Regulation of liver regeneration and hepatocarcinogenesis by suppressor of cytokine signaling 3

Abstract

Suppressor of cytokine signaling 3 (SOCS3) down-regulates several signaling pathways in multiple cell types, and previous data suggest that SOCS3 may shut off cytokine activation at the early stages of liver regeneration (Campbell, J.S., L. Prichard, F. Schaper, J. Schmitz, A. Stephenson-Famy, M.E. Rosenfeld, G.M. Argast, P.C. Heinrich, and N. Fausto. 2001.J. Clin. Invest. 107:1285-1292). We developed Socs3 hepatocyte-specific knockout (Socs3 h-KO) mice to directly study the role of SOCS3 during liver regeneration after a two-thirds partial hepatectomy (PH). Socs3 h-KO mice demonstrate marked enhancement of DNA replication and liver weight restoration after PH in comparison with littermate controls. Without SOCS3, signal transducer and activator of transcription 3 (STAT3) phosphorylation is prolonged, and activation of the mitogenic extracellular signal-regulated kinase 1/2 (ERK1/2) is enhanced after PH. In vitro, we show that SOCS3 deficiency enhances hepatocyte proliferation in association with enhanced STAT3 and ERK activation after epidermal growth factor or interleukin 6 stimulation. Microarray analyses show that SOCS3 modulates a distinct set of genes, which fall into diverse physiological categories, after PH. Using a model of chemical-induced carcinogenesis, we found that Socs3 h-KO mice develop hepatocellular carcinoma at an accelerated rate. By acting on cytokines and multiple proliferative pathways, SOCS3 modulates both physiological and neoplastic proliferative processes in the liver and may act as a tumor suppressor.

Figure 1.

Liver regeneration is enhanced in Socs3 h-KO mice. (A) Staining for BrdU indicates increased DNA synthesis in Socs3 h-KO mice. (B) Increased mitotic figures in hepatocytes lacking Socs3. (C) More rapid return to preoperative liver weight in the absence of Socs3. (D) Increased expression of cyclin A and p107 in Socs3 h-KO mice after PH, demonstrated by immunoblotting. β-Actin and β-tubulin are shown as loading controls. (E) Confirmation of lack of Socs3 induction after PH in Socs3 h-KO mice by Northern blotting, with cyclophilin as a loading control. (F) Northern blotting demonstrates no compensatory up-regulation of Socs1 in the absence of Socs3. Data shown are representative of three to six mice per genotype per time point and are presented as mean ± SEM. *, P < 0.05.

Figure 2.

Activation of STAT3 and ERK1/2 are enhanced in Socs3 h-KO mice. (A) Serum IL-6 levels after PH are unchanged in Socs3 h-KO mice (ELISA). (B) Lack of SOCS3 in hepatocytes leads to prolonged phosphorylation of STAT3 (Y705) after PH, demonstrated by immunoblotting. (C) Northern blot analysis indicates prolonged expression of Cd14 Socs3 h-KO mice. (D) Saa2 expression is also prolonged in Socs3 h-KO mice. (E) Socs3 deficiency leads to enhanced phosphorylation of ERK1/2 after PH, demonstrated by immunoblotting. All demonstrated data are representative of three to six mice per genotype per time point and are presented as mean ± SEM. *, P < 0.05.

Figure 3.

SOCS3 down-regulates hepatocyte proliferation and growth factor signaling in vitro. (A) [³H]Thymidine incorporation by primary hepatocytes in the absence of growth factors 36 h after isolation and initiation of cell culture. (B) Hepatocyte proliferation in response to 20 ng/ml EGF. (C) Pi-STAT3^Y705 is enhanced in response to EGF in Socs3-deficient hepatocytes. (D) Similar enhancement of Pi-ERK1/2 in response to EGF in Socs3 KO hepatocytes. All data are representative of at least three separate experiments, with four to six replicates in each ³H experiment, and are presented as mean ± SEM. *, P < 0.05.

Figure 4.

Inhibition of JAK or MEK in culture inhibits hepatocyte proliferation. (A) AG490 inhibits tyrosine phosphorylation of STAT3. (B) AG490 inhibits hepatocyte proliferation ([³H]thymidine assay). (C) U0126 inhibits ERK activation in response to EGF. (D) U0126 inhibits hepatocyte proliferation ([³H]thymidine assay). Data are representative of three independent experiments and are presented as mean ± SEM. *, P < 0.05.

Figure 5.

Activation of STAT3, ERK1/2, and Akt are prolonged in Socs3 KO hepatocytes after IL-6 treatment. (A) Pi-STAT3^Y705 is prolonged in response to IL-6 in Socs3 KO hepatocytes. (B) Activation of ERK1/2 is enhanced in response to IL-6 in Socs3 KO hepatocytes. (C) Prolonged activation of Pi-S6 ^S240/244 ribosomal protein in Socs3 KO versus littermate hepatocytes. β-Actin is shown as a loading control. (D) No difference in activation of Akt in Socs3 versus control hepatocytes. Data are representative of three to five independent experiments.

Figure 6.

Oligonucleotide microarrays demonstrate broad and diverse effects of SOCS3 on gene expression after PH. (A) Heatmap demonstrating a global picture of differential gene expression in Socs3 h-KO mice versus control littermates at 18 h after PH. (B) NIH DAVID analysis of a differentially regulated gene list reveals that multiple pathways are differentially regulated during regeneration in Socs3 h-KO mice. (C) Confirmed up-regulation of c-jun, Haptoglobin, Hif1α, IκBα, Pdgf-c, and Pdgfrα in Socs3 h-KO mice versus control littermates at 18 h after PH, demonstrated by real-time RT-PCR. Data are presented as fold change using the values from nonoperated mice as the denominator and are presented as mean ± SEM. *, P < 0.05; ***, P < 0.001.

Figure 7.

Analysis of early response to DEN injection in Socs3 h-KO mice versus littermates. Mice were injected with 100 μg DEN per gram of body weight, and liver tissues were harvested 24 and 48 h later. (A) DEN injection induces caspase 3 activity at 48 h but similarly in Socs3 h-KO (hatched bars) and littermate (black bars) mice. (B) Serum IL-6 levels are elevated in Socs3 h-KO versus littermate mice 24 h after DEN injection. (C) Enhanced activation of STAT3 after DEN injection in Socs3 h-KO mice. Data presented are representative of six mice per genotype per time point and are presented as mean ± SEM. *, P < 0.05.