Sex Difference in the Response of Melanin-Concentrating Hormone Neurons in the Lateral Hypothalamic Area to Glucose, as Revealed by the Expr
http://www.100md.com
内分泌学杂志 2005年第8期
Department of Neuroendocrinology (K.M., T.F., D.M., H.H., F.K.), Yokohama City University Graduate School of Medicine, Kanazawa-ku, Yokohama 236-0004; and International University of Health and Welfare (F.K.), Amity-Nogizaka Minamiaoyama, Minato-ku, Tokyo 107-0062, Japan
Address all correspondence and requests for reprints to: Dr. Toshiya Funabashi, Department of Neuroendocrinology, Yokohama City University Graduate School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama 236-0004, Japan. E-mail: toshiya@med.yokohama-cu.ac.jp.
Abstract
Because there are sex differences in feeding behavior in rats, we looked for a possible sex difference in the response to glucose of melanin-concentrating hormone (MCH) neurons in the lateral hypothalamic area using phosphorylated cAMP response element-binding protein (pCREB) as a marker of neural activity. Intact male rats and female rats at diestrus 2, proestrus, or estrus were fed normally or fasted for 48 h and injected with saline or glucose (400 mg/kg). Thereafter, preparations were subjected to immunohistochemical processing for the double staining of MCH and pCREB. Fasting increased the ratio of MCH neurons with pCREB (double-stained cells) in both male and female rats. In fasted rats, glucose injection decreased the ratio of double-stained cells more promptly in females than in males. The magnitude of decrease caused by glucose was greater at proestrus and estrus than at diestrus 2. Gonadectomy in males enhanced and in females attenuated the response of MCH neurons to glucose. Testosterone and estrogen replacement in males and females, respectively, restored the response of MCH neurons to glucose. The demonstrated sex differences in the response of MCH neurons to glucose correlated well with the gonadal steroid milieu; thus, MCH neurons may play an important role in sex differences in feeding behavior.
Introduction
SEX DIFFERENCES IN feeding behavior have long been recognized in many species. For example, analyses of spontaneous patterns of feeding have shown that the size of a single meal is smaller in females than in males, and the total volume of daily food intake is also smaller in females than in males (1, 2). In addition, the size of a single meal varies during the estrous cycle, decreasing around the time of ovulation in females, including humans (1, 3, 4). Although these sex differences are attributable to gonadal steroid hormones such as testosterone and/or estrogen (5, 6, 7, 8, 9), the neural mechanism is still unclear. For example, although cholecystokinin has been reported to be involved in estrogen-induced decrease in food intake (10), there is still controversy about whether cholecystokinin mediates estrogen-induced anorexia (11, 12). It is also not known whether cholecystokinin mediates testosterone-induced hyperphagia.
Recent studies have identified distinct neurons in the lateral hypothalamic area (LHA) containing melanin-concentrating hormone (MCH) neurons (13) that generate hunger (14, 15). Neuropeptide Y (NPY), which is produced by neurons in the arcuate nucleus of the hypothalamus (16), is another neuropeptide that prominently induces feeding (17, 18). The hypothalamic circuit between the LHA and the arcuate nucleus is thus an important component of the central system controlling feeding behavior (19). In this circuit, judging from studies that genetically disrupted the orexigenic neuropeptides, MCH neurons in the LHA play a key role. Mice lacking the gene encoding MCH are lean (20), whereas mice overexpressing the MCH gene are susceptible to obesity (21). It has also been suggested that MCH neurons are involved in the glucosensing system, because hypothalamic MCH mRNA levels are increased by hypoglycemic conditions such as fasting, insulin injection, and 2-deoxy-D-glucose treatment (14, 22, 23). In contrast to MCH, mice lacking the gene encoding prepro-orexin or NPY display normal growth (24, 25, 26).
It is probable that sex differences in feeding behavior are at least in part attributable to sex differences in the neurons involved in feeding behavior. Because MCH neurons in the LHA play an important role in the regulation of feeding behavior, we hypothesized that there might be a sex difference in the activity of MCH neurons in the LHA during fasting and/or after glucose injection. In support of this view, hypothalamic MCH neurons are suggested to be involved in estrogen-induced weight loss in male rats (27). We therefore examined the response of MCH neurons to an iv injection of glucose after 48 h of fasting in adult male and female rats, using the expression of phosphorylated cAMP response element-binding protein (pCREB) immunoreactivity as a marker of neural activity. On the assumption that a sex difference in MCH neurons existed, we also examined whether it could be affected by gonadal steroid hormones such as testosterone and/or estrogen.
In the present study, we used pCREB as a marker for neural activity increased of the generally used c-Fos, because an increase in the expression of c-Fos needs transcription and/or translation to occur (28), it takes a long time to see changes, and it would be difficult to examine acute changes in response to glucose. In contrast, CREB is a constitutively expressed transcription factor and can be immediately activated by phosphorylation via a cAMP- and calcium-dependent cascade (29, 30). Thus, we thought that pCREB was a suitable marker to examine neural activity that changes within a short period in response to dynamic changes such as blood glucose concentrations.
Materials and Methods
Animals
Adult male and female Wistar rats (Charles River Laboratories, Yokohama, Japan) were maintained at a constant temperature at 24–26 C under controlled lighting conditions (lights on, 0500–1900 h) with food and water available ad libitum. Daily vaginal smears were taken from all female rats, and those exhibiting two or more consecutive 4-d estrous cycles were used in experiments 1 and 2. Forty-eight hours before saline or glucose injection, a silicone cannula for iv injection was implanted into the right atrium of all rats under ether anesthesia. This procedure did not interfere with the estrous cycle. The number of rats used for each group is indicated in the figures. All animal housing and surgical procedures were in accord with the guidelines laid down by the institutional animal care and use committee of Yokohama City University School of Medicine.
Determination of blood glucose and insulin concentrations
To determine changes in blood glucose and insulin concentrations after glucose injection (400 mg/kg), intact males, orchidectomized (ORX) males, females at proestrus and diestrus 2, and ovariectomized (OVX) females were implanted with a silicone cannula in the right atrium under ether anesthesia. Male and female rats were castrated 4 wk before the experiment. After cannula implantation, they were fasted for 48 h and injected with 400 mg/kg glucose dissolved in 200 μl saline after the first blood sampling of 100 μl. Control rats were injected with the same volume of saline. Thereafter, blood was sampled 5, 10, 15, 30, 60, and 120 min after the injection though the same cannula. An equal volume of heparinized saline (2 IU/ml) was injected after each sampling. As another control, blood samples were obtained from intact males, ORX males, females at proestrus, and OVX females that were not fasted. Blood glucose concentrations were determined with an instant blood glucose assay apparatus (ACCU-CHEK Comfort, Roche, Tokyo, Japan). Serum insulin concentrations were measured with an ELISA kit (Morinaga Institute of Biological Sciences, Yokohama, Japan).
Experiment 1: sex difference in the activity of MCH neurons
To determine whether there is a sex difference in the activity of MCH neurons, some male and female rats were fed normally and injected iv with 200 μl saline (Fed+Saline group), and other male and female rats were fasted for 48 h and given an iv injection of 200 μl saline (Fast+Saline group) or glucose (Fast+Glucose group; 400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. The injection was made through the silicon cannula in the right atrium as described above. The dose of glucose used in the present study was based on a previous study which showed that this treatment could reverse the inhibition of pulsatile LH secretion caused by insulin-induced hypoglycemia (31). For females, three groups, composed of rats at diestrus 2 (D2 group), proestrus (P group), or estrus (E group), were prepared to check the effect of the stage of the estrous cycle. The rats in the D2 group had been fasted from the day of estrus, those in the P group had been fasted from the day of last diestrus 1, and those in the E group had been fasted from the day of last diestrus 2. Thus, 12 groups of rats (four groups containing three treatment groups each) were prepared, and they were processed blindly for immunohistochemistry.
In all groups, rats were killed 5 min after saline or glucose injection by an iv injection of an overdose of pentobarbital (100 mg/kg). Perfusion through the cardiac ventricle was quickly started with 2% paraformaldehyde and 4% acrolein in phosphate buffer (PB; pH 7.5) at approximately 4 C. After perfusion, the brains were removed from the cranium, fixed overnight at 4 C in PB containing 2% paraformaldehyde, and incubated overnight at 4 C in 30% sucrose in PB. The brains were then frozen with powdered dry ice and stored at –70 C until immunohistochemical processing.
Experiment 2: time-dependent changes in the activity of MCH neurons after glucose injection
To examine time-dependent changes in the activity of MCH neurons after iv glucose injection, male rats were fasted for 48 h and injected iv with glucose (400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. Rats were killed 15, 30, and 60 min after glucose injection (Fast+Glucose 15 min group, Fast+Glucose 30 min group, and Fast+Glucose 60 min group) by iv injection of an overdose of pentobarbital. Female rats at diestrus 2, which had been fasted for 48 h from the day of estrus, were similarly killed. All subsequent procedures to obtain brains were as described for experiment 1.
Experiment 3: effects of testosterone and estrogen on the response of MCH neurons to glucose
To determine whether gonadal steroid hormones affected the response of MCH neurons to glucose, male and female rats were castrated under ether anesthesia and used for the experiment 4 wk later. ORX and OVX rats were implanted sc with a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 30 mm) containing 100% testosterone crystals (ORX+T group and OVX+T group), a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 15 mm) containing 20% 17?-estradiol crystals (ORX+E2 group and OVX+E2 group), or a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 30 mm) containing cholesterol alone (ORX group and OVX group) as a control. All silicone tubes had been previously soaked in saline for more than 24 h to facilitate rapid hormone release. It was shown previously that a tube containing this dose of 17?-estradiol produced a serum 17?-estradiol concentration equivalent to that observed at proestrus in intact female rats (32), and a tube containing testosterone produced a serum testosterone concentration equivalent to that observed in intact male rats (33). After implantation of the tubes, some rats were fed normally, and other rats were fasted for 48 h. Normally fed rats were injected iv with 200 μl saline (Fed+Saline group), and fasted rats were injected iv with 200 μl saline (Fast+Saline group) or glucose (Fast+Glucose group; 400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. All rats were killed 5 min after iv injection. Thus, a total of 18 groups of rats (six groups containing three treatment groups each: ORX, ORX+T, ORX+E2, OVX, OVX+T, and OVX+E2 groups x Fed+Saline, Fast+Saline, and Fast+Glucose groups) was prepared. All subsequent procedures to obtain brains were as described for experiment 1.
