Substantial evidence suggests that poor intrauterine milieu elicited by maternal nutritional disturbance may programme susceptibility in the fetus to later development of chronic diseases, such as obesity, hypertension, cardiovascular disease and diabetes. One of the most interesting features of fetal programming is the evidence from several studies that the consequences may not be limited to the first-generation offspring and that it can be passed transgenerationally. In the present study, female rats (F0) were fed either a normal-protein diet [control diet (C); 19 g of protein/100 g of diet] or a low-protein diet [restricted diet (R); 5 g of protein/100 g of diet]. The offspring were termed according to the period and the types of diet the dams were fed, i.e. CC, RC, CR and RR (first letter indicates the diet during gestation and the second the diet during lactation). At 3 months of age, F1 females were bred to proven males, outside the experiment, to produce F2 offspring. At weaning, F2 offspring were divided by gender. RC1 offspring (with the number indicating the filial generation) were born with low birthweight, but afterwards they had catch-up growth, reaching the weight of the CC1 offspring. The increased glycaemia in RC1 offspring was associated with insulin resistance. CR1 and RR1 offspring had impaired growth with no changes in glucose metabolism. RC2 offspring had high BM (body mass) at birth, which was sustained over the whole experiment in male offspring. The F2 generation had more alteration in glucose metabolism than the F1 generation. CR2 and RC2 offspring had hyperglycaemia accompanied by hyperinsulinaemia and insulin resistance in both genders. CR2 offspring had an increase in body adiposity with hyperleptinaemia. In conclusion, low protein during gestation improves BM, fat mass and growth rate in F1 rats, but has adverse effects on glucose and leptin metabolism, resulting in insulin resistance in adult F1 and F2 offspring. Low protein during lactation has adverse effects on glucose, insulin and leptin metabolism, resulting in insulin resistance in adult F2 offspring. These findings suggest that low protein during gestation and/or lactation can be passed transgenerationally to the second generation.

INTRODUCTION

Numerous experimental and epidemiological investigations have highlighted the role of early reduced calorie and protein nutrition on decreased birthweight and the increased likelihood of elevated blood pressure and cardiovascular diseases in later life [1]. This association is explained by the ‘fetal programming hypothesis’, proposing that adverse environmental influences during critical stages of fetal development, such as insufficient nutrient supply, lead to fetal adaptations that result in both reduced fetal growth and permanent changes in the activity of functional systems [2].

Animal models of intrauterine growth retardation, such as maternal protein restriction, are an invaluable tool to address the genetic, molecular and cellular events that determine fetal growth and development. Rodents are a mammalian system with similar embryology, anatomy and physiology to humans. Utilization of these systems has led to a greater understanding of the pathophysiology and consequences of intrauterine growth retardation. These observations are comparable with that observed in humans born small-for-gestational age, and are of interest because of the known association between poor fetal growth and development of adult disease [3].

The mechanisms through which specific tissues could be affected permanently by nutritional perturbations in early life is still a matter of debate, and include epigenetic changes in gene regulation [4], variations in organ structure [5], alteration in cell number [6], and apoptotic remodelling and metabolic differentiation [7].

There is increasing evidence that some elements of heritable or familial component of susceptibility to chronic non-communicable diseases, such as Type 2 diabetes, obesity and cardiovascular disease, are transmitted non-genomically [8]. The processes whereby environmental influences act during early development to shape disease risk in later life can have effects beyond a single generation. Such heritability may operate through epigenetic mechanisms involving the regulation of either imprinted or non-imprinted genes, but also through broader mechanisms related to parental physiology or behaviour [9]. The fundamental concepts consist of the progression from a healthy phenotype to a chronic disease phenotype due to changes in gene expression or by differences in activities of proteins and enzymes, and that dietary chemicals directly or indirectly regulate the expression of genomic information [10].

The effects of fetal programming may not be limited to the first-generation offspring. One of the most interesting and important features of developmental programming is the evidence from several studies that the consequences of an altered intrauterine environment can be passed transgenerationally from mother (F0) to daughter (F1) to the F2 progeny [1113]. One general mechanism by which prenatal and postnatal exposures could be linked to phenotypic changes in later life is an alteration of epigenetic markers, which have a central role in determining the functional output of the information that is stored in the genome [14]. Epigenetic factors have a potential importance in the intergenerational inheritance of the ‘programming phenotype’ and provide the basis for the inherited association between low birthweight and cardiovascular risk factors [12].

The present study investigates the effects of protein restriction during pregnancy and/or lactation in F0 dams on overall growth, blood pressure and metabolic phenotype of F1 and F2 offspring. We aimed to study whether (i) there are any transgenerational effects on F2 offspring, (ii) these effects are dimorphic, and (iii) they are dependent on the period of development in which protein restriction occurs.

MATERIALS AND METHODS

F1 generation

Virgin female Wistar rats, 12 weeks of age, were obtained from colonies maintained at the State University of Rio de Janeiro. Rats were maintained in a temperature (21±2 °C)- and humidity (60±10%)-controlled room, with a 12h-dark/light cycle (artificial lights, 07.00–19.00 hours) and an air-exhaustion cycle of 15 min/h. All procedures were carried out in accordance with the Conventional Guidelines for Experimentation with Animals (NIH Publication No. 85-23, revised 1996). The Ethics Committee for Animal Experimentation at the State University of Rio de Janeiro approved the experimental procedures used in the study.

Females (F0) were mated with proven male breeders, and the day on which spermatozoa were present in a vaginal smear was designated as the day of conception. Only females that were pregnant within 5 days of the introduction of the male were retained in the study. Pregnant rats were transferred to individual cages and allocated at random to one of two groups to be fed either a normal-protein diet [control diet (C), with 19 g of protein/100 g of diet] or a low-protein diet [restricted diet (R), with 5 g of protein/100 g of diet). Both diets were isoenergetic (1900 kJ/100 g of diet). The restricted diet was compensated by the addition of carbohydrates. The mineral and vitamin contents of the two diets were identical and in accordance with the American Institute of Nutrition's recommendation (AIN 93G) [15]. Diets were produced by Rhoster, and their compositions are shown in Table 1.

Table 1
Composition of the experimental diets

Vitamin and mineral mixes were formulated to meet the American Institute of Nutrition AIN 93G recommendation for rodents.

ControlRestricted
Nutrient(normal-protein) diet(low-protein) diet
Carbohydrate (g/100 g)   
 Corn starch 33.5 47.7 
 Sucrose 20.0 20.0 
Casein (g/100 g) 19 
Cystine (g/100 g) 0.3 0.15 
Choline (g/100 g) 0.2 0.2 
Fat (soya bean oil) (g/100 g) 16.0 16.0 
Fibre (cellulose) (g/100 g) 5.0 5.0 
Mineral mix (g/100 g) 1.0 1.0 
Vitamin mix (g/100 g) 5.0 5.0 
Energy (kJ/100 g of diet) 1900 1900 
ControlRestricted
Nutrient(normal-protein) diet(low-protein) diet
Carbohydrate (g/100 g)   
 Corn starch 33.5 47.7 
 Sucrose 20.0 20.0 
Casein (g/100 g) 19 
Cystine (g/100 g) 0.3 0.15 
Choline (g/100 g) 0.2 0.2 
Fat (soya bean oil) (g/100 g) 16.0 16.0 
Fibre (cellulose) (g/100 g) 5.0 5.0 
Mineral mix (g/100 g) 1.0 1.0 
Vitamin mix (g/100 g) 5.0 5.0 
Energy (kJ/100 g of diet) 1900 1900 

All F0 rats delivered the F1 generation by spontaneously vaginal delivery. The F1 litter sizes and pup BMs (body masses) were recorded at birth, and, to standardize food supply, litters were adjusted to six pups/dam (1:1 gender ratio where possible) [16]. The pups were accompanied until weaning when only one pup/litter was randomly assigned to form the F1 groups of study [17]. The offspring were then termed according to the period and the type of diet the dams were fed, i.e. CC1 offspring (F1 offspring from dams fed the control diet in both gestation and lactation); RC1 offspring (F1 offspring from dams fed the restricted diet during gestation and the control diet during lactation); CR1 offspring (F1 offspring from dams fed the control diet during gestation and the restricted diet during lactation); and RR1offspring (F1 offspring from dams fed the restricted diet in both gestation and lactation). After weaning, pups had free access to the control diet and water, and continued to be measured weekly.

