Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2010 Dec 15;108(1):355–360. doi: 10.1073/pnas.1016785108

Multinodal regulation of the arcuate/paraventricular nucleus circuit by leptin

Masoud Ghamari-Langroudi a,1, Dollada Srisai b, Roger D Cone a,1
PMCID: PMC3017160  PMID: 21169216

Abstract

Melanocortin-4 receptor (MC4R) is critical for energy homeostasis, and the paraventricular nucleus of the hypothalamus (PVN) is a key site of MC4R action. Most studies suggest that leptin regulates PVN neurons indirectly, by binding to receptors in the arcuate nucleus or ventromedial hypothalamus and regulating release of products like α-melanocyte-stimulating hormone (α-MSH), neuropeptide Y (NPY), glutamate, and GABA from first-order neurons onto the MC4R PVN cells. Here, we investigate mechanisms underlying regulation of activity of these neurons under various metabolic states by using hypothalamic slices from a transgenic MC4R-GFP mouse to record directly from MC4R neurons. First, we show that in vivo leptin levels regulate the tonic firing rate of second-order MC4R PVN neurons, with fasting increasing firing frequency in a leptin-dependent manner. We also show that, although leptin inhibits these neurons directly at the postsynaptic membrane, α-MSH and NPY potently stimulate and inhibit the cells, respectively. Thus, in contrast with the conventional model of leptin action, the primary control of MC4R PVN neurons is unlikely to be mediated by leptin action on arcuate NPY/agouti-related protein and proopiomelanocortin neurons. We also show that the activity of MC4R PVN neurons is controlled by the constitutive activity of the MC4R and that expression of the receptor mRNA and α-MSH sensitivity are both stimulated by leptin. Thus, leptin acts multinodally on arcuate nucleus/PVN circuits to regulate energy homeostasis, with prominent mechanisms involving direct control of both membrane conductances and gene expression in the MC4R PVN neuron.

Keywords: melanocortin signaling, electrophysiology, obesity

Neurons expressing melanocortin-4 receptor (MC4R) in the paraventricular nucleus of the hypothalamus (PVN) play a crucial role in energy homeostasis. The genetic and pharmacological disruption of MC4R increases energy intake and decreases thermogenesis (1, 2). Thus, these neurons sense peripheral signals of adiposity and maintain energy homeostasis by coordinating energy intake and expenditure (2, 3). The adipocyte hormone leptin relays information on changes in peripheral energy stores to melanocortin and other circuits in the brain to regulate energy homeostasis (4). Neurons in the arcuate nucleus of the hypothalamus (ARC) and other nuclei, including the ventromedial nucleus (VMH), dorsomedial nucleus, and lateral nucleus, express leptin receptors and play an important role in transmitting the leptin signal to PVN neurons (5–7). Despite functional evidence suggesting expression of the leptin receptor in PVN (8), direct action of leptin on PVN neurons has not been thoroughly investigated, perhaps because of the low density of leptin-receptor expression (9, 10).

Circulating leptin, by affecting the activity of neuropeptide Y (NPY) and proopiomelanocortin (POMC) neurons in the ARC, regulates synthesis of NPY/agouti-related protein (AgRP) and α-melanocyte-stimulating hormone (α-MSH) as well as their putative release from nerve endings onto neurons in PVN (11, 12). The synthesis and release of inhibitory products of NPY/AgRP neurons—NPY, AgRP, and GABA—are suppressed by leptin (13), and the products of ARC POMC neurons then maintain an excitatory effect on PVN neurons (14). Removal of that excitatory effect pharmacologically, or by destruction of ARC neurons (15, 16), presumably decreases the activity of PVN neurons, resulting in increases in energy intake. Despite advances in our understanding of the regulation of ARC POMC and NPY/AgRP neurons, mechanisms involved in regulation of downstream effector PVN neurons are less well understood. Recent studies have shown that deletion of leptin receptors on ARC POMC neurons only partially reduces the extreme obesity seen with global deletion of leptin-receptor signaling (17), suggesting involvement of other hypothalamic nuclei in relaying the peripheral signals of adiposity. In fact, deletion of leptin receptors in VMH neurons also increases body weight (18), further underscoring a role of leptin-receptor signaling in “non-ARC” neuronal centers. These studies thus suggest involvement of multiple neuronal centers in sensing and relaying information of energy status to the CNS (19). In this study, we investigate mechanisms involved in regulation of downstream effector PVN neurons in vivo by analyzing how metabolic state, leptin, and products of ARC neurons regulate the activity of MC4R effector neurons in PVN.

Results

MC4R PVN Neurons Are Tonically Regulated by Metabolic State.

We sought to understand how the activity of MC4R PVN effector neurons is regulated in different metabolic states by recording from cells expressing GFP under the control of the MC4R promoter (20), primarily in the mid- to posterior region of the mouse PVN. We first measured action-potential firing frequency of these neurons in hypothalamic slices obtained from mice either fed ad libitum or fasted (16 h). Using loose-patch recordings, we obtained results from 23 mice that indicate that fasting significantly increased mean frequency of action-potential firing of these neurons compared with the ad libitum–fed state (2.7 ± 0.2 Hz in fasted mice, n = 134, 1.9 ± 0.2 Hz in fed mice, n = 116, P < 0.005, unpaired t test, Fig. 1A).

Fig. 1.

Fig. 1.

Open in a new tab

Leptin tonically inhibits firing of MC4R PVN neurons. (A) Average ± SEM of action potential, firing frequency, and frequency distribution (i) of MC4R PVN neurons obtained from mice subjected to 16 h of fasting (n = 134) or fed ad libitum (n = 116, *P < 0.005). (B) Average ± SEM of firing frequency and frequency distribution (ii) of these neurons from 16-h fasted mice that were injected i.p. 3 h before decapitation with 3 μg/g leptin (n = 100) or saline (n = 99, *P < 0.005). (C) Average ± SEM of frequency of firing of MC4R PVN neurons obtained from ob/ob mice that were subjected to 16 h of fasting (n = 105), fed ad libitum (n = 132, P > 0.5), or injected i.p. with leptin 3 h before decapitation after 16 h of fasting (n = 100, *P < 0.0001).

