Protective role of HO-1 and carbon monoxide in ethanol-induced hepatocyte cell death and liver injury in mice

Abstract

Background & aims: Alcoholic liver disease is associated with inflammation and cell death. Heme oxygenase-1 (HO-1) is a stress-inducible enzyme with anti-apoptotic and anti-inflammatory properties. Here we tested the hypothesis that induction of HO-1 or treatment with a carbon monoxide releasing molecule (CORM) during chronic ethanol exposure protects and/or reverses ethanol-induced liver injury.

Methods: Female C57BL/6J mice were allowed free access to a complete liquid diet containing ethanol or to pair-fed control diets for 25days. Mice were treated with cobalt protoporphyrin (CoPP) to induce HO-1 expression during ethanol feeding or once liver injury had been established. Mice were also treated with CORM-A1, a CO-releasing molecule (CORM), after ethanol-induced liver injury was established. The impact of HO-1 induction on ethanol-induced cell death was investigated in primary cultures of hepatocytes.

Results: Induction of HO-1 during or after ethanol feeding, as well as treatment with CORM-A1, ameliorated ethanol-induced increases in AST and expression of mRNAs for inflammatory cytokines. Treatment with CoPP or CORM-A1 also reduced hepatocyte cell death, indicated by decreased accumulation of CK18 cleavage products and reduced RIP3 expression in hepatocytes. Exposure of primary hepatocyte cultures to ethanol increased their sensitivity to TNFα-induced cell death; this response was attenuated by necrostatin-1, an inhibitor of necroptosis, but not by caspase inhibitors. Induction of HO-1 with CoPP or CORM-3 treatment normalized the sensitivity of hepatocytes to TNFα-induced cell death after ethanol exposure.

Conclusions: Therapeutic strategies to increase HO-1 and/or modulate CO availability ameliorated chronic ethanol-induced liver injury in mice, at least in part by decreasing hepatocellular death.

Keywords: Alcoholic liver disease; Apoptosis; Heme oxygenase-1; Hepatocytes; Necroptosis.

Figure 1. Induction of hemeoxygenase-1 (HO-1) with cobalt protoporphyrin (CoPP) treatment during chronic ethanol feeding to mice prevents the development of liver injury

C57BL/6 mice were allowed free access to ethanol containing diets or pair-fed control diets for 25d. Mice were treated with 5 mg/kg CoPP or vehicle (saline) three times at 7 day intervals on d5, d12 and d19 of the ethanol feeding protocol. (A) Plasma AST activity and (B) liver triglycerides were measured by biochemical assay. (C) Expression of mRNA for TLR-4, TNFα, IL6 and CXCL10 was measured by qRT-PCR. Values are expressed as the percent increase over pair-fed controls. (D) Immunoreactive 4-HNE adducts, (E) TUNEL positive nuclei and (F) HO-1 expression were visualized and semi-quantified in paraffin-embedded liver sections. Images were acquired using a 10× (4HNE and HO-1) or 20x (TUNEL) objective. For panels A, B, D, E, F, values with different superscripts are significantly different from each other (P < 0.05). For panel C, *P<0.05 compared to pair-fed mice.

Figure 2. Induction of hemeoxygenase-1 (HO-1) with cobalt protoporphyrin (CoPP) treatment after the development of chronic ethanol feeding to mice reverses liver injury

C57BL/6 mice were allowed free access to ethanol containing diets or pair-fed control diets for 25d. On d25 of the ethanol feeding protocol, mice were treated with 5 mg/kg CoPP or vehicle (saline) via intraperitoneal injection and euthanized 24h later. (A) Plasma AST activity and (B) liver triglycerides were measured by biochemical assay. (C) Expression of mRNA for TLR-4, TNFα, IL6 and CXCL10 was measured by qRT-PCR. Values are expressed as the percent increase over pair-fed controls. (D) M30, a caspase-cleavage product of CK18, was visualized in paraffin-embedded livers. M30-positive cells were counted and expressed as total number of cells per 20X frame. (E/F) Paraffin-embedded livers were immunostained for (E) RIP3 and (F) HO-1 expression. For panels A, B, D, E, F, values with different superscripts are significantly different from each other (P < 0.05). For panel C, *P<0.05 compared to pair-fed mice.

Figure 3. Treatment with the CO-releasing molecule, CORM-A1, after the development of chronic ethanol feeding to mice reverses liver injury

C57BL/6 mice were allowed free access to ethanol containing diets or pair-fed control diets for 25d. Mice were treated with 3mg/kg CORM-A1, a CO releasing molecule, via intraperitoneal injection five times at 24 h intervals on days 21–25 of the ethanol feeding protocol. (A) Plasma AST activity and (B) liver triglycerides were measured. (C) Expression of mRNA for TLR-4, TNFα, IL6 and CXCL10 was measured by qRT-PCR. Values are expressed as the percent increase over pair-fed controls. (D) Ly6C⁺ cells were visualized by immunostaining OCT-embedded frozen liver sections and positive cells were counted per 40X frame. (E) Paraffin-embedded liver sections were immunostained for isolevuglandin-2 (iso[4]LGE₂). (F) M30, a caspase-cleavage product of CK18, was visualized in paraffin-embedded livers. M30-positive cells were counted and expressed as total number of cells per 20X frame. (G) Paraffin-embedded livers were immunostained for RIP3. For panel C, *P<0.05 compared to pair-fed mice. For all other panels, values with different superscripts are significantly different from each other (P < 0.05).

Figure 4. Sensitization of primary hepatocytes to TNFα-induced cell death after exposure to ethanol

Primary hepatocytes were cultured in the presence or absence of 50 mM ethanol for 24h, then treated with or without inhibitors (20 µM Z-VAD-FMK, 20µM necrostatin-1, 4µM CoPP, 10µM CORM-3) for 4 h prior to challenge with 30 ng/ml TNFα for a further 24 h. (A, C, F) TNFα-induced cytotoxicity was assessed by MTT assay. (B) FACS analysis of propidium iodide and annexin V. Histograms are representative of 5 independent experiments. (D/E) Hepatocyte lysates were used to assess (D) RIP3-FADD interaction in a FADD-pull down assay (normal IgG was used as a negative control) or (E) probed for HO-1 expression in response to CoPP (HSC70 was used as a loading control). (F) AML12 cells were transfected with an HO-1 expression vector or empty vector, cultured with ethanol and TNFα-induced cytotoxicity assessed by MTT assays. Values represent means ± SEM, n=4–6, except for panel F, where n=2. *p<0.05 compared to cells not exposed to ethanol, †p<0.05 compared to cells not treated with TNF-α and **p<0.05 compared to cells treated with TNF-α within an ethanol/no ethanol treatment group.