Heme oxygenase-1 suppresses hepatitis C virus replication and increases resistance of hepatocytes to oxidant injury

Abstract

Oxidative injury to hepatocytes occurs as a result of hepatitis C virus (HCV) infection and replication. Modulation of host cell antioxidant enzymes such as heme oxygenase-1 (HO-1) may be useful therapeutically to minimize cellular injury, reduce viral replication, and attenuate liver disease. In this report, we evaluated the effects of HO-1 overexpression on HCV replication and hepatocellular injury. Full-length (FL) (Con1) or nonstructural (NS) replicons (I 389 NS3-3') were transfected with complete human HO-1 sequences or empty vector for control. Cell lines overexpressing HO-1 (twofold to sixfold above basal values) or empty vector were isolated, and their HCV RNA synthesis, pro-oxidant levels, and resistance to oxidative injury were assessed. HO-1 overexpression decreased HCV RNA replication in both FL and NS replicons without affecting cellular growth or DNA synthesis. The attenuation of HCV replication was significantly reversed in both replicon systems with HO-1 small interfering RNA (siRNA) knockdown. Both FL and NS replicons that overexpress HO-1 showed reduced prooxidant levels at baseline and increased resistance to oxidant-induced cytotoxicity. HO-1 induction with hemin also markedly decreased HCV replication in both parental FL and NS replicon cell lines. Conversely, knockdown of HO-1 messenger RNA (mRNA) by siRNA in parental FL or NS replicons did not significantly affect HCV replication, suggesting that less than basal levels of HO-1 had minimal effect on HCV replication.

Conclusion: Overexpression or induction of HO-1 results in decreased HCV replication as well as protection from oxidative damage. These findings suggest a potential role for HO-1 in antiviral therapy and therapeutic protection against hepatocellular injury in HCV infection.

Figure 1. Western blots of antioxidative enzymes in replicon parental lines and subclones overexpressing HO-1 or vector only controls

Whole cell lysates were electrophoresed on SDS gels and immunoblots performed using appropriate antibodies. A. Analysis of antioxidant enzymes in Huh5-15NS and Huh7.5FL replicon and parental cell lines. Lanes for HO-1 detection contained 15ug of cellular protein, while other lanes contained 30ug of cellular protein. B. Huh5-15NS subclones overexpressing HO-1 (515ZH7, 8, 15 and 16) vs vector only controls (515Z7 and 8). All lanes contained 30ug of cellular protein. C. Huh7.5FL subclones overexpressing HO-1 (Huh75ZH2, 3 and 4) vs vector only controls (Huh75Z1 and 2). All lanes contained 30ug of cellular protein.

Figure 2. Relative quantification of HCV replication in Huh5-15NS and Huh7.5FL replicons that overexpress HO-1 or vector only

A and B. HCV RNA was measured in HO-1 overexpressing or control subclones of Huh5-15NS (A) or Huh7.5FL (B) replicons using Real Time RT-PCR and comparative cycle threshold methodology as described in Methods. ANOVA was used to assess differences between the amount of HCV RNA in replicons overexpressing HO-1 as compared to empty vector. Each point is the mean of 9 determinations and plotted as +/- SEM of the group mean. Huh5-15NS vector only vs HO-1 overexpressing (* p < 0.01). Huh7.5FL vector only vs HO-1 overexpressing (** p < 0.05). C and D. HCV RNA was quantified using Real Time RT-PCR after HO-1 RNA knockdown in replicon cells as described in the Methods. Results shown are for HO-1 overexpressing Huh5-15NS (C) or Huh7.5FL (D) subclones after 72hr or 72-120 hr knockdown respectively. The amount of HCV RNA after knockdown (HO-1 siRNA or scramble RNA) in the HO-1 overexpressing subclones was quantified relatively to the amount of HCV RNA in replicon subclones expressing empty vector using the comparative cycle threshold method as described in Methods. ANOVA was used to assess differences between HO-1 siRNA and scramble RNA. Each point is the mean of 9 determinations and plotted as +/− SEM of the group mean. C and D Lower panels. Immunoblots of HO-1 protein in lysates of subclonal lines following HO-1 siRNA or scramble RNA knockdown.

