Chronic Leptin Administration in Developing Rats Reduces Stress Responsiveness Partly through Changes in Maternal Behavior

Hormones and Behavior 37, 366 –376 (2000) doi:10.1006/hbeh.2000.1578, available online at http://www.idealibrary.com on

Chronic Leptin Administration in Developing Rats Reduces Stress Responsiveness Partly through Changes in Maternal Behavior Max Oates, Barbara Woodside,* and Claire-Dominique Walker Department of Psychiatry, Douglas Hospital Research Center, McGill University, 6875 Lasalle Boulevard, Montreal, PQ H4H 1R3; and *CSBN, Department of Psychology, Concordia University, Montreal, Canada Received April 21, 1999; revised December 17, 1999; accepted January 19, 2000

In adult rodents, leptin has been shown to significantly alter the activity of several neuroendocrine functions, including the activity of the hypothalamic–pituitary–adrenal (HPA) axis. Leptin is generally believed to be inhibitory to HPA activity in adults. Developing rat pups have high circulating levels of leptin, which begs the question of leptin’s physiological role in controlling basal and stress-induced adrenocortical activity in neonatal rats. In this study, we treated rat pups daily from days 2–9 (or 6 –10) of life with either vehicle or leptin (1 or 3 mg/kg body wt, ip) and determined the effects on body weight gain, fat pad deposits, and HPA activity in 10-day-old pups. We measured hypothalamic CRF mRNA levels in vehicle- and leptin-treated pups by in situ hybridization and determined plasma ACTH, corticosterone, and leptin concentrations under basal conditions or following exposure to a 3-min ether stress. Because leptin activates sympathetic activity and energy expenditure in adults and possibly also in rat pups, and because litter temperature is an important determinant of maternal behavior, we also investigated whether chronic leptin administration would modify aspects of maternal care that are important for the maintenance of HPA function. Chronic leptin treatment increased circulating levels of leptin and had significant dose-related metabolic effects, including reduced body weight gain and fat pad weight in 10-day-old pups. Basal expression of CRF mRNA in the PVN or secretion of ACTH and corticosterone was not modified by leptin treatment. In contrast, chronically elevated leptin concentrations during the neonatal period significantly lowered CRF expression in the PVN 60 min after stress and reduced the duration of the ACTH response to stress in pups, suggesting that glucocorticoid feedback on the HPA axis might be altered by this treatment. In addition, mothers caring for pups injected with leptin displayed longer bouts of anogenital licking of pups than mothers of vehicle-treated rats. Given that this particular type of pup stimulation

366

has been shown to influence stress responsiveness, it is possible that the maternal response modulates the effects of exogenous leptin treatment. In conclusion, our results demonstrate that the leptin signal is functional during the early developmental period and that leptin can modulate the hormonal response to stress in young rats either by a direct effect on the HPA axis or indirectly through changing some aspects of maternal behavior. © 2000 Academic Press

Key Words: leptin administration; developing rats; stress responses; maternal behavior.

In most mammals, a bidirectional communication between systems maintaining energy balance and neuroendocrine function is well documented. In particular, recent evidence demonstrates that leptin directly affects the activity of the hypothalamic–pituitary–adrenal (HPA) axis at multiple levels. Leptin has been shown to modify ACTH secretion (Raber, Chen, Mucke, and Feng, 1997) and to inhibit adrenal glucocorticoid release in several adult species (Bornstein, Uhlmann, Haidan, Ehrhart-Bornstein, and Scherbaum, 1997). Furthermore, a daily rhythm of plasma leptin concentrations is inversely correlated with diurnal variations in glucocorticold levels in mice as well as in humans (Korbonits, Trainer, Little, Edwards, Kopelman, Besser, Svec and Grossman, 1997; Lincio, Mantzoros, Negrao et al., 1997). More controversial is the effect of leptin on the production and secretion of CRH from the paraventricular nucleus of the hypothalamus (PVN). While in vitro studies have demonstrated a stimulatory effect of leptin on CRH release (Costa, Poma, Martignoni, Nappi, Ur, and Grossman, 1997), most of the in vivo work suggests that leptin is inhibitory to CRH production in the PVN

0018-506X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

367

Effects of Leptin in Rat Pups

of rodents following exposure to a stressor (Heiman, Ahima, Craft, Schoner, Stephens, and Flier, 1997; Huang, Rivest, and Richard, 1998). Leptin represents a metabolic index of energy balance which is strongly influenced by diet (Zhang, Proenca, Maffei, Barone, Leopold, and Feidman, 1994). Increased fat intake in adult humans and rodents results in elevated leptin secretion from increased fat deposits (Ahren, Mansson, Gingrich, and Havel, 1997) and produces significant changes in the activity of the HPA axis (Tannenbaum, Tannenbaum, Brindley, Dallman, McArthur, and Meaney, 1997). Although leptin is present in mothers’ milk and therefore can be transferred to the offspring (Houseknecht, Portocarrero, McGuire, and Beerman, 1997; Casabiell, Pinero, Tome, Peino, Dieguez, and Casanueva, 1997) very little is known about its effect during development. Interestingly, circulating leptin concentrations are elevated over adult values in developing mice and rats (Ahima, Prabakaran, and Flier, 1998) as well as in newborns (Hassink, deLancey, Sheslow, Smith-Kirwin et al., 1997). Leptin has been closely associated with a number of developmental processes, including the onset of reproductive capabilities in rodents (Cheung, Thornton, Kuijper, Weigle, Clifton, and Steiner, 1997) and in children (Blum, Englaro, Hanitsch, Juul et al., 1997). The influence of leptin levels on the stress axis during infancy, however, has not been fully explored. In rodents, the neonatal period constitutes an important period of programming in the HPA axis, during which stressful events can have long-lasting consequences until adulthood (Ladd, Huot, Thrivikraman, Nemeroff, Meaney, and Plotsky, 1999). It is therefore advantageous to maintain a tight regulation over stress responsiveness during this period (Walker, Scribner, Cascio, and Dallman, 1991) and the contribution of elevated leptin concentrations might be crucial in this process. In our previous studies, we have demonstrated that during development, increased fat composition in the milk can similarly raise leptin levels in suckling pups and reduce the magnitude of their responses to stress (Trottier, Koski, Brun, Toufexis, Richard, Walker, 1998), pointing to an effect of leptin on HPA activity during development. These studies provided only indirect evidence for a role of leptin in maintaining low stress responses in developing pups, however. In the present studies, therefore, we sought to investigate a direct role of leptin on body weight gain and fat accumulation as well as on regulation of the HPA axis in developing rats. We investigated whether exogenous leptin is biologically active in neonatal rats and can affect the activity of the HPA axis

