Emerging evidence suggests that paternal obesity plays an important role in offspring health. Our previous work using a rodent model of diet-induced paternal obesity showed that female offspring from high-fat diet (HFD)-fed fathers develop glucose intolerance due to impairment of pancreatic insulin secretion. Here, we focused on the health outcomes of male offspring from HFD-fed fathers. Male Sprague-Dawley rats (3 wk old) were fed control (CD-F0) or HFD (HFD-F0) for 12 wk before mating with control-fed females. Male offspring were fed control diets for up to 8 wk or 6 mo. Although male offspring from HFD-F0 did not develop any obvious glucose metabolism defects in this study, surprisingly, a growth deficit phenotype was observed from birth to 6 mo of age. Male offspring from HFD-F0 had reduced birth weight compared with CD-F0, followed by reduced postweaning growth from 9 wk of age. This resulted in 10% reduction in body weight at 6 mo with significantly smaller fat pads and skeletal muscles. Reduced circulating levels of growth hormone (GH) and IGF-I were detected at 8 wk and 6 mo, respectively. Expression of adipogenesis markers was decreased in adipose tissue of HFD-F0 offspring at 8 wk and 6 mo, and expression of growth markers was decreased in muscle of HFD-F0 offspring at 8 wk. We propose that the reduced GH secretion at 8 wk of age altered the growth of male offspring from HFD-F0, resulting in smaller animals from 9 wk to 6 mo of age. Furthermore, increased muscle triglyceride content and expression of lipogenic genes were observed in HFD-F0 offspring, potentially increasing their metabolic risk.
it is well recognized that obesity is a worldwide epidemic, but it has only recently become evident that its incidence is growing most in young adults, with 30.3% of 20- to 39-yr-olds being obese in the US and 36.4% of 18- to 24-yr-olds overweight or obese in Australia (3, 59). This is of particular concern because it is within this age group that most pregnancies are conceived, and parental obesity has been shown to be a major risk factor for offspring metabolic disease (44, 60). It is known that a propensity to gain weight has a strong familial component. Although genome-wide association studies have shown that a number of gene variants are associated with increased BMI, no single gene or multiple gene effect discovered to this date sufficiently accounts for the rise in obesity that has occurred in the last two decades (2). Childhood obesity and metabolic complications such as type 2 diabetes have been associated with obesity in either parent, and obese children tend to become obese adults (30, 46). This forms an intergenerational cycle that promotes the obesity epidemic. This phenomenon has been termed “programming of obesity” and has been incorporated into the Developmental Origins of Health and Disease hypothesis (5, 58). This hypothesis is supported by evidence demonstrating that insults during critical periods of development are able to permanently alter an offspring in ways that affect their predisposition to disease. Monozygotic twin studies have shown that such insults could lead to developmental programming of metabolic defects independently of the offspring genetic background (8).
It is now well established that maternal obesity promotes metabolic complications in offspring such as greater body mass index (BMI), hyperlipidemia, and hyperglycemia as well as greater incidence of cardiovascular diseases (44, 58). In recent years, the focus has moved toward the role of paternal obesity in offspring metabolic programming. In humans, several epidemiological studies have shown that a father’s BMI or body weight is independently associated with offspring BMI (46, 58). Some studies even reveal a stronger father-son BMI association than mother-son (18, 47). Paternal adiposity has also been associated with increased adiposity in offspring (10, 13, 15) as well as insulin resistance incidence (26, 57). Thus, such data provide evidence of transmission of metabolic disease from father to offspring.
In human studies, it is often difficult to discriminate between the role of paternal preconception metabolic status and the influence of a shared unhealthy lifestyle between parents and offspring. Furthermore, it is rare that only one parent is obese or overweight; thus it is difficult in the human situation to study paternal obesity in isolation (32). To this end, we developed a model of diet-induced paternal obesity in the rat and were the first to show a direct impact of paternal obesity; female offspring from high-fat diet (HFD)-fed fathers developed glucose intolerance from 6 wk of age (42), which is linked to a defect in insulin secretion by pancreatic islets in response to glucose. Furthermore, we identified common pathways in adipose tissue and pancreatic islets impacted by paternal obesity in female offspring, suggesting that the programming event occurs very early during development (43). Effects of paternal obesity have also been studied in a model of obese but not diabetic mouse, where offspring developed glucose intolerance and females showed increased body weight and adiposity (20).
