Effects of Dietary Restriction on Regulation of Energy Metabolism in Male Wistar Rats (Rattus norvegicus)

Introduction

The most common way of feeding laboratory rodents is ad libitum (AL). In some cases, dietary restriction (DR) is used. There are several reasons for using DR in laboratory rodents. First, the feeding method may be an integral part of the experimental setup. DR-fed rodents can be used as animal models of the effects of weight reduction and maintenance in obesity-related diseases. There are other fields of research where DR is often used to provide appropriate experimental designs for example in the study of eating disorders and other forms of malnutrition. Second, DR may be implemented when the mortality or morbidity of the experimental animals needs to be decreased. It has been repeatedly shown that AL feeding increases the mortality and morbidity of laboratory rodents compared with DR-fed counterparts.^1–7 In toxicology and regulatory testing, the short lifespan and high incidence of neoplastic diseases in AL-fed rodents are problems, for which DR has been recommended as a solution.^1,8

A third reason to use DR is to control interindividual variability and to standardize the food intake of the experimental animals. In AL feeding, food intake is highly variable between different individuals, whereas DR can offer the possibility to control the food intake.^9–11 A fourth reason to use DR is to establish a stronger motivation to eat. This is required in some behavioral tests, where food is used as a reward.

Obesity caused by AL may lead to increased numbers of animals needed in the studies either via poor survival or increased interindividual variability. A potential increase in each and any parameter variance leads to an exponential increase in the number of animals required for the same study outcome. This has been shown to be the case with organ weights but not with other parameters.^11–13

Regardless of the rationale behind the choice of the feeding method, a common way to implement DR in laboratory rodents is to provide them with a precalculated portion of food once a day, during working hours. To avoid aggressive competition for food, the animals are often single housed.¹⁴ This method of DR has shortcomings with respect to both scientific integrity and animal welfare. Social isolation is a potent stressor for rats and can lead to decreased welfare.^15–19 Current European legislation recommends that rats be housed individually only if there is a justification based on veterinary, welfare, or experimental grounds.²⁰ Another concern is the diurnal rhythms of the experimental animals. When fed AL, rats eat several small meals during the dark period.^21,22 The current methods of DR disrupt the diurnal rhythms of feeding and consequently interfere with many other physiologic variables.^23–25 One might hypothesize that it is very difficult to differentiate the effects of the stress of social isolation and the altered diurnal rhythms from those of the actual caloric restriction. This may confound the interpretation of research results.

Our research group has developed a feeding device, the diet board, to solve some of the problems associated with the current methods of DR in laboratory rats. The advantages of the diet board include the possibility of group housing the animals and allowing them to maintain normal diurnal eating rhythms while subjecting them to moderate levels of DR.^26–28

The purpose of this study was to determine the diet board’s effects on the serum levels and variance of the hormones ghrelin, leptin, insulin, and adiponectin or liver triglycerides compared with AL feeding. These hormones reflect the levels of food intake and adiposity of the individual.^29,30 It is not sufficient that the diet board simply produces similar effects on weight gain and thus shows face validity to the current methods of DR. To be an acceptable alternative, the diet board should possess true construct validity; that is, the same mechanisms should be responsible for the similarities in the observed outcomes.^31–33 Our goal was to assess how the diet board would affect the hormone and liver triglyceride levels, and their variance compared with AL feeding.

Materials and Methods

The diet board.

The diet board (Figure 1A and B) consisted of 2 aspen boards (35.0 × 12.2 × 2.7 cm); 2 corners of each board were removed (6.0 × 6.0 cm) to facilitate the rats’ movement within the cage. Each board had 20 vertical drill holes (Ø 12.5 mm) with a 2- to 3-mm slot open to the side of the board. The holes were filled with food pellets that were fixed in place by autoclaving the board (121 °C, 20 min, 220 kPa; Finn-Aqua 121821 D; Steris Finn-Aqua). The control animals had similar autoclaved boards of the same size but without the drill holes and food. The boards were placed into the cage in the form of a cross made by intersecting 2 boards providing 4 separate compartments. The boards were made of aspen, which is the same material as the bedding and nesting material, and presumably had the same emission profile but at lower concentrations than the nonautoclaved aspen bedding material.³⁵

View Figure

• Download as Powerpoint
• Download Figure

Results

Detailed results on growth and adiposity have been published elsewhere. They showed that at the end of the 10-wk experiment with the diet board (15% DR as compared with AL), the rat weights (mean ± SD) were 321 ± 38 g for the DR group and 360 ± 35 g for the AL group, a difference with P < 0.0001. Moreover, the weight of gonadal fat, postmortem, was 4.47 ± 0.90 g in DR rats and 5.66 ± 1.06 g in AL rats, a difference with P < 0.0001.²⁶

Leptin, adiponectin, ghrelin, and insulin.

