Maternal high-fat diet promotes body length increases and insulin insensitivity in second-generation mice

Abstract

Maternal obesity and diet consumption during pregnancy have been linked to offspring adiposity, cardiovascular disease, and impaired glucose metabolism. Furthermore, nutrition during development is clearly linked to somatic growth. However, few studies have examined whether phenotypes derived from maternal high-fat diet exposure can be passed to subsequent generations and by what mechanisms this may occur. Here we report the novel finding of a significant body length increase that persisted across at least two generations of offspring in response to maternal high-fat diet exposure. This phenotype is not attributable to altered intrauterine conditions or maternal feeding behavior because maternal and paternal lineages were able to transmit the effect, supporting a true epigenetic manner of inheritance. We also detected a heritable feature of reduced insulin sensitivity across two generations. Alterations in the GH secretagogue receptor (GHSR), the GHSR transcriptional repressor AF5q31, plasma IGF-I concentrations, and IGF-binding protein-3 (IGFBP3) suggest a contribution of the GH axis. These studies provide evidence that the heritability of body length and glucose homeostasis are modulated by maternal diet across multiple generations, providing a mechanism where length can increase rapidly in concert with caloric availability.

Figure 1

Breeding scheme for first- and second-generation offspring. C57BL/6:129 dams (top lines) from the chow (mCh) group consumed a 12% fat house chow diet from 4 wk before conception and throughout pregnancy and the nursing period, whereas dams from the high-fat (mHF) group consumed a 45% fat high-fat diet during this time. All first-generation offspring were weaned onto chow. Approximately two thirds of first-generation mice were fed a high-fat diet for 8 wk starting at 10 wk of age to examine physiological responses, but these mice were not used to produce the second generation. The remaining mice (naive to high-fat diet as adults) were bred to produce second-generation offspring, one where the maternal [2HF(M)], the paternal [2HF(P)], or both lineages [2HF(MP)] had been exposed to maternal high-fat diet. All second-generation mice consumed a high-fat diet as adults, commencing at 10 wk.

Figure 2

Exposure to a maternal high-fat diet results in increased body weight and length in first-generation offspring with increased body length transmitted to the second generation through maternal or paternal lineages. A, First-generation E17 embryos exhibited increased body length after exposure to maternal high-fat diet (1HF, 45% fat) relative to those exposed to chow (1Ch, 12% fat) *, P = 0.05; #, P = 0.018, main effect of diet. B, These body length differences persisted into adulthood in 1HF male and female offspring. #, P = 0.041, main effect of diet. C, Body length increases persisted in second-generation offspring from the maternal lineage [2HF(M); **, P = 0.0016; ***, P = 0.0002], the paternal lineage [2HF(P); **, P < 0.005], and in the maternal/paternal lineage [2HF(MP); ***, P < 0.0001] compared with second-generation controls (2Ch). 1HF male (D) and female (E) offspring exhibit increased body weights (**, P < 0.05; ***, P < 0.0005) relative to 1Ch controls. This increase does not persist in 2HF males (F) and females (G). Hatched bars indicate time of high-fat diet consumption. Data are mean ± sem.

Figure 3

Reduced insulin sensitivity during house chow diet consumption is inherited by the second generation from maternal or paternal first-generation high-fat offspring. GTTs are not significantly different between chow (1Ch) and high-fat diet (1HF) lines in either first-generation male (A) or female (B) adult offspring. Second-generation male (C) and female (D) offspring from HF maternal [2HF(M)], paternal [2HF(P)], or maternal and paternal lineages [2HF(MP)] have normal glucose tolerance relative to controls (2Ch). ITTs reveal impaired insulin sensitivity in 1HF male (E) but not female (F) offspring. *, P = 0.041. Insulin sensitivity is significantly impaired in 2HF(P) males (G) (*, P < 0.05) and 2HF(P) female (H) offspring (30 min; *, P < 0.05; **, P < 0.005). ITT points are normalized to baseline glucose levels at 0 min. Data are mean ± sem.

Figure 4

Sex-dependent programming of the GH pathway through changes in IGF-I, IGFBP-3, GHSR, and growth factors by maternal high-fat diet. A, Plasma insulin is not changed in adult first-generation (1HF, black bars) or second-generation [2HF(M, P, and MP), black bars] males or females from the high-fat lineage relative to controls [Chow, white bars, combined from first-generation (1Ch) and second-generation (2Ch) mice]. B, Adult plasma leptin levels are reduced in 1HF (**, P = 0.001; ***, P = 0.0002) and 2HF offspring from the maternal [2HF(M); *, P = 0.019; **, P = 0.0035], paternal [2HF(P); *, P = 0.017; **, P = 0.0025], and maternal/paternal lines [2HF(MP); *, P = 0.0065; **, P = 0.0006] relative to controls. C, IGF-I protein levels are significantly increased in 1HF, 2HF(P), and 2HF(MP) females. *, P < 0.05; **, P < 0.005. D, IGFBP-3 transcript levels are significantly increased at postnatal d 1 (PN1) in 1HF male and female offspring. #, P = 0.021, overall effect of diet. E, Trend toward increased GHSR expression in 1HF females in PN1 brain homogenates by quantitative RT-PCR. F, Trend toward increased GHSR expression in 2HF(MP) adult females by quantitative RT-PCR in micropunches from the arcuate nucleus of the hypothalamus. G, Significant reduction in transcript levels of a GHSR transcriptional repressor (AF5q31) in 1HF female PN1 brains. *, P = 0.05. H, Significantly altered growth factor expression from a PCR array illustrate sex differences in response to maternal high-fat diet exposure in 1HF male and female PN1 brains. *, P < 0.05; **, P < 0.005. Data are presented as fold change in 1HF males and females relative to their respective 1Ch controls. All data are mean ± sem.

Figure 5

Second-generation male and female adult offspring exhibit decreased methylation at the GHSR promoter. Percent methylation assayed by bisulfite conversion and pyrosequencing at 14 cytosines of the GHSR promoter-associated CpG island in microdissected arcuate nuclei in second-generation adult male (A) and female (B) offspring is significantly altered between the chow (2Ch) or high-fat groups from the maternal [2HF(M), black and white hatched bars], paternal [2HF(P); gray and black hatched bars], or maternal/paternal lineages [2HF(MP); black bars]. *, P < 0.05. Data are mean ± sem.