Abstract
We compared the central mechanisms of feeding suppression by the anorexigenic hormones neuromedin U (NMU) and neuromedin S (NMS) in rats. I.c.v. injection of either NMU or NMS dose dependently decreased 3-h food intake during the first quarter of a dark period. Pretreatment involving i.c.v. injection of a specific anti-NMS IgG blocked the suppression of food intake induced by i.c.v.- and i.p.-injected leptin, but anti-NMU IgG elicited no blockade. Quantitative PCR analysis revealed that i.c.v. injection of NMU or NMS caused a dose-dependent increase in CRH and proopiomelanocortin mRNA expression in the paraventricular nucleus (PVN) and arcuate nucleus (Arc) respectively. In tissue cultures of the Arc, secretion of α-melanocyte-stimulating hormone was stimulated by NMU and NMS, with more potent stimulation by NMS. The time-course curves of the increase in neuronal firing rate in Arc slices in response to NMU and NMS showed almost the same pattern, with a peak 10–15 min after treatment, whereas the time-course curves for the PVN slices differed between NMU and NMS. These results suggest that NMS and NMU may share anorexigenic effects, depending on physiological conditions.
Introduction
Neuromedin U (NMU) is a highly conserved brain-gut peptide that was first isolated from the porcine spinal cord (Minamino et al. 1985) and later from the brain, spinal cord, and intestine of other species (Domin et al. 1986, 1989, O'Harte et al. 1991, Austin et al. 1995). Thus far, the only known physiological role of NMU is contraction of smooth muscles in blood vessels, the uterus, and the gastrointestinal tract (Minamino et al. 1985). Two G-protein-coupled receptors (NMUR1 and NMUR2) are thought to bind to NMU and mediate its physiological effects (Fujii et al. 2000, Hedrick et al. 2000, Hosoya et al. 2000, Howard et al. 2000, Kojima et al. 2000, Raddatz et al. 2000, Szekeres et al. 2000). NMUR1 (formerly FM-3/GPR66) is located in a wide range of peripheral tissues, such as intestine, testis, pancreas, uterus, lung, and kidney. In contrast, expression of NMUR2 (formerly FM-4/TGR-1) is limited to areas of the brain such as the paraventricular nucleus (PVN) along the wall of the third ventricle in the hypothalamus and the CA1 region of the hippocampus (Howard et al. 2000, Raddatz et al. 2000, Guan et al. 2001, Graham et al. 2003).
Neuromedin S (NMS), consisting of 36 amino acids, has been identified in the rat brain by a reverse-pharmacological technique as another endogenous ligand for FM-3/GPR66 and FM-4/TGR-1 (Mori et al. 2005, 2008). This neuropeptide is expressed mainly in the suprachiasmatic nucleus (SCN). Although NMS shares a C-terminal core structure (7-amino acid residues) with NMU and activates recombinant NMUR1 and NMUR2 expressed in Chinese hamster ovary cells with almost the same affinity as NMU, NMS is not a splice variant of NMU because the NMS and NMU genes have been mapped to discrete chromosomes (Mori et al. 2005, 2008). In addition, although NMU mRNA has been detected in peripheral and central organs (Howard et al. 2000, Yu et al. 2003, Ivanov et al. 2004), the distribution of NMS is limited to the testis, spleen, and SCN (Mori et al. 2005).
Evidence accumulated over the last decade has suggested that NMU and/or NMS are involved in the central regulation of feeding, energy homeostasis, stress, and circadian rhythms, probably through NMUR2 (Nakazato et al. 2000, Hanada et al. 2001, Ivanov et al. 2002, Nakahara et al. 2004a,b, Ida et al. 2005, Mori et al. 2005, Novak et al. 2006, 2007, Zeng et al. 2006, Novak 2009, Peier et al. 2009). I.c.v. administration of NMU or NMS suppresses both dark-phase food intake and fasting-induced feeding (Howard et al. 2000, Kojima et al. 2000, Ida et al. 2005). Transgenic overexpression of NMU promotes leanness and hypophagia in mice (Kowalski et al. 2005). Conversely, disruption of the NMU gene in mice (producing NMU KO mice) results in severe obesity by increasing food intake (Hanada et al. 2004). Central administration of leptin decreases body weight in NMU KO mice, suggesting that NMU-induced anorexia is independent of the leptin signaling pathway (Hanada et al. 2004). In contrast, immunoblockade of endogenous NMU by anti-NMU IgG partly inhibits leptin-induced suppression of feeding, suggesting that NMU is located downstream in the leptin signaling pathway (Jethwa et al. 2005). Although the reason for the discrepancy between these findings is unknown, the anti-NMU IgG used in the above immunoblockade experiment might have bound to NMS as well as to NMU because it would have recognized NMU-8, which shares an almost common C-terminal region with NMU and NMS (Wren et al. 2002, Jethwa et al. 2005, Mori et al. 2005).
Although the mechanism of NMU- and NMS-induced suppression of food intake is known to involve the hypothalamic anorexigenic peptides proopiomelanocortin (POMC) in the arcuate nucleus (Arc) or corticotrophin-releasing hormone (CRH) in the PVN, or both, several incongruous problems related to this food intake suppression remain to be solved. NMU KO mice develop obesity characterized by a decrease in the expression of mRNA for POMC in the Arc and CRH in the PVN. However, i.c.v. injection of NMU in rats does not affect POMC mRNA expression in the Arc but augments CRH mRNA expression in the PVN (Hanada et al. 2004). In contrast, i.c.v. injection of NMS into rats also induces anorexia by increasing the expression of POMC mRNA in the Arc and CRH mRNA in the PVN, and pretreatment with α-melanocyte-stimulating hormone (α-MSH) and CRH antagonists blocks NMS-induced suppression of food intake (Ida et al. 2005). Hanada et al. (2003) reported that CRH KO mice showed no reduction in food intake after NMU injection. Thompson et al. (2004) demonstrated that chronic administration of NMU into the PVN stimulated the hypothalamus–pituitary–adrenal axis but did not influence food intake or body weight.
