Impact of Diet Composition in Adult Offspring is Dependent on Maternal Diet during Pregnancy and Lactation in Rats

The Thrifty Phenotype Hypothesis proposes that the fetus takes cues from the maternal environment to predict its postnatal environment. A mismatch between the predicted and actual environments precipitates an increased risk of chronic disease. Our objective was to determine if, following a high fat, high sucrose (HFS) diet challenge in adulthood, re-matching offspring to their maternal gestational diet would improve metabolic health more so than if there was no previous exposure to that diet. Animals re-matched to a high prebiotic fiber diet (HF) had lower body weight and adiposity than animals re-matched to a high protein (HP) or control (C) diet and also had increased levels of the satiety hormones GLP-1 and PYY (p < 0.05). Control animals, whether maintained throughout the study on AIN-93M, or continued on HFS rather than reverting back to AIN-93M, did not differ from each other in body weight or adiposity. Overall, the HF diet was associated with the most beneficial metabolic phenotype (body fat, glucose control, satiety hormones). The HP diet, as per our previous work, had detrimental effects on body weight and adiposity. Findings in control rats suggest that the obesogenic potential of the powdered AIN-93 diet warrants investigation.


Introduction
While in utero the fetus is subjected to numerous cues about the environment it will encounter in postnatal life. The fetus is completely dependent on maternal nutrient supply for normal growth and development and therefore many of the environmental cues it encounters are linked to maternal diet. When a mismatch occurs between the nutritional environment predicted by the fetus and the actual environment encountered postnatally, the risk of developing chronic diseases, such as metabolic syndrome or type 2 diabetes, hypertension or cardiovascular disease, is increased [1]. More recent evidence also shows detrimental effects on appetite regulation and glucose and lipid metabolism in response to nutritional mismatch [2]. This programming effect has often been examined in terms of an adverse maternal environment, such as energy or protein restriction, drug treatment, or a maternal high fat diet which is then mismatched in offspring with provision of a control diet or a high energy density diet, either at weaning or in adulthood. Even small degrees of mismatch can have negative consequences on offspring, although in general the greater the mismatch the greater the consequences [3]. While it is typical for a weaning diet high in fat to have negative effects on offspring health regardless of maternal diet, there are some metabolic markers that have been shown to improve with consistent dietary exposure from the pre-to post-natal period. Offspring of high fat-diet fed Pups were weighed on the day after birth, and litters then culled to 10 pups with equal numbers of males and females where possible. Offspring were weighed weekly for the remainder of the study. Food intake was measured for 5 consecutive days out of every 20 days by subtracting the weight of the cup and diet from the previous days' weight. At weaning (3 weeks), 1 male and 1 female pup were randomly selected from each litter to continue in the study until 28 weeks of age. By selecting one male and one female from each litter we examined n = 10 individual rats per sex that were each derived from a different litter, minimizing the effect of any single dam. Therefore, the total number of offspring is n = 10 males and n = 10 females, each from different dams who consumed either control (n = 10), high protein (n = 10) or high fiber (n = 10). Pups were weaned onto AIN-93G control diet [11]. Offspring were then switched to AIN-93M (15.07 kJ/g) for maintenance at 10 weeks of age. At 14 weeks of age, offspring were fed a high fat, high sucrose (HFS) diet (19.26 kJ/g) for 8 weeks (Table 1). Thirteen males and 13 females were kept on AIN-93M as a reference group (R). This reference group, matched for age and sex to the intervention groups, provides a standard of normal growth in these rats. After 8 weeks on HFS, rats were re-matched to the diet of their respective dams for 6 weeks and thereafter identified as C1, HP1 or HF1 (Table 1). Four males and females continued to consume the HFS diet for the final 6 weeks (H). The overall study design is depicted in Figure 1. Pups were weighed on the day after birth, and litters then culled to 10 pups with equal numbers of males and females where possible. Offspring were weighed weekly for the remainder of the study. Food intake was measured for 5 consecutive days out of every 20 days by subtracting the weight of the cup and diet from the previous days' weight. At weaning (3 weeks), 1 male and 1 female pup were randomly selected from each litter to continue in the study until 28 weeks of age. By selecting one male and one female from each litter we examined n = 10 individual rats per sex that were each derived from a different litter, minimizing the effect of any single dam. Therefore, the total number of offspring is n = 10 males and n = 10 females, each from different dams who consumed either control (n = 10), high protein (n = 10) or high fiber (n = 10). Pups were weaned onto AIN-93G control diet [11]. Offspring were then switched to AIN-93M (15.07 kJ/g) for maintenance at 10 weeks of age. At 14 weeks of age, offspring were fed a high fat, high sucrose (HFS) diet (19.26 kJ/g) for 8 weeks (Table 1). Thirteen males and 13 females were kept on AIN-93M as a reference group (R). This reference group, matched for age and sex to the intervention groups, provides a standard of normal growth in these rats. After 8 weeks on HFS, rats were re-matched to the diet of their respective dams for 6 weeks and thereafter identified as C1, HP1 or HF1 (Table 1). Four males and females continued to consume the HFS diet for the final 6 weeks (H). The overall study design is depicted in Figure 1.    HFS, High fat, high sucrose. Inulin supplied as Orafti Raftiline HP and oligofructose as Orafti Raftilose P95 in a 1:1 blend by weight (Quadra Chemicals Ltd., Burlington, ON, Canada). The energy value of the fiber blend (1.5 kcal/g) was used to substitute for an equicaloric amount of cornstarch in the high fiber diet.