and if so, whether changes in HPA activity are mediated by modifications in maternal behavior. We found that exogenous leptin administration produced weight loss and decreased fat deposits in neonates, as documented in adults, and that such treatment did not significantly alter maternal behavior, except for increased anogenital licking prior to nursing. In addition, we confirmed that leptin has a direct effect on the duration, but not the magnitude, of the stress response in neonates, thus participating in the complex and critical regulation of the HPA axis during this period.

MATERIALS AND METHODS Animals Pregnant Sprague–Dawley female rats were received in our animal facility on day 15–16 of gestation (Charles River, St. Constant, Canada) and singly housed in our animal facility in clear plastic cages with food (Purina chow diet, Ralston-Purina, St. Louis, MO) and water available ad libitum. The conditions of the animal facility were kept constant at an ambient temperature of 22–25°C, humidity of 70 – 80%, and with a 12:12 h light:dark cycle (lights on at 08:00 h). The day of parturition was designated day 0, and litters were culled to 10 –11 pups per mother on day 2. Since we have been consistently unable to find sex differences in the hormonal response to stress in developing pups and that circulating leptin levels were similar between male and female pups (male/vehicle, 4.05 ⫾ 0.8, n ⫽ 11, female/vehicle ⫽ 4.43 ⫾ 0.8, n ⫽ 12; ns), we did not manipulate the sex ratio of the litters. Litters were assigned to receive either leptin or vehicle, with dose and frequency depending on the experimental conditions. Litter weight was measured each day at the time of the pups’ injection. All protocols were approved by the Animal Care Committee at McGill University and followed ethical guidelines from the CCAC. Leptin Administration Murine leptin was obtained lyophilized from Peprotech Inc. (Rocky Hill, NJ) and reconstituted in 10 mM Tris buffer at a pH of 9.5. After dissolution, the pH was readjusted to 7.4 by the addition of HCl. Leptin or vehicle (10 mM Tris–HCl, pH 7.4) was injected intraperitoneally in a volume of 50 ␮l, with all injections given in the morning between 08:00 and 10:00 am. In experiment 1, vehicle (three litters) or

368

leptin (1 mg/kg body wt, three litters, or 3 mg/kg body wt, two litters) was administered daily from day 2 to day 9 of life for the determination of body weight gain, fat pad weight, and plasma leptin levels on day 10 of life (i.e., 24 h after the last leptin injection). In this experimental series, litters receiving vehicle or the low dose of leptin (1 mg/kg body wt) were videotaped for analysis of maternal behavior and tested in the stress paradigm on day 10 of life as described below. In experiment 2, we kept the same experimental design for the 1 mg/kg dose (four litters) and the vehicle group (four litters), but in order to prevent excessive reduction in weight gain, pups receiving the high dose of leptin were injected only from days 6 –10 of life and were tested 3 h after the last treatment (three litters). A matching group receiving vehicle from days 6 –10 of life was also included (four litters). We opted for this experimental design in the high- dose group to test stress responses at a time when leptin levels are maximally elevated (3 h compared to 24 h after the last injection). In this experimental series, rats were tested for their stress response and hypothalamic neuropeptide expression as well as their body weight gain and fat pad weight. Litters from the vehicle and low dose of leptin groups were also videotaped for analysis of maternal behavior. Doses of the treatment (either 1 or 3 mg/kg body wt) were calculated daily based on the weight of the litter in the previous day and adjusted for 24 h of pup growth. Maternal Behavior Maternal behavior was recorded on day 8 –9 of lactation and each mother was recorded only once. Mothers from vehicle (n ⫽ 7)- and leptin-injected pups (1 mg/kg body wt, n ⫽ 7) were kept in a quiet room during the entire duration of the experiment and access to the room was prevented during recording sessions. Videotape recordings started 1 h after administration of the treatment to the pups and continued for a period of 8 h during the light portion of the light: dark cycle. Tapes were analyzed to determine the total time each dam spent with her pups (total nesting time divided by 8 h, min/h), the frequency of nesting bouts (total number of nesting bouts divided by 8 h), the average duration of each bout (min/bout), and the average time spent in anogenital grooming of the pups prior to the onset of a nesting bout (total grooming time divided by 8 h, min/h). In addition, milk ejection episodes were identified by recording the stretching reflex of the pups and reorganization of the litter to the nipples that typically occur following a milk ejection

Oates, Woodside, and Walker

reflex (MER). The average frequency of MER was determined per nursing bout for two to three bouts per mother within the first 3 h of recording. Stress Testing and Blood and Tissue Collection All experiments were conducted between 10:00 h and 13:00 h to minimize the effect of daily hormonal fluctuations on stress responses and pups were tested on day 10 of life. Pups from the different treatment groups (VEH days 2–9 ⫽ seven litters, 17–18 pups/ time point; 1 mg/kg body wt dose days 2–9 ⫽ seven litters, 17–18 pups/time point; 3 mg/kg body wt dose days 6 –10 ⫽ three litters, 6 –7 pups/time point; VEH days 6 –10 ⫽ three litters, 6 –7 pups/time point) were separated from their mothers 20 –30 min prior to the onset of stress and placed in a normal housing cage kept in a quiet room. This allowed randomization of the litters within one experimental group and reduction of the disturbances generally associated with repeated intrusion into the mother-litter’s cage. As much as possible, all disturbances prior to and during testing were avoided. Control animals were rapidly weighed and sacrificed within 5 s of removal from their cage. Experimental pups were weighed and exposed to ether vapors for a total duration of 3 min, including 1 min in a glass jar saturated with ether vapors and 2 min under a nose cone containing cotton impregnated with ether. After completion of the 3-min exposure to ether, pups were returned to clean cages and sacrificed at 5, 30, or 60 min after the onset of the stressor. Trunk blood was collected in Eppendorf tubes containing 10 ␮l of EDTA (60 mg/ml) and plasma was kept frozen at ⫺20°C prior to being assayed for plasma leptin, ACTH, and corticosterone concentrations. The left retroperitoneal fat pad was dissected and weighed. Brains from the control group (0 min) and 60-min time points (VEH and 3 mg/kg body wt dose) were rapidly dissected and postfixed in a solution of 4% paraformaldehyde in phosphate buffer (0.05 M, pH 7.4, 4°C) for 4 days followed by immersion in a solution of 10% sucrose in phosphate buffer for 2 days at 4°C. Brains were then frozen at ⫺80°C prior to being processed for CRF in situ hybridization. In Situ Hybridization of CRF mRNA in the PVN In situ hybridization for CRF mRNA was performed according to a protocol described earlier for similar neuropeptides (Laurent-Huck and Felix, 1991). The CRF probe was a 45-base oligomer, complementary to

