Glucose-regulated protein 78 is an intracellular antiviral factor against hepatitis B virus

Abstract

Hepatitis B virus (HBV) infection is a global public health problem that plays a crucial role in the pathogenesis of chronic hepatitis, cirrhosis, and hepatocellular carcinoma. However, the pathogenesis of HBV infection and the mechanisms of host-virus interactions are still elusive. In this study, two-dimensional gel electrophoresis and mass spectrometry-based comparative proteomics were applied to analyze the host response to HBV using an inducible HBV-producing cell line, HepAD38. Twenty-three proteins were identified as differentially expressed with glucose-regulated protein 78 (GRP78) as one of the most significantly up-regulated proteins induced by HBV replication. This induction was further confirmed in both HepAD38 and HepG2 cells transfected with HBV-producing plasmids by real time RT-PCR and Western blotting as well as in HBV-infected human liver biopsies by immunohistochemistry. Knockdown of GRP78 expression by RNA interference resulted in a significant increase of both intracellular and extracellular HBV virions in the transient HBV-producing HepG2 cells concomitant with enhanced levels of hepatitis B surface antigen and e antigen in the culture medium. Conversely overexpression of GRP78 in HepG2 cells led to HBV suppression concomitant with induction of the positive regulatory circuit of GRP78 and interferon-beta1 (IFN-beta1). In this connection, the IFN-beta1-mediated 2',5'-oligoadenylate synthetase and RNase L signaling pathway was noted to be activated in GRP78-overexpressing HepG2 cells. Moreover GRP78 was significantly down-regulated in the livers of chronic hepatitis B patients after effective anti-HBV treatment (p = 0.019) as compared with their counterpart pretreatment liver biopsies. In conclusion, the present study demonstrates for the first time that GRP78 functions as an endogenous anti-HBV factor via the IFN-beta1-2',5'-oligoadenylate synthetase-RNase L pathway in hepatocytes. Induction of hepatic GRP78 may provide a novel therapeutic approach in treating HBV infection.

Fig. 1.

Protein profile differences between tetracycline-treated and non-tetracycline-treated HepAD38 cells. A and B, representative silver-stained 2-DE maps of proteins from tetracycline-treated (HBV suppression, Tet⁺; A) and non-tetracycline-treated (HBV induction, Tet⁻; B) HepAD38 cells (n = 3). Twenty-three differentially expressed spots that were identified by MALDI-TOF-MS and MS/MS scanning are marked with numbers. C, enlarged sections of the 2-DE maps showing different expressions of spot 1 (PCNA), spot 17 (LMNB1), spot 26 (GRP78), spot 34 (cathepsin D preproprotein (CTSD)), and spot 36 (PDIA3) between tetracycline-treated (Tet⁺) and non-tetracycline-treated (Tet⁻) HepAD38 cells.

Fig. 2.

Identification of GRP78 (spot 26) by MALDI-TOF MS and MS/MS analysis. A, MALDI-TOF MS spectrum of GRP78 labeled with masses detected and peptide assignments. The precursor ion 1934.04 m/z, highlighted by an open circle, was submitted for MS/MS scanning. Mox, oxidized methionine. B, MS/MS spectra of the precursor ion 1934.04 m/z for peptide DNHLLGTFDLTGIPPAPR of GRP78. b ions and y ions with corresponding peak values are marked.

Fig. 3.

Confirmation of GRP78 overexpression in HBV-replicating HepAD38 and pHBV-transfected HepG2 cells. A, a time-dependent decrease of GRP78 expression was revealed by Western blot in HepAD38 cells when HBV replication was suppressed by tetracycline. Tet⁺, tetracycline treatment. B, elevated mRNA and protein (inset) expression of GRP78 was confirmed by quantitative PCR and Western blotting, respectively, in HepG2 cells 48 h postnucleofection. GAPDH was used as the internal control. Data are expressed as mean ± S.D. Error bars represent S. D. (n = 3; *, p < 0.01).

Fig. 4.

