Effects of Food Restriction on Testis and Accessory Sex Glands in Maturing Rats

Abstract

Reduced food consumption and associated lower body weights may occur in subacute toxicity studies. The short-term effects of food restriction (FR) on body and reproductive organ weights, hormones, and testis histology were assessed in Sprague-Dawley rats fed 20% to 36% less (21 g feed/day) than rats fed ad libitum (AL) starting at six, eight, ten, or twelve weeks of age for two or six weeks. Body weight and relative seminal vesicle, ventral prostate, and/or epididymis weights were reduced in rats FR for two or six weeks. Degeneration of stage VII pachytene spermatocytes was seen in rats FR for six weeks when initiated at eight, ten, and twelve weeks of age. Plasma testosterone concentrations were lower in rats FR at ages six to eight weeks, eight to ten weeks, six to twelve weeks, and eight to fourteen weeks. Luteinizing hormone was not statistically different in FR rats compared with AL counterparts. Therefore, duration of lower food intake had a greater impact on spermatogenesis, whereas a younger initial age of lower food intake was more influential on testosterone levels. These interactions are important in the interpretation of subacute toxicology studies employing FR or when test articles lower food consumption relative to AL-fed rats.

Introduction

Food restriction (FR) in rat chronic toxicity studies prevents the development of obesity and delays the onset of inherent disease (Keenan et al. 1999; Keenan et al. 2005; Molon-Noblot et al. 2003; Nold et al. 2001). Food restriction is usually implemented when rats are approximately eight weeks of age, prior to full sexual maturity. In previous subchronic toxicity studies, we implemented the same reduced-feeding regimen used in chronic studies and noted that diverse compounds caused identical, dose-dependent lesions in the testis in one-month toxicity studies, which triggered the current investigation.

Significant FR (≥ 50%) has clearly been shown to decrease testosterone and luteinizing hormone (LH) levels leading to effects on androgen-dependent male rat reproductive organs, including reduced organ weights and testicular degeneration (Glass et al. 1986; Grewall et al. 1971; Herbert 1985; Howland 1975; Levin et al. 1993; Santos et al. 2004). Short-term toxicity studies may frequently be associated with decreased body weight gain or with body weight loss. It was shown that even mild FR may lower testosterone and seminal vesicle and ventral prostate weights and pose a dilemma in determining whether effects are related to the test article or food consumption (Trentacoste et al. 2001). Studies demonstrating the effects of altered nutrition on results of spermatogenic staging in rats have not been conducted (Szczech and Russell 1997).

Therefore, a study was designed to assess effects of FR during different periods of maturation in rats on testicular histology, reproductive organ weight, and plasma hormones.

Materials and Methods

Study Design

Approximately 160 male Sprague-Dawley virus-antibody–free rats [Crl: CD^® (SD) IGS BR] (Charles River Laboratories, Raleigh, NC) with unique identification transponders were obtained for this study (Guide for the Care and Use of Laboratory Animals, 1996). The rats were not littermates and were four, six, eight, or ten weeks of age at the time of receipt. Following a two-week quarantine period, rats were weighed, subjected to a detailed clinical examination, and randomized to groups by computer using ranked body weight and age. Rats were housed individually in stainless-steel cages in a controlled environment (64°F–79°F; 30% to 70% relative humidity) with a twelve-hour light/twelve-hour dark cycle. Certified Rodent Diet 4.2EXT-5L35 (PMI Feeds, Inc., St. Louis, MO) was provided ad libitum (AL) to all rats during the quarantine period (approximately two weeks). The macronutrient content of this diet is as follows: 20% protein, 4.5% fat, and 55% carbohydrates, with a gross energy of 4.04 kcal/g. Following randomization, rats in Groups 1, 3, 5, 7, 9, 11, 13, and 15 were provided feed AL, and rats in Groups 2, 4, 6, 8, 10, 12, 14, and 16 were given a daily allotment of approximately 21 g/day for FR to according to the schedule listed in the study design (Table 1). The amount of 21 g feed had previously been employed in FR studies. Individual pellets were extruded to weigh approximately 2.6 g, new batches were routinely controlled, and rats were given eight pellets per day. Filtered tap water was provided from an automatic watering system or water bottles. Available information indicated that no substance was present in the diet, drinking water, or bedding at a concentration likely to influence the outcome of this study. Rats were weighed once during quarantine, prior to randomization, and daily following initiation of the diet regimen. Food consumption was checked daily for FR rats and measured daily for AL animals following initiation of the diet regimen. One Group 11 rat (AL 10–16 weeks) was euthanized four weeks after the initiation of the ad libitum feeding regimen because of a fractured incisor, which resulted in decreased food consumption and body weight loss. No other deaths occurred prior to scheduled termination in this study. The study was conducted in accordance with current guidelines for animal welfare (Guide for the Care and Use of Laboratory Animals 1996; Animal Welfare Act 1966, as amended in 1970, as amended in 1976, as amended in 1985, 9 CFR Parts 1, 2, 3). Procedures used in these studies were reviewed and approved by the internal Institutional Animal Care and Use Committee.