Oral Glucose Tolerance Test and Tissue Sampling
Four days before the end of the study, rats were fasted overnight and an oral glucose tolerance test (OGTT) performed. Blood was sampled from the tip of the tail in conscious rats followed by an oral glucose gavage (2 g/kg). At 15, 30, 60 and 90 min post-glucose gavage, additional blood was sampled from the tail and immediately analyzed using a blood glucose meter (Accu-Chek Blood Glucose Meter, Laval, QC, Canada). One day prior to study termination rats underwent a DXA scan (Hologic ODR 4500; Hologic Inc., Bedford, MA, USA) while lightly anesthetized using isoflurane. Hologic QDR software for small animals was used to determine lean mass, fat mass and bone mineral density. A second OGTT for satiety hormone analysis was performed at the time of terminal tissue collection and blood sampled at 0, 15, 30, 60 and 90 min into tubes containing diprotinin-A, Sigma protease inhibitor and Roche Pefabloc at concentrations previously described [12]. Plasma was stored at´80˝C until analysis. The OGTT was a terminal procedure and after the 90 min blood collection rats were killed via over-anesthetization and aortic cut. The stomach, small intestine, cecum, colon and liver were measured and a sample of each was snap frozen in liquid nitrogen and stored at´80˝C.

Hepatic Triglyceride Analysis
Triglyceride content of the liver was quantified using 25 mg of tissue according to the manufacturer guidelines of the GPO reagent set (Pointe Scientific Inc., Lincoln Park, MI, USA).

RNA Extraction and Real-Time PCR
Total RNA was extracted from the liver using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). Reverse transcription was performed with an input of 1 µg of total RNA using the 1st strand cDNA synthesis kit for RT-PCR (Invitrogen) with oligo d(T)15 as a primer. The cDNA was amplified using primers synthesized by the University of Calgary Core DNA Services (Calgary, AB, Canada) and analyzed by real time PCR. Primer sequences for Acetyl Co-A Carboxylase (ACC), Fatty Acid Synthase (FAS), sterol regulatory element binding protein-1c (SREBP1c), glucose-6-phosphatase and AMP-activated protein kinase alpha-1 (AMPKα1) were according to our previous work [10]. A melt curve showed the melting point of the PCR product of interest. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was verified as a suitable housekeeping gene for the tissues of interest and GAPDH primers included as an internal control in the reactions. The 2´∆ CT method (∆CT = CT (gene of interest)-CT (reference gene)) was utilized for the data analysis where threshold cycle (CT) indicates the fractional cycle number at which the amount of amplified target reaches a fixed threshold [13]. The ∆CT is the difference in threshold cycles for the gene of interest and GAPDH.

