NRF2 Serves a Critical Role in Regulation of Immune Checkpoint Proteins (ICPs) During Trophoblast Differentiation

Abstract

Using cultured human trophoblast stem cells (hTSCs), mid-gestation human trophoblasts in primary culture, and gene-targeted mice, we tested the hypothesis that the multinucleated syncytiotrophoblast (SynT) serves a critical role in pregnancy maintenance through production of key immune modulators/checkpoint proteins (ICPs) under control of the O2-regulated transcription factor, NRF2/NFE2L2. These ICPs potentially act at the maternal-fetal interface to protect the hemiallogeneic fetus from rejection by the maternal immune system. Using cultured hTSCs, we observed that several ICPs involved in the induction and maintenance of immune tolerance were markedly upregulated during differentiation of cytotrophoblasts (CytTs) to SynT. These included HMOX1, kynurenine receptor, aryl hydrocarbon receptor, PD-L1, and GDF15. Intriguingly, NRF2, C/EBPβ, and PPARγ were markedly induced when CytTs fused to form SynT in a 20% O2 environment. Notably, when hTSCs were cultured in a hypoxic (2% O2) environment, SynT fusion and the differentiation-associated induction of NRF2, C/EBPβ, aromatase (CYP19A1; SynT differentiation marker), and ICPs were blocked. NRF2 knockdown also prevented induction of aromatase, C/EBPβ and the previously mentioned ICPs. Chromatin immunoprecipitation-quantitative PCR revealed that temporal induction of the ICPs in hTSCs and mid-gestation human trophoblasts cultured in 20% O2 was associated with increased binding of endogenous NRF2 to putative response elements within their promoters. Moreover, placentas of 12.5 days postcoitum mice with a global Nrf2 knockout manifested decreased mRNA expression of C/ebpβ, Pparγ, Hmox1, aryl hydrocarbon receptor, and Nqo1, another direct downstream target of Nrf2, compared with wild-type mice. Collectively, these compelling findings suggest that O2-regulated NRF2 serves as a key regulator of ICP expression during SynT differentiation.

Keywords: C/EBPβ; NRF2; immune modulators/checkpoint proteins (ICPs); trophoblast.

Figure 1.

Hypoxia inhibits cell fusion and mRNA expression of SynT-specific markers in cultured hTSCs. hTSCs were cultured in medium containing 2 μM forskolin to promote SynT differentiation. The cells were cultured in 20% O₂ or 2% O₂ for 1, 3, or 5 days. (A-C) Hypoxia inhibited the temporal induction of SynT differentiation markers (aromatase, CGB3, and PSG4), compared with cells cultured in 20% O₂. (D) Plakoglobin immunostaining (green) of hTSCs after 5 days of culture in 20% O₂ or 2% O₂. Nuclei were stained with DAPI (blue). Shown is a representative confocal fluorescence image (magnification: 400×). ImageJ was used to quantify fused cells in 10 fields in each condition. Fusion index = (N − S)/T×100%. N refers to the number of nuclei in the syncytia, S refers to the number of syncytia, and T refers to the total number of nuclei. Scale bar: 40 μm. Values are the mean ± SEM (n ≥ 3); *P < 0.05, **P < 0.01, ***P < 0.001 vs 0 d; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs 20% O₂.

Figure 2.

Hypoxia inhibits expression of transcription factors NRF2 and C/EBPβ in cultured hTSCs. hTSCs cultured in SynT differentiation medium containing forskolin in 20% O₂ or 2% O₂ for 1, 3, or 5 days were harvested for analysis by RT-qPCR and western blotting. mRNA (A) and nuclear protein levels (D, E) of NRF2 increased markedly during culture of hTSCs in 20% O₂ and were inhibited in response to culture in a hypoxic (2% O₂) environment after 3 and 5 days of culture. (A, D, E). Keap1 protein expression was reduced during SynT differentiation after 5 days of culture in 20% oxygen and remained increased at levels similar to those of undifferentiated CytT after 5 days of culture in 2% O₂ (D, F). Expression of C/EBPβ at both mRNA (B) and protein (G, H) levels was markedly increased after 3 and 5 days of culture in SynT differentiation medium in 20% O₂ and was inhibited in response to culture in 2% O₂. Shown are representative images of immunoblots (D, G) and quantifications of the scans of blots (E-I). Values are the means ±SEM of data from 3 to 5 independent experiments; *P < 0.05; **P < 0.01; ***P < 0.001 vs 0 d; ^#P < 0.05; ^##P < 0.01; ^###P < 0.001 vs 20% O₂ at 5 days.