Here, therefore, we compared the molecular and electrophysiological mechanisms involved in the effects of NMU and NMS on feeding suppression in the Arc and PVN, and we examined which of NMU and NMS was involved in leptin-induced suppression of food intake.
Materials and Methods
Animals
Male Wistar rats weighing 260–300 g were maintained in individual cages under controlled temperature (21–23 °C) and light (lights on from 0700 to 1900 h) conditions with food and water made available ad libitum. All procedures were performed in accordance with the Japanese Physiological Society's guidelines for animal care for studies involving the use of laboratory animals.
Preparation of anti-rat NMU and NMS antibodies
Polyclonal antibodies were raised against the specific N-terminal portions of rat NMU and NMS because the 7-residue C-terminal amidated sequences of rat NMS and NMU are identical (Mori et al. 2005). Antiserum was obtained by using a protocol reported previously (Hosoda et al. 2000). In brief, a synthetic peptide, (Cys0)-rat NMU (1–23) or (Cys0)-rat NMS (1–20), was conjugated to maleimide-activated mariculture keyhole limpet hemocyanin (Pierce, Rockford, IL, USA). New Zealand White rabbits were immunized by s.c. injection of these conjugates emulsified with Freund's complete adjuvant. Antibody specificity and titer were confirmed by RIA (Hosoda et al. 2000). Anti-rat NMU antibody did not crossreact with rat NMS, and anti-rat NMS antibody did not crossreact with NMU. Neutralizing activity was verified by calcium mobilization assay using Chinese hamster ovary cells stably expressing NMU or NMS receptor (Mori et al. 2005).
Feeding experiments
A stainless steel cannula was implanted into one of the lateral cerebral ventricles by a method that has been described previously (Ida et al. 2005). Each rat was sham injected with saline before the study and weighed and handled daily. Only animals demonstrating progressive weight gain following surgery were used for subsequent experiments. First, to compare the potencies of the acute suppressive effects of NMU and NMS on feeding, sham was injected separately at a dose of 0.1, 0.5, or 1.0 nmol/10 μl saline into the lateral cerebral ventricle of free-moving rats (n=6/group), with saline only as a control, via a 27-gauge cannula connected to a 50 μl Hamilton syringe at 1845 h. The 3-h food intake during the first quarter of the dark period (1900–2200 h) was then examined. Although we were unsure whether these doses were physiological, we did not use a dose exceeding 1.0 nmol in this study, as we had found beforehand that this produced abnormal behavior. Secondly, to determine whether NMU and NMS acted downstream of the leptin signaling pathway, rats were pretreated with i.c.v. injection of anti-NMU IgG (6.5 nmol) or anti-NMS IgG (5.5 nmol), with almost equivalent neutralizing activity, or with normal rabbit IgG (6.5 nmol) as a control, 30 min before leptin treatment. Leptin (i.c.v.: 3.5 or 10 μg/rat, i.p.: 1.0 or 3.0 mg/kg body weight) was administered i.c.v. or i.p. at 1845 h, and the 12-h food intake was then measured. Thirdly, to further confirm the effect of anti-NMU IgG pretreatment on the leptin-induced decrease of food intake, 13 nmol anti-NMU IgG or 13 nmol normal rabbit IgG were injected i.c.v. 30 min before leptin treatment. For this experiment, leptin was administered i.p. (3 mg/kg body weight) or i.c.v. (10 μg/rat).
Quantification of mRNAs
Each of six rats was killed by decapitation 2 h after i.c.v. injection of saline, 0.5 nmol NMU, or 0.5 nmol NMS. All experiments were performed twice using new rats (n=6/group). The Arc and PVN were collected by punch-out from frozen brain slices using a method described previously (Mori et al. 2005). Levels of expression of mRNAs for agouti-related peptide (AGRP) and POMC in the Arc and CRH in the PVN were measured by real-time quantitative PCR, as described previously (Nakahara et al. 2004a,b). AGRP mRNA was evaluated as one of the control for no influence by i.c.v. injection of NMU and NMS. Total RNA was extracted from each tissue by using an RNeasy Micro kit (Qiagen) and synthesized into first-strand cDNA by using an iScript cDNA Synthesis kit (Bio-Rad Laboratories). A single tissue sample was sufficient for measuring the mRNA levels. An aliquot of first-strand cDNA (40–100 ng tissue equivalent) was quantified on an iCycler (Bio-Rad Laboratories) by using iQ SYBR Green Supermix (Bio-Rad Laboratories) with primers to amplify glyceraldehyde 3-phosphate dehydrogenase (GAPDH), AGRP, POMC, and CRH specifically. The primer sets used for rat AGRP, POMC, and CRH, and GAPDH were as follows: AGRP: sense, 5′-TCTGAAGAAGACAGCAGCAGACCGA-3′, antisense, 5′-AGCGACGCGGAGAA CGAGACT-3′; POMC: sense, 5′-GACCTCACCACGGAA AGCAACCTG-3′, antisense, 5′-ACTTCCGGGGATTTTCAGTCAAGGG-3′; CRH: sense, 5′- ATCTCACCTTCCAC CTTCTG-3′, antisense, 5′-GTGTGCTAAATGCAG AATCG-3′, and GAPDH: sense, 5′-CGGCAAGTTCAACGGCACA-3′, antisense, 5′-AGACGCCAGTAGACTCCA CGACA-3′.