Human studies showed clearly that paternal BMI can influence offspring developmental trajectories. But whether these factors can be attributed to direct influence of paternal obesity or shared nutritional environment and lifestyle is not known. It has been demonstrated that an aberrant growth trajectory of an offspring is associated with later development of metabolic disease (5, 24). Offspring from undernourished mothers, presenting with small birth weight and postnatal catchup growth, went on to develop greater adiposity and increased risk for metabolic and cardiovascular disease later in life (6, 45, 58). Our group showed a reduction in birth weight in the female offspring from obese fathers (42). In this study, we monitored the growth trajectory of male offspring from obese fathers and looked at both metabolic and anthropometric parameters. We followed offspring body weight from birth to 6 mo of age and measured growth endocrine regulator secretion, as well as expression of muscle growth and adipogenesis markers, to assess the effect of paternal obesity on offspring growth and adiposity development.
MATERIALS AND METHODS
All animal procedures were approved by the Animal Experimentation Ethics Committee of the University of New South Wales (ACEC No. 11/82B). All animals were housed under controlled conditions under a 12:12-h light-dark cycle.
Three-week-old F0 male Sprague-Dawley rats (Animal Research Centre, Perth, Australia) were split into two groups of equal average body weight (n = 14/14). Control rats were fed a control chow diet (CD) (11 kJ/g, 13% fat, 22% protein, and 65% carbohydrate as precent energy; Gordon’s Stockfeeds, Sydney, Australia). The high-fat diet (HFD) group was offered control chow and two commercial high-fat diets (SF03-020: 20 kJ/g, 43% fat, 17% protein, and 40% carbohydrate; SF01-025: 18.3 kJ/g, 44% fat, 17% protein, 39% carbohydrate; Specialty Feeds, Glen Forest, Australia) because variety of diet has been shown to promote food intake in rat model of diet-induced obesity (33, 50). One week before mating, a fasted blood sample was collected from the tail vein of F0 males to allow assessment of blood glucose, plasma insulin, leptin, and triglyceride concentrations (assays explained below). The F0 males were mated after 13–14 wk of diet with 12-wk-old females (ARC) consuming control chow. Females did not differ in body weight, fasting blood glucose, or plasma insulin concentration between F0 groups, as measured 1 wk before mating (Table 1).
During mating, one male and one female were housed together with ad libitum access to control chow during the daylight period only and then returned to their home cages to continue their assigned diet. Females were fed control chow throughout pregnancy and lactation. F0 males and females were culled after overnight fasting shortly after litters were weaned.
From 28 mated pairs, 16 resulted in a successful pregnancy. Eight litters per group from eight different male and female F0 pairings were born. Offspring were weighed on day 1 after birth and every 3 days after birth. To control for postnatal nutrition, litter size was reduced to 12 pups where necessary. No litter was fewer than 12 pups, and no cross-fostering was performed. Male offspring were weaned at 21 days of age onto control chow and housed four per cage until 8 wk of age, then two per cage. Offspring body weight was measured twice weekly. The average 24-h food intake was calculated weekly.
Depending on litter size, one to three male offspring per litter were culled at 8 or 27 wk after overnight fasting. Blood and tissues were collected the morning after, between 9 AM and 1 PM, with groups randomized across this time period to account for differences in fasted state and diurnal hormone secretion. Naso-anal length was measured on anesthetized animals. Abdominal circumference was measured at the widest part of the abdomen when rats were placed in ventral position. After anesthesia, blood was collected by cardiac puncture into heparin-coated tubes, and then the animals were decapitated. Plasma samples, obtained after whole blood centrifugation, were aliquoted and stored at −20°C. Liver, muscle, and adipose tissue samples were collected, snap-frozen in liquid nitrogen, and then stored at −80°C.
Glucose and insulin tolerance tests.