The serum hormone concentrations were analyzed from terminal blood samples taken at the end of test week 10. The serum ghrelin concentrations were lower in the DR animals. The model-based estimates of the geometric means for the groups were 1,700 pg/mL (95% CI [1,380 to 2,100]) in DR rats and 2,400 pg/mL (95% CI [1,940 to 2,970]) in AL rats. The ratio of the geometric means was 0.71 (95% CI [0.607 to 0.83], P < 0.001). No statistically significant differences were found between the DR and AL groups in the serum levels of leptin, adiponectin, or insulin (P > 0.05) (Figure 2; Table 1).

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Values and distributions of hormones and adiponectin at the end of the 10-wk study.

	Diet board				Ad libitum
Group	Mean	Quartile 1	Median	Quartile 3	Mean	Quartile 1	Median	Quartile 3
Ghrelin (pg/mL)	1,764	1,405	1,741	2,147	2,705	1,624	2,175	3,680
Leptin (ng/mL)	3.5	2.4	3.2	4.7	4.6	1.4	4.6	6.6
Insulin (µg/L)	1.2	0.8	1.2	1.5	1.6	0.7	1.2	2.7
Adiponectin (µg/mL)	13.9	11.9	13.3	16.8	14.9	12.6	13.7	16.1

View Fullscreen

Discussion

The diet board is a method of restricting the weight gain of laboratory rats. It allows the rats to be group-housed and does not expose the animals to periods of fasting. The diurnal rhythms of physical activity, blood pressure, and heart rate remain undisturbed.²⁸ The purpose of this study was to validate the diet board as an alternative method of DR in studies on energy metabolism.

The diet board significantly reduces the food consumption of rats.²⁶ The difference in absolute food consumption is approximately 15% during the sixth week of diet board feeding. When related to body weight, diet board and AL animals consume an equal amount of food per gram of body weight. The caloric intake is roughly estimated as 0.2 kcal/g body weight/day in both groups. This phenomenon is well known in DR research; the energy intake/body weight ratio remains the same on different levels of food intake.³⁵ These results suggest that the effects of the diet board are a result of true DR and not attributable to increased caloric expenditure due to physical exertion (that is gnawing).

The 15% difference in food intake corresponds with the 15% difference in body weight after 10 wk of diet board feeding. The diet board-fed animals also have 30% less gonadal fat and 3% shorter tibiae.²⁶ Thus, the diet board decreases fat accumulation quite effectively but disturbs skeletal growth only minimally. This is in accordance with other studies, where the lean body mass³⁸ and musculoskeletal system³⁹ were only marginally affected by DR.

Obesity in rats is typically a problem of outbred rats, and the majority of DR studies are done with Wistar or Sprague–Dawley stocks. DR procedures applied to rats are variable, as duration ranges from 10 d to 6 mo, severity ranges from 20% to 75% reduction from AL amounts,^40–43 and food presentation requires either permanent or temporary solitary housing. None of these come without consequences. Feeding time may derail diurnal rhythm,²⁵ and DR severity and duration may have a major impact on results and their interpretation.

The current methods of DR have extensive effects on lipid metabolism in rodents; DR has been repeatedly shown to decrease serum cholesterol, triglyceride, and free fatty acids (FFA) levels.^{4,40,42,44,45} The diet board-fed animals also have lower levels of serum triglycerides, FFA, and cholesterol.²⁶ Liver triglycerides have been reported to be unaffected by DR,⁴² which was also the case of diet board.²⁶

Leptin and adiponectin are produced and secreted from fat tissue, insulin from the pancreas, and ghrelin from the gastric mucosa.³⁰ The circulating levels of leptin, adiponectin, and insulin reflect the fat content of the body.^29,46–49 The current methods of DR have been associated with decreased serum levels of leptin and insulin and increased levels of adiponectin.^{40,42,48,50–53} In this study, however, no statistically significant differences were found between the DR and AL animals in their serum levels of leptin, insulin, or adiponectin (P > 0.05). The degree of DR (85% of AL) achieved with the diet board was relatively mild, thus possibly not eliciting as prominent changes as more severe regimes of DR. Furthermore, in the current methods of DR, the animals are often food deprived for long periods every day, which is not the case in diet board feeding.