369

Effects of Leptin in Rat Pups

bases 523 to 667 of the second exon of the CRF gene (Sheldon Biotechnology Center, Montreal, PQ). The probes were 3⬘-end labeled with 35S and terminal deoxynucleotydltransferase using a kit from Boerhinger Manheim (Laval, Quebec) and purified on Nensorb columns (Dupont NEN, Boston, MA). Brain sections (25 ␮m) of vehicle- and leptin-treated rats were collected onto slides coated with poly-L-lysine and stored at ⫺80°C until hybridization. Sections were fixed with 4% paraformaldehyde in 0.1 M phosphate buffer for 10 min and dehydrated in graded ethanol prior to being submitted to a series of washes in saline sodium citrate (SSC) 4⫻ containing 1% Denhardt’s solution (1⫻ 1 h), 0.2 M triethanolamine (TEA) and 18% NaCl (1⫻ 5 min), 0.2 M TEA, 18% NaCl, and 0.25% acetic anhydride (1 ⫻ 10 min), and 2⫻SSC (3 ⫻ 5 min). The sections were then dehydrated in graded ethanol, rinsed in chloroform, followed by ethanol 100 and 95%, and air dried. Sections were incubated with 75 ␮l of hybridization solution containing 7.5–9.0 ⫻ 10 5 cpm of the 35S-labeled probe and coverslipped before being incubated overnight at 42°C. The hybridization solution consisted of 0.6 M NaCl, 0.01 M Tris buffer, 500 ␮l/ml formamide, Denhardt’s solution (1⫻), 0.1 M phosphate buffer, Sarkosyl (1⫻), 1 mM EDTA, 0.5 mg/ml tRNA, and 0.25 mg/ml salmon sperm DNA. The imperfect hybrids were disrupted by successive washes in 4⫻ SSC, 1⫻SSC, 0.5⫻SSC, and 2⫻SSC and the sections were dehydrated again in graded ethanol and air dried before exposure to Beta-Max Hyperfilm (Amersham, Arlington Heights, IL) for 6 days at ⫺80°C. Radioactive standards prepared from brain paste with high activity 3H and 14C were exposed simultaneously. Hybridization signal on the autoradiograms was quantified from sections placed in the medial portion of the PVN and using computerized densitometry by means of an MCID image analyzer system (Imaging Research Inc., Ste. Catherine, ON). Hormone Assays Plasma ACTH levels were measured by specific radioimmunoassay as described previously (Walker, Akana, Cascio, and Dallman, 1990). The limit of detection of the assay was 15.6 pg/ml and the inter- and intraassay variabilities were 26 and 8%, respectively. Plasma corticosterone concentrations were determined by RIA using a kit from ICN Biomedicals (Costa Mesa, CA) with small modifications. The limit of detection was 0.2 ␮g/dl and the inter- and intraassay variabilities were 12 and 3%, respectively. Plasma leptin levels were measured in basal samples (0 min)

FIG. 1. Plasma concentrations of leptin in 10-day-old pups injected with either vehicle (VEH: 10 mM Tris–HCl, open bars) or leptin at doses of 1 mg/kg body wt (hatched bars) or 3 mg/kg body wt (dark bars) from days 2 to 9 of life. Leptin concentrations were obtained 24 h after the last treatment. Values are the means ⫾ SEM of 6 –12 animals per group taken from the 0-min time point. **P ⬍ 0.01 compared to vehicle-injected pups (one-way ANOVA). #P ⬍ 0.05 compared to 1 mg/kg body wt dose.

by specific RIA using a kit from Linco Research (St. Charles, MO). The limit of detection was 0.5 ng/ml and the interassay variability was 9%. Statistics All results were analyzed using Student’s t test or analysis of variance (ANOVA) where appropriate. Significant interactions were determined by F tests for simple main effect, with pairwise comparisons performed using Tukey’s honestly significant difference test. The level of significance was set as P ⬍ 0.05. All values are expressed as means ⫾ SEM.

RESULTS Effect of Leptin Injection on Plasma Leptin Levels Chronic injection of leptin in pups between days 2 and 9 of life significantly elevated plasma leptin levels measured 24 h after the last injection (Fig. 1). There was a dose-related increase in plasma leptin concentrations between the 1 mg/kg (8.04 ⫾ 1.49 ng/ml) and the 3 mg/kg (13.6 ⫾ 2.22 ng/ml) dose compared to vehicle-injected (3.27 ⫾ 0.69 ng/ml) pups. In pups receiving a chronic leptin treatment (1 mg/kg body wt) from days 2–10, the last injection of the leptin treatment on day 10, 3 h prior to sacrifice, resulted in elevated leptin concentrations compared to those of vehicle-injected pups (vehicle, 4.11 ⫾ 0.65 ng/ml; leptin, 70.2 ⫾ 8.0, P ⬍ 0.001). A similar increase in leptin levels could be observed in pups injected with the high dose of leptin (3 mg/kg body wt) from days 6 –10

370

Oates, Woodside, and Walker

tivity of leptin in pups was confirmed by measurement of retroperitoneal fat deposition in these pups, which was significantly and dose dependently reduced by leptin treatment (Fig. 2, bottom, normalized to body weight; P ⬍ 0.001). When fat pad was expressed in absolute values, a similar effect was observed (P ⬍ 0.001, data not shown). Maternal Behavior toward Vehicle- or Leptin-Injected Litters Because leptin was affecting fat pad weight and body weight gain in the pups, we hypothesized that an increased energy expenditure induced by leptin treatment could modify some aspects of maternal behavior. As seen in Fig. 3, the amount of time spent in anogenital grooming was significantly greater for dams caring for leptin-injected pups than for those nursing vehicle-injected pups (Fig. 3 top; P ⬍ 0.05).