F0 fathers and offspring underwent glucose tolerance test (GTT) 2 wk before cull and insulin tolerance test (ITT) 1 wk before cull. One representative male per litter per time point was chosen randomly to undergo GTT, followed by ITT. GTT was performed following an overnight fast (15 h). Two grams of glucose/kg body weight (50% wt/vol glucose injectable solution; Phebra, Sydney, Australia) was administered intraperitoneally, and blood glucose concentration was measured at 0, 15, 30, 45, 60, 90, and 120 min using an Accu-check Performa glucose meter (Roche Diagnostics, Sydney, Australia). Blood samples were collected on heparin for insulin measurement at 0, 15, 30, 60, and 120 min. Plasma samples were stored at −20°C. ITT was performed 6 h after food removal. One IU/kg body weight of insulin (Actrapid; NovoNordisk) was administered intraperitoneally and blood glucose concentration measured at 0, 15, 30, 45, 60, 90, and 120 min.
Plasma hormones and triglyceride assays.
Plasma leptin and insulin concentrations were analyzed using commercially available radioimmunoassays according to the manufacturer’s instructions (Merck Millipore, Billerica, MA) and counted on a WIZARD2 Automatic Gamma Counter (PerkinElmer, Melbourne, Australia).
Triglycerides were measured in muscle and liver samples after lipid extraction from ground tissues using the Folch method (17). Tissue and plasma triglycerides were measured colorimetrically (490 nm; iMark Microplate Absorbance Reader; Bio-Rad, Sydney, Australia) using a commercially available triglyceride reagent (GPO-PAP; Roche Diagnostics) and glycerol standard (Sigma-Aldrich, Sydney, Australia).
Growth hormone and IGF-I ELISA assays.
Commercially available ELISA kits were used to measure plasma growth hormone (GH; Invitrogen-Life Technologies, Mulgrave, Australia) and insulin-like growth factor I (IGF-I) (R & D, Sapphire Bioscience, Sydney, Australia) concentrations in one offspring per litter. Absorbance at 450 nm was measured on an iMark Microplate Absorbance Reader (Bio-Rad).
Taqman gene expression assay.
Total RNA was extracted from ground tissue using TRI reagent (Sigma-Aldrich). After extraction, samples underwent DNAse I treatment (DNA-Free Turbo kit, Ambion-Life Technologies, Scoresby, Australia), followed by on-column RNeasy Micro Clean-up (Qiagen, Melbourne, Australia). After each step, the RNA concentration and quality of the samples (260:280 and 260:230 ratios, respectively) were assessed using a Shimadzu BioSpec-nano spectrophotometer (Shimadzu Scientific Instruments, Columbia, MD). Two micograms of RNA were reverse-transcribed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems-Life Technologies).
TaqMan gene expression assays (Applied Biosystems-Life Technologies) were performed using TaqMan Fast Advanced Master Mix on a Quantstudio DX Real-Time PCR Instrument (Applied Biosystems-Life Technologies). Gene expression was measured in one representative offspring per litter. Gene expression assays were carried out according to the MIQE Guidelines: Minimum Information for Publication of Quantitative real-time PCR Experiments (9). Two reference genes were chosen for each tissue after the expression stability of six potential reference genes was tested across groups, using NormFinder software (1). Differential gene expression was determined by calculation of fold change.
Anthropometric and metabolic measures were expressed as average per litter (n = 8 litters/group). ELISAs and Taqman gene expression assays were performed in one representative offspring per litter. Offspring were chosen based on their metabolic phenotype being representative of the group; the subgroup showed similar phenotypic changes as the data generated from average per litter.
Results were expressed as means ± SD. Data were analyzed using SPSS version 20 (SPSS, Chicago, IL). Normality of the data was tested using the Shapiro-Wilk test, and data were transformed when needed. Body weight data were analyzed using ANOVA with repeated measurements, with age as “within-subject factor” the and father’s diet as “between-subject factor”. All other data were analyzed by Student’s two-tailed t-test.
Metabolic profile of F0 fathers.
At the time of mating, after 12 wk of diet, the F0 male rats on HFD (HFD-F0) were >30% heavier than CD males (CD-F0) (Table 2). Measurements of plasma collected at the time of mating showed that HFD-F0 rats were hyperglycemic, hyperinsulinemic, hypertriglyceridemic, and hyperleptinemic compared with CD rats (Table 2). After mating, F0 male rats underwent GTT after 17 wk and ITT after 18 wk of diet. F0 males on HFD showed an increased area under curve for both GTT and ITT compared with CD males, indicating that they were glucose intolerant and insulin resistant. At cull, after 20 wk of diet, HFD-F0 males had increased fat pad mass, confirming increased adiposity compared with CD-F0.