F2 generation

At 3 months of age, F1 females were bred to proven males, outside the experiment, to produce F2 offspring. During the pregnancy and for the rest of the study, F1 females were fed the control diet. F1 females underwent vaginal delivery spontaneously. The F2 litter size and pup BMs were recorded at birth and after weaning. At birth, the litter size was adjusted to six pups/dam, maintaining as close to a 1:1 gender ratio as possible. The animals were accompanied until weaning when only one pup/litter was randomly assigned to form the F2 groups. At weaning, F2 offspring were divided by gender and termed according to the period and the type of diet F0 dams were fed, i.e. CC2, CR2, RC2 and RR2. After weaning, F2 pups had free access to the control diet and water, and continued to be measured weekly.

Biometry

BM and NAL (naso-anal length) were measured at the same time every week. These procedures were the same for both F1 and F2 offspring.

Termination of the experiment

Animals of both genders of F1 and F2 offspring were followed until 6 months old. At 24 h before the end of the experiment, animals were kept in metabolic cages and were food-deprived overnight. Animals were deeply anaesthetized (by an intraperitoneal injection of sodium pentobarbital), the thorax was opened and then blood samples were rapidly obtained from puncture of the right atrium. Afterwards, the abdomen was opened and the retroperitoneal and genital fat pads were completely removed on both sides of the animal and were then weighed. The retroperitoneal fat pad was taken as the distinct deposit around each kidney along the lumbar muscles. Genital fat pad (ovarian for female and epididymal for male) included adipose tissue surrounding the ureters, bladder, epididymis, ovaries, oviducts and uterus.

Biochemical analyses and serum hormone concentrations

After blood collection, serum was obtained by centrifugation (120 g for 15 min) at room temperature (21 °C) and was stored individually at −20 °C until assayed. Fasting glucose concentrations were determined at the moment of death, using the glucose-oxidase method and a glucometer reader (Accu-chek; Roche Diagnostic). Fasting insulin and leptin concentrations were measured by RIA [RI-13K for insulin (intra-assay coefficient of variation was 1.4%) and RL-83K for leptin (intra-assay coefficient of variation was 4.2%); Linco Research].

HOMA (homoeostasis model assessment) for insulin resistance was calculated with baseline values: HOMA=(fasting glucose×fasting insulin)/22.5.

Data analysis

Values are expressed as means±S.E.M. Differences among male groups and among female groups in the same generation were analysed by one-way ANOVA and Tukey’s post-hoc test. Differences between genders in the same generation or between F1 and F2 offspring were analysed with an unpaired Student’s t test with Welch's correction. Three-way (2×2×2 factorial) ANOVA was performed to analyse interactions among effects (gender compared with type of diet during gestation compared with type of diet during lactation) (Statistica, version 7.0; Statsoft).

The correlation of the BM (dependent variable) with age (independent variable) was tested for the different groups. The bivariate study used log-transformed data and the allometric model log y=log a+(b)log x [18]. Because of the problem of biased estimates of slopes of y on x when both variables are subject to measurement error, the slope of the principal axis of the standardized variables, i.e. reduced major axis, was computed. A Student's t test was used to test the significance departure from a predicted slope examined with residual analysis [19]. The Pearson product-moment correlation coefficient and the linear regression were established for each age group and, in the cases where it could be relevant, we compared the two regressions (comparison of slopes) (Primer of Biostatistics, version 5.0; McGraw-Hill). In all analyses, a P value of 0.05 was considered statistically significant.

RESULTS

F1 generation

No significant differences were observed between pregnancies in dams fed the control and restricted diet in terms of litter size number, the number of dead pups at birth and the composition of the litter gender (Table 2).

Table 2
Biometrical results for offspring from the F1 and F2 generations

Values are means±S.E.M. *Significantly different from C1; †significantly different from CC offspring in the same generation; ‡ significantly different from CR offspring in the same generation; § significantly different from RC offspring in the same generation; ∥significantly different from the corresponding male offspring; ¶ significantly different from the corresponding F1 group.

Offspring
At birthC1R1
Litter size (n10±0.4 10.4±0.5   
Pups dead (n  
Litter male/female distribution 1:2 1:1   
Male BM (g) 6.2±0.2 4.1±0.2*   
Female BM (g) 5.9±0.2 4.4±0.1*   
Male NAL (cm) 5.1±0.2 4.5±0.3*   
Female NAL (cm) 4.9±0.1 4.4±0.3*   
Offspring
From weaning to 6 months of ageCC1CR1RC1RR1
Male     
 BM at weaning (g) 32.4±1.0 16.4±1.0† 32.6±1.3‡ 11.2±0.5†‡§ 
 NAL at weaning (cm) 10.3±0.2 8.3±0.1† 10.3±0.1‡ 7.4±0.1†‡§ 
 BM at 6 months of age (g) 385.2±16.2 337.2±7.4† 376.0±14.9 306.0±4.7†§ 
 NAL at 6 months of age (cm) 24.2±0.3 23.7±0.1 24.0±0.4 22.7±0.2†‡§ 
Female     
 BM at weaning (g) 42.8±0.8∥ 16.8±0.8† 31.6±1.2†‡ 10.4±0.8†‡§ 
 NAL at weaning (cm) 11.5±0.1∥ 8.1±0.2† 10.0±0.1†‡ 7.0±0.1†‡§ 
 BM at 6 months of age 267.6±6.0∥ 228.8±7.9†∥ 250.8±8.4‡∥ 219.2±7.4†§∥ 
 NAL at 6 months of age (cm) 21.0±0.1∥ 20.1±0.2†∥ 20.8±0.2†∥ 20.0±0.2†§∥ 
(B) F2 generation
Offspring
At birthCC2CR2RC2RR2
Litter size (n13±1.1 11±0.7 8±2.3 12±3.3 
Pups dead (n
Litter male/female distribution 1:2 1:1 1:1 1:1 
Male BM (g) 6.1±0.2 5.7±0.4 7.0±0.3†‡ 5.8±0.3§ 
Female BM (g) 6.0±0.1 5.9±0.2 6.7±0.2†‡ 5.1±0.2†‡§ 
Male NAL (cm) 4.9±0.04 4.8±0.1 5.1±0.1‡ 4.9±0.01 
Female NAL (cm) 4.9±0.04 4.7±0.1 5.1±0.1‡ 4.9±0.01 
Offspring
From weaning to 6 months of ageCC2CR2RC2RR2
Male     
 BM at weaning (g) 36.4±0.8 32.8±3.0¶ 39.2±1.7¶ 32.4±1.5¶ 
 NAL at weaning (cm) 10.9±0.2 10.3±0.2 10.4±0.1 10.4±0.1 
 BM at 6 months of age (g) 370.4±11.9 384.0±10.4¶ 447.6±8.2†‡¶ 378.4±8.3§¶ 
 NAL at 6 months of age (cm) 23.7±0.2 24.2±0.1¶ 24.6±0.1† 24.0±0.2¶ 
Female     
 BM at weaning (g) 38.0±2.2 33.6±1.7¶ 38.8±0.5¶ 33.6±2.8¶ 
 NAL at weaning (cm) 10.5±0.2 10.3±0¶ 10.7±0.2 10.2±0.2¶ 
 BM at 6 months of age (g) 246±4.3∥ 270.5±11.7∥¶ 266.7±9.7∥ 236.4±7.8∥ 
 NAL at 6 months of age (cm) 20.5±0.1 20.8±0.1¶ 20.9±0.2 21.0±0.1¶ 
Offspring
At birthC1R1
Litter size (n10±0.4 10.4±0.5   
Pups dead (n  
Litter male/female distribution 1:2 1:1   
Male BM (g) 6.2±0.2 4.1±0.2*   
Female BM (g) 5.9±0.2 4.4±0.1*   
Male NAL (cm) 5.1±0.2 4.5±0.3*   
Female NAL (cm) 4.9±0.1 4.4±0.3*   
Offspring
From weaning to 6 months of ageCC1CR1RC1RR1
Male     
 BM at weaning (g) 32.4±1.0 16.4±1.0† 32.6±1.3‡ 11.2±0.5†‡§ 
 NAL at weaning (cm) 10.3±0.2 8.3±0.1† 10.3±0.1‡ 7.4±0.1†‡§ 
 BM at 6 months of age (g) 385.2±16.2 337.2±7.4† 376.0±14.9 306.0±4.7†§ 
 NAL at 6 months of age (cm) 24.2±0.3 23.7±0.1 24.0±0.4 22.7±0.2†‡§ 
Female     
 BM at weaning (g) 42.8±0.8∥ 16.8±0.8† 31.6±1.2†‡ 10.4±0.8†‡§ 
 NAL at weaning (cm) 11.5±0.1∥ 8.1±0.2† 10.0±0.1†‡ 7.0±0.1†‡§ 
 BM at 6 months of age 267.6±6.0∥ 228.8±7.9†∥ 250.8±8.4‡∥ 219.2±7.4†§∥ 
 NAL at 6 months of age (cm) 21.0±0.1∥ 20.1±0.2†∥ 20.8±0.2†∥ 20.0±0.2†§∥ 
(B) F2 generation
Offspring
At birthCC2CR2RC2RR2
Litter size (n13±1.1 11±0.7 8±2.3 12±3.3 
Pups dead (n
Litter male/female distribution 1:2 1:1 1:1 1:1 
Male BM (g) 6.1±0.2 5.7±0.4 7.0±0.3†‡ 5.8±0.3§ 
Female BM (g) 6.0±0.1 5.9±0.2 6.7±0.2†‡ 5.1±0.2†‡§ 
Male NAL (cm) 4.9±0.04 4.8±0.1 5.1±0.1‡ 4.9±0.01 
Female NAL (cm) 4.9±0.04 4.7±0.1 5.1±0.1‡ 4.9±0.01 
Offspring
From weaning to 6 months of ageCC2CR2RC2RR2
Male     
 BM at weaning (g) 36.4±0.8 32.8±3.0¶ 39.2±1.7¶ 32.4±1.5¶ 
 NAL at weaning (cm) 10.9±0.2 10.3±0.2 10.4±0.1 10.4±0.1 
 BM at 6 months of age (g) 370.4±11.9 384.0±10.4¶ 447.6±8.2†‡¶ 378.4±8.3§¶ 
 NAL at 6 months of age (cm) 23.7±0.2 24.2±0.1¶ 24.6±0.1† 24.0±0.2¶ 
Female     
 BM at weaning (g) 38.0±2.2 33.6±1.7¶ 38.8±0.5¶ 33.6±2.8¶ 
 NAL at weaning (cm) 10.5±0.2 10.3±0¶ 10.7±0.2 10.2±0.2¶ 
 BM at 6 months of age (g) 246±4.3∥ 270.5±11.7∥¶ 266.7±9.7∥ 236.4±7.8∥ 
 NAL at 6 months of age (cm) 20.5±0.1 20.8±0.1¶ 20.9±0.2 21.0±0.1¶ 