Statistical Analysis
All data are presented as mean˘SEM. Data collected from the dams was analyzed with one-way ANOVA with Tukey's post hoc analysis. In offspring, a two-way ANOVA was used to compare the main effects of diet and sex and their interaction. Only when a significant interaction effect was identified were all groups, as applicable, compared using a one-way ANOVA with Tukey's post hoc analysis. p ď 0.05 was considered significant. Statistical analysis was performed using SPSS v 19.0 software (Chicago, IL, USA).

Offspring Re-Matched to Maternal HP, HF or C
The following results describe the outcomes in rats re-matched to the diet they were exposed to in utero, either control AIN-93 (C1), HF (HF1), or HP (HP1).

Growth and Energy Intake
The growth trajectory of the pups to the end of the HFS period (22 weeks) has been previously reported [8]. Among animals being re-matched to the HF, HP or C diets, body weight was significantly affected by the interaction of time and diet (p < 0.001) (Figure 2). At weeks 26 and 27, HF1 rats weighed less than C1 rats for both males and females. When re-matched to their maternal diet following the HFS challenge (22-28 weeks), HF1 rats lost weight resulting in a final body weight that was significantly lower than that at week 23. There was also an independent effect of sex with males weighing more than females (p < 0.001). When expressed as change in weight from 23 to 28 weeks, male HP1 and HF1 lost weight and C1 gained weight while in females only the weight loss in HF1 offspring was significantly different from the weight gain in C1.
Energy intake was affected by the interaction of diet and sex at 24 weeks of age with male HP1 consuming more energy than male HF1 (p = 0.034) (Figure 3). At 27 weeks there was a significant sex effect with males consuming more energy than females (p < 0.001). GAPDH primers included as an internal control in the reactions. The 2 −∆CT method (ΔCT = CT (gene of interest)-CT (reference gene)) was utilized for the data analysis where threshold cycle (CT) indicates the fractional cycle number at which the amount of amplified target reaches a fixed threshold [13]. The ∆CT is the difference in threshold cycles for the gene of interest and GAPDH.

Statistical Analysis
All data are presented as mean ± SEM. Data collected from the dams was analyzed with one-way ANOVA with Tukey's post hoc analysis. In offspring, a two-way ANOVA was used to compare the main effects of diet and sex and their interaction. Only when a significant interaction effect was identified were all groups, as applicable, compared using a one-way ANOVA with Tukey's post hoc analysis. p ≤ 0.05 was considered significant. Statistical analysis was performed using SPSS v 19.0 software (Chicago, IL, USA).

Offspring Re-Matched to Maternal HP, HF or C
The following results describe the outcomes in rats re-matched to the diet they were exposed to in utero, either control AIN-93 (C1), HF (HF1), or HP (HP1).

Growth and Energy Intake
The growth trajectory of the pups to the end of the HFS period (22 weeks) has been previously reported [8]. Among animals being re-matched to the HF, HP or C diets, body weight was significantly affected by the interaction of time and diet (p < 0.001) (Figure 2). At weeks 26 and 27, HF1 rats weighed less than C1 rats for both males and females. When re-matched to their maternal diet following the HFS challenge (22-28 weeks), HF1 rats lost weight resulting in a final body weight that was significantly lower than that at week 23. There was also an independent effect of sex with males weighing more than females (p < 0.001). When expressed as change in weight from 23 to 28 weeks, male HP1 and HF1 lost weight and C1 gained weight while in females only the weight loss in HF1 offspring was significantly different from the weight gain in C1.
Energy intake was affected by the interaction of diet and sex at 24 weeks of age with male HP1 consuming more energy than male HF1 (p = 0.034) (Figure 3). At 27 weeks there was a significant sex effect with males consuming more energy than females (p < 0.001). Values are means ± SEM, n = 10 per group except n = 4 for group H. * represents a significant (p < 0.05) difference between HF1 and C1. ** represents a significant (p < 0.05) difference within HF1 between 23 and 28 weeks. For weight change, treatments with different letters are significantly different between diets (p < 0.05): superscripts x,y indicate significant difference between C1, HP1, HF1. There were no differences between R, H and C1. Values are means ± SEM, n = 10 per group except n = 4 for group H. * represents a significant (p < 0.05) difference between HF1 and C1. ** represents a significant (p < 0.05) difference within HF1 between 23 and 28 weeks. For weight change, treatments with different letters are significantly different between diets (p < 0.05): superscripts x,y indicate significant difference between C1, HP1, HF1. There were no differences between R, H and C1.