Body weight and growth of male offspring.
Male offspring from HFD-F0 (HFD-F0 offspring) presented a lower birth weight than those from CD-F0 (CD-F0 7.2 ± 0.9 g, HFD-F0 6.5 ± 0.3 g, P = 0.025; n = 8 litters/group). No difference in litter size between groups was observed. During the lactation period, HFD-F0 offspring remained significantly lighter than their control counterparts (Fig. 1A). From weaning to 9 wk of age, they caught up with the CD-F0 group, with no difference in body weight between groups (Fig. 1B). From 9 wk of age, HFD-F0 offspring showed a significant reduction in body weight compared with CD-F0 (Fig. 1B), remaining significantly lower for the duration of the study.
Offspring culled at 8 wk showed no significant difference in body weight, but compared with CD-F0, HFD-F0 offspring showed a significant decrease in adiposity, which was expressed as the sum of the fat pads collected. At 6 mo, HFD-F0 offspring presented a significant reduction in body weight of >10%, which was associated with a reduced body circumference (Table 3). A significant decrease in fat mass and a trend for decrease in muscle mass were seen in HFD-F0 offspring compared with CD-F0 (Table 3).
No significant difference in metabolic markers was detected (Table 3), but at 8 wk of age we noticed a trend (P = 0.09) toward decreased fasting plasma insulin in HFD-F0 offspring compared with CD-F0.
No significant difference in daily energy intake was detected between groups [average daily energy intake (kJ/day) at 8 wk of age: CD-F0 323 ± 15, HFD-F0 312 ± 20; at 6 mo of age: CD-F0 373 ± 69, HFD-F0 379 ± 44; average cumulative energy intake over the study (from 2 to 20 wk of diet protocol; kJ): CD-F0 43,068 ± 6619, HFD-F0 44,249 ± 3,517].
GH and IGF-I secretion in male offspring.
GH concentration was significantly decreased in plasma of HFD-F0 offspring at 8 wk compared with CD-F0 (Fig. 2A). At 6 mo, GH was secreted at very low levels and no difference between groups was detectable (Fig. 2A).
At 8 wk, IGF-I concentration in plasma of HFD-F0 offspring showed a non-significant reduction compared with CD-F0. At 6 mo, IGF-I levels were higher in both groups compared with 8-wk-old offspring, and HFD-F0 offspring secreted significantly less IGF-I than CD-F0 (Fig. 2B).
Growth pathways in skeletal muscle and adipose tissue of male offspring.
Given the lower adiposity in HFD-F0 offspring, adipogenesis markers were measured in retroperitoneal fat pads.
At 8 wk, adipose tissue from HFD-F0 offspring showed a decrease in the expression of peroxisome proliferator-activated receptor-γ (PPARγ), a key gene involved in adipogenesis (Fig. 3A). They also presented an increase in Igf1r expression. At 6 mo, other markers of adipogenesis showed decreased expression: adiponectin, Cidea (cell death-inducing DFF45-like effector A), and CCAAT/enhancer binding protein-a (Fig. 3B). Leptin expression was also significantly downregulated in HFD-F0 offspring.
At 6 mo, HFD-F0 offspring appeared to have smaller muscle than CD-F0 (Table 2). The reduction in muscle mass reached significance when all offspring from a litter were considered (P = 0.012; n = 13/group, 1–3 animals/litter). Gene analysis revealed a decrease in the expression of key muscle growth markers in tibialis anterior from HFD-F0 offspring at 8 wk: GH receptor (GHr), Igf1, mammalian target of rapamycin (mTOR), and myogenic differentiation factor 1 (Myod1) (Fig. 3C). No changes in expression in muscle growth markers were detected at 6 mo (Fig. 3D).
Lipid metabolism in skeletal muscle of male offspring.
At 6 mo, soleus but not tibialis anterior muscle from HFD-F0 offspring showed an increase in triglyceride content compared with CD-F0 (Table 2). When genes involved in lipid metabolism were examined, an increase in expression in key regulators of lipogenesis, sterol response element-binding factor 1 (Srebf1), and fatty acid synthase (Fasn) was found in soleus from HFD-F0 compared with CD-F0 offspring at 8 wk and 6 mo of age (Fig. 4, A and B).
