Progesterone-Mediated Enhancement of Hepatitis E Virus Replication in Human Liver Cells

Abstract

Progesterone is crucial for the maintenance of pregnancy. During pregnancy hepatitis E virus (HEV) infection is associated with increased fulminant hepatic failure and mortality rates. In this study, we determined whether progesterone modulates HEV replication and HEV-induced innate cytokine response in Huh7-S10-3 human liver cells. We first demonstrated that Huh7-S10-3 liver cells expressed SH3-domain-containing progesterone receptor membrane component (PGRMC)1/2 receptors involved in the progesterone nonclassical signaling pathway, while the classical progesterone receptor isoforms progesterone receptor-A and -B protein levels were undetectable. We showed that the genotype 3 HEV (strain P6) induced mRNA expression of type III interferon (IFN-λ1), but not other innate cytokines in Huh7-S10-3 cells. Pretreatment with progesterone at concentrations of 80 nM, 160 nM, or 480 nM, which are the physiological concentrations typically seen in the first- to third-trimester during pregnancy, significantly increased HEV replication in Huh7-S10-3 cells. However, pretreatment of cells with progesterone (80 nM) did not affect the level of HEV-induced IFN-λ1 mRNA expression. We further showed that loss of PGRMC1/2 receptors by small interfering RNA (siRNA) knockdown leads to an increase in HEV-induced IFN-λ1 expression levels at early time points via the extracellular signal-regulated kinase pathway and thus resulted in a reduced level of HEV replication. Collectively, the results indicated that progesterone-mediated modulation of HEV replication in human liver cells is plausibly through SH3-domain containing proteins such as PGRMC1/2, but not likely through immunomodulation of HEV-induced interferon response in liver cells. The results have important implications in understanding the underlying mechanisms of high mortality and fulminant hepatitis in HEV-infected pregnant women. IMPORTANCE Hepatitis E is usually a self-limiting acute disease; however, during pregnancy, a severe form of fulminant hepatic failure and high mortality rate are associated with hepatitis E virus (HEV) infection. Increased levels of progesterone and HEV RNA are observed in pregnant women with fulminant hepatic failures. Since progesterone is crucial for maintenance of pregnancy, we investigated the potential role of progesterone in HEV replication and disease pathogenesis. We demonstrated that progesterone at a concentration seen during pregnancy enhances HEV replication in human liver cells, but did not modulate HEV-induced interferon response in human liver cells. We also showed that loss of the progesterone nonclassical receptor, progesterone receptor membrane component (PGRMC)1/2, leads to a reduced level of HEV replication and an increased level of HEV-induced type III interferon (IFN-λ1) mRNA expression via the extracellular signal-regulated kinase pathway. The results from this study will aid our understanding of the underlying mechanism of pathogenesis and HEV-associated severe disease during pregnancy.

Keywords: PGRMC1/2 receptor; hepatitis E virus; pregnancy; progesterone; type III interferon; virus replication.

FIG 1

Expression profile of progesterone receptor in human liver cells. (A) Western blot analysis of progesterone receptors, both classical receptor (PR-A, and PR-B) and nonclassical receptor (PGRMC1/2), in Huh7-S-10-3 liver cells and HepG2-C3A liver cells in the presence and absence of various concentrations of progesterone. (B) RT-PCR detection of progesterone receptor PR-B CDS, as well as the control G protein-coupled estrogen receptor (GPER) CDS in Huh7-S10-3 liver cells.

FIG 2

Progesterone enhances HEV replication at a concentration seen in the first trimester of pregnancy and in a dose-dependent manner. (A) Experimental regimen of progesterone treatment. (B) Intracellular HEV RNA levels. (C) Extracellular HEV RNA levels in HEV-P6 transfected Huh7-S10-3 cells, at day 5 (D5) post-HEV transfection, in the presence or absence of progesterone as determined by HEV RT-qPCR. (D) Experimental regimen of progesterone treatment during growth kinetics of HEV (D7 and D11). (E) Intracellular HEV RNA levels. (F) Extracellular HEV RNA levels, at D7 and D11 post-HEV transfection, in the presence of various concentrations of progesterone as determined by HEV RT-qPCR. Progesterone concentrations in final culture volume are 0.8 nM PRO = 0.25 ng/ml; 8 nM PRO = 2.5 ng/ml; 80 nM PRO = 25 ng/ml. (G) HEV negative-strand RNA levels in progesterone-treated cells compared to HEV-transfected cells without progesterone treatment as determined by HEV negative-strand RT-qPCR. Data represent average ± standard error of the mean (SEM) from panels B and C, representative of 3 independent experiments; for panels E, F, and G, n = 2 independent experiments; Student’s t test; *, P ≤ 0.05; **, P ≤ 0.01.

