INTRODUCTION

Top Abstract Introduction Materials Results Discussion References

FOOD AVAILABILITY PLAYS an important role in the control of reproductive rate in many species (6). If energetic demands are larger than the energy resources available, reproductive function is suppressed until food becomes more abundant. This adaptive mechanism has been illustrated in both field studies (e.g., Refs. 11, 15) and in laboratory experiments in which food restriction or food deprivation is imposed on animals in various reproductive states (see Refs. 26 and 27 for review). For example, food restricting female prepubertal rats delays the onset of their first estrus (6). Similarly, a 48-h food deprivation regimen suppresses luteinizing hormone (LH) levels in adult female rats (7, 16) and disrupts the estrous cycle of female hamsters (24).

During lactation, female mammals show a period of infertility that is also sensitive to energy availability (4, 25). Studies conducted in our laboratory have determined that rat dams that are food restricted to 50% of the ad libitum ration during the first 2 wk of lactation prolong their period of lactational infertility for ~1 wk (29). The extended period of lactational infertility seen after 2 wk of food restriction appears to be related to a prolonged suppression in circulating levels of LH, which has been attributed to a decrease in gonadotropin-releasing hormone (GnRH) release from the hypothalamus (28), and it is not dependent on increases in suckling stimulation from the pups (30). It is not known, however, whether a more acute 48-h fast similar to that used by Cagampang et al. (7) in rats and by Schneider and Wade (24) in hamsters would also prolong the length of lactational diestrus. Determining the effects of such an acute energy intake manipulation on lactational diestrus would provide more information about the parameters of food restriction necessary to prolong lactational infertility and provide insight into the underlying mechanisms. In experiment 1, therefore, lactating rats were food deprived for 48 h at the end of the second week of lactation and their length of lactational diestrus was compared with that of ad libitum-fed (AL) counterparts. In addition, food-deprived (FD) and AL dams matched for day of parturition had their litters switched to evaluate the contribution of suckling to the length of lactational diestrus seen in the FD groups.

It has been suggested that the GnRH suppression that is observed after periods of food restriction or food deprivation is caused by changes in metabolic fuel availability (24, 26, 27). Animals treated with pharmacological agents that block glucose utilization and fatty acid oxidation, such as 2-deoxy-D-glucose and insulin, show suppressed LH pulses and disrupted estrous cycles (18, 19, 24). More recently, it has become evident that circulating signals from adipose tissue also regulate food intake and reproductive function.

Leptin, the protein product of the ob gene expressed in adipocytes, produces a marked decrease in food intake and body weight in genetically obese and wild-type mice when administered systemically and centrally (8, 12, 20). In addition, leptin treatment accelerates puberty in female mice (9) and participates in the mechanisms that control the onset of puberty in female rats (10). Moreover, systemic leptin treatment corrects some of the reproductive deficits seen in genetically obese mice (ob/ob) (3) and in mice food deprived for 48 h (2). It is therefore reasonable to propose that leptin may act as a signal that could attenuate any effects that food restriction and food deprivation may have on the length of lactational infertility. In experiment 2, we explored the possibility that leptin treatment concurrent with 48 h of food deprivation at the end of the second week of lactation would reverse any effect of fasting on the length of lactational infertility. Finally, because it has recently been reported that leptin levels are low in lactating AL animals relative to nonlactating females (5), we determined the effect of leptin administration on the length of lactational diestrus in AL lactating rats and obtained preliminary data on circulating leptin levels in lactating rats after a 48-h fast as well as in AL lactating and nonlactating rats.

	METHODS AND MATERIALS

Top Abstract Introduction Materials Results Discussion References

Subjects

General Procedures

Experiment 1. On the morning of day 13 pp, animals were assigned to either the AL or 48-h FD condition, and food was removed from the cages of the rats in the FD group. On the morning of day 15, all animals were fed ad libitum, and the litters of some of the animals in the FD condition were exchanged with age-matched litters of animals in the AL condition, creating four groups: AL dam/AL pups (AL/AL; n = 7), FD dam/FD pups (FD/FD; n = 8), AL dam/FD pups (AL/FD; n = 6) and FD dam/AL pups(FD/AL; n = 6). Litter weight was recorded daily until day 20 pp, and dam weight was recorded daily until the end of the study.

Experiment 2. On day 13 pp, animals were assigned to one of three groups: 1) a 48-h FD group (FD + leptin; n = 5) that received intraperitoneal injections of murine leptin (2.5 mg · kg¹ · day¹; Preprotech; pH 7.35) every 12 h during the deprivation period, 2) a 48-h FD group (FD + Veh) that received injections of the vehicle (10 mM Tris buffer), and 3) an AL group that also received vehicle injections (AL + Veh). Data collection procedures were as described in experiment 1, with the exception that food intake was also measured during the 2 days that followed the diet-injection regimen (days 15 and 16 pp).