Figure 3.

Expression of ICPs was increased during hTSC differentiation to SynT and was inhibited by hypoxia. hTSCs were cultured for up to 5 days in SynT differentiation medium in 20% or 2% O₂ and analyzed for HMOX1, AhR, PD-L1, and GDF15 mRNA (A-D) and protein (E-I) expression. (A-D) mRNA expression of the ICPs increased significantly during SynT differentiation in 20% O₂ and was inhibited by culture in 2% O₂. (E) Representative immunoblots of HMOX1, AhR, PD-L1, and GDF15 protein levels in hTSCs before (CytT) and after culture for 5 days in 20% or 2% O₂. (F-I) Quantifications of scans of immunoblots from at least 3 independent experiments. (J) Consistent with GDF15 protein levels in cell lysates, the secreted form of GDF15, measured by human GDF15 ELISA of the culture medium, was increased during SynT differentiation in 20% O₂, but inhibited by culture in 2% O₂. Values are the means ± SEM of data from ≥ 3 independent experiments; *P < 0.05; **P < 0.01; ***P < 0.001 vs 0 d (CytT); ^#P < 0.05; ^##P < 0.01; ^###P < 0.001 vs 20% O₂.

Figure 4.

NRF2 knockdown (KD) inhibits induction of C/EBPβ, aromatase, CGB3, and ICPs during hTSC differentiation. Inhibition of NRF2 by transfection of NRF2 siRNA (A) inhibited mRNA expression of C/EBPβ (B), SynT differentiation markers aromatase and CGB3 (D, E), and ICPs (F-I). Values are the means ± SEM of data from at least 3 independent experiments (n ≥ 3); *P < 0.05, ***P < 0.001 vs 0 day; ^##P < 0.01, ^###P < 0.001 vs transfection of nontargeting control.

Figure 5.

Binding of endogenous NRF2 to promoters of ICPs in hTSCs increases during SynT differentiation. ChIP-qPCR was used to analyze binding of endogenous NRF2 to promoters of genes encoding HMOX1 (A), AhR (B), PD-L1 (C), and GDF15 (D) before (0 days) and after 1 and 3 days of culture in SynT differentiation medium. Nonimmune IgG was used as a nonspecific control. Values are shown as percentage input. Values are the means ± SEM of data from at least 3 independent experiments (n ≥ 3); *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs NRF2 binding at 0 hours.

Figure 6.

Placental immune modulator gene expression is significantly decreased, whereas proinflammatory IL-1β is increased in Nrf2 global KO mice at E12.5. (A) Abrogated mRNA expression of Nrf2 in placentas of Nrf2 KO mice was confirmed by qRT-PCR. (B, C) Placentas of Nrf2 KO mice at E12.5 displayed a significant decrease in C/EBPβ, PPARγ, HMOX1, AhR, and another direct downstream target of Nrf2, Nqo1. (D) IL-1β mRNA was significantly increased in placentas of Nrf2 KO mice, compared with wild-type (WT). (n ≥ 8 for WT; n ≥ 12 for Nrf2 KO.) *P < 0.05; **P < 0.01; ***P < 0.001 vs WT.

Figure 7.

Increased O₂ tension promotes SynT differentiation with induction of the redox-sensitive transcription factor, NRF2, which induces expression of C/EBPβ, aromatase/CYP19a1 and ICPs through binding to their promoters. In this manner, NRF2 and downstream ICP target genes serve a critical role in the maintenance of immune tolerance at the maternal-fetal interface.