Neuropeptide W (NPW) is an endogenous ligand for G protein-coupled receptor 7 (GPR7). There are two forms of the peptide, designated as neuropeptide W-23 (NPW23) and neuropeptide W-30 (NPW30). In the current study we found that intracerebroventricular administration of NPW23 increased c-Fos immunoreactivity (IR) in a variety of brain sites, many of which are involved in the regulation of feeding. In particular, we noted that c-Fos IR levels were increased in hypocretin-expressing neurons in the perifornical region of the lateral hypothalamus (LH). We then studied whether injection of NPW23 into the paraventricular nucleus of the hypothalamus (PVN) and the LH increased food intake over a 24-h time period. Intra-PVN injection of NPW23 at doses ranging from 0.1 to 3 nmol increased feeding for up to 4 h, and doses ranging from 0.3 to 3 nmol increased feeding for up to 24 h. In contrast, only the 3-nmol dose of NPW23 increased feeding after administration into the LH. Together, these data suggest a modulatory role for NPW in the control of food intake.
- ingestive behavior
the orphan G protein-coupled receptor 7 (GPR7) is synthesized in the amygdala, hippocampus, cortex, and hypothalamus of rats (15, 18). One of the endogenous ligands for this receptor, neuropeptide W (NPW), is distributed both peripherally and centrally in the rat (4, 22). The gene encoding NPW is mainly expressed in the hypothalamus and hippocampus of rat brains. Another neuropeptide, NPB, which bears significant sequence similarity to NPW, has also been shown to bind GPR7 (7).
There are two forms of the peptide, designated as neuropeptide W-23 (NPW23) and neuropeptide W-30 (NPW30), both of which bind to the GPR7 receptor with nanomolar affinities (22). NPW23 is the shorter form equivalent to the NH2-terminal sequence of NPW30. Dun et al. (5) reported that immunoreactive NPW23 cells were found in several nuclei of the hypothalamus, including the paraventricular nucleus (PVN), supraoptic nucleus, accessory neurosecretory nuclei, dorsal and lateral hypothalamic (LH) areas, perifornical nucleus, arcuate nucleus, and anterior and posterior pituitary. Intracerebroventricular (icv) administration of NPW23 in rats increases food intake at a dose as low as 3 nmol but only for ∼2 h after injection (2, 22). In contrast, Mondal et al. (17) noted that icv administered NPW decreased food intake in deprived rats and during dark-phase feeding. Interestingly, centrally administered NPW activates the hypothalamic-pituitary-adrenal (HPA) axis, an effect that may be mediated by increased release of CRF (25). These changes in the HPA axis could potentially impact feeding behavior (25). Injection of NPW23 icv stimulates prolactin and corticosterone levels and decreases growth hormone levels in the circulation (2, 22). Icv administration of NPB, which binds to the orphan receptors GPR7 and GPR8, was also shown recently to decrease growth hormone and to elevate prolactin and corticosterone circulating levels (21). Recently Ishii et al. (8) reported that gold-thioglucose lesions in the ventromedial nucleus of the hypothalamus, which increased food intake, resulted in downregulation of the GPR7 receptor. GPR7-null male mice are hyperphagic and show altered energy metabolism (2).
In the current study we mapped the brain areas activated by icv administration of NPW with c-Fos immunocytochemistry. We determined whether injection of NPW into the PVN and LH, two sites known to be involved in food intake regulation, altered food intake over a 24-h period. Opioid pathways have been implicated in feeding stimulated by other orexigenic peptides (16, 24). Thus we also studied the effect of parenchymal injection of naltrexone (NTX), an opioid antagonist, on NPW-induced feeding.
MATERIALS AND METHODS
All experiments were carried out in accordance with institutional guidelines for animal care. Two sets of adult male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) were used. The first set of rats weighed between 250 and 300 g at the beginning of the study and were used for c-Fos studies. The second set weighed between 375 and 450 g at the beginning of the experiment and were used for the feeding studies. Animals were housed individually in suspended wire-mesh cages in a temperature- and humidity-controlled room under a standard 12:12-h light-dark schedule (lights on at 0700). Water and rodent chow (Teklad, Indianapolis, IN) were available ad libitum except where noted.
