Human ESCs (H1, WA01) were cultured in Matrigel (BD Bioscience)-coated, six-well tissue culture plates (Thermo Scientific) under an atmosphere of 5% (vol/vol) CO2/air at 37°C in mTeSR1 medium (STEMCELL Technologies). The culture medium was changed daily and the cells were passaged every 5-6 days. The method for TB differentiation has been described previously (19). Briefly, on the day after passaging onto Matrigel-coated dishes at 1.2 × 10⁴ cells/cm², the culture medium was changed to DMEM/F-12 medium (Thermo Scientific) with knock-out serum replacement (KOSR, Invitrogen) that had been conditioned by mouse embryonic fibroblasts (MEF) and supplemented with FGF2 (4 ng/mL). After 24 h, the conditioned medium was replaced with daily changes of nonconditioned DME/F12/KOSR medium lacking FGF2, but containing BMP4 (10 ng/mL), A83-01 (1 μM), and PD173074 (0.1 μM) (BAP treatment) for up to 8 days. Control cultures were maintained in conditioned medium containing 4 ng/mL FGF2.

Human subjects were recruited for participation under an IRB approved protocol (University of Missouri, Columbia, Office of Human Research Protection Program, Medical IRB Committee-1 #2017804). Each subject provided written informed consent prior to enrollment. Peripheral blood was collected from 13 healthy women in the first trimester of pregnancy (6-12 wks estimated gestational age). The PBMC fraction was isolated from peripheral blood by using Ficoll-Paque Premium (GE Healthcare, Pittsburgh, PA) and centrifugation at 400×g for 30 min at 20°C; cells were cryopreserved until use.

PBMCs were thawed in pre-warmed (37°C) FBS. PBMC viability was over 80% as judged by Trypan blue (Sigma-Aldrich, St. Louis, MO) exclusion. After washing, PBMCs were resuspended at 5 X 10⁵ cells/mL in nonconditioned DME/F12/KOSR medium containing IL-2 (100U/ml) and co-cultured with BAP treated TB cells for 72 h (BAP d5-d8) at 37°C.

BAP-primed hESC cells were dissociated with ACCUTASE™ (STEMCELL™ TECHNOLOGIES) at 37°C for 10 min. Cells were washed in culture medium and passed through a 40-µm cell strainer to remove the larger STB population (Fisher Scientific). Cells were placed in 3% (wt/vol) BSA in PBS for 30 min to minimize non-specific interactions with antibody and stained with APC-conjugated anti-human HLA-G monoclonal mouse antibody (BioLegend, 335910). After being washed twice with PBS, cells were fixed in 4% (vol/vol) paraformaldehyde (PFA) in PBS. APC-conjugated mouse IgG2a Isotype control (BioLegend, 40222) antibody was employed to account for nonspecific staining. Cells were analyzed in a BD Accuri™ flow cytometer (BD Biosciences, San Jose, CA) with subsequent FlowJo software (Tree Star, Ashland, OR). Suspended cells from the co-culture were harvested for flow cytometry 72 h after co-culture establishment. Cells were washed and stained with the following antibodies: PE/Dazzle™ 594 conjugated anti-CD3 Ab (BioLegend 300336), APC-conjugated anti-CD56 Ab (BioLegend 362504) and phycoerythrin (PE)-conjugated anti-CD16 Ab (BioLegend 302008) for 30 min at 4°C. After washing twice with PBS, cells were fixed in the Foxp3/transcription factor staining buffer (eBioscience) according to manufacturer’s instructions. Matched fluorescence-labeled isotype antibodies were used as a negative control to account for nonspecific staining. Immunostained cells were analyzed by flow cytometry as described above.

For small interfering RNA (siRNA)-mediated knockdown of VTCN1, differentiated hESCs were transfected with either a VTCN1 human siRNA oligo duplex (Origene, SR312516, 10nM) or a control, Trilencer-27 Universal scrambled negative control siRNA duplex (Origene, SR30004, 10 nM) on BAP treatment d 3 by using Lipofectamine 2000 (Invitrogen). After 6 h, the medium was replaced with BAP medium. Total RNA was collected 24 h, 48 h and 72 h after transfection. Protein was collected 72h after transfection. At BAP treatment d 6, i.e., 72 h after the BAP 3 days, cells colonies were stained with crystal violet solution [0.2% wt/vol crystal violet in an aqueous solution containing 20% (vol/vol) ethanol and 10% (vol/vol) formaldehyde] for 15 min. After washing with water, images of individual cell colonies were captured with a Leica M205 stereoscope.