Immunohistochemistry
Thirty-micrometer frozen coronal sections were cut with a Bright cryostat (Jencons Ltd., Leighton Buzzard, UK), and every fourth section was used for the study. The sections were incubated with 1% sodium borohydride in PBS and then with 0.3% H2O2 in 20% methanol in PBS. They were incubated overnight with rabbit polyclonal antibody to pCREB diluted 1:900 (Cell Signaling Technology, Beverly, MA) in PBS containing 1.5% normal goat serum and 0.1% Triton X-100. The next day they were incubated with biotinylated antirabbit IgG diluted 1:200 in PBS containing 1.5% normal goat serum and 0.05% Triton X-100, and thereafter incubated with streptavidin-biotin-peroxidase complex (Vectastain Elite ABC Kit, Vector Laboratories, Inc., Burlingame, CA). Bound peroxidase was visualized by incubating sections for 30 min in 0.05% 3,3'-diaminobenzidine with H2O2. For double staining, sections were then incubated overnight with rabbit polyclonal antibody to MCH diluted 1:1800 (Phoenix Pharmaceuticals, Belmont, CA) in PBS containing 1.5% normal goat serum and 0.1% Triton X-100. They were additionally incubated with biotinylated antirabbit IgG diluted 1:200 in PBS containing 1.5% normal goat serum and 0.05% Triton X-100, then with Cy3-labeled streptavidin (Amersham Biosciences, Little Chalfont, UK). Finally, they were mounted on glass slides, dehydrated in graded alcohol, cleared in xylene, and coverslipped with Permount (Fisher Scientific, Fairlawn, NJ). A total of eight series of immunohistochemical procedures was performed, and in each of them, the incubation time of 3,3'-diaminobenzidine staining and other experimental conditions were checked carefully for reproducibility.
Histological analysis
The counting was performed by an investigator blind to the experimental conditions and expectations. We initially used every fourth section through the LHA, and thus approximately 10 sections/rat were stained with MCH and pCREB antibodies. Using a fluorescence microscope, five sections per rat were finally selected for counting MCH-immunoreactive cells (MCH neurons) and pCREB-immunoreactive cells (pCREB cells) in the dorsolateral hypothalamic group of the LHA, because the distribution of MCH neurons varies from anterior to posterior in the diencepalon, and this group of the LHA comprises the numerically predominant grouping of MCH-expressing cells in the diencephalon (13). Sections were carefully matched across all animals in all experimental groups. The number of MCH neurons in the LHA (square region of 1.7 x 1.0 mm) was counted bilaterally with a fluorescence microscope at 20 x 10 magnification. MCH cells in which a nucleus was clearly visible and surrounded by fluorescent cytoplasmic staining were counted. In the same sections, pCREB cells were also observed, and the number of MCH neurons expressing pCREB-immunoreactive nuclei (double-stained neurons; Fig. 1) was counted. Cells were defined as double stained for MCH and pCREB when a blue-black nucleus was surrounded by fluorescent cytoplasm at 20 x 10 magnification. Then the average ratio (percentage) of the number of double-stained neurons to the number of MCH neurons in each rat was calculated. Both the average number of MCH neurons per section in each rat and the average percentage of double-stained neurons per section in each rat were used for statistical analysis. In male rats and female rats of the P group in experiment 1, we counted the number of pCREB cells in the LHA.
FIG. 1. Representative photographs of double-stained neurons in the LHA. Fluorescent MCH-immunoreactive neurons (top, arrowheads) and pCREB-immunoreactive nuclei (bottom, arrowheads) in the same tissue section (magnification, x400) in the Fast+Saline group (left panels) and Fast+Glucose group (right panels) of male and female (P) rats in experiment 1 are shown.
Data were analyzed by one-way ANOVA, followed by Fisher’s protected least significant difference post hoc test for comparisons of treatment groups in each animal group. Differences were considered significant at P < 0.05.
Results
In control (i.e. not fasted) intact males, ORX males, females at proestrus, and OVX females, blood glucose and insulin levels were 103.8 ± 1.5 mg/dl (n = 4) and 1273 ± 54 pg/ml, 107.4 ± 2.4 (n = 5) mg/dl and 1433 ± 121 pg/ml, 101.3 ± 2.4 mg/dl (n = 4) and 1306 ± 113 pg/ml, and 105.0 ± 2.7 mg/dl (n = 4) and 1323 ± 143 pg/ml, respectively. No significant differences in glucose (F3,13 = 0.811; P > 0.1) or insulin (F3,13 = 0.398; P > 0.5) were observed. As shown in Table 1, one-way ANOVA in each saline-injected group showed no significant effect of saline injection on blood glucose concentrations: the male group (F6,21 = 0.811; P > 0.5), the ORX group (F6,21 = 0.61; P > 0.5), the P group (F6,21 = 1.61; P > 0.1), and the OVX group (F6,28 = 2.29; P > 0.05), suggesting that the control procedure used in the present study did not affect blood glucose concentrations. However, blood glucose concentrations were significantly increased after glucose injection in all groups. One-way ANOVA at each time point showed that there was no significant difference in blood glucose concentrations among the groups (Table 2). In addition, the area under the curve (AUC), which was calculated for each rat, was not significantly different among the groups (F4,19 = 0.562; P > 0.5). This suggests that changes in blood glucose concentrations after glucose injection were uniform among the groups. Insulin concentrations were also significantly increased after glucose injection in all groups, but one-way ANOVA at each time point showed that there was a significant difference in insulin concentrations among the groups (Table 2). That is, insulin concentrations in the intact male and the ORX groups were significantly higher than those in the D2, P, and OVX groups at some time points after glucose injection. In fact, the AUC of insulin was significantly different among the groups (F4,19 = 9.09; P < 0.002). Post hoc comparison showed that the AUC in the intact male and ORX groups was significantly greater than in the D2, P, and OVX groups (P < 0.05). This result suggested that insulin secretion in response to glucose in male rats was greater than that in female rats and was in good accord with previous reports of a sex difference in glucose tolerance (34, 35)
View this table:
[in this window]
[in a new window]
TABLE 1. Changes in blood glucose concentrations (milligrams per deciliter) after saline injection
TABLE 2. Changes in blood glucose and insulin concentrations after 400 mg/kg glucose injection
In all rats, the overall distribution of MCH neurons with red fluorescent cytoplasm was in good accord with a previous report (13) in all experiments. The average number of MCH neurons per section was approximately 500 in all groups, and this number did not vary in any experiment (data not shown). In the intact male and P groups, the number of pCREB-expressing cells was slightly larger than that of MCH neurons expressing pCREB (Table 3), but changes in the number of pCREB cells were similar to those in MCH neurons expressing pCREB, as indicated below. These results suggest that MCH neurons were not the only population of cells expressing pCREB in the LHA.
TABLE 3. Changes in the number of pCREB-expressing cells in the lateral hypothalamic area of male rats and female rats at proestrus
Experiment 1
All female rats in the E and D2 groups exhibited a regular 4-d estrous cycle, as monitored by vaginal smear during the experiments. Almost 40% of female rats fasted for 48 h in the P group, however, had a diestrus-like vaginal smear during the day on which they were supposed to be in proestrus. Data for rats showing a perturbed cycle were discarded, and only rats showing a proestrous vaginal smear were included in the P group.
In all Fed+Saline groups, pCREB cells with blue-black nuclei were scattered in the LHA and the arcuate nucleus of the hypothalamus (data not shown), and the ratio of double-stained neurons was approximately 10% (Fig. 2).
FIG. 2. Effects of 48-h fasting or glucose injection after 48 h of fasting on the ratio (percentage) of MCH-immunoreactive neurons expressing pCREB-immunoreactive nuclei in experiment 1. In all groups, rats were examined 5 min after saline or glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. other treatment groups. See text for details of statistics. D2, Diestrus 2; P, proestrus; E, estrus.
In all Fast+Saline groups, a large number of pCREB cells was observed in the LHA, but not in the arcuate nucleus of the hypothalamus (data not shown), and the ratio of double-stained neurons was approximately 40% (Fig. 2). One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in each animal group: the intact male group (F2,12 = 7.45; P < 0.01), the P group (F2,12 = 9.35; P < 0.01), the E group (F2,12 = 5.53; P < 0.05), and the D2 group (F2,11 = 11.38; P < 0.01). In all groups, post hoc comparison showed a significantly greater ratio of double-stained neurons in the Fast+Saline group than in the Fed+Saline group (Fig. 2; P < 0.05). These results indicated that 48-h fasting significantly increased the ratio of double-stained neurons in all animal groups.
In the Fast+Glucose group, the number of pCREB cells in the LHA varied among animal groups (data not shown). Post hoc comparison in the male group showed a significantly greater ratio of double-stained neurons in the Fast+Glucose group than in the Fed+Saline group (P < 0.05), but there was no significant difference between the Fast+Glucose group and the Fast+Saline group, indicating that glucose injection did not affect the increase in the ratio of double-stained neurons caused by fasting in intact male rats (Fig. 2). In contrast, pCREB cells were relatively few in the LHA in all female groups after glucose injection (data not shown). Post hoc comparisons in all female groups showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose group than in the Fast+Saline group (Fig. 2; P < 0.05). In both the P and E groups, the ratio of double-stained neurons did not differ between the Fast+Glucose group and the Fed+Saline group, indicating that glucose injection completely blocked the fasting-induced increase in the ratio of double-stained neurons (Fig. 2). In the D2 group, in contrast, the ratio of double-stained neurons was significantly greater in the Fast+Glucose group than in the Fed+Saline group (Fig. 2; P < 0.05), indicating that 48 h after fasting, glucose injection was less effective in the D2 group than in the P and E groups.
Experiment 2
The data from experiment 2 were analyzed with the data from experiment 1 (the Fast+Saline and Fast+Glucose groups of intact male rats, and the D2 group of female rats; Fig. 3).
FIG. 3. Time-dependent changes in the ratio (percentage) of MCH-immunoreactive cells expressing pCREB-immunoreactive nuclei in the Fast+Saline and Fast+Glucose groups of male and female (D2) rats in experiment 2. In all groups, rats were fasted for 48 h and then examined after 5 min of saline injection or after 15, 30, and 60 min of glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. Fast+Saline groups.
In intact male rats, pCREB cells in the LHA were few in the Fast+Glucose 15 min group and the Fast+Glucose 30 min group, and the ratio of double-stained neurons was less than 10% (Fig. 3, upper panel). The number of pCREB cells was relatively large in the Fast+Glucose 60 min group. One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in intact male rats (F4,20 = 5.45; P < 0.01). Post hoc comparison in intact male rats showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose 15 min group and the Fast+Glucose 30 min group than in the Fast+Saline group (P < 0.05). These results indicated that glucose injection significantly decreased the ratio of double-stained neurons 15 and 30 min after injection in intact male rats, but the effect disappeared 60 min after injection.
In the D2 group of female rats, pCREB cells in the LHA were few in the Fast+Glucose 15 min group, and the ratio of double-stained neurons was approximately 10% (Fig. 3, lower panel). The number of pCREB cells was relatively large in the Fast+Glucose 30 min group and the Fast+Glucose 60 min group. One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in the D2 group (F4,19 = 11.30; P < 0.01). Post hoc comparison in the D2 group showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose 5 min group and the Fast+Glucose 15 min group than in the Fast+Saline group (P < 0.05). These results indicated that glucose injection significantly decreased the ratio of double-stained neurons 5 and 15 min after injection in female rats, but the effect disappeared 30 min after injection.