Rats (first group of rats; c-Fos studies) were implanted under halothane anesthesia (1–2%) with a chronic stainless steel guide cannula aseptically into the right lateral ventricle of rats (n = 6). The guide cannula was placed according to the Paxinos and Watson (19) stereotaxic coordinates (AP −0.34, ML ± 1.0, DV +2.0). After surgical implantation and appropriate wound closure, rats were housed in individual Plexiglas recording cages placed in environmentally controlled chambers (Tech/Serv model EPC-010). Animals were maintained at a 12:12-h light-dark cycle at an ambient temperature of 25 ± 1°C throughout the 10-day recovery period and the experiment. Six days after surgery, the patency and free drainage of the guide cannula were assessed in animals implanted with an icv cannula by microinjection of 5 μl of pyrogen-free isotonic NaCl (PFS; Abbott Laboratories, North Chicago, IL). After the 10-day recovery period rats were treated with 5 μg of NPW23 (2 nmol) in 5 μl of vehicle. Rats were anesthetized with isoflurane 1 h after the injection and perfused intracardially with 4% paraformaldehyde.
After a 2-wk acclimation period, rats (second group; feeding studies) were anesthetized with Nembutal (60 mg/kg ip) and stereotactically implanted with a 26-gauge guide cannula (Plastics One, Roanoke, VA). The coordinates used were derived from the atlas of Paxinos and Watson (19). PVN coordinates were 0.5 mm lateral, −1.9 mm AP, and −7.3 mm DV from bregma. Coordinates used for LH placement were 1.9 mm lateral, −2.2 mm A/P, and −7.2 mm DV to bregma. For both PVN and LH sites the injector tip extended 1.0 mm beyond the cannula tip. Two stainless steel screws and dental acrylic were used to secure the cannula to the skull surface. Rats were allowed 10 days to recover from surgery before first injection. Cannula placement was histologically verified; data from rats with misplaced cannulas were deleted from statistical analyses.
Drugs and injections.
NPW23 (>95% purity by C18 reverse-phase HPLC; NH2: WYKHVASPRYHTVGRASGLLMGL-AMIDE), synthesized by L. de Lecea, was diluted with 0.9% NaCl, further diluted, and aliquoted for future use. Peptide identity was assessed by mass spectrometry and amino acid analysis. NTX hydrochloride (Sigma, St. Louis, MO) was also diluted with saline but was prepared fresh each day just before injection.
Food intake study: dose-response effect of NPW23.
After the nocturnal feeding period, animals received NPW23 in a repeated-measures dosing design: 3, 1, 0.3, or 0.1 nmol or saline in 1 μl, injected into the PVN over 60 s, between 1000 and 1200. Food intake, corrected for spillage through the cage floor, was measured 1, 2, 4, and 24 h after injection. A minimum of 2 days was allowed between injections. LH-cannulated rats were tested similarly. Although NPW was administered in a relatively large amount of vehicle (1 μl), it is unlikely that the injection directed at the PVN diffused to the LH or vice versa, because the PVN and LH are fairly distant from one another.
Food intake study: effect of NTX on NPW23-induced food intake.
For the next study, PVN rats were injected with saline + 3 nmol NPW23, 79 nmol NTX + 3 nmol NPW23, or saline + saline in a repeated-measures design. NTX was injected over 60 s just before injection of NPW23; both were in 1-μl volumes, and food intake was measured at 1, 2, 4, and 24 h after injection. PVN rats were also deprived of food overnight and injected with 79 nmol NTX, and food intake was measured at 1, 2 and 4, h after injection.
Data were evaluated by ANOVA, and means were compared with Fisher's least significant difference test. Data are shown as means ± SE, and P < 0.05 is considered statistically significant.
c-Fos immunoreactivity was monitored 1 h after injection in two sets of animals: 1) a control group of rats receiving an icv injection of saline (n = 3) vs. 2) a treated group receiving 5 μg NPW23 icv (2-nmol dose; n = 3) as previously described (3). Briefly, free-floating sections (30 μm) were rinsed in 0.1 M PBS pH 7.4 and then treated 30 min in H2O2 (0.3%)-PBS solution. Sections were then incubated overnight at 4°C in blocking solution containing antisera directed against hypocretin-1 (goat polyclonal antibody raised against Orexin-A; Santa Cruz Biotechnology, Santa Cruz, CA) and c-Fos (rabbit polyclonal antibody; Oncogene Research Products, San Diego, CA) diluted 1:1,500 and 1:12,000, respectively. These sections were incubated for 45 min at room temperature in blocking solution containing a biotinylated donkey anti-rabbit IgG (1:500; Jackson Immunoresearch Laboratories, West Grove, PA) and were processed in an avidin-biotin-peroxidase complex (ABC) solution (Vector Laboratories, Burlingame CA) for another 45 min. Sections were rinsed three times in PBS, and c-Fos immunoreactivity could be visualized as a black or brown reaction product after 7-min detection in a 0.04% 3,3′-diaminobenzidine tetrahydrochloride solution containing 0.01% H2O2 with (double-labeled brain sections) or without (brain stem sections) 0.4% nickel-chloride. Brain sections were further incubated with a donkey polyclonal anti-goat antibody (1:250, Jackson Immunoresearch Laboratories) for 45 min and processed in the ABC solution for an additional 45 min. Sections were analyzed by standard light microscopy with a Zeiss Axioplan microscope (Zeiss). The following criteria were used to estimate c-Fos density labeling: low, <10 c-Fos-positive cells/brain area; moderate, 10–30 cells/area; high, >30 cells/area.
