript or the supplementary materials.

Stay Connected to Science

Navigate This Article