We next tested whether leptin mediated the fasting-induced increase in activity of MC4R PVN neurons by comparing the firing activity of neurons from fasted mice administered 3 mg/kg leptin i.p. or the same volume of saline (12). Using loose-patch recording to sample firing frequency in MC4R PVN neurons from 18 mice, we found that leptin suppressed the activity of MC4R PVN neurons to ad libitum–fed levels (2.4 ± 0.2 Hz in saline-injected mice, n = 100, to 1.6 ± 0.2 Hz in leptin-injected mice, n = 99, P < 0.005, unpaired t test, Fig. 1B). Frequency analysis indicates that fasting activates a subset of MC4R neurons firing at frequencies between 2.5 and 5.5 Hz, and leptin replacement reversed the fasting-induced increase in this subset of neurons (Fig. 1 Ai and Bii). To confirm that the fasting-induced activation of MC4R PVN neurons was mediated by endogenous leptin, we examined the firing activity of these neurons in leptin-deficient ob/ob mice. We generated double-mutant MC4R-GFP ob/ob mice and then compared firing activity of MC4R PVN neurons in these animals, fasted or fed ad libitum. The mean firing frequency of neurons obtained from fed MC4R-GFP × ob/ob mice was not significantly different from that of 16-h fasted MC4R-GFP × ob/ob mice (3.9 ± 0.2 Hz in fed, n = 132; 3.8 ± 0.2 Hz in fasted mice, n = 105; unpaired t test, P > 0.5). The firing frequency of MC4R PVN neurons significantly decreased when fasted MC4R-GFP × ob/ob mice were injected with leptin (3.8 ± 0.2 Hz in fasted mice, n = 105, to 2.1 ± 0.2 Hz in fasted, leptin-injected mice, n = 100, P < 0.0001, unpaired t test, Fig. 1C).

α-MSH Augments and AgRP Decreases Firing Activity of MC4R PVN Neurons Through MC4R Signaling.

Leptin has been shown to affect neuronal activity of NPY/AgRP and POMC neurons in ARC and to regulate the release of their products (5, 11, 12, 21). To study the actions of these ARC peptides on MC4R PVN neurons, we first compared average firing frequencies measured over 4–6 min in individual neurons in control conditions with frequencies observed in the presence of the melanocortin agonist MTII. Results obtained from 11 neurons (Fig. S1) indicate that bath application of 100 nM MTII significantly increased the firing rate in all neurons tested (1.4 ± 0.2 Hz in control conditions to 3.2 ± 0.5, n = 11 neurons from 11 mice, P < 0.001), in agreement with other reports of effects of MTII on PVN or ARC neurons (20, 22). Furthermore, this effect of MTII was abolished when neurons were pretreated with an MC4R antagonist, 100 nM SHU9119 (23) (1.7 ± 0.9 Hz in SHU9119 alone to 1.8 ± 0.7 Hz in SHU9119 + MTII, P > 0.05, n = 7, Fig. S1).

We then investigated the effects on firing rate of the endogenous agonist, α-MSH, and the endogenous inverse agonist, AgRP, of MC4R signaling. Using loose-patch recordings, we tested the effects of 250 nM α-MSH (24) on MC4R PVN neurons. In 13 neurons tested, bath applications of α-MSH significantly and reversibly increased by 78% the frequency of action-potential firing (from 1.7 ± 0.4 Hz in control conditions to 3.1 ± 0.5 Hz with α-MSH treatment, n = 13, P < 0.001, Fig. 2 A and B). Measured ∼30–40 min after wash out, the firing frequency of MC4R PVN neurons was significantly reduced to values different from those obtained in the presence of α-MSH (2.1 ± 0.6, n = 13, P < 0.01). Using loose patch recording in seven MC4R PVN neurons, bath application of 100 nM AgRP significantly decreased (to 53% of control) the frequency of action-potential firing (from 2.7 ± 0.4 Hz in control to 1.4 ± 0.5 Hz with AgRP treatment, n = 7, P < 0.01, Fig. 2 C and D).

Fig. 2.

Fig. 2.

Open in a new tab

α-MSH augments and AgRP and NPY decrease firing activity of MC4R PVN neurons. (A) Bath application of 250 nM α-MSH augments firing frequency of MC4R PVN neurons recorded using the loose-patch technique. (B) Average ± SEM of effect of 250 nM concentration of this peptide obtained from 13 MC4R PVN neurons (*P < 0.001). (C) Bath application of 100 nM AgRP significantly inhibits firing frequency of PVN neurons obtained by loose-patch recordings. (D) Average ± SEM of effect of 100 nM AgRP from seven PVN neurons (*P < 0.01). (E) A whole-cell recording from a spontaneously firing PVN neuron indicates that application of 100 nM NPY generates a significant inhibition of firing activity associated with hyperpolarization of membrane potential. (F and G) Average ± SEM of effect of 100 nM NPY on firing frequency (*P < 0.05) and membrane potentials (*P < 0.001) of eight MC4R PVN neurons.

We next investigated whether the action of α-MSH is associated with changes in membrane potential of PVN neurons by using whole-cell recordings. The membrane potential of PVN neurons was held near threshold for action potentials (25) between −55 and −50 mV by injecting −20–0 pA of constant DC current, allowing cells to fire action potentials spontaneously. The firing frequency and membrane potential of these neurons were then monitored for 20–30 min under control conditions, before bath application of 250 nM α-MSH for 7–12 min, followed by a longer period of wash out (>20 min). Firing frequency and membrane potential were compared for a period of 4–6 min in control and for an identical period during α-MSH addition. Our results obtained from 24 MC4R PVN neurons tested under these conditions indicate that bath application of 250 nM α-MSH significantly increased firing frequency in 21 of 24 neurons (from 0.9 ± 0.2 Hz to 3.0 ± 0.6 Hz, n = 21, P < 0.0001, Fig. S2 A–D). Furthermore, our results indicate that 23 of 24 PVN cells examined were depolarized by α-MSH. When all neurons were included, α-MSH induced significant depolarization of membrane potential (from −54.2 ± 1.1 mV to −46.3 ± 0.9 mV, n = 24, P < 0.0001, Fig. S2 A–C and E).