FIG. 2. Effects of chronic vehicle (open bars) or leptin injection (1 mg/kg body wt, hatched bars and 3 mg/kg body wt, dark bars) between days 2 and 9 of life on daily weight gain (top) and retroperitoneal fat pad weight (bottom). Daily weight gain represents an average of seven values obtained over 24-h periods (days 2–9) for two litters in the 3 mg/kg body wt regimen and seven litters in the other groups. Retroperitoneal fat pad weight was measured on day 10 of life, i.e., 24 h after the last injection. Values for the fat pad were normalized for body weight and represent the means ⫾ SEM of 12–38 determinations per group. *P ⬍ 0.05; ***P ⬍ 0.001 compared to vehicle-treated pups (one-way ANOVA).

of life (vehicle, 3.24 ⫾ 0.7 ng/ml; leptin, 260 ⫾ 26.4, P ⬍ 0.001). Effect of Leptin Administration on Fat Pad and Body Weight Daily weight gain was calculated as the average of seven values obtained over 24-h periods between days 2 and 9. For each treatment (VEH, seven litters; 1 mg/kg dose, seven litters; 3 mg/kg dose, two litters) litter weight rather than individual pup’s weight was recorded. As shown in Fig. 2 (top), leptin injection from days 2–9 at the 1 mg/kg body wt dose significantly reduced daily body weight gain of the pups compared to that of vehicle injection groups (P ⬍ 0.05). For the 3 mg/kg body wt dose, we observed a similar effect; however, because we had only two litters for this group we were unable to perform statistical analysis including this group. The biological ac-

FIG. 3. Effects of chronic vehicle (open bars) or leptin (1 mg/kg body wt, dark bars) injection of pups on maternal behavior. Treatment was administered from days 2 to 9 of life and maternal behavior was recorded for 8 h during the light period on days 8 and 9 of lactation. All litters were homogeneous for treatment and each mother was only recorded once. Values represent the means ⫾ SEM of seven mothers per group. *P ⬍ 0.05 compared to mothers from vehicle-injected pups (Student’s t test).

Effects of Leptin in Rat Pups

371

FIG. 4. Plasma ACTH response to a 3-min ether stress in 10-day-old pups chronically injected with either vehicle (open circles) or leptin (closed squares). (Top) Pups received either VEH or 1 mg/kg body wt of leptin between days 2 and 9 of life and (bottom) pups received either VEH or 3 mg/kg body wt of leptin between days 6 and 10 of life. In the left panels, the time course of the ACTH response is depicted and in the right panels, the area under the curve (AUC) for each treatment is shown. Leptin had a significant effect on ACTH secretion at both doses. At the 1 mg/kg body wt dose, ANOVAs showed time (P ⬍ 0.0001) and treatment (P ⬍ 0.0001) effects as well as a time ⫻ treatment interaction (P ⬍ 0.01). At the 3 mg/kg body wt dose, ANOVAs showed time (P ⬍ .0001) and treatment (P ⬍ 0.001) effects as well as a time ⫻ treatment interaction (P ⬍ 0.01). Values represent the means ⫾ SEM of 18 –20 pups/time point under the 1 mg/kg body wt condition (three experiments pooled) and 6 –7 pups/time point under the 3 mg/kg body wt condition (two experiments pooled). **P ⬍ 0.01 compared to vehicle-treated pups (ANOVA, Tukey HSD test).

There were no significant differences between treatment groups in any other behavioral parameters measured, including the number of milk ejection reflexes per nesting bout (vehicle, 2.9 ⫾ 0.4; leptin, 2.8 ⫾ 0.4 MER/bout, ns), total nesting time per h (vehicle, 40.58 ⫾ 4.4 min/h; leptin, 44.4 ⫾ 3.3, ns), and the average duration of each nesting bout (Fig. 3, bottom). However, for this last measure, there was a trend for mothers caring for leptin-injected pups to spend more time in the nest than their vehicle counterparts. Effect of Leptin on Hormonal Responses to Stress and Hypothalamic CRF mRNA Levels Plasma ACTH responses to ether stress in 10-dayold pups injected chronically with either vehicle or

leptin are displayed in Fig. 4. Measurements shown in Fig. 4 (top) are from pups that received leptin at a dose of 1 mg/kg body wt from days 2–9 of life (three experiments pooled) and those in Fig. 4 (bottom), from pups that received leptin at a dose of 3 mg/kg body wt from days 6 –10 of life (two experiments pooled). There was a significant main effect of dose (1 mg/kg, P ⬍ 0.0001; 3 mg/kg, P ⬍ 0.001) and of time (P ⬍ 0.0001), as well as a significant interaction between dose and time (P ⬍ 0.01, ANOVA) for both treatment regimens. With both doses, the peak ACTH response was not altered, but leptin-injected pups showed a faster return to baseline levels than vehicle-injected pups. The reduced magnitude of the total pituitary response to stress in leptin-treated pups was obvious

372

Oates, Woodside, and Walker

TABLE 1 Basal and Stress-Induced Plasma Corticosterone Concentrations in 10-Day-Old Pups Treated Chronically with Either Vehicle or Leptin Group

Minutes

1 mg/kg days 2–9

3 mg/kg days 6–10

Vehicle

0 5 30 60 0 5 30 60

1.24 ⫾ 0.11 2.31 ⫾ 0.22 5.52 ⫾ 0.60 4.41 ⫾ 0.63 1.12 ⫾ 0.11 2.01 ⫾ 0.25 4.94 ⫾ 0.60 6.01 ⫾ 1.33