BM and NAL (Figure 1, Table 2 and Table 3)

At birth, BM of the R1 offspring were significantly lower than the C1 offspring (both genders P<0.0001, as determined by one-way ANOVA). Similar results were obtained with the NAL, with the R1 offspring having a shorter NAL than the C1 offspring (both genders P<0.0001, as determined by one-way ANOVA).

Progression of BM in the F1 and F2 generations
Figure 1
Progression of BM in the F1 and F2 generations

Values are means±S.E.M. Weaning is marked with an arrow on the x-axis.

Figure 1
Progression of BM in the F1 and F2 generations

Values are means±S.E.M. Weaning is marked with an arrow on the x-axis.

Table 3
ANOVA to test the effects and interaction of gender, gestation diet and lactation diet on BM in the offspring of the F1 and F2 generations

One-, two- or three-way ANOVA was used as appropriate. G, gender, GD, gestational diet; LD, lactation diet; ns, not significant.

BM
GenerationBirthWeaning6 Months of age
F1    
 G effect ns P<0.0001 P<0.0001 
 GD effect P<0.0001 P<0.0001 P<0.0001 
 LD effect ns P<0.0001 P<0.0001 
 G×GD interaction P<0.01 P<0.0001 ns 
 G× LD interaction ns P<0.0001 ns 
 GD×LD interaction ns ns ns 
 G×GD×LD interaction ns P<0.0001 ns 
F2    
 G effect ns ns P<0.0001 
 GD effect ns P<0.0001 P<0.0001 
 LD effect P<0.0001 ns P<0.0001 
 G×GD interaction ns ns P<0.0001 
 G×LD interaction ns ns P<0.0001 
 GD×LD interaction ns ns P<0.0001 
 G×GD×LD interaction ns ns ns 
BM
GenerationBirthWeaning6 Months of age
F1    
 G effect ns P<0.0001 P<0.0001 
 GD effect P<0.0001 P<0.0001 P<0.0001 
 LD effect ns P<0.0001 P<0.0001 
 G×GD interaction P<0.01 P<0.0001 ns 
 G× LD interaction ns P<0.0001 ns 
 GD×LD interaction ns ns ns 
 G×GD×LD interaction ns P<0.0001 ns 
F2    
 G effect ns ns P<0.0001 
 GD effect ns P<0.0001 P<0.0001 
 LD effect P<0.0001 ns P<0.0001 
 G×GD interaction ns ns P<0.0001 
 G×LD interaction ns ns P<0.0001 
 GD×LD interaction ns ns P<0.0001 
 G×GD×LD interaction ns ns ns 

The RC1 offspring had a catch-up of BM when compared with CC1 offspring. In contrast, the CR1 and RR1 offspring had impaired growth, thus they were lighter than the CC1 and RC1 offspring. When compared with the CC1 offspring, the BM was 65% less in male RR1 offspring and 50% less in CR1 offspring at weaning (P<0.0001, as determined by one-way ANOVA). At 6 months of age, BM was 20% less in male RR1 offspring (P<0.0001, as determined by one-way ANOVA) and 10% less in CR1 offspring (P<0.001, as determined by one-way ANOVA). BM was 30% less in male RR1 offspring compared with male CR1 offspring at weaning (P<0.0001, as determined by one-way ANOVA), but no difference was found at 6 months of age. Females had comparable results.

At birth, BM was influenced by the interaction between gender and gestation diet, and between gender and lactation diet (Table 3). At weaning, BM was also influenced by gender, gestation diet and lactation diet in isolation and also by the interaction between gender and gestation diet, gender and lactation diet, and gender, gestation diet and lactation diet (Table 3). Therefore these interactions were no longer evident at 6 months of age (Table 3).

At weaning, the NAL of male CR1 offspring was 20% less and of male RR1 offspring was 30% less than that of both the CC1 and RC1 offspring (P<0.0001, as determined by one-way ANOVA). At 6 months of age, the NAL remained significantly shorter in RR1 offspring than in all of the other groups (P<0.01, as determined by one-way ANOVA). Females had comparable results (Table 2).

Growth rate (Table 4)

The increase in BM in the offspring in a specific age group was investigated. The analysis, based on the slope magnitude of the allometric equation, indicated the growth rate for that period. RC1 offspring has the highest growth rate among the groups during lactation period.

Table 4
Comparative growth rates using the allometric model

Slope (S.E.M.) calculated with the reduced major axis approximation based on the linear regression of BM (in g) as the dependent variable (y) against offspring age (in days) as the independent variable (x), fitted using the equation log y=log a+b×log x. †Significantly different from CC offspring in the same generation; ‡significantly different from CR offspring in the same generation; §significantly different from RC offspring in the same generation; ∥significantly different from the corresponding male offspring; ¶significantly different from the corresponding F1 group; *significantly different from the previous time period in the same group.