Collection of Specimens

According to the schedule in Table 1, rats were killed by CO₂ asphyxiation and exsanguination. Blood samples (approximately 1–3 mL) from all animals (Groups 1 to 16) were obtained from the vena cava at necropsy for measurement of plasma hormone concentrations. The blood samples were transferred into heparinized blood collection tubes and immediately placed on crushed water ice prior to centrifugation. Following centrifugation, the resultant plasma was separated and transferred into uniquely labeled polypropylene tubes and frozen immediately over solid carbon dioxide.

A necropsy limited to reproductive organs was conducted. Each testis, paired epididymides, paired seminal vesicles, and the ventral prostate were weighed before fixation. Prior to removal, the seminal vesicles were clamped with hemostats to prevent the loss of secretory fluids. Testes were fixed in Bouin’s solution for approximately twenty-four hours and then transferred to 70% ethanol. One testis was trimmed sagitally and the other transversally to obtain sections through the rete testis in both (Foley 2001). Testes were routinely processed, sectioned, and stained with hematoxylin and eosin/phloxine or periodic acid Schiff (PAS) for examination by light microscopy.

Hormone Assays

Testosterone (T) and luteinizing hormone (LH) were measured in plasma samples using commercially available double-antibody radioimmunoassay kits from MP Biomedicals (T) and GE Healthcare BioSciences (LH). Plasma samples were centrifuged at high speed to condense fibrin clots and were measured in singlet. Plasma volume of 0.050 mL or 0.100 mL was used in the T assays. The standard curve ranged from 2.5 pg/tube to 500 pg/tube for all assays. The lowest detectable hormone level (with 95% confidence) was 0.01 pg/tube to 2.53 pg/tube. The interassay variations were from 7% to 12%. Plasma volume for the LH assay was 0.100 mL. The standard curve ranged from 0.08 ng/tube to 5 ng/tube. The lowest detectable hormone level (with 95% confidence) was from 0.06 ng/tube to 0.10 ng/tube. The interassay variation was from 1% to 17%.

Statistical Analysis

Body weight data were statistically analyzed for group differences via the parametric F test. The t test was then used to test pairwise comparisons between the mean values for each FR group compared to the corresponding ad libitum-fed control group.

Organ weights, absolute and relative to terminal body weight, were analyzed for group differences using the Kruskal-Wallis test, followed by the Wilcoxon test comparing each diet-optimized group to its corresponding ad libitum–fed control group. Statistical tests were performed using statistical options in the DATATOX System. Statistically significant data (p ≤ .05) are reported from these overall and pairwise tests.

To assess differences in hormone levels in male rats that were fed ad libitum (concurrent control) or a restricted diet, a nonparametric Kruskal-Wallis analysis of variance (ANOVA) was used followed by multiple comparisons of ranks. If the overall p value indicated significance (p ≤ .05), Kruskal-Wallis/Independent Comparisons (KWIC) Test was used to determine final significance. All descriptive and statistical analyses for hormone data were performed using the statistical software Statistica, version 6.90 (StatSoft Inc., Tulsa, OK).

Results

Food Consumption

In AL groups, mean food consumption increased with increasing age from 26 g/day for rats aged six to eight weeks to approximately 33 g/day for the oldest rats (seventeen to eighteen weeks; individual data not shown). The 21 g allotment provided to FR rats was consumed daily and accounted for approximately 80% of the food consumed by the six- to eight-week-old rats fed AL and about 64% to 70% of the food consumed by the eight- to eighteen-week-old rats fed AL (Table 2).

