Mutual antagonism between circadian protein period 2 and hepatitis C virus replication in hepatocytes

Abstract

Background: Hepatitis C virus (HCV) infects approximately 3% of the world population and is the leading cause of liver disease, impacting hepatocyte metabolism, depending on virus genotype. Hepatic metabolic functions show rhythmic fluctuations with 24-h periodicity (circadian), driven by molecular clockworks ticking through translational-transcriptional feedback loops, operated by a set of genes, called clock genes, encoding circadian proteins. Disruption of biologic clocks is implicated in a variety of disorders including fatty liver disease, obesity and diabetes. The relation between HCV replication and the circadian clock is unknown.

Methods: We investigated the relationship between HCV core infection and viral replication and the expression of clock genes (Rev-Erbα, Rorα, ARNTL, ARNTL2, CLOCK, PER1, PER2, PER3, CRY1 and CRY2) in two cellular models, the Huh-7 cells transiently expressing the HCV core protein genotypes 1b or 3a, and the OR6 cells stably harboring the full-length hepatitis C genotype 1b replicon, and in human liver biopsies, using qRT-PCR, immunoblotting, luciferase assays and immunohistochemistry.

Results: In Huh-7 cells expressing the HCV core protein genotype 1b, but not 3a, and in OR6 cells, transcript and protein levels of PER2 and CRY2 were downregulated. Overexpression of PER2 led to a consistent decrease in HCV RNA replicating levels and restoration of altered expression pattern of a subset of interferon stimulated genes (ISGs) in OR6 cells. Furthermore, in liver biopsies from HCV genotype 1b infected patients, PER2 was markedly localized to the nucleus, consistent with an auto-inhibitory transcriptional feedback loop.

Conclusions: HCV can modulate hepatic clock gene machinery, and the circadian protein PER2 counteracts viral replication. Further understanding of circadian regulation of HCV replication and rhythmic patterns of host-hosted relationship may improve the effectiveness of HCV antiviral therapy. This would extend to hepatic viral infections the current spectrum of chronotherapies, implemented to treat metabolic, immune related and neoplastic disease.

Figure 1. qRT-PCR analysis of clock gene mRNA expression in OR6 control cells (cured, not expressing the HCV 1b full replicon) and in HCV replicating OR6 cells (HCV).

Control (cured) OR6 cells and OR6 cells replicating HCV genotype 1b were serum shocked and RNA was extracted every 1 h, 4 h, 10 h, 16 h, 22 h and 28 h after serum shock. mRNA levels of Rev-Erbα, Rorα, ARNTL, ARNTL2, CLOCK, PER1, PER2, PER3, CRY1 and CRY2 genes were assessed by qRT-PCR (A and B). Values were normalized against TBP as housekeeping control gene. Results are expressed as means ± SE of three independent experiments. * = p<0.05 in HCV replicating cells versus control cured OR6 cells.

Figure 2. Mutual antagonism between HCV genotype 1b replication and PER2 expression in OR6 cells.

(A) Representative immunoblots of PER2 and CRY2 protein expression in HCV replicating OR6 cells versus control (cured) cells. 1 out of 3 blots is showed. β-actin was used as a loading control. (B) PER2 overexpression in HCV replicating OR6 cells using a Flag-tagged construct. HCV replicating OR6 cells or control OR6 cells were electroporated with Flag-PER2, 24 hours after electroporation protein lysates were processed for immunoblotting and Flag levels were detected with a specific antibody. β-actin was used as a loading control. (C) HCV replication was detected by a luciferase reporter assay in HCV replicating OR6 cells 24 hours post electroporation with a control (GFP) or Flag-PER2 construct. (D) qRT-PCR to detect HCV RNA levels upon control (GFP) or Flag-PER2 plasmid electroporation in HCV replicating OR6 cells HCV RNA values were normalized against TBP as housekeeping control gene (C and D). Results are expressed as means ± SE of three independent experiments. * = p<0.05, ** = p<0.01 in HCV replicating OR6 cells overexpressing Flag-PER2 versus HCV replicating OR6 cells overexpressing GFP.

Figure 3. qRT-PCR analysis of interferon stimulated genes (ISGs: OAS1, Mx1, IRF9, PKR) mRNAs expression levels in control (cured) OR6 cells and HCV replicating OR6 cells electroporated with GFP (cured GFP, HCV GFP) or with Flag-PER2 (cured PER2, HCV PER2) A, B, C and D.

mRNA was extracted 24 hours after electroporation with the respective constructs and ISGs levels were normalized to TBP as housekeeping gene. Results are expressed as means ± SE of three independent experiments. * = p<0.05, ** = p<0.01 in HCV replicating OR6 cells overexpressing GFP versus OR6 control cells overexpressing GFP or Flag-PER2. # = p<0.05 in HCV replicating OR6 cells overexpressing Flag-PER2 versus HCV replicating OR6 cells overexpressing GFP.

Figure 4. qRT-PCR analysis of clock gene mRNAs expression levels in Huh-7 cells overexpressing the HCV core proteins genotype 1b and 3a.

Huh-7 cells were transiently transfected with HCV core proteins 1b and 3a as previously described , or with GFP. 48 hours after transfection mRNA levels of Rev-Erbα, Rorα, ARNTL, ARNTL2, CLOCK, PER1, PER2, PER3, CRY1 and CRY2 genes were assessed by qRT-PCR. Values were normalized against TBP as housekeeping control gene. Light gray: GFP transfected cells; dark gray: HCV core protein genotype 3a transfected cells; black: HCV core protein genotype 1b transfected cells. Results are expressed as means ± SE of three independent experiments. * = p<0.05 in HCV core proteins transfected versus GFP transfected control cells.

Figure 5. Immunoblot detection of circadian proteins in Huh-7 cells expressing the HCV core protein genotype 1b or 3a and GFP-expressing control cells.

(A) 48 hours after transfection cells were lysed and equal amounts of proteins were loaded on a 10% polyacrylamide gel, separated by electrophoresis and immunoblotted with specific Rev-Erbα, Rorα, CLOCK, ARNTL, ARNTL2, PER1, PER2, CRY1 and CRY2 primary antibodies. β-actin expression served as loading control. (B) Densitometric quantification of CRY2, PER2 and CLOCK proteins normalized to β-actin expression of three different experiments.

Figure 6. Representative panel of immunostainings performed for PER2 in samples of normal liver, hepatitis related to HCV (genotype 1b positive) and cirrhosis arisen on HCV infection (genotype 1b positive)

(A). Positivity of nuclei of hepatocytes in hepatitis was significantly higher compared with those in cirrhosis. The latter was comparable to that of hepatocytes in normal liver. In cirrhosis, PER2 Immunopositivity is also observed in the cytoplasm of hepatocytes. Positive control of esophagus showed immunopositivity in the nuclei of all cells (C). Pictures showed the same area observed with a lower (above) and higher (below) magnification. (B) Histogram shows statistical results for the evaluation of immunopositivity for PER2. Significant differences (** = p<0.005) in the percentage of positive nuclei were found between biopsies of hepatitis and either cirrhosis and normal liver.

Figure 7. Scheme illustrating the antagonism between HCV genotype 1b replication and PER2 in hepatocytes.

HCV core protein genotype 1b, via a yet undefined mechanism, induces a downregulation of PER2 mRNA. PER2 protein produced in the cytoplasm accumulates in the nucleus, as observed in human liver biopsies infected with HCV genotype 1b, where could further inhibit the production of its own RNA , . Alteration of this equilibrium by exogenous overexpression of PER2 protein hampers HCV genotype 1b replication.