DayMonth
Offspring1–211–22–33–44–55–6
F1 male       
 CC1 0.55 (0.02) 1.69 (0.05)* 1.17 (0.12)* 0.49 (0.15)* 1.15 (0.34) 0.81 (0.27) 
 CR1 0.33 (0.02)† 2.09 (0.11)†* 1.12 (0.04)* 0.58 (0.04)* 0.81 (0.11)* 0.50 (0.12)* 
 RC1 0.71 (0.03)†‡ 1.76 (0.10)‡* 0.97 (0.10)* 0.51 (0.11)* 0.75 (0.24) 0.71 (0.24) 
 RR1 0.35 (0.02)†§ 2.19 (0.10)†‡* 1.30 (0.07)‡§* 0.58 (0.18)* 0.79 (0.20) 0.30 (0.09) 
F1 female       
 CC1 0.67 (0.01)§ 0.95 (0.06)∥* 0.68 (0.08)∥* 0.34 (0.05)* 0.42 (0.14) 0.45 (0.14) 
 CR1 0.35 (0.02)† 1.90 (0.07)†* 0.83 (0.12)†* 0.57 (0.16)* 0.99 (0.34) 0.77 (0.25) 
 RC1 0.67 (0.02)‡ 1.30 (0.11)†‡∥* 1.25 (0.12)† 0.51 (0.12)* 0.75 (0.25) 0.59 (0.20) 
 RR1 0.29 (0.03)†‡§ 1.94 (0.09)†§∥* 1.05 (0.12)†§* 0.58 (0.12)* 0.79 (0.23)§ 0.54 (0.19) 
F2 male       
 CC2 0.62 (0.02)¶ 1.57 (0.07)* 1.09 (0.12)* 0.49 (0.12)* 0.84 (0.26) 0.69 (0.21) 
 CR2 0.60 (0.04)¶ 1.62 (0.06)¶* 1.15 (0.10)* 0.64 (0.09)¶* 0.69 (0.21) 0.57 (0.20) 
 RC2 0.59 (0.02)¶ 1.96 (0.09)†‡* 0.78 (0.06)‡* 0.56 (0.08) 0.60 (0.15) 0.39 (0.11) 
 RR2 0.59 (0.02)¶ 1.49 (0.09)§¶* 0.95 (0.13)¶* 0.61 (0.08)* 0.59 (0.15)¶ 0.41 (0.14) 
F2 female       
 CC2 0.63 (0.02) 1.40 (0.03)∥¶* 0.59 (0.06)∥* 0.38 (0.08) 0.48 (0.15) 0.36 (0.12) 
 CR2 0.59 (0.02)¶ 1.37 (0.09)∥¶* 0.79 (0.11)∥* 0.49 (0.16) 0.72 (0.29) 0.50 (0.17) 
 RC2 0.60 (0.01)¶ 1.52 (0.09)∥* 0.66 (0.15)¶* 0.32 (0.14) 0.75 (0.28) 0.68 (0.24) 
 RR2 0.64 (0.03)¶ 1.46 (0.08)¶* 0.54 (0.09)∥¶* 0.21 (0.11)§* 0.44 (0.29) 0.51 (0.12) 
DayMonth
Offspring1–211–22–33–44–55–6
F1 male       
 CC1 0.55 (0.02) 1.69 (0.05)* 1.17 (0.12)* 0.49 (0.15)* 1.15 (0.34) 0.81 (0.27) 
 CR1 0.33 (0.02)† 2.09 (0.11)†* 1.12 (0.04)* 0.58 (0.04)* 0.81 (0.11)* 0.50 (0.12)* 
 RC1 0.71 (0.03)†‡ 1.76 (0.10)‡* 0.97 (0.10)* 0.51 (0.11)* 0.75 (0.24) 0.71 (0.24) 
 RR1 0.35 (0.02)†§ 2.19 (0.10)†‡* 1.30 (0.07)‡§* 0.58 (0.18)* 0.79 (0.20) 0.30 (0.09) 
F1 female       
 CC1 0.67 (0.01)§ 0.95 (0.06)∥* 0.68 (0.08)∥* 0.34 (0.05)* 0.42 (0.14) 0.45 (0.14) 
 CR1 0.35 (0.02)† 1.90 (0.07)†* 0.83 (0.12)†* 0.57 (0.16)* 0.99 (0.34) 0.77 (0.25) 
 RC1 0.67 (0.02)‡ 1.30 (0.11)†‡∥* 1.25 (0.12)† 0.51 (0.12)* 0.75 (0.25) 0.59 (0.20) 
 RR1 0.29 (0.03)†‡§ 1.94 (0.09)†§∥* 1.05 (0.12)†§* 0.58 (0.12)* 0.79 (0.23)§ 0.54 (0.19) 
F2 male       
 CC2 0.62 (0.02)¶ 1.57 (0.07)* 1.09 (0.12)* 0.49 (0.12)* 0.84 (0.26) 0.69 (0.21) 
 CR2 0.60 (0.04)¶ 1.62 (0.06)¶* 1.15 (0.10)* 0.64 (0.09)¶* 0.69 (0.21) 0.57 (0.20) 
 RC2 0.59 (0.02)¶ 1.96 (0.09)†‡* 0.78 (0.06)‡* 0.56 (0.08) 0.60 (0.15) 0.39 (0.11) 
 RR2 0.59 (0.02)¶ 1.49 (0.09)§¶* 0.95 (0.13)¶* 0.61 (0.08)* 0.59 (0.15)¶ 0.41 (0.14) 
F2 female       
 CC2 0.63 (0.02) 1.40 (0.03)∥¶* 0.59 (0.06)∥* 0.38 (0.08) 0.48 (0.15) 0.36 (0.12) 
 CR2 0.59 (0.02)¶ 1.37 (0.09)∥¶* 0.79 (0.11)∥* 0.49 (0.16) 0.72 (0.29) 0.50 (0.17) 
 RC2 0.60 (0.01)¶ 1.52 (0.09)∥* 0.66 (0.15)¶* 0.32 (0.14) 0.75 (0.28) 0.68 (0.24) 
 RR2 0.64 (0.03)¶ 1.46 (0.08)¶* 0.54 (0.09)∥¶* 0.21 (0.11)§* 0.44 (0.29) 0.51 (0.12) 

During the lactation period, CR1 and RR1 offspring (both genders) had significantly smaller growth rates than CC1 and RC1 offspring. In the post-weaning period, this trend was altered. In the period from 1–2 months of age, the CR1 and RR1 offspring (both genders) now had greater growth rates than CC1 and RC1 offspring, indicating catch-up growth in these groups during this period. This trend disappeared in the subsequent months of the post-weaning period.

Fat mass, and glucose, insulin and leptin levels (Table 5)

Fat mass

Male RC1 offspring had 30% more fat mass than CC1 offspring (P<0.001, as determined by one-way ANOVA), 50% more fat mass than CR1 offspring (P<0.0001, as determined by one-way ANOVA) and 150% more fat mass than RR1 offspring (P<0.0001, as determined by one-way ANOVA). Male RR1 offspring, however, had 50% less fat mass than CC1 offspring and 40% less of fat mass than CR1 offspring (P<0.0001, as determined by one-way ANOVA). In the corresponding female groups, the differences were less marked than in the male groups. Fat mass was affected by the interactions between gender and lactation diet and between gestation diet and lactation diet (Table 5).

Table 5
Hormone concentrations, HOMA and percentage fat mass in 6-month-old F1 offspring

Values are means±S.E.M. Blood was collected from the right atrium by cardiac puncture. †Significantly different from CC offspring; ‡ significantly different from CR offspring; § significantly different from RC offspring; ∥significantly different from the corresponding male offspring. For ANOVA, one-, two- or three-way ANOVA was used as appropriate. G, gender, GD, gestational diet; LD, lactation diet; ns, not significant.

GroupFat mass (%)Glucose (mmol/l)Insulin (pmol/l)HOMALeptin (ng/ml)
Males      
 CC1 2.9±0.2 4.9±0.2 94.5±8.9 18.2±1.0 1.0±0.1 
 CR1 2.5±0.2 5.0±0.2 87.5±13.6 16.4±0.9 1.1±0.1 
 RC1 3.8±0.2†‡ 6.1±0.2†‡ 112.0±21.2 30.1±5.6†‡ 2.0±0.2†‡ 
 RR1 1.5±0.2†‡§ 5.3±0.1§ 66.5±17.0 15.8±4.0§ 1.4±0.1§ 
Female      
 CC1 2.5±0.3 4.6±0.2 77.0±16.2 15.8±8.1 1.0±0.1 
 CR1 4.0±0.6†∥ 4.4±0.2 80.4±19.6 16.1±4.0 1.4±0.2 
 RC1 4.5±0.3† 5.5±0.2†‡∥ 42.0±4.3§ 10.2±1.1§ 1.1±0.1∥ 
 RR1 2.7±0.5§ 5.0±0.3 70.0±0.5 15.7±0.8 1.5±0.2 
ANOVA      
 G effect P<0.001 P<0.001 P<0.01 P<0.001 ns 
 GD effect ns P<0.0001 ns ns P<0.001 
 LD effect P<0.01 ns ns ns ns 
 G×GD interaction ns ns ns ns ns 
 G×LD interaction P<0.01 ns P<0.05 ns ns 
 GD×LD interaction P<0.0001 P<0.01 ns ns ns 
 G×GD×LD interaction ns ns ns P<0.05 P<0.05 
GroupFat mass (%)Glucose (mmol/l)Insulin (pmol/l)HOMALeptin (ng/ml)
Males      
 CC1 2.9±0.2 4.9±0.2 94.5±8.9 18.2±1.0 1.0±0.1 
 CR1 2.5±0.2 5.0±0.2 87.5±13.6 16.4±0.9 1.1±0.1 
 RC1 3.8±0.2†‡ 6.1±0.2†‡ 112.0±21.2 30.1±5.6†‡ 2.0±0.2†‡ 
 RR1 1.5±0.2†‡§ 5.3±0.1§ 66.5±17.0 15.8±4.0§ 1.4±0.1§ 
Female      
 CC1 2.5±0.3 4.6±0.2 77.0±16.2 15.8±8.1 1.0±0.1 
 CR1 4.0±0.6†∥ 4.4±0.2 80.4±19.6 16.1±4.0 1.4±0.2 
 RC1 4.5±0.3† 5.5±0.2†‡∥ 42.0±4.3§ 10.2±1.1§ 1.1±0.1∥ 
 RR1 2.7±0.5§ 5.0±0.3 70.0±0.5 15.7±0.8 1.5±0.2 
ANOVA      
 G effect P<0.001 P<0.001 P<0.01 P<0.001 ns 
 GD effect ns P<0.0001 ns ns P<0.001 
 LD effect P<0.01 ns ns ns ns 
 G×GD interaction ns ns ns ns ns 
 G×LD interaction P<0.01 ns P<0.05 ns ns 
 GD×LD interaction P<0.0001 P<0.01 ns ns ns 
 G×GD×LD interaction ns ns ns P<0.05 P<0.05 
Glucose

Among the groups, fasting glucose levels varied significantly with gender and gestation diet (Table 5). Gestation diet and lactation diet interacted influencing fasting glucose (Table 5). Male RC1 offspring had the highest fasting glucose levels, 20% more than both the CC1 and CR1 offspring, and 15% more than RR1 offspring (P<0.001, as determined by one-way ANOVA). Fasting glucose levels in females were also higher in RC1 offspring, 20% more than in CC1 offspring and 25% more than CR1 offspring (P<0.01, as determined by one-way ANOVA).