To examine whether i.p. administration of leptin affected the expression of NMS mRNA, hypothalamic levels of NMS mRNA were quantified 2 and 12 h after i.p. injection of leptin (3 mg/kg body weight) or saline, using real-time quantitative PCR by the method described above with some modifications. Each of six rats was killed by decapitation 2 and 12 h after leptin or saline treatment. Total RNA was extracted and synthesized into first-strand cDNA using a High-capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA). A single tissue sample was sufficient for measuring the level of mRNA. An aliquot of first-strand cDNA was quantified on a 7300 Real-time PCR System (Applied Biosystems) using TaqMan Gene Expression Master Mix (Applied Biosystems) with primers to amplify β-actin and NMS specifically. For these two genes, probe/primer kits were purchased from Applied Biosystems (TaqMan Gene Expression Assay ID: Rn00667869_m1, GenBank NM: NM031144 for β-actin and Assay ID: Rn02349491, GenBank NM: NM_001012233 for NMS).
Measurement of α-MSH secretion in Arc tissue cultures
Adult male Wistar rats were killed by decapitation at 1845 h. Each brain was quickly removed and cut with a Vibratome (Leica Microsystems GmbH, Wetzlar, Germany: VT 1000S) into 750 μm thick coronal slices including both sides of the Arcs. Using a Pasteur pipette with an inner diameter of 1.2 mm, the Arcs were dissected out from each brain slice under microscopic observation. Three Arc explants/well were incubated at 37 °C for 30 min under 95% air and 5% CO2 in each of 5 wells/group in 96-well plates with 200 μl oxygen-saturated DMEM (Gibco, Invitrogen) containing 10 mM HEPES, 2% FCS, and penicillin (100 U/ml)/streptomycin (100 μg/ml). Then the medium was changed to fresh medium with or without NMU or NMS at either 0.5 or 1.0 nmol/ml. After 1 h of incubation with slight shaking at 37 °C, all the culture medium was collected and centrifuged at 5000 g for 10 min at 4 °C. The supernatant was used to measure α-MSH with an ELISA kit (Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA).
In vitro recording of neuronal firing activity
In vitro multiple unit activity was recorded from the Arc and PVN in hypothalamic slice cultures plated on a microelectrode array dish (MEAD: Multi Channel Systems MCS GmbH, Reutlingen, Germany). A MEAD biosensor has a square recording area with sides 1000 μm long. Within this area, 60 electrodes are aligned in an 8×8 layout grid; the electrode diameter is 10 μm, and the inter-electrode distance is 100 μm. Each Arc and PVN slice was carefully cut with a respective thickness of 700 and 500 μm with a Vibratome slicer under cooling, and the upper surface of the slice was placed downward on the MEAD biosensor. The slice position was adjusted to ensure that the Arc and PVN were located over the electrode array and held lightly from above by mesh. DMEM containing 10 mM HEPES was then put carefully into the dish, and the biosensor was connected to a computer via an input amplifier and the MEAD connector housed on a custom-built microscope stage. The incubator consisted of a 16.5×16.5 cm steel base and an aluminum block through which the temperature was controlled (TC02, Multi Channel System MCSs GmbH). Records were taken corresponding to the site of each microelectrode. The neuronal firing activity frequencies were calculated at 30-s intervals and displayed on the computer screen using software developed by Multi Channel Systems MCS GmbH. After stable multiple unit activity recordings had been obtained, either NMU or NMS was added to the culture medium at a final concentration of 0.5 nmol/ml, and the recording was continued. After 20 min of incubation, the medium was changed to fresh medium without NMU or NMS to confirm the recovery of multiple unit activity.
Statistical analysis
The data (means±s.e.m.) were analyzed statistically by ANOVA with the post hoc Fisher's test at a significance level of P<0.05. In addition, the data (food intake, NMS mRNA, α-MSH secretion, and multiple unit activity) were evaluated using two-way ANOVA (NMU and NMS).
Results
I.c.v. injection of either NMU or NMS decreased 3-h food intake during the first quarter of the dark period in a dose-dependent manner (Fig. 1). Although NMS treatment elicited suppressive effects with smaller doses than NMU, there was no significant difference between NMS and NMU (F (1,30)=3.375, P=0.07).
Comparison of food intake in rats after i.c.v. injection of neuromedin U (NMU) and neuromedin S (NMS). Each bar and vertical line represents the mean±s.e.m. (n=6). Food intake of free-feeding rats was examined for 3 h during the first quarter of a dark period (from 1900 to 2200 h) in NMU- or NMS-treated rats. Various doses of NMU or NMS were injected at 1845 h. Asterisks indicate significant differences (P<0.05) between reagent-treated and saline-treated groups.
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
There were no significant differences in 12-h food intake between saline-treated rats preinjected i.c.v. with normal rabbit IgG, anti-NMU IgG, or anti-NMS IgG (Fig. 2A and B). I.c.v. injection of leptin caused a significant and dose-dependent decrease of food intake in rats that had been pretreated with normal rabbit IgG or anti-NMU IgG, but not with anti-NMS IgG. No significant difference was observed in these decreases between rats pretreated with normal rabbit IgG and those pretreated with anti-NMU IgG (Fig. 2A). I.p. injection of leptin at 3 mg/kg, but not at 1 mg/kg, also caused a significant decrease of food intake in rats pretreated with normal rabbit IgG or anti-NMU IgG, but not in rats pretreated with anti-NMS IgG (Fig. 2B). No significant difference in these decreases was observed between rats pretreated with normal rabbit IgG and those pretreated with anti-NMU IgG (Fig. 2B). As anti-NMU IgG did not suppress the leptin-induced decrease of food intake, we confirmed this using a larger dose of anti-NMU IgG. As shown in Fig. 2C, although i.c.v. pretreatment with a large dose of anti-NMU IgG in saline-treated rats resulted in a tendency for an increase of food intake during the dark period in comparison with pretreatment using the same dose of normal rabbit IgG, the difference between them was not significant. Pretreatment with anti-NMU IgG did not suppress the leptin-induced decrease in food intake because no significant difference was observed in these decreases between rats pretreated with normal rabbit IgG and those pretreated with anti-NMU IgG (Fig. 2C). The expression of NMS mRNA was higher at 12 h than at 2 h in both saline- and leptin-treated mice (F (1,20)=15.72, P=0.0008). On the other hand, there was no significant difference in NMS mRNA levels between leptin and saline treatment (F (1,20)=1.208, P=0.2848) (Fig. 2D).
(A and B) Effect of pretreatment with normal rabbit IgG (6.5 nmol), anti-NMU IgG (6.5 nmol) and anti-NMS IgG (5.5 nmol) on leptin-induced suppression of food intake. Specific antibody for NMU or NMS, or normal rabbit IgG as a control, was injected i.c.v. 30 min before administration of leptin. Leptin or saline was administered i.c.v. (A) or i.p. (B) just before the dark period, and the 12-h food intake was then measured. (C) Effect of i.c.v. pretreatment with normal rabbit IgG (13 nmol: white bar) and anti-NMU IgG (13 nmol: gray bar) on the leptin-induced suppression of food intake. Each bar and vertical line represents the mean±s.e.m. (n=6/group). Asterisks indicate significant differences (*P<0.05, ** P<0.01). (D) Effect of i.p. administration of leptin on the expression of hypothalamic NMS mRNA. Leptin (3 mg/kg body weight) was administered i.p. just before the dark period. Hypothalamic NMS mRNA levels were measured 2 and 12 h after leptin treatment (n=6/group). Asterisks indicate significant differences (*P<0.05).
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
Quantitative PCR analysis revealed that the expression of mRNAs for CRH and POMC was increased by i.c.v. injection of NMU or NMS (Fig. 3). There was no significant change in AGRP mRNA levels. Expression of POMC mRNA was significantly increased by NMU at 0.5 and 1.0 nmol and by NMS at a level as low as 0.1 nmol. CRH mRNA expression tended to be increased more by NMU than by NMS, but this difference was not significant.
Effects of i.c.v. injection of saline (white bar), NMU (gray bar), and NMS (black bar) on CRH mRNA expression in the paraventricular nucleus (PVN) and on AGRP and POMC mRNA expression in the arcuate nucleus (Arc). Samples were collected 2 h after i.c.v. injection. Each bar and vertical line represents the mean±s.e.m. (n=16). Asterisks indicate significant differences (P<0.05 versus saline-treated group).
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
When Arc explants were incubated with medium containing NMU and NMS, secretion of α-MSH was significantly increased by NMU at 1.0 nmol/ml or NMS at 0.5 or 1.0 nmol/ml, but not by NMU at 0.5 nmol/ml (Fig. 4). There was significant difference between NMS and NMU (F (1,20)=9.587, P=0.0057).
Effects of NMU and NMS on α-MSH secretion from cultured Arc explants. Each bar and vertical line represents the mean±s.e.m. Asterisks indicate significant differences (P<0.05 versus control group).
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
In vitro electrophysiological analysis using MEAD showed that NMU and NMS increased the number of spikes in both Arc (Fig. 5) and PVN (Fig. 6) slices. In each experiment, the effects of NMU and NMS on neuronal activity were evaluated by calculation of the average number of spikes at 5-min intervals from each electrode because all NMU- or NMS-responsive neurons do not respond with a similar temporal profile, and by excluding the data from the electrode showing no change in firing rate within 20 min after NMU or NMS treatment. When the average number of spikes per 30 s was calculated at 5-min intervals from each electrode, the first significant increase in neuronal activity in the Arc slice was observed during 5–10 min after NMU or NMS treatment, and the highest activity with both the treatments was observed between 10 and 15 min (Fig. 5). With both the treatments, the increase in the number of spikes was greatest near the outside of the Arc. There was no change in neuronal activity near the central area (data not shown). On the other hand, the first significant increase of neuronal activity in the PVN slice was observed at 0–5 min after NMU or NMS treatment, and the highest activity was observed at 5–10 min after NMU treatment and at both 0–5 and 15–20 min after NMS treatment (Fig. 6). As shown in Figs 5 and 6, there was no significant difference between NMU and NMS (Fig. 5, F (1,24)=3.236, P=0.08; Fig. 6, F (1,24)=0.01, P=0.94).
Effects of NMU and NMS on in vitro neuronal activity (multiple unit activity; MUA) in cultured Arc slices plated on a microelectrode array dish. (A) The photo represents the physical relationship between the electrodes on the dish and the hypothalamic slice (represented by hind gray shadow) including Arc (broad outline shown by the white curve). The insets are two samples of MUA recording for 20 min after NMS treatment. Each record corresponds to the sites of microelectrode shown in the white circles. The black arrow in each recording indicates the point of NMS treatment. Upper recording indicates an increase in firing rate, and lower one shows no change in firing rate. (B and C) Effects of NMU (gray bars) and NMS (black bars) on the total average number of spikes recorded at 5-min intervals in four individual experiments. In each experiment, the average spike number for each 5-min interval was first calculated, with the inclusion of all data that showed an increase in firing rate after NMU or NMS treatment. Then the total average spike number was obtained for the four experiments. Each bar and vertical line represents the mean±s.e.m. (n=4). Asterisks indicate significant differences (P<0.05 versus pretreatment).
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
Effect of NMU and NMS on in vitro neuronal activity (multiple unit activity; MUA) in cultured PVN slices plated on a microelectrode array dish. (A) The photo represents the physical relationship between the electrodes on the dish and the hypothalamic slice (represented by hind gray shadow) including PVN (broad outline shown by white curve) slice. The insets are two samples of MUA recording for 20 min after NMU treatment. Other details are the same as in Fig. 5. (B and C) Effects of NMU (gray bars) and NMS (black bars) on the total average number of spikes recorded at 5-min intervals in four individual experiments. Total average spike numbers were obtained as described in the caption to Fig. 5. Each bar and vertical line represents the mean±s.e.m. (n=4). Asterisks indicate significant differences (P<0.05 versus pretreatment).
Citation: Journal of Endocrinology 207, 2; 10.1677/JOE-10-0081
Discussion
We previously reported that i.c.v. injection of NMS decreased 12-h food intake during the dark period in rats, and that this anorexigenic effect was more potent and persistent than that observed with the same dose of NMU (Ida et al. 2005). In this study, i.c.v. injection of NMU or NMS decreased 3-h food intake during the first quarter of the dark period. This acute inhibitory effect of NMU and NMS was dose dependent and was more potent with NMS than with NMU because a smaller dose of NMS significantly suppressed food intake. These results, together with our previous observations (Ida et al. 2005), indicate that NMS-induced suppression of food intake is more potent both acutely and sustainably than NMU. As NMS and NMU activate recombinant NMUR1 and NMUR2 expressed in Chinese hamster ovary cells with almost the same affinity (Mori et al. 2005), the difference in their potencies of food intake suppression is not due to a difference in their affinities for NMUR1 and NMUR2.
We previously reported that the anorexigenic effect of NMU is independent of leptin in NMU KO mice because NMU reduces food intake in leptin-deficient mice (ob/ob mice) and leptin reduces intake in NMU KO mice (Hanada et al. 2004). However, leptin stimulates the secretion of NMU in hypothalamic explants in vitro (Wren et al. 2002). In addition, leptin-induced reduction of food intake is partly attenuated by i.c.v. injection of anti-NMU IgG (Jethwa et al. 2005). These latter results suggest that NMU may act partly downstream from the action of leptin. However, the anti-NMU IgG used in the previous reports (Wren et al. 2002, Jethwa et al. 2005) recognized the C-terminal regions of NMU, suggesting that this antibody binds to not only NMU but also NMS. If so, NMS also may act partly downstream from the action of leptin. To examine this possibility, we produced antibodies specific to NMU and NMS recognizing their N-terminal regions with little homology. The decrease of food intake induced by i.c.v. and i.p. treatment with leptin was blocked by pretreatment with the antibody specific to NMS but not by the antibody specific to NMU. This lack of effect in the latter case did not seem to be due simply to the dosage of anti-NMU because a large dose of anti-NMU IgG also failed to block it. These results suggest that NMS, but not NMU, acts downstream of the leptin signaling pathway.
When the hypothalamic expression of NMS mRNA was evaluated by real-time PCR, basal levels were significantly higher at 12 h (0700 h) than at 2 h (2100 h) after saline treatment, suggesting a circadian rhythm of NMS mRNA levels (Mori et al. 2005). On the other hand, there was no significant difference in NMS mRNA levels between leptin and saline treatment. Therefore, leptin might stimulate NMS release, but not synthesis. The source of the NMS released upon leptin stimulation was unclear. Granted that leptin stimulates the release of NMS from the SCN, there is no evidence for any leptin receptor in the SCN. Further studies will be required to elucidate the relationship between leptin and NMS or NMU.
The target of NMU and NMS for suppression of feeding seems to be the Arc and/or PVN. Radiolabeled NMU has been shown to bind to the Arc and PVN in vitro by receptor autoradiography (Mangold et al. 2008), and expression of NMUR2 mRNA has been confirmed in the PVN (Sakamoto et al. 2007). Therefore, it has been assumed that cFos expression in both nuclei after i.c.v. injection of NMU and NMS may be due to direct action on them (Ida et al. 2005).
Although the mechanism of NMU- and NMS-induced suppression of food intake is known to involve the hypothalamic anorexigenic peptides POMC in the Arc or CRH in the PVN, or both, several incongruous problems with regard to this suppression of food intake remain to be solved. Our quantitative analysis of mRNA expression and our in vitro electrophysiological analysis in the Arc and PVN demonstrated that both NMU and NMS increased POMC mRNA expression in the Arc and CRH mRNA expression in the PVN and increased the neuronal firing rates in both nuclei. Although it is not clear whether the increase in firing rate in the Arc and PVN in response to NMU and NMS resulted in the increase in mRNA expression in these nuclei, it is likely that a direct action of NMU and NMS on neuronal firing rate in the Arc and PVN and an increase in POMC and CRH mRNAs are involved in NMU- and NMS-induced suppression of food intake. In these analyses, however, we observed subtle differences in the effects of NMU and NMS. Expression of POMC mRNA was stimulated more by NMS than by NMU, and expression of CRH mRNA was stimulated more (although not significantly) by NMU than by NMS. Although the time-dependent pattern of the increase in neuronal activity caused by NMU in the PVN was similar to that in the Arc, the time-course patterns of the NMS-induced changes in neuronal activity differed between the Arc and the PVN.
Hanada et al. (2004) reported previously that i.c.v. injection of NMU in rats did not affect the expression of POMC mRNA in the Arc but augmented CRH mRNA expression in the PVN. The reasons for the differences in the observed effects of NMU on POMC mRNA expression between this study and the previous one are unknown. One possible explanation may be that the in situ hybridization method used in the previous study (Hanada et al. 2004) might have been incapable of detecting the small change in the level of POMC mRNA after NMU injection. Although the expression of POMC mRNA was increased by NMU at 0.5 and 1.0 (but not 0.1) nmol, NMS increased it at 0.1 nmol. In addition to these differences in POMC mRNA expression, α-MSH secretion from Arc explants in culture was significantly increased by NMU at 1.0 (but not 0.5) nmol, whereas NMS increased it at both 0.5 and 1.0 nmol. These results indicate that the more potent suppression of food intake by NMS than by NMU may be due to more potent stimulation of α-MSH synthesis and release in the Arc because CRH mRNA expression was not greater with NMS than with NMU. In this study, we do not know which is the primary instrument of NMU-induced suppression of food intake between α-MSH and CRH. It has been shown that food intake in CRH KO mice is not reduced after NMU injection (Hanada et al. 2003). In contrast, Thompson et al. (2004) have demonstrated that chronic administration of NMU into the PVN stimulates the hypothalamus–pituitary–adrenal axis but does not influence food intake or body weight.
Our in vitro electrophysiological analysis indicates that NMU and NMS directly stimulate the neuronal firing in the Arc and PVN. In the Arc, a significant increase in firing rate in response to both NMU and NMS started after 5 min of treatment. In the PVN, on the other hand, it was observed within 5 min after treatment. In addition to CRH neurons, the PVN includes vasopressin and oxytocin neurons. Both NMU and NMS stimulate the secretion of these hormones through direct action on NMUR2 in vasopressin and oxytocin neurons (Sakamoto et al. 2007, 2008). In this case, too, the action of NMS is more potent than that of NMU (Sakamoto et al. 2007, 2008). Therefore, the differences in the time course or strength of neuronal firing in response to NMU and NMS in the Arc and PVN might arise from differences in the populations of neurons responding to NMU and NMS in these nuclei.
In conclusion, we have demonstrated that NMU and NMS increase the expression of POMC and CRH mRNA in the Arc and PVN respectively and increase the neuronal firing rates in these nuclei with different potencies. We also showed that NMS, but not NMU, may act downstream of the leptin signaling pathway. These results may partly explain the difference in potency of feeding suppression between NMU and NMS in vivo, and suggest that NMS and NMU may share an anorexigenic action that is dependent on physiological conditions.
However, a paradox in the NMU and NMS observations remains to be explained: i.e., why the receptors for NMS and NMU are the same but the downstream mechanisms of feeding regulation by NMS and NMU are different? In addition, it is also unclear why NMU KO mice develop obesity (Hanada et al. 2004), whereas NMUR2 KO mice are lean and smaller than normal (Zeng et al. 2006, Novak 2009, Peier et al. 2009). Of course, as mentioned above, NMS may act on an undefined receptor other than NMU1R and NMU2R. Alternatively, NMU may act on an undefined receptor other than these two. Several studies have demonstrated, on the other hand, that the mRNAs encoding NMU, NMS, NMU1R, and NMU2R each have an intrinsic rhythmic expression in the SCN or hypothalamus with a different circadian pattern (Graham et al. 2003, Nakahara et al. 2004a,b, Mori et al. 2005, Jethwa et al. 2006). As the SCN sends neural projections into the PVN and Arc (Vrang et al. 1995, Saeb-Parsy et al. 2000), these different rhythmic expressions may be related to the different effects of NMS and NMU.
Recently, Egecioglu et al. (2009) demonstrated that long-term central NMU treatment reduced body weight, food intake, and adiposity in diet-induced obese mice, but not in lean mice fed a standard diet, and that female (but not male) NMUR2 KO mice fed a high-fat diet were protected from central NMU-induced body weight loss in comparison with wild-type littermates. In addition, Bechtold et al. (2009) have demonstrated that proNMU(104–136) is a novel modulator of energy balance since central administration of proNMU(104–136) caused a significant but transient (∼4 h) increase in feeding, although both food intake and body weight decreased over the following 24 h. Further studies will be required to elucidate the mechanism responsible for regulation of feeding and energy metabolism by NMU, NMS, and proNMU, including their receptors.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was supported in part by grants-in-aid from the Ministry of Education, Science, Sports, and Culture of Japan and by the Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN).
References
Austin C, Lo G, Nandha KA, Meleagros L & Bloom SR 1995 Cloning and characterization of the cDNA encoding the human neuromedin U (NMU) precursor: NMU expression in the human gastrointestinal tract. Journal of Molecular Endocrinology 14 157–169 doi:10.1677/jme.0.0140157.
Bechtold DA, Ivanov TR & Luckman SM 2009 Appetite-modifying actions of pro-neuromedin U-derived peptides. American Journal of Physiology. Endocrinology and Metabolism 297 E545–E551 doi:10.1152/ajpendo.00255.2009.
Domin J, Ghatei MA, Chohan P & Bloom SR 1986 Characterization of neuromedin U like immunoreactivity in rat, porcine, guinea-pig and human tissue extracts using a specific radioimmunoassay. Biochemical and Biophysical Research Communications 140 1127–1134 doi:10.1016/0006-291X(86)90752-7.
Domin J, Yiangou YG, Spokes RA, Aitken A, Parmar KB, Chrysanthou BJ & Bloom SR 1989 The distribution, purification, and pharmacological action of an amphibian neuromedin U. Journal of Biological Chemistry 264 20881–20885.
Egecioglu E, Ploj K, Xu X, Bjursell M, Salomé N, Andersson N, Ohlsson C, Taube M, Hansson C & Bohlooly-Y M et al. 2009 Central NMU signaling in body weight and energy balance regulation: evidence from NMUR2 deletion and chronic central NMU treatment in mice. American Journal of Physiology. Endocrinology and Metabolism 97 E708–E716 doi:10.1152/ajpendo.91022.2008.
Fujii R, Hosoya M, Fukusumi S, Kawamata Y, Habata Y, Hinuma S, Onda H, Nishimura O & Fujino M 2000 Identification of neuromedin U as the cognate ligand of the orphan G protein-coupled receptor FM-3. Journal of Biological Chemistry 275 21068–21074 doi:10.1074/jbc.M001546200.
Graham ES, Turnbull Y, Fotheringham P, Nilaweera K, Mercer JG, Morgan PJ & Barrett P 2003 Neuromedin U and neuromedin U receptor-2 expression in the mouse and rat hypothalamus: effects of nutritional status. Journal of Neurochemistry 87 1165–1173 doi:10.1046/j.1471-4159.2003.02079.x.
Guan XM, Yu H, Jiang Q, Van Der Ploeg LH & Liu Q 2001 Distribution of neuromedin U receptor subtype 2 mRNA in the rat brain. Brain Research. Gene Expression Patterns 1 1–4 doi:10.1016/S1567-133X(00)00002-8.
Hanada R, Nakazato M, Murakami N, Sakihara S, Yoshimatsu H, Toshinai K, Hanada T, Suda T, Kangawa K & Matsukura S et al. 2001 A role for neuromedin U in stress response. Biochemical and Biophysical Research Communications 289 225–228 doi:10.1006/bbrc.2001.5945.
Hanada T, Date Y, Shimbara T, Sakihara S, Murakami N, Hayashi Y, Kanai Y, Suda T, Kangawa K & Nakazato M 2003 Central actions of neuromedin U via corticotropin-releasing hormone. Biochemical and Biophysical Research Communications 311 954–958 doi:10.1016/j.bbrc.2003.10.098.
Hanada R, Teranishi H, Pearson JT, Kurokawa M, Hosoda H, Fukushima N, Fukue Y, Serino R, Fujihara H & Ueta Y et al. 2004 Neuromedin U has a novel anorexigenic effect independent of the leptin signaling pathway. Nature Medicine 10 1067–1073 doi:10.1038/nm1106.
Hedrick JA, Morse K, Shan L, Qiao X, Pang L, Wang S, Laz T, Gustafson EL, Bayne M & Monsma FJ Jr 2000 Identification of a human gastrointestinal tract and immune system receptor for the peptide neuromedin U. Molecular Pharmacology 58 870–875.
Hosoda H, Kojima M, Matsuo H & Kangawa K 2000 Ghrelin and des-acyl ghrelin: two major forms of rat ghrelin peptide in gastrointestinal tissue. Biochemical and Biophysical Research Communications 279 909–913 doi:10.1006/bbrc.2000.4039.
Hosoya M, Moriya T, Kawamata Y, Ohkubo S, Fujii R, Matsui H, Shintani Y, Fukusumi S, Habata Y & Hinuma S et al. 2000 Identification and functional characterization of a novel subtype of neuromedin U receptor. Journal of Biological Chemistry 275 29528–29532 doi:10.1074/jbc.M004261200.
Howard AD, Wang R, Pong SS, Mellin TN, Strack A, Guan XM, Zeng Z, Williams DL, Feighner SD & Nunes CN et al. 2000 Identification of receptors for neuromedin U and its role in feeding. Nature 406 70–74 doi:10.1038/35017610.
Ida T, Mori K, Miyazato M, Egi Y, Abe S, Nakahara K, Nishihara M, Kangawa K & Murakami N 2005 Neuromedin S is a novel anorexigenic hormone. Endocrinology 146 4217–4223 doi:10.1210/en.2005-0107.
Ivanov TR, Lawrence CB, Stanley PJ & Luckman SM 2002 Evaluation of neuromedin U actions in energy homeostasis and pituitary function. Endocrinology 143 3813–3821 doi:10.1210/en.2002-220121.
Ivanov TR, Le Rouzic P, Stanley PJ, Ling WY, Parello R & Luckman SM 2004 Neuromedin U neurones in the rat nucleus of the tractus solitarius are catecholaminergic and respond to peripheral cholecystokinin. Journal of Neuroendocrinology 16 612–619 doi:10.1111/j.1365-2826.2004.01210.x.
Jethwa PH, Small CJ, Smith KL, Seth A, Darch SJ, Abbott CR, Murphy KG, Todd JF, Ghatei MA & Bloom SR 2005 Neuromedin U has a physiological role in the regulation of food intake and partially mediates the effects of leptin. American Journal of Physiology. Endocrinology and Metabolism 289 301–305 doi:10.1152/ajpendo.00404.2004.
Jethwa PH, Smith KL, Small CJ, Abbott CR, Darch SJ, Murphy KG, Seth A, Semjonous NM, Patel SR & Todd JF et al. 2006 Neuromedin U partially mediates leptin-induced hypothalamo-pituitary adrenal (HPA) stimulation and has a physiological role in the regulation of the HPA axis in the rat. Endocrinology 147 2886–2892 doi:10.1210/en.2005-0983.
Kojima M, Haruno R, Nakazato M, Date Y, Murakami N, Hanada R, Matsuo H & Kangawa K 2000 Purification and identification of neuromedin U as an endogenous ligand for an orphan receptor GPR66 (FM3). Biochemical and Biophysical Research Communications 276 435–438 doi:10.1006/bbrc.2000.3502.
Kowalski TJ, Spar BD, Markowitz L, Maguire M, Golovko A, Yang S, Farley C, Cook JA, Tetzloff G & Hoos L et al. 2005 Transgenic overexpression of neuromedin U promotes leanness and hypophagia in mice. Journal of Endocrinology 185 151–164 doi:10.1677/joe.1.05948.
Mangold C, Ksiazek I, Yun S-W, Berger E & Binkert C 2008 Distribution of neuromedin U binding sites in the rat CNS revealed by in vitro receptor autography. Neuropeptides 42 377–386 doi:10.1016/j.npep.2008.04.008.
Minamino N, Kangawa K & Matsuo H 1985 Neuromedin U-8 and U-25: novel uterus stimulating and hypertensive peptides identified in porcine spinal cord. Biochemical and Biophysical Research Communications 130 1078–1085 doi:10.1016/0006-291X(85)91726-7.
Mori K, Miyazato M, Ida T, Murakami N, Serino R, Ueta Y, Kojima M & Kangawa K 2005 Identification of neuromedin S and its possible role in the mammalian circadian oscillator system. EMBO Journal 24 325–335 doi:10.1038/sj.emboj.7600526.
Mori K, Miyazato M & Kangawa K 2008 Neuromedin S: discovery and functions. Results and Problems in Cell Differentiation 46 201–212 doi:10.1007/400_2007_054.
Nakahara K, Hanada R, Murakami N, Teranishi H, Ohgusu H, Fukushima N, Moriyama M, Ida T, Kangawa K & Kojima M 2004a The gut–brain peptide neuromedin U is involved in the mammalian circadian oscillator system. Biochemical and Biophysical Research Communications 318 156–161 doi:10.1016/j.bbrc.2004.04.014.
Nakahara K, Kojima M, Hanada R, Egi Y, Ida T, Miyazato M, Kangawa K & Murakami N 2004b Neuromedin U is involved in nociceptive reflexes and adaptation to environmental stimuli in mice. Biochemical and Biophysical Research Communications 323 615–620 doi:10.1016/j.bbrc.2004.08.136.
Nakazato M, Hanada R, Murakami N, Date Y, Mondal MS, Kojima M, Yoshimatsu H, Kangawa K & Matsukura S 2000 Central effects of neuromedin U in the regulation of energy homeostasis. Biochemical and Biophysical Research Communications 14 191–194 doi:10.1006/bbrc.2000.3669.
Novak CM 2009 Neuromedin S and U. Endocrinology 150 2985–2987 doi:10.1210/en.2009-0448.
Novak CM, Zhang M & Levine JA 2006 Neuromedin U in the paraventricular and arcuate hypothalamic nuclei increases non-exercise activity thermogenesis. Journal of Neuroendocrinology 18 594–601 doi:10.1111/j.1365-2826.2006.01454.x.
Novak CM, Zhang M & Levine JA 2007 Sensitivity of the hypothalamic paraventricular nucleus to the locomotor-activating effects of neuromedin U in obesity. Brain Research 1169 57–68 doi:10.1016/j.brainres.2007.06.055.
O'Harte F, Bockman CS, Zeng W, Abel PW, Harvey S & Conlon JM 1991 Primary structure and pharmacological activity of a nonapeptide related to neuromedin U isolated from chicken intestine. Peptides 12 809–812 doi:10.1016/0196-9781(91)90138-F.
Peier A, Kosinski J, Cox-York K, Qian Y, Desai K, Feng Y, Trivedi P, Hastings N & Marsh DJ 2009 The antiobesity effects of centrally administered neuromedin U and neuromedin S are mediated predominantly by the neuromedin U receptor 2 (NMUR2). Endocrinology 150 3101–3109 doi:10.1210/en.2008-1772.
Raddatz R, Wilson AE, Artymyshyn R, Bonini JA, Borowsky B, Boteju LW, Zhou S, Kouranova EV, Nagorny R & Guevarra MS et al. 2000 Identification and characterization of two neuromedin U receptors differentially expressed in peripheral tissues and the central nervous system. Journal of Biological Chemistry 275 32452–32459 doi:10.1074/jbc.M004613200.
Saeb-Parsy K, Lombardelli S, Khan FZ, McDowall K, Au-Yong IT & Dyball RE 2000 Neural connections of hypothalamic neuroendocrine nuclei in the rat. Journal of Neuroendocrinology 12 635–648 doi:10.1046/j.1365-2826.2000.00503.x.
Sakamoto T, Nakahara K, Mori K, Miyazato M, Kangawa K, Sameshima H & Murakami N 2007 Neuromedin S exerts an antidiuretic action in rats. Biochemical and Biophysical Research Communications 361 457–461 doi:10.1016/j.bbrc.2007.07.036.
Sakamoto T, Mori K, Miyazato M, Kangawa K, Sameshima H, Nakahara N & Murakami N 2008 Involvement of neuromedin S in the oxytocin release response to suckling stimulus. Biochemical and Biophysical Research Communications 375 49–53 doi:10.1016/j.bbrc.2008.07.124.
Szekeres PG, Muir AI, Spinage LD, Miller JE, Butler SI, Smith A, Rennie GI, Murdock PR, Fitzgerald LR & Wu H et al. 2000 Neuromedin U is a potent agonist at the orphan G protein-coupled receptor FM3. Journal of Biological Chemistry 275 20247–20250 doi:10.1074/jbc.C000244200.
Thompson EL, Murphy KG, Todd JF, Martin NM, Small CJ, Ghatei MA & Bloom SR 2004 Chronic administration of NMU into the paraventricular nucleus stimulates the HPA axis but does not influence food intake or body weight. Biochemical and Biophysical Research Communications 323 65–71 doi:10.1016/j.bbrc.2004.08.058.
Vrang N, Larsen PJ, Moller M & Mikkelsen JD 1995 Topographical organization of the rat suprachiasmatic–paraventricular projection. Journal of Comparative Neurology 353 585–603 doi:10.1002/cne.903530409.
Wren AM, Small CJ, Abbott CR, Jethwa PH, Kennedy AR, Murphy KG, Stanley SA, Zollner AN, Ghatei MA & Bloom SR 2002 Hypothalamic actions of neuromedin U. Endocrinology 143 4227–4234 doi:10.1210/en.2002-220308.
Yu XH, Cao CQ, Mennicken F, Puma C, Dray A, O'Donnell D, Ahmad S & Perkins M 2003 Pro-nociceptive effects of neuromedin U in rat. Neuroscience 120 467–474 doi:10.1016/S0306-4522(03)00300-2.
Zeng H, Gragerov A, Hohmann JG, Pavlova MN, Schimpf BA, Xu H, Wu LJ, Toyoda H, Zhao MG & Rohde AD et al. 2006 NeuromedinU receptor 2-deficient mice display differential responses in sensory perception, stress, and feeding. Molecular and Cellular Biology 26 9352–9363 doi:10.1128/MCB.01148-06.
