We examined the effect of six doses of dexamethasone (Dex) administered daily (2–7 days of age) to postnatal rats on body weight gain, food and water intake, peripheral hormonal/metabolic milieu, and hypothalamic neuropeptides that regulate food intake. We observed a Dex-induced acute (3 days of age) suppression of endogenous corticosterone and an increase in circulating leptin concentrations that were associated with a decrease in body weight in males and females. Followup during the suckling, postsuckling, and adult stages (7–120 days of age) revealed hypoleptinemia in males and females, and hypoinsulinemia, a relative increase in the glucose-to-insulin ratio, and a larger increase in skeletal muscle glucose transporter (GLUT 4) concentrations predominantly in the males, reflective of a catabolic state associated with a persistent decrease in body weight gain. The increase in the glucose-to-insulin ratio and hyperglycemia was associated with an increase in water intake. In addition, the changes in the hormonal/metabolic milieu were associated with an increase in hypothalamic neuropeptide Y content in males and females during the suckling phase, which persisted only in the 120-day-old female with a transient postnatal decline in α-melanocyte-stimulating hormone and corticotropin-releasing factor. This increase in neuropeptide Y (NPY) during the suckling phase in males and females was associated with a subsequent increase in adult food intake that outweighed the demands of body weight gain. In contrast to the adult hypothalamic findings, cerebral ventricular dilatation was more prominent in adult males. We conclude that postnatal Dex treatment causes permanent sex-specific changes in the adult phenotype, setting the stage for future development of diabetes (increased glucose:insulin ratio), obesity (increased NPY and food intake), and neurological impairment (loss of cerebral volume).
- food intake
- neuropeptide Y
- glucose transporters
postnatal glucocorticoid therapy constituted a standard intervention to combat chronic lung disease in the newborn premature infant (18, 40). Recent reports of adverse outcomes such as cerebral palsy have cautioned the clinicians into judicious use of this drug (2, 17, 50). However, there still continue to be certain infants with chronic lung changes who are resilient to conventional modes of therapy warranting the use of postnatal glucocorticoids as a life-saving intervention (19, 42). Furthermore, despite the absence of well-controlled trials, glucocorticoids are still used in term infants for differing indications other than prevention of chronic lung disease. While the adverse effects of antenatal glucocorticoid therapy have been examined and long-term changes reported in animals (13, 20, 33, 37, 38), limited studies exist in response to repetitive postnatal glucocorticoid use (15, 16).
Antenatal studies to date have reported the development of hypertension and hyperglycemia in the postnatal phase of development (13, 20, 33, 37, 38), and longer-term studies conducted in mice and rats have revealed diabetes in adult life (37, 38). While some of the mechanisms by which such changes occur have been reported, which include an increase in glucocorticoid receptors and hepatic phosphoenolpyruvate carboxykinase expression (37), other mechanisms that can substantially contribute to this ultimate phenotype remain unexplored.
Postnatal glucocorticoid use in the human causes a catabolic state (22, 24). However, the mechanisms that contribute toward this postnatal weight loss remain unclear. Food intake and substrate utilization constitute important mechanisms that underlie body weight gain (7). Previous studies in the adult have demonstrated that glucocorticoids can alter certain hypothalamic neuropeptides that control food intake and body weight gain (34, 52). In addition, glucocorticoids alter skeletal muscle and fat insulin-responsive glucose transporter (GLUT 4) concentration (12, 21), suggesting that glucocorticoids have both a central and peripheral effect that is capable of altering the ultimate body habitus.
To test the hypothesis that both peripheral metabolic and central hypothalamic mechanisms contribute toward the adverse effects of postnatal glucocorticoid administration, we investigated the effect of exogenous repetitive glucocorticoid treatment on hypothalamic neuropeptides that regulate food intake and peripheral mechanisms that underlie substrate utilization such as GLUT 4 in skeletal muscle.
MATERIALS AND METHODS
Gestationally timed pregnant Sprague-Dawley rats (Taconic Farm's, Germantown, NY) were housed in individual cages, exposed to 12:12-h light-dark cycles at 21–23°C, and allowed access to standard rat chow (Purina, St. Louis, Mo) ad libitum. As approved by the Magee-Womens Research Institute's Animal Care and Use Committee at the University of Pittsburgh, the guidelines of the National Institutes of Health were followed. The animals were allowed to deliver, and the pups per litter were culled to 10 to minimize the effect of litter size on nutrition and body weight.