Insulin

Plasma insulin levels were not significantly different among the groups in both the male and female F1 offspring.

HOMA

HOMA indicated an insulin-resistant state in male RC1 offspring when compared with CC1, CR1 and RR1 offspring (P<0.01, as determined by one-way ANOVA), with a significant three-factor interaction (gender, gestation diet and lactation diet; Table 5).

Leptin

Male RC1 offspring had 100% more circulating plasma leptin than CC1 offspring (P<0.0001, as determined by one-way ANOVA), 80% more than CR1 offspring (P<0.0001, as determined by one-way ANOVA), and 40% more than RR1 offspring (P<0.001, as determined by one-way ANOVA). ANOVA revealed a significant three-factor interaction between gender, gestation diet and lactation diet (Table 5). In contrast, females had no evidence of hyperleptinaemia.

F2 generation

There were no differences in the number of pups born dead, litter size and litter gender composition among the groups (Table 2).

BM and NAL (Figure 1 and Table 2)

At birth, both genders of RC2 offspring were significantly heavier, and had longer NAL, than CC2 offspring (both genders; P<0.01, as determined by one-way ANOVA). At weaning, the BM differences disappeared among the groups. At 6 months of age, male RC2 offspring were approx. 20% heavier than CC2, CR2 and RR2 offspring (P<0.001, as determined by one-way ANOVA). Females did not have any difference in BM among the groups. BM was affected by the interaction between gender and gestation diet, gender and lactation diet, and gestation diet and lactation diet (Table 3).

Comparing F1 and F2 generations at weaning, both genders of CR2 (100% more), RC2 (20% more) and RR2 (200% more) offspring were heavier than the respective F1 groups (P<0.001, as determined by one-way ANOVA). At 6 months of age, male CR2 (15% more), RC2 (20% more) and RR2 (25% more) offspring were heavier than the corresponding F1 groups (P<0.001, as determined by one-way ANOVA). Female CR2 offspring were 20% heavier than female CR1 offspring (P<0.01, as determined by one-way ANOVA). No difference in NAL was observed among the groups.

Growth rate (Table 4)

When compared with the F1 offspring, F2 offspring did not have any difference in growth rates at weaning among the groups, but they were different from the corresponding F1 offspring, with the exception of female CC2 offspring. Starting in the post-weaning period, differences among the groups were distinct in male RC2 offspring at 1–2 months of age and 2–3 months of age.

Fat mass, and glucose, insulin and leptin levels (Table 6)

Fat mass

Male CR2 offspring had 35% more fat mass than CC2 offspring (P<0.01, as determined by one-way ANOVA) and 75% more fat mass than RR2 offspring (P<0.0001, as determined by one-way ANOVA). RC2 offspring had 45% more fat mass than RR2 offspring (P<0.01, as determined by one-way ANOVA). Females had comparable results.

Table 6
Hormone concentrations, HOMA and percentage fat mass in 6-month-old F2 offspring

Values are means±S.E.M. Blood was collected from the right atrium by cardiac puncture. Significances between groups were determined by one-way ANOVA, followed by Tukey’s post-hoc test. *Significantly different from the F1 group; †significantly different from CC offspring; ‡significantly different from CR offspring; §significantly different from RC offspring; ∥significantly different from the corresponding male offspring; For the ANOVA, one-, two- or three-way ANOVA was used as appropriate. G, gender, GD, gestational diet; LD, lactation diet; ns, not significant.

OffspringFat mass (%)Glucose (mmol/l)Insulin (pmol/l)HOMALeptin (ng/ml)
Male      
 CC2 2.6±0.1 5.1±0.2 65.3±3.9 16.0±4.3 1.2±0.1 
 CR2 3.5±0.2† 7.5±0.3†* 171.9±8.4†* 57.3±3.1†* 2.7±0.1†* 
 RC2 2.9±0.3* 6.3±0.2†‡ 248.4±32.0†‡* 70.2±11.3†* 1.6±0.1‡ 
 RR2 2.0±0.2‡§ 6.1±0.1†‡* 83.9±15.1‡§ 23.1±4.6‡§ 1.2±0.2‡ 
Female      
 CC2 2.7±0.3 4.9±0.02 57.2±3.5 12.4±0.8 1.0±0.2 
 CR2 4.4±0.4†∥ 6.4±0.1†∥* 107.3±19.6∥ 30.1±5.3†∥* 1.8±0.2†∥ 
 RC2 3.2±0.3‡* 6.4±0.6†* 77.0±17.2∥ 20.5±2.9†∥* 1.7±0.1†* 
 RR2 2.2±0.2‡ 5.6±0.3 74.5±14.0 18.2±2.9 0.9±0.1‡§ 
ANOVA      
 G effect P<0.05 P<0.05 P<0.0001 P<0.0001 P<0.001 
 GD effect ns ns ns ns P<0.001 
 LD effect P<0.001 P<0.0001 ns ns P<0.01 
 G×GD interaction ns ns ns ns P<0.01 
 G×LD interaction ns ns P<0.05 ns P<0.001 
 GD×LD interaction P<0.001 P<0.0001 P<0.0001 P<0.0001 P<0.0001 
 G×GD×LD interaction ns ns P<0.0001 P<0.0001 ns 
OffspringFat mass (%)Glucose (mmol/l)Insulin (pmol/l)HOMALeptin (ng/ml)
Male      
 CC2 2.6±0.1 5.1±0.2 65.3±3.9 16.0±4.3 1.2±0.1 
 CR2 3.5±0.2† 7.5±0.3†* 171.9±8.4†* 57.3±3.1†* 2.7±0.1†* 
 RC2 2.9±0.3* 6.3±0.2†‡ 248.4±32.0†‡* 70.2±11.3†* 1.6±0.1‡ 
 RR2 2.0±0.2‡§ 6.1±0.1†‡* 83.9±15.1‡§ 23.1±4.6‡§ 1.2±0.2‡ 
Female      
 CC2 2.7±0.3 4.9±0.02 57.2±3.5 12.4±0.8 1.0±0.2 
 CR2 4.4±0.4†∥ 6.4±0.1†∥* 107.3±19.6∥ 30.1±5.3†∥* 1.8±0.2†∥ 
 RC2 3.2±0.3‡* 6.4±0.6†* 77.0±17.2∥ 20.5±2.9†∥* 1.7±0.1†* 
 RR2 2.2±0.2‡ 5.6±0.3 74.5±14.0 18.2±2.9 0.9±0.1‡§ 
ANOVA      
 G effect P<0.05 P<0.05 P<0.0001 P<0.0001 P<0.001 
 GD effect ns ns ns ns P<0.001 
 LD effect P<0.001 P<0.0001 ns ns P<0.01 
 G×GD interaction ns ns ns ns P<0.01 
 G×LD interaction ns ns P<0.05 ns P<0.001 
 GD×LD interaction P<0.001 P<0.0001 P<0.0001 P<0.0001 P<0.0001 
 G×GD×LD interaction ns ns P<0.0001 P<0.0001 ns 

Fat mass varied significantly with gender and lactation diet (Table 6). Gestation diet and lactation diet interacted influencing fat mass (Table 6), indicating that the diet of the original F0 mothers influenced fat mass in the F2 generation. Comparing F1 and F2 generations, there was a difference between male and female RC offspring. The fat mass was 25% less in male RC2 offspring compared with the corresponding RC1 offspring (P<0.01, as determined by one-way ANOVA), and 30% less in female RC2 offspring compared with the corresponding RC1 offspring (P<0.05, one-way ANOVA).