We also examined whether effects of α-MSH persisted in the absence of GABA(A) and ionotropic glutamate neurotransmission recorded from hypothalamic slices pretreated with 200 μM picrotoxin (PTX) and 1 mM kynurenic acid (KYN) (26). Application of 250 nM α-MSH significantly increased firing frequency in six MC4R PVN neurons under these conditions (from 1.2 ± 0.1 Hz to 2.7 ± 0.6 Hz, P < 0.05, n = 6, Fig. S2 F and G). This increase in firing activity was associated with a significant depolarization of membrane potentials (from −55.4 ± 1.1 mV to −49.2 ± 2.1 mV, P < 0.005, n = 9, Fig. S2H). These findings are consistent with α-MSH acting independently of ionotropic GABAergic or glutamatergic neurotransmission to depolarize MC4R PVN neurons. However, they do not exclude a role of other neurotransmitters in mediating the observed effects of α-MSH (27). We hence tested effects of bath applications of α-MSH on membrane potential of PVN neurons, as described above, in the absence of all action potential–dependent synaptic transmission by recording from hypothalamic slices pretreated with 0.5 μM tetrodotoxin. Application of α-MSH significantly depolarized membrane potential of PVN neurons pretreated with tetrodotoxin (from −56.8 ± 1.1 mV to −50.7 ± 2.1 mV, n = 6, P < 0.01, Fig. S2H).

NPY Inhibits Firing in MC4R PVN Neurons.

Because NPY neurons in ARC project to MC4R PVN neurons, where Y receptors have been found in abundance (28), we also examined effects of NPY. Using whole-cell recordings, we examined effects of this peptide on firing frequency as well as membrane potentials of MC4R PVN neurons by comparing these parameters in control solution and in the presence of 100 nM NPY. Bath application of NPY consistently and reversibly decreased firing frequency (from 3.9 ± 1.5 Hz to 0.5 ± 0.2 Hz, n = 8, P < 0.05, Fig. 2 E and F). Furthermore, the decrease in firing frequency was associated with significant hyperpolarization of membrane potentials (from −46.6 ± 2.0 mV to −56.0 ± 2.1 mV, n = 8, P < 0.001, Fig. 2 E and G). Moreover, we tested the effects of NPY on activity of MC4R PVN neurons in external solutions with low concentrations of Ca+2 relative to Mg+2 (Ca+2/Mg+2 = 0.2) in which the probability of synaptic release is significantly diminished. In loose-patch recordings under these conditions, bath application of NPY reversibly inhibited firing rates in PVN neurons (from 8.1 ± 1.2 Hz to 3.98 ± 0.93 Hz, n = 12, P < 0.0001, Fig. S3), suggesting that the observed effects of NPY are mediated through postsynaptic mechanisms.

Effects of Leptin on MC4R PVN Neurons.

Because fasting is known to be associated with increases and decreases in the release of NPY and α-MSH, respectively, from ARC, the increases in the firing frequency of MC4R PVN neurons observed in fasted mice cannot be explained by virtue of ARC neuropeptide inputs only (Fig. 1A). We thus tested whether MC4R PVN neurons can directly respond to leptin. Effects of bath applications of 35–50 nM leptin on firing activity of MC4R PVN neurons were examined with loose-patch and whole cell recordings (Fig. 3). Leptin inhibited firing activity of 14 neurons and increased firing activity in 2 other neurons. Interestingly, both neurons that were excited by leptin application were located in the anterior PVN (−0.58 to −0.70 mm from bregma), whereas all neurons located in the mid- to posterior part of the PVN (−0.70 to −1.20 mm from bregma; ref. 29) were inhibited by leptin. In work published elsewhere, we have demonstrated that >90% of anterior thyrotropin-releasing hormone (TRH)-positive PVN neurons are activated by both leptin and α-MSH (30).

Fig. 3.

Fig. 3.

Open in a new tab

In vitro effects of leptin on firing activity of MC4R neurons in the mid- to posterior and anterior PVN. (A) Frequency histogram of effects of bath application of 50 nM leptin on firing activity of a MC4R PVN neuron recorded by loose-patch technique. (B) Average ± SEM of this effect obtained from 14 neurons recorded from mid- to posterior PVN (n = 14, *P < 0.0001). (C) A whole-cell recording indicates effects of 50 nM leptin on firing activity and membrane potentials of a spontaneously firing MC4R PVN neuron. (D and E) Average ± SEM of effects of bath application of 35–50 nM leptin on firing frequency (*P < 0.0001) and membrane potentials (*P < 0.0005) of 17 mid- to posterior MC4R PVN neurons. (F) A whole-cell recording indicates effects of 50 nM leptin on firing activity and membrane potential of a MC4R PVN neuron pretreated with 200 μM PTX and 1 mM KYN. (G and H) Average ± SEM of effects of 35–50 nM bath applied leptin on firing frequency (*P < 0.001) and membrane potential (*P < 0.05) of nine mid- to posterior MC4R PVN neurons pretreated with 200 μM PTX and 1 mM KYN. (I–L) Leptin activates action-potential firing activity of MC4R neurons in anterior PVN. (I) A whole-cell recording from a spontaneously firing MC4R neuron indicates that bath application of 35 nM leptin induces an increase in firing activity associated with depolarization of membrane potential. (J) The frequency histogram of this effect. (K and L) The bar graphs indicate average ± SEM of effects of 35–50 nM leptin on action-potential firing frequency (K) and membrane potential (L) of seven neurons tested (in both K and L, *P < 0.005).