At 8 wk, the lipid transporter fatty acid-binding protein 4 (Fabp4) was also upregulated in HFD-F0 offspring, as well as acetyl-CoA acetyltransferase 1, which was involved in β-oxidation (Fig. 4A). Carnitine palmitoyl transferase I (Cpt1b), which was also involved in β-oxidation, did not show any difference in expression between groups at 8 wk or 6 mo (Fig. 4).
Epidemiological studies show that paternal BMI can influence both growth trajectory and metabolic risk of offspring (10, 18, 25, 26, 57). Several human studies have shown a stronger father-son than father-daughter association (10, 18, 47); thus we characterized the effects of paternal obesity on male offspring in our rat model of diet-induced obesity. Male offspring from obese fathers showed reduced birth weight, followed by a postweaning catchup growth. Then, unexpectedly, from 9 wk of age, they exhibited reduced growth. This was associated with reduced circulating levels of GH and IGF-I resulting in smaller animals, with smaller fat pads and muscles at 6 mo of age. The downregulation of adipogenesis and muscle growth regulators may account for this phenotype. Furthermore, male offspring from obese fathers showed disturbance in muscle lipid metabolism, which may increase their metabolic risk.
Effects of paternal obesity on the growth trajectory of male offspring.
Findings from our previous study (42) showed reduced birth weight in females from HFD-F0, followed by a catchup growth before weaning, with no difference in body weight from weaning until 14 wk of age. Male offspring from HFD-F0 in this study also showed low birth weight, but their phenotype differed after birth from earlier observations in females. They were smaller during the suckling period. After a short postweaning catchup growth, they were smaller than control from 9 wk until 6 mo of age. In contrast, studies on mouse models of paternal obesity reported no difference in birth weight, followed by no difference or increased postweaning body weight of male offspring from HFD-fed fathers (19, 20, 34), with unchanged or increased adiposity. The difference in paternal effect on offspring body weight between studies may relate to differences in species but also the difference in metabolic phenotype of the fathers at the time of mating. Indeed, the model of diet-induced obesity used by Fullston and colleagues (19, 20) resulted in male mice that were overweight after 10 wk of diet (+15% body weight vs. control-fed animals) but showed no sign of impaired glucose homeostasis. In contrast to Fullston’s mouse model, male rats from our model fed a HFD for 12 wk showed more severe metabolic changes, being 30% heavier than control, severely glucose intolerant, and insulin resistant. Depending on the severity of the metabolic disease, paternal obesity could have different consequences on programming of the offspring.
In our model, low birth weight of offspring from obese fathers reproduces human observation where paternal obesity has been associated with reduced birth weight and fetal growth restriction (10, 26). Low birth weight is an independent risk factor for cardiovascular and metabolic disease in later life (5, 6, 24, 58). A study reported preclinical evidence of insulin resistance in low-birth weight offspring from obese fathers (26). Animal models of intrauterine growth restriction showed that low-birth weight offspring undergo a catchup growth in the preweaning period, resulting in increased adiposity and development of metabolic syndrome later in life (24, 45, 58). In our model, offspring from obese fathers, despite showing a lower birth weight, showed a growth deficit and reduced adiposity with no signs of metabolic syndrome at 6 mo of age. Other models of intrauterine growth restriction did not develop metabolic defects before 18 mo of age (21, 58); this may be similar in our model.
Effects of paternal obesity on the GH/IGF-I axis.
Somatic growth is under control of the somatotropic axis, which involves pituitary GH as a main regulator and IGF expressed in peripheral tissues (29). Normal development involves a pubertal growth spurt driven by a rise in GH secretion (35, 49, 55). In rats, the growth spurt is activated by increased GH expression between 4 and 12 wk of age (27). Whereas offspring in our model followed a typical growth pattern with a postweaning increase in growth rate, HFD-F0 offspring seemed to exit the growth spurt earlier than their control counterparts. This concurred with reduced circulating GH observed at 8 wk. GH secretion decreases during senescence (53), as observed in our 6-mo-old offspring. Most studies have described a similar pattern for plasma IGF-I (4, 27, 53). Surprisingly, in our hands, 6-mo-old rats showed greater plasma concentration of IGF-I than at 8 wk.