Glucose

Fasting glucose levels varied significantly with gender and lactation diet (Table 6). The gestation diet and lactation diet interacted altering the fasting glucose (Table 6). Compared with CC2 offspring, fasting glucose was increased by 50% in CR2 offspring, 25% in RC2 offspring and 20% in RR2 offspring (P<0.0001, as determined by one-way ANOVA). The plasma glucose level in CR2 offspring was 20% more than both the RC2 and RR2 offspring (P<0.001, as determined by one-way ANOVA). Females had corresponding results.

Comparing F1 and F2 generations, we observed differences in both genders of CR2 and RR2 offspring. The glucose levels were increased 50% in male CR2 offspring (P<0.0001, as determined by one-way ANOVA) and 15% in male RR2 offspring (P<0.001, as determined by one-way ANOVA) compared with matched F1 offspring. The glucose levels increased 45% in female CR2 offspring (P<0.0001, as determined by one-way ANOVA) and 15% in female RC2 offspring (P<0.05, as determined by one-way ANOVA).

Insulin

Fasting insulin levels increased significantly, with 160% more in CR2 offspring and 280% in RC2 offspring compared with CC2 offspring (P<0.0001, as determined by one-way ANOVA). Moreover, fasting plasma insulin levels were increased 100% in CR2 offspring compared with RR2 offspring (P<0.001, as determined by one-way ANOVA) and 200% in RC2 offspring compared with RR2 offspring (P<0.0001, as determined by one-way ANOVA) (Table 6).

Interactions were found to influence fasting insulin levels between gender and lactation diet, and between gestation diet and lactation diet (Table 6). ANOVA revealed a significant three-factor interaction between gender, gestation diet and lactation diet (Table 6), reinforcing that either diet fed to the original F0 mothers altered plasma insulin in the F2 generation, and that male offspring were altered more than female offspring.

Comparing F1 and F2 generations, fasting insulin levels varied between male CR and RC offspring. It was 100% higher in male CR2 offspring (P<0.001, as determined by one-way ANOVA) and 120% higher in RC2 offspring (P<0.001, as determined by one-way ANOVA) compared with the corresponding F1 offspring. Females did not show any difference among the groups.

HOMA

HOMA indicated an insulin-resistant state in male CR2 and RC2 offspring compared with CC2 offspring (P<0.001, as determined by one-way ANOVA). CR2 and RC2 offspring had an insulin-resistant state compared with RR2 offspring (P<0.001, as determined by one-way ANOVA). In females, HOMA was significantly higher in CR2 (P<0.001) and RC2 (P<0.05) offspring compared with CC2 offspring (as determined by one-way ANOVA) (Table 6).

We observed an interaction between gestation diet and lactation diet in the original F0 mothers, and an interaction between gender, gestation diet and lactation diet (Table 6) altering HOMA in the F2 generation. These results indicate that the male offspring were more sensitive than females, as the peripheral insulin-resistant state may be a consequence of the diet of original F0 mothers.

Comparing F1 and F2 generations, HOMA values were significantly higher in both genders of CR2 and RC2 offspring compared with the corresponding F1 offspring (P<0.001, as determined by one-way ANOVA).

Leptin

Circulating plasma leptin concentration was highest in both genders of CR2 offspring. In male CR2 offspring, it was increased 120% more than CC2 offspring, 70% more than RC2 offspring, and 125% more than RR2 offspring (P<0.001, as determined by one-way ANOVA). In females, the plasma leptin concentration did not show any difference between CR2 and RC2 offspring, but, in CR2 offspring, it was increased 80% more than CC2 offspring and 100% more than RR2 offspring (P<0.001, as determined by one-way ANOVA) (Table 6).

The plasma leptin varied significantly with gender, gestation diet or lactation diet (Table 6). Moreover, there were significant interactions between gender and gestational diet, between gender and lactation diet, and between gestation diet and lactation diet (Table 6), indicating that males were altered more than females and that hyperleptinaemia may be a consequence of the diet of the original F0 mothers.

Comparing F1 and F2 generations, we observed differences in plasma leptin levels between male CR2 and CR1 offspring (145% more in CR2 offspring; P<0.0001, as determined by one-way ANOVA). In females, we observed differences between RC2 and RC1 offspring (55% higher in RC2; P<0.001, as determined by one-way ANOVA).

DISCUSSION

In the present study, the lower birthweight of the offspring of mothers fed on a restricted (low-protein) diet became apparent at birth, as both genders of the R1 offspring had significantly lower birthweight. Moreover, changes in growth and BM were not limited to low protein during gestational period. Postnatal nutrition appeared to play a major role in determining subsequent growth. This was reflected by the fact that the postnatal growth either faltered or accelerated irrespective of the BM at day 21, and it was maintained through life. As evidence of this, RC1 animals had rapid catch-up during lactation, reaching the BM of the CC1 offspring within 21 days, whereas postnatal growth retardation was evident in both genders of CR1 and RR1 offspring. The poor fetal growth followed by rapid postnatal catch-up amplifies the risk of developing chronic diseases and is associated with decreased average longevity in mice [20]. These results demonstrate that maternal nutrition during critical periods of development has a major impact on longevity as well as quality of life [21]. Catch-up growth immediately after early malnutrition could be a key point for the programming of obesity in later life [22].

Previous studies have shown that the major adverse effect on post-weaning weight resulted from restriction of maternal protein intake during lactation [6]. In the present study, return to an optimal diet at weaning did not correct the body deficits in CR1 and RR1 offspring, even though these groups had a greater growth-rate curve within the first 3 months. Therefore our results agree with previous suggestions that the postnatal period, rather than the prenatal period, is critical for overall growth [7,13].

In the present study, nutrient restriction to which F1 mothers themselves were exposed was transferred across to the next generation of their offspring. Experimental evidence has demonstrated that a low-protein diet in early, middle or late gestation resulted in a relative deficiency in pancreatic β-cells following birth, due to a failure to develop larger islets, with females being particularly susceptible in mid-gestation and males in late gestation [23]. The adult offspring of the mothers fed a low-protein diet had deficient β-cell mass and low pancreatic insulin content, resulting in a latent gestational diabetogenic tendency [24]. This latent diabetogenic tendency and the metabolic stress of pregnancy may result in gestational diabetes; the hyperglycaemic intrauterine milieu of mothers with diabetes stimulates the fetal endocrine pancreas to increase insulin production and accelerated anabolism, resulting in fetal and neonatal macrosomia in the next generation [11]. In the F2 generation in the present study, we observed macrosomia in the RC2 offspring, which led to a persistent high BM in the male group throughout the experiment, even though no differences in growth rate were shown. This transgenerational effect on the F2 generation, from the diabetic mother on the F2 and even F3 generation, fetus and adults is only transmitted via the maternal line: female offspring of mothers fed the restricted-protein diet developed gestational diabetes and induced the effect on their fetuses, and thereby in the next generation. Male offspring have impaired glucose tolerance, but do not transmit the effect to their offspring [25].

Besides the metabolic alterations observed in F1 females while they were in utero, transgenerational epigenetic inheritance could explain the transgenerational effects observed. Nutrition might induce, at some loci, epigenetic or other changes that could be transmitted to the next generation affecting health [26]. Therefore transgenerational response to the nutrition of ancestors is now considered the main influence on longevity [27].

The best-characterized epigenetic modification of DNA is the methylation of cytosine residues. Early nutrition may influence the establishment and maintenance of cytosine methylation [4]. It has been shown that the epigenetic state, and associated phenotype, may lead to programmed alterations in metabolism and can be inherited transgenerationally, resulting in some memory of the epigenetic state persisting in the next generation [28]. Another hypothesis is the alteration of the mitochondrial DNA of the female fetus, resulting from the challenge experienced during development. If such mitochondrial DNA changes occur in the female gametes, they would be passed to the offspring [13]. The environmental contamination by endocrine-disrupting chemicals that represent an unappreciated force in sexual selection is a good example of the role of epigenetics as a determinant factor in evolution [29].

