Growth hormone upregulates the pro-tumorigenic galectin 1 in mouse liver

Abstract

Transgenic mice overexpressing growth hormone (GH) spontaneously develop liver tumors, including hepatocellular carcinoma (HCC), within a year. The preneoplastic liver pathology in these mice recapitulates that observed in humans at high risk of developing hepatic cancer. Although increased expression of galectin 1 (GAL1) in liver tissue is associated with HCC aggressiveness, a link between this glycan-binding protein and hormone-related tumor development has not yet been explored. In this study, we investigated GAL1 expression during liver tumor progression in mice continuously exposed to high levels of GH. GAL1 expression was determined by Western blotting, RT-qPCR and immunohistochemistry in the liver of transgenic mice overexpressing GH. Animals of representative ages at different stages of liver pathology were studied. GAL1 expression was upregulated in the liver of GH-transgenic mice. This effect was observed at early ages, when animals displayed no signs of liver disease or minimal histopathological alterations and was also detected in young adults with preneoplastic liver pathology. Remarkably, GAL1 upregulation was sustained during aging and its expression was particularly enhanced in liver tumors. GH also induced hepatic GAL1 expression in mice that were treated with this hormone for a short period. Moreover, GH triggered a rapid increment in GAL1 protein expression in human HCC cells, denoting a direct effect of the hormone on hepatocytes. Therefore, our results indicate that GH upregulates GAL1 expression in mouse liver, which may have critical implications in tumorigenesis. These findings suggest that this lectin could be implicated in hormone-driven liver carcinogenesis.

Keywords: galectin 1; growth hormone; hepatocellular carcinoma; liver cancer.

Figure 1

GAL1 expression is increased in the liver of young GH-transgenic mice. Liver extracts of control (C) and GH-transgenic (T) mice of three representative ages: 2 week-old (2 w), 4 week-old (4 w) and 9 week-old (9 w, considered young adults and used as reference) were evaluated for GAL1 protein expression by Western blotting and densitometric analysis in females and males (panels A and B, respectively). Protein loading control was performed by Coomassie blue staining (CBS) of PVDF membranes. Liver Gal1 mRNA levels relative to Cyclophilin A were determined by RT-qPCR analysis in females and males (panels C and D, respectively). Data are the mean ± s.e.m. of the indicated number of samples per group (n), each one representing a different animal. Different letters denote significant differences by age (P < 0.05), small letters correspond to normal mice and capital letters to transgenic animals; asterisks indicate differences between genotypes analyzed by two-way ANOVA followed by Bonferroni post-test (*P < 0.05, **P < 0.01, ***P < 0.001). GAL1 expression and cellular localization were evaluated by immunohistochemistry in the liver of 9-week-old animals (E). Three independent experiments were performed with reproducible results and representative microphotographs are shown for liver sections of female mice. Similar results were obtained for males (Supplementary Fig. 2A). Negative controls were performed in parallel in all sections by incubation without primary antibody (Supplementary Fig. 2B).

Figure 2

GAL1 is upregulated in the liver and in hepatic tumors of old GH-transgenic mice. GAL1 expression was assessed in the liver of control (C) and GH-transgenic (T) mice of 10–13 months old by Western blotting and densitometric analysis (A) and RT-qPCR assay (B). Protein loading control was performed by Coomassie blue staining (CBS) of PVDF membranes. Tumoral (Tu) and non-tumoral (NTu) zones for the same GH-transgenic animal were processed in parallel and GAL1 protein content (C) and mRNA levels relative to Cyclophilin A (D) were determined. Data are the mean ± s.e.m. of the indicated number of samples per group (n), each one representing a different animal. The same letter denotes no significant difference between sexes, small letters correspond to normal mice and capital letters to transgenic animals; the symbol ^# indicates differences between genotypes analyzed by two-way ANOVA followed by Bonferroni post-test (^###P<0.001). Asterisks indicate differences analyzed by Student’s t-test (*P < 0.05, ***P < 0.001). Liver sections from control and transgenic mice were processed for immunohistochemistry to analyze GAL1 expression and localization. Tumoral (Tu) and non-tumoral (NTu) tissues were processed in parallel (E). Three independent experiments were performed with reproducible results and representative microphotographs are shown for liver sections of female mice. Similar results were obtained for males (data not shown). Negative controls were performed by omitting primary antibody (Supplementary Fig. 2C).

Figure 3

Prolonged exposure to GH does not increase GAL1 expression in the kidney, muscle or heart of GH-transgenic mice. Kidney (A), heart (B) and skeletal muscle (C) extracts from adult control (C) and GH-transgenic (T) females were evaluated for GAL1 expression by Western blotting and analyzed by densitometry. Protein loading control was performed by Coomassie blue staining (CBS) of PVDF membranes. Data are the mean ± s.e.m. of the indicated number of samples per group (n), each one representing a different animal. No significant differences were obtained when analyzed by Student’s t test.

Figure 4

GAL1 expression is induced in the liver of GH-treated Swiss-Webster mice. Liver extracts of female and male mice treated for 5 days with GH (1 mg/kg of body weight per day) by continuous infusion with an osmotic minipump (GH) or their respective controls (C) were evaluated for GAL1 protein expression by Western blotting and analyzed by densitometry. Protein loading control was performed by β-actin immunoblotting. Data are the mean ± s.e.m. of the indicated number of samples per group (n), each one representing a different animal. Asterisks indicate differences induced by GH stimulus analyzed by Student’s t-test (*P < 0.05).

Figure 5

GAL1 expression is induced by GH in human HCC cells in vitro. Western blotting and densitometric analysis showing GAL1 protein expression in HepG2 and HuH-7 HCC cells treated with serum-free media containing GH (1 µg/ml) or vehicle (C, control cells used as reference) for 30 min (A). Protein loading control was performed by β-actin immunoblotting. GAL1 mRNA levels relative to RPL13A were determined by RT-qPCR analysis in cells cultured in the presence of GH (1 µg/ml) (HepG2-GH and HuH-7-GH) and in control cells (HepG2-C and HuH-7-C) for 15 min or 3 h (B). Results are expressed as fold change versus values measured for untreated cells at the beginning of the experiment (t0). Data are the mean ± s.e.m. of the indicated number of different experiments (n). Asterisks indicate differences induced by GH-stimulus analyzed by Student’s t-test (*P < 0.05) (A). No significant differences were found by two-way ANOVA followed by Bonferroni post-test analysis (B).