Body and Organ Weights

Rats fed AL gained body weight throughout the study, with weekly increases in body weight that decreased in magnitude as rats matured (Figure 1; Table 3). Rats placed on FR at the age of six weeks also gained body weight throughout the study, whereas rats placed on FR at a later age initially lost body weight for one to two weeks before resuming growth. At all times, FR rats gained less body weight than the AL counterparts, such that terminal body weights of FR rats for the two-week regimen and 22% to 27% less for the six-week regimen (Tables 3a and 3b).

Testes and epididymal weights were analyzed as absolute weights because these are less dependent on body weight than seminal vesicles and ventral prostate, which are presented adjusted for body weight (Figure 2). There was no difference between left and right testis (data not shown), and FR had no or only a negligible effect on absolute testicular weight. Slight but statistically significant decreases in epididymal weights were seen in rats FR for six weeks at most ages tested. Marked decreases in seminal vesicle weights (normalized to body weight) occurred in FR rats on both the two-week regimen and the six-week regimen. Decreases in relative ventral prostate weights also occurred in FR rats, but these changes were less marked than for seminal vesicle weight changes and affected only certain age groups.

Microscopic Observations of the Testis

The most common finding associated with FR was the occurrence of pachytene spermatocyte degeneration in stage VII of the spermatogenic cycle (Figure 3). These cells were characterized by dark basophilic nuclei; shrunken, deeply eosinophilic cytoplasm, consistent with apoptosis; and they were located at the base of the tubules. Only cases with bilateral occurrence were diagnosed and graded according to severity as minimal, mild, or moderate (Tables 4a and 4b). Sporadic, unilateral, degenerative changes were not included. Most commonly, degenerating pachytene spermatocytes in stage VII were seen in rats that had been FR for six weeks (16/40) compared to FR for two weeks (5/40) and were rare in AL rats (1/79). FR for six weeks when initiated at the ages of six, eight, ten, or twelve weeks was associated with degeneration of pachytene spermatocytes in 1/10, 7/10, 4/10, and 4/10 rats, respectively (Table 4). In one rat FR from eight to fourteen weeks of age, degeneration of pachytene spermatocytes was noted in stages VII and VIII accompanied by retention of elongated step 19 spermatids in stages VIII to XII (Figure 4). The most severe case of testicular changes was noted in a rat FR from twelve to eighteen weeks of age showing tubular degeneration and loss of germ cells, including degeneration of pachytene spermatocytes and spermatid retention (Figure 5). Food restriction for either only two weeks or six weeks, when initiated at six weeks of age, was associated with a lower incidence of rats with degenerating pachytene spermatocytes in stage VII.

Hormone Assays for Testosterone and Luteinizing Hormone

Food restriction resulted in significant changes in plasma T concentrations. Specifically, median plasma T concentrations were lower in rats when FR was initiated at six or eight weeks of age; when initiated at later ages (ten or twelve weeks), no effects on plasma T levels were observed. Therefore, the age at which FR was started was a more important determinant of plasma T levels than the two- or six-week duration of FR in this study. In rats FR at six weeks of age, plasma T concentrations were reduced by 72% in rats FR at the age of six to eight weeks, by 76% at eight to ten weeks (p ≤ .05), by 81% at six to twelve weeks (p ≤ .05), and by 88% at eight to fourteen weeks (p ≤ .05; Tables 5a and 5b). Circulating plasma LH levels appeared to be unaffected by FR.

Discussion

The current study demonstrates that mild reductions in food intake not only lowered body weight gain, reduced reproductive organ weights (seminal vesicles, ventral prostate, epididymis), and decreased serum testosterone, but also caused specific stage-dependent, microscopic testicular alterations.

Food restriction of 20% in rats aged six to eight weeks and 30% to 36% in rats aged eight to eighteen weeks resulted in approximately 15% or 25% lower terminal body weight for the two- or six-week regimen, respectively, compared with AL-fed rats. Compared with older animals, it appears that younger (smaller) rats were able to adapt more readily to the nutritional change and did not lose body weight in the first week of FR, which may have modulated the effects on male reproductive organs and resulted in fewer changes. Food restriction initiated in six-week-old rats, therefore controlled growth, whereas food restriction started in larger rats at least eight weeks old caused a loss of body weight during the first week, which could have triggered responses resulting eventually in a lowering of testosterone.

It has been recommended to use absolute testis weight rather than relative testis weight to assess effects, because the testis, like the brain, is conserved despite body weight loss (Creasy 2003).