In the present study, we observed alterations in glucose metabolism in both F1 and F2 generations. Maternal F0 protein restriction during gestation had a more pronounced hyperglycaemic effect on offspring glucose metabolism in the F1 male generation than restriction during lactation. Interestingly, maternal F0 protein restriction during gestation as well as during lactation had a similar hyperglycaemic effect on offspring glucose metabolism in the F2 males and females. Restriction of dietary protein during development in rats results in a reduction in β-cell mass, insulin content in the offspring and glucose intolerance in adulthood, and in early insulin hypersensitivity followed by late insulin resistance [23]. Insulin hypersensitivity gives way to resistance earlier in male than in female rats [30]. The male RC1 offspring in the present study had high levels of glucose and insulin resistance. HOMA reinforces that males were more affected than females and that the peripheral insulin-resistant state may be a consequence of early growth restriction followed by rapid catch-up growth. On the contrary, we observed normal glycaemia with normal insulin sensitivity in both genders of offspring from mothers fed a protein-restricted diet during lactation. The increased HOMA in the male RC1 offspring when compared with RR1 and CR1 offspring also supports the view that glucose regulatory mechanisms are at least, to some extent, programmed by the level of protein availability prenatally. However, other reports have shown low glycaemia and high insulin sensitivity in adult offspring whose mothers were submitted to protein restriction during lactation [7,31]. This apparent discrepancy between these studies and the present findings can be explained, as the previous studies used a moderate protein-restriction diet, whereas we have used a severe protein-restriction diet. A severe protein-restriction diet is known to enhance the offspring response, displaying gender dimorphism when it exists [32]. In addition, rats have an age-dependent impairment in glucose tolerance (by 15 months of age, they have marked worsening glucose tolerance and, by 17 months, the male low-protein offspring are diabetic with elevated insulin levels) [33]. Normal glucose tolerance in the presence of low insulin secretion in juvenile animals could be due to an adaptation of peripheral tissue through an increased number of insulin receptors in the liver, adipose tissue and muscle and through other peripheral adaptations [34]. The transgenerational presence of aberrant glucose/insulin metabolism and skeletal muscle insulin signalling in the adult female F2 offspring from dams fed a protein-restricted diet supports nutritionally induced heritable mechanisms contributing to the epidemic of Type 2 diabetes mellitus [35].

Altered maternal/fetal metabolism appears to be associated with a diabetogenic effect in adult offspring and some adaptations later in life may take place, but stress situations such as pregnancy and aging precipitate the animals to glucose intolerance and insulin resistance. The next generation could be affected by the impairment in the development of the endocrine pancreas [25]. In the present study, it is noteworthy that F1 females did not have glucose intolerance during their own pregnancies, but glucose metabolism in the F2 generation was more affected, with offspring whose mothers were exposed to a low-protein diet during perinatal life having high glucose levels and high insulin levels with consequent insulin resistance. The findings indicate that adverse second-generation effects can follow protein restriction in both pregnancy and lactation. Although, when protein restriction persisted in both periods (RR2), F2 offspring did not have any alterations in glucose metabolism.

Adipocyte development begins in the fetus and, in contrast with all other tissues whose growth ceases in late juvenile life, it has the capacity for ‘unlimited’ growth [36]. In normal healthy individuals, the increase in fat mass with age is accompanied by a parallel increase in cortisol sensitivity, i.e. increased glucocorticoid receptor abundance and increased activity of the enzyme 11β-HSD1 (type 1 11β-hydroxysteroid dehydrogenase) [37]. Enhanced adipocyte sensitivity to cortisol is promoted in offspring born to mothers that were nutrient-restricted in utero in conjunction with increased PPARα (peroxisome-proliferator-activated-receptor α). This adaptation only appears to be associated with greater fat mass in offspring when maternal nutrient restriction is confined to late gestation, coincidental with the period of maximal fetal growth [38]. In the present study, it appeared that maternal protein restriction had a greater effect on male and female F1 offspring when restriction occurred during pregnancy (RC1), and increased fat mass was accompanied by glucose intolerance and insulin resistance observed in these animals. The increased fat mass in the male and female CR2 offspring when compared with CC2 offspring indicates that maternal protein restriction had a greater effect on male and female F2 offspring when restriction occurred during lactation.

Adipose-tissue-derived leptin is pivotal in the central nervous control of appetite and energy balance [39]. As expected, our present results showed that leptin concentrations correlated with the percentage of fat mass in both generations. Increased leptin concentrations were observed in male RC1 offspring, male and female CR2 offspring and female RC2 offspring. These findings indicate that adverse second-generation effects can follow protein restricted in both pregnancy and lactation. Differences in leptin concentrations have been observed in human studies and other examples of developmental programming. In humans, leptin levels increase throughout gestation, peak in the second trimester and remain high until delivery. These elevated leptin levels are not associated with negative energy balance as might be expected, perhaps because leptin resistance in pregnancy sustains maternal food intake for fetal growth [40]. An early peak in neonate leptin in rodents may play a role in programming appetite in later life [41]. Leptin levels in neonates decrease rapidly following birth, and this rapid decline may be a stimulus for feeding behaviour in early life, explaining why growth-restricted babies with low leptin levels at birth have increased growth rates in early neonatal life [33]. Postnatal leptin levels, which are predominantly regulated by nutrition, also play a key role in the permanent effects on adiposity and body composition [42]. The leptin concentrations measured in RC1 offspring suggest that dietary influences during development may be involved in leptin sensitivity in F1 and F2 generations.

In conclusion, low protein during gestation improves BM, fat mass and growth rate in F1 rats and has adverse effects on glucose and leptin metabolism, resulting in insulin resistance in the adult F1 and F2 offspring. Low-protein during lactation has adverse effects on glucose, insulin and leptin metabolism, resulting in insulin resistance in adult F2 offspring. These findings suggest that low protein during gestation and/or lactation can be passed transgenerationally to the second generation.

Abbreviations

     
  • BM

    body mass

  •  
  • C1

    F1 from dams fed the control diet in gestation

  •  
  • CC1 (CC2)

    F1 (F2) from dams fed the control diet in both gestation and lactation

  •  
  • CR1 (CR2)

    F1 (F2) from dams fed the control diet during gestation and the restricted diet during lactation

  •  
  • HOMA

    homoeostasis model assessment

  •  
  • NAL

    naso-anal length

  •  
  • RC1 (RC2)

    F1 (F2) from dams fed the restricted diet during gestation and the control diet during lactation

  •  
  • R1

    F1 from dams fed the restricted diet in gestation

  •  
  • RR1 (RR2)

    F1 (F2) offspring from dams fed the restricted diet in both gestation and lactation

We thank Mrs Thatiany S. Marinho for her technical assistance. This study was partially supported by CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) and FAPERJ (Fundação de Amparo à Pesquisa do Estado de Rio de Janeiro).

References

References
1
Barker
 
D. J.
 
In utero programming of cardiovascular disease
Theriogenology
2000
, vol. 
53
 (pg. 
555
-
574
)
2
Frasch
 
M. G.
Muller
 
T.
Wicher
 
C.
, et al 
Fetal body weight and the development of the control of the cardiovascular system in fetal sheep
J. Physiol.
2007
, vol. 
579
 (pg. 
893
-
907
)
3
Vuguin
 
P. M.
 
Animal models for small for gestational age and fetal programming of adult disease
Horm. Res.
2007
, vol. 
68
 (pg. 
113
-
123
)
4
Waterland
 
R. A.
Jirtle
 
R. L.
 
Early nutrition, epigenetic changes at transposons and imprinted genes, and enhanced susceptibility to adult chronic diseases
Nutrition
2004
, vol. 
20
 (pg. 
63
-
68
)
5
Almeida
 
J. R.
Mandarim-de-Lacerda
 
C. A.
 
Overweight is gender-dependent in prenatal protein–calorie restricted adult rats acting on the blood pressure and the adverse cardiac remodeling
Life Sci.
2005
, vol. 
77
 (pg. 
1307
-
1318
)
6
Pires
 
K. M. P.
Aguila
 
M. B.
Mandarim-de-Lacerda
 
C. A.
 
Early renal structure alteration in rat offspring from dams fed low protein diet
Life Sci.
2006
, vol. 
79
 (pg. 
2128
-
2134
)
7
Zambrano
 
E.
Bautista
 
C. J.
Deas
 
M.
, et al 
A low maternal protein diet during pregnancy and lactation has sex- and window of exposure-specific effects on offspring growth and food intake, glucose metabolism and serum leptin in the rat
J. Physiol.
2006
, vol. 
571
 (pg. 
221
-
230
)
8
Skinner
 
M. K.
 
Endocrine disruptors and epigenetic transgenerational disease etiology
Pediatr. Res.
2007
, vol. 
61
 (pg. 
48R
-
50R
)
9
Gluckman
 
P. D.
Hanson
 
M. A.
Beedle
 
A. S.
 
Non-genomic transgenerational inheritance of disease risk
BioEssays
2007
, vol. 
29
 (pg. 
145
-
154
)
10
Kaput
 
