| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Neurosciences Institute, Division of Pathology and Neuroscience, Ninewells Hospital and Medical School, University of Dundee, Dundee, UK
2 Centre for Diabetes and Endocrinology, Rayne Institute, University College London, London, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 17 August 2006;
accepted after revision 19 October 2006;
first published online 26 October 2006)
Corresponding author M. Ashford: Division of Pathology and Neuroscience, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, UK. Email: m.l.j.ashford{at}dundee.ac.uk
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-melanocyte-stimulating hormone (-MSH – a POMC gene product) suppresses, food intake (Ramos et al. 2005). Agouti-related protein (AgRP) is predominantly localized to NPY neurons, is a natural antagonist of -MSH on melanocortin 3 (MC3R) and melanocortin 4 (MC4R) receptors and stimulates a long-lasting increase in food intake (Hagan et al. 2000). The NPY/AgRP and POMC systems alter metabolic homeostasis by regulation of gene transcription, excitability and synaptic transmission (Cowley, 2003; Cone, 2005). They project to areas such as the paraventricular nucleus (PVN) and lateral hypothalamic area (LHA), where further integration occurs and outputs from these and the ARC extend to extra-hypothalamic centres. Hence, sites extrinsic to the ARC are thought to be where melanocortin receptors predominantly influence circuits responsible for energy homeostasis. Alterations of the melanocortin pathway, predominantly via the MC4R, have a major influence on energy homeostasis. Global deletion of the NPY gene (Erickson et al. 1996) produces a weak phenotype in comparison to transgenes that target the melanocortin system. Notably the MC4R knockout mouse (Huszar et al. 1997) and mice overexpressing agouti (Fan et al. 1997) or AgRP (Graham et al. 1997) are obese. Furthermore, selective ablation of AgRP and POMC neurons induces anorexia and hyperphagia, respectively, (Gropp et al. 2005). The MC4R displays constitutive activity, which appears to be essential for body weight maintenance (Srinivasan et al. 2004). AgRP is an inverse agonist and suppresses the intrinsic activity of the MC4R (Haskell-Luevano & Monck, 2001), indicating that AgRP may increase food intake independently of melanocortin ligands. Moreover, approximately 5% of severe human obesity has been ascribed to MC4R deficiency (Farooqi et al. 2000), and the melanocortin system, including AgRP, is implicated in anorexia (Kas et al. 2003), cachexia (Lechan & Tatro, 2001) and type 2 diabetes (Bonilla et al. 2006).
Although melanocortin receptors are key elements in energy homeostasis control, relatively little is known about the electrophysiological properties of MC3R- and MC4R-expressing neurons in these hypothalamic circuits. Previous studies have shown that ARC NPY neurons are insensitive to the mixed MC3R and MC4R agonist, MTII (Roseberry et al. 2004) and POMC neurons are inhibited by an MC3R agonist (Cowley et al. 2001). In addition to these neurons, an independent ARC neuronal population, identified by the rat insulin 2 promoter (RIPCre) transgene expression was demonstrated to differ from POMC neurons in its response to insulin and leptin. POMC neurons are hyperpolarized by insulin and depolarized by leptin, whereas RIPCre neurons are depolarized by insulin but are insensitive to leptin (Choudhury et al. 2005). Additionally, the mixed MC3R/MC4R agonist MTII depolarized RIPCre neurons. Consequently, we have investigated the mechanisms by which melanocortin agonists and AgRP alter the excitability of POMC and RIPCre neurons. We show here that melanocortin agonists depolarize and induce excitation of both POMC and RIPCRe neurons and that AgRP inhibits both neuron populations by membrane hyperpolarization.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
As previously described (Choudhury et al. 2005), we have used two Cre recombinase transgenic lines, RIPCre and POMCCre and intercrossed these with the ZEG indicator mouse to generate mice expressing green fluorescent protein (GFP) in selective hypothalamic neuronal populations. All procedures conformed to the UK Animals (Scientific Procedures) Act 1986, and were approved by our institutional ethical review committee. RIPCreZEG and POMCCreZEG mice (8-16 weeks old) were killed by cervical dislocation, the brain rapidly removed and submerged in an ice-cold slicing solution containing (mM): KCl 2.5, NaH2PO4 1.25, NaHCO3 28, CaCl2 0.5, MgCl2 7, D-glucose 7, ascorbate 1, pyruvate 3 and sucrose 235, equilibrated with 95% O2–5% CO2 to give a pH of 7.4. Hypothalamic coronal slices (350 µm), containing the ARC, were cut using a Vibratome, transferred and kept at room temperature (22–25°C) in an external solution containing (mM): NaCl 125, KCl 2.5, NaH2PO4 1.25, NaHCO3 25, CaCl2 2, MgCl2 1, D-glucose 10, D-mannitol 15, ascorbate 1 and pyruvate 3, equilibrated with 95% O2–5% CO2; pH 7.4, 
320 mosmol l–1.