Experiment 3
In all Fed+Saline groups and Fast+Saline groups, the ratios of double-stained neurons were approximately 10% and 40%, respectively (Fig. 4). One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in each animal group: the ORX group (F2,10 = 15.14; P < 0.001), the ORX+T group (F2,10 = 4.61; P < 0.05), the ORX+E2 group (F2,12 = 38.14; P < 0.0001), the OVX group (F2,9 = 6.62; P < 0.05), the OVX+T group (F2,11 = 11.03; P < 0.005), and the OVX+E2 group (F2,9 = 6.41; P < 0.05). In all animal groups, post hoc comparison showed a significantly greater ratio of double-stained neurons in the Fast+Saline group than in the Fed+Saline group (Fig. 4; P < 0.05). Because the ratio of double-stained neurons in the Fed+Saline groups and the Fast+Saline groups in experiment 3 was comparable to that in experiment 1, these results indicated that 48-h fasting significantly increased the ratio of double-stained neurons regardless of the presence or absence of gonadal steroid hormones.
FIG. 4. Effects of testosterone or estrogen on the ratio (percentage) of MCH-immunoreactive neurons expressing pCREB-immunoreactive nuclei in the Fed+Saline, Fast+Saline, and Fast+Glucose groups in experiment 3. In all groups, rats were examined 5 min after saline or glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. other treatment groups. See text for details of statistics. E2, Implantation of 17?-estradiol crystals; T, implantation of testosterone crystals.
Effects of glucose injection varied among the groups. In the ORX and ORX+E2 groups, post hoc comparisons showed that the ratio of double-stained neurons in the Fast+Glucose group was significantly smaller than that in the Fast+Saline group (P < 0.001) and that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fed+Saline group (P > 0.5). In the ORX+T groups, post hoc comparison showed that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fast+Saline group and the Fed+Saline group (P > 0.5). These results indicated that orchidectomy and estrogen replacement decreased the ratio of double-stained neurons 5 min after glucose injection, but testosterone replacement blocked the orchidectomy-induced decrease in the ratio of double-stained neurons in male rats.
In the OVX+T and OVX+E2 groups, post hoc comparisons showed that the ratio of double-stained neurons in the Fast+Glucose group was significantly smaller than that in the Fast+Saline group (P < 0.05) and that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fed+Saline group (P > 0.5). In the OVX groups, post hoc comparison showed that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fast+Saline group (P > 0.5) and that the ratio of double-stained neurons in the Fast+Glucose group was significantly larger than that in the Fed+Saline group (P < 0.05). These results indicated that ovariectomy increased the ratio of double-stained neurons 5 min after glucose injection, but both testosterone and estrogen replacement blocked the ovariectomy-induced increase in the ratio of double-stained neurons in female rats.
Discussion
The present study demonstrated sex differences in the response of MCH neurons to glucose, using the expression of pCREB as a marker of neural activity. Fasting for 48 h activated MCH neurons in a similar manner in both male and female rats, but inactivation of MCH neurons by glucose after 48 h of fasting occurred more rapidly in female rats than in male rats. These sex differences could be attributed to gonadal steroid hormones, including testosterone and estrogen. Taking into account both the orexigenic effect of MCH (14, 15) and the glucostatic theory of feeding (36), our findings provide a molecular/neurotransmitter basis for neurons that are controlled by glucose in the LHA and suggest that MCH neurons in the LHA play an important role in sex differences in feeding behavior, acting through the CREB pathway (29, 30).
Although hypoglycemia caused by treatments such as fasting, insulin injection, or 2-deoxy-D-glucose injection increases the expression of hypothalamic MCH mRNA (14, 22, 23), there is little information on the second messenger system that would be stimulated by hypoglycemic conditions. For example, Fos expression was increased in the LHA by insulin-induced hypoglycemia, but there were few MCH neurons that expressed Fos protein (22, 37). Thus, it is unclear whether Fos proteins are involved in the control of the expression of MCH mRNA in the LHA by hypoglycemia. In experiment 1, we showed that under normal feeding conditions in male and female rats at all stages of the estrous cycle examined, approximately 10% of MCH neurons expressed pCREB-immunoreactive nuclei in the LHA, indicating that only a small population of MCH neurons is active in the LHA. However, we found that fasting for 48 h increased the ratio of double-stained neurons to approximately 40% in both sexes. Furthermore, we showed that this fasting-induced increase was independent of the presence of testosterone or estrogen. This is the first demonstration that MCH neurons express pCREB in their nuclei in response to fasting and is in good accord with the results of a previous study showing that fasting for 48 h increased pCREB levels in nuclear extracts from the rat hypothalamus, although which neurons expressed the pCREB was not shown (38). An analog of cAMP and agents that increase levels of endogenous cAMP and promote phosphorylation of CREB (29) were reported to elicit a vigorous feeding response when microinjected into the LHA (39). Thus, pCREB can be a reliable marker of activation of MCH neurons by 48-h fasting, and phosphorylation of CREB, probably promoted by an increase in cAMP, may contribute at least in part to the orexigenic activity of MCH neurons. We thus suggest that there is no sex difference in the fasting-induced activation of MCH neurons.
We also examined changes in the activity of MCH neurons after glucose injection in 48-h fasted rats in experiment 1 and found that a significant decrease in the ratio of double-stained neurons was induced after 5 min only in female rats at all stages of the estrous cycle examined. In contrast to female rats, MCH neurons in male rats failed to respond to glucose within 5 min. Additional time-course studies in experiment 2 demonstrated that MCH neurons in male rats responded to glucose 15–30 min after glucose injection, whereas those in female rats, at the stage that was most insensitive to glucose, MCH neurons responded after 5–15 min. In this study we found a clear sex difference in the response of MCH neurons to glucose; the response of MCH neurons to glucose was delayed in male rats compared with that in female rats. This is not due to differences in blood glucose concentrations, because there was no sex difference in blood glucose concentrations after glucose injection between intact male and female rats. Furthermore, this is not due to sex differences in insulin secretion in response to glucose, because there was no sex difference in the effect of castration on insulin secretion and thus it could not explain the results of experiment 3. One possible reason for this delay in the decrease in pCREB may be that the sensitivity of MCH neurons to glucose in male rats is less than that in female rats, and thus MCH neurons do not respond to an increase in glucose concentrations in blood, resulting in a delay of the decrease in pCREB. According to the glucostatic theory that an increase in glucose in blood is a signal for meal termination (33), the dietary glucose may inactivate the orexigenic activity of MCH neurons and induce meal termination more rapidly in female rats than in male rats. This may account for the sex difference in meal size, which is smaller in females than in males (1, 2). Although glucose-sensing neurons may respond to physiological changes in blood glucose and regulate integrative metabolic functions, such as counterregulatory responses, food intake, and metabolic rate (40), the glucostatic theory by itself is not sufficient to explain the complex regulation of feeding behavior (41). Glucose is, obviously, only one of several peripheral metabolic signals involved in the regulation of feeding behavior under physiological conditions (42). Hence, we speculate that the inactivation of MCH neurons by metabolic signals after feeding is faster in females than in males.
We currently do not know how glucose acts on MCH neurons. However, judging from an electrophysiological study showing that some neurons in the LHA respond to changes in glucose concentration in vitro (43), a direct effect of glucose on MCH neurons is likely. Effects of afferent signals from glucose sensors localized at different central and peripheral structures are also likely (41, 44). In the hypothalamus, NPY neurons in the arcuate nucleus have been shown to be glucose-sensitive neurons in an in vitro electrophysiological study (45). Considering the hypothalamic circuit involving the LHA and arcuate nucleus (19), it is probable that glucose affects MCH neurons indirectly via NPY neurons in the arcuate nucleus.
We also found in the present study that the response of MCH neurons to glucose varied depending on the stage of the estrous cycle, i.e. the magnitude of decrease in the number of double-stained neurons after glucose injection was greater at proestrus and estrus than at diestrus 2. An additional time-course study confirmed that the ratio of double-stained neurons at diestrus 2 decreased 15 min after glucose injection to approximately 10%, which was similar to the levels at proestrus and estrus 5 min after glucose injection, indicating that the response of MCH neurons to glucose at diestrus 2 was the most delayed. As discussed above, if the decrease in the activity of MCH neurons is related to meal termination, it follows that meal termination occurs more rapidly at proestrus and estrus than at diestrus 2. In support of this, the meal size was smaller at proestrus or estrus than at diestrus (2, 4). Therefore, MCH neurons may also participate in the fluctuation of feeding behavior during the estrous cycle.
Our findings in experiment 3 indicate that the sex difference in MCH neurons is at least in part attributable to the presence of gonadal steroid hormones in adults. That is, the presence of testosterone in males and that of estrogen in females are critical to the male- and female-type responses of MCH neurons to glucose. In male rats, we found that orchidectomy significantly enhanced the response of MCH neurons to glucose, but testosterone replacement attenuated the effect. Because orchidectomy decreased food intake (46), and this effect was reversed by testosterone replacement in ORX rats (7, 47, 48), we speculate that testosterone attenuates the response of MCH neurons to metabolic signals and then increases food intake in males. In contrast, estrogen replacement did not cause any significant effect on orchidectomy-induced enhancement of the response to glucose in male rats. This result is in good accord with a report that food intake in ORX rats is relatively insensitive to the influence of estrogen compared with that in OVX rats (8). The results of the present study in male rats therefore suggest that estrogen could no longer enhance the response of MCH neurons to glucose, because orchidectomy caused the response to reach its maximum.
In female rats, ovariectomy attenuated the response of MCH neurons to glucose, and estrogen replacement prevented ovariectomy-induced attenuation of the response. These results are in good accord with previous studies in which ovariectomy in adult female rats increased food intake (7, 46), and this effect was reversed by estrogen replacement (6). Thus, it can be easily hypothesized that estrogen enhances the response of MCH neurons to metabolic signals in females and causes decreases in food intake and body weight. In support of this, estrogen inhibited the expression of MCH mRNA in the LHA of female (49) and male (27) rats. In contrast to the effect of testosterone in ORX rats, testosterone enhanced the response of MCH neurons in OVX rats as estrogen did. This effect of testosterone in OVX rats could be explained by postulating that testosterone enhances the response of MCH neurons after being converted to estrogen. In support of this speculation, it has been suggested that in male rats, the reduced food intake and weight gain seen during high-dose testosterone treatment are due to aromatization of testosterone to estrogen (8). In light of previous reports that females are more sensitive to the effect of estrogen on food intake than males (8, 50) and males are more sensitive to the effect of testosterone on food intake (47, 50), the present study indicates that the dominant effect of testosterone on the response of MCH neurons in females is not testosterone-induced attenuation, but is estrogen-induced enhancement. In males, in contrast, the dominant effect of testosterone on the response of MCH neurons is not estrogen-induced enhancement, but is testosterone-induced attenuation. Hence, the present study may provide a neural basis for the sex difference in feeding behavior, although the mechanisms of action of estrogen and testosterone need to be determined in future studies.
