Metformin suppresses intestinal polyp growth in ApcMin/+ mice

Abstract

Metformin is a biguanide derivative that is widely used in the treatment of diabetes mellitus. One of the pharmacological targets of metformin is adenosine monophosphate-activated protein kinase (AMPK). We investigated the effect of metformin on the suppression of intestinal polyp formation in Apc(Min/+) mice. Administration of metformin (250 mg/kg) did not reduce the total number of intestinal polyp formations, but significantly reduced the number of intestinal polyp formations larger than 2 mm in diameter in Apc(Min/+) mice. To examine the indirect effect of metformin, the index of insulin resistance and serum lipid levels in Apc(Min/+) mice were assessed. These factors were not significantly attenuated by the treatment with metformin, indicating that the suppression of polyp growth is not due to the indirect drug action. The levels of tumor cell proliferation as determined by 5-bromodeoxyuridine and proliferating cell nuclear antigen immunohistochemical staining, and apoptosis, via transferase deoxytidyl uridine end labeling staining, in the polyps of metformin-treated mice were not significantly different in comparison to those of control mice. Gene expression of cyclin D1 and c-myc in intestinal polyps were also not significantly different between those two groups. In contrast, metformin activated AMPK in the intestinal polyps, resulting in the inhibition of the activation of mammalian target of rapamycin, which play important roles in the protein synthesis machinery. Metformin suppressed the polyp growth in Apc(Min/+) mice, suggesting that it may be a novel candidate as a chemopreventive agent for colorectal cancer.

Figure 1

(a) Experimental design for this study. Nine‐week‐old Apc ^Min/+ mice (n = 32; 20 for histological analysis, 12 for protein or RNA expression analysis) were divided into two groups: with or without metformin (250 mg/kg) treatment for 10 weeks. (b) Mean weights of Apc ^Min/+ mice in the control and metformin‐treated groups. Bodyweight was measured once a week for 10 weeks. No statistically significant differences in bodyweight between metformin‐treated and control groups were observed.

Figure 2

Effect of metformin on the number and the size distribution of intestinal polyps in Apc ^Min/+ mice. Apc ^Min/+ mice were fed either a basal diet or a diet containing metformin (250 mg/kg) for 10 weeks. (a) The number of polyps per mouse in each size class is given as a mean value. (b) Comparison of polyp size below 2 mm and over 2 mm in diameter. Bars represent standard error. *^a P = 0.034; *^b P = 0.032; *^c P = 0.019.

Figure 3

Immunohistochemistry for β‐catenin, cell proliferation assay using 5‐bromodeoxyuridine (BrdU) and proliferating cell nuclear antigen (PCNA), apoptosis assay, and gene expression analyses of cyclin D1 and c‐myc in the tumor cells. (a) HE staining and immunohistochemical staining for β‐catenin, BrdU and PCNA in intestinal tumors from Apc ^Min/+ mice fed a basal diet or treated with metformin. Transferase deoxytidyl uridine end labeling staining in tumors are also demonstrated. (b,c,d) Calculations for the BrdU, PCNA and apoptotic indices are detailed in Materials and Methods section. No significant difference was observed in each index category between metformin‐treated and control groups. Gene expression of (e) cyclin D1 and (f) c‐myc in intestinal polyps using real time reverse transcription polymerase chain reaction analysis are also shown. β‐Actin was used as the internal control. No significant difference was observed in these gene expression levels between metformin‐treated and control groups. Bars represent standard error.

Figure 4

Plasma levels of insulin, triglyceride, cholesterol and glucose. Apc ^Min/+ mice were fed a basal diet (control) or a diet containing metformin (250 mg/kg). Plasma and blood were obtained after 12 h fasting. (a) Cholesterol, (b) triglyceride, (c) insulin, (d) blood glucose and (e) homeostasis model assessment of insulin resistance (HOMA‐IR) levels are shown. HOMA‐IR was used to calculate insulin resistance using the following formula: HOMA‐IR = fasting plasma immunoreactive insulin (IRI; µU/mL) × fasting plasma glucose (FBG; mg/dL)/405. No statistical significance was observed in the plasma levels of cholesterol, triglyceride and HOMA‐IR between metformin‐treated and control groups. Bars represent standard error.

Figure 5

Treatment with metformin activates adenosine monophosphate‐activated protein kinase (AMPK) and downregulates the mammalian target of rapamycin (mTOR)/S6K pathway. Apc ^Min/+ mice were treated with a basal diet (control, n = 6) and a diet containing metformin (250 mg/kg, n = 6) for 10 weeks. Western blotting was performed using intestinal tumors (>2 mm in diameter) from both groups; five polyps from each mice were analyzed (a total of 60 polyps were assessed). Representative western blotting images are demonstrated. (a) Immunoblotting using antibodies against phosphorylated AMPK and total AMPK. The relative activity of phosphorylated AMPK against total AMPK is demonstrated. (b) Phospho‐mTOR, mTOR and relative activity of mTOR. (c) Phopho‐S6K/S6. (d) Phospho‐p70 protein S6 (S6)/total S6. (e) Western blotting analysis of the expression of AMPK in small intestine, colon and liver from Apc ^Min/+ mice. glyceraldehyde 3‐phosphate dehydrogenase (GAPDH) is shown as a loading control. Bars represent standard error. *^a P = 0.0016; *^b P = 0.0009; *^c P = 0.0103; *^d P = 0.0054.