Electrophysiology
Individual arcuate neurons were identified by epifluorescence and differential interference contrast optics using an upright Zeiss Axioskop-2 FS plus microscope. Slices were continually perfused with a modified external solution (0.5 mM CaCl2 and 2.5 mM MgCl2, no ascorbate and pyruvate) at a flow rate of 5–10 ml min–1 and a bath temperature of 33°C. For experiments in high-potassium solution, the normal external solution was replaced by a solution containing (mM): NaCl 130, KCl 20, CaCl2 0.5, MgCl2 2.5, D-glucose 10, D-mannitol 15 and Hepes 10; pH 7.4, 320 mosmol l–1. Patch-clamp recordings were performed using borosilicate patch pipettes (4–8 M
) filled with an internal solution containing (mM): potassium gluconate 130, KCl 10, EGTA 0.5, Hepes 10, NaCl 1, CaCl2 0.28, MgCl2 3, Na2ATP 3, Tris-GTP 0.3 and phosphocreatine 14; pH 7.2, 310 mosmol l–1. Whole-cell series resistance (Rs) was compensated using an Axopatch 200B amplifier in current- (Ifast) and voltage-clamp modes (Rs, 30–60 and 10–30 M, respectively). Note that any remaining uncompensated capacitance, produced by the voltage steps, was digitally subtracted in Igor Pro using a –90 mV voltage step as a template. Voltage and current commands were manually or externally driven using pCLAMP 9.2 software and injected into neurons via the patch-clamp amplifier. Under current clamp, hyperpolarizing current pulses (5–20 pA, 200 ms, at a frequency of 0.05 Hz) were used to monitor input and series resistance. Whole-cell currents and voltages were filtered at 5 and 2 kHz, respectively, and digitized at 50 kHz using pCLAMP 9.2 software. Miniature excitatory postsynaptic currents (mEPSCs) were recorded in a standard external solution (2 mM CaCl2 and 1 mM MgCl2) and in the presence of (+)bicuculline (20 µM), whereas miniature inhibitory postsynaptic currents (mIPSCs) were recorded in the presence of 2 mM kynurenic acid. Neurons were voltage clamped at –70 mV using a CsCl-based (130 mM) internal solution containing 5 mM QX314 to block regenerative sodium spikes. Unitary events of uniform rise and decay kinetics were identified by the template detection function in Clampex 9.2 and were analysed for peak amplitude and frequency. All data were stored unsampled on digital audio tape for off-line analysis using Clampex 9.2 or Igor Pro. Membrane potentials were either replayed unsampled on an EasyGraph TA240 chart recorder (Gould), or digitized and imported into Adobe illustrator for illustration purposes.
Unless stated otherwise, drugs were added to the external solution and applied to the slice via the perfusion system. Occasionally, drugs were locally applied using a broken-tipped pipette (4 µm diameter) positioned above the recording neurons, as previously described (Choudhury et al. 2005). At least 10 min of stable control data were recorded before the application of any drug, and, wherever possible, complete reversibility upon washout was sought. However, some responses appeared not to reverse completely as commonly found with peptides, even following extensive washout (
1 h). Note that neuronal integrity was assessed by all of the following: small holding current (
20 pA at –70 mV) when voltage clamped, high input resistance (1 G), large amplitude rebound spikes, the ability to fire and lack of obvious morphological deterioration (i.e. lack of blebbing and nucleus not visually present).
Chemicals
Kynurenic acid, (+)bicuculline, tetraethylammonium chloride (TEA), tolbutamide, 4-aminopyridine (4-AP), nimodipine, QX314 and mibefradil dihydrochloride were purchased from Sigma (Dorset, UK). Neuropeptide Y (NPY), JKC363, SHU9119, AgRP (82–131)-amide, MC4R agonist (cyclo(
-Ala-His-D-Phe-Arg-Trp-Glu)-NH2) and melanotan II (MTII) were obtained from Phoenix Pharmaceuticals Inc. (Belmont, CA, USA). Tetrodotoxin (TTX), r-tertiapin-Q, r-heteropodatoxin-2 (HpTX) and 
-conotoxin GVIA were purchased from Alomone Laboratories Ltd (Jerusalem, Israel). D-Trp8-
-MSH was a gift from Dr V. J. Hruby (University of Arizona).
Statistics
Statistical significance was determined on all neurons examined within a data set, using Student's two-tailed paired t test. However, as described by Cowley et al. (2001), we found that POMC neuronal responsiveness was not homogeneous. Therefore, as described by Dhillon et al. (2006), we divided responding neurons from non-responding neurons based on the criterion that the change (in membrane potential or current amplitude) induced by the drug challenge was ± three times the standard deviation prior to drug addition. The mean ± standard error of these responses is presented, with the number of cells studied.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
, n
= 62; POMC, 1.5 ± 0.05 G, n
= 50) and spontaneously fired sodium-mediated action potentials from a mean resting membrane potential (Vm, range –40 to –60 mV) of –50 ± 1 mV (RIPCre; n
= 62) and –50 ± 1 mV (POMC; n
= 50). Selective MC3R and MC4R agonists depolarize POMC and RIPCre neurons
In contrast to a previous report (Cowley et al. 2001), the selective MC3 receptor agonist (D-Trp8--MSH, 10–50 nM) depolarized (P < 0.05, n
= 11) POMC neurons (Fig. 1A) by +5.9 ± 1.2 mV (n
= 7 out of 11), although a lower concentration (5 nM) had no effect on POMC neuronal excitability (n
= 5, data not shown). D-Trp8--MSH (10–20 nM) also depolarized RIPCre neurons (Fig. 1B) by +9.6 ± 3.2 mV (P < 0.05, n
= 5 out of 5). Moreover, the selective MC4 receptor agonist cyclo(-Ala-His-D-Phe-Arg-Trp-Glu)-NH2 (10–20 nM; Bednarek et al. 2001) depolarized a majority of POMC neurons (Fig. 1C, P < 0.05, n
= 6) by +2.0 ± 0.4 mV (n
= 4 out of 6) and RIPCre neurons (Fig. 1D, P < 0.05, n
= 17) by +5.4 ± 1.2 mV (n
= 7 out of 17).