Recent reports demonstrating that estrogen attenuates neuroendocrine responses to hypoglycemia in women (51) and female rats (52) suggest that estrogen enhances the effects of glucose in the brain. In light of the results of the present study, estrogen may enhance glucose transport and its metabolism in the brain (53, 54, 55). This suggests that one mechanism by which estrogen exerts its widely reported protective effects in the brain (56) is by enhancing glucose transport and metabolism.
References
de Castro JM 1987 Circadian rhythms of the spontaneous meal pattern, macronutrient intake, and mood of humans. Physiol Behav 40:437–446
Laviano A, Meguid MM, Gleason JR, Yang ZJ, Renvyle T 1996 Comparison of long-term feeding pattern between male and female Fischer 344 rats: influence of estrous cycle. Am J Physiol 270:R413–R419
Dye L, Blundell JE 1997 Menstrual cycle and appetite control: implications for weight regulation. Hum Reprod 12:1142–1151
Eckel LA, Houpt TA, Geary N 2000 Spontaneous meal patterns in female rats with and without access to running wheels. Physiol Behav 70:397–405
Bell DD, Zucker I 1971 Sex differences in body weight and eating: organization and activation by gonadal hormones in the rat. Physiol Behav 7:27–34
Blaustein JD, Wade GN 1976 Ovarian influences on the meal patterns of female rats. Physiol Behav 17:201–208
Gentry RT, Wade GN 1976 Androgenic control of food intake and body weight in male rats. J Comp Physiol Psychol 90:18–25
Gentry RT, Wade GN 1976 Sex differences in sensitivity of food intake, body weight, and running-wheel activity to ovarian steroids in rats. J Comp Physiol Psychol 90:747–754
Varma M, Chai JK, Meguid MM, Laviano A, Gleason JR, Yang ZJ, Blaha V 1999 Effect of estradiol and progesterone on daily rhythm in food intake and feeding patterns in Fischer rats. Physiol Behav 68:99–107
Geary N 1999 Estradiol, CCK and satiation. Peptides 22:1251–1263
Wager-Srdar SA, Gannon M, Levine AS 1987 The effect of cholecystokinin on food intake in gonadectomized and intact rats: the influence of sex hormones. Physiol Behav 40:25–28
Flanagan-Cato LM, King JL, Blechman JG, O’Brien MP 1998 Estrogen reduces cholecystokinin-induced c-Fos expression in the rat brain. Neuroenedocrinology 67:384–391
Bittencourt JC, Presse F, Arias C, Peto C, Vaughan J, Nahon JL, Vale W, Sawchenko PE 1992 The melanin-concentrating hormone system of the rat brain: an immuno- and hybridization histochemical characterization. J Comp Neurol 319:218–245
Qu D, Ludwig DS, Gammeltoft S, Piper M, Pelleymounter MA, Cullen MJ, Mathes WF, Przypek R, Kanarek R, Maratos-Flier E 1996 A role for melanin-concentrating hormone in the central regulation of feeding behaviour. Nature 380:243–247
Rossi M, Choi SJ, O’Shea D, Miyoshi T, Ghatei MA, Bloom SR 1997 Melanin-concentrating hormone acutely stimulates feeding, but chronic administration has no effect on body weight. Endocrinology 138:351–355
Allen YS, Adrian TE, Allen JM, Tatemoto K, Crow TJ, Bloom SR, Polak JM 1983 Neuropeptide Y distribution in the rat brain. Science 221:877–879
Clark JT, Kalra PS, Crowley WR, Kalra SP 1984 Neuropeptide Y and human pancreatic polypeptide stimulate feeding behavior in rats. Endocrinology 115:427–429
Stanley BG, Magdalin W, Seirafi A, Thomas WJ, Leibowitz SF 1993 The perifornical area: the major focus of (a) patchily distributed hypothalamic neuropeptide Y-sensitive feeding system(s). Brain Res 604:304–317
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Shimada M, Tritos NA, Lowell BB, Flier JS, Maratos-Flier E 1998 Mice lacking melanin-concentrating hormone are hypophagic and lean. Nature 396:670–674
Ludwig DS, Tritos NA, Mastaitis JW, Kulkarni R, Kokkotou E, Elmquist J, Lowell B, Flier JS, Maratos-Flier E 2001 Melanin-concentrating hormone overexpression in transgenic mice leads to obesity and insulin resistance. J Clin Invest 107:379–386
Bahjaoui-Bouhaddi M, Fellmann D, Griffond B, Bugnon C 1994 Insulin treatment stimulates the rat melanin-concentrating hormone-producing neurons. Neuropeptides 27:251–258
Sergeyev V, Broberger C, Gorbatyuk O, Hokfelt T 2000 Effect of 2-mercaptoacetate and 2-deoxy-D-glucose administration on the expression of NPY, AGRP, POMC, MCH and hypocretin/orexin in the rat hypothalamus. Neuroreport 11:117–121
Erickson JC, Clegg KE, Palmiter RD 1996 Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 381:415–421
Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato M, Hammer RE, Saper CB, Yanagisawa M 1999 Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98:437–451
Willie JT, Chemelli RM, Sinton CM, Yanagisawa M 2001 To eat or to sleep? Orexin in the regulation of feeding and wakefulness. Annu Rev Neurosci 24:429–458
Mystkowski P, Seeley RJ, Hahn TM, Baskin DG, Havel PJ, Matsumoto AM, Wilkinson CW, Peacock-Kinzig K, Blake KA, Schwartz MW 2000 Hypothalamic melanin-concentrating hormone and estrogen-induced weight loss. J Neurosci 20:8637–8642
Ji RR, Rupp F 1997 Phosphorylation of transcription factor CREB in rat spinal cord after formalin-induced hyperalgesia: relationship to c-fos induction. J Neurosci 17:1776–1178
Montminy MR, Bilezikjian L 1987 Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature 328:175–178
Dash PK, Karl KA, Colicos MA, Prywes R, Kandel ER 1991 cAMP response element-binding protein is activated by Ca2+/calmodulin- as well as cAMP-dependent protein kinase. Proc Natl Acad Sci USA 88:5061–5065
He D, Funabashi T, Sano A, Uemura T, Minaguchi H, Kimura F 1999 Effects of glucose and related substrates on the recovery of the electrical activity of gonadotropin-releasing hormone pulse generator which is decreased by insulin-induced hypoglycemia in the estrogen-primed ovariectomized rat. Brain Res 820:71–76
Mitsushima D, Tin-Tin-Win-Shwe, Kimura F 2003 Sexual dimorphism in the GABAergic control of gonadotropin release in intact rats. Neurosci Res 46:399–405
Steiner RA, Bremner WJ, Clifton DK 1982 Regulation of luteinizing hormone pulse frequency and amplitude by testosterone in the adult male rat. Endocrinology 111:2055–2061
Crace CJ, Swenne I, Milner RDG 1989 Long-term follow-up after early protein-calorie malnutrition in young rats: sex difference in glucose tolerance and serum insulin levels. Metabolism 38:933–938
Galipeau DM, Yao L, McNeill JH 2002 Relationship among hyperinsulinemia, insulin resistance, and hypertension is dependent on sex. Am J Phyisol 283:H562–H567
Mayer J 1953 Glucostatic mechanism of regulation of food intake. N Engl J Med 249:13–16
Bahjaoui-Bouhaddi M, Fellmann D, Bugnon C 1994 Induction of Fos-immunoreactivity in prolactin-like containing neurons of the rat lateral hypothalamus after insulin treatment. Neurosci Lett 168:11–15
Sheriff S, Chance WT, Fischer JE, Balasubramaniam A 1997 Neuropeptide Y treatment and food deprivation increase cyclic AMP response element-binding in rat hypothalamus. Mol Pharmacol 51:597–604
Gillard ER, Khan AM, Grewal RS, Mouradi B, Wolfsohn SD, Stanley BG 1998 The second messenger cAMP elicits eating by an anatomically specific action in the perifornical hypothalamus. J Neurosci 18:2646–2652
Yang X–J, Funabashi T, Kow L–M, Mobbs CV 2004 A glucokinase/AP-1 glucose transduction mechanism in the ventromedial hypothalamic satiety center. In: Matschinsky FM, Magnuson MA, eds. Frontiers in diabetes. Vol 16. Basel: Karger; 313–326
Levin BE, Dunn-Meynell AA, Routh VH 1999 Brain glucose sensing and body energy homeostasis: role in obesity and diabetes. Am J Physiol 276:R1223–R1231
Havel PJ 2001 Peripheral signals conveying metabolic information to the brain: short-term and long-term regulation of food intake and energy homeostasis. Exp Biol Med 226:963–977
Orsini JC, Armstrong DL, Wayner MJ 1992 Responses of lateral hypothalamic neurons recorded in vitro to moderate changes in glucose concentration. Brain Res Bull 29:503–505
Niijima A 1986 Neural control of blood glucose level. Jpn J Physiol 36:827–841
Muroya S, Yada T, Shioda S, Takigawa M 1999 Glucose-sensitive neurons in the rat arcuate nucleus contain neuropeptide Y. Neurosci Lett 264:113–116
Kakolewski JW, Cox VC, Valenstein ES 1968 Sex differences in body-weight changes following gonadectomy of rats. Psychol Rep 22:547–554
Rowland DL, Perrings TS, Thommes JA 1980 Comparison of androgenic effects on food intake and body weight in adult rats. Physiol Behav 24:205–209
Chai J–K, Blaha V, Meguid MM, Laviano A, Yang Z–J, Varma M 1999 Use of orchidectomy and testosterone replacement to explore number-to-meal size relationship in male rats. Am J Physiol 276:R1366–R1373
Murray JF, Baker BI, Levy A, Wilson CA 2000 The influence of gonadal steroids on pre-pro melanin-concentrating hormone mRNA in female rats. J Neuroendocrinol 12:53–59
Czaja JA 1984 Sex differences in the activational effects of gonadal hormones on food intake and body weight. Physiol Behav 33:553–558
Sandoval DA, Ertl AC, Richardson A, Tate DB, Davis SN 2003 Estrogen blunts neuroendocrine and metabolic responses to hypoglycemia. Diabetes 52:1749–1755
Adams JM, Legan SJ, Ott CE, Jackson BA 2005 Modulation of hypoglycemia-induced increases in plasma epinephrine by estrogen in the female rat. J Neurosci Res 79:360–367
Shi J, Zhang YQ, Simpkins JW 1997 Effects of 17b-estradiol on glucose transporter 1 expression and endothelial cell survival following focal ischemia in the rats. Exp Brain Res 117:200–206
Luine VN, Khylchevskaya RI, McEwen BS 1975 Effect of gonadal hormones in enzyme activities in brain and pituitary of male and female rats. Brain Res 86:283–292
Hamba H, Sokoloff L 1984 Acute administration of high doses of estrogen increases glucose utilization throughout brain. Brain Res 291:391–394
Wise PM 2003 Estrogens: protective or risk factors in brain function? Prog Neurobiol 69:181–191(Kazutaka Mogi, Toshiya Fu)
Address all correspondence and requests for reprints to: Dr. Toshiya Funabashi, Department of Neuroendocrinology, Yokohama City University Graduate School of Medicine, 3-9 Fukuura, Kanazawa-ku, Yokohama 236-0004, Japan. E-mail: toshiya@med.yokohama-cu.ac.jp.
Abstract
Because there are sex differences in feeding behavior in rats, we looked for a possible sex difference in the response to glucose of melanin-concentrating hormone (MCH) neurons in the lateral hypothalamic area using phosphorylated cAMP response element-binding protein (pCREB) as a marker of neural activity. Intact male rats and female rats at diestrus 2, proestrus, or estrus were fed normally or fasted for 48 h and injected with saline or glucose (400 mg/kg). Thereafter, preparations were subjected to immunohistochemical processing for the double staining of MCH and pCREB. Fasting increased the ratio of MCH neurons with pCREB (double-stained cells) in both male and female rats. In fasted rats, glucose injection decreased the ratio of double-stained cells more promptly in females than in males. The magnitude of decrease caused by glucose was greater at proestrus and estrus than at diestrus 2. Gonadectomy in males enhanced and in females attenuated the response of MCH neurons to glucose. Testosterone and estrogen replacement in males and females, respectively, restored the response of MCH neurons to glucose. The demonstrated sex differences in the response of MCH neurons to glucose correlated well with the gonadal steroid milieu; thus, MCH neurons may play an important role in sex differences in feeding behavior.
Introduction
SEX DIFFERENCES IN feeding behavior have long been recognized in many species. For example, analyses of spontaneous patterns of feeding have shown that the size of a single meal is smaller in females than in males, and the total volume of daily food intake is also smaller in females than in males (1, 2). In addition, the size of a single meal varies during the estrous cycle, decreasing around the time of ovulation in females, including humans (1, 3, 4). Although these sex differences are attributable to gonadal steroid hormones such as testosterone and/or estrogen (5, 6, 7, 8, 9), the neural mechanism is still unclear. For example, although cholecystokinin has been reported to be involved in estrogen-induced decrease in food intake (10), there is still controversy about whether cholecystokinin mediates estrogen-induced anorexia (11, 12). It is also not known whether cholecystokinin mediates testosterone-induced hyperphagia.
Recent studies have identified distinct neurons in the lateral hypothalamic area (LHA) containing melanin-concentrating hormone (MCH) neurons (13) that generate hunger (14, 15). Neuropeptide Y (NPY), which is produced by neurons in the arcuate nucleus of the hypothalamus (16), is another neuropeptide that prominently induces feeding (17, 18). The hypothalamic circuit between the LHA and the arcuate nucleus is thus an important component of the central system controlling feeding behavior (19). In this circuit, judging from studies that genetically disrupted the orexigenic neuropeptides, MCH neurons in the LHA play a key role. Mice lacking the gene encoding MCH are lean (20), whereas mice overexpressing the MCH gene are susceptible to obesity (21). It has also been suggested that MCH neurons are involved in the glucosensing system, because hypothalamic MCH mRNA levels are increased by hypoglycemic conditions such as fasting, insulin injection, and 2-deoxy-D-glucose treatment (14, 22, 23). In contrast to MCH, mice lacking the gene encoding prepro-orexin or NPY display normal growth (24, 25, 26).
It is probable that sex differences in feeding behavior are at least in part attributable to sex differences in the neurons involved in feeding behavior. Because MCH neurons in the LHA play an important role in the regulation of feeding behavior, we hypothesized that there might be a sex difference in the activity of MCH neurons in the LHA during fasting and/or after glucose injection. In support of this view, hypothalamic MCH neurons are suggested to be involved in estrogen-induced weight loss in male rats (27). We therefore examined the response of MCH neurons to an iv injection of glucose after 48 h of fasting in adult male and female rats, using the expression of phosphorylated cAMP response element-binding protein (pCREB) immunoreactivity as a marker of neural activity. On the assumption that a sex difference in MCH neurons existed, we also examined whether it could be affected by gonadal steroid hormones such as testosterone and/or estrogen.
In the present study, we used pCREB as a marker for neural activity increased of the generally used c-Fos, because an increase in the expression of c-Fos needs transcription and/or translation to occur (28), it takes a long time to see changes, and it would be difficult to examine acute changes in response to glucose. In contrast, CREB is a constitutively expressed transcription factor and can be immediately activated by phosphorylation via a cAMP- and calcium-dependent cascade (29, 30). Thus, we thought that pCREB was a suitable marker to examine neural activity that changes within a short period in response to dynamic changes such as blood glucose concentrations.
Materials and Methods
Animals
Adult male and female Wistar rats (Charles River Laboratories, Yokohama, Japan) were maintained at a constant temperature at 24–26 C under controlled lighting conditions (lights on, 0500–1900 h) with food and water available ad libitum. Daily vaginal smears were taken from all female rats, and those exhibiting two or more consecutive 4-d estrous cycles were used in experiments 1 and 2. Forty-eight hours before saline or glucose injection, a silicone cannula for iv injection was implanted into the right atrium of all rats under ether anesthesia. This procedure did not interfere with the estrous cycle. The number of rats used for each group is indicated in the figures. All animal housing and surgical procedures were in accord with the guidelines laid down by the institutional animal care and use committee of Yokohama City University School of Medicine.
Determination of blood glucose and insulin concentrations
To determine changes in blood glucose and insulin concentrations after glucose injection (400 mg/kg), intact males, orchidectomized (ORX) males, females at proestrus and diestrus 2, and ovariectomized (OVX) females were implanted with a silicone cannula in the right atrium under ether anesthesia. Male and female rats were castrated 4 wk before the experiment. After cannula implantation, they were fasted for 48 h and injected with 400 mg/kg glucose dissolved in 200 μl saline after the first blood sampling of 100 μl. Control rats were injected with the same volume of saline. Thereafter, blood was sampled 5, 10, 15, 30, 60, and 120 min after the injection though the same cannula. An equal volume of heparinized saline (2 IU/ml) was injected after each sampling. As another control, blood samples were obtained from intact males, ORX males, females at proestrus, and OVX females that were not fasted. Blood glucose concentrations were determined with an instant blood glucose assay apparatus (ACCU-CHEK Comfort, Roche, Tokyo, Japan). Serum insulin concentrations were measured with an ELISA kit (Morinaga Institute of Biological Sciences, Yokohama, Japan).
Experiment 1: sex difference in the activity of MCH neurons
To determine whether there is a sex difference in the activity of MCH neurons, some male and female rats were fed normally and injected iv with 200 μl saline (Fed+Saline group), and other male and female rats were fasted for 48 h and given an iv injection of 200 μl saline (Fast+Saline group) or glucose (Fast+Glucose group; 400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. The injection was made through the silicon cannula in the right atrium as described above. The dose of glucose used in the present study was based on a previous study which showed that this treatment could reverse the inhibition of pulsatile LH secretion caused by insulin-induced hypoglycemia (31). For females, three groups, composed of rats at diestrus 2 (D2 group), proestrus (P group), or estrus (E group), were prepared to check the effect of the stage of the estrous cycle. The rats in the D2 group had been fasted from the day of estrus, those in the P group had been fasted from the day of last diestrus 1, and those in the E group had been fasted from the day of last diestrus 2. Thus, 12 groups of rats (four groups containing three treatment groups each) were prepared, and they were processed blindly for immunohistochemistry.
In all groups, rats were killed 5 min after saline or glucose injection by an iv injection of an overdose of pentobarbital (100 mg/kg). Perfusion through the cardiac ventricle was quickly started with 2% paraformaldehyde and 4% acrolein in phosphate buffer (PB; pH 7.5) at approximately 4 C. After perfusion, the brains were removed from the cranium, fixed overnight at 4 C in PB containing 2% paraformaldehyde, and incubated overnight at 4 C in 30% sucrose in PB. The brains were then frozen with powdered dry ice and stored at –70 C until immunohistochemical processing.
Experiment 2: time-dependent changes in the activity of MCH neurons after glucose injection
To examine time-dependent changes in the activity of MCH neurons after iv glucose injection, male rats were fasted for 48 h and injected iv with glucose (400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. Rats were killed 15, 30, and 60 min after glucose injection (Fast+Glucose 15 min group, Fast+Glucose 30 min group, and Fast+Glucose 60 min group) by iv injection of an overdose of pentobarbital. Female rats at diestrus 2, which had been fasted for 48 h from the day of estrus, were similarly killed. All subsequent procedures to obtain brains were as described for experiment 1.
Experiment 3: effects of testosterone and estrogen on the response of MCH neurons to glucose
To determine whether gonadal steroid hormones affected the response of MCH neurons to glucose, male and female rats were castrated under ether anesthesia and used for the experiment 4 wk later. ORX and OVX rats were implanted sc with a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 30 mm) containing 100% testosterone crystals (ORX+T group and OVX+T group), a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 15 mm) containing 20% 17?-estradiol crystals (ORX+E2 group and OVX+E2 group), or a silicone tube (inside diameter, 2.0 mm; outside diameter, 3.0 mm; length, 30 mm) containing cholesterol alone (ORX group and OVX group) as a control. All silicone tubes had been previously soaked in saline for more than 24 h to facilitate rapid hormone release. It was shown previously that a tube containing this dose of 17?-estradiol produced a serum 17?-estradiol concentration equivalent to that observed at proestrus in intact female rats (32), and a tube containing testosterone produced a serum testosterone concentration equivalent to that observed in intact male rats (33). After implantation of the tubes, some rats were fed normally, and other rats were fasted for 48 h. Normally fed rats were injected iv with 200 μl saline (Fed+Saline group), and fasted rats were injected iv with 200 μl saline (Fast+Saline group) or glucose (Fast+Glucose group; 400 mg/kg) dissolved in 200 μl saline at 1000–1200 h. All rats were killed 5 min after iv injection. Thus, a total of 18 groups of rats (six groups containing three treatment groups each: ORX, ORX+T, ORX+E2, OVX, OVX+T, and OVX+E2 groups x Fed+Saline, Fast+Saline, and Fast+Glucose groups) was prepared. All subsequent procedures to obtain brains were as described for experiment 1.
Immunohistochemistry
Thirty-micrometer frozen coronal sections were cut with a Bright cryostat (Jencons Ltd., Leighton Buzzard, UK), and every fourth section was used for the study. The sections were incubated with 1% sodium borohydride in PBS and then with 0.3% H2O2 in 20% methanol in PBS. They were incubated overnight with rabbit polyclonal antibody to pCREB diluted 1:900 (Cell Signaling Technology, Beverly, MA) in PBS containing 1.5% normal goat serum and 0.1% Triton X-100. The next day they were incubated with biotinylated antirabbit IgG diluted 1:200 in PBS containing 1.5% normal goat serum and 0.05% Triton X-100, and thereafter incubated with streptavidin-biotin-peroxidase complex (Vectastain Elite ABC Kit, Vector Laboratories, Inc., Burlingame, CA). Bound peroxidase was visualized by incubating sections for 30 min in 0.05% 3,3'-diaminobenzidine with H2O2. For double staining, sections were then incubated overnight with rabbit polyclonal antibody to MCH diluted 1:1800 (Phoenix Pharmaceuticals, Belmont, CA) in PBS containing 1.5% normal goat serum and 0.1% Triton X-100. They were additionally incubated with biotinylated antirabbit IgG diluted 1:200 in PBS containing 1.5% normal goat serum and 0.05% Triton X-100, then with Cy3-labeled streptavidin (Amersham Biosciences, Little Chalfont, UK). Finally, they were mounted on glass slides, dehydrated in graded alcohol, cleared in xylene, and coverslipped with Permount (Fisher Scientific, Fairlawn, NJ). A total of eight series of immunohistochemical procedures was performed, and in each of them, the incubation time of 3,3'-diaminobenzidine staining and other experimental conditions were checked carefully for reproducibility.
Histological analysis
The counting was performed by an investigator blind to the experimental conditions and expectations. We initially used every fourth section through the LHA, and thus approximately 10 sections/rat were stained with MCH and pCREB antibodies. Using a fluorescence microscope, five sections per rat were finally selected for counting MCH-immunoreactive cells (MCH neurons) and pCREB-immunoreactive cells (pCREB cells) in the dorsolateral hypothalamic group of the LHA, because the distribution of MCH neurons varies from anterior to posterior in the diencepalon, and this group of the LHA comprises the numerically predominant grouping of MCH-expressing cells in the diencephalon (13). Sections were carefully matched across all animals in all experimental groups. The number of MCH neurons in the LHA (square region of 1.7 x 1.0 mm) was counted bilaterally with a fluorescence microscope at 20 x 10 magnification. MCH cells in which a nucleus was clearly visible and surrounded by fluorescent cytoplasmic staining were counted. In the same sections, pCREB cells were also observed, and the number of MCH neurons expressing pCREB-immunoreactive nuclei (double-stained neurons; Fig. 1) was counted. Cells were defined as double stained for MCH and pCREB when a blue-black nucleus was surrounded by fluorescent cytoplasm at 20 x 10 magnification. Then the average ratio (percentage) of the number of double-stained neurons to the number of MCH neurons in each rat was calculated. Both the average number of MCH neurons per section in each rat and the average percentage of double-stained neurons per section in each rat were used for statistical analysis. In male rats and female rats of the P group in experiment 1, we counted the number of pCREB cells in the LHA.
FIG. 1. Representative photographs of double-stained neurons in the LHA. Fluorescent MCH-immunoreactive neurons (top, arrowheads) and pCREB-immunoreactive nuclei (bottom, arrowheads) in the same tissue section (magnification, x400) in the Fast+Saline group (left panels) and Fast+Glucose group (right panels) of male and female (P) rats in experiment 1 are shown.
Data were analyzed by one-way ANOVA, followed by Fisher’s protected least significant difference post hoc test for comparisons of treatment groups in each animal group. Differences were considered significant at P < 0.05.
Results
In control (i.e. not fasted) intact males, ORX males, females at proestrus, and OVX females, blood glucose and insulin levels were 103.8 ± 1.5 mg/dl (n = 4) and 1273 ± 54 pg/ml, 107.4 ± 2.4 (n = 5) mg/dl and 1433 ± 121 pg/ml, 101.3 ± 2.4 mg/dl (n = 4) and 1306 ± 113 pg/ml, and 105.0 ± 2.7 mg/dl (n = 4) and 1323 ± 143 pg/ml, respectively. No significant differences in glucose (F3,13 = 0.811; P > 0.1) or insulin (F3,13 = 0.398; P > 0.5) were observed. As shown in Table 1, one-way ANOVA in each saline-injected group showed no significant effect of saline injection on blood glucose concentrations: the male group (F6,21 = 0.811; P > 0.5), the ORX group (F6,21 = 0.61; P > 0.5), the P group (F6,21 = 1.61; P > 0.1), and the OVX group (F6,28 = 2.29; P > 0.05), suggesting that the control procedure used in the present study did not affect blood glucose concentrations. However, blood glucose concentrations were significantly increased after glucose injection in all groups. One-way ANOVA at each time point showed that there was no significant difference in blood glucose concentrations among the groups (Table 2). In addition, the area under the curve (AUC), which was calculated for each rat, was not significantly different among the groups (F4,19 = 0.562; P > 0.5). This suggests that changes in blood glucose concentrations after glucose injection were uniform among the groups. Insulin concentrations were also significantly increased after glucose injection in all groups, but one-way ANOVA at each time point showed that there was a significant difference in insulin concentrations among the groups (Table 2). That is, insulin concentrations in the intact male and the ORX groups were significantly higher than those in the D2, P, and OVX groups at some time points after glucose injection. In fact, the AUC of insulin was significantly different among the groups (F4,19 = 9.09; P < 0.002). Post hoc comparison showed that the AUC in the intact male and ORX groups was significantly greater than in the D2, P, and OVX groups (P < 0.05). This result suggested that insulin secretion in response to glucose in male rats was greater than that in female rats and was in good accord with previous reports of a sex difference in glucose tolerance (34, 35)
View this table:
[in this window]
[in a new window]
TABLE 1. Changes in blood glucose concentrations (milligrams per deciliter) after saline injection
TABLE 2. Changes in blood glucose and insulin concentrations after 400 mg/kg glucose injection
In all rats, the overall distribution of MCH neurons with red fluorescent cytoplasm was in good accord with a previous report (13) in all experiments. The average number of MCH neurons per section was approximately 500 in all groups, and this number did not vary in any experiment (data not shown). In the intact male and P groups, the number of pCREB-expressing cells was slightly larger than that of MCH neurons expressing pCREB (Table 3), but changes in the number of pCREB cells were similar to those in MCH neurons expressing pCREB, as indicated below. These results suggest that MCH neurons were not the only population of cells expressing pCREB in the LHA.
TABLE 3. Changes in the number of pCREB-expressing cells in the lateral hypothalamic area of male rats and female rats at proestrus
Experiment 1
All female rats in the E and D2 groups exhibited a regular 4-d estrous cycle, as monitored by vaginal smear during the experiments. Almost 40% of female rats fasted for 48 h in the P group, however, had a diestrus-like vaginal smear during the day on which they were supposed to be in proestrus. Data for rats showing a perturbed cycle were discarded, and only rats showing a proestrous vaginal smear were included in the P group.
In all Fed+Saline groups, pCREB cells with blue-black nuclei were scattered in the LHA and the arcuate nucleus of the hypothalamus (data not shown), and the ratio of double-stained neurons was approximately 10% (Fig. 2).
FIG. 2. Effects of 48-h fasting or glucose injection after 48 h of fasting on the ratio (percentage) of MCH-immunoreactive neurons expressing pCREB-immunoreactive nuclei in experiment 1. In all groups, rats were examined 5 min after saline or glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. other treatment groups. See text for details of statistics. D2, Diestrus 2; P, proestrus; E, estrus.
In all Fast+Saline groups, a large number of pCREB cells was observed in the LHA, but not in the arcuate nucleus of the hypothalamus (data not shown), and the ratio of double-stained neurons was approximately 40% (Fig. 2). One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in each animal group: the intact male group (F2,12 = 7.45; P < 0.01), the P group (F2,12 = 9.35; P < 0.01), the E group (F2,12 = 5.53; P < 0.05), and the D2 group (F2,11 = 11.38; P < 0.01). In all groups, post hoc comparison showed a significantly greater ratio of double-stained neurons in the Fast+Saline group than in the Fed+Saline group (Fig. 2; P < 0.05). These results indicated that 48-h fasting significantly increased the ratio of double-stained neurons in all animal groups.
In the Fast+Glucose group, the number of pCREB cells in the LHA varied among animal groups (data not shown). Post hoc comparison in the male group showed a significantly greater ratio of double-stained neurons in the Fast+Glucose group than in the Fed+Saline group (P < 0.05), but there was no significant difference between the Fast+Glucose group and the Fast+Saline group, indicating that glucose injection did not affect the increase in the ratio of double-stained neurons caused by fasting in intact male rats (Fig. 2). In contrast, pCREB cells were relatively few in the LHA in all female groups after glucose injection (data not shown). Post hoc comparisons in all female groups showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose group than in the Fast+Saline group (Fig. 2; P < 0.05). In both the P and E groups, the ratio of double-stained neurons did not differ between the Fast+Glucose group and the Fed+Saline group, indicating that glucose injection completely blocked the fasting-induced increase in the ratio of double-stained neurons (Fig. 2). In the D2 group, in contrast, the ratio of double-stained neurons was significantly greater in the Fast+Glucose group than in the Fed+Saline group (Fig. 2; P < 0.05), indicating that 48 h after fasting, glucose injection was less effective in the D2 group than in the P and E groups.
Experiment 2
The data from experiment 2 were analyzed with the data from experiment 1 (the Fast+Saline and Fast+Glucose groups of intact male rats, and the D2 group of female rats; Fig. 3).
FIG. 3. Time-dependent changes in the ratio (percentage) of MCH-immunoreactive cells expressing pCREB-immunoreactive nuclei in the Fast+Saline and Fast+Glucose groups of male and female (D2) rats in experiment 2. In all groups, rats were fasted for 48 h and then examined after 5 min of saline injection or after 15, 30, and 60 min of glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. Fast+Saline groups.
In intact male rats, pCREB cells in the LHA were few in the Fast+Glucose 15 min group and the Fast+Glucose 30 min group, and the ratio of double-stained neurons was less than 10% (Fig. 3, upper panel). The number of pCREB cells was relatively large in the Fast+Glucose 60 min group. One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in intact male rats (F4,20 = 5.45; P < 0.01). Post hoc comparison in intact male rats showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose 15 min group and the Fast+Glucose 30 min group than in the Fast+Saline group (P < 0.05). These results indicated that glucose injection significantly decreased the ratio of double-stained neurons 15 and 30 min after injection in intact male rats, but the effect disappeared 60 min after injection.
In the D2 group of female rats, pCREB cells in the LHA were few in the Fast+Glucose 15 min group, and the ratio of double-stained neurons was approximately 10% (Fig. 3, lower panel). The number of pCREB cells was relatively large in the Fast+Glucose 30 min group and the Fast+Glucose 60 min group. One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in the D2 group (F4,19 = 11.30; P < 0.01). Post hoc comparison in the D2 group showed a significantly smaller ratio of double-stained neurons in the Fast+Glucose 5 min group and the Fast+Glucose 15 min group than in the Fast+Saline group (P < 0.05). These results indicated that glucose injection significantly decreased the ratio of double-stained neurons 5 and 15 min after injection in female rats, but the effect disappeared 30 min after injection.
Experiment 3
In all Fed+Saline groups and Fast+Saline groups, the ratios of double-stained neurons were approximately 10% and 40%, respectively (Fig. 4). One-way ANOVA showed a significant effect of treatment on the ratio of double-stained neurons in each animal group: the ORX group (F2,10 = 15.14; P < 0.001), the ORX+T group (F2,10 = 4.61; P < 0.05), the ORX+E2 group (F2,12 = 38.14; P < 0.0001), the OVX group (F2,9 = 6.62; P < 0.05), the OVX+T group (F2,11 = 11.03; P < 0.005), and the OVX+E2 group (F2,9 = 6.41; P < 0.05). In all animal groups, post hoc comparison showed a significantly greater ratio of double-stained neurons in the Fast+Saline group than in the Fed+Saline group (Fig. 4; P < 0.05). Because the ratio of double-stained neurons in the Fed+Saline groups and the Fast+Saline groups in experiment 3 was comparable to that in experiment 1, these results indicated that 48-h fasting significantly increased the ratio of double-stained neurons regardless of the presence or absence of gonadal steroid hormones.
FIG. 4. Effects of testosterone or estrogen on the ratio (percentage) of MCH-immunoreactive neurons expressing pCREB-immunoreactive nuclei in the Fed+Saline, Fast+Saline, and Fast+Glucose groups in experiment 3. In all groups, rats were examined 5 min after saline or glucose injection. Columns and vertical lines indicate the mean and SEM, respectively. Numbers in parentheses refer to the number of rats. Numbers under columns are body weights (mean ± SEM) at the time of testing. *, P < 0.05 vs. other treatment groups. See text for details of statistics. E2, Implantation of 17?-estradiol crystals; T, implantation of testosterone crystals.
Effects of glucose injection varied among the groups. In the ORX and ORX+E2 groups, post hoc comparisons showed that the ratio of double-stained neurons in the Fast+Glucose group was significantly smaller than that in the Fast+Saline group (P < 0.001) and that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fed+Saline group (P > 0.5). In the ORX+T groups, post hoc comparison showed that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fast+Saline group and the Fed+Saline group (P > 0.5). These results indicated that orchidectomy and estrogen replacement decreased the ratio of double-stained neurons 5 min after glucose injection, but testosterone replacement blocked the orchidectomy-induced decrease in the ratio of double-stained neurons in male rats.
In the OVX+T and OVX+E2 groups, post hoc comparisons showed that the ratio of double-stained neurons in the Fast+Glucose group was significantly smaller than that in the Fast+Saline group (P < 0.05) and that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fed+Saline group (P > 0.5). In the OVX groups, post hoc comparison showed that the ratio of double-stained neurons in the Fast+Glucose group was not different from that in the Fast+Saline group (P > 0.5) and that the ratio of double-stained neurons in the Fast+Glucose group was significantly larger than that in the Fed+Saline group (P < 0.05). These results indicated that ovariectomy increased the ratio of double-stained neurons 5 min after glucose injection, but both testosterone and estrogen replacement blocked the ovariectomy-induced increase in the ratio of double-stained neurons in female rats.
Discussion
The present study demonstrated sex differences in the response of MCH neurons to glucose, using the expression of pCREB as a marker of neural activity. Fasting for 48 h activated MCH neurons in a similar manner in both male and female rats, but inactivation of MCH neurons by glucose after 48 h of fasting occurred more rapidly in female rats than in male rats. These sex differences could be attributed to gonadal steroid hormones, including testosterone and estrogen. Taking into account both the orexigenic effect of MCH (14, 15) and the glucostatic theory of feeding (36), our findings provide a molecular/neurotransmitter basis for neurons that are controlled by glucose in the LHA and suggest that MCH neurons in the LHA play an important role in sex differences in feeding behavior, acting through the CREB pathway (29, 30).
Although hypoglycemia caused by treatments such as fasting, insulin injection, or 2-deoxy-D-glucose injection increases the expression of hypothalamic MCH mRNA (14, 22, 23), there is little information on the second messenger system that would be stimulated by hypoglycemic conditions. For example, Fos expression was increased in the LHA by insulin-induced hypoglycemia, but there were few MCH neurons that expressed Fos protein (22, 37). Thus, it is unclear whether Fos proteins are involved in the control of the expression of MCH mRNA in the LHA by hypoglycemia. In experiment 1, we showed that under normal feeding conditions in male and female rats at all stages of the estrous cycle examined, approximately 10% of MCH neurons expressed pCREB-immunoreactive nuclei in the LHA, indicating that only a small population of MCH neurons is active in the LHA. However, we found that fasting for 48 h increased the ratio of double-stained neurons to approximately 40% in both sexes. Furthermore, we showed that this fasting-induced increase was independent of the presence of testosterone or estrogen. This is the first demonstration that MCH neurons express pCREB in their nuclei in response to fasting and is in good accord with the results of a previous study showing that fasting for 48 h increased pCREB levels in nuclear extracts from the rat hypothalamus, although which neurons expressed the pCREB was not shown (38). An analog of cAMP and agents that increase levels of endogenous cAMP and promote phosphorylation of CREB (29) were reported to elicit a vigorous feeding response when microinjected into the LHA (39). Thus, pCREB can be a reliable marker of activation of MCH neurons by 48-h fasting, and phosphorylation of CREB, probably promoted by an increase in cAMP, may contribute at least in part to the orexigenic activity of MCH neurons. We thus suggest that there is no sex difference in the fasting-induced activation of MCH neurons.
We also examined changes in the activity of MCH neurons after glucose injection in 48-h fasted rats in experiment 1 and found that a significant decrease in the ratio of double-stained neurons was induced after 5 min only in female rats at all stages of the estrous cycle examined. In contrast to female rats, MCH neurons in male rats failed to respond to glucose within 5 min. Additional time-course studies in experiment 2 demonstrated that MCH neurons in male rats responded to glucose 15–30 min after glucose injection, whereas those in female rats, at the stage that was most insensitive to glucose, MCH neurons responded after 5–15 min. In this study we found a clear sex difference in the response of MCH neurons to glucose; the response of MCH neurons to glucose was delayed in male rats compared with that in female rats. This is not due to differences in blood glucose concentrations, because there was no sex difference in blood glucose concentrations after glucose injection between intact male and female rats. Furthermore, this is not due to sex differences in insulin secretion in response to glucose, because there was no sex difference in the effect of castration on insulin secretion and thus it could not explain the results of experiment 3. One possible reason for this delay in the decrease in pCREB may be that the sensitivity of MCH neurons to glucose in male rats is less than that in female rats, and thus MCH neurons do not respond to an increase in glucose concentrations in blood, resulting in a delay of the decrease in pCREB. According to the glucostatic theory that an increase in glucose in blood is a signal for meal termination (33), the dietary glucose may inactivate the orexigenic activity of MCH neurons and induce meal termination more rapidly in female rats than in male rats. This may account for the sex difference in meal size, which is smaller in females than in males (1, 2). Although glucose-sensing neurons may respond to physiological changes in blood glucose and regulate integrative metabolic functions, such as counterregulatory responses, food intake, and metabolic rate (40), the glucostatic theory by itself is not sufficient to explain the complex regulation of feeding behavior (41). Glucose is, obviously, only one of several peripheral metabolic signals involved in the regulation of feeding behavior under physiological conditions (42). Hence, we speculate that the inactivation of MCH neurons by metabolic signals after feeding is faster in females than in males.
We currently do not know how glucose acts on MCH neurons. However, judging from an electrophysiological study showing that some neurons in the LHA respond to changes in glucose concentration in vitro (43), a direct effect of glucose on MCH neurons is likely. Effects of afferent signals from glucose sensors localized at different central and peripheral structures are also likely (41, 44). In the hypothalamus, NPY neurons in the arcuate nucleus have been shown to be glucose-sensitive neurons in an in vitro electrophysiological study (45). Considering the hypothalamic circuit involving the LHA and arcuate nucleus (19), it is probable that glucose affects MCH neurons indirectly via NPY neurons in the arcuate nucleus.
We also found in the present study that the response of MCH neurons to glucose varied depending on the stage of the estrous cycle, i.e. the magnitude of decrease in the number of double-stained neurons after glucose injection was greater at proestrus and estrus than at diestrus 2. An additional time-course study confirmed that the ratio of double-stained neurons at diestrus 2 decreased 15 min after glucose injection to approximately 10%, which was similar to the levels at proestrus and estrus 5 min after glucose injection, indicating that the response of MCH neurons to glucose at diestrus 2 was the most delayed. As discussed above, if the decrease in the activity of MCH neurons is related to meal termination, it follows that meal termination occurs more rapidly at proestrus and estrus than at diestrus 2. In support of this, the meal size was smaller at proestrus or estrus than at diestrus (2, 4). Therefore, MCH neurons may also participate in the fluctuation of feeding behavior during the estrous cycle.
Our findings in experiment 3 indicate that the sex difference in MCH neurons is at least in part attributable to the presence of gonadal steroid hormones in adults. That is, the presence of testosterone in males and that of estrogen in females are critical to the male- and female-type responses of MCH neurons to glucose. In male rats, we found that orchidectomy significantly enhanced the response of MCH neurons to glucose, but testosterone replacement attenuated the effect. Because orchidectomy decreased food intake (46), and this effect was reversed by testosterone replacement in ORX rats (7, 47, 48), we speculate that testosterone attenuates the response of MCH neurons to metabolic signals and then increases food intake in males. In contrast, estrogen replacement did not cause any significant effect on orchidectomy-induced enhancement of the response to glucose in male rats. This result is in good accord with a report that food intake in ORX rats is relatively insensitive to the influence of estrogen compared with that in OVX rats (8). The results of the present study in male rats therefore suggest that estrogen could no longer enhance the response of MCH neurons to glucose, because orchidectomy caused the response to reach its maximum.
In female rats, ovariectomy attenuated the response of MCH neurons to glucose, and estrogen replacement prevented ovariectomy-induced attenuation of the response. These results are in good accord with previous studies in which ovariectomy in adult female rats increased food intake (7, 46), and this effect was reversed by estrogen replacement (6). Thus, it can be easily hypothesized that estrogen enhances the response of MCH neurons to metabolic signals in females and causes decreases in food intake and body weight. In support of this, estrogen inhibited the expression of MCH mRNA in the LHA of female (49) and male (27) rats. In contrast to the effect of testosterone in ORX rats, testosterone enhanced the response of MCH neurons in OVX rats as estrogen did. This effect of testosterone in OVX rats could be explained by postulating that testosterone enhances the response of MCH neurons after being converted to estrogen. In support of this speculation, it has been suggested that in male rats, the reduced food intake and weight gain seen during high-dose testosterone treatment are due to aromatization of testosterone to estrogen (8). In light of previous reports that females are more sensitive to the effect of estrogen on food intake than males (8, 50) and males are more sensitive to the effect of testosterone on food intake (47, 50), the present study indicates that the dominant effect of testosterone on the response of MCH neurons in females is not testosterone-induced attenuation, but is estrogen-induced enhancement. In males, in contrast, the dominant effect of testosterone on the response of MCH neurons is not estrogen-induced enhancement, but is testosterone-induced attenuation. Hence, the present study may provide a neural basis for the sex difference in feeding behavior, although the mechanisms of action of estrogen and testosterone need to be determined in future studies.
Recent reports demonstrating that estrogen attenuates neuroendocrine responses to hypoglycemia in women (51) and female rats (52) suggest that estrogen enhances the effects of glucose in the brain. In light of the results of the present study, estrogen may enhance glucose transport and its metabolism in the brain (53, 54, 55). This suggests that one mechanism by which estrogen exerts its widely reported protective effects in the brain (56) is by enhancing glucose transport and metabolism.
References
de Castro JM 1987 Circadian rhythms of the spontaneous meal pattern, macronutrient intake, and mood of humans. Physiol Behav 40:437–446
Laviano A, Meguid MM, Gleason JR, Yang ZJ, Renvyle T 1996 Comparison of long-term feeding pattern between male and female Fischer 344 rats: influence of estrous cycle. Am J Physiol 270:R413–R419
Dye L, Blundell JE 1997 Menstrual cycle and appetite control: implications for weight regulation. Hum Reprod 12:1142–1151
Eckel LA, Houpt TA, Geary N 2000 Spontaneous meal patterns in female rats with and without access to running wheels. Physiol Behav 70:397–405
Bell DD, Zucker I 1971 Sex differences in body weight and eating: organization and activation by gonadal hormones in the rat. Physiol Behav 7:27–34
Blaustein JD, Wade GN 1976 Ovarian influences on the meal patterns of female rats. Physiol Behav 17:201–208
Gentry RT, Wade GN 1976 Androgenic control of food intake and body weight in male rats. J Comp Physiol Psychol 90:18–25
Gentry RT, Wade GN 1976 Sex differences in sensitivity of food intake, body weight, and running-wheel activity to ovarian steroids in rats. J Comp Physiol Psychol 90:747–754
Varma M, Chai JK, Meguid MM, Laviano A, Gleason JR, Yang ZJ, Blaha V 1999 Effect of estradiol and progesterone on daily rhythm in food intake and feeding patterns in Fischer rats. Physiol Behav 68:99–107
Geary N 1999 Estradiol, CCK and satiation. Peptides 22:1251–1263
Wager-Srdar SA, Gannon M, Levine AS 1987 The effect of cholecystokinin on food intake in gonadectomized and intact rats: the influence of sex hormones. Physiol Behav 40:25–28
Flanagan-Cato LM, King JL, Blechman JG, O’Brien MP 1998 Estrogen reduces cholecystokinin-induced c-Fos expression in the rat brain. Neuroenedocrinology 67:384–391
Bittencourt JC, Presse F, Arias C, Peto C, Vaughan J, Nahon JL, Vale W, Sawchenko PE 1992 The melanin-concentrating hormone system of the rat brain: an immuno- and hybridization histochemical characterization. J Comp Neurol 319:218–245
Qu D, Ludwig DS, Gammeltoft S, Piper M, Pelleymounter MA, Cullen MJ, Mathes WF, Przypek R, Kanarek R, Maratos-Flier E 1996 A role for melanin-concentrating hormone in the central regulation of feeding behaviour. Nature 380:243–247
Rossi M, Choi SJ, O’Shea D, Miyoshi T, Ghatei MA, Bloom SR 1997 Melanin-concentrating hormone acutely stimulates feeding, but chronic administration has no effect on body weight. Endocrinology 138:351–355
Allen YS, Adrian TE, Allen JM, Tatemoto K, Crow TJ, Bloom SR, Polak JM 1983 Neuropeptide Y distribution in the rat brain. Science 221:877–879
Clark JT, Kalra PS, Crowley WR, Kalra SP 1984 Neuropeptide Y and human pancreatic polypeptide stimulate feeding behavior in rats. Endocrinology 115:427–429
Stanley BG, Magdalin W, Seirafi A, Thomas WJ, Leibowitz SF 1993 The perifornical area: the major focus of (a) patchily distributed hypothalamic neuropeptide Y-sensitive feeding system(s). Brain Res 604:304–317
Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG 2000 Central nervous system control of food intake. Nature 404:661–671
Shimada M, Tritos NA, Lowell BB, Flier JS, Maratos-Flier E 1998 Mice lacking melanin-concentrating hormone are hypophagic and lean. Nature 396:670–674
Ludwig DS, Tritos NA, Mastaitis JW, Kulkarni R, Kokkotou E, Elmquist J, Lowell B, Flier JS, Maratos-Flier E 2001 Melanin-concentrating hormone overexpression in transgenic mice leads to obesity and insulin resistance. J Clin Invest 107:379–386
Bahjaoui-Bouhaddi M, Fellmann D, Griffond B, Bugnon C 1994 Insulin treatment stimulates the rat melanin-concentrating hormone-producing neurons. Neuropeptides 27:251–258
Sergeyev V, Broberger C, Gorbatyuk O, Hokfelt T 2000 Effect of 2-mercaptoacetate and 2-deoxy-D-glucose administration on the expression of NPY, AGRP, POMC, MCH and hypocretin/orexin in the rat hypothalamus. Neuroreport 11:117–121
Erickson JC, Clegg KE, Palmiter RD 1996 Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 381:415–421
Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato M, Hammer RE, Saper CB, Yanagisawa M 1999 Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98:437–451
Willie JT, Chemelli RM, Sinton CM, Yanagisawa M 2001 To eat or to sleep? Orexin in the regulation of feeding and wakefulness. Annu Rev Neurosci 24:429–458
Mystkowski P, Seeley RJ, Hahn TM, Baskin DG, Havel PJ, Matsumoto AM, Wilkinson CW, Peacock-Kinzig K, Blake KA, Schwartz MW 2000 Hypothalamic melanin-concentrating hormone and estrogen-induced weight loss. J Neurosci 20:8637–8642
Ji RR, Rupp F 1997 Phosphorylation of transcription factor CREB in rat spinal cord after formalin-induced hyperalgesia: relationship to c-fos induction. J Neurosci 17:1776–1178
Montminy MR, Bilezikjian L 1987 Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature 328:175–178
Dash PK, Karl KA, Colicos MA, Prywes R, Kandel ER 1991 cAMP response element-binding protein is activated by Ca2+/calmodulin- as well as cAMP-dependent protein kinase. Proc Natl Acad Sci USA 88:5061–5065
He D, Funabashi T, Sano A, Uemura T, Minaguchi H, Kimura F 1999 Effects of glucose and related substrates on the recovery of the electrical activity of gonadotropin-releasing hormone pulse generator which is decreased by insulin-induced hypoglycemia in the estrogen-primed ovariectomized rat. Brain Res 820:71–76
Mitsushima D, Tin-Tin-Win-Shwe, Kimura F 2003 Sexual dimorphism in the GABAergic control of gonadotropin release in intact rats. Neurosci Res 46:399–405
Steiner RA, Bremner WJ, Clifton DK 1982 Regulation of luteinizing hormone pulse frequency and amplitude by testosterone in the adult male rat. Endocrinology 111:2055–2061
Crace CJ, Swenne I, Milner RDG 1989 Long-term follow-up after early protein-calorie malnutrition in young rats: sex difference in glucose tolerance and serum insulin levels. Metabolism 38:933–938
Galipeau DM, Yao L, McNeill JH 2002 Relationship among hyperinsulinemia, insulin resistance, and hypertension is dependent on sex. Am J Phyisol 283:H562–H567
Mayer J 1953 Glucostatic mechanism of regulation of food intake. N Engl J Med 249:13–16
Bahjaoui-Bouhaddi M, Fellmann D, Bugnon C 1994 Induction of Fos-immunoreactivity in prolactin-like containing neurons of the rat lateral hypothalamus after insulin treatment. Neurosci Lett 168:11–15
Sheriff S, Chance WT, Fischer JE, Balasubramaniam A 1997 Neuropeptide Y treatment and food deprivation increase cyclic AMP response element-binding in rat hypothalamus. Mol Pharmacol 51:597–604
Gillard ER, Khan AM, Grewal RS, Mouradi B, Wolfsohn SD, Stanley BG 1998 The second messenger cAMP elicits eating by an anatomically specific action in the perifornical hypothalamus. J Neurosci 18:2646–2652
Yang X–J, Funabashi T, Kow L–M, Mobbs CV 2004 A glucokinase/AP-1 glucose transduction mechanism in the ventromedial hypothalamic satiety center. In: Matschinsky FM, Magnuson MA, eds. Frontiers in diabetes. Vol 16. Basel: Karger; 313–326
Levin BE, Dunn-Meynell AA, Routh VH 1999 Brain glucose sensing and body energy homeostasis: role in obesity and diabetes. Am J Physiol 276:R1223–R1231
Havel PJ 2001 Peripheral signals conveying metabolic information to the brain: short-term and long-term regulation of food intake and energy homeostasis. Exp Biol Med 226:963–977
Orsini JC, Armstrong DL, Wayner MJ 1992 Responses of lateral hypothalamic neurons recorded in vitro to moderate changes in glucose concentration. Brain Res Bull 29:503–505
Niijima A 1986 Neural control of blood glucose level. Jpn J Physiol 36:827–841
Muroya S, Yada T, Shioda S, Takigawa M 1999 Glucose-sensitive neurons in the rat arcuate nucleus contain neuropeptide Y. Neurosci Lett 264:113–116
Kakolewski JW, Cox VC, Valenstein ES 1968 Sex differences in body-weight changes following gonadectomy of rats. Psychol Rep 22:547–554
Rowland DL, Perrings TS, Thommes JA 1980 Comparison of androgenic effects on food intake and body weight in adult rats. Physiol Behav 24:205–209
Chai J–K, Blaha V, Meguid MM, Laviano A, Yang Z–J, Varma M 1999 Use of orchidectomy and testosterone replacement to explore number-to-meal size relationship in male rats. Am J Physiol 276:R1366–R1373
Murray JF, Baker BI, Levy A, Wilson CA 2000 The influence of gonadal steroids on pre-pro melanin-concentrating hormone mRNA in female rats. J Neuroendocrinol 12:53–59
Czaja JA 1984 Sex differences in the activational effects of gonadal hormones on food intake and body weight. Physiol Behav 33:553–558
Sandoval DA, Ertl AC, Richardson A, Tate DB, Davis SN 2003 Estrogen blunts neuroendocrine and metabolic responses to hypoglycemia. Diabetes 52:1749–1755
Adams JM, Legan SJ, Ott CE, Jackson BA 2005 Modulation of hypoglycemia-induced increases in plasma epinephrine by estrogen in the female rat. J Neurosci Res 79:360–367
Shi J, Zhang YQ, Simpkins JW 1997 Effects of 17b-estradiol on glucose transporter 1 expression and endothelial cell survival following focal ischemia in the rats. Exp Brain Res 117:200–206
Luine VN, Khylchevskaya RI, McEwen BS 1975 Effect of gonadal hormones in enzyme activities in brain and pituitary of male and female rats. Brain Res 86:283–292
Hamba H, Sokoloff L 1984 Acute administration of high doses of estrogen increases glucose utilization throughout brain. Brain Res 291:391–394
Wise PM 2003 Estrogens: protective or risk factors in brain function? Prog Neurobiol 69:181–191(Kazutaka Mogi, Toshiya Fu)