The functional significance of GRP78 in HBV replication. The HBV titers (HBV/luciferase) of both cytoplasm and supernatant (A1) and the levels (absorbance/luciferase) of both secreted HBsAg and HBeAg (A2) were significantly increased when GRP78 was knocked down. A3, down-regulation of GRP78 by siGRP78-1 was confirmed by Western blotting in HepG2 cells 48 h post-transfection (n = 3). In contrast the HBV titers (HBV/luciferase) of both cytoplasm and supernatant (B1) and the levels (absorbance/luciferase) of both the secreted HBsAg and HBeAg (B2) were markedly decreased during the overexpression of GRP78. B3, up-regulation of GRP78 was validated by Western blotting in pGRP78-transfected HepG2 cells 48 h post-transfection (n = 3). Data are expressed as mean ± S.D. (n = 5; *, p < 0.05; **, p < 0.01).

Fig. 5.

Kinetics of HBV and GRP78 mRNA levels in HepAD38 cells. A, the intracellular HBV virions in siSARS- and siGRP78-1-treated HepAD38 cells. B, the mRNA levels of GRP78 in siSARS- and siGRP78-1-treated HepAD38 cells. The values were normalized with GAPDH mRNA levels (loading control) and expressed as -fold changes to base-line level (at time point 0 h), which was set as 1. Data are expressed as mean ± S.D. (n = 3; *, p < 0.01, siGRP78-1 versus siSARS at the same time point).

Fig. 6.

GRP78 is down-regulated in postlamivudine treatment liver biopsies. A, serum HBV DNA levels of 19 CHB patients pretreatment (median, 8.47 log₁₀ copies/ml; range, 6–9.37 log₁₀ copies/ml) and post-treatment (median, 6.18 log₁₀ copies/ml; range, 2–8.33 log₁₀ copies/ml; p < 0.001) with lamivudine. B, GRP78 staining scores of liver biopsies from 19 CHB patients pretreatment (median, 2.5; range, 2–4) and post-treatment (median, 2.5; range, 1–3.5; p = 0.019) with lamivudine. The box plots display the median (bold middle line), the 25th and 75th percentiles (box margins), and the minimum and maximum values (whiskers). Outlying values are expressed as circles, and p values are indicated. C1–C3, representative microphotographs of GRP78 immunochemical staining in a pair of liver biopsies before (C1) and after (C2) lamivudine treatment. C3 is the negative control without primary antibody (magnification, ×200).

Fig. 7.

Effects of GRP78 overexpression on IFNs and IFN-inducible gene mRNA expression and IFN stimulation on GRP78 expression in HepG2 cells. A, mRNA expression levels of GRP78, IFNs (IFN-α1 and IFN-β1), and IFN-inducible genes (ISG15, Mx1, OAS1, OAS2, OAS3, OASL, RNase L, and PKR measured by real time RT-PCR. B1, dose-dependent effect of IFN-β1a on the GRP78 expression in HepG2 cells. Total RNA was isolated from HepG2 cells 8 h after the treatment with varying concentrations of IFN-β1a (0, 1, 10, 100, and 1000 units/ml), and real time RT-PCR was performed to measure GRP78 expression. B2, time course change of the mRNA expression of GRP78 after IFN-β1a treatment. Total RNA was extracted from HepG2 cells treated with 1000 units/ml IFN-β1a at the indicated incubation times (0, 2, 4, 6, 8, 10, and 12 h) and used for real time RT-PCR analysis. The expression levels of GRP78, IFNs, and IFN-inducible genes were normalized to GAPDH mRNA level and are expressed as -fold changes to corresponding controls. Data are expressed as mean ± S.D. (n = 3; *, p < 0.05; **, p < 0.01). C1, time-dependent elevation of GRP78 protein expression induced by IFN-β1a (1000 units/ml) (n = 3). C2, GRP78 protein expression in response to IFN-αA (1000 units/ml) and IFN-β1a (1000 units/ml) treatment, respectively, for 24 h (n = 3).

Fig. 8.

Proposed mechanisms for the anti-HBV activity of GRP78 in HepG2 cells. The mutual activation of GRP78 and IFN-β1 may compose a positive feedback loop that amplifies signal by inducing OAS1 and OAS2 overexpression. The antiviral effector RNase L is then activated, thereby suppressing HBV replication.