VTCN1 has five isoforms and our study showed CRISPR gRNA/Cas9 at a single site mutation was not sufficient to knockout the isoform products. To overcome it, two gRNAs were used to delete the whole coding sequence. The VTCN1 ^-/- cell line was generated from the hESC by CRISPR nuclease-induced, targeted, double-strand breakage at the Genome Engineering and iPSC Center (GEiC), Washington University in St. Louis. Briefly, gRNAs were designed to target an exon common to all transcripts, where an indel can lead to nonsense mediated decay of the messenger RNAs. The parental H1 cells were nucleofected with the CRISPR gRNA/Cas9 ribonucleoprotein complex, and single cell H1 clones were screened for indels or biallelic deletions in the target gene (VTCN1). The whole coding sequence of VTCN1 was targeted with two guide RNAs: sgRNA1: 5’-AGCCAGTACCCAGATACGCT-3’ and sgRNA2: 5’-GAGATTAATCACAAATAGTG-3’ ( Figure 1 ). Genomic PCR determined biallelic deletion of VTCN1 gene.

RNA was extracted in STAT60 containing 1-bromo-3-chloro-propane (Sigma-Aldrich), and DNA removed by using the TURBO-DNA-free kit (Ambion Inc.). A total of 650ng of each RNA sample was included for complementary DNA synthesis by using the iScript™ cDNA Synthesis kit (BioRad). Primers were designed based on published sources and synthesized by Integrated DNA Technologies; primer sequences are summarized in Table 1 . Quantitative RT-PCR (qRT-PCR) was performed with a SYBR Green qPCR Master Mix (MedChemExpress) on the BioRad CFX Connect Real-Time PCR Detection System. Thermal cycling conditions were as follows: 95°C for 5 min, followed by 45 cycles of: denaturation at 95 °C for 10 sec, annealing at 56 °C for 10 sec and extension at 72 °C for 10 sec. PCR normalization of data to that of the endogenous control (GAPDH) and fold-change values calculated by the 2^−ΔΔCT procedure have been described elsewhere (20). The final PCR results were expressed as relative expression compared to individual control sample in each assay. Data followed over time were subjected to two-way ANOVA, followed by the Bonferroni test for pairwise comparisons. Values of p<0.05 were considered to support the conclusion that differences were statistically significant.

RNA was obtained from VTCN1 knock down or negative control siRNA transfected cells 24 h, 48 h and 72 h after transfection. The experiment was performed three times to provide a total of 18 RNA samples. RNA quantitation and quality control was performed on a Fragment Analyzer (Advanced Analytical) and cDNA libraries were constructed by standard methods (Illumina TruSeq mRNA stranded kit) with index adapters (Illumina TruSeq indexes). DNA was then sequenced as single-end, 50 base-length reads on a NovaSeq6000 instrument (Illumina, Inc.) with an average read count of 24-34 million reads per sample. Initial base calling and quality filtering of the mRNA-seq reads generated with the Illumina analysis pipeline (fastQ format) were performed with the FastxToolkit (http://hannonlab.cshl.edu/fastx_toolkit/) and adapters trimmed by using CutAdapt (https://cutadapt.readthedocs.org/en/stable/). Reads that mapped either to the human mitochondrial genome or to the internal Phix standard were removed through use of Bowtie software (http://bowtie-bio.sourceforge.net/index.shtml). Reads were mapped (TopHat v2.0.13; http://ccb.jhu.edu/software/tophat/index.shtml) to the reference genome (from Ensembl, v78; http://www.ensembl.org/index.html; Homo_sapiens.GRCh38.dna.primary assembly.fa with Homo_sapiens.GRCh38.78.gtf as annotation). Tests for differential expression were made by using cuffdiff v2.2.1. Uniquely mapped reads were counted with htseq-count v.011.1. Gene expression levels were calculated in FPKM (fragments per kilobase of exon per million mapped fragments), considering gene length as a sum of all exonic nonoverlapping sequences of all isoforms of a given gene. Regularized log transformed (rlog) counts output from DESeq2 to plot heatmaps and gene expression. Tests for differential expression (q-value < 0.01, Log2(foldchange) > 1.5; the Benjamini–Hochberg P value adjustment method was used with DESeq2 within Rstudio (RStudio, Inc.). To broadly analyze pathways affected by VTCN1 suppression, regulated genes with identifiable alterations upon VTCN1 knock-down were further analyzed by KEGG pathway enrichment with the web-based DAVID tool (Database for Annotation, Visualization, and Integrated Discovery v6.8). Further clustering and pathway analyses were performed by Ingenuity Pathway Analysis (Qiagen, Hilden, Germany).

