Introduction
Obesity is a disease classically characterized by an excess of body weight and fat accumulation, strongly associated with an increased risk of metabolism diseases, leading to loss of disease-free years at adulthood and premature death. 1,Reference Nyberg, Batty and Pentti2 Epidemiological and experimental studies highlight the role of overnutrition, among other nutritional disturbances, in early developmental stages of life, increasing the susceptibility and prevalence of obesity and non-communicable diseases (NCDs) in adulthood. Reference de Almeida, Fabricio and Trombini3–Reference Plagemann, Harder and Schellong5
Excessive body weight is related to low aerobic fitness and lower values of maximum oxygen consumption during exercise (VO2max). Reference Bassett and Howley6 VO2max is considered as the best index available for assessing aerobic exercise capacity. Among different factors, visceral obesity, age, training status, and peripheral insulin resistance have been shown to affect VO2max. Reference Nyholm, Nielsen and Kristensen7
Moderate exercise, with an intensity of 55%–65% of VO2max, for more than 30 min per session is able to promote high levels of mobilization of fat stores and other important peripheral adaptations that contribute to the loss of body fat and improve tissue sensitivity to insulin. Reference Achten and Jeukendrup8 In human and animal models, moderate physical training has been classically identified as a valuable non-pharmacological intervention to mitigate the effects of obesity on body weight, fat accumulation, and blood glucose homeostasis. Reference Scomparin, Grassiolli and Gomes9,Reference Sung and Bae10
The literature suggests that early postnatal overfeeding (PO) induces persistent changes in energy expenditure and decreases spontaneous physical activity. Reference Zhu, Eclarinal and Baker11 Although the effects of physical exercise on obese subjects have been widely studied, there is little data on how these individuals adapt to exercise or their physical fitness, especially when it comes to developmental obesity. As early overfeeding in rats promotes long-term obesity and unmistakable metabolic disorders, it is possible that it also affects the adaptation to physical exercise, reducing the benefits of physical training interventions. In this way, the present study aimed to investigate the effects of early PO on body weight, blood glucose, and physical fitness of adult male Wistar rats submitted to a moderate intensity with low-frequency training program.
Methods
Animals
Seventy-day-old female and 80-day-old male Wistar rats were obtained from the Central Animals Facility (Maringa State University) and housed in the Sectoral Animal Facility under standard conditions of light (12 h dark/12 h light cycle) and temperature (22°± 2°C). Animals had ad libitum access to standard commercial chow (Nuvital®, Curitiba, Brazil) and water through all the experimental periods. After 5 days of adaptation, three females and one male were placed in a breeding cage. Pregnant rats (eight for NL and eight for SL groups) were placed in individual cages until natural delivery (postnatal day zero – PN0). Considering the issue of sexual dimorphism and that litter size manipulation has been shown to have significant effects on male rats compared to females, Reference Bassett and Craig12 in the present study, only male rats were used. On PN2, the number of pups was adjusted to nine per dam for normal litters (NL) and three pups per dam for small litters (SL) to induce early overfeeding (Fig. 1a). On PN21, animals were weaned, group housed (three per cage for all groups), and randomly assigned to either, exercise or sedentary groups: NL sedentary (NL-Se), NL exercised (NL-Ex), SL sedentary (SL-Se), and SL exercised (SL-Ex). After weaning, each group had a number of 12 animals from 4 litters.
Effects of early postnatal overfeeding on animals’ adaptation to exercise: (a) Experimental timeline, (b) relative values of VO2max at PN30, (c) relative values of VO2max at PN90, (d) Creatine Kinase (CK) activity at PN90 and (e) Lactate Dehydrogenase (LDH) activity at PN91.
