Impaired liver regeneration in Nrf2 knockout mice: role of ROS-mediated insulin/IGF-1 resistance

Abstract

The liver is frequently challenged by surgery-induced metabolic overload, viruses or toxins, which induce the formation of reactive oxygen species. To determine the effect of oxidative stress on liver regeneration and to identify the underlying signaling pathways, we studied liver repair in mice lacking the Nrf2 transcription factor. In these animals, expression of several cytoprotective enzymes was reduced in hepatocytes, resulting in oxidative stress. After partial hepatectomy, liver regeneration was significantly delayed. Using in vitro and in vivo studies, we identified oxidative stress-mediated insulin/insulin-like growth factor resistance as an underlying mechanism. This deficiency impaired the activation of p38 mitogen-activated kinase, Akt kinase and downstream targets after hepatectomy, resulting in enhanced death and delayed proliferation of hepatocytes. Our results reveal novel roles of Nrf2 in the regulation of growth factor signaling and in tissue repair. In addition, they provide new insight into the mechanisms underlying oxidative stress-induced defects in liver regeneration. These findings may provide the basis for the development of new strategies to improve regeneration in patients with acute or chronic liver damage.

Figure 1

Liver regeneration is impaired in Nrf2-deficient mice. (A) Liver sections of Nrf2 knockout mice (ko) and wild-type (wt) littermates 48 h after hepatectomy were stained with hematoxylin/eosin. Arrows indicate lipid droplets in hepatocytes. (B) Proliferation was assessed by BrdU incorporation. Representative sections from injured liver (48 h after hepatectomy) are shown. (C) The percentage of proliferating cells was determined by counting 3–5 independent microscopic fields per liver at × 200 magnification, n (number of mice) >6 per genotype and time point. (D) Cryosections were stained for cleaved caspase-3. The percentage of stained cells 6 h after hepatectomy was determined by counting 3–5 independent microscopic fields ( × 200 magnification, n=4 per genotype). Bars represent mean±s.e.m.; ^*P<0.05; ^**P<0.01.

Figure 2

Reduced mRNA levels of Nrf2 target genes and enhanced oxidative stress in normal and hepatectomized livers of Nrf2 knockout mice. (A) Total cellular RNA (10 μg from pooled livers of at least four mice per time point and genotype) was analyzed by RPA for transcripts encoding Nrf2 target genes. The time after hepatectomy or sham surgery is indicated on top of each lane; 0 h indicates resting liver, which was removed during hepatectomy; 20 μg tRNA served as a negative control and 1000 c.p.m. of the hybridization probes were used as a size marker. Hybridization with a GAPDH riboprobe served as a loading control. Representative autoradiograms of two independent experiments are shown. (B) Total lysates from resting and injured livers 72 h after partial hepatectomy were assayed for GST activity. The formation of glutathione/1-chloro-2,4-dinitrobenzene conjugates was measured spectrophotometrically. Bars represent mean±s.e.m.; ^**P<0.01 (n=6). (C) Hepatocytes were analyzed for GST activity. Bars represent mean±s.e.m.; ^*P<0.05; ^**P<0.01 (n=6). (D) Freshly prepared cryosections (7 μm) from non-injured and injured livers (72 h after hepatectomy) were stained with DEH and analyzed by fluorescence microscopy (n⩾5). (E) Primary hepatocytes from Nrf2 knockout and wild-type mice were treated with H₂DCFH-DA and analyzed by flow cytometry for the levels of intracellular ROS.

Figure 3

Altered JNK and p38 activation but normal AP-1 activity in injured liver of Nrf2 knockout mice. (A, B) Total protein (60 μg) from liver lysates of Nrf2 knockout and wild-type mice (from pooled livers of at least four mice per time point after hepatectomy and genotype) was analyzed by immunoblotting for the levels of phosphorylated and total Erk, p38 and JNK (A), and phosphorylated c-Jun (B). Staining of the membrane with antibodies to GAPDH or Lamin A was used as a loading control. Representative blots from two independent experiments with lysates from different hepatectomy experiments are shown. (C) Radiolabeled oligonucleotides containing AP-1-binding sites were incubated with total protein lysates (20 μg) from normal and hepatectomized livers, and EMSAs were performed. Addition of an excess of non-labeled oligonucleotides (lanes labeled: cold comp.) inhibited mobility shifts, whereas addition of non-labeled oligonucleotides with mutated AP-1-binding sites had no effect (lanes labeled: mut. cold comp.).