Experiment 3. Animals in this study were assigned either to a leptin-treated (AL + leptin; n = 6) or vehicle-treated (AL + Veh; n = 6) group on day 13 pp. Animals in the AL + leptin group received intraperitoneal injections of murine leptin (2.5 mg · kg¹ · day¹; Preprotech; pH 7.35) every 12 h for 48 h beginning at 0800 on day 13 pp. Those in the AL + Veh group received vehicle injections at the same time intervals. Data collection was as described above.

Experiment 4. Nonlactating (n = 7) and lactating (n = 12) females served as subjects in this experiment. The lactating females were assigned to an AL (n = 8) and an FD (n = 4) group. Females in the FD group were food deprived for 48 h beginning on the morning of day 13 pp. Blood samples were taken via indwelling jugular catheters on day 15 pp in the lactating animals and once estrous cyclicity had been established in the nonlactating females, which were fed ad libitum throughout the study. Rats in the FD group were not refed before sampling.

Cannulation. Females were equipped with a jugular catheter 2 days before blood sampling. After anesthesia with a ketamine-xylazine mixture (6 mg ketamine and 0.86 mg xylazine/100 g body wt), a 26-mm length of Silastic tubing (Dow Corning, Midland, MI; outer and inner diameters 0.047 and 0.025 in., respectively) was introduced into the right jugular vein to the atrium of the heart, secured to the vein, and exteriorized between the scapulae. The cannula was filled with heparinized sterile saline (5,000 U heparin/100 ml), and the wound was rapidly sutured. A topical antibiotic (Cicatrin, Wellcome, Kirkland, PQ, Canada) was applied to prevent infection at the wound site.

Sampling. The day before sampling, dams and their litters or nonlactating females were transferred to experimental cages (similar to housing cages) and kept in a quiet experimental room. The next morning (0800-0900), the Silastic jugular cannula was connected to ~70 cm (200 µl of dead volume) of PE50 tubing (Becton Dickinson, Parsippany, NJ), which was exteriorized out of the cage, and rats were left undisturbed with their litters for 2-3 h before sampling. Sampling was initiated between 1100 and 1200, when blood (300-400 µl) was withdrawn through the extension tubing. The sampling procedure did not appear to disturb the behavior of the animals, most of which slept or nursed their litters throughout the sampling process. Blood was collected on ice and centrifuged for 3 min at 13,000 rpm, and plasma was stored at 20°C before hormonal determination.

RIA. Plasma leptin concentrations were determined using a rat leptin RIA kit obtained from Linco Research (St. Charles, MO) Determinations for each sample were made in duplicate within one assay. Intra-assay variability was <8%, and nonspecific binding was <5%.

Statistical Analyses

Experiment 1. Differences in the length of lactational diestrus among the four groups were assessed using a two-way between-groups ANOVA (with maternal diet condition and pup diet condition as the between-group factors). Differences in dam weight change and litter weight gain measures during and after the deprivation period were evaluated using three-way repeated measures ANOVAs (with maternal diet condition and pup diet condition as the between-group factors and days as the within-group factor).

Experiment 2. A between-groups ANOVA was used to analyze the differences in the length of lactational infertility among the three groups. Food intake measures were averaged to obtain a mean posttreatment (days 16 and 17 pp) score for each group, and these means were compared using a between-groups ANOVA. Repeated-measures ANOVAs (group × day) were conducted to examine the change in dam weight and litter weight during and after the experimental treatment.

Experiment 3. Independent t-tests were used to compare length of lactational diestrus, dam weight change during treatment, and daily pup growth before treatment. Repeated- measures ANOVAs were used to examine the effects of treatment on food intake and pup growth.

Experiment 4. A between-groups ANOVA was used to compare circulating leptin levels among the experimental groups.

	RESULTS

Top Abstract Introduction Materials Results Discussion References

Experiment 1

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Length of lactational diestrus (means ± SE) of ad libitum-fed (AL) and food-deprived (FD) females either nursing same litters throughout experiment (AL/AL, FD/FD) or whose litters were exchanged with females in the other diet condition (AL/FD, FD/AL) on day 15 postpartum (pp). No. of rats used is given in parentheses above each bar.

