Overactivation of the nuclear factor (erythroid-derived 2)-like 2-antioxidant response element pathway in hepatocytes decreases hepatic ischemia/reperfusion injury in mice

Abstract

Hepatic ischemia/reperfusion injury (IRI) is a critical component of hepatic surgery. Oxidative stress has long been implicated as a key player in IRI. In this study, we examine the cell-specific role of the nuclear factor (erythroid-derived 2)-like 2 (Nrf2)-antioxidant response element pathway in warm hepatic IRI. Nrf2 knockout (KO) and wild-type (WT) animals and novel transgenic mice expressing a constitutively active nuclear factor (erythroid-derived 2)-like 2 (caNrf2) mutant in hepatocytes (AlbCre+/caNrf2+) and their littermate controls underwent partial hepatic ischemia or sham surgery. The animals were killed 6 hours after reperfusion, and their serum and tissue were collected for analysis. As compared to WT animals after ischemia/reperfusion (IR), Nrf2 KO mice had increased hepatocellular injury with increased serum alanine aminotransferase and aspartate aminotransferase, Suzuki score, apoptosis, an increased inflammatory infiltrate, and enhanced inflammatory cytokine expression. On the other hand, AlbCre+/caNrf2+ that underwent IR had significantly reduced serum transaminases, less necrosis on histology, and a less pronounced inflammatory infiltrate and inflammatory cytokine expression as compared to the littermate controls. However, there were no differences in apoptosis. Taken together, Nrf2 plays a critical role in our murine model of warm hepatic IRI, with Nrf2 deficiency exacerbating hepatic IRI and hepatocyte-specific Nrf2 overactivation providing protection against warm hepatic IRI.

Figure 1. Nrf2-dependent gene expression and serum ALT/AST levels following hepatic IRI

Hepatic mRNA levels of Nqo1, GstA2, GstA4, GCLC, and GCLM in (A) AlbCre+/caNrf2- and AlbCre+/caNrf2+ mice subjected to sham and IR surgery. Serum transaminases following sham or IR surgery in (B) Nrf2 WT and KO mice and (C) AlbCre+/caNrf2- and AlbCre+/caNrf2+ mice (AlbCre+/caNrf2+ Sham: n=3; AlbCre+/caNrf2- Sham: n=3; AlbCre+/caNrf2+ IR: n=8; AlbCre+/caNrf2- IR: n=6; WT Sham: n=4; KO Sham: n=4; WT IR: n=13; KO IR: n=16). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 2. Nrf2 deficiency increases damage following IRI

Representative H&E staining in livers of Nrf2 WT (A and B) and Nrf2 KO (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery. (E) Suzuki scoring of the H&E liver sections for congestion, vacuolization, necrosis, and additive total score (WT Sham: n=4; KO Sham: n=4; WT IR: n=13; KO IR: n=16). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 3. Hepatocyte-specific Nrf2 over-activation reduces necrosis following IRI

Representative H&E staining in livers of AlbCre+/caNrf2- (A and B) and AlbCre+/caNrf2+ (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery. (E) Suzuki scoring of the H&E liver sections for congestion, vacuolization, necrosis, and additive total score (AlbCre+/caNrf2+ Sham: n=3; AlbCre+/caNrf2- Sham: n=3; AlbCre+/caNrf2+ IR: n=8; AlbCre+/caNrf2- IR: n=6). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 4. Nrf2 deficiency increases apoptosis following IRI

Representative cleaved Caspase 3 staining in livers of Nrf2 WT (A and B) and Nrf2 KO (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery (WT Sham: n=4; KO Sham: n=4; WT IR: n=13; KO IR: n=16). (E) Western blot analysis and (F) densitometry of cleaved Caspase 3 protein following sham or IR surgery.* P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 5. Nrf2 over-activation in hepatocytes does not ameliorate the increased apoptosis following IRI

Representative cleaved Caspase 3 staining in livers of AlbCre+/caNrf2- (A and B) and AlbCre+/caNrf2+ (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery (AlbCre+/caNrf2+ Sham: n=3; AlbCre+/caNrf2- Sham: n=3; AlbCre+/caNrf2+ IR: n=8; AlbCre+/caNrf2- IR: n=6). (E) Western blot analysis and (F) densitometry of cleaved Caspase 3 protein following sham or IR surgery. * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 6. Nrf2 deficiency increases neutrophil infiltration following IRI

Representative Ly6G staining in livers of Nrf2 WT (A and B) and Nrf2 KO (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery. (E) Ly6G-positive cells were quantified from four random fields of each section at 20X magnification (WT Sham: n=4; KO Sham: n=4; WT IR: n=13; KO IR: n=16). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 7. Nrf2 over-activation in hepatocytes reduces neutrophilic infiltration following IRI

Representative Ly6G staining in livers of AlbCre+/caNrf2- (A and B) and AlbCre+/caNrf2+ (C and D) mice subjected to either sham (A and C) or IR (B and D) surgery. (E) Ly6G-positive cells were quantified from four random fields of each section at 20X magnification (AlbCre+/caNrf2+ Sham: n=3; AlbCre+/caNrf2- Sham: n=3; AlbCre+/caNrf2+ IR: n=8; AlbCre+/caNrf2- IR: n=6). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.

Figure 8. Nrf2 attenuates the upregulation of pro-inflammatory gene expression following hepatic IRI

Hepatic mRNA levels of IL-1β, IL-6, TNF-α, CCL2, CXCL10, and ICAM-1 in (A) Nrf2 WT and KO mice and (B) AlbCre+/caNrf2- and AlbCre+/caNrf2+ mice subjected to either sham (WT: n=4; KO: n=4; AlbCre+/caNrf2+: n=3; AlbCre+/caNrf2-: n=3) or IR surgery (WT: n=13; KO: n=16; AlbCre+/caNrf2+: n=8; AlbCre+/caNrf2-: n=6). * P<0.05 compared to sham surgery of the same genotype. # P<0.05 compared to control genotype of the same surgery.