Dose-dependent effects of calorie restriction on gene expression, metabolism, and tumor progression are partially mediated by insulin-like growth factor-1

Abstract

The prevalence of obesity, an established risk and progression factor for breast and many other cancer types, remains very high in the United States and throughout the world. Calorie restriction (CR), a reduced-calorie dietary regimen typically involving a 20-40% reduction in calorie consumption, prevents or reverses obesity, and inhibits mammary and other types of cancer in multiple tumor model systems. Unfortunately, the mechanisms underlying the tumor inhibitory effects of CR are poorly understood, and a better understanding of these mechanisms may lead to new intervention targets and strategies for preventing or controlling cancer. We have previously shown that the anticancer effects of CR are associated with decreased systemic levels of insulin-like growth factor-1 (IGF-1), the primary source of which is liver. We have also reported that CR strongly suppresses tumor development and growth in multiple mammary cancer models. To identify CR-responsive genes and pathways, and to further characterize the role of IGF-1 as a mediator of the anticancer effects of CR, we assessed hepatic and mammary gland gene expression, hormone levels and growth of orthotopically transplanted mammary tumors in control and CR mice with and without exogenous IGF-1. C57BL/6 mice were fed either control AIN-76A diet ad libitum (AL), subjected to 20%, 30%, or 40% CR plus placebo timed-release pellets, or subjected to 30% or 40% CR plus timed-release pellets delivering murine IGF-1 (mIGF-1, 20 μg/day). Compared with AL-fed controls, body weights were decreased 14.3% in the 20% CR group, 18.5% in the 30% CR group, and 38% in the 40% CR group; IGF-1 infusion had no effect on body weight. Hepatic transcriptome analyses indicated that compared with 20% CR, 30% CR significantly modulated more than twice the number of genes and 40% CR more than seven times the number of genes. Many of the genes specific to the 40% CR regimen were hepatic stress-related and/or DNA damage-related genes. Exogenous IGF-1 rescued the hepatic expression of several metabolic genes and pathways affected by CR. Exogenous IGF-1 also rescued the expression of several metabolism- and cancer-related genes affected by CR in the mammary gland. Furthermore, exogenous IGF-1 partially reversed the mammary tumor inhibitory effects of 30% CR. We conclude that several genes and pathways, particularly those associated with macronutrient and steroid hormone metabolism, are associated with the anticancer effects of CR, and that reduced IGF-1 levels can account, at least in part, for many of the effects of CR on gene expression and mammary tumor burden.

Keywords: Cancer biology; cancer prevention; carcinogenesis; nutrition.

Figure 1

Regulation of gene expression by calorie restriction (CR). (A) Venn diagram comparing 20%, 30%, and 40% CR. (B) Gene ontology of genes common to all. (C) Expression of Fabp5. (D) Expression of Acot1. (E) Expression of Gck. (F) Gene ontology of genes unique to 40% CR. (G) Expression of Cdkn1. Data shown are mean ± SEM, n = 5 per group. Significance (P < 0.05) between groups is denoted by different letters.

Figure 2

Regulation of gene expression by 30% CR and 30% CR + IGF-1. (A) Venn diagram comparing number of genes expressed by 30% CR and 30% CR + IGF-1 compared to control. (B) Gene ontology of genes specific to 30% CR + IGF-1. (C) Gene ontology of genes common to 30% CR and 30% CR + IGF-1. (D) Expression of Serpina12. Data shown are mean ± SEM. Significance (P < 0.05) between groups is denoted by different letters. CR, calorie restriction; IGF-1, insulin-like growth factor-1.

Figure 3

Regulation of gene expression by 40% CR and 40% CR + IGF-1. (A) Venn diagram comparing number of genes differentially expressed for 40% CR and 40% CR + IGF-1, compared with control. (B) Gene ontology of genes uniquely differentially expressed between 40% CR and control, but not 40% CR + IGF-1 and control. (C) Expression of Hnf4α. (D) Gene ontology of genes unique to 40% CR + IGF-1. (E) Expression of Scd1. Data shown are mean ± SEM. Significance (P < 0.05) between groups is denoted by different letters.

Figure 4

Gene expression changes between 30% CR and 30% CR + IGF-1 in the mammary fat pad. (A) Expression of Dusp14. (B) Expression of Fbp1. (C) Expression of Irs2. Data shown are mean ± SEM. Significance (P < 0.05) between groups is denoted by different letters.

Figure 5

Effects of 30% CR and 30% CR + IGF-1 on mammary tumor growth. (A) Serum leptin levels. (B) Serum total IGF-1 levels. (C) Tumor weight at the end of the study. Data shown are mean ± SEM for serum hormones. Significance (P < 0.05) between groups is denoted by different letters. (D) CR tumors showed decreased IGF-1R activation (pIGF-1R) whereas control and 30% CR + IGF-1 showed increased IGF-1R activation. There were no changes in total IGF-1R, pAkt, total Akt, CD31, and Ki67 between the groups. Scale bar = 100 μm.