Body Composition
Male and female HF1 rats had lower body weight and percent body fat than C1 and HP1 rats at 28 weeks of age (p = 0.009) (Figure 3). Small intestine length, colon length and weight, and cecum weight were greater in HF1 than C1 and HP1 rats at euthanization at 28 weeks of age (p < 0.001) ( Table 2). Kidney weight was greater in HP1 than HF1 and C1 (p = 0.037).

Plasma Satiety Hormones and Blood Glucose
In males, fasting insulin was lower in HF1 and HP1 compared to C1 (Table 3). In males and females, fasting PYY and GLP-1 were higher in HF1 versus C1 and HP1 while leptin was lower in HF1 compared to C1 and HP1 (p < 0.05). The interaction between diet and sex affected HOMA-IR scores, which were lower in male HF1 and HP1 than male C1 rats (p = 0.012) ( Table 3).  Glucose AUC was lower in HF1 than C1 animals at 28 weeks of age (p = 0.001) while AUC for insulin was lower in HF1 and HP1 compared to C1 (Figure 4). The interaction between diet and sex affected leptin AUC (p = 0.033) with male C1 rats having higher leptin than male HP1 and HF1 ( Figure 4). AUC for the gut satiety hormones PYY and GLP-1 was higher in HF1 rats compared to C1 and HP1 (p < 0.001) ( Figure 5). Conversely, HF1 rats had lower GIP AUC compared to C1 at 28 weeks of age (p = 0.026). Serial concentrations of glucose over the course of the OGTT are presented in Figure 6. There was a significant effect of time (p < 0.05) but no significant interaction effect between time and diet or sex during the course of the OGTT. Glucose AUC was lower in HF1 than C1 animals at 28 weeks of age (p = 0.001) while AUC for insulin was lower in HF1 and HP1 compared to C1 (Figure 4). The interaction between diet and sex affected leptin AUC (p = 0.033) with male C1 rats having higher leptin than male HP1 and HF1 ( Figure 4). AUC for the gut satiety hormones PYY and GLP-1 was higher in HF1 rats compared to C1 and HP1 (p < 0.001) ( Figure 5). Conversely, HF1 rats had lower GIP AUC compared to C1 at 28 weeks of age (p = 0.026). Serial concentrations of glucose over the course of the OGTT are presented in Figure 6. There was a significant effect of time (p < 0.05) but no significant interaction effect between time and diet or sex during the course of the OGTT.  at 28 weeks. Values are means ± SEM, n = 10 per group except n = 4 for group H. Treatments with different letters are significantly different between diets: superscripts x,y indicate significant difference between C1, HP1, HF1; superscripts a,b indicate significant difference between R, H and C1 (p < 0.05).

Liver Triglyceride
Diet and sex independently affected liver triglyceride concentrations (p = 0.001). Liver triglycerides were higher in males than females, and lower in HF1 than C1 and HP1 rats at 28 weeks of age (Table 3).