As expected, dams in the FD groups (FD/FD and FD/AL) lost weight during the food deprivation period, whereas the AL dams showed a slight increase in body weight during the same period (AL, 9 ± 2.67 g; FD, 94.34 ± 4.1 g). By day 20 pp, FD dams had not gained as much weight as that gained by animals in the AL groups [F(1,23) = 19.84, P < 0.05].

Overall, pups in the FD dam groups gained less weight than pups in the AL dam groups [main effect for maternal diet condition; F(8,184) = 87.23, P < 0.05]. By day 17 pp, all litters were gaining weight at a similar rate [diet × time interaction, F(8,184) = 76.91, P < 0.05]. As seen in Fig. 2, both the nutritional history of the dams and that of their litters contributed to the variation in litter growth on days 15 and 16 pp [diet × litter switch × time interaction, F(8,184) = 8.30, P < 0.05]. Simple effects analysis on these days showed that on day 15 pp, pups being nursed by AL mothers gained more weight than pups nursed by FD mothers. In addition, previously FD pups gained more weight than pups nursed by AL dams throughout the experiment. On day 16 pp, litters in the AL/FD group gained more weight than those in any other group.

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Daily weight gain (means ± SE) of litters nursed by females in each of 4 experimental conditions.

Experiment 2

View larger version (10K): [in this window] [in a new window]   Fig. 3.   Length of lactational diestrus (means ± SE) of AL females, FD females, and FD females treated with leptin (2.5 mg · kg1 · day1) during food deprivation period. Veh, vehicle. No. of rats used is given in parentheses above each bar.

Leptin treatment had no effect on the food consumed by FD animals in the 2 days following the fast; animals in the AL group consumed more food than animals in both FD groups during this time [means ± SE: FD + leptin, 53.58 ± 3.39 g; FD + Veh, 56.92 ± 2.17 g; AL, 65.08 ± 1.96 g; F(2,19) = 5.82, P < 0.05].

As expected, FD dams lost a significant amount of weight during the deprivation period (AL, 4.95 ± 5.16 g; FD, 97.12 ± 3.67 g) and overall gained less weight than AL dams from day 13 to day 20 pp [F(2,19) = 28.39, P < 0.05]. Interestingly, FD dams treated with leptin had not gained as much weight on day 17 pp as FD dams treated with vehicle [F(2,19) = 10.85, P < 0.05] and remained different from AL dams until day 18 pp [F(2,19) = 4.22, P < 0.05].

Leptin administration had no effect on litter weight change. Litters nursed by FD dams gained less weight during the deprivation period and on days 15 and 16 pp than those nursed by AL dams but gained more weight than litters nursed by AL dams from day 17 to day 19 of lactation [statistically significant group × day interaction, F(14, 133) = 25.44, P < 0.05].

Experiment 3

Leptin administration had no effect on maternal body weight. Dams in the leptin-treated group lost an average of 3.08 + 4.62 g during treatment, whereas those in the vehicle-treated group lost an average of 1.38 + 2.89 g [t(10)= 0.31, P > 0.05]. As can be seen in Fig. 4A, maternal food intake was reduced by leptin treatment. Although average food intake was similar for both groups before treatment (day 12 pp), females in the AL + leptin group ate less than those in the AL + Veh group during treatment (days 13 and 14 pp). This difference disappeared on the first day after treatment [day 15 pp; statistically significant group × days interaction, F(3,30) = 4.48, P = 0.05]. There was a marginally significant effect of leptin treatment on litter weight gain (see Fig. 4B). There was no difference in litter growth between the groups before treatment [t(10) = 0.35, P < 0.05], but litters nursed by dams receiving leptin injections gained less weight on days 13, 14, and 15 pp than did those nursed by vehicle-treated dams [statistically significant main effect for groups, F(1,10) = 4.57, P < 0.06].

View larger version (12K): [in this window] [in a new window]   Fig. 4.   A: daily food intake (means ± SE) of vehicle- and leptin-treated dams. B: daily litter growth of pups nursed by vehicle- and leptin-treated AL dams.

Experiment 4

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Circulating leptin levels (means ± SE) of AL nonlactating female rats and AL and FD lactating rats on day 15 pp.

	DISCUSSION

Top Abstract Introduction Materials Results Discussion References

The present study demonstrates that 48 h of food deprivation at the end of the second week of lactation suppresses reproductive function in lactating rats. These data are consistent with those of previous studies in rats and hamsters demonstrating that a similar food deprivation schedule disrupts estrous cyclicity (7, 16, 24). Interestingly, the ~7-day increase in length of lactational diestrus observed in the current study is similar to that of dams restricted to 50% of ad libitum food intake for the first 2 wk of lactation (29, 30).