|
As the response to the MC4R agonist was unexpected, we examined whether AgRP could oppose the constitutive receptor activity associated with this receptor type (Haskell-Luevano & Monck, 2001). In the absence of bath-applied melanocortin agonist and consistent with its inverse agonist properties, AgRP (10–20 nM) hyperpolarized RIPCre (change in Vm (
Vm), –18.3 ± 4.8 mV, P < 0.05, n
= 4 out of 4) and POMC (Vm, –2.4 ± 0.7 mV, n
= 5 out of 10; P < 0.05, n
= 10) neurons (Fig. 2A and B). However, AgRP did not significantly affect input resistance (% of control: RIPCre, 78 ± 15%; POMC, 93 ± 7%; P > 0.1). Furthermore, AgRP did not exert its neuromodulatory effects on these ARC neurons indirectly by alteration of synaptic activity as RIPCre neuron membrane potential was hyperpolarized by AgRP (Fig. 2C; n
= 3 out 3) in the presence of TTX (0.5 µM).
|
Vm, –8.0 ± 2.8 mV, n
= 5). MTII also depolarized POMC neurons (P < 0.05, n
= 11) by a similar magnitude to RIPCre neurons (Vm, +5.2 ± 1.1 mV, n
= 5 out 11) and this effect was prevented by SHU9119 (SHU9119, –55 ± 3 mV; SHU9119 + MTII, –55 ± 3 mV; n
= 5, Fig. 2G). In contrast to SHU9119, application of the selective MC4R antagonist JKC363 (10–20 nM) had no effect on the membrane potential of RIPCre neurons, but did reverse the depolarization (P < 0.05, n
= 10) induced by the selective MC4R agonist (Fig. 2H). The MC4 receptor agonist depolarized RIPCre neurons by +4.4 ± 1.6 mV (P < 0.05, n
= 10) and the subsequent application of JKC363, in the continuous presence of MC4R agonist, repolarized these neurons by –4.4 ± 1.6 mV (P < 0.05), although again a significant delay in reversibility was observed (20–80 min). These data suggest that SHU9119 may possess some intrinsic inverse agonist activity independent of MC4R. MTII and AgRP directly modulate POMC and RIPCre neuronal activity
While the effects of MTII (Choudhury et al. 2005) and AgRP were reproducible in the presence of TTX, it is possible that spontaneous miniature synaptic events (that are insensitive to TTX; Roseberry et al. 2004) are modulated by these peptides and contribute to the responses. Thus, the frequency and amplitude of spontaneous mIPSCs and mEPSCs were recorded in the absence and presence of bath and/or locally applied MTII. The frequencies of these spontaneous events were not affected by 100 nM MTII in RIPCre (mIPSC: control, 0.6 ± 0.2 Hz; MTII, 0.4 ± 0.1 Hz; n
= 4, Fig. 3A; mEPSC: control, 4.1 ± 1.3 Hz; MTII, 4.1 ± 1.2 Hz; n
= 5, Fig. 3B) and POMC (mIPSC: control, 0.5 ± 0.1 Hz; MTII, 0.7 ± 0.1 Hz; n
= 5, Fig. 3E; mEPSC: control, 0.9 ± 0.2 Hz; MTII, 0.4 ± 0.1 Hz; n
= 5, Fig. 3F) neurons. Moreover, the cumulative frequency–amplitude distributions did not differ between control events and those recorded in the presence of MTII for mEPSCs or mIPSCs in RIPCre neurons (Fig. 3C and D) or for mEPSCs in POMC neurons (Fig. 3G). However, in POMC neurons, mIPSC amplitude was reversibly increased by MTII (change in amplitude, +6.1 ± 0.9 pA, n
= 5, P < 0.05) and the cumulative frequency distribution shifted in the presence of MTII (Fig. 3H). These data indicate that MTII increases GABAA receptor tone on POMC neurons, as previously reported (Cowley et al. 2001). In contrast, the mechanism of action appears to differ, such that D-Trp8--MSH was reported to increase mIPSC frequency (Cowley et al. 2001), whereas in this study MTII increased synaptic amplitude.
|
Block of K+ channels prevents MTII depolarization
As both POMC and RIPCre neurons displayed responses to melanocortin-selective ligands, which were indistinguishable from one another qualitatively, we did not discriminate between these populations when examining the mechanisms underlying these electrical changes, and consequently utilized MTII as agonist. Initially, we attempted to block the depolarizing response of MTII by the presence, in the bath solution, of various heavy metal cations (0.1–1 mM Gd3+, La3+, Cs+, Cd2+ or Ni2+), which block various non-selective cation and calcium channels (Hille, 2001). However, these and indeed more selective calcium channel blockers such as -conotoxin GVIA (100 nM), nimodipine (10 µM) and mibefradil (10 µM) did not alter the membrane potential or prevent MTII-induced depolarization of RIPCre neurons (data not shown). These data, while not conclusive, indicate that calcium and non-selective cation conductances are unlikely to be responsible for the MTII-induced depolarization. Thus, we assessed whether blocking potassium conductance could mimic or prevent the depolarization elicited by MTII on ARC neurons.
The presence of the ATP-sensitive potassium (KATP) channel inhibitor tolbutamide (200 µM) did not significantly affect membrane potential and failed to prevent depolarization of RIPCre neurons by 100 nM MTII (n = 3 out of 3; data not shown). Ba2+ (100 µM), a non-selective inward rectifier potassium (KIR) channel blocker (Takano & Ashcroft, 1996; Liu et al. 2001), induced a small depolarization of RIPCre and POMC neurons (Fig. 4A) by +2.5 ± 0.5 mV (n = 4 out of 4, P < 0.05). The voltage-dependent transient outward potassium (‘IA-type’ currents) channel blocker (Wang et al. 1996), 4-AP (4 mM) depolarized RIPCre and POMC neurons by +5.6 ± 0.5 mV (P < 0.05, n = 5 out of 5; Fig. 4B). These actions of Ba2+ and 4-AP were reproducible in the presence of inhibitors of glutamatergic (kynurenic acid) and GABAergic ((+)bicuculline) fast neurotransmission (data not shown). Consequently, hyperpolarizing current was injected into neurons in the presence of either Ba2+ or 4-AP to reset the membrane potential to the preblocker value prior to challenge with MTII (see Fig. 4B and C). Subsequent application of 100 nM MTII, in the presence of either Ba2+ (control, –52 ± 2 mV; Ba2+, –52 ± 2 mV; n = 4 out of 4) or 4-AP (control, –57 ± 4 mV; 4-AP, –58 ± 6 mV; n = 4 out of 4) alone did not affect the resting membrane potential (Fig. 4A and B). Although Ba2+ was difficult to wash out from the slice, on the occasions where clear reversibility was obtained, re-application of 100 nM MTII induced a depolarizing response (e.g. Fig. 4A). Thus, the presence of either Ba2+ or 4-AP alone is capable of preventing neuron depolarization by 100 nM MTII, although MTII increased spike frequency from control by 69 ± 20% (n = 4, P < 0.05) and 36 ± 5% (n = 4, P < 0.05) in the presence of 4-AP and Ba2+, respectively (Fig. 4A and B). However, in the combined presence of Ba2+ and 4-AP, 100 nM MTII was unable to alter the membrane potential (control, –55 ± 2 mV; MTII, –55 ± 4 mV, n = 5 out of 5) or firing frequency (control, 3.4 ± 0.5 Hz; MTII, 3.6 ± 0.7 Hz, n = 5) of POMC neurons (Fig. 4C), suggesting that both conductances (the transient outward current (henceforth termed IA for convenience) and KIR) contribute to the MTII response.
|
In order to identify the potassium conductance modulated by MTII and AgRP, RIPCre neurons were voltage clamped in an external solution containing 10 µM bicuculline, 2 mM kynurenic acid and 1 µM TTX to block synaptic transmission and regenerative Na+ spikes. Neurons were held at –70 mV and voltage pulses (500 ms duration) were stepped from –90 to –10 mV in 5 mV increments, with a 5 ms prepulse stepped to –170 mV (5 ms duration) to deactivate voltage-dependent potassium conductances (Fig. 5A). Initial studies with an external calcium concentration of 0.5 mM, 40 mM external tetraethylammonium (TEA) and internal K+ replaced by Cs+ (in order to inhibit potassium conductance), demonstrated that no significant calcium current could be observed (data not shown). In contrast, with normal external saline and potassium gluconate in the electrode solution, depolarizing command potentials (above –45 mV) elicited outward current, characterized by an initial rapid transient component that inactivated to reveal an underlying steady-state current (Fig. 5B). Such current kinetics are indicative of the IA and delayed rectifier (IK) potassium conductances, respectively. To confirm this interpretation we examined the effect of the IA channel blocker 4-AP (4 mM). Accordingly, at –10 mV, 4-AP blocked the peak current by 56 ± 6% (P < 0.05, n = 4 out of 4) but also partially blocked the steady-state (measured at the end of the pulse) current (Fig. 5B) by 28 ± 6% (P < 0.05, n = 4 out of 4). Application of 40 mM TEA resulted in a 68 ± 4% (n = 4 out of 4, P < 0.05) depression in the amplitude of the delayed rectifier current (data not shown).
|
In contrast with MTII, local application of 10 nM AgRP to voltage-clamped RIPCre neurons caused an increase in IA amplitude (P < 0.05, n = 9) at all potentials where this current was visible (Fig. 6A). This action of AgRP was not immediately reversed on washout (Fig. 6B and C). For example, at a potential of –15 mV, AgRP increased peak amplitude by 19 ± 7% (n = 7 out of 9). Although on some occasions AgRP apparently increased the current through IK (Fig. 6A), this was not a consistent finding and, on average, there was no significant effect on the steady-state outward current amplitude in the presence of either MTII or AgRP. As with MTII, the voltage-dependent activation curve of IA was not affected by AgRP such that V0.5 values were unaltered (control, –31 ± 1 mV; AgRP, –32 ± 1 mV, n = 7; Fig. 6D). These data suggest that MTII and AgRP modulate IA independent of changes in voltage-dependent gating.
|
|
-MSH reduced the peak amplitudes of IA (P < 0.05, n
= 12, Fig. 7F), as did the MC4R selective agonist, 10 nM (cyclo(-Ala-His-D-Phe-Arg-Trp-Glu)-NH2) (P < 0.05, n
= 14, Fig. 7I). Thus, at a test voltage of –15 mV, IA peak amplitude was decreased by 8 ± 3% (n
= 8 out of 12) and 12 ± 3% (n
= 7 out of 14) in the presence of the MC3 (Fig. 7G) and MC4 (Fig. 7J) receptor agonists, respectively. Furthermore, the selective MC3R (P < 0.05, n
= 12) and MC4R (P
= 0.07, n
= 14) agonists decreased KIR currents (Fig. 7F and I), although the latter response was not significant at the 95% level of confidence. For RIPCre neurons that produced a response to these agonists, the steady-state slope conductance (measured at the end of each voltage step between –90 and –50 mV) was decreased by 18 ± 5% (n
= 8 out 12) and 28 ± 8% (n
= 7 out of 14) by application of the MC3 (Fig. 7H) and MC4 (Fig. 7K) receptor agonists, respectively. These data indicate that both melanocortin receptor subtypes contribute to melanocortin agonist modulation of IA and perhaps KIR conductances in ARC neurons. |
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Previous studies demonstrate a wide CNS distribution of MC3R and MC4R, with moderate MC3R and low MC4R density in the arcuate (Mountjoy et al. 1994; Harrold et al. 1999; Liu et al. 2003). The PVN and dorsomedial nucleus (DMH) express much higher levels of MC4R (Liu et al. 2003) and are the hypothalamic areas with the greatest sensitivity to melanocortin agonists and AgRP (Kim et al. 2000). The location and effectiveness of MC4R modulation in hypothalamic circuits is particularly important given that MC4R knockout mice display a more severe obese phenotype (Huszar et al. 1997) than MC3R knockout mice (Butler et al. 2000). Moreover, AgRP and MTII do not significantly affect food intake or body weight in MC4R knockout mice (Fekete et al. 2004). It is interesting that using a Cre-mediated reactivation approach, Balthasar et al. 2005) showed that MC4R outside the PVN and amygdala is required to replicate the phenotype of the MC4R global null mouse. Clearly, MC4R is crucial for energy homeostasis, but questions remain as to the cellular mechanisms by which melanocortin agonists and AgRP modulate hypothalamic neuron excitability and the role of these receptors in the ARC.
We have demonstrated that two distinct arcuate neuron populations, POMC and RIPCre, display sensitivity to MC3R and MC4R selective agonists and to the inverse agonist, AgRP. A previous study on POMC neurons reported that the selective MC3R agonist D-Trp8--MSH hyperpolarized the majority of POMC neurons, although it was unclear whether this was due to indirect (MC3R autoreceptor) or direct action (Cowley et al. 2001). As we had previously reported that the mixed MC3R/MC4R agonist MTII depolarized RIPCre neurons (Choudhury et al. 2005), we expected the MC4R agonist to depolarize these arcuate neurons. However, MC3R and MC4R agonists depolarize both POMC and RIPCre neurons, and MTII induces depolarization in the presence of TTX. Furthermore, we found no evidence of altered synaptic drive (inhibitory or excitatory), causing a change of membrane potential, or of direct hyperpolarization of POMC neurons. These results are consistent with a recent report that MTII depolarizes MC4R-positive neurons in the DMH/PVN (Liu et al. 2003). Nevertheless, we observed that MTII drives a small increase in mIPSC amplitude in POMC neurons, indicating that under specific ionic conditions, MTII could hyperpolarize POMC neurons via an indirect, presynaptic mechanism. However, MTII failed to hyperpolarize POMC or RIPCre neurons under control conditions or when the MC receptor signalling system in the postsynaptic neuron was blocked with Rp-cAMP. Following dialysis of the neuron with our internal recording solution (chloride reversal potential is –53 mV), an increase in GABAA activity is unlikely to have much influence on membrane potential (Vm, –50 mV). Using the perforated-patch technique, which does not disrupt the intracellular environment, we observed similar MTII-induced depolarizations. Increasing mIPSC amplitude could reduce input resistance and shunt the resting potassium conductance, which would result in depolarization. However, no significant change in input resistance was observed in this study, perhaps due to the combination of opening GABAA channels, inhibition of resting potassium conductances, and opening of voltage-gated conductances that accompany neuronal depolarization.
Does MC4R constitutive activity contribute to neuron excitability?
The MC4R displays constitutive activity in the absence of ligand, which is inhibited by AgRP binding (Nijenhuis et al. 2001). The long-term increase in food intake induced by AgRP administration may be a result of its inverse agonist properties because JKC363 (a selective MC4R antagonist) does not replicate this action (Kim et al. 2002). POMC and RIPCre neuronal hyperpolarization by AgRP in the absence of exogenously applied melanocortin agonist is consistent with inverse agonism, whereby the intrinsic activity of the receptor contributes a tonic depolarizing influence. Alternative explanations for the hyperpolarization are that AgRP acts on a separate (non-melanocortin) receptor (Hagan et al. 2000; Kim et al. 2002) or behaves as a competitive antagonist at MC3R and MC4R (Yang et al. 1999) to block the actions of endogenously released -MSH. The latter possibility, although supported by the observation that SHU9119 causes RIPCre neuron hyperpolarization, is not substantiated by the lack of effect of JKC363, and the finding that AgRP induces hyperpolarization in the presence of TTX.
Mechanisms of melanocortin and AgRP actions on POMC and RIPCre neurons
The K+ channel blocker data indicate that reduced postsynaptic potassium conductance underlies the depolarization elicited by melanocortin agonists. Voltage-clamp recordings from arcuate neurons demonstrate that melanocortin agonists reduced IA and, to a lesser extent, KIR conductances. Indeed, to completely inhibit the depolarizing/excitatory influence of MTII on RIPCre and POMC neurons required the simultaneous presence of inhibitors of these conductances, 4-AP (IA) and Ba2+ (KIR). This supports the notion that multiple K+ conductances control the excitability of arcuate neurons. Previous studies have suggested that GIRK channels underlie the resting potassium conductance in POMC neurons (Cowley et al. 2001; Ibrahim et al. 2003; Roseberry et al. 2004). However, the concentrations (> 100 µM) of barium used to substantiate this claim block most of the KIR family. Indeed, the GIRK channel inhibitor r-tertiapin-Q toxin did not block the resting inward rectifier conductance in this, and a recent (Acuna-Goycolea & van den Pol, 2005), study indicating that another strong inward rectifier (e.g. KIR2 family) may underlie this conductance. The finding that melanocortin agonists, and AgRP, alter IA in POMC and RIPCre neurons indicates that this conductance is a common target for transmitter receptors in hypothalamic neurons (Burdakov & Ashcroft 2002; Tang et al. 2005). The mechanism by which melanocortin receptors alter IA is unclear. We have no evidence for modulation of voltage-dependent gating to explain the modification of IA and suggest that a change in channel open state probability or availability may be responsible.
Arcuate neurons have high input resistances (1–2 G) as a result of a very low resting conductance, thus a change of a few picoamps in resting KIR and/or IA window current is more than sufficient to produce millivolt changes in Vm. Additionally, because these neurons have a membrane potential that straddles the threshold for action potential firing, such modest changes in conductance and membrane potential can have a significant effect on spike frequency. A recent report shows that although NPY and peptide YY (PYY3-36) inhibited resting calcium conductances in POMC and NPY neurons, this did not affect membrane potential (Acuna-Goycolea & van den Pol, 2005). However, NPY, PYY3-36 and µ-opioid receptor stimulation activate a GIRK, r-tertiapin-Q toxin-sensitive (Acuna-Goycolea & van den Pol, 2005) conductance in identified arcuate neurons (Cowley et al. 2001; Ibrahim et al. 2003; Roseberry et al. 2004). This causes significant hyperpolarization with a large increase in resting potassium conductance. Consequently, to repolarize arcuate neurons when GIRK channels are activated, significant neuronal excitation or the closure of GIRK channels is required. Such responses are quite distinct from melanocortin receptor agonist actions described here, which use alternative conductances (IA and KIR) to produce modest changes in membrane potential (5 mV) and resting conductance (20% of control), and perhaps are best described as neuromodulatory rather than having the ability to turn neurons ‘on’ or ‘off’.
Implications for the local arcuate circuit model
The presence of functional MC4R and MC3R on POMC neurons suggests that both could amplify melanocortin peptide release, in a local arcuate positive feedback circuit. In contrast, AgRP acting on POMC neurons would inhibit this positive feedback loop, by antagonism of MC3R and through inverse agonism at MC4R, and provide local inhibition of the melanocortin network even in the absence of significant -MSH release. This would reinforce orexigenic outputs and limit satiety signals at the level of the arcuate circuitry in addition to AgRP limiting melanocortin signals at second-order neurons in the PVN and LHA. The functional activation of melanocortin receptors on RIPCre neurons suggest that this neuronal type may also integrate the gene products locally released from both POMC- and AgRP-expressing neurons. Therefore, we suggest that the two-neuron model of ARC circuitry involved in control of energy homeostasis (Cowley et al. 2001; Cowley, 2003; Cone, 2005) may need to be extended in order to encompass additional neuronal types such as RIPCre.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Balthasar N, Dalgaard LT, Lee CE, Yu J, Funahashi H, Williams T et al. (2005). Divergence of melanocortin pathways in the control of food intake and energy expenditure. Cell 123, 493–505.[CrossRef][Medline]
Bays HE (2004). Current and investigational anti-obesity agents and obesity therapeutic treatment targets. Obes Res 12, 1197–1211.[Medline]
Bednarek MA, MacNeil T, Tang R, Kalyani RN, van der Ploeg LH & Weinberg DH (2001). Potent and selective peptide agonists of -melanotropin action at human melanocortin receptor 4: their synthesis and biological evaluation in vitro. Biochem Biophys Res Commun 286, 641–645.[CrossRef][Medline]
Bonilla C, Panguluri RK, Taliaferro-Smith L, Argyropoulos G, Chen G, Adeyemo AA et al. (2006). Agouti-related protein promoter variant associated with leanness and decreased risk for diabetes in West Africans. Int J Obes 30, 715–721.[CrossRef][Medline]
Burdakov D & Ashcroft FM (2002). Cholecystokinin tunes firing of an electrically distinct subset of arcuate nucleus neurons by activating A-type potassium channels. J Neurosci 22, 6380–6387.