Hepatic Gene Expression
Diet had a significant effect on expression of hepatic SREBP-1c, ACC, FAS, and AMPKα1 (p < 0.03) with HP1 rats having higher SREBP1c mRNA levels than HF1 and C1; increased ACC and AMPKα1 mRNA levels than HF1; and increased FAS compared to C1. Sex had a significant effect on SREBP1c and FAS mRNA levels with males having higher expression than females (p < 0.03). The

Liver Triglyceride
Diet and sex independently affected liver triglyceride concentrations (p = 0.001). Liver triglycerides were higher in males than females, and lower in HF1 than C1 and HP1 rats at 28 weeks of age (Table 3).

Hepatic Gene Expression
Diet had a significant effect on expression of hepatic SREBP-1c, ACC, FAS, and AMPKα1 (p < 0.03) with HP1 rats having higher SREBP1c mRNA levels than HF1 and C1; increased ACC and AMPKα1 mRNA levels than HF1; and increased FAS compared to C1. Sex had a significant effect on SREBP1c and FAS mRNA levels with males having higher expression than females (p < 0.03). The interaction of diet and sex affected PGC1α (p < 0.01) and glucose-6-phosphatase mRNA levels (p < 0.02) with higher levels in HP1 versus C1 and HF1 (Table 4). It should be noted that many of the genes measured have post-translational regulatory factors, resulting in potential differences between mRNA expression and physiological function. Table 4. Hepatic gene expression in the offspring of dams fed a control, high-protein or high-prebiotic fiber diet during pregnancy and lactation and then re-matched to their respective diets in adulthood.

Effect of HFS Exposure and Reversibility
The following results describe the outcomes in offspring that were never exposed to the HFS diet and consumed the control AIN-93 diet throughout the study called the reference group (R); offspring continued on HFS for an additional 6 weeks post-HFS challenge (H); or re-matched to the control AIN-93 diet for 6 weeks following the HFS challenge (C1).

Growth and Energy Intake
For the control rats that either continued on AIN-93 control diet (R), switched back to AIN-93 diet following HFS (C1) or continued on HFS (H), there was a significant effect of time but not diet on body weight (p < 0.001) ( Figure 2). Energy intake at 24 weeks was greater in C1 than H (p = 0.025) (Figure 3). At 27 weeks, energy intake was greater in R than C1 and was also greater in males than females (p = 0.037, p < 0.001).

Body Composition
There was no statistical difference in percent body fat at 28 weeks of age between the three groups of offspring from control dams; namely C1, R or H (Figure 3). Lean mass (p = 0.009) was lowest in C1 for males and females and BMD (p = 0.004) was lower in C1 males than R and H. Diet had a significant effect on organ weight measured at 28 weeks of age (Table 2). Stomach weight (p = 0.003) was lower in C1 while cecum weight (p < 0.001) was greater in R than C1 and H. There was also a significant sex effect for these parameters, with males having higher values than females (p < 0.001).

Plasma Satiety Hormones and Blood Glucose
The interaction between diet and sex affected fasting amylin and insulin wherein amylin was higher in male H than C1 and R rats (p < 0.001), and fasting insulin was higher in male C1 than R and H rats (p = 0.006) ( Table 3). Diet had a significant effect on fasting glucose (Table 3) and AUC for glucose, amylin and ghrelin (Figures 4 and 5). C1 rats had higher fasting glucose than R and H rats (p = 0.001) and greater glucose AUC (p < 0.001) and amylin AUC (p < 0.001) than R rats. GLP-1, fasting and AUC, was higher in R than C1 females (p = 0.03) ( Figure 5). Ghrelin AUC (p = 0.021) was greater in R than C1.