Whereas GH secretion decreases with aging, IGF-I expression is maintained in peripheral tissues to preserve cell regeneration (29). At 6 mo, circulating IGF-I levels were lower in offspring from HFD-F0 than CD-F0, potentially contributing to their lower body weight. Insulin also promotes somatic growth (49, 55); in HFD-F0 offspring, we observed a small decrease in fasting insulin plasma level at 8 wk, which may have contributed to the decreased body weight. Other models of parental programming have reported reduced offspring growth, but very few studies have characterized GH and IGF-I profiles. Offspring from maternal low-protein rat models showed a similar profile with low birth weight and reduced postweaning growth, which was associated with decreased postnatal IGF-I secretion (21, 38).
Taken together, we believe that the observed change in growth trajectory in male offspring from obese fathers is the result of a programmed alteration in the GH/IGF-I axis, causing decreased secretion of GH during the growth spurt. This led to impaired development of fat pads and muscles, resulting in smaller animals, which is discussed further below.
Effects of paternal obesity on adipose tissue.
Increased GH during the pubertal growth spurt stimulates the proliferation of preadipocytes and promotes their differentiation (48). The early decrease in GH secretion in offspring from obese fathers likely reduced the number and differentiation of preadipocytes, culminating in reduced numbers of mature adipocytes and smaller fat pads. GH promotes growth and differentiation of preadipocytes both directly and indirectly via IGF-I, which stimulates adipogenesis (7). GH modifies the expression profile of growth factor receptor subtypes on preadipocytes (41), allowing them to respond to IGF-I. Thus, in the absence of GH, preadipocytes are unresponsive to IGF-I. Therefore, although 8-wk-old HFD-F0 offspring had normal levels of IGF-I, this could not compensate for the lack of GH, explaining the impaired fat pad growth. GH is also responsible for establishing a pool of preadipocytes required to allow adipose tissue expansion in adulthood. So a reduced GH in the prepubertal period may impair the lipid storage capacity of adipose tissue in our HFD-F0 offspring. Other tissues may then have to buffer excess lipids causing ectopic lipid accumulation, such as those seen in soleus muscle of HFD-F0 offspring at 6 mo.
Both GH and IGF-I stimulate the expression of adipogenic factors (7, 41) such as CCAAT/enhancer-binding protein (C/EBP)-a and -b and PPARγ to regulate adipocyte differentiation (14). PPARγ, the main effector of adipogenesis, controls the expression of adipose tissue-specific genes such as leptin and adiponectin, the lipid droplet-associated protein Cidea, and genes involved in lipogenesis (56). It has been demonstrated that GH stimulates adipogenesis through activation of PPARγ (28). Thus, the low GH levels at 8 wk may have led to the decreased PPARγ gene expression in adipose tissue of HFD-F0 offspring. IGF-I can induce differentiation of preadipocytes by stimulating expression of C/EBPa (23, 52). Reduced circulating IGF-I levels in 6-mo-old HFD-F0 offspring could participate in the reduced C/EBPa gene expression in adipose tissue as well as leptin, adiponectin, and Cidea. Insulin also promotes differentiation of adipocytes (7, 14, 52), so low plasma insulin at 8 wk may have contributed to inhibiting adipogenesis.
Although we detected decreased expression of leptin in HFD-F0 offspring at 6 mo, this was not reflected in plasma leptin concentrations. A 40% reduction in mRNA may not manifest as a difference in secretion, as other regulatory mechanisms might compensate for the lack of expression. Also, the expression data reported are from retroperitoneal fat pads, and other fat pads might compensate. We noticed increased Igf1r gene expression in adipose tissue of 8-wk-old HFD-F0 offspring, which may be a counterregulatory response to low circulating levels of GH and IGF-I.
Effects of paternal obesity on skeletal muscle.
The GH/IGF-I axis is a major regulator of postnatal muscle hyperplasia and hypertrophy (11, 16, 40, 54). Adults with GH-deficiency have reduced muscle mass, and mice lacking GHr had reduced muscle mass due to defects in myofiber differentiation and growth (11, 36). Therefore, it is not surprising that we observed smaller muscles in HFD-F0 offspring. In parallel with a decreased GH secretion at 8 wk, offspring from obese fathers showed a downregulation of the GHr gene, likely contributing to reduced size of muscle. The effects of GH on muscle are also mediated largely by local production of IGF-I. Studies show that the autocrine/paracrine effect of IGF-I on muscle cells is even more important for muscle growth than circulating levels of IGF-I (11, 12, 29, 54). In this study, HFD-F0 offspring showed a reduced muscle expression of IGF-I at 8 wk, which may have been sufficient to directly impact muscle development in the absence of altered circulating levels of plasma IGF-I. IGF-I induces myogenesis by stimulating the expression of myogenic factors such as MyoD (12, 16), which is also downregulated in the muscle of HFD-F0 offspring at 8 wk.