Butler AA, Kesterson RA, Khong K, Cullen MJ, Pelleymounter MA, Dekoning J, Baetscher M & Cone RD (2000). A unique metabolic syndrome causes obesity in the melanocortin-3 receptor-deficient mouse. Endocrinology 141, 3518–3521.
Choudhury AI, Heffron H, Smith MA, Al-Qassab H, Xu AW, Selman C et al. (2005). The role of insulin receptor substrate 2 in hypothalamic and cell function. J Clin Invest 115, 940–950.[CrossRef][Medline]
Cone RD (2005). Anatomy and regulation of the central melanocortin system. Nat Neurosci 8, 571–578.[CrossRef][Medline]
Cowley MA (2003). Hypothalamic melanocortin neurons integrate signals of energy state. Eur J Pharmacol 480, 3–11.[CrossRef][Medline]
Cowley MA, Smart JL, Rubinstein M, Cerdán MG, Diano S, Horvath TL, Cone RD & Low MJ (2001). Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 411, 480–484.[CrossRef][Medline]
Dhillon H, Zigman JM, Ye C, Lee CE, McGovern RA, Tang V et al. (2006). Leptin directly activates SF1 neurons in the VMH, and this action by leptin is required for normal body-weight homeostasis. Neuron 49, 191–203.[CrossRef][Medline]
Erickson JC, Clegg KE & Palmiter RD (1996). Sensitivity to leptin and susceptibility to seizures of mice lacking neuropeptide Y. Nature 381, 415–418.[CrossRef][Medline]
Fan W, Boston BA, Kesterson RA, Hruby VJ & Cone RD (1997). Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 385, 165–168.[CrossRef][Medline]
Farooqi IS, Yeo GS, Keogh JM, Aminian S, Jebb SA, Butler G, Cheetham T & O'Rahilly S (2000). Dominant and recessive inheritance of morbid obesity associated with melanocortin 4 receptor deficiency. J Clin Invest 106, 271–279.[Medline]
Fekete C, Marks DL, Sarkar S, Emerson CH, Rand WM, Cone RD & Lechan RM (2004). Effect of agouti-related protein in regulation of the hypothalamic-pituitary-thyroid axis in the melanocortin 4 receptor knockout mouse. Endocrinology 145, 4816–4821.
Graham M, Shutter JR, Sarmiento U, Sarosi I & Stark KL (1997). Over-expression of Agrt leads to obesity in transgenic mice. Nat Genet 17, 273–274.[CrossRef][Medline]
Gropp E, Shanabrough M, Borok E, Xu AW, Janoschek R, Buch T et al. (2005). Agouti-related peptide-expressing neurons are mandatory for feeding. Nat Neurosci 8, 1289–1291.[CrossRef][Medline]
Hagan MM, Rushing PA, Pritchard LM, Schwartz MW, Strack AM, Van der Ploeg LH, Woods SC & Seeley RJ (2000). Long-term orexigenic effects of AgRP-(83–132) involve mechanisms other than melanocortin blockade. Am J Physiol Regul Integr Comp Physiol 279, R47–R52.
Harrold JA, Widdowson PS & Williams G (1999). Altered energy balance causes selective changes in melanocortin-4 (MC4-R), but not melanocortin-3 (MC3-R), receptors in specific hypothalamic regions: further evidence that activation of MC4-R is a physiological inhibitor of feeding. Diabetes 48, 267–271.[Abstract]
Haskell-Luevano C & Monck EK (2001). Agouti-related protein functions as an inverse agonist at a constitutively active brain melanocortin-4 receptor. Regul Pept 99, 1–7.[CrossRef][Medline]
Hille B (2001). Ion Channels of Excitable Membrane, 3rd edn. Sinauer Associates Inc., Massachusetts, USA.
Huszar D, Lynch C, Rair-Huntress V, Dunmore JH, Fang Q, Berkemeier LR et al. (1997). Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 88, 131–141.[CrossRef][Medline]
Ibrahim N, Bosch MA, Smart JL, Qiu J, Rubinstein M, Rønnekleiv OK, Low MJ & Kelly MJ (2003). Hypothalamic proopiomelanocortin neurons are glucose responsive and express KATP channels. Endocrinology 144, 1331–1340.