Giving FD females AL litters to nurse after the food deprivation period decreased the length of lactational infertility, suggesting that, with this paradigm, the nutritional history of the dam and of the litter that she suckles after the food deprivation period both contribute to the effects of food availability on reproductive function. These data contrast with previous results showing that the length of lactational infertility of chronically food-restricted dams is unaffected by the nutritional history of the pups (30). Thus it appears that both acute and chronic shortages of food extend the length of lactational infertility but that they do so by somewhat different mechanisms.

Leptin administration during the fasting period eliminated the effect of food deprivation on length of lactational diestrus without changing maternal or litter weight loss. There was, however, a slight retardation of maternal weight gain during the postfasting period in females in the FD + leptin group. These data are consistent with those of Ahima et al. (2) demonstrating that leptin administration reversed the effects of food deprivation on estrous cyclicity in mice. Unlike these results, however, no differences in food intake between leptin-treated and vehicle-treated FD groups were observed in the postfasting phase in the current study, and both groups ate less than the AL animals.

Our preliminary data suggest that circulating leptin levels in lactating animals were reduced by food deprivation. Thus the demonstration here that leptin administration is able to eliminate the effects of food deprivation on the length of lactational infertility suggests that circulating leptin levels may play a physiological role in this phenomenon. These data join a growing body of evidence to suggest that leptin influences reproductive function in a variety of situations. Leptin treatment has been used to restore fertility in genetically obese (ob/ob) mice (3), to induce the early onset of puberty in lean female mice (9), and to reduce the effects of food deprivation on estrous cyclicity in mice (2).

What role if any the low level of circulating leptin observed in AL lactating rats plays in the suppression of reproductive function at this time remains to be determined. Brogan et al. (5) reported that lactating rats on day 10 pp had lower levels of circulating leptin than nonlactating rats, and here we obtained similar results when we compared nonlactating rats with dams on day 15 pp. These data suggest that AL lactating animals have chronically low leptin levels. In the current study we found that leptin administration to AL lactating rats on days 13 and 14 pp had only a small, statistically nonsignificant effect on length of lactational infertility. A more prolonged period of leptin administration may be necessary to affect reproductive function in AL lactating rats.

The pattern of leptin administration used in this experiment was sufficient to reduce food intake in AL lactating rats and to suppress litter growth. The pathway through which this latter effect was achieved is not clear. Litter growth could be affected through a reduction in milk production consequent on the decrease in food intake of the dams or through a direct effect on the mammary gland or maternal behavior. Interestingly, recent data suggest that leptin is detectable in human breast milk and that the levels in milk reflect the levels in the maternal circulation (13). Thus the reduction in litter growth produced by leptin administration might also reflect a direct effect of leptin on the pups.

Although the ability of leptin to modulate the reproductive axis is clear (2, 3, 9), the mechanism through which the effects of leptin administration are produced is less well defined. Leptin receptors have been localized in many endocrine and neuroendocrine tissues (32). Furthermore, the results of in vitro studies suggest that leptin influences ovarian function, increases the release of LH and follicle-stimulating hormone from the pituitary, and stimulates GnRH release from arcuate nucleus/median eminence hypothalamic explants (31).

It is also possible that leptin acts indirectly to modulate reproductive function. Leptin decreases neuropeptide Y (NPY) synthesis in the arcuate nucleus in both satiated and FD animals and NPY itself has been shown to affect reproductive function (17). For example, the prolonged period of lactational diestrus observed in food-restricted dams is mediated by a prolonged suppression of LH levels that persists for at least 5 days after the dams are refed (28). These low levels of LH are thought to be accompanied by suppression of GnRH release from the hypothalamus (28) and to be related to food restriction-induced increases in NPY release (1). It has also been suggested that leptin administration changes levels of circulating glucose and free fatty acids (21, 22) and modulates glucose metabolism (14). Such changes could affect reproductive function by changing the activity of metabolic fuel receptors.

Perspectives

Leptin administration during the period of food deprivation is sufficient to completely prevent the prolongation of lactational infertility. Hence increasing leptin levels in FD dams overrides the effect of nursing previously underfed pups as well as the effects of food deprivation on the dam herself. Furthermore, the finding that FD lactating rats have lower plasma leptin levels than AL dams suggests that the effect of leptin in this paradigm is physiologically relevant. AL dams have lower levels of circulating leptin than nonlactating females, but the pattern of leptin administration used here was not sufficient to shorten lactational infertility in these animals, although it did reduce food intake and litter growth. Thus the question of what role, if any, low circulating leptin levels may play in the suppression of fertility during lactation remains to be elucidated.