In contrast to the response in MC4R neurons of the anterior PVN, MC4R neurons of the mid- to posterior PVN responded uniformly to leptin application with an inhibition of action-potential firing (from 2.2 ± 0.3 Hz to 1.6 ± 0.3 Hz, n = 14, P < 0.0001, Fig. 3 A and B). Because most (80–93%) of the MC4R neurons recorded from various metabolic states (Fig. 1) were located from mid- to posterior PVN, these neurons were binned separately for further analyses. In a separate set of experiments using whole cell recording, leptin decreased firing activity of the mid- to posterior PVN neurons (from 2.4 ± 0.2 Hz to 1.1 ± 0.2 Hz, P < 0.0001, n = 17). This decrease in firing activity was associated with hyperpolarization of membrane potential (from −46.8 ± 2.0 mV to −55.8 ± 2.0 mV, n = 17, P < 0.0005, Fig. 3 C–E). Next, we tested whether the observed inhibitory effects of leptin result from postsynaptic or presynaptic actions by examining the effect of leptin on posterior PVN MC4R neurons while blocking ionotropic glutamate and GABA(A) receptors with PTX (200 μM) and KYN (1 mM) in whole-cell recording. Application of 35–50 nM leptin under these conditions resulted in a decrease in spontaneous firing frequency (from 2.5 ± 0.2 Hz to 1.1 ± 0.19 Hz, P < 0.001, n = 9), which was associated with hyperpolarization of membrane potentials (from −46.5 ± 1.7 mV to −55.9 mV, P < 0.05, n = 9, Fig. 3 F–H). We then repeated this experiment in some midposterior PVN MC4R neurons in the presence of tetrodotoxin to block all activity-dependent neurotransmitter release. Again, leptin caused a hyperpolarization under these conditions (from −47 ± 1.2 to −57 ± 1.4 mV, P < 0.05, n = 5, Fig. 3H). Thus, leptin hyperpolarized and inhibited spontaneous firing in mid- to posterior PVN MC4R neurons by a postsynaptic mechanism.

However, in MC4R neurons from anterior PVN, leptin consistently increased the spontaneous firing rate of these neurons (from 1.2 ± 0.3 Hz to 2.8 ± 0.5 Hz, n = 7, P < 0.005). This increase in firing frequency was associated with depolarization of membrane potential (from −47.0 ± 1.0 mV to −41.2 ± 1.2 mV, n = 7, P < 0.005, Fig. 3 I–L). These data further indicate that the leptin-induced responses of neurons in the anterior PVN differ from those in mid- to posterior part of the nucleus, consistent with in our study of TRH-expressing PVN neurons (30). We tested the hypothesis that the discrepancy in neuronal response to leptin and α-MSH in midposterior MC4R PVN neurons may be related to their functional role by investigating their neuroanatomical and neurochemical properties. We thus sought to identify MC4R PVN neurons involved in regulation of autonomic function by examining their ability to take up the retrograde dye, fluorogold, injected unilaterally into the medial nucleus tractus solitarius and the T2 level of the medial spinal cord of MC4R-GFP mice. Similarly, we used systemic administration of fluorogold to label median eminence–projecting neurons (31). In both cases, the neurochemical identity of PVN MC4 neurons as positive for TRH, corticotropin-releasing factor (CRF), or oxytocin/vasopressin was also determined. These data suggest that the MC4R neurons in the mid- to posterior PVN characterized in this study are primarily brainstem-projecting neurons that coexpress oxytocin/vasopressin and CRF, whereas MC4R neurons in the anterior PVN are primarily median eminence–projecting neurons that coexpress TRH and CRF (see SI Text, Fig. S4, and Table S1).

Circulating Leptin Increases Responsiveness of MC4R PVN Neurons to α-MSH.

Previous studies have shown that leptin increases responsiveness to intracerebroventricular administration of MC4R agonists, as measured by reduction of food intake (32, 33). Furthermore, leptin can modulate the density of NPY-mediated current responses in ARC neurons (34). We therefore investigated whether levels of leptin in vivo can affect responsiveness of MC4R-expressing neurons to α-MSH. We thus compared the magnitude of responses of MC4R PVN neurons to applications of α-MSH in hypothalamic slices obtained from 16-h fasted mice injected i.p. with either leptin or saline 3 h before decapitation. Repeating the protocol in Fig. 1B, basal firing activity of PVN neurons from fasted animals was significantly lower when mice were injected with leptin compared with saline. Furthermore, bath application of α-MSH induced excitatory responses in these neurons that were significantly different in posterior PVN MC4R neurons from fasted mice injected with leptin compared with those treated with saline (one-way ANOVA, P < 0.0005, Fig. 4A). Furthermore, when we compared the magnitude of the α-MSH–induced excitatory responses, neurons that were pretreated with leptin displayed a significantly greater response (∼5.6-fold increase, n = 17) compared with those with saline (∼1.5-fold increase, n = 15, Fig. 4B). These results suggest that in vivo exposure to leptin modifies the responsiveness of MC4R PVN neurons to α-MSH.

Fig. 4.

Fig. 4.

Open in a new tab

Fasting reduces responsiveness of MC4R PVN neurons to bath application of α-MSH in vitro. (A) Effect of bath applications of 250 nM α-MSH on firing frequency of MC4R PVN neurons obtained from 16-h fasted mice injected 3 h before decapitation with either 3 mg/kg leptin or saline (one-way ANOVA, *P = 0.0002). (B) Average ± SEM of magnitude of α-MSH–induced response indicates that this response is greater in neurons from mice treated with leptin (∼5.6-fold increase, n = 17) than those treated with saline (∼1.5-fold increase, n = 15, *P < 0.0005, one-way ANOVA). (C) Hypothalamic expression of MC4R gene normalized to β-actin gene obtained from fasted mice (20 h) that were injected i.p. 5 h before decapitation with either saline (n = 20) or leptin (n = 19, *P < 0.01). Data were obtained using quantitative real-time PCR. (D) Hypothalamic expression of MC4R normalized to β-actin gene obtained from MC4R−/− (KO, n = 5), MC4R−/+ (HET, n = 5), and MC4R+/+ (WT, n = 12) adult male mice fed with normal chow. Asterisk indicates all groups are significantly different (P < 0.0001, one-way ANOVA). Data were obtained using quantitative real-time PCR. (E) Average ± SEM of body weight of corresponding groups of mice as in D (*P < 0.0001, one-way ANOVA).

Leptin Increases Expression of MC4R mRNA.