Training protocol
Adaptation to treadmill training was performed from PN22 to PN28, in short sessions (from 10 to 22 min) at low intensity (16 cm/s). Exercise training was performed on a treadmill proper for rodents (LE8710R 76–0308; rat 5-lane; Panlab, Harvard Apparatus®). On PN30, NL and SL animals were submitted to an effort test to determine VO2max, using a gas analyzer coupled to an individual rodent treadmill (Panlab, Harvard Apparatus®, Cornellà- Barcelona – Spain). The test started with a warm-up (5 min/10 cm/s/0º of inclination), followed by an increase in speed of 9 cm/s every 3 min until the animal’s exhaustion, determined by their inability to keep up with the treadmill. VO2max was considered as the value achieved when an increase in workload did not affect the consumption of O2 ± 5%. Reference Rodrigues, De Souza and Da Silva13 The final workload (FWL) was considered to be the workload where O2 consumption was not modified even as an increase in maximal speed occurred, being used to determine the training workload (TWL) for NL-Ex and SL-Ex rats. Exercised groups performed the effort test at PN30, PN45, PN60, and PN75 to adjust the TWL and monitoring physical progress and at PN90 to determine their final physical status. Sedentary rats were subjected to the effort test at PN30 and PN90 for standardization. Training sessions were performed three times per week for 40 min, with a progressive TWL from 55% to 65% of the FWL achieved in the previous effort test.
Biometric measures
Control and SL rats were weighed at weaning (PN21), and then body weight was assessed every 10 days from PN30 to PN90. On PN92, after a 48-hour interval for recovery from the last effort test, animals in the four groups that did not undergo surgery were weighed and euthanized (eight per group from four different litters). Retroperitoneal fat was dissected and weighed on a precision scale.
Intravenous glucose tolerance test (ivGTT) and HOMA-IR index
On PN91, four rats from four litters in each group were weighed and anesthetized (ketamine–xylazine, at doses of 3 mg and 0.6 mg/100 g of BW, respectively), and a silicone cannula was surgically implanted in the right jugular vein. Twenty-four hours after surgery (PN92), following an overnight fast of 12 h (20:00–8 h 00 min), the IvGTT was performed. The first blood sample (300–350 µl) collected through the cannula was considered time “0” (fasting glycemia and insulinemia). Then, a glucose bolus (1 g/kg/BW) was injected. Blood samples (300–350 µl) were collected at 5, 15, 30, and 45 min after glucose solution administration. Reference Venci, Ramos and Martins14 Blood glucose was assessed by the glucose oxidase method by spectrophotometer (Semi-Automatic Biochemical Analyzer, BIO 200FL, Bio Plus®, São Paulo/SP, Brazil) utilizing a commercial kit (Gold Analisa®, Belo Horizonte/MG, Brazil). Insulin concentrations were determined by an iodine-125-labeled insulin radioimmunoassay with human insulin as the standard and an antibody for rat insulin. Increases in total glycemia and insulinemia were calculated using the area under the curve (AUC) for the 45 min of the IvGTT. The glucose Homeostasis Model Assessment Insulin Resistance (HOMA-IR) index was calculated as previously described. Reference Pacini and Mari15
Creatine kinase (CK) and lactate dehydrogenase (LDH) assessment
Blood samples were collected from the animal’s tail vein for CK and LDH measurements in the four groups. To assess the short-term muscle damage, immediately after the last effort test (PN90), 75 µl of blood was withdrawn to determine the serum activity of CK (12 animals from 4 litters per group). On PN91, 75 µl of blood was collected from animals that did not undergo surgery to determine serum LDH activity (eight animals from four litters per group). Serum levels of enzymes were determined using a commercial kit (Gold Analisa, Belo Horizonte, MG, Brazil) by spectrophotometry at 340 nm in a bioanalyzer (BIOPLUS, Bio-200). The reaction was linear until 2000 U/L.
Statistical analysis
For body weight at weaning and the first effort test (PN30), before the exercise protocol, Student’s t-test was used to compare NL and SL groups. After the exercise protocol (PN90 and PN92), the results were compared by two-way analysis of variance and source of variation: Early PO, exercise (E), and interaction (I –PO vs. exercise). Tukey’s post hoc test was used to multiple comparisons between groups using the statistical program GraphPad Prism 6.01. A - value < 0.05 was considered significant.