The serum levels of ghrelin were significantly (P < 0.001) lower in DR rats compared with AL rats, and the decrease from 2,400 to 1,700 pg/mL is so large that it cannot be considered a false positive result (Figure 2; Table 1). This result is at odds with the known functions of ghrelin. Ghrelin is an orexigenic hormone, and decreased food intake and low body weight are associated with elevated levels of ghrelin.^29,54–56 Increased levels of both plasma and gastric ghrelin have been observed in rats subjected to DR ranging in severity from 30% to 75% of AL food intake.^41,42,54 There are, however, examples where DR has failed to increase serum ghrelin⁴³ or even resulted in decreased ghrelin levels.⁴⁰ A possible explanation behind the decreased ghrelin levels in DR rats observed in this study could be the timing of the meals. Even though diet-board-fed rats show similar diurnal rhythms of activity compared with the AL rats,²⁸ it could be that the diet board rats distribute their food intake in a more prolonged and scattered manner throughout the night, thus having eaten more recently before the blood sampling than the AL animals. The decreased levels of FFA in the terminal blood samples collected in the morning provide support for the notion that the diet board rats’ eating activity is more pronounced during the end of the dark phase compared with the AL-fed rats.²⁶

Increased secretion of corticosterone is an integral part of the organism’s metabolic response to decreased energy intake, and elevated levels of corticosterone have been repeatedly observed in rodents subjected to DR.^45,56–61 Serum corticosterone concentrations are likewise elevated in the diet board rats.²⁷ In line with our results, DR has been reported to increase corticosterone levels in the absence of significant effects on adiponectin or ghrelin levels.⁴³

DR has been reported to decrease result variation^{11,13,62–64} and has been named a “powerful Reduction tool”.⁶⁵ Diet board feeding, however, does not seem to possess this kind of reduction potential. No differences were found in the variation of the investigated variables, neither in this study nor in our previous studies²⁶ (Table 1). Perhaps the most striking difference to the earlier studies is having rats group housed all the time. In the study of Moraal et al.,¹¹ the DR rats were denied access to food during the day and received a restricted amount of food during the night while single housed. Furthermore, they¹¹ were able to feed the animals with precalculated doses while in our study the rats were able to eat at will provided they work (gnaw) for food. The likely explanation is that the diet board feeding does not offer any more control over the animals’ food intake than AL feeding.

Moraal et al.,¹¹ showed that food restriction feeding leads to a significantly reduced variation in body weight, growth, and lung weight, thereby indicating the potential for reduction. With respect to the 9 blood parameters tested, they found no differences in the variation, indicating that a 25% reduction in food intake may not affect the variation of clinical chemistry values.¹¹ These findings are in line with our results although diet board leads to only a 15% reduction in food intake.

The diet board does have shortcomings. It is not possible to regulate or register the food intake of individual animals. Rats younger than 7 wk are unable to get enough food from the diet board, and the animals require at least a one-week period of learning and adjustment with the diet board before the food intake and behavior has stabilized. The diet board cannot be used on animals with dental problems or other difficulties in eating.

In conclusion, the diet board offers a method of achieving moderate DR. The weight gain and adiposity are moderately decreased. The levels of the adiposity-related hormones leptin, insulin, and adiponectin are not significantly decreased. The decreased serum levels of the hunger signal ghrelin and FFA in the morning suggest that the diet board rats’ eating activity is more pronounced during the end of the dark phase compared with the AL-fed rats. Diet board feeding offers a low-tech and pragmatic method for DR with group housing in rats, which does not cause disruption to the animals’ diurnal rhythms. The diet board does not possess reduction potential for tested energy metabolism parameters.

Conflict of Interest

The authors have no conflicts of interest to declare.

Funding

This study was financially supported by the Finnish Ministry of Education, the Academy of Finland, the European College of Laboratory Animal Medicine and European Society of Laboratory Animal Veterinarian Foundations, Universities Federation for Animal Welfare, the Research and Science Foundation of Farmos, the Finnish Foundation of Veterinary Sciences, and the Oskar Öflund Foundation.