2.05 ⫾ 0.24 2.02 ⫾ 0.07 4.76 ⫾ 0.73 6.19 ⫾ 1.04 3.13 ⫾ 0.45 3.78 ⫾ 0.38 6.49 ⫾ 0.96 4.02 ⫾ 0.91

Leptin

Note. Values represent the mean ⫾ SEM of 18 –20 pups/time point in the 1 mg/kg body wt group and 6 –7 pups/time point in the 3 mg/kg body wt group. There were no significant differences in basal or stimulated corticosterone secretion between treatments. ANOVA showed a significant time effect for both doses (P ⬍ 0.01), but no treatment effect (P ⬎ 0.05). Time ⫻ treatment interaction was only significant for the highest dose (3 mg/kg body wt).

when expressed as a function of the area under the curve (AUC, Fig. 4, right panels). Although leptin treatment significantly reduced the ACTH response to stress, it did not affect plasma ACTH and corticosterone under resting conditions. Plasma corticosterone levels under basal conditions or following ether stress were not different between treatment groups in either leptin regimen (Table 1). At the 0-min time point, VEH-injected pups on the days 6 –10 injection schedule (Fig. 4, bottom, and Table 1) exhibited higher levels of ACTH and corticosterone than those from the days 2–9 injection regimen (Fig. 4, top, and Table 1). This is likely to reflect a residual effect of the injection given 3 h prior to the 0-min time point in the days 6 –10 regimen. To determine whether the effect of leptin on stress responsiveness requires a chronic administration, we tested the ACTH response of 10-day-old pups to ether stress following acute leptin administration (injections at ⫺18 and ⫺3 h prior to stress, 3 mg/kg dose) in earlier experiments. We found no significant differences between leptin- or vehicle-injected pups in the magnitude of the response (peak ACTH secretion at 5 min: vehicle, 162.6 ⫾ 29.2 pg/ml; leptin, 183.9 ⫾ 36.2, ns) or the duration of the response (60-min time point: vehicle, 158.5 ⫾ 42.6 pg/ml; leptin, 149.8 ⫾ 9.0, ns). Although circulating levels of leptin were also elevated in this acute injection paradigm (vehicle, 3.38 ⫾ 0.35 ng/ml; leptin, 55.4 ⫾ 6.5, P ⬍ 0.001), this experiment confirmed that expression of the effects of leptin

on ACTH secretion require a chronic exposure to leptin as described in Fig. 4 (bottom). As for basal ACTH and corticosterone secretion, basal CRF mRNA expression in the hypothalamic PVN was not affected by leptin treatment (VEH days 6 –10, 189.4 ⫾ 20.4 nCi/g; leptin, 3 mg/kg days 6 –10, 173.9 ⫾ 13.2, ns). In contrast, 60 min after the onset of stress, stimulated expression of CRF mRNA was significantly reduced by leptin administration in the PVN of neonatal rats (Fig. 5). Because in situ hybridization for the 0- and 60-min time points were not performed in the same series, we are only able to compare treatment effect and not time effect on CRF mRNA levels.

DISCUSSION In these studies, we demonstrate that leptin modulates indices of both energy balance and neuroendocrine functions in neonatal rats. Chronic leptin administration during the first 9 or 10 days of life significantly decreased fat deposition and body weight gain and reduced the hypothalamic–pituitary response to ether stress in neonates. In developing rats, the first 2–3 weeks of life represent a critical period for a number of maturational processes which are strongly favored by a positive energy balance (Vitiello, DiBenedetta, Cioffi, and Gombos, 1980). Intact pups steadily increase their body weight (Walker and Aubert, 1988) and fat deposits during the suckling period, despite high levels

FIG. 5. Leptin reduces the stress-induced expression of CRF mRNA in the paraventricular nucleus of 10-day-old pups. Neonates were treated from days 6 to 10 of life with either vehicle (VEH) or leptin (3 mg/kg body wt) and brains were collected for in situ hybridization 60 min following exposure to ether stress. Leptin treatment did not affect basal CRF mRNA levels (see Results). Values represent means ⫾ SEM of 7–12 sections from each of 5– 6 pups/group. *P ⬍ 0.05 (Student’s t test).

Effects of Leptin in Rat Pups

of circulating leptin (Trottier et al., 1998). In our experiments, we found that exogenous administration of leptin in pups during the first 9 or 10 days of life significantly reduced body weight gain and fat pad weight. This decrease in body weight gain caused by pharmacological increases in circulating leptin levels in suckled pups could result either from a decrease in the amount of milk ingested or from an increased energy expenditure. From our maternal behavior data, we found no evidence for differences in milk ejection frequency in mothers who were nursing leptin-injected litters compared to those nursing vehicle-injected litters. Provided that pup treatment did not alter the amount of milk available per milk ejection episode, this result indicates that milk availability was not modified by the treatment and also that the intensity of the suckling stimulus was probably comparable in both groups. These results are in agreement with earlier studies conducted in 9- to 14-day-old fooddeprived pups, which failed to detect any differences in acute food intake between leptin- and vehicletreated rats (Stehling, Doring, Ertl, Preibisch, and Schmidt, 1996). In this report, it was also demonstrated that increased energy expenditure induced by leptin administration in artificially reared pups was responsible for the reduction in body weight gain and fat deposition and also led to an increase in core body temperature (Stehling et al., 1996). It is well known that changes in litter temperature in the nest can have important consequences on maternal behavior, in particular on nesting bout duration (Woodside and Leon 1980). We thus hypothesized that leptin administration in pups would lead to an increase in litter temperature and decrease nesting bout duration as well as total nesting time. This was not the case, however, since dams nursing leptin-treated pups showed a tendency to have longer nesting bouts than controls and were not different in their total nesting time in an 8-h period. Our observation that exogenous leptin treatment causes a dramatic decrease in retroperitoneal fat pad weight is particularly interesting because we previously showed that pups nursing on high-fat milk have circulating levels of leptin similar to those obtained in our chronic paradigm, but continued to increase their body weight and deposit more fat than pups nursing on control diet milk (Trottier et al., 1998). The fact that exogenous leptin appears to have the opposite effect on fat deposition than endogenous increases due to a high-fat diet lends itself to various interpretations. First, the decreased fat pad in leptin-injected animals could be due to increased energy expenditure in these

373

animals that is not compensated by ingestion of a higher fat diet, as it is for pups nursing on a high-fat milk. Thus, a negative energy balance is induced by exogenous leptin administration, but not in pups fed a high-fat milk. Second, the magnitude and pattern of leptin exposure might be important in determining the direction of changes in energy balance. In the case of exogenous leptin administration, we observed a large increase in circulating leptin concentrations (in the pharmacological range) 3 h after an injection and plasma levels that were still two- to three-fold higher than those of vehicle-injected pups 24 h after an injection. This pattern of changes in circulating leptin levels is likely very different from that of tonically elevated endogenous physiological production and it is possible that the high levels of leptin present immediately after injection may be sufficient to induce specific receptor changes or affect other metabolic systems that do not occur with the more consistent levels seen in a high-fat milk paradigm. Finally, it is possible that factors other than leptin which are present in high-fat milk may counteract the effect of leptin on fat deposition in pups. In addition to its effects on weight gain and fat deposition, leptin has been documented to modify the activity of the HPA axis in adult rodents (Heiman et al., 1997). However, nothing is known about leptin’s effects in neonatal animals. In developing pups, we observed an inverse relationship between circulating leptin concentrations and the magnitude of the stress response (Trottier et al., 1998), which suggested that leptin could be a direct modulator of the activity of the HPA axis. Thus, the present studies were designed to test the direct effect of exogenous leptin administration on basal and stimulated HPA activity in rat pups. We found that exogenous leptin did not modify the peak ACTH response to ether stress in 10-day-old pups, as we previously reported for increased endogenous secretion, but induced a significantly faster return of plasma ACTH to baseline levels after termination of the stressor. In addition, chronic leptin treatment significantly reduced CRF expression in the PVN 60 min after the onset of stress. Despite the higher levels of leptin in the 3 mg/kg dose compared to the 1 mg/kg dose at the time of testing, we still observed similar effects of both dosages on ACTH secretion. We believe that expression of the effect of leptin on HPA activity is not necessarily related to circulating levels of leptin at the time of stress, but is more dependent upon a chronic elevation in circulating leptin levels over at least 4 –7 days prior to testing. This is in contrast with the inhibitory effect of leptin

374

on the HPA axis following an acute injection demonstrated in adult rats (Heiman et al., 1997). In neonates, earlier experiments testing the acute effect of leptin in 10-day-old rats and achieving plasma leptin levels around 55 ng/ml failed to show a significant effect on basal or stimulated ACTH and corticosterone secretion. In our chronic experiments, the effect of leptin was specific to stimulated HPA activity since we did not see any differences either in basal CRF synthesis in the PVN or in basal ACTH or corticosterone levels in leptin-treated animals compared to the vehicle-treated group. The faster return of plasma ACTH secretion to baseline levels after exposure to stress is reminiscent of a number of situations where changes in glucocorticoid feedback efficiency have been documented (Meaney, Aitken, Sharma, Viau, and Sarrieau, 1989). Therefore, one possible mechanism mediating the changes in stress-induced ACTH response in leptin-treated animals might be an enhanced glucocorticoid feedback system. If these pups are indeed in a state of negative energy balance due to leptin-induced increases in energy expenditure, it is reasonable to suppose that a more efficient glucocorticoid feedback system would be adaptive in preventing prolonged corticosterone release and exacerbation of this negative energy balance condition. Such an adaptation of the HPA axis would compensate for the metabolic effects of leptin. Although we found no differences between treatment groups with respect to basal or stimulated corticosterone levels, glucocorticoid feedback efficiency could still be enhanced by changes in circulating levels of corticosterone-binding-globulin (CBG) (Sakly and Koch, 1983) or a change in the number or sensitivity of glucocorticoid receptors in the brain and pituitary (Meaney, O’Donnell, Viau, Bhatnagar, Sarrieau, Smythe, Shanks, and Walker, 1993). It is noteworthy that the effects of leptin in a state of negative energy balance are to limit the duration of glucocorticoid secretion, while in a state of positive energy balance, elevated leptin secretion reduces the magnitude rather than the duration of ACTH secretion (Trottier et al., 1998). These observations suggest that additional metabolic factors or neuropeptides secreted in response to variations in energy balance modulate the effect of leptin on the HPA axis. Alternatively, the reduced expression of CRF and secretion of ACTH at the 60min time point in leptin-injected pups might suggest that synthesis of CRF in these pups cannot replenish secreted stores of CRF as rapidly as in vehicle-injected rats. Indeed, leptin has been suggested to have a dual effect on CRF in adult rodents (Heiman et al., 1997),

Oates, Woodside, and Walker

one to rapidly decrease the readily releasable storage pool (which would reduce ACTH secretion) and a second effect to inhibit CRF synthesis in the PVN (which would decrease levels of CRF mRNA). It is possible that both of these effects of leptin on CRF are already observed in our young neonates. One important finding of our studies is that leptin treatment of the litter elicited significantly more anogenital licking from the mother compared to litters injected with vehicle. As demonstrated for the genderrelated differences in maternal licking (Moore, 1982), it is possible that olfactory cues or altered locomotion in leptin-treated pups might enhance maternal licking behavior toward these pups. Alternatively, increased ultrasonic vocalizations in leptin-treated pups might signal increased sympathetic activity and thermogenesis in these animals and stimulate maternal attention (Hofer, and Shair, 1993). Maternal licking and grooming behavior has long been thought to be important in developmental programming processes, including the development of the HPA axis (Rosenfeld, Suchecki, and Levine, 1992; Hofer, 1994). More recently, it was shown that rat dams who display more licking/ grooming behavior toward their litters have pups whose HPA response to stress is decreased as adults (Liu, Diorio, Tannenbaum, Caldji, Freedman, Sharma, Pearson, Plotsky, and Meaney, 1997). Thus, one of the beneficial effects of elevated levels of leptin during development might be to elicit specific behavioral responses in the mother which are directed toward reducing stress responsiveness in the adult offspring. In conclusion, we have shown that exogenously administered leptin is biologically active in neonatal rat pups and that it alters energy balance, resulting in a decrease in fat deposition and body weight gain. As demonstrated indirectly by the absence of differences in milk ejection frequency, these effects of leptin are likely mediated by increased energy expenditure rather than by decreased food intake. We have further demonstrated that leptin can alter stress responsiveness in neonates, by causing a faster return of stimulated ACTH secretion to baseline. This suggests that glucocorticoid feedback efficiency is enhanced by elevated leptin concentrations, an adaptive mechanism which might prove to be of critical importance in a state of negative energy balance. Further elucidation of the expression pattern and signaling of the leptin receptor and investigation of CBG and glucocorticoid receptor levels in this paradigm will bring us closer to a definitive role for leptin during rat development. Finally, leptin-induced increase in anogenital licking of the pups might participate in the long-term regula-

Effects of Leptin in Rat Pups

tion of adrenocortical activity in rats from development to adulthood.

ACKNOWLEDGMENTS We thank Ms. Ning Huang for her help in the animal room for some of the experiments and Dr. W. Engeland (University of Minnesota, Mineapolis, MN) for his gift of ACTH antiserum. We are grateful to Dr. D. Richard (University Laval, Quebec, Canada) for helping us with the early phase of the in situ hybridization technique. This work was supported by a grant from the Medical Research Council of Canada to C.D.W. (No. MT14994).

REFERENCES Ahima, R. S., Prabakaran, D., and Flier, J. S. (1998). Postnatal leptin surge and regulation of circadian rhythm of leptin by feeding: Implications for energy homeostasis and neuroendocrine function. J. Clin. Invest. 101, 1020 –1027. Ahren, B., Mansson, S., Gingrich, R. L., and Havel, P. J. (1997). Regulation of plasma leptin in mice: Influence of age, high-fat diet, and fasting. Am. J. Physiol. 273, R113–R120. Blum, W. F., Englaro, P., Hanitsch, S., Juul, A., Hertel, N. T., Muller, J., Skakkebaek, N. E., Heiman, M. L., Birkett, M., Attanasio, A. M., Kiess, W., and Rascher, W. (1997). Plasma leptin levels in healthy children and adolescents: Dependence on body mass index, body fat mass, gender, pubertal stage, and testosterone. J. Clin. Endocrinol. Metab. 82, 2904 –2910. Bornstein, S. R. (1997). Is leptin a stress related peptide? Nature Med. 3, 937. Bornstein, S. R., Uhlmann, K., Haidan, A., Ehrhart-Bornstein, M., and Scherbaum, W.A. (1997). Evidence for a novel peripheral action of leptin as a metabolic signal to the adrenal gland. Diabetes 46, 1235–1238. Casabiell, X., Pineiro, V., Tome, M. A., Peino, R., Dieguez, C., and Casanueva, F. F. (1997). Presence of leptin in colostrum and/or breast milk from lactating mothers: A potential role in the regulation of neonatal food intake. J. Clin. Endocrinol. Metab. 82, 4270 – 4273. Cheung, C. C., Thornton, J. E., Kuijper, J. L., Weigle, D. S., Clifton, D. K., and Steiner, R.A (1997). Leptin is a metabolic gate for the onset of puberty in the female rat. Endocrinology 138, 855– 858. Costa, A., Poma, A., Martignoni, E., Nappi, G., Ur, E., and Grossman, A. (1997). Stimulation of corticotropin-releasing hormone release by the obese (ob) gene product, leptin, from hypothalamic explants. Neuroreport 8, 1131–1134. Frederich, R. C., Hamann, A., Anderson, S., Lollmann, B., Lowell, B. B., and Flier, J. S. (1995). Leptin levels reflect body lipid content in mice: Evidence for diet-induced resistance to leptin action. Nature Med. 1, 1311–1314. Hassink, S. G., deLancey, E., Sheslow, D. V., Smith-Kirwin, S. M., O’Connor, D. M., Considine, R.V., Opentanova, I., Dostal, K., Spear, M. L., Leef, K., Ash, M., Spitzer, A. R., and Funanage, V. L. (1997). Placental leptin: An important new growth factor in intrauterine and neonatal development? Pediatrics 100, 1–10. Heiman, M. L., Ahima, R. S., Craft, L. S., Schoner, B., Stephens, T. W., and Flier, J. S. (1997). Leptin inhibition of the hypothalamic-