We next tested the hypothesis that leptin might increase α-MSH responsiveness by increasing MC4R with quantitative real-time PCR of hypothalamic MC4R mRNA in 20-h fasted mice 5 h after i.p. injection with saline or 3 μg/g of body weight of leptin. Results indicate that i.p. administration of leptin significantly increased (∼24%) expression of MC4R mRNA compared with the saline-injected group (n = 19–20 mice, P < 0.01, Fig. 4C). These findings support the hypothesis that leptin increases the responsiveness of PVN neurons to α-MSH by increasing the expression of the MC4R gene and possibly increasing receptor density on neurons (35, 36). To test the impact of MC4R mRNA levels on energy homeostasis, we examined levels of MC4R expression in hypothalami of MC4R−/− (KO), MC4R +/− (HET), and MC4R +/+ (WT) age-matched male mice fed with normal chow (1). Our results indicate that the levels of MC4R mRNA, normalized to that of β-actin, in HET mice (0.57 ± 0.09, n = 5) were significantly lower that than in WT mice (1.01 ± 0.04, n = 12) and higher than those in KO mice (0.03 ± 0.01, n = 5, one-way ANOVA, P < 0.0001). Our data further indicate that the averages of body weight of HET mice (31.0 ± 0.7) were significantly higher than WT (26.6 ± 1.0) and lower than KO mice (42.1 ± 1.0, one-way ANOVA, P < 0.0001, Fig. 4 D and E), demonstrating a sharp inverse correlation between levels of MC4R mRNA and body weight.

Discussion

Analysis of action-potential firing frequency indicates that fasting activates MC4R neurons in the mid- to posterior PVN. Leptin replacement (3 h) in fasted mice reversed the fasting-induced increase in firing activity, suggesting that the increased firing frequency is caused by the reduction in serum leptin levels. These data imply that midposterior MC4R PVN neurons are under tonic inhibition by leptin. The absence of an increase in firing frequency of MC4R PVN neurons upon fasting in the ob/ob mouse further supports this hypothesis.

A commonly described model for leptin action in the regulation of PVN neurons invokes leptin-mediated augmentation of production and release of α-MSH as well as inhibition of release of NPY, AgRP, and GABA after leptin action on ARC POMC and NPY/AgRP neurons, respectively (13). Previously, we had demonstrated that α-MSH acts presynaptically to activate GABAergic inputs innervating medial parvocellular PVN neurons (37). Although presynaptic activity was also observed in this study, we report here a dominant and direct postsynaptic effect of α-MSH on MC4R PVN neurons, increasing the firing activity of these cells. Furthermore, NPY potently and reversibly inhibited neuronal firing activity of MC4R PVN neurons associated with hyperpolarization of membrane potentials. Thus, the tonic inhibition of firing frequency mediated by serum leptin cannot be explained by leptin-mediated increase in α-MSH and/or suppression of NPY release because we show that α-MSH stimulates and NPY potently inhibits firing frequency of these cells. AgRP was also found to potently inhibit the firing of MC4R PVN neurons (Fig. 2 C and D). This finding suggests that our preparation retains endogenous α-MSH release from projections onto the cells and/or that the MC4R retains constitutive activity in the preparation (38). These data suggest that the basal firing activity of energy homeostasis circuits is regulated not only by leptin but also in part by the constitutive activity of the MC4R.

Our results also indicate that leptin caused significant reversible inhibition of firing activity associated with hyperpolarization of membrane potential in all MC4R neurons recorded in mid- to posterior PVN (Fig. 4 A–E) and that these inhibitory effects were mediated through postsynaptic mechanisms. These findings, in addition to the lack of fasting-induced increase in firing activity of PVN neurons observed in leptin-deficient mice, suggest that leptin directly acts on PVN neurons to modulate their activity. These results clearly require a high-affinity leptin binding site because experiments were typically performed using doses as low as 50 nM leptin. The actions of leptin observed in vivo and in vitro, along with the observed effects of arcuate peptides, are consistent with a direct action of circulating leptin on MC4R PVN neurons, in contrast to the prevailing view of the dominance of the ARC and VMH inputs. Deletion of leptin receptors from POMC and VMH neurons, jointly, still does not cause the magnitude of obesity seen in the ob/ob mice (18). Of course, the leptin receptor acts in many other brain regions, including additional PVN-projecting sites such as the dorsomedial nucleus and lateral nucleus (9, 39–41). Thus, although we report here that nearly all midposterior MC4R PVN neurons are under tonic inhibition by leptin and exhibit an inhibitory postsynaptic response to leptin, the overall regulation of these neurons in vivo, of course, may involve the integration of direct leptin actions on PVN with leptin-regulated inputs from multiple additional nuclei other than ARC.

More than 90% of TRH-expressing neurons in the anterior PVN also exhibit a direct postsynaptic depolarizing response to both leptin and α-MSH (30). This study also suggests that direct leptin action at TRH MC4R neurons in PVN may be required for leptin replacement to maximally restore serum T4 levels during a fast. More generally, hypothalamo-pituitary disconnection in the sheep, which ablates the ARC, suggests that the ARC is required for excitation of neurons in the lateral nucleus but not in the PVN, VMH, or dorsomedial nucleus after i.v. leptin administration (42). Thus, MC4R PVN neurons, regardless of location or function, appear capable of responding to leptin in the absence of the ARC (Fig. S5).

Our in vivo data indicated that MC4R PVN neurons activated by fasting were located mostly (80–92%) in the mid- to posterior region of the nucleus. In vitro results recapitulated that finding by showing that neurons in this region were inhibited by bath application of leptin through postsynaptic mechanisms. Revealed by tracing techniques, a significant fraction of MC4R neurons in this region project to the hindbrain, coexpress oxytocin/vasopressin and/or CRF, and may mediate anorexic signals (Fig. S4). Previous studies have demonstrated a role of these peptides in relaying satiety signals from the PVN to the brainstem (43, 44). In fact, based on characterization of electrophysiological features reported for rat PVN neurons (25), our results indicate that ∼56% of these neurons are putative nonneuroendocrine brainstem-projecting cells involved in satiety and autonomic control (Table S2). In contrast, MC4R neurons activated by leptin were mostly located in the anterior PVN. These neurons are activated in fed states and inhibited during fasting states, properties consistent with up-regulation of the hypothalamic–pituitary–thyroid axis by fasting. Consistent with this model, our tracing and immunohistochemistry studies performed on TRH-cre mice (Table S1) indicated that this population of anterior MC4R neurons project to the median eminence and coexpress TRH (30).