One of the most sensitive reproductive endpoints was lower seminal vesicle weights relative to body weight. The greater sensitivity of seminal vesicle compared with prostate is likely because of its larger proportion of glandular luminal contents relative to organ mass.

Food restriction for six weeks (except FR for the six-week period of six to twelve weeks) had a greater effect on testicular histology than the two-week regimen. These observations are consistent with reports showing that decreased testosterone can result in specific stage-dependent testicular changes including degeneration (apoptosis) of pachytene spermatocytes in stage VII tubules and later may also impact pachytene spermatocytes in stage VIII tubules, with retention of step 19 spermatids and degeneration of step 7 round spermatids (Creasy 2001; Russell et al. 1981). In toxicity studies, it will have to be assessed on a case-by-case basis whether this observation constitutes an adaptive phenomenon resulting from decreased weight gain or an effect attributable to actions of the test article.

In the current study, plasma testosterone levels were decreased in rats when FR started at six or eight weeks of age, suggesting that FR initiated between the ages of six and eight weeks can cause decreases in testosterone. Interestingly, this is also the time in rats with normal caloric intake when testicular T levels undergo a series of sharp rises (Herbert 1985). These testicular T-level rises occur between the fifth and the tenth weeks of age and coincide with puberty and the initiation of spermatogenesis. Data in this study suggest that FR may blunt this rise in T and this effect is sustained throughout the period of FR. However, when FR is initiated later in life, that is, after ten weeks of age, Leydig cells apparently are no longer as sensitive to diet or caloric restriction.

Surprisingly, there was no change in the circulating plasma LH levels in our study, although others (Chacon et al. 2004) have shown a decrease in the twenty-four-hour LH profile, which is most apparent in the dark period. As LH values in this study were obtained from a single time point during the daytime period, it is possible that decreases in LH may have been missed. The effect of lowering LH may, however, also depend on the level of food restriction. Chronic dietary restriction showed, at the age of fifty-three weeks, a lowering of serum LH only in markedly restricted rats (48% of adult AL amount), but not in rats fed 68% to 79% of adult AL amount (Molon-Noblot et al. 2003).

Decreases in circulating plasma T levels did not strictly correlate with the organ weight changes and effects on spermatogenesis in this study. For example, decreased seminal vesicle weights were observed under nearly all FR conditions, whereas testicular changes were most commonly observed when FR started at eight weeks of age and later. Additionally, decreased T levels were observed only when FR was started at six or eight weeks of age. However, an association between lowered T and effects on spermatogenesis cannot be completely discounted, as the group with highest incidence of spermatocyte degeneration (Group 8, FR from age eight to fourteen weeks) also had the greatest decreases in plasma T.

Lack of an effect on prostate weights likely reflects the absence of an effect of FR on intraprostatic DHT levels. Using selective 5-α-reductase, it has been demonstrated that in the presence of high testosterone levels, lowering of intraprostatic dihydrotestosterone levels is sufficient to reduce prostate weights (Prahalada et al. 1998). In some groups, seminal vesicle weights were decreased even in the absence of any testosterone lowering, despite known androgen dependence (Marty et al. 2003), but this finding is not unprecedented as others have reported decreased seminal vesicle weights in the absence of testosterone changes in diet-restricted rats (O’Connor et al. 2000). It is likely that diet restriction alters other biochemical and metabolic processes that may affect seminal vesicle weights and may lower testosterone levels.

Over an extended period, older rats appear to adjust to reduced caloric intake, since no sex organ weight changes have been reported in studies of seventeen weeks duration (Chapin et al. 1993), and Glass et al. (1986) showed normalization of spermatogenesis with prolonged undernutrition starting at weaning. Rebounding of serum testosterone levels and accessory sex organ weights also occurred when FR was initiated at ten weeks of age and continued for twenty-four weeks (Glass et al. 1986). Therefore, in the course of a two-year carcinogenicity study, reduced food consumption will not have a negative impact on the male reproductive system (Keenan et al. 2005; Seki et al. 1997).

In summary, FR for two or six weeks caused a reduction in body and reproductive organ weights, lower plasma testosterone levels, and degeneration of pachytene spermatocytes in stage VII. The duration of lower food intake (six weeks vs. two weeks) was the greatest determinant of testicular degeneration, whereas a younger initial age of lower food intake (six or eight weeks vs. ten or twelve weeks) was most influential on testosterone levels. These interactions can be important points in the interpretation of subacute toxicology studies employing FR or when test articles lower food consumption relative to control-fed rats.