Pups within a litter from a total of 52 litters were arbitrarily divided into two major groups; one received intraperitoneal dexamethasone (Dex; 0.2 μg/g) daily between 2 and 7 days of age (n = 260), and the second group received the same volume of vehicle (Veh; n = 260). The 2 to 7 days of age was chosen for intervention because this is a critical period of hypothalamic development during which permanent effects lasting into adulthood have been previously observed (24, 48). Individual pups in each litter were weighed daily between 8 and 10 AM. On weaning of pups on day 21, body weight was assessed once every 10 days until 120 days of age. Food intake was measured over a 24-h period by weighing the rat chow at the beginning and end of the 24-h period after accounting for spillage and evaporation. Similarly, water intake was assessed over a 24-h period by measuring the water at the beginning and end of the 24-h period after accounting for evaporation.
Animals were randomly assigned for study at preweaning and postweaning ages. While some litters were predominantly examined during the suckling phase where pups from the same litter belonging to the same treatment group were pooled to generate an n = 1, other litters were not examined during the suckling phase but only in the postsuckling and adult stages. Each n at these later stages was one or two pooled animals that were from the same litter and were randomly assigned to study of certain outcome variables. Each n in a group represented animals that came from different litters, for example, an n = 5 represents one or two pooled animals/litter that arose from five separate litters.
Animals were euthanized with intraperitoneal pentobarbital sodium (100 mg/kg); blood was collected from the left ventricle, and the plasma was separated and aliquoted for measurement of glucose by the glucose oxidase method (Sigma Diagnostics, St. Louis, MO; sensitivity = 0.1 mM with an intra-assay coefficient of variation = 1.2%) (48). Insulin, leptin, and corticosterone were quantified by double-antibody RIAs using rat standards and anti-rat insulin, anti-rat leptin (Linco Research, St. Charles, MO), or anti-rat corticosterone (Amersham Life Science, Buckinghamshire, UK) antibodies (sensitivity: insulin = 0.1 ng/ml, leptin = 0.5 ng/ml, corticosterone = 0.06 ng/ml). Leutinizing hormone (sensitivity = 0.005 ng/ml) and estrogen (sensitivity = 1.2 pg/ml) were also assessed by RIAs using rat specific standards and antibodies as previously described (48).
Hypothalamic tissue assays.
The hypothalamus was obtained as a frontal slide by vertical cuts 1 mm anterior to the body of the optic chiasm and 1 mm posterior to the mammillary bodies. The tissue block was weighed and extracted in 4 vol of 0.1 N HCl (wt/vol). The extract was sonicated for 10 s (Sonic dismembrator, Fisher Scientific) using 10-W output power. The sonicated acid extracts were centrifuged at 10,000 rpm for 10 min to remove the tissue debris. The supernatant was freeze dried. The freeze-dried extracts were reconstituted in 0.05 M Tris·HCl buffer containing 0.1% BSA (pH 7.8) for neuropeptide Y (NPY) measurements by RIA. NPY was assessed by an RIA that employed a polyclonal rabbit anti-rat NPY antibody and rat NPY standards (Peninsula Laboratories, Belmont, CA). NPY was expressed as picograms per milligram hypothalamic protein (43) that was assessed by the Bradford dye-binding assay (3).