J.
Rodriguez
 
R. L.
 
Nutritional genomics: the next frontier in the postgenomic era
Physiol. Genomics
2004
, vol. 
16
 (pg. 
166
-
177
)
11
Aerts
 
L.
Van Assche
 
F. A.
 
Intra-uterine transmission of disease
Placenta
2003
, vol. 
24
 (pg. 
905
-
911
)
12
Drake
 
A. J.
Walker
 
B. R.
Seckl
 
J. R.
 
Intergenerational consequences of fetal programming by in utero exposure to glucocorticoids in rats
Am. J. Physiol. Regul. Integr. Comp. Physiol.
2005
, vol. 
288
 (pg. 
R34
-
R38
)
13
Zambrano
 
E.
Martinez-Samayoa
 
P. M.
Bautista
 
C. J.
, et al 
Sex differences in transgenerational alterations of growth and metabolism in progeny (F2) of female offspring (F1) of rats fed a low protein diet during pregnancy and lactation
J. Physiol.
2005
, vol. 
566
 (pg. 
225
-
236
)
14
Jirtle
 
R. L.
Skinner
 
M. K.
 
Environmental epigenomics and disease susceptibility
Nat. Rev. Genet.
2007
, vol. 
8
 (pg. 
253
-
262
)
15
Reeves
 
P. G.
Nielsen
 
F. H.
Fahey
 
G. C.
 
AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet
J. Nutr.
1993
, vol. 
123
 (pg. 
1939
-
1951
)
16
Langley
 
S. C.
Jackson
 
A. A.
 
Increased systolic blood pressure in adult rats induced by fetal exposure to maternal low protein diets
Clin. Sci.
1994
, vol. 
86
 (pg. 
217
-
222
)
17
Wainwright
 
P. E.
 
Issues of design and analysis relating to the use of multiparous species in developmental nutritional studies
J. Nutr.
1998
, vol. 
128
 (pg. 
661
-
663
)
18
Jolicoeur
 
P.
Heusner
 
A. A.
 
Log-normal variation belts for growth curves
Biometrics
1986
, vol. 
42
 (pg. 
785
-
794
)
19
Sokal
 
R. R.
Rohlf
 
F. J.
 
The Principles and Practice of Statistics in Biological Research
Biometry
1995
3rd ed.
New York
Freeman
pg. 
887
 
20
Ozanne
 
S. E.
Hales
 
C. N.
 
Lifespan: catch-up growth and obesity in male mice
Nature
2004
, vol. 
427
 (pg. 
411
-
412
)
21
Ozanne
 
S. E.
Hales
 
C. N.
 
Poor fetal growth followed by rapid postnatal catch-up growth leads to premature death
Mech. Ageing Dev.
2005
, vol. 
126
 (pg. 
852
-
854
)
22
Bieswal
 
F.
Ahn
 
M. T.
Reusens
 
B.
, et al 
The importance of catch-up growth after early malnutrition for the programming of obesity in male rat
Obesity
2006
, vol. 
14
 (pg. 
1330
-
1343
)
23
Chamson-Reig
 
A.
Thyssen
 
S. M.
Arany
 
E.
Hill
 
D. J.
 
Altered pancreatic morphology in the offspring of pregnant rats given reduced dietary protein is time and gender specific
J. Endocrinol.
2006
, vol. 
191
 (pg. 
83
-
92
)
24
Dahri
 
S.
Reusens
 
B.
Remacle
 
C.
Hoet
 
J. J.
 
Nutritional influences on pancreatic development and potential links with non-insulin-dependent diabetes
Proc. Nutr. Soc.
1995
, vol. 
54
 (pg. 
345
-
356
)
25
Aerts
 
L.
Van Assche
 
F. A.
 
Animal evidence for the transgenerational development of diabetes mellitus
Int. J. Biochem. Cell. Biol.
2006
, vol. 
38
 (pg. 
894
-
903
)
26
Csaba
 
G.
Kovacs
 
P.
Pallinger
 
E.
 
Transgenerational effect of neonatal vitamin A or D treatment (hormonal imprinting) on the hormone content of rat immune cells
Horm. Metab. Res.
2007
, vol. 
39
 (pg. 
197
-
201
)
27
Kaati
 
G.
Bygren
 
L. O.
Pembrey
 
M.
Sjostrom
 
M.
 
Transgenerational response to nutrition, early life circumstances and longevity
Eur. J. Hum. Genet.
2007
, vol. 
15
 (pg. 
784
-
790
)
28
Rakyan
 
V.
Whitelaw
 
E.
 
Transgenerational epigenetic inheritance
Curr. Biol.
2003
, vol. 
13
 pg. 
R6
 
29
Crews
 
D.
Gore
 
A. C.
Hsu
 
T. S.
, et al 
Transgenerational epigenetic imprints on mate preference
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
5942
-
5946
)
30
Fernandez-Twinn
 
D. S.
Wayman
 
A.
Ekizoglou
 
S.
Martin
 
M. S.
Hales
 
C. N.
Ozanne
 
S. E.
 
Maternal protein restriction leads to hyperinsulinemia and reduced insulin-signaling protein expression in 21-mo-old female rat offspring
Am. J. Physiol. Regul. Integr. Comp. Physiol.
2005
, vol. 
288
 (pg. 
R368
-
R373
)
31
Moura
 
A. S.
Caldeira Filho
 
J. S.
Mathias
 
P. C.
de Sa
 
C. C.
 
Insulin secretion impairment and insulin sensitivity improvement in adult rats undernourished during early lactation
Res. Commun. Mol. Pathol. Pharmacol.
1997
, vol. 
96
 (pg. 
179
-
192
)
32
Woods
 
L. L.
Ingelfinger
 
J. R.
Rasch
 
R.
 
Modest maternal protein restriction fails to program adult hypertension in female rats
Am. J. Physiol. Regul. Integr. Comp. Physiol.
2005
, vol. 
289
 (pg. 
R1131
-
R1136
)
33
Stocker
 
C. J.
Arch
 
J. R.
Cawthorne
 
M. A.
 
Fetal origins of insulin resistance and obesity
Proc. Nutr. Soc.
2005
, vol. 
64
 (pg. 
143
-
151
)
34
Holness
 
M. J.
Langdown
 
M. L.
Sugden
 
M. C.
 
Early-life programming of susceptibility to dysregulation of glucose metabolism and the development of Type 2 diabetes mellitus
Biochem. J.
2000
, vol. 
349
 (pg. 
657
-
665
)
35
Thamotharan
 
M.
Garg
 
M.
Oak
 
S.
, et al 
Transgenerational inheritance of the insulin-resistant phenotype in embryo-transferred intrauterine growth-restricted adult female rat offspring
Am J. Physiol. Endocrinol. Metab.
2007
, vol. 
292
 (pg. 
E1270
-
E1279
)
36
Gimeno
 
R. E.
Klaman
 
L. D.
 
Adipose tissue as an active endocrine organ: recent advances
Curr. Opin. Pharmacol.
2005
, vol. 
5
 (pg. 
122
-
128
)
37
Desbriere
 
R.
Vuaroqueaux
 
V.
Achard
 
V.
, et al 
11β-Hydroxysteroid dehydrogenase type 1 mRNA is increased in both visceral and subcutaneous adipose tissue of obese patients
Obesity
2006
, vol. 
14
 (pg. 
794
-
798
)
38
Budge
 
H.
Gnanalingham
 
M. G.
Gardner
 
D. S.
Mostyn
 
A.
Stephenson
 
T.
Symonds
 
M. E.
 
Maternal nutritional programming of fetal adipose tissue development: Long-term consequences for later obesity
Birth Defects Res. C Embryo Today
2005
, vol. 
75
 (pg. 
193
-
199
)
39
Kershaw
 
E. E.
Flier
 
J. S.
 
Adipose tissue as an endocrine organ
J. Clin. Endocrinol. Metab.
2004
, vol. 
89
 (pg. 
2548
-
2556
)
40
Seeber
 
R. M.
Smith
 
J. T.
Waddell
 
B. J.
 
Plasma leptin-binding activity and hypothalamic leptin receptor expression during pregnancy and lactation in the rat
Biol. Reprod.
2002
, vol. 
66
 (pg. 
1762
-
1767
)
41
Unger
 
R. H.
 
Lipotoxic diseases
Annu. Rev. Med.
2002
, vol. 
53
 (pg. 
319
-
336
)
42
Ehrhardt
 
R. A.
Greenwood
 
P. L.
Bell
 
A. W.
Boisclair
 
Y. R.
 
Plasma leptin is regulated predominantly by nutrition in preruminant lambs
J. Nutr.
2003
, vol. 
133
 (pg. 
4196
-
4201
)