In agreement with previous studies that circulating leptin increases the anorexigenic activity of MTII (32, 33), our results suggest that fasting in fact decreases the responsiveness of MC4R PVN neurons to applications of α-MSH. Our results indicate that fasting decreases hypothalamic expression of MC4R mRNA in a leptin-dependent manner. Furthermore, these findings provide evidence for plasticity of melanocortin signaling as a function of energy state. In parallel, there is precedent for a role of leptin in modulating the density of NPY-mediated membrane current in ARC neurons (34). Indeed, we determined that the quantity of expression of MC4R mRNA has a rough correlation with the severity of obesity in that haploinsufficiency of the MC4R, which causes an intermediate degree of obesity relative to wild-type and knockout mice, reduces mRNA expression by ∼50%.

In conclusion, this study provides a detailed analysis of the regulation of MC4R PVN neurons likely to be involved in feeding behavior. These data suggest that ARC to PVN circuits involved in energy homeostasis are directly and tonically controlled by leptin in a multinodal fashion. Additionally, we identify three regulatory mechanisms important for the control of these PVN-MC4 neurons: (i) direct postsynaptic modulation by leptin, (ii) regulation of MC4R mRNA expression and α-MSH responsiveness by leptin, and (iii) regulation by the constitutive activity of the MC4R signaling. The observation that anorexigenic α-MSH and leptin act in opposing direction to stimulate and inhibit midposterior MC4R PVN neurons, respectively, contradicts the commonly accepted model of regulation of the PVN by leptin, which argues for control via leptin action on a homogeneous population of PVN-projecting POMC and NPY/AgRP ARC neurons. Recent data, however, demonstrate that POMC neurons, for example, are quite heterogeneous both neurochemically and in terms of leptin responsiveness (45, 46), and the PVN-projecting subset of POMC neurons remain to be characterized.

Materials and Methods

Animals and Housing.

In all electrophysiological experiments, 26- to 60-d-old MC4R-Tau-Sapphire transgenic (MC4R-GFP) mice and leptin-deficient ob/ob mice on a C57BL/6J background were used (20). Animal husbandry is described in SI Materials and Methods. All animal experiments were approved by the University Animal Care and Use Committee.

Electrophysiology.

MC4R-GFP mice were deeply anesthetized with isoflurane before decapitation. The brain was entirely removed and immediately submerged in ice-cold, gassed (95% O2 and 5% CO2) artificial cerebrospinal fluid containing (in mM): 126.2 NaCl, 3.1 KCl, 2 Ca Cl2, 1 Mg Cl2, 1 NaH2PO4, 26.2 NaHCO3, 10 glucose, and 16.2 sucrose (pH 7.39, when gassed with 95% O2 and 5% CO2 at room temperature). Cell recordings were performed by using patch pipettes of 2.4-MΩ to 5-MΩ resistance when filled with a solution containing (in mM): 125 K gluconate, 8 KCl, 5 MgCl2, 10 Hepes, 5 NaOH, 4 Na2ATP, 0.4 Na3GTP, 15.4 sucrose, and 7 KOH, which resulted in a pH ∼7.23 and osmolarity of 295–300 mosM/kg. Preparation and use of drugs is described in SI Materials and Methods.

Quantitative Real-Time-PCR.

Total RNA was extracted from hypothalamic tissue with the RNeasy mini kit (Qiagen) according to the manufacturer's instruction, and gene expression analysis was performed in 96-well plates using TaqMan universal PCR master mix (Applied Biosystems) in a Stratagene Mx3000p. Details are provided in SI Materials and Methods.

Immunohistochemistry.

MC4R-GFP mice were anesthetized with 2% Avertin and were injected with 200 nl of 2% fluorogold into the nucleus tractus solitarius and spinal cord at T2. Mice were allowed to recover for 1 wk and then were injected with 20 μg of colchicine i.c.v. in 1 μL of sterile distilled water. After 24 h, mice were then deeply anesthetized and underwent tissue fixation via transcardial perfusion with 0.9% saline followed by ice-cold fixative (4% paraformaldehyde in 0.01 M PBS). Immunohistochemistry was then performed as described in SI Materials and Methods.

Statistical Analysis.

All data are presented as average ± SEM, and statistical significance was determined by using paired t test, except where indicated, with P = 0.05 as the threshold for statistical significance.

Supplementary Material

Supporting Information
supp_108_1_355__index.html (776B, html)

Acknowledgments

We thank Dr. J. Friedman (Rockefeller University, New York) for providing MC4R-Tau-Sapphire transgenic mice, A. Hollenberg (Beth Israel Deaconess Medical Center, Boston) for TRH-cre mice, and Dr. H. Gainer (National Institutes of Health, Bethesda, MD) for providing us with a monoclonal anti-neurophysin antibody. This work was supported by National Institutes of Health Grant DK070332 (to R.D.C.) and Canadian Institutes of Health Research Fellowship Award 129207 (to M.G.-L.).

Footnotes

The authors declare no conflict of interest.