375

pituitary-adrenal axis in response to stress. Endocrinology 138, 3859 –3863. Hofer, M. A. (1994). Early relationships as regulators of infant physiology and behavior. Acta Paediatr. Suppl. 397, 9 –18. Hofer, M. A., and Shair H. N. (1993). Ultrasonic vocalization, laryngeal braking, and thermogenesis in rat pups: A reappraisal. Behav. Neurosci. 107, 354 –362. Houseknecht, K. L., McGuire, M. K., Portocarrero C. P., McGuire, M.A., and Beerman, K. (1997). Leptin is present in human milk and is related to maternal plasma leptin concentration and adiposity. Biochem. Biophys. Res. Commun. 240, 742–747. Huang, Q., Rivest, R., and Richard, D. (1998). Effects of leptin on corticotropin-releasing factor (CRF) synthesis and CRF neuron activation in the paraventricular hypothalamic nucleus of obese (ob/ob) mice. Endocrinology 139, 1524 –1532. Korbonits, M., Trainer, P. J., Little, J. A., Edwards, R., Kopelman, P. G., Besser, G. M., Svec, F., and Grossman, A.B. (1997). Leptin levels do not change acutely with food administration in normal or obese subjects, but are negatively correlated with pituitaryadrenal activity. Clin. Endocrinol. 46, 751–757. Ladd, C. O., Huot, R. L., Thrivikraman, K. V., Nemeroff, C. B., Meaney, M. J., and Plotsky, P. M. (1999). Long-term behavioral and neuroendocrine adaptations to adverse early experience. In E. Mayer and C. Saper (Eds.), The Biological Basis for Mind Body Interactions, Progress in Brain Research. Elsevier, Amsterdam (in press). Laurent-Huck, F. M., and Felix, J. M. (1991). Measurement of oxytocin and vasopressin gene expression by in situ hybridization. Methods Neurosci. 5, 159 –182. Lincio, J., Mantzoros, C., Negrao, A. B., Giovani, C., Wong, M.-L., Bongiorno, P. B., Chrousos, G. P., Karp, B., Allen, C., Lier, J. S., and Gold, P. W. (1997). Human leptin levels are pulsatile and inversely related to pituitary-adrenal function. Nature Med. 3, 575–579. Liu, D., Diorio, J., Tannenbaum, B., Caldji, C., Francis, D., Freedman, A., Sharma, S., Pearson, D., Plotsky, P. M., and Meaney, M. J. (1997). Maternal care, hippocampal glucocorticoid receptors, and hypothalamic-pituitary-adrenal responses to stress. Science 277, 1659 –1662. Meaney, M. J., Aitken, D. H., Sharma, S., Viau, V., and Sarrieau, A. (1989). Postnatal handling of rats alters adrenocortical negative feedback: A model for individual differences in the neuroendocrine response to stress. Neuroendocrinol. 50, 597– 604. Meaney, M. J., O’Donnell, D., Viau, V., Bhatnagar, S., Sarrieau, A., Smythe, J., Shanks, N., and Walker, C.-D. (1993). Corticosteroid receptors in the rat brain and pituitary during development and hypothalamic-pituitary-adrenal function. In Ian Zagon and Patricia McLaughlin (Eds.), Receptors in the Developing Nervous System, Vol. 1: Growth Factors and Hormones. Chapman & Hall, London. Moore, C .L. (1982). Maternal behavior of rats is affected by hormonal condition of pups. J. Comp. Physiol. Psychol. 96(1), 123–129. Raber, J., Chen, S., Mucke, L., and Feng, L. (1997). Cortocotropinreleasing factor and adrenocorticotropic hormone as potential mediators of Ob effects. J. Biol. Chem. 272, 15057–15060. Rosenfeld, P., Suchecki, D., and Levine, S. (1992). Multifactorial regulation of the hypothalamic-pituitary adrenal axis during development. Neurosci. Biobehav. Rev. 16, 553–568. Sakly, M., and Koch, B. (1983). Ontogenetical Variations of transcortin modulate glucocorticoid receptor function and corticotropic activity in the pituitary gland. Horm. Metab. Res. 15, 92–96. Schrauwen, P., van Marken, Lichtenbelt, W. D., Westerterp, K. R.,

376

and Saris, W. H. M. (1997). Effect of diet composition on leptin concentration in lean subjects. Metabolism 46, 420 – 424. Spinedi, E., and Gaillard, R. C. (1998). A regulatory loop between the hypothalamo-pituitary-adrenal (HPA) axis and circulating leptin: A physiological role of ACTH. Endocrinology 139, 4016 – 4020. Stehling, O., Doring, H., Ertl, J., Preibisch, G., and Schmidt, I. (1996). Leptin reduces juvenile fat stores by altering the circadian cycle of energy expenditure. Am. J. Physiol. 271, R1770 –1774. Tannenbaum, B. M., Tannenbaum, G. S., Brindley, D. N., Dallman, M. F., McArthur, M. D., and Meaney, M. J. (1997). High-fat feeding impairs both basal and stress-induced hypothalamic-pituitary-adrenal responsiveness in the rat. Am. J. Physiol. 273, E1168 –E1177. Trottier, G., Koski, K. G., Brun, T., Toufexis, D. J., Richard, D., and Walker. C.-D. (1998). Increased fat intake during lactation modifies hypothalamic-pituitary-adrenal responsiveness in developing rat pups: A possible role for leptin. Endocrinology 139, 3704 –3711. Van Dijk, G., Donahey, J. C. K., Thiele, T. E., Scheurink, A. J. W., Steffens, A. B., Wilkinson, C. W., Tenenbaum, R., Campfield, L. A., Burn, P., Seely, R. J., and Woods, S. C. (1997). Central leptin stimulates corticosterone secretion at the onset of the dark phase. Diabetes 46, 1911–1946. Vitiello, F., Di Benedetta, C., Cioffi, L. A., and Gombos, G. (1980). Malnutrition and brain development. In C. Di Benedetta, (Ed.),

Oates, Woodside, and Walker

Multidisciplinary Approach to Brain Development. pp. 293. Elsevier/ North Holland Biomedical Press. Wabitsch, M., Jensen, P. B., Blum, W. F., Christoffersen, C. T., Englaro, P., Heinze, E., Rascher, W., Teller, W., Tornqvist, H., and Hauner, H. (1996). Insulin and cortisol promote leptin production in cultured human fat cells. Diabetes 45, 1435–1438. Walker, C.-D., and Aubert, M. L. (1988). Effects of early undernutrition and handling on the adrenocortical activity of neonatal rats. Life Sci. 43, 1983–1990. Walker, C.-D., Akana, S. F., Cascio, C. S., and Dallman, M. F. (1990). Adrenalectomy in the neonate: Adult-like adrenocortical system responses to both removal and replacement of corticosterone. Endocrinology 127, 832– 842. Walker, C.-D., Scribner, K. A., Cascio, C. S., and Dallman, M. F. (1991). The pituitary-adrenocortical system of neonatal rats is responsive to stress throughout development in a time-dependent and stressor specific fashion. Endocrinology 128, 1385–1395. Woodside, B., and Leon, M. (1980). Thermoendocrine influences on maternal behavior in rats. J. Comp. Physiol. Psychol. 94, 41– 60. Zarjewska, K. E., Cusin, I., Sainsbury, A., Rohner-Jeanrenaud, F., and Jeanrenaud, B. (1997). Glucocorticoids as counterregulatory hormones of leptin. Diabetes 46, 717–719. Zhang, Y. Y., Proenca, R., Maffei, M., Barone, M., Leopold, L., and Friedman, J. M. (1994). Positional cloning of the mouse obese gene and its human homolog. Nature 372, 425– 432.