Figure 3. Effects of HO-1 induction (A) with hemin or HO-1 siRNA knockdown (B) on HCV replication

A. HO-1 mRNA was quantified using CT methodology in replicon cells after incubation with control vehicle or 25 uM hemin overnight. B. HCV RNA was quantified similarly after 48 hr of HO-1 mRNA knockdown in replicon cells. The relative quantity of HCV RNA in either un-induced vs hemin treated cells (A) or siRNA vs scramble RNA knockdown (B) were compared. Each point is the mean of 9 determinations and plotted as +/− SEM of the group mean.After either treatment, cells were also assayed for HO-1, core protein, or NS5A protein by Western blots.

Figure 4. Proliferation of Huh5-15NS and Huh7.5FL clonal replicons with HO-1 overexpression or vector only controls

Clonal lines were seeded into wells and proliferation determined by either MTT dye (A, B) or ³H-thymidine incorporation (C) at various times during growth of cultures. Each point is the mean of 6 determinations and plotted as +/- SEM of the group mean. There was no statistical difference in growth between cells overexpressing HO-1 as compared to vector only control cells [p > 0.05]. Clonal lines of Huh5-15NS overexpressing vector only (5-15Z7 and 5-15Z8) or HO-1 (5-15ZH15 and 5-15ZH18) as well as Huh7.5FL overexpressing vector only (75Z1 and 75Z2) or HO-1 (75ZH2 and 75ZH3) are as described in the text.

Figure 5. Susceptibility of Huh5-15NS or Huh7.5FL replicons with or without HO-1 overexpression to peroxide injury

Semiconfluent cultures of Huh5-15NS (A), or Huh7.5FL(B) replicons with or without HO-1 overexpression were treated with varying concentrations of tBOOH for three hours, then viability determined with MTT dye assay. Toxicity was compared in Huh5-15NS replicons overexpressing HO-1 or vector only controls (5-15ZH15 and 16 vs 5-15Z7 and 8) and Huh7.5FL replicons overexpressing HO-1 or vector only controls (75ZH2 and 3 vs 75Z1 and 2). Data are plotted as the group mean of 9 determinations per point +/− SEM. Differences between groups were determined with ANOVA and t-test. * 515ZH16 vs 515Z7 p < 0.05 ** 75ZH2 or 75ZH3 vs 75Z1 or 75Z2 p < 0.01

Figure 6. Effects of HO-1 knockdown on susceptibility to peroxide cytotoxicity

HO-1 knockdown was assessed after transfection of HO-1 siRNA or scrambled control RNA into parental Huh5-15NS and Huh7.5FL replicons. 48 hours after transfection, cells were assayed for HO-1 expression with either Western blots (A) or treated with tBOOH (400 uM) for three hours, then cytotoxicity assessed with MTT assay (B) as described in Methods. The percentage differences were calculated with respect to control incubations that received vehicle only. Each value is the mean of 9 determinations +/− SEM normalized to a group coefficient of variation which is plotted. Differences between groups were determined with ANOVA and t-test. *Increased susceptibility to peroxide toxicity with siRNA knockdown of HO-1 vs scrambled RNA in Huh 5-15NS or Huh 7.5FL replicons [p < 0.05 and p < 0.01 respectively]. **Increased tolerability of Huh5-15NS replicons to HO-1 knockdown and peroxide toxicity as compared to Huh7.5FL: [p < 0.01].

Figure 7. Intracellular prooxidant production in HO-1 overexpression vs control cells

Replicons either with HO-1 overexpression or vector only control were assayed for prooxidant activity using DCF-DA methodology as described in Methods. Data are plotted as the mean of 9 determinations per point +/− SEM. Differences between groups were determined with ANOVA and t-test. **Significant reduction of DCF activity in HO-1 transfected Huh5-15NS or Huh7.5FL cells as compared with vector only control cells (** p<0.01).