The number of hypocretin-immunoreactive neurons displaying a concomitant c-Fos-immunoreactive nucleus was counted bilaterally in consecutive sections in a minimum of 12 different sections from 3 animals at the level of the perifornical area of the dorsolateral hypothalamus based on the Paxinos and Watson atlas (19). Data are expressed as means ± SE. NPW23 effect was analyzed by one-way ANOVA (P < 0.05).
NPW activates c-Fos expression in hypothalamus and several brain stem nuclei.
To determine the extent of neuronal activation after icv NPW injection, we used the early gene product c-Fos as an immunohistochemical marker. The distribution of c-Fos immunoreactivity was analyzed throughout the central nervous system. We only report regions for which immunoreactive cells were observed (Table 1). Rats injected with vehicle had only a few c-Fos-positive cells in the hypothalamic PVN and in the LH area. Injection of NPW increased the number of immunoreactive c-Fos nuclei in several brain areas. Numerous cells for c-Fos were detected in the hypothalamus, thalamus and brain stem. Indeed, c-Fos-positive cells induced by NPW injection were localized predominantly in the LH area, arcuate nucleus, ventromedial and dorsomedial nuclei, as well as periventricular nucleus and PVN of the hypothalamus. NPW injection increased the number of c-Fos-immunoreactive nuclei in the thalamic centromedial and paraventricular nuclei. In the mesencephalon, c-Fos immunoreactivity was observed in the zona incerta, central gray, median and dorsal raphe nuclei, raphe pontis nucleus, as well as cuneiform nucleus and pedunculopontine nucleus. NPW injection also increased c-Fos-positive cells in the dorsolateral tegmental nucleus, locus coeruleus, dorsal parabrachial nucleus, lateral and medial superior olive nuclei, and medial vestibular nucleus. In addition, very few scattered c-Fos-positive nuclei were observed in layers II-III and V-VI of the cortex, in caudate putamen, globus pallidus, as well as central and basolateral nuclei of the amygdala.
The hypocretinergic system in the LH is a well-known modulator of the stability of the sleep/wakefulness cycle, activity, and food intake (13, 23, 26). To determine whether NPW activates hypocretinergic neurons, we carried out double staining with c-Fos and hypocretin. Analysis of double-labeled cells demonstrated that hypocretinergic neurons are activated on icv NPW injection (Fig. 1). Cell counting in the perifornical region of the LH area showed that the number of hypocretinergic neurons expressing c-Fos immunoreactivity was significantly increased by icv NPW injection (F(1,74) = 281.301; P < 0.0001; Fig. 1D).
Local injection of NPW in hypothalamus stimulates feeding.
Injection of NPW23 into the PVN increased feeding significantly 1, 2, 4, and 24 h after initial drug administration (Fig. 2). The highest dose (3 nmol) increased food intake more than fourfold by the second hour after NPW23 injection. At the 4 -h time point 0.1 nmol of NPW23 more than doubled food intake. Although intra-PVN injection of NPW23 did not significantly increase food intake during the 4–to 24-h time point, feeding was still significantly increased during the 0–24 h time period. Injection of 3 nmol NPW23 into the LH significantly increased feeding; however, lower doses had no effect (Fig. 3).
Intra-PVN injection of the opioid antagonist, NTX, which decreases feeding induced by a variety of peptides and food deprivation (16), failed to inhibit NPW23-induced feeding. However, this same dose of NTX injected into the PVN potently decreased deprivation-induced food intake (Table 2).