Discussion
Our first aim was to determine if the effects of a transient HFS diet could be attenuated by re-matching offspring in adulthood to the diet they were exposed to in utero. Re-matching to a diet high in prebiotic fiber (HF1) reduced body weight and percent body fat compared to both C1 and HP1. We previously reported that a maternal diet high in prebiotic fiber protected female offspring against diet-induced obesity from a HFS challenge, while a maternal diet high in protein pre-disposed female offspring to greater fat mass accumulation [8]. The protection is similar when the diet is introduced at weaning and rats that consume a high prebiotic diet from weaning to adulthood are more resistant to diet-induced obesity compared to control and high protein fed animals [7]. In the current study, the decreased body weight and adiposity in HF1 rats is likely due in part to increased levels of the satiety hormones GLP-1 and PYY. These hormones have frequently been reported to decrease food intake and increase feelings of fullness [14].
Insulin resistance, as measured by HOMA-IR scores, was lower in HF1 and HP1 males than C1 males. Previously, we reported that HOMA-IR scores were lower in HP1 animals than C1 at the end of the HFS period [8]. This protection appears to persist when the rats are re-matched to the HP diet. While a meta-analysis of the long-term effects of low versus high protein diets on metabolic risk factors in human clinical studies [15] did not identify any difference in fasting glucose or glycosylated hemoglobin, fasting insulin was significantly reduced with the high versus low protein diet which is consistent with our data in males. The lower HOMA-IR in the HF animals is consistent with previous reports of improved glycemic and insulinemic response with prebiotic intake [16].
HF1 animals had decreased liver triglyceride concentration compared to HP1, retaining the same pattern as previously reported following the HFS challenge [8]. HP1 animals in addition to having greater hepatic triglyceride levels also demonstrated increased gene expression of the lipogenic enzymes acetyl Co-A carboxylase (ACC), fatty acid synthase (FAS) and sterol regulatory element-binding protein-1c (SREBP1c) in the liver. These enzymes are integrally involved with de novo fat synthesis and have been shown to increase with certain dietary modifications [17]. For example, fatty liver induced by fructose feeding is associated with increased SREBP-1c and FAS mRNA levels and has been shown to be produced by both insulin-dependent and independent pathways [18,19]. Furthermore, high fat diets also increase the expression of ACC, FAS and SREBP-1c [20]. While studies evaluating the effect of high protein diets are more limited, there is evidence that supplemental protein in yaks increases body weight, intramuscular fat and expression of FAS, ACC and SREBP-1c [21].
It has previously been shown that supplementation of the prebiotic fructooligosaccharide results in enlargement of the cecum [22]. As the cecum is the location for fermentation of non-digestible components of the diet, increased fermentation will increase the size of the cecum [23]. We expected and confirmed that HF1 animals had increased cecal mass due to the fermentation of the prebiotic. One of the bacterial species increased with prebiotic fiber intake is Bifidobacterium. This species is strongly associated with gut health and has also been found to have an impact on whole body metabolic health [24]. Specifically, increases in Bifidobacterium spp. are associated with improved measures of glycemia and insulinemia [25]. Increases in bifidobacteria with the HF diet, which we have previously shown [26,27] are likely to explain some of the beneficial effects on glucose tolerance and gut satiety hormones seen when rats were re-matched to the HF diet.
Our second aim was to examine the effects of transient versus persistent exposure to a high fat, high sucrose diet in offspring whose mothers consumed a control diet. We compared a variety of metabolic outcomes in animals that were never exposed to HFS, exposed for the final 14 weeks and those that were re-matched to control diet following the 6 week HFS challenge. Perhaps not surprising given evidence that a diet high in fat and sugar will increase glycemia [28,29], the animals that were never exposed to HFS diet had lower fasting glucose and glucose AUC compared to those that transiently consumed the HFS diet for 8 weeks (i.e., C1). Females in the R group who were exposed solely to control diet pre-and postnatally also had higher fasting GLP-1 and GLP-1 AUC than those with transient exposure to the HFS diet. GLP-1 is an incretin hormone with glucose-lowering properties, secreted from the L-cells of the distal small intestine and proximal colon [30]. This hormone has been shown to be negatively impacted by high fat diets wherein obesity-prone rats consuming a high fat diet have lower plasma GLP-1 levels and a decreased number of L-cells in the distal small intestine [31]; therefore higher GLP-1 levels in the R group would be expected.