The GH/IGF-I axis also regulates protein synthesis through mTOR kinase (12, 40, 51). In muscle of HFD-F0 offspring, mTOR expression was reduced at 8 wk, potentially participating in reduced muscle size. Because insulin also stimulates protein synthesis via the mTOR pathway (51, 54), a lower level of insulin at 8 wk in offspring from HFD-F0 may also have slowed muscle growth. No change in expression of these markers was observed at 6 mo. We believe that the reduction in muscle size seen a 6 mo in HFD-F0 offspring results from impaired muscle growth caused by a decrease in circulating GH during the pubertal growth spurt detected at 8 wk.
Lipid metabolism in muscle is also under the influence of the GH/IGF-I axis. GH has been shown to stimulate lipolysis and decrease lipogenesis (11, 29). Low circulating GH level observed at 8 wk could participate in increasing expression of the genes involved in lipid transport and synthesis: Fabp4, Srebf1, and Fasn in soleus muscle from HFD-F0 offspring. This could explain the accumulation of triglycerides observed in soleus muscle of those offspring. The transcription factor SREBP1, encoded by the Srebf1 gene, is the main regulator of Fasn expression, coding for fatty acid synthase (22), both of which are upregulated in muscle of HFD-F0 offspring. SREBP1 has also been shown to inhibit myogenesis through downregulation of Myod1 expression (31); thus the observed increase in its expression might also have participated in reducing muscle mass.
Lipid metabolism was altered in soleus (red, oxidative) and not tibialis anterior (mixed, fast glycolytic). Red-type muscles store more triglycerides to support their oxidative capacity, and therefore, they are more susceptible to ectopic lipid accumulation (37). As a consequence, red oxidative muscles such as soleus are the first to develop insulin resistance due to lipotoxicity (39). Accumulation of triglycerides in the soleus observed in HFD-F0 offspring might be an early sign of metabolic syndrome.
Having consumed a control diet for their whole lives, the only difference between the HFD-F0 and control offspring is the environment to which their fathers’ sperm was exposed. Yet these offspring showed moderate but undeniable programmed alterations. We propose that the altered growth trajectory observed in offspring from obese rat fathers is driven mainly by a decrease in GH secretion, altering the pubertal growth spurt, as observed at 8 wk of age. This resulted in impaired adipogenesis and muscle growth, leading to smaller muscles and fat pads in HFD-F0 offspring at 6 mo of age maintained by decreased IGF-I levels. The reduced lipid storage capacity of adipose tissue of these animals may cause ectopic lipid accumulation, such as that seen in soleus muscle of 6-mo-old HFD-F0 offspring. This could potentially lead to muscle lipotoxicity causing alteration of insulin sensitivity and ultimately metabolic syndrome when offspring age further.
Given their reduced adipose storage capacity and decreased expression of muscle growth factors, It is critical to further investigate the response of these offspring to postnatal challenges, such as HFD consumption or exercise. Another important question we are currently investigating is whether offspring of obese fathers can influence health outcomes of the next generation, providing evidence for transgenerational transmission of obesity.
This work was supported by a project grant from the Australian National Health and Medical Research Council awarded to M. J. Morris and a general grant from the Diabetes Australian Research Trust awarded to V. Lecomte. V. Lecomte was supported by a University of New South Wales Vice Chancellor’s Fellowship.
No conflicts of interest, financial or otherwise, are declared by the authors.
V.L., C.A.M., and K.W.W. performed experiments; V.L., C.A.M., and K.W.W. analyzed data; V.L, C.A.M., and M.J.M. interpreted results; V.L. and C.A.M. drafted manuscript; V.L., C.A.M., and M.J.M. edited and revised manuscript; V.L., C.A.M., K.W.W., and M.J.M. approved final version of manuscript.
- Copyright © 2017 the American Physiological Society