Results
Effect of early postnatal overfeeding and exercise on body weight, fat accumulation, and glucose homeostasis
Table 1 shows the effect of PO on body weight at weaning and biometric parameters of sedentary and exercised animals on PN92. The PO by litter reduction increased the body weight on PN21 (NL 46.4 ± 0.31 vs. SL 56.9 ± 0.72, p < 0.0001) and in adult life (NL-Se 387.5 ± 11.60 vs. 402.6 ± 9.21, p PO < 0.05). Fat accumulation on PN92 also was increased by PO (NL-Se 1.1 ± 0.08 vs. SL-Se 1.6 ± 0.06, p PO < 0.001), while the exercise protocol reduced body weight (NL-Se 387.5 ± 11.60 vs. NL-Exe 346.0 ± 7.63, SL-Se 402.6 ± 9.21 vs. SL-Ex 382.8 ± 5.15, p E < 0.001) and retroperitoneal fat pad (NL-Se 1.1 ± 0.08 vs. NL-Exe 0.84 ± 0.07, SL-Se 1.6 ± 0.06 vs. SL-Ex 0.99 ± 0.07, p E < 0.001). The exercise effects on fat accumulation reflect a significant interaction between exercise and early PO (p I < 0.05).
Effect of early postnatal overfeeding and exercise on body weight, fat accumulation, and glucose homeostasis
Values are the mean ± SEM. AUC, area under the curve; BW, body weight; E, exercise factor; I, interaction factor, NL, normal litter; ns, not significant; PO, factor postnatal overfeeding; SL, small litter.
P-values indicate the level of statistical significance (P < 0.05; P < 0.01; P < 0.001; P < 0.0001) of postnatal overfeeding based on Student’s t-test (PN21) and postnatal overfeeding, exercise versus sedentary treatment and the interaction among the source of variations on two-way ANOVA (PN92).
* P < 0.05, ** P < 0.01, *** P < 0.001 indicates the statistical significance of differences between NL and SL sedentary groups.
# P < 0.05, ## P < 0.01, and ### P < 0.001 indicate the statistical significance of exercise versus sedentary treatment for groups from the same litter size, based on Tukey’s multiple comparisons posttest.
Table 1 also shows the effects of PO and exercise on glucose homeostasis on PN92. While fasting glycemia was not altered by PO or exercise, fasting insulinemia was affected by both treatments (NL-Se 0.124 ± 0.01 vs. SL-Se 0.338 ± 0.01, p PO < 0.0001, SL-Se 0.338 ± 0.01 vs. SL-Ex 0.123 ± 0.01, p E < 0.0001) with a significant interaction (p I < 0.0001). The AUC for the glucose during the ivGTT showed the main effect of PO (NL-Se 565.6 ± 10.40 vs. SL-Se 615.1 ± 5.31, p PO < 0.001). Exercise attenuated AUC for glycemia only in SL animals (SL-Se 615.1 ± 5.31 SL-Ex 566.3 ± 8.30, p E < 0.001) with significant interaction (p I < 0.05). The blood insulin AUC during the ivGTT demonstrates the main effect of PO (NL-Se 0.59 ± 0.03 vs. SL-Se 1.88 ± 0.18, p PO < 0.0001). A significant interaction (p I < 0.05) is observed, and the posttest indicates a difference between SL groups (SL-Se 1.88 ± 0.18 vs. SL-Ex 1.36 ± 0.13, P < 0.05). The HOMA-IR was affected by PO (NL-Se 0.44 ± 0.12 vs. SL-Se 1.25 ± 0.10, p PO < 0.01) and exercise (SL-Se 1.25 ± 0.10 vs. SL-Ex 0.47 ± 0.16, p E < 0.05) in SL animals, leading to a significant interaction (p I < 0.05).