Kanjhan R, Coulson EJ, Adams DJ & Bellingham MC (2005). Tertiapin-q blocks recombinant and native large conductance K+ channels in a use-dependent manner. J Pharmacol Exp Ther 314, 1353–1361.
Kas MJ, van Dijk G, Scheurink AJ & Adan RA (2003). Agouti-related protein prevents self-starvation. Mol Psychiatry 8, 235–240.[CrossRef][Medline]
Kim MS, Rossi M, Abbott CRAI, Ahmed SH, Smith DM & Bloom SR (2002). Sustained orexigenic effect of Agouti related protein may not be mediated by the melanocortin receptor. Peptides 23, 1069–1076.[CrossRef][Medline]
Kim MS, Rossi M, Abusnana S, Sunter D, Morgan DG, Small CJ et al. (2000). Hypothalamic localization of the feeding effect of agouti-related peptide and alpha-melanocyte-stimulating hormone. Diabetes 49, 177–182.[Abstract]
Lechan RM & Tatro JB (2001). Hypothalamic melanocortin signaling in cachexia. Endocrinology 142, 3288–3291.
Liu GX, Derst C, Schlichthörl G, Heinen S, Seebohm G, Brüggemann A, Kummer W, Veh RW, Daut J & Müller RP (2001). Comparison of cloned Kir2 channels with native inward rectifier K+ channels from guinea-pig cardiomyocytes. J Physiol 532, 115–126.
Liu H, Kishi T, Roseberry AG, Cai X, Lee CE, Montez JM, Friedman JM & Elmquist JK (2003). Transgenic mice expressing green fluorescent protein under the control of the melanocortin-4 receptor promoter. J Neurosci 23, 7143–7154.
Mountjoy KG, Mortrud MT, Low MJ, Simerly RB & Cone RD (1994). Localization of the melanocortin-4 receptor (MC4-R) in neuroendocrine and autonomic control circuits in the brain. Mol Endocrinol 8, 1298–1308.[Abstract]
Nijenhuis WA, Oosterom J & Adan RA (2001). AgRP (83-132) acts as an inverse agonist on the human melanocortin-4 receptor. Mol Endocrinol 15, 164–171.
Plum L, Ma X, Hampel B, Balthasar N, Coppari R, Münzberg H et al. (2006). Enhanced PIP3 signaling in POMC neurons causes KATP channel activation and leads to diet-sensitive obesity. J Clin Invest 116, 1886–1901.[CrossRef][Medline]
Ramos EJ, Meguid MM, Campos AC & Coelho JC (2005). Neuropeptide Y, -melanocyte-stimulating hormone, and monoamines in food intake regulation. Nutrition 21, 269–279.[CrossRef][Medline]
Roseberry AG, Liu H, Jackson AC, Cai X & Friedman JM (2004). Neuropeptide Y-mediated inhibition of proopiomelanocortin neurons in the arcuate nucleus shows enhanced desensitization in ob/ob mice. Neuron 41, 711–722.[CrossRef][Medline]
Srinivasan S, Lubrano-Berthelier C, Govaerts C, Picard F, Santiago P, Conklin BR & Vaisse C (2004). Constitutive activity of the melanocortin-4 receptor is maintained by its N-terminal domain and plays a role in energy homeostasis in humans. J Clin Invest 114, 1158–1164.[CrossRef][Medline]
Takano M & Ashcroft FM (1996). The Ba2+ block of the ATP-sensitive K+ current of mouse pancreatic beta-cells. Pflugers Arch 431, 625–631.[Medline]
Tang SL, Tran V & Wagner EJ (2005). Sex differences in the cannabinoid modulation of an A-type K+ current in neurons of the mammalian hypothalamus. J Neurophysiol 94, 2983–2986.
Wang XY, McKenzie JS & Kemm RE (1996). Whole-cell K+ currents in identified olfactory bulb output neurones of rats. J Physiol 490, 63–77.[Medline]
Wikberg JE (1999). Melanocortin receptors: perspectives for novel drugs. Eur J Pharmacol 375, 295–310.[CrossRef][Medline]
Yang YK, Thompson DA, Dickinson CJ, Wilken J, Barsh GS, Kent SB & Gantz I (1999). Characterization of agouti-related protein binding to melanocortin receptors. Mol Endocrinol 13, 148–155.
|
Acknowledgements |
|---|
-MSH. Research support was provided by The Wellcome Trust (M.L.J.A. and D.J.W.), Medical Research Council (D.J.W.) and Biotechnology and Biological Sciences Research Council (D.J.W.). This article has been cited by other articles:
|
|
 |
|
  L.-Y. Fu and A. N. van den Pol Agouti-Related Peptide and MC3/4 Receptor Agonists Both Inhibit Excitatory Hypothalamic Ventromedial Nucleus Neurons J. Neurosci., May 21, 2008; 28(21): 5433 - 5449. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Fujimoto, T. Shiuchi, T. Miki, Y. Minokoshi, Y. Takahashi, A. Takeuchi, K. Kimura, M. Saito, T. Iwanaga, and S. Seino Dmbx1 is essential in agouti-related protein action PNAS, September 25, 2007; 104(39): 15514 - 15519. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||