The present study demonstrates that icv injection of NPW induced neuronal activation in several regions of the central nervous system as revealed by the induction of c-Fos expression 1 h after peptide administration. The overall distribution of c-Fos immunoreactivity is consistent with the distribution of GPR7 receptors (15, 18). c-Fos-positive cells were localized mainly in hypothalamic nuclei (i.e., periventricular, paraventricular and arcuate hypothalamic nuclei, LH area, ventro- and dorsomedial nuclei) and brain stem, including raphe nuclei, locus coeruleus, and surrounding nuclei (Table 1). Indeed, Lee et al. (15) reported a very dense expression of GPR7 mRNA in the paraventricular, supraoptic, ventromedial, and dorsomedial nuclei of the hypothalamus. The regions activated on NPW injection are in accordance with nuclei that have been suggested to be involved in the regulation of energy balance, including the PVN and LH. Whether the recruitment of c-Fos-positive cells occurred through mono- or polysynaptic connections still remains to be evaluated.
NPW is a ligand for the orphan receptors GPR7 and GPR8 (GPR8 has not been found in rat brain; Ref. 15). Baker et al. (2) reasoned that because GPR7 receptors are expressed in the hypothalamus NPW might be involved in neuroendocrine systems. They found that icv administration of NPW23 elevated prolactin and corticosterone and lowered growth hormone circulating levels. They also noted that a 3-nmol icv injection of NPW23 increased food intake, an effect that lasted ∼2 h. Shimomura et al. (22) also reported an increase in food intake after icv injection of NPW23, but at the 10-nmol dose. We found that NPW23 increased feeding after injection into the PVN of the hypothalamus. As expected, lower doses of NPW23 than were administered icv increased feeding after PVN injection. At the 4-h time point 0.1 nmol of NPW23 increased food intake markedly. However, higher doses were necessary to increase food intake at earlier time points. For example, only the 3-nmol dose increased feeding 1 h after injection, and 1 and 3 nmol of NPW23 increased feeding 2 h after injection. Of note is the observation that 1 and 3 nmol of NPW23 increased cumulative feeding for up to 24 h. Although LH injection of NPW23 increased feeding, its effect only lasted for 4 h and only at the highest dose, 3 nmol, a dose that also increased feeding when administered icv (2). Also, the 3-nmol dose of NPW, when injected into the LH, decreased feeding at the 24-h time point. It is unclear why this occurred; however, high doses of neuropeptides can be aversive and result in a decrease in food intake. The opioid antagonist NTX had no effect on PVN NPW23-induced feeding but potently decreased deprivation-induced feeding. A similar observation has been described for neuropeptide Y (NPY)-induced feeding (12). However, injection of NTX into the hindbrain or into the periphery decreases NPY-induced feeding. Such a study with NPW23 will need to be conducted to thoroughly study whether opioid pathways are involved in NPW23-related feeding. The current results suggest that NPW23-induced feeding involves a different mechanism than feeding stimulated by energy needs.
Although the GPR7 receptor is present in both the PVN and the LH, we found that intra-PVN injection of NPW23 into the PVN was effective at lower doses than needed to stimulate feeding in the lateral ventricle or the LH. An interesting observation of the present study was the activation of hypocretinergic neurons in response to NPW injection. Hypocretins are peptides that have been shown to play a key role in arousal as well as in regulation of energy balance, and the effect of NPW on food intake could be mediated through the activation of the hypocretinergic system. Interestingly, hypocretin neurons are activated by stress and CRF can modulate the activity of these neurons, probably through CRF-R1 receptors (27). Furthermore, activation of the hypocretinergic neurons is impaired in CRF-R1-deficient mice after foot shock and restraint stress, suggesting that hypocretinergic neurons are downstream of CRF in the acute stress response (27). In addition, infusion of hcrt1 stimulates the HPA axis (9–11, 14, 20). As hypothesized elsewhere, hypocretinergic neurons may integrate information from multiple circuits, including CRF and NPW peptidergic systems, to provide a coherent output to stabilize arousal networks (23). Recent observations have shown that NPW stimulates corticosterone secretion (25), suggesting that some effects of NPW may be mediated by enhanced corticosterone release. Interaction of NPW with glucocorticoid-sensitive neurons in the hypothalamus may account for the discordant effects of NPW in other studies (17). Furthermore, the hypocretinergic system could be a component of the circuitry that contributes to a NPW-modulating effect, in particular on feeding behavior or in the context of acute stress response. The fact that we did not observe orexigenic effects in the LH could indicate that hypocretinergic neurons need coincident signals to trigger food intake. Interestingly, it has been suggested that distinct subpopulations of hypocretinergic neurons could be activated on different stimulation, which is consistent with the hypothesis of several functional subsets of hypocretinergic neurons (1, 6). Together, our data suggest a modulatory role of NPW on hypothalamic control of food intake.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society