The 7-day-old (n = 3–5), 21-day-old (n = 4–6), and 120-day-old (n = 3–5) rats from each Dex and Veh treatment groups were initially anesthetized by a combination of ketamine (40 mg/kg) and xylazine (8 mg/kg), and their brains were perfused and fixed as previously described (39, 48). Serial rostrocaudal floating microtome coronal brain sections were obtained (35 μm) and subjected to immunohistochemical analysis as previously reported (39, 48). Rabbit anti-rat NPY (1:8,000; Peninsula Laboratories, Belmont, CA) (39), sheep anti-α-melanocyte stimulating hormone (α-MSH; 1:10,000) (14), or rabbit anti-rat corticotropin-releasing factor (CRF; 1:500; Peninsula Laboratories, Belmont, CA) (39) IgGs served as the primary antibody. PBS alone, PBS buffer containing appropriate dilutions of normal rabbit or sheep serum and lacking the primary antibody, preimmune serum, and the peptide preabsorbed antibodies were used as appropriate controls. Biotinylated goat anti-rabbit (or donkey anti-sheep) IgG (H + L) was employed as the secondary antibody in a 1:600 dilution (Sigma Chemical) at room temperature for 60 min followed by incubation for an additional 60 min at room temperature in the avidin-biotin-peroxidase complex at a 1:200 dilution (Vector Laboratories, Burlingame, CA). Immunolabeling was produced with 3′,3′-diaminobenzidine and enhanced with 2.5% nickel sulfate in 0.175 mM sodium acetate. The sections were mounted in rostrocaudal sequence on charged ProbeOn Plus glass slides (Fisher Scientific, Pittsburgh, PA) and coverslipped under Histomount (National Diagnostics, Atlanta, GA). Sections containing the hypothalamic region were subjected to image analysis under a ×40 magnification using the Simple C-32 software program (C-imaging series SIMPLE 32 Compix Imaging Systems, Cranberry, PA). After subtracting the background, a gray scale was developed based on the intensity of the immunoreactivity. This gray scale provided the relative intensity of the neuropeptide immunoreactivity. In addition, the area of the immunoreactivity was circumscribed and measured. The measured intensity multiplied by the area of the neuropeptide immunoreactivity was equal to the amount of total peptide immunoreactivity expressed in arbitrary units per section (48). A total of five sections per brain was analyzed to obtain a mean value of an n = 1.
Western blot analysis.
The soleus and extensor digitorum longus (EDL) skeletal muscle homogenates were sonicated and centrifuged at 10,000 g at 4°C for 10 min, and the supernatant was saved for Western blot analysis. Predetermined optimal protein concentrations of the homogenates (25 μg) were subjected to discontinuous 10% SDS-polyacrylamide gel electrophoresis followed by electroblot transfer to nitrocellulose filters (Nytran, Schleicher and Schuell, Keene, NH). The filters were incubated for 1–2 h at 23°C with an affinity-purified rabbit anti-rat GLUT 4 (1:500 dilution). The filters were subsequently treated with a peroxidase-linked goat anti-rabbit IgG and subsequently exposed to a chemiluminescence reagent (Amersham Life Science, Little Chalfont, Buckinghamshire, UK). The chemiluminescence was captured by autoradiography over the predetermined optimal exposure time. GLUT 4 protein concentrations were assessed by quantification of the protein bands by densitometry. The presence of linearity between the time of autoradiographic exposure and the optical density of the GLUT 4 bands was initially ensured (47).
Data are expressed as means ± SE. The Wilcoxon-Mann-Whitney test was used to make intergroup comparisons, with simultaneous intergroup and interage comparisons being made using the Friedman's extension of ANOVA for nonparametric data. Using the Friedman's approach, the model included group (Dex vs. Veh), age, and group-by-age interaction terms. These models were fit to all the outcome variables. An overall F test for the significance of the model was computed in addition to an F test for the individual effects. For all comparisons, an exact significance level was computed. This level of statistical significance is based on all possible outcomes of the data and does not rely on large sample sizes for the computation. This method is more conservative than methods based on approximate significance levels. However, due to the small sample sizes, the large sample approximation is inappropriate. Significance levels on adjustment for multiple comparisons within each category would be lowered to 0.05/3 for many of the comparisons presented. The design of the study included power computations, indicating a minimum sample size of five animals per group, per sex, and at each age. These computations were based on a detectable level of difference in terms of the SD obtained from preliminary experimentation.
Body weight changes are depicted in Fig. 1. Because no sex-related changes in body weight gain pattern were observed before day 21, male and female values of the Veh-treated group were pooled at a given age (Fig. 1A). Compared with the Veh-treated group, postnatal glucocorticoids (Dex) caused a decrease in postnatal body weight gain to a similar extent in both the male and female; hence these data were pooled in the Dex-treated group as well. This Dex-related decrease in the body weight gain pattern manifested in 24 h after initiation of the Dex treatment. During the treatment period (2–7 days of age), the decrease in body weight gain continued and was observed even after cessation of treatment. Furthermore, in the adult, both in males (Fig. 1B) and females (Fig. 1C), diminution in the body weight gain pattern persisted at all ages until 120 days of age.
Assessment of food intake (in g/day) revealed a decrease in the Dex-treated adult males (Fig. 2A) and females (Fig. 2B) at 21 days of age (immediately postweaning) and at 35 days of age. By 60 days and at 120 days of age, food intake was no different between the two treatment groups in both males and females. The Dex-associated decrease in food intake compared with the age-matched Veh group appeared more exaggerated in females (∼45%; P < 0.05) compared with males at 21 and 35 days of age (∼15%). Because food intake assessed in a day may only reflect the demand of body weight gain, food intake when expressed per unit body weight per day was higher in both males (Fig. 2C) and females (Fig. 2D) compared with the age-matched and sex-matched Veh-treated group. The only exception was at 21 days of age, where the males and females demonstrated no statistical difference in food intake when expressed per constant unit (100 g) of body weight and compared with the age-matched Veh-treated group. Hence food intake exceeded that expected based merely on the body weight gain pattern in the Dex-treated group.
Water intake (measured in ml/day) increased in the Dex-treated 120-day-old males (Fig. 3A) with no change in the females (Fig. 3B) after the initial decrease at 21 days of age in both sexes compared with the age-matched and sex-matched Veh group. When water intake (in ml/day) was standardized against a constant unit of body weight (100 g), an increase in water intake was evident in males (Fig. 3C) while an increase was only noted at 120 days of age in females (Fig. 3D). This apparent increase in water intake in males at 21, 35, 60, and 120 days of age was not related to basal hyperglycemia at least at 21 days of age; however, serum glucose concentrations in the 120-day-old males were higher than the corresponding Veh-treated group (Table 1). The insulin concentrations were lower in the 14-, 21-, and 120-day-old Dex-treated animals, with no such change in the females compared with the age- and sex-matched Veh-treated group (Table 1). The ratios of circulating glucose to insulin were higher in the Dex-treated group in the 14-, 21-, and 120-day-old males, as well as the 120-day-old females (Table 1). Postnatal Dex led to an acute increase (24 h) in circulating leptin concentrations at 3 days of age followed by a persistent decrease in both males and females at 7 through 120 days of age. Endogenous corticosterone concentrations were acutely suppressed by the exogenous Dex treatment at 3 days alone (Table 1). Postnatal Dex treatment also led to a decrease in plasma estrogen concentrations in the 21-day-old males and females with no change in the leutinizing hormone (LH) concentrations (Table 2). In contrast, at 120 days of age in both the males and females, no change in plasma estrogen concentrations with a decrease in LH levels was observed (Table 2).
An increase in skeletal muscle GLUT 4 concentrations was noted in the 7-day-old males and females (Fig. 4), with the males demonstrating a greater difference compared with the age-matched females (P < 0.01). At 120 days of age, while the Dex male soleus (oxidative) and EDL (glycolytic) exhibited no change in GLUT 4, in the Dex female a decrease in the soleus and an increase in the EDL GLUT 4 concentrations were detected (Fig. 4).
When total hypothalamic NPY content was assessed by RIA, no differences were observed at the ages investigated (Table 2). Thus hypothalamic neuropeptides were assessed specifically in the paraventricular (PVN) and arcuate (ARC) nuclei because these two nuclei are intricately involved in appetite control and are responsive to metabolic changes. Postnatal Dex treatment led to an increase in NPY within the PVN and ARC nuclei at 7 and 21 days of age in both males and females and only in the 120-day-old females, with the 120-day change being reflected only in the PVN (Fig. 5; Table 3). PVN and ARC nuclear CRF immunoreactivity was decreased in the 21-day-old in the Dex-treated group with no change at any other age (Fig. 6, E and F; Table 3). In contrast, α-MSH (Fig. 6, C and D; Table 3) was no different at all ages except for a decrease in the PVN at 7 days of age in the Dex-treated group. No Dex-induced change in the leptin receptor concentration was observed at all ages examined (Table 3; Fig. 6, A and B). Focusing on the ventricular regions of the brain, both in the 120-day-old males and females, an increase in the ventricular size was noted in the postnatal Dex treatment group (Fig. 7, A and C) compared with the Veh treatment group (Fig. 7, B and D). This increase was more prominent in males compared with females (P < 0.01; Fig. 7, A, C, and E).
We have investigated the effects of repetitive (6 doses) systemic administration of Dex postnatally on the suckling and adult phenotype. Dex was associated with a lower body weight gain pattern at all ages in both males and females. This decrease in body weight gain initially (16) may be due to a decrease in milk intake, an increase in energy expenditure (41), or inadequate substrate availability/utilization (23). Although we could not accurately assess milk intake during the suckling phase, we observed an increase in the hypothalamic orexigenic NPY concentrations as early as 7 days of age. This increase in NPY was associated with a transient decrease in anorexigenic α-MSH and CRF. The balance between the orexigenic and anorexigenic peptides determines the ultimate appetite (7). The postnatal changes we observed in the hypothalamus set the stage for subsequent development of Dex-associated “catch-up” in food intake at 35, 60, and 120 days of age in the males and only at 60 and 120 days of age in the females compared with the controls. When expressed per unit body weight, both males and females exhibited an increase in food intake at 35, 60, and 120 days of age. This may be subsequent to an increase in hypothalamic NPY concentrations noted at 7 and 21 days of age in both the sexes. At 120 days of age, however, an increase in hypothalamic NPY concentrations is noted only in females, an observation that is not consistent with the increase in food intake seen in both the 120-day-old males and females. This increase in hypothalamic NPY in the 120-day-old females may predict the subsequent persistence of hyperphagia in females alone. Previous investigations in adult rats revealed that systemic glucocorticoids increased hypothalamic NPY content (34, 52) similar to our present findings in postnatal rats. Separate studies employing only two doses of a significantly higher dose (10-fold) of postnatal glucocorticoids (35) observed a decrease in food intake at 60 days of age, while prenatal glucocorticoid administration made no difference to the food intake at 70 to 80 days of age (45). Both these investigations demonstrated results that mimic our present observations. However, both the reports presented food intake per day and did not normalize it to unit body weight, which in turn may have demonstrated an increase in food intake.
This postnatal change in hypothalamic neuropeptides could be a direct effect of Dex (34, 52) or an indirect effect of the altered hormonal milieu secondary to postnatal Dex administration. Postnatal Dex treatment while acutely not affecting plasma insulin concentrations at 3 days of age caused a suppression of plasma insulin concentrations later (7 to 21 days of age) particularly in males, reflective of a catabolic state (49). The decrease in circulating insulin concentrations during the suckling phase is associated with an increase in insulin receptors (9, 10, 29) and total insulin-responsive GLUT 4 concentrations suggestive of improved insulin sensitivity, glucose transport, and utilization by the skeletal muscle. Increased insulin sensitivity at 70 days of age was confirmed by the hyperinsulinemic euglycemic clamp technique after postnatal glucocorticoid administration (35). Previous studies in adults have demonstrated a Dex-responsive increase in adipocytic and skeletal muscle GLUT 4 expression and concentration (12, 21). A similar effect was observed with the placental glucose transporter isoforms (GLUT 1 and GLUT 3) (28). The decrease in circulating insulin concentrations was associated with a propensity toward glucose intolerance in the postsuckling males (glucose:insulin ratio) that culminated in hyperglycemia at 120 days of age. This finding may underlie the increase in water intake noted predominantly in the postsuckling and adult males. Our present observations are different from findings of hyperglycemia and hyperinsulinemia in the adult progeny that had received antenatal glucocorticoids (37, 38). Furthermore, antenatal steroid administration to sheep caused hyperglycemia in the fetus and newborn (20, 33). Whether insulin resistance will develop subsequently after 120 days of age is unknown at the present time.
In contrast to the changes noted with circulating insulin, Dex led to an acute increase in plasma leptin concentrations at 3 days of age; however, from 7 to 120 days of age (in both females and males) leptin concentrations were decreased. Previous in vitro and in vivo experiments have demonstrated that Dex acutely increases adipocytic leptin synthesis and secretion (25, 52). Thus at 3 days of age, exogenous Dex may have increased adipocytic production of leptin while suppressing endogenous corticosterone concentrations. Hyperleptinemia can suppress hypothalamic NPY concentrations and diminish milk intake (1, 6, 51). This decrease in calorie intake can potentially initiate postnatal growth retardation (51). Beginning at 7 days of age, a more chronic decrease in plasma leptin concentrations with no change in the endogenous corticosterone concentrations is reflective of a diminished adipocytic mass paralleling the persistence of a lower body weight gain pattern in both sexes (5, 35). A decrease in circulating leptin and insulin concentrations is indicative of a catabolic state and is known to increase hypothalamic NPY while suppressing α-MSH and CRF concentrations (7, 52). This catabolic state may have decreased circulating estrogen (26) initially, which in turn subsequently decreased LH concentrations, a finding not observed with fewer postnatal (35) or antenatal glucocorticoid doses (45). This decrease in estrogen concentrations may cause a delay in sexual maturation particularly in the females. However, we did not monitor the related physical characteristics in the present investigation.
Thus, despite the hypothalamic changes geared toward increasing calorie intake, the set point for weight gain in the Dex group was not reset. The Dex-treated animals continued to gain weight at a lower set point than the Veh-treated group. Once body weight gain is adversely affected during a critical developmental stage (fetal/postnatal), recovery through the remaining trajectory of life is compromised (27, 49) despite galvanizing compensatory mechanisms. In addition to mechanisms responsible for regulating calorie intake that we investigated here, other mechanisms involved in increasing energy expenditure (41) or breakdown of substrates (e.g., increased proteolysis) (23) may also contribute toward the Dex-associated catabolic state.
While the focus of the present study was on the hypothalamus, we serendipitously observed an increase in ventricular size associated with postnatal Dex treatment. This increase in ventricular size is exaggerated in the adult males compared with the females. This observation reflects a loss of brain tissue volume rather than hydrocephalus because a decrease in total brain weight due to postnatal Dex administration has previously been reported (16). Furthermore, separate investigations demonstrated DNA fragmentation of brain cells and white matter injury due to an increase in endogenous corticosterone levels caused by maternal separation (53). Functional studies demonstrate aberrations in the hypothalamic-pituitary-adrenal axis and neurobehavior due to postnatal Dex administration (16, 31, 32).
While both sexes demonstrated some changes in the hypothalamus and hormonal/metabolic milieu associated with postnatal Dex treatment, some associations were more exaggerated in one sex vs. the other. The adult females demonstrated a persistent hypothalamic change targeted toward developing hyperphagia, while the males expressed at an earlier age a more profound catabolic state seen as low circulating insulin concentrations and a propensity toward glucose intolerance (glucose:insulin ratios). Similarly ventriculomegaly was more pronounced in the adult males. Whether these sex-specific differences will persist with age (beyond 120 days of age) remains to be explored.
Our current Dex-associated changes are consistent with that reported previously in the intrauterine growth-restricted (IUGR) progeny (8, 39). In the IUGR progeny, calorie restriction begins in utero, resulting in persistence of growth failure into adulthood (27, 49). The IUGR offspring has low insulin (39) and leptin concentrations in utero (8), while the impact on endogenous corticosterone concentrations has been controversial, with some studies demonstrating no effect (4, 36) and others showing an increase that may cause the programming effects observed in the adult (30, 33). In conjunction with these hormonal effects, a sustained postnatal increase in hypothalamic neuropeptide Y concentrations is observed in the IUGR (39) and in the maternally deprived offspring (24). Thus similar to the IUGR offspring (49), the postnatal Dex-treated growth-restricted offspring demonstrates compensatory hyperphagia, which in time may result in obesity, insulin resistance, and diabetes mellitus. Furthermore, the Dex-associated cerebral ventriculomegaly mimics the aberrations observed in the IUGR offspring (11, 46).
We conclude that postnatal glucocorticoid treatment acutely alters the postnatal phenotype consistent with an initial diminution in calorie intake. These postnatal changes alter the entire trajectory, causing sex-specific aberrations in the adult phenotype. The phenotypic changes consistent with a catabolic state cause cerebral ventriculomegaly, affect the hormonal/metabolic milieu, and are associated with compensatory hypothalamic perturbations, which in turn set the stage for hyperphagia. These changes over the long term may predispose toward the future development of cerebral palsy/cognition defects, diabetes (pancreatic failure and insulin resistance), and obesity (hyperphagia and visceral adiposity), imitating the concept of “fetal/neonatal programming” that has been described in the adult IUGR offspring. While animal data cannot be extrapolated to the human directly, our current investigations further caution against the indiscriminate use of glucocorticoids for any indication whether in the premature or term infant.
This work was supported by National Institutes of Health Grants HD-25024, HD-41230, and HD-33997.
We thank Dr. J. B. Tatro, Dept. of Medicine, Tufts University School of Medicine and the New England Center, Boston, MA for the generous gift of the anti-α-MSH antibody (BA387) that was used in this study.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society