References

  • 1.Huszar D, et al. Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell. 1997;88:131–141. doi: 10.1016/s0092-8674(00)81865-6. [DOI] [PubMed] [Google Scholar]
  • 2.Fan W, Boston BA, Kesterson RA, Hruby VJ, Cone RD. Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature. 1997;385:165–168. doi: 10.1038/385165a0. [DOI] [PubMed] [Google Scholar]
  • 3.Cone RD. Anatomy and regulation of the central melanocortin system. Nat Neurosci. 2005;8:571–578. doi: 10.1038/nn1455. [DOI] [PubMed] [Google Scholar]
  • 4.Zhang Y, et al. Positional cloning of the mouse obese gene and its human homologue. Nature. 1994;372:425–432. doi: 10.1038/372425a0. [DOI] [PubMed] [Google Scholar]
  • 5.Fekete C, et al. α-Melanocyte-stimulating hormone is contained in nerve terminals innervating thyrotropin-releasing hormone-synthesizing neurons in the hypothalamic paraventricular nucleus and prevents fasting-induced suppression of prothyrotropin-releasing hormone gene expression. J Neurosci. 2000;20:1550–1558. doi: 10.1523/JNEUROSCI.20-04-01550.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bell ME, Bhatnagar S, Akana SF, Choi S, Dallman MF. Disruption of arcuate/paraventricular nucleus connections changes body energy balance and response to acute stress. J Neurosci. 2000;20:6707–6713. doi: 10.1523/JNEUROSCI.20-17-06707.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Baker RA, Herkenham M. Arcuate nucleus neurons that project to the hypothalamic paraventricular nucleus: Neuropeptidergic identity and consequences of adrenalectomy on mRNA levels in the rat. J Comp Neurol. 1995;358:518–530. doi: 10.1002/cne.903580405. [DOI] [PubMed] [Google Scholar]
  • 8.Nillni EA, et al. Leptin regulates prothyrotropin-releasing hormone biosynthesis. J Biol Chem. 2000;275:36124–36133. doi: 10.1074/jbc.M003549200. [DOI] [PubMed] [Google Scholar]
  • 9.Elmquist JK, Bjørbaek C, Ahima RS, Flier JS, Saper CB. Distributions of leptin receptor mRNA isoforms in the rat brain. J Comp Neurol. 1998;395:535–547. [PubMed] [Google Scholar]
  • 10.Leshan RL, Björnholm M, Münzberg H, Myers MG., Jr Leptin receptor signaling and action in the central nervous system. Obesity (Silver Spring) 2006;14(Suppl 5):208S–212S. doi: 10.1038/oby.2006.310. [DOI] [PubMed] [Google Scholar]
  • 11.Elias CF, et al. Leptin differentially regulates NPY and POMC neurons projecting to the lateral hypothalamic area. Neuron. 1999;23:775–786. doi: 10.1016/s0896-6273(01)80035-0. [DOI] [PubMed] [Google Scholar]
  • 12.Takahashi KA, Cone RD. Fasting induces a large, leptin-dependent increase in the intrinsic action potential frequency of orexigenic arcuate nucleus neuropeptide Y/Agouti-related protein neurons. Endocrinology. 2005;146:1043–1047. doi: 10.1210/en.2004-1397. [DOI] [PubMed] [Google Scholar]
  • 13.Stephens TW, et al. The role of neuropeptide Y in the antiobesity action of the obese gene product. Nature. 1995;377:530–532. doi: 10.1038/377530a0. [DOI] [PubMed] [Google Scholar]
  • 14.Sarkar S, Légrádi G, Lechan RM. Intracerebroventricular administration of α-melanocyte stimulating hormone increases phosphorylation of CREB in TRH- and CRH-producing neurons of the hypothalamic paraventricular nucleus. Brain Res. 2002;945:50–59. doi: 10.1016/s0006-8993(02)02619-7. [DOI] [PubMed] [Google Scholar]
  • 15.Olney JW. Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science. 1969;164:719–721. doi: 10.1126/science.164.3880.719. [DOI] [PubMed] [Google Scholar]
  • 16.Li AJ, Dinh TT, Ritter S. Hyperphagia and obesity produced by arcuate injection of NPY-saporin do not require upregulation of lateral hypothalamic orexigenic peptide genes. Peptides. 2008;29:1732–1739. doi: 10.1016/j.peptides.2008.05.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Balthasar N, et al. Leptin receptor signaling in POMC neurons is required for normal body weight homeostasis. Neuron. 2004;42:983–991. doi: 10.1016/j.neuron.2004.06.004. [DOI] [PubMed] [Google Scholar]
  • 18.Dhillon H, et al. Leptin directly activates SF1 neurons in the VMH, and this action by leptin is required for normal body-weight homeostasis. Neuron. 2006;49:191–203. doi: 10.1016/j.neuron.2005.12.021. [DOI] [PubMed] [Google Scholar]
  • 19.Myers MG, Jr, Münzberg H, Leinninger GM, Leshan RL. The geometry of leptin action in the brain: More complicated than a simple ARC. Cell Metab. 2009;9:117–123. doi: 10.1016/j.cmet.2008.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Liu H, et al. Transgenic mice expressing green fluorescent protein under the control of the melanocortin-4 receptor promoter. J Neurosci. 2003;23:7143–7154. doi: 10.1523/JNEUROSCI.23-18-07143.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Cowley MA, et al. Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature. 2001;411:480–484. doi: 10.1038/35078085. [DOI] [PubMed] [Google Scholar]
  • 22.Smith MA, et al. Melanocortins and agouti-related protein modulate the excitability of two arcuate nucleus neuron populations by alteration of resting potassium conductances. J Physiol. 2007;578:425–438. doi: 10.1113/jphysiol.2006.119479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hruby VJ, et al. Cyclic lactam α-melanotropin analogues of Ac-Nle4-c[Asp4,D-Phe7, Lys10]α-MSH(4-10)-NH2 with bulky aromatic amino acids at position 7 show high antagonist potency and selectivity at specific melanocortin receptors. J Med Chem. 1995;38:3454–3461. doi: 10.1021/jm00018a005. [DOI] [PubMed] [Google Scholar]
  • 24.Fong TM, Van der Ploeg LH. A melanocortin agonist reduces neuronal firing rate in rat hypothalamic slices. Neurosci Lett. 2000;283:5–8. doi: 10.1016/s0304-3940(00)00823-5. [DOI] [PubMed] [Google Scholar]