Figure 4

Reduced activation of the PI3K/Akt signaling pathway in injured liver of Nrf2 knockout mice. Total protein (60 μg) from liver lysates of Nrf2 knockout and wild-type mice (from pooled livers of at least four mice per time point after hepatectomy and genotype) was analyzed by immunoblotting for the levels of the PI3K p85α subunit, phosphorylated and total Akt, GSK-3β and S6 kinase (A), non-phosphorylated and phosphorylated Bad, GAPDH and Lamin A (B). Representative blots from two to four experiments with lysates from different hepatectomy experiments are shown.

Figure 5

Reduced IGF-1R/IR signaling in injured liver of Nrf2 knockout mice. (A) Liver lysates (60 μg protein) of Nrf2 knockout and wild-type mice at different time points after hepatectomy were analyzed by immunoblotting for the levels of the total IGF-1R and IR, phosphorylated IGF-1R/IR, phosphorylated and total Akt, phosphorylated IRS-1/2 (Tyr608) and total IRS-1, and GAPDH. Representative blots from two to three experiments with lysates from different hepatectomy experiments are shown. (B) Total liver lysates from injured liver (3 h after hepatectomy) of Nrf2-deficient and wild-type mice were subjected to immunoprecipitation using antibodies against IR. Precipitates were analyzed by immunoblotting using antibodies against IR or pIR/IGF-1R.

Figure 6

Oxidative stress impairs insulin-induced Akt activation in hepatocytes. Primary murine hepatocytes from wild-type mice were pretreated for 2 h with 50 or 100 mU/ml GO and subsequently stimulated for 15 min with 100 nM insulin. Activation of IGF-1R/IR signaling was monitored by immunoblotting of 30 μg of total protein with phosphospecific antibodies to IGF-1R/IR and Akt. Staining of the membrane with antibodies against total Akt and GAPDH served as controls. Representative blots from three experiments are shown.

Figure 7

Nrf2 deficiency in hepatocytes causes insulin resistance. (A) Primary hepatocytes from Nrf2 knockout or wild-type mice were serum-starved, and treated for the indicated time with 100 nM insulin. Protein lysates (40 μg) were analyzed by western blotting using phosphospecific antibodies to IGF-1R/IR and Akt, or antibodies to total IR, Akt and GAPDH. Representative blots from four independent experiments are shown. (B) Primary hepatocytes from Nrf2 knockout or wild-type mice were serum-starved, and treated for 15 min with insulin. Lysates were subjected to immunoprecipitation with an IR antibody, and precipitates were analyzed by western blotting using antibodies against IR or pIR/IGF-1R. (C) Alternatively, lysates were subjected to immunoprecipitation with an IRS-1 antibody, and association of IRS-1 with PI3K-p85α was monitored by immunoblotting (upper panel). To ensure equal loading, 30 μg of the lysates used for immunoprecipitation was analyzed by western blotting for the levels of total IRS-1, PI3K-p85α and GAPDH (lower panel). Representative blots from two independent experiments are shown. (D) Liver samples from Nrf2 knockout mice and wild-type control animals were harvested 3 h after hepatectomy. Lysates were immunoprecipitated with an antibody to IRS-1. Precipitates were analyzed by western blotting for the presence of PI3K (p85α subunit). Analysis of IRS-1 levels in the precipitate served as a control. In addition, levels of IRS-1, PI3K (p85α subunit), p-Akt and GAPDH were analyzed in the lysate before immunoprecipitation (input: western blots shown in the lower panel). Representative blots from three experiments are shown.

Figure 8

Schematic representation of Nrf2 action in the injured liver. Nrf2 induces the expression of ROS-detoxifying enzymes, which prevents ROS accumulation. Its deficiency causes oxidative stress, resulting in reduced tyrosine phosphorylation of IRS-1 and -2, most likely by ROS-activated JNK and possibly other Ser/Thr IRS kinases. Since tyrosine phosphorylation of IRS-1 is required for PI3K activation, phosphorylation of Akt and downstream targets is reduced.