Effects of early postnatal overfeeding and exercise on animals’ adaptation to exercise
On PN30 (Fig. 1b), no difference is observed in the VO2max between NL and SL rats (P = 0.74). In adult animals (Fig. 1c – PN90), the moderate-intensity and low-frequency exercise induced a significant improvement in VO2max (p E < 0.001), but only in NL animals (posttest results: NL-Se 19.9 ± 0.9 vs. NL-Ex 26.2 ± 1.0, P < 0.0001; SL-Se 21.6 ± .01 vs. SL-Ex 22.9 ± 0.9, P = 0.90), leading to a significant interaction in exercise vs. PO (p I < 0.05).
On PN90 (Fig. 1d), exercised animals showed reduced CK activity (NL-Se 265,8 ± 11.2 vs. NL-Ex 233.1 ± 12.5, SL-Se 282.1 ± 15.4 vs. SL-Ex 237.7 ± 20.0, p E < 0.05) in both, control and postnatal overfed animals. Exercise reduced LDH activity (Fig. 1e) in NL animals (NL-Se 722.8 ± 96.2 vs. NL-Ex 647 ± 32.0, p E < 0.05), while early PO increased the LDH activity, with greater effect in SL-Ex animals when compared with all other groups (NL-Se 722.8 ± 96.2 vs. NL-Ex 647 ± 32.0 vs. SL-Se 705.1 ± 85.7 vs. SL-Se 1200 ± 174.4, p PO < 0.05), resulting in a significant interaction in PO vs. exercise (p I < 0.01).
Discussion
The lactation period is a sensitive stage of development and early overnutrition has been shown to induce obesity with further impairment of metabolism. Reference Plagemann, Harder and Schellong5 As expected, in the present study, the early PO had a negative impact on the biometric parameters and glucose metabolism of adult male Wistar rats. The moderate intensity with low-frequency exercise training was able to reduce BW and retroperitoneal fat accumulation and ameliorate glucose homeostasis, with greater effect in SL rats. Nevertheless, the present results also show, for the first time, that early PO can mitigate the positive gains of moderate exercise on the animals’ physical fitness, as the SL-Ex rats exhibited lower VO2max during exercise and higher levels of post-exercise LDH.
The effects of exercise depend on the duration, frequency, and intensity of training. Reference Moghetti, Bacchi and Brangani16 Here, the positive effects of moderate-intensity exercise on BW and retroperitoneal fat accumulation were observed mainly in PO animals. Previous studies have shown that the overweight and increased fat accumulation is often related to an altered blood lipid profile, mainly elevated triglycerides and total cholesterol. Reference Habbout, Li and Rochette17 Moderate exercise is shown to increase fat oxidation and improve the lipid profile in humans and in rodent models. Reference Achten and Jeukendrup8,Reference Scomparin, Grassiolli and Gomes9 Although the present limitation for not evaluating the animal’s lipid parameters, the effects of exercise reported on blood lipids may underlie the improvements observed here in BW and in the accumulation of body fat.
The SL-Ex rats’ biometric changes were accompanied by reduced fasting insulinemia, HOMA-IR index, and AUC for blood glucose in the ivGTT compared to SL-Se. Such changes were not observed between NL groups. This improvement was previously reported in models of obese rodents, Reference Boaventura, Casimiro-Lopes and Pazos-Moura18 but generally on higher doses of exercise. Achieving body weight reduction with less fat accumulation and glycemic normalization through an exercise program with less frequency and moderate intensity indicates a viable intervention to reduce health risks in obese individuals.
Interestingly, even though the present exercise program improved SL blood glucose homeostasis, it does not affect insulin levels in the ivGTT. Although the SL-Ex insulinemia during the test was reduced in the posttest compared to the SL-Se group, it was still elevated compared to the NL groups, with an interaction between factors pointing that early PO hyperinsulinemia was not overcome with moderate exercise. On the one hand, these results corroborate that hyperinsulinemia induced by early PO is not easily overcome later in life. Reference Habbout, Guenancia and Lorin19 On the other hand, the improvement in the SL-Ex glycemia in the absence of insulin greater changes suggests that the present exercise training increased non-insulin-dependent glucose uptake and improved insulin sensitivity, Reference O’Neill20 which is reinforced by the normalized HOMA-IR observed in SL-Ex animals. Such aspects are important components for glucose homeostasis.