Cells were grown on coverslips coated with Matrigel and placed in six-well tissue culture plates as described previously (19). After fixing the cells in 4% paraformaldehyde (PFA) in PBS for 10 min and permeabilizing them in 1.0% Triton X-100/PBS for 30 min, coverslips were placed in 5% (vol/vol) goat serum/5% (wt/vol) BSA in PBS for 1 h. Cells were then incubated with appropriately diluted primary antibodies ( Table 2 ) overnight at 4°C. Secondary antibody staining was performed with either Alexa Fluor 568-, 647-, or 488-labeled detection antibodies at a 1:300 dilution for 2 hours at room temperature ( Table 2 ). Images were captured under a Zeiss Axiovert 200M with a Leica DFC290 color camera.

Proteins were extracted from cells in radioimmunoprecipitation assay (RIPA) buffer [10 mM Tris·HCl (pH 7.2), 1 mM EDTA, 1% Triton X-100, 0.1% SDS, 0.1% sodium deoxycholate, and 100 mM NaCl], then fractionated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane by means of a transfer apparatus according to the manufacturer’s protocols (Bio-Rad). After incubation with 5% nonfat milk in Tris-buffered saline/Tween 20 (TBST; 10 mM Tris, pH 8.0, 150 mM NaCl, 0.5% Tween 20) for 60 min, the membrane was washed once with TBST, and target proteins detected by incubating with the primary antibodies described in Table 3 at 4°C for 12 h. Membranes were washed three times for 10 min and incubated with a 1:3000 dilution of horseradish peroxidase-conjugated anti-mouse or anti-rabbit antibodies for 2 h. Blots were washed with TBST three times and developed with the UVP imaging system (AnalytikJena) according to the manufacturer’s protocols. Pairwise comparisons were made with the Student’s t test. Values of p<0.05 were considered to support the conclusion that differences were statistically significant.

An aliquot of 5 x 10⁴ H1 cells was transferred to invasion chamber wells as previously described (21). After 24 h from initial plating, the medium was changed (2 ml in each chamber) to BAP treatment for differentiation. The medium was then changed daily for 3 days. On the third day (4 days after initial plating), the cells were transfected with VTCN1 knock-down or scrambled-control siRNA for 6 h. Then the medium was replaced with BAP treatment medium and the cells cultured for another 3 days. On the sixth day the cells on the lower surface of the membrane were fixed in 4% PFA and then stained with DAPI (diluted in PBS, 1:750). Each membrane was imaged by taking 12 images per membrane at 10 x magnification and the number of nuclei counted by means of Image J software. Pairwise comparisons were determined using the Student’s t test, with a significance of p<0.05 indicated by asterisks.

After 6 days of BAP treatment with siRNA transfection as above, spent culture medium was collected to measure hCG (human chorionic gonadotrophin) levels (Genway Biotech, LLC; GWB-BQK0F2). Separate but identically treated wells had spent culture medium collected on culture days 7 and 8. Total DNA was isolated (Promega, #A2360) from the respective cultures in order to normalize immunoassay results to cell density levels. ELISAs were performed by following the manufacturer-recommended protocols. Samples were collected in three independent experiments for each treatment. Data followed over time were subjected to two-way ANOVA, followed by the Bonferroni test for pairwise comparisons. Values of p<0.05 were considered to support the conclusion that differences were statistically significant.

H1 (WA01) ESCs were differentiated to TB with BMP4, A83-01, and PD173074 (BAP treatment) as previously described for up to 8 days (10 total days in culture) (19). Total protein was collected from BAP cells on treatment days 3 through 8 ( Figure 2A ). Western immunoblotting demonstrated marked increases in VTCN1 in BAP treated cells beginning 4 days after exposure and continuing through d8 (10 days of culture) ( Supplementary Figure 1A ).

To determine possible roles for VTCN1 in general TB development and function, we used siRNA to block VTCN1 transcription ( Figure 2A ). VTCN1 mRNA knock-down (KD) was confirmed by qPCR and decreases in protein expression demonstrated by western blotting and immunofluorescence histochemistry (IHC, Figure 2B , Supplementary Figures 1B-D ). We began our analyses on the effects of VTCN1 KD by assessing typical BAP-induced TB developmental milestones. To determine whether VTCN1 regulated the syncytialization of TB cells, we transfected BAP cells with VTCN1-specific and scrambled control siRNA on treatment d3 ( Figure 2A ). Direct phase contrast imaging and crystal violet staining, which preferentially highlights syncytialized patches of cells within the colonies, demonstrated impairment of syncytialization in VTCN1 siRNA- but not in scrambled control siRNA-exposed cells ( Figure 2C ). Fluorescence immunohistochemistry was used to assess the effects of VTCN1 suppression on two protein markers of TB cell syncytialization, CGB and CGA ( Supplementary Figure 1E , 2D , respectively). VTCN1 KD did not affect CGB expression on BAP d6, which was expected, as CGB expression in BAP cells is typically first detected later in the differentiation progression (19). In contrast, CGA expression, which typically increases in the days prior to the initiation of syncytialization in BAP cultures, decreased upon VTCN1 KD ( Figure 2D ), consistent with the visual reductions in STB areas within colonies discussed above ( Figure 2C ).

We next used time-course experiments beginning on d6 of BAP treatment (3 days after VTCN1 KD) to document the effects of VTCN1 KD on secretion of hCG, whose concentrations were normalized to DNA content of the respective cultures to adjust for possible treatment-related differences in cell growth and proliferation. Confirming the IHC results, BAP treated TB cells exposed to VTCN1 siRNA produced similar amounts of hCG on d6 when compared to scrambled siRNA controls ( Figure 2E ).

The noted decrease in syncytialization imparted by VTCN1 KD led us to hypothesize that the BAP cells were differentiating through an alternative developmental pathway, most likely conversion into invasive EVT-like cells (22). In experiments replicating the BAP and transfection methods described above, BAP treated TB cells were exposed to VTCN1 or scrambled control siRNA but were then cultured under conditions that allowed specific assessment of TB cell invasion (invasion assays). The invasive capacity of BAP cells increased after VTCN1 KD when compared to scrambled siRNA-transfected cells (p=0.032) ( Figure 2F ).

We next performed RNA-seq on BAP cells at 24, 48 and 72 h after VTCN1 KD to define pathways regulated by VTCN1 better. A total of 61 genes were up-regulated and 155 were down-regulated 24 h after transfection. At this time point, 21 up-regulated pathways were enriched with a criterion of p<0.05 while no down-regulated pathways were enriched ( Supplementary Figure 2A , Supplementary Table 1 ). Upregulated, differentially expressed genes at the 24 h time point were highly associated with pathways that included: herpes simplex infection, influenza A and cytokine receptor interaction. Of the 21, 20 were associated with immune responses. Analyses of the 48 h and 72 h post-transfection samples identified a further 12 and seven upregulated and enriched pathways, respectively ( Supplementary Figure 2B, C , Supplementary Table 1 ), with upregulated differentially expressed genes highly associated with pathways that again included responses to virus.

To assess the effect of VTCN1 KD at the single gene level in more detail, we generated a clustering heatmap of differentially expressed genes. The top 20 genes identified in this analysis are shown in the heatmap in Figure 3A , among which 16 genes are related to type I IFN responses, which are known to mediate innate immune responses to viruses. The genes most altered by VTCN1 suppression when compared to scrambled siRNA exposed controls were next analyzed by using QIAGEN Ingenuity Pathway Analysis software (QIAGEN IPA). The affected functional categories with highest significance 24 h, 48 h and 72 h after transfection are presented in Figure 3A, B , Supplementary Figure 3A and Supplementary Figure 4A . Potential upstream regulators of the differentially expressed genes as predicted from the literature by IPA (Ingenuity Pathway Analysis) are shown in Figure 3C , Supplementary Figure 3B and Supplementary Figure 4B .

RNA-seq data rlog counts normalized to scrambled siRNA controls demonstrated a decrease in MAPK and an increase in STAT1 transcripts in response to VTCN1 siRNA knock-down ( Figures 4A, B ). Additionally, the expression of EVT pro-invasion markers, Integrin Subunit Alpha 5 (ITGA5) and matrix metalloproteinases (MMP)-12, were significantly increased after VTCN1 had been knocked down ( Supplementary Table 2 ). To verify the documented transcriptional changes seen with bulk RNA-seq as well as activation of the MAPK/ERK1/2 and JAK/STAT signaling pathways upon VTCN1 KD, protein levels of MAPK, phospho-MAPK (pMAPK) and phospho-Stat1 (pSTAT1) were examined by western blotting. As with the transcriptional changes in STAT1, pSTAT1 level increased in TB cells upon VTCN1 KD ( Figure 4D ). While the direction of change in pMAPK protein expression upon VTCN1 KD followed that of the transcriptional changes, protein levels of pMAPK, using an antibody detecting p44/42 MAPK, were significantly elevated ( Figure 4C ).

RNAseq data from VTCN1 KD cells also highlighted an increase in IFITM1 transcripts when compared to scrambled control siRNA exposed cells. ( Figure 4E ). The IFN-induced transmembrane protein (IFITM) family includes members (IFITM1, -2, and -3) that protect multiple cell types from viral infections by preventing viral membrane fusion with cells and by inhibiting the syncytialization of infected cells (23–25). Since syncytialization is central to TB development, we compared IFITM1 protein dynamics by Western blotting to placental TB developmental changes in BAP primed TB with or without VTCN1. As expected, IFITM1 increased as syncytialization decreased when VTCN1 was knocked down in BAP cells ( Figure 4F ).

RNA-seq analyses provided evidence that the mRNA levels of classical MHC class I molecules, HLA-A, HLA-B and HLA-C increase with VTCN1 suppression, while HLA-G transcripts remain relatively unchanged. The changes in HLA-A, -B, -C and -G expression were validated by RT-PCR and correlated well with protein data acquired by western immunoblotting, fluorescence ICC and flow cytometry ( Figures 5 , 6 ). HLA-A, HLA-B and HLA-C transcripts increased on d5 and d6 of BAP treatment (d2 and d3 post VTCN1 KD) ( Figure 5A ) and protein expression increased on BAP d6/VTCN1 KD d3 ( Figures 5B, C ). The effects of VTCN1 KD on the transcript and protein levels of HLA-G, a marker for EVT, were similarly analyzed and were not affected by treatment.

To identify possible off target gene silencing during siRNA treatment and/or the possibility of a purely interferon-stimulated increase in classical MHC class I expression secondary to a cellular response to the siRNA, we monitored HLA-A protein levels and IFN-stimulated genes expression in a VTCN1 CRISPR-Cas9 knock-out (KO) hESC cell line after BAP treatment. HLA-A protein on d4, d6 and d8 of BAP treated VTCN1 KO cells and IFN-stimulated genes including ISG15, OAS1 and MX1 mRNA levels on d6 of BAP treated VTCN1 KO cells increased when compared to BAP-exposed wild type H1 cells ( Figures 5D, E ), replicating findings in our siRNA-induced transient knock-down models ( Figure 3A , Figures 5A-C ).

To further verify a relationship between VTCN1 and MHC class I gene alterations during primitive TB differentiation, we used an alternative model and analyzed a publicly-available single cell RNA-seq database generated by West et al. for TB cells isolated from embryonic day (D) 8, D 10 and D 12 human embryos during blastocyst extended culture (26). Although the transcripts for VTCN1, HLA-A and HLA-B were all expressed at low levels in cytotrophoblast (CTB) at D 8, D 10 and D 12, the expression of VTCN1 progressively decreased as the expression of HLA-A and HLA-B increased over time in culture, suggesting a possible association between B7-H4 and MHC class I during early placental development. In contrast, HLA-C and the non-classical MHC I molecules HLA-E and HLA-G genes were more robustly expressed throughout culture, and expression increased dramatically from D 10 to D 12 in culture ( Supplementary Figure 5 ), likely a reflection of an increasing proportion of cells that are migratory (EVT-like) in the embryos.

We hypothesized that alterations in MHC cell surface expression in invasive primitive trophoblast of the implanting embryo might play a role in possible phenotype switching of peripheral maternal CD3-CD56^dimCD16⁺ NK cells toward the CD3^-CD56^brightCD16^- NK cells typical of the decidua (27). Since VTCN1 KD increased the surface expression of classical MHC class I molecules, including HLA-C, on BAP primed TB cells, we examined the effects of such changes on peripheral immune cells from pregnant women in a co-culture system. Cryopreserved PBMCs isolated from peripheral blood of pregnant women (5-12 weeks gestation) were thawed and cocultured for 3 days (d5 – d8 of BAP differentiation/d3–d5 post VTCN1 KD) with control and VTCN1 KD BAP-treated human ES cells. Non-adherent cells in the co-culture, largely PBMCs, were then isolated and subjected to flow cytometry. Lymphocytes were identified by FSC/SSC light scatter. The gating strategy for CD56 and CD16 expression on all CD3-negative cells is presented in Figure 7A . The percentage of CD3-CD56^dimCD16⁺ peripheral-type NK cells among CD3 negative lymphocytes increased (p=0.043, Figure 7B ) and the percentage of CD3^-CD56^brightCD16^- decidual NK-like cells decreased (p = 0.019, Figure 7C ) in PBMCs cocultured with VTCN1 KD TBs when compared to those cocultured with scrambled control siRNA-exposed TB.