In contrast to what one might expect with prolonged exposure to a HFS diet [28,29], animals that remained on HFS for the final 14 weeks of the study in fact had lower fasting glucose compared to the C1 animals in which the exposure to the HFS diet was transient. Insulin levels of animals switched from a cafeteria diet to chow have previously been reported to be three times higher than those never consuming a cafeteria diet [32]. Similarly, our C1 rats re-matched to the control diet after 8 weeks of HFS consumption had higher fasting insulin levels than reference R rats. However, contrary to results from South et al. [32] where any exposure to the cafeteria diet resulted in elevated insulin, our male rats maintained on the HFS diet did not have significantly higher fasting insulin levels than those never exposed to HFS. Fasting glucose was also highest in C1 animals and therefore reflective of HOMA-IR scores which point to greater insulin resistance with transient exposure versus no exposure or prolonged exposure to HFS diet. In a similar diet switch study, rats exposed transiently to a low protein diet showed higher insulin resistance, hyperglycemia and body fat compared to rats fed normal protein continuously, an observation linked to altered sympathetic and vagus tonus [33]. Our results suggest that transient exposure to HFS diet can impair glucose handling and this effect is still evident 6 weeks after switching back to a control diet.
Despite the differences observed in glucose tolerance, rats that were re-matched to control diet after the HFS challenge did not display any differences in body weight or body composition from rats that continued to consume HFS over the final 6 weeks re-matching period. This could be considered "persistent" obesity, as has been reported elsewhere and points to a persistent elevation of body weight and fat mass after the removal of obesogenic diets [34,35]. This is in contrast to other studies that have shown the reversibility of diet-induced obesity and reported weight loss following a switch from a high fat diet to a control chow diet [36,37]. Whether the persistence of obesity occurred in our animals is difficult to conclude due to the similar body weight and body composition to those rats never exposed to the HFS diet (i.e., R rats). The control diet used in this study was AIN-93. Unpublished data from our laboratory has shown that percent body fat is higher in rats consuming the purified powdered AIN-93 diet compared to a standard pelleted chow diet and is not statistically different from fat mass in rats fed a HFS diet for 8 weeks (Neustadter and Reimer, unpublished results [38]). Rodent diet provided in powdered versus pellet form is easier for animals to eat and typically contains ingredients that make it more palatable, such as Dyetrose, a selectively depolymerized corn starch rich in tetrasaccharides. In a comparison of soft and hard food, Sako et al. [39] found that when given the choice, rats would consume more soft food than hard pellets. The lack of weight loss in our C1 animals compared to other studies that switch from a HFS to control diet [36,37] could be due to the palatability and softness of our powdered diet compared to chow pellets which likely impacted food intake and subsequent body weight. Indeed reference animals consumed more energy during the final food intake measurement period than C1 animals, neither of which was different from H animals. The similarity in palatability of our control AIN-93 diet and HFS diets may have played a role in the persistence of obesity given a previous report that decreased food intake and weight loss occurred only when there was a drastic difference in palatability of the diet from liquid Ensure to chow [40].
A limitation of this study design is the inability to distinguish between the lasting effects of maternal diet during pregnancy and maternal diet during suckling. As organ and metabolic systems are in different developmental phases in the offspring in the fetal and early postnatal periods, there is the potential for differential effects of maternal diet that are unique in each period and this should be examined further.
In conclusion, changes in diet composition across the lifespan are a common occurrence [41,42]. Evidence suggests that when the dietary mismatch across periods is great, the susceptibility to chronic disease increases [43]. This study suggests that re-matching offspring to a maternal diet high in prebiotic fiber has clear benefits for reducing the negative impact of a transient HFS diet, including reduced body weight and body fat and improved satiety hormone response in adulthood. Although re-matching offspring to a maternal high protein diet resulted in significantly higher body fat than HF1 rats, they were no worse than C1 animals and the males in fact had better HOMA-IR scores than C1. The surprising lack of difference in body fat between C1, R and H animals suggests that the obesogenic potential of the AIN-93 diet warrants further investigation.