The metabolic and biometric improvements promoted by exercise in SL rats are important to improve health conditions, however, they were not fully accompanied by fitness improvements in postnatal overfed trained animals. Even though VO2max was not different between the NL and SL groups before the exercise program, at PN30, or comparing sedentary groups (NL-Se vs. SL-Se) at PN90, the VO2max gain promoted by exercise in the NL-Ex group was not observed in the SL-Ex rats, indicating that the improvement in VO2max by moderate exercise training was blocked by early PO.
Hyperinsulinemia, as found in SL rats, has been identified as a possible factor for worsening VO2max. Reference Bassett and Howley6 Wisløff and colleagues subjected a model of rats with metabolic syndrome and low aerobic capacity Reference Koch and Britton21 to a program of five sessions per week at 70% of VO2max, and observed that there was no increase in oxygen consumption during exercise, despite the reduction in adipose tissue, Reference Haram, Kemi and Lee22 corroborating with the present data. Moreover, early PO was showed to impair the mitochondrial respiratory rate with losses in bioenergetic capacity. Reference Moura Freitas, Nascimento and Braz23 Together, the hyperinsulinemia induced by PO, which in the present study persists in SL-Ex rats, less physical capacity, and mitochondrial impairment could be the basis for the blocked VO2max gains in SL-Ex rats.
In addition to the limited physical fitness, indicated by the lack of VO2max gain, the SL-Ex animals also showed changes in LDH activity. LDH and CK are cytosolic enzymes that leak when the muscle cell membrane is damaged. Reference Dawson, Biasetti and Messina24 Therefore, persistent high levels of LDH in the serum are strongly associated with muscle damage.
In a recent study, our group showed that the present exercise protocol did not cause muscle damage in male control rats. Reference Moreira, da Silva Franco and Prates25 This observation was confirmed in the present study when assessing CK and LDH levels in NL-Ex rats. Nonetheless, while serum LDH levels were reduced in NL-Ex rats, it was increased in the SL-Ex group compared to all other groups. Accordingly, early weaning rats submitted to a running wheel exercise protocol also showed high levels of LDH. Reference Boaventura, Casimiro-Lopes and Pazos-Moura18 These results indicate that developmental stress can increase the susceptibility to muscle damage in exercised male rats, regardless of the exercise method used. Additionally, the literature points that the same processes that lead to metabolic diseases can also induce a type of barrier to adapting to exercise as a way to save energy, Reference Zhu, Eclarinal and Baker11 which may be associated with poor adaptation to physical exercise and decreased physical fitness in early obese animals.
In summary, the present data show that despite moderate physical training mitigate obesity and improved glycemia homeostasis in postnatal overfed male rats, PO-impaired VO2max during exercise and increased muscle damage in the SL-Ex group. Although there is a limitation in establishing the mechanisms underlying impairment of adaptation to exercise in postnatal overfed animals, this study reinforces the importance of the lactation period for physiological development and the risk of early overnutrition stress for long-term health. We can suggest that, although exercise in moderate intensity at low frequency is still a viable non-pharmacological intervention to attenuate the progression of overweight and associated metabolic diseases, early overnutrition during lactation can restrict the benefits of exercise training, therefore, nutritional care at such a sensitive stage of development should not be overlooked.
Acknowledgments
We thank Ms Maroly Pinto and Ms Marli Licero for helping to care for the rats in the animal facility.
Financial support
This work was supported by Brazilian Governmental Agency: Coordenação de Aperfeiçoamento de Pessoal de Nível superior (CAPES foundation), (process numbers: 10040/13-7 and 7006/14-4).
Conflict of interest
None.
Ethical standards
All the experimental protocols conducted in this study comply with the Brazilian Association for Animal Experimentation (COBEA) guidelines, Brazilian Federal Law and the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. The Animal Ethics Committee of Maringa State University approved all the procedures (protocol number 013/2014).