  • 25.Tasker JG, Dudek FE. Electrophysiological properties of neurones in the region of the paraventricular nucleus in slices of rat hypothalamus. J Physiol. 1991;434:271–293. doi: 10.1113/jphysiol.1991.sp018469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ganong AH, Lanthorn TH, Cotman CW. Kynurenic acid inhibits synaptic and acidic amino acid-induced responses in the rat hippocampus and spinal cord. Brain Res. 1983;273:170–174. doi: 10.1016/0006-8993(83)91108-3. [DOI] [PubMed] [Google Scholar]
  • 27.Ludwig M, Leng G. Dendritic peptide release and peptide-dependent behaviours. Nat Rev Neurosci. 2006;7:126–136. doi: 10.1038/nrn1845. [DOI] [PubMed] [Google Scholar]
  • 28.Kishi T, et al. Neuropeptide Y Y1 receptor mRNA in rodent brain: Distribution and colocalization with melanocortin-4 receptor. J Comp Neurol. 2005;482:217–243. doi: 10.1002/cne.20432. [DOI] [PubMed] [Google Scholar]
  • 29.Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 2nd Ed. Sydney: Academic; 1986. [DOI] [PubMed] [Google Scholar]
  • 30.Ghamari-Langroudi M, et al. Regulation of thyrotropin-releasing hormone-expressing neurons in the paraventricular nucleus of the hypothalamus by signals of adiposity. Mol Endocrinol. 2010;24:2366–2381. doi: 10.1210/me.2010-0203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Merchenthaler I. Neurons with access to the general circulation in the central nervous system of the rat: A retrograde tracing study with fluoro-gold. Neuroscience. 1991;44:655–662. doi: 10.1016/0306-4522(91)90085-3. [DOI] [PubMed] [Google Scholar]
  • 32.Pierroz DD, et al. Effects of acute and chronic administration of the melanocortin agonist MTII in mice with diet-induced obesity. Diabetes. 2002;51:1337–1345. doi: 10.2337/diabetes.51.5.1337. [DOI] [PubMed] [Google Scholar]
  • 33.Blüher S, et al. Responsiveness to peripherally administered melanocortins in lean and obese mice. Diabetes. 2004;53:82–90. doi: 10.2337/diabetes.53.1.82. [DOI] [PubMed] [Google Scholar]
  • 34.Roseberry AG, Liu H, Jackson AC, Cai X, Friedman JM. Neuropeptide Y-mediated inhibition of proopiomelanocortin neurons in the arcuate nucleus shows enhanced desensitization in ob/ob mice. Neuron. 2004;41:711–722. doi: 10.1016/s0896-6273(04)00074-1. [DOI] [PubMed] [Google Scholar]
  • 35.Gout J, et al. Leptin infusion and obesity in mouse cause alterations in the hypothalamic melanocortin system. Obesity (Silver Spring) 2008;16:1763–1769. doi: 10.1038/oby.2008.303. [DOI] [PubMed] [Google Scholar]
  • 36.Enriori PJ, et al. Diet-induced obesity causes severe but reversible leptin resistance in arcuate melanocortin neurons. Cell Metab. 2007;5:181–194. doi: 10.1016/j.cmet.2007.02.004. [DOI] [PubMed] [Google Scholar]
  • 37.Cowley MA, et al. Integration of NPY, AGRP, and melanocortin signals in the hypothalamic paraventricular nucleus: Evidence of a cellular basis for the adipostat. Neuron. 1999;24:155–163. doi: 10.1016/s0896-6273(00)80829-6. [DOI] [PubMed] [Google Scholar]
  • 38.Nijenhuis WA, Oosterom J, Adan RA. AgRP(83-132) acts as an inverse agonist on the human-melanocortin-4 receptor. Mol Endocrinol. 2001;15:164–171. doi: 10.1210/mend.15.1.0578. [DOI] [PubMed] [Google Scholar]
  • 39.Håkansson ML, Meister B. Transcription factor STAT3 in leptin target neurons of the rat hypothalamus. Neuroendocrinology. 1998;68:420–427. doi: 10.1159/000054392. [DOI] [PubMed] [Google Scholar]
  • 40.Perello M, Stuart RC, Nillni EA. The role of intracerebroventricular administration of leptin in the stimulation of prothyrotropin releasing hormone neurons in the hypothalamic paraventricular nucleus. Endocrinology. 2006;147:3296–3306. doi: 10.1210/en.2005-1533. [DOI] [PubMed] [Google Scholar]
  • 41.Mercer JG, et al. Localization of leptin receptor mRNA and the long form splice variant (Ob-Rb) in mouse hypothalamus and adjacent brain regions by in situ hybridization. FEBS Lett. 1996;387:113–116. doi: 10.1016/0014-5793(96)00473-5. [DOI] [PubMed] [Google Scholar]
  • 42.Qi Y, Henry BA, Oldfield BJ, Clarke IJ. The action of leptin on appetite-regulating cells in the ovine hypothalamus: Demonstration of direct action in the absence of the arcuate nucleus. Endocrinology. 2010;151:2106–2116. doi: 10.1210/en.2009-1283. [DOI] [PubMed] [Google Scholar]
  • 43.Lu XY, Barsh GS, Akil H, Watson SJ. Interaction between α-melanocyte-stimulating hormone and corticotropin-releasing hormone in the regulation of feeding and hypothalamo-pituitary-adrenal responses. J Neurosci. 2003;23:7863–7872. doi: 10.1523/JNEUROSCI.23-21-07863.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Blevins JE, Schwartz MW, Baskin DG. Evidence that paraventricular nucleus oxytocin neurons link hypothalamic leptin action to caudal brain stem nuclei controlling meal size. Am J Physiol Regul Integr Comp Physiol. 2004;287:R87–R96. doi: 10.1152/ajpregu.00604.2003. [DOI] [PubMed] [Google Scholar]
  • 45.Williams KW, et al. Segregation of acute leptin and insulin effects in distinct populations of arcuate proopiomelanocortin neurons. J Neurosci. 2010;30:2472–2479. doi: 10.1523/JNEUROSCI.3118-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Hentges ST, Otero-Corchon V, Pennock RL, King CM, Low MJ. Proopiomelanocortin expression in both GABA and glutamate neurons. J Neurosci. 2009;29:13684–13690. doi: 10.1523/JNEUROSCI.3770-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting Information
supp_108_1_355__index.html (776B, html)
1016785108_pnas.201016785SI.pdf (1.3MB, pdf)

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES