Distinct Effects of Leptin and a Melanocortin Receptor Agonist Injected Into Medial Hypothalamic Nuclei on Glucose Uptake in Peripheral Tissues

Male FVB/N Jcl mice (CLEA Japan, Tokyo, Japan) were studied at 12–16 weeks of age. The animals were housed individually in plastic cages at 24 ± 1°C with lights on from 0600 to 1800 h, and they were maintained with free access to a laboratory diet (Oriental Yeast, Tokyo, Japan) and water. Mice were anesthetized by intraperitoneal injection of ketamine (100 mg/kg body mass) and xylazine (10 mg/kg), and a chronic double-walled stainless steel cannula was implanted stereotaxically and unilaterally into the right side of the VMH, ARC, DMH, or PVH or into the lateral ventricle according to the atlas of Franklin and Paxinos (26). The stereotaxic coordinates were AP 1.3 (1.3 mm anterior to the bregma), L 0.3 (0.3 mm lateral to the bregma), and H 5.8 (5.8 mm below the bregma on the surface of the skull) for the VMH; AP 1.6, L 0.2, and H 6.1 for the ARC; AP 1.6, L 0.3, and H 5.5 for the DMH; AP 0.75, L 0.2, and H 4.9 for the PVH; and AP 0.3, L 1.0, and H 2.25 for the lateral ventricle. Cannulas were anchored firmly to the skull. In experiments featuring administration of the MCR antagonist SHU9119, cannulas were implanted both into VMH and intracerebroventricularly.

Three days before determination of tissue glucose uptake, a silicone catheter was implanted into the external jugular vein. Animals were handled repeatedly during the recovery period (2 weeks) after cannula implantation to habituate them to the injection and blood-sampling procedures. Correct placement of the cannula tips was verified microscopically in brain sections in all experiments, with >90% of animals manifesting correct placement. Tyrosine phosphorylation of signal transducer and activator of transcription 3 (STAT3) in each medial hypothalamic nucleus was examined after injection of leptin, as described below. Food was removed just before the administration of the agents. All animal experiments were performed in accordance with institutional guidelines for the care and handling of experimental animals, and they were approved by the ethics committee for animal experiments of the National Institute for Physiological Sciences.

Leptin (5 ng) (National Hormones and Pituitary Program, Torrance, CA) or MT-II (10 ng) (Phoenix Pharmaceuticals, Burlingame, CA) dissolved in 0.1 μl physiological saline was injected with the use of a Hamilton microsyringe into the right side of the VMH, DMH, PVH, or ARC of freely moving mice through the unilateral cannula implanted into the corresponding nucleus. The concentrations of leptin (3 μmol/l) and MT-II (100 μmol/l) dissolved in 0.1 μl physiological saline injected into the medial hypothalamic nuclei are at least 10 times higher than those necessary for the maximum activation of leptin receptor and MCRs (27,–29). Alternatively, MT-II (3 μg) in 0.5 μl saline was injected into the lateral ventricle. SHU9119 (1 μg) (Phoenix Pharmaceuticals) in 0.5 μl saline was also injected intracerebroventricularly immediately before leptin injection. The dose of MT-II and SHU9119 injected intracerebroventricularly altered food intake significantly (30,31). Control animals received 0.1 μl saline delivered into the various nuclei or 0.5 μl saline delivered intracerebroventricularly, respectively.

The rate constant of net tissue uptake of 2-[³H]deoxy-d-glucose (2[³H]DG) in peripheral tissues was determined, as described previously (32,33), by injecting a mixture of 6.25 μCi 2[³H]DG (10 Ci/mmol) and 1.25 μCi [¹⁴C]sucrose (10 Ci/mmol) (American Radiolabeled Chemicals, St. Louis, MO) through the jugular vein catheter 3 or 6 h after the microinjection of leptin or MT-II. Blood was collected 0, 10, 15, and 20 min after injection of the radioactive tracers. Immediately after collection of the final blood sample (20 min), an overdose of pentobarbital sodium (100 mg/kg) was injected through the jugular vein catheter and the mice were rapidly decapitated. Hypothalamic nuclei were then rapidly dissected as described below and frozen in liquid nitrogen for subsequent immunoblot analysis. Skeletal muscle (soleus, red and white portions of the gastrocnemius, and extensor digitorum longus [EDL]) as well as interscapular BAT, heart, spleen, and epididymal white adipose tissue (WAT) were rapidly dissected, weighed, and assayed for the radioactivity. Plasma samples were also analyzed for glucose (Glucose CII Test; Wako, Osaka, Japan) and insulin (mouse insulin ELISA kit [U-Type]; Shibayagi, Gunma, Japan) concentrations. The rate constant of net tissue uptake of 2[³H]DG was calculated as described previously (32,33). The radioactivity of [¹⁴C]sucrose in tissues was used to calculate the 2[³H]DG radioactivity remaining in the extracellular space (33).

Medial hypothalamic nuclei were isolated as described previously (34). The accuracy of the dissection was assessed by measurement of mRNAs for neuropeptides or transcription factors such as corticotropin-releasing factor (CRF) mRNA for the PVH, POMC and NPY mRNAs for the ARC, SF1 mRNA for the VMH, and the absence of these various mRNAs for the DMH. The right side of the PVH, ARC, VMH, or DMH was dissected from a 1-mm-thick sagittal section prepared from the midline of the fresh brain (online appendix supplemental Fig. 1 [available at http://diabetes.diabetesjournals.org/cgi/content/full/db09-0638/DC1]).

Hypothalamic nuclei were homogenized at 4°C in PBS containing 1% Nonidet P-40. The homogenates were centrifuged, and the resulting supernatants (5 μg of protein) were fractionated by SDS-PAGE. Immunoblot analysis was then performed with antibodies (1 μg/ml), including those to the Tyr⁷⁰⁵-phosphorylated or total forms of STAT3 (Cell Signaling Technology, Danvers, MA), to c-Fos (Santa Cruz Biotechnology, Santa Cruz, CA), or to β-actin (Cell Signaling Technology). Immune complexes were visualized with horseradish peroxidase–conjugated secondary antibodies (Santa Cruz Biotechnology) and enhanced chemiluminescence reagents (GE Healthcare, Tokyo, Japan). Protein bands were quantified using Image J software (National Institutes of Health, http://rsbweb.nih.gov/ij).

Total RNA was isolated from hypothalamic and peripheral tissue with the use of Isogen (Nippon Gene, Wako, Japan), and portions of the RNA (300 ng) were subjected to reverse transcription with an oligo(dT) primer and avian myeloblastosis virus reverse transcriptase (Takara, Shiga, Japan). The resulting cDNA was subjected to the PCR with LA Taq (Takara) and primers obtained from Sigma Genosys (Ishikari, Japan). For quantitative real-time PCR, cDNA was amplified using SYBR Green PCR Master Mix with an ABI 7500 real-time PCR system (Applied Biosystems, Tokyo, Japan). Data were normalized by the amount of eukaryotic elongation factor 2 (eEF2) mRNA. The sequences of the primers are shown in supplemental Table 1.

Data are presented as means ± SE. Statistical analysis of STAT3 phosphorylation, c-Fos expression, and real-time PCR was performed by Student t test, and analysis for other experiments was performed by ANOVA followed by Dunnett test. A P value of <0.05 was considered statistically significant.

Microinjection of leptin (5 ng) into the VMH induced a significant increase in the rate constant of 2[³H]DG uptake in the red type of skeletal muscle (soleus and red portion of gastrocnemius), mixed type of skeletal muscle (EDL), BAT, and heart but not in spleen or epididymal WAT (Fig. 1). Glucose uptake in red or mixed skeletal muscle was significantly increased at 6 h after leptin injection (Fig. 1,A), whereas that in BAT and heart was increased at both 3 and 6 h (Fig. 1,B). Glucose uptake in the white portion of the gastrocnemius showed a tendency to increase in response to leptin, but the change was not statistically significant. Glucose uptake in peripheral tissues at 3 h after saline injection into the VMH did not differ from that apparent at 6 h (data not shown). Injection of leptin into the VMH significantly increased GLUT4 mRNA in BAT and heart but not in soleus muscle at 6 h after leptin injection (Fig. 1,E). It also increased uncoupling protein 1 (UCP1) mRNA in BAT. Hexokinase II mRNA did not change in soleus, BAT, or heart (Fig. 1 E).

Injection of maximal dose of leptin into the ARC induced a small but significant increase in the rate constant of 2[³H]DG uptake in BAT at 6 h after injection, but it had no effect on glucose uptake in skeletal muscle, heart, spleen, or WAT (Fig. 2). Injection of leptin into the DMH or PVH had no effect on glucose uptake in peripheral tissues (Fig. 2). Glucose uptake in peripheral tissues after saline injection into the DMH or PVH did not differ from that apparent after saline injection into the ARC (data not shown). Plasma glucose and insulin concentrations were not affected by leptin injection into the VMH or other hypothalamic nuclei (supplemental Table 2), consistent with previous observations (13,15).

To determine whether leptin activated Ob-Rb in the hypothalamic nuclei, we examined the tyrosine phosphorylation of STAT3 in tissue samples therefrom. RT-PCR analysis confirmed that the isolated PVH, ARC, and VMH specimens were enriched in CRF mRNA, POMC and NPY mRNAs, and SF1 mRNA, respectively, and that the DMH was largely devoid of these mRNAs (Fig. 3,A). The medial hypothalamic nuclei were thus accurately dissected. Microinjection of leptin into the VMH, DMH, or PVH preferentially increased the tyrosine phosphorylation of STAT3 to the similar extent in the corresponding nucleus at 6 h after injection (Fig. 3,B–D). These results confirmed that leptin was correctly injected into these various hypothalamic nuclei. Injection of leptin into the ARC increased the tyrosine phosphorylation of STAT3 in the DMH as well as in the ARC (Fig. 3 E), possibly as a result of leakage of leptin into the DMH through the surface of the injection cannula, given that the ARC cannula passed through the DMH. Together, these results suggested that the VMH is a key target of leptin in its regulation of glucose uptake in skeletal muscle, heart, and BAT, whereas the leptin receptor in the ARC mediates stimulation of glucose uptake in BAT.

We next examined whether leptin injection into the VMH might increase neuronal activity in other hypothalamic nuclei by measuring expression of the transcription factor c-Fos. Leptin injection into the VMH significantly increased c-Fos expression in the ARC as well as in the VMH at 6 h after injection (Fig. 3 F). Leptin injection into the VMH increased the phosphorylation of STAT3 in the nucleus preferentially at 1 h after the injection, similar to that at 6 h. However, it did not increase c-Fos expression in any nucleus at 1 h (supplemental Fig. 2), as described previously (35). Leptin injection into the VMH did not increase POMC mRNA in the ARC at 6 h after injection (supplemental Fig. 2).

We next examined the role of MCRs in glucose uptake in peripheral tissues induced by injection of leptin into the VMH (Fig. 4). Injection of the MCR antagonist SHU9119 (1 μg) into the lateral ventricle (intracerebroventricularly) abolished the increase in 2[³H]DG uptake in peripheral tissues normally apparent at 6 h after the injection of leptin into the VMH. Injection of SHU9119 alone did not affect glucose uptake in peripheral tissues. Plasma glucose and insulin levels were also not changed in response to intracerebroventricular injection of SHU9119 (supplemental Table 2).

We tested the effects of intracerebroventricular injection of the MCR agonist MT-II on glucose uptake in peripheral tissues (Fig. 5). The intracerebroventricular injection of MT-II (3 μg) increased 2[³H]DG uptake in BAT, heart, and all types of skeletal muscle, including the white portion of the gastrocnemius, but not in spleen or epididymal WAT, at 3 or 6 h after injection. Glucose uptake in peripheral tissues at 3 h after saline injection did not differ from that apparent at 6 h (data not shown). The intracerebroventricular injection of MT-II increased the plasma glucose level at both 3 and 6 h after injection (supplemental Table 2). These results thus suggested that intracerebroventricular injection of MT-II promotes glucose production as well as glucose uptake in certain peripheral tissues. In contrast, plasma insulin concentration did not change after intracerebroventricular injection of MT-II, despite the associated hyperglycemia, suggesting that MT-II inhibits insulin secretion from pancreatic β-cells.

Finally, we examined the effects of direct injection of MT-II into the individual medial hypothalamic nuclei on glucose uptake in peripheral tissues (Fig. 6). Microinjection of MT-II (10 ng) into the VMH increased 2[³H]DG uptake in BAT, heart, and all types of skeletal muscle but not in spleen or epididymal WAT. In contrast, injection of MT-II into the PVH increased glucose uptake only in BAT, and that into the DMH or ARC did not affect glucose uptake in any of the peripheral tissues examined. Plasma glucose and insulin levels did not change in response to injection of MT-II into any of the hypothalamic nuclei (supplemental Table 2).

Leptin is a physiologically and clinically important hormone that regulates glucose metabolism in peripheral tissues. We have previously shown that microinjection of leptin into the VMH and nearby medial hypothalamic area preferentially increased glucose uptake in skeletal muscle, heart, and BAT (14,–16). In the present study, we found that activation of the leptin receptor specifically in the VMH, as detected by measurement of the tyrosine phosphorylation of STAT3, resulted in a marked increase in glucose uptake in those peripheral tissues, similar to the effects of intracerebroventricular or peripheral administration of leptin (13,36). In contrast, injection of leptin into the ARC increased glucose uptake only in BAT. Injection of leptin into the DMH or PVH had no effect on glucose uptake in any of the peripheral tissues examined. The present data thus suggest that the leptin receptor in the VMH and ARC regulates glucose uptake in different peripheral tissues.

Our present results further indicate that MCR activation is necessary for the increase in glucose uptake in peripheral tissues induced by injection of leptin into the VMH. The intracerebroventricular injection of SHU9119 thus abolished the effect of leptin injected into the VMH on peripheral glucose uptake, whereas intracerebroventricular injection of MT-II increased glucose uptake in peripheral tissues. Moreover, injection of MT-II into VMH, but not that into the DMH or ARC, increased glucose uptake in skeletal muscle, heart, and BAT, whereas injection of MT-II into the PVH increased glucose uptake preferentially in BAT. The VMH expresses abundantly MC3R as well as MC4R (23,25). Both MCRs in the VMH and PVH may play an important role in the regulation of glucose uptake in peripheral tissues.

POMC neurons in the ARC receive strong excitatory input from the dorsomedial region of the VMH (37). The restoration of Ob-Rb expression in the VMH by adeno-associated virus–mediated gene transfer in Koletsky rats increased the amount of POMC mRNA in the ARC (20). We have now shown that injection of leptin into the VMH increased c-Fos expression in the ARC without an effect on STAT3 phosphorylation in the ARC. Furthermore, the effect of leptin injected into the VMH on glucose uptake in peripheral tissues was found to be dependent on MCRs in the brain. We therefore propose that stimulation of VMH neurons by leptin results in activation of a set of POMC neurons in the ARC and thereby increases glucose uptake in skeletal muscle, heart, and BAT. MCRs in the VMH and PVH contribute to the upregulation of glucose uptake in peripheral tissues induced by injection of leptin into the VMH. Although c-Fos expression did not increase in the PVH in response to leptin injection into the VMH, this may have been due to the operation of γ-aminobutyric acid (GABA)-mediated neurotransmission in the PVH (6). The mechanism by which injection of leptin into the ARC increased glucose uptake specifically in BAT remains unclear. One possible explanation for this observation is that the injection of leptin into the ARC activated a selective set of POMC neurons in the ARC that regulate glucose uptake in BAT alone, with POMC neurons being abundant in both the anterior and posterior regions of the ARC (38). It is also possible that POMC neurons in the ARC that regulate glucose uptake in skeletal muscle and heart require excitatory input as well as leptin for their full activation.

The intracerebroventricular injection of MT-II increased glucose uptake in BAT to a markedly greater extent than did the maximum dose of injection of MT-II into the VMH or PVH. MC4R-expressing neurons in several regions of the brain stem as well as in the hypothalamic nuclei connect polysynaptically with interscapular BAT (39). Direct injection of MT-II into the raphe pallidus induces a thermogenic response in interscapular BAT (40). Glucose uptake in BAT might thus be regulated by MCRs in multiple brain regions. To date, there is little histological evidence of a connection between VMH neurons and interscapular BAT, although injection of leptin into the VMH increases plasma catecholamine levels more effectively than does that into other hypothalamic regions (41). We previously showed that injection of leptin into the medial hypothalamus increased glucose uptake in peripheral tissues through the activation of sympathetic nerves (14,15). Injection of leptin into the medial hypothalamus also increased insulin sensitivity in peripheral tissues by a β-adrenergic mechanism (14). Whereas the molecular mechanism remains elusive, a direct effect of β-adrenergic receptors expressed in peripheral tissues as well as a β-adrenergic receptor–dependent increase in blood flow appear to contribute to leptin-induced glucose uptake in these tissues.

The DMH is an important hypothalamic nucleus in the regulation of thermogenesis in BAT (42). Injection of a GABA type_A (GABA_A) receptor antagonist into the DMH thus increased thermogenic activity in BAT (42). Moreover, injection of a GABA_A receptor agonist into the DMH blocked sympathetic, thermogenic, and cardiovascular responses induced either by injection of prostaglandin E₂ into the medial preoptic area (43) or by skin cooling (44). Both Ob-Rb and MC4R are abundant in the DMH (7). Furthermore, DMH neurons, including those expressing MC4R, connect polysynaptically with interscapular BAT (39). However, we have now shown that injection of leptin or MT-II into the DMH did not increase glucose uptake in the peripheral tissues examined. Although we cannot exclude the possibility that injection of leptin or MT-II activated only a subset of DMH neurons expressing Ob-Rb or MCR, our results suggest that the leptin receptor and MCR in the DMH have other roles, such as the regulation of food intake or modulation of BAT thermogenesis in response to cold stimuli.

Skeletal muscle is a central player in glucose homeostasis in mammals. Our present results indicate that the VMH regulates glucose uptake in skeletal muscle as well as in heart and BAT, accompanying increased GLUT4 mRNA in BAT and heart. Thus, increased expression of GLUT4 may involve the acute increase in glucose uptake in BAT and heart in response to leptin, while other mechanism might be involved in the leptin-induced glucose uptake in skeletal muscle. Intracerebroventricular injection of MT-II has been shown to increase GLUT4 mRNA expression in skeletal muscle at 24 h after the injection (22). Gene expression in skeletal muscle in response to leptin and MT-II may require >24 h.

The present results showed that injection of leptin or MT-II in some medial hypothalamic nuclei increased glucose uptake in certain peripheral tissues, while it did not alter plasma glucose level. These results suggest that leptin and MT-II stimulates hepatic glucose production as well as glucose utilization in peripheral tissues. The result was supported by the previous report (13) showing that leptin increased plasma glucose turnover in mice at 6 h after the injection. Moreover, Gutierrez-Juarez et al. (45) has shown that intracerebroventricular infusion of MCR agonist increased hepatic glucose production in lean rats at 6 h after the start of the infusion, while it inhibited hepatic glucose production at 7 days (46). These results suggest that the effects of leptin and melanocortin receptor agonist on hepatic glucose production alter time dependently.

Recent study suggests that MT-II seems to be a promising agent to bypass leptin resistance and to improve energy homeostasis in diet-induced obese mice (47). We have shown that intracerebroventricular injection of MT-II but not leptin activates AMP-activated protein kinase in skeletal muscle in diet-induced obese mice (48). The PVH and VMH may be target sites of MT-II to improve glucose metabolism in these mice.

Together, our results suggest that the VMH plays a key role in leptin- and MT-II–induced glucose uptake in skeletal muscle, heart, and BAT, whereas the leptin receptor in the ARC and MCRs in the PVH regulate glucose uptake in BAT. The medial hypothalamic nuclei thus appear to play distinct roles in the regulation of glucose uptake in peripheral tissues by leptin and MT-II.

This study was supported by grants-in-aid for scientific research (B) (16390062 and 19390059, to Y.M.) and a grant-in-aid for exploratory research (17659069, to Y.M.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; by the Specific Research Fund of the National Institutes for Natural Sciences (to Y.M.); and by the Research Program of Hayama Center for Advanced Studies of Sokendai (to Y.M.).

No potential conflicts of interest relevant to this article were reported.

We thank Kumiko Saito and Nahomi Kawai for helpful assistance, the Center for Analytical Instruments at the National Institute for Basic Biology (NIBB) (Okazaki) for real-time PCR, and the Center for Radioisotope Facilities at the NIBB for measurement of in vivo glucose uptake.