FIG 3

Progesterone pretreatment of Huh7-S10-3 human liver cell enhanced HEV replication as determined by IFA of HEV capsid protein and virus infectivity assay. (A) Representative images of immunofluorescence (two different fields) of HEV ORF2 capsid protein in HEV-P6 transfected Huh7-S10-3 cells, at D5 post-HEV transfection, in the presence or absence of 80 nM PRO. Nuclei are counterstained using DAPI. The inset bar diagram represents intracellular HEV RNA levels determined during the experiment. CC, “cell-control” without HEV transfection. (B and D) Infectious HEV titers as determined by an HEV infectivity assay. (C) HEV negative-strand RNA levels as determined by HEV negative-strand RNA RT-qPCR. (E) Cell proliferation level determined by a WST-1 assay. Data represent an average ± SEM from panels B (n = 2), C (n = 5), D (n = 3 independent experiments), and E (n = 6 replicates); Student’s t test; *, P ≤ 0.05; **, P ≤ 0.01 compared to P6. Progesterone final concentrations are 80 nM PRO = 25 ng/ml, 160 nM PRO = 50ng/ml, 480 nM = 150 ng/ml.

FIG 4

Progesterone pretreatment is required for enhanced HEV replication in human liver cells. (A) Experimental regimen of progesterone pretreatment (80 nM PRO Pre) and posttreatment (80 nM PRO Post). (B) HEV infectious titer. (C) HEV negative-strand RNA copy numbers. (D) Fold change (in percentage) in HEV-P6 transfected Huh7-S10-3 cells during pre- or posttreatment with progesterone. Data represent an average ± SEM from panels B (n = 3) and C and D (n = 7 independent experiments). Statistical analyses were performed with Student’s t test (panel B), one-way ANOVA with post hoc Tukey’s test (panel C), and the Friedman test (panel D) of differences among the multiple groups, which rendered a chi-square value of 7.71, which was significant (P = 0.02). *, P ≤ 0.05; **, P ≤ 0.01 compared to HEV-P6 by Student’s t test. 80 nM PRO = 25 ng/ml progesterone final concentration.

FIG 5

Innate cytokine mRNA profile induced by a known IFN inducer (HEV SL3, positive control) and by HEV-P6. (A) IFN-λ1 mRNA levels. (B) Other cytokine (IL-6, TNF-α, IL-22, IFN-α, IL-1β, and IL-8) mRNA levels. (C) IFN-β mRNA levels induced by HEV-P6 3′ UTR PAMP SL3-transfected (black-filled bars) Huh7-S10-3 cells compared to the mock cell control (CC, unfilled bars). (D) IFN-λ1, IL-6, TNF-α, and IFN-β levels in Huh7-S10-3 cells transfected with genomic RNA transcripts of HEV-P6 (gray-filled bars) compared to the mock cell control (CC). Cytokine mRNA levels were determined using gene-specific cytokine RT-qPCR. The fold change was calculated using the 2^-ddCt method. The data represent an average ± SEM from panels A and B (n = 2 independent experiments) and from panels C and D (n = 4 independent experiments); Student’s t test. N.D, not detected; N.S, nonsignificant.

FIG 6

Progesterone treatment had no effect on the HEV-induced IFN-λ1 innate immune response. (A) Progesterone pretreatment (n = 4 independent experiments); fold change in IFN-λ1 mRNA levels in HEV-P6 transfected Huh7-S10-3 cells. (B) Progesterone posttreatment (n = 4 independent experiments); fold change in IFN-λ1 mRNA levels in HEV-P6 transfected Huh7-S10-3 cells. The data represent an average ± SEM; Student’s t test. N.S, nonsignificant; 80 nM PRO = 25 ng/ml progesterone final concentration.

FIG 7

Loss of PGRMC1/2 by siRNA knockdown during progesterone treatment leads to a significantly decreased level of HEV replication. (A) Experimental regimen. (B) Percentage fold change in HEV negative-strand RNA levels; the Friedman test of differences among the multiple groups rendered a chi-square value of 8.40, which was significant (P = 0.03). (C) Infectious HEV titer (FFU/ml). The data represent an average ± SEM of panels B (n = 4) and C (n = 3 independent experiments); Student’s t test. N.S, nonsignificant; siCnt, scramble control siRNA transfected cells; siPGRMC1 + 2, PGRMC1/2 siRNA transfected cells; 80 nM PRO = 25 ng/ml progesterone final concentration.

FIG 8

Loss of PGRMC1/2 by siRNA knockdown during progesterone treatment increased the levels of HEV-induced IFN-λ1 expression via ERK. (A) Experimental regimen. (B) Fold change in IFN-λ1 mRNA levels in various conditions tested. (C and D) Western blot analysis of phosphorylated ERK (pERK) and total ERK (tERK) levels during progesterone treatment. (C) PGRMC1/2 knockdown cells. (D) ERK inhibitor U0126 (U01) treated cells. (E) Fold change in IFN-λ1 mRNA levels in the presence or absence of ERK inhibitor U0126 (U01). The data represent an average ± SEM of 2 independent experiments; Student’s t test, N.S, nonsignificant; siCnt, scramble control siRNA transfected cells; siPGRMC1 + 2, PGRMC1/2 siRNA transfected cells; U01, 10 μM ERK inhibitor U0126; 80 nM PRO = 25 ng/ml progesterone final concentration.