Ghrelin, an Endogenous Growth Hormone Secretagogue, Is a Novel Orexigenic Peptide That Antagonizes Leptin Action Through the Activation of Hypothalamic Neuropeptide Y/Y1 Receptor Pathway

Seven-week-old male Sprague-Dawley rats (200-220 g) and 9- to 11-week-old spontaneous dwarf rats (SDRs) (100-110 g)(42,43)were purchased from Japan SLC (Hamamatsu, Japan). They were housed in a temperature-, humidity-, and light-controlled room (12-h light/12-h dark cycle) and allowed free access to standard rat food (CE-2,352 kcal/100 g;Japan CLEA, Tokyo), unless otherwise indicated. All experimental procedures were approved by the Kyoto University Graduate School of Medicine Committee on Animal Research.

Intracerebroventricular injection experiments. A stainless steel intracerebroventricular (ICV) cannula (outer diameter 1.09 mm) (Becton Dickinson, Sparks, MD) was implanted under anesthesia in the skull of rats 5-7 days before the injection experiments, using coordinates (6.5 mm anterior to the lambdoidal suture; ± 1.4 mm lateral to the midline; 4.5 mm from the dural surface)(24,44). The ICV cannula placement was confirmed in all rats by introducing Evans blue after the experiments. Only the animals that showed the correct ICV cannula placement were included in this study.

ICV injection of ghrelin in Sprague-Dawley rats and SDRs.For the ICV injection of human ghrelin alone (5-5,000 ng/10 μl/rat)(11), cumulative food intake was measured in Sprague-Dawley rats and SDRs during 4 h at the early light phase (10:00-14:00) after the ICV injection. It was reported that a single ICV injection of ghrelin at doses of 10-200 pmol/rat (∼33-660 ng/rat)stimulates pituitary GH release dose-dependently(45).

ICV co-injection of ghrelin and NPY Y1 receptor antagonist in Sprague-Dawley rats. To examine the involvement of NPY/Y1 pathway in ghrelin's orexigenic effect, we used an NPY Y1 receptor antagonist (J-115814)(Banyu Pharmaceuticals, Ibaraki, Japan). In vitro binding analysis revealed that J-115814 displaces [¹²⁵I]peptide YY binding to the cloned human and rat Y1 receptors with inhibitory constant (K_i)values of 1.4 and 1.8 nmol/l, respectively, whereas it shows low affinities for human Y2 (K>10,000 nmol/l), Y4 (K_i>620 nmol/l), Y5 receptors (K >6,000 nmol/l) (M. Hata, S. Mashiko, A. Ishihara, O. Okamoto, J. Haga, T. Ohe, T. Kanno, N. Murai, Y. Ishii, T. Fukuroda, T. Fukami, M. Ihara, unpublished observations). Feeding induced by NPY (5 μg/rat) was inhibited dose-dependently by ICV co-injection of J-115814 (10-100 μg/rat) (data not shown).

For the ICV co-injection of ghrelin and J-115814, cumulative food intake was measured in Sprague-Dawley rats for 4 h at the early light phase(10:00-14:00) after a single ICV co-injection of ghrelin and J-115814 (500 ng+ 10-30 μg/10 μl/rat) or vehicle (10 μl).

ICV co-injection of leptin and ghrelin in Sprague-Dawley rats. For the ICV co-injection of leptin and ghrelin, cumulative food intake was measured in Sprague-Dawley rats for a period of 4 h (19:00-23:00)at the onset of the dark phase after a single ICV co-injection of leptin and ghrelin (2 μg + 5-500 ng/10 μl/rat) or vehicle (10 μl).

Total RNA extraction and RNA analysis. Total RNA was extracted from the whole hypothalami (24) obtained from Sprague-Dawley rats 4 h after a single ICV injection of ghrelin and/or leptin or vehicle. A 232-bp rat NPY cDNA fragment was prepared by reverse transcriptase-polymerase chain reaction using two oligonucleotide primers(sense 5′-CTGTGTGGACTGACCCTCGC-3′ and antisense 5′-CATTTTCTGTGCTTTCTCTC-3′). Northern blot analysis of NPY mRNA was performed (24) using rat NPY cDNA as a probe. A human β-actin genomic probe (Wako Pure Chemical,Osaka, Japan) was used to monitor the amount of RNA in each sample. We confirmed that the intensity of hybridization signals shows a linear relation to the amount of total RNA used in Northern blot analysis (data not shown). The hybridization signal intensity was quantitated using an image analyzer BAS-2500 (Fuji Photo Film, Tokyo) and normalized for the β-actin signal intensity.

Statistical analysis. All values were expressed as means ±SE. Statistical significance of difference in mean values was assessed by Duncan's multiplerange test following one-way analysis of variance.

ICV injection of ghrelin in Sprague-Dawley rats. A single ICV injection of ghrelin (5-5,000 ng/rat) resulted in a significant and dose-related increase in cumulative food intake in 8-week-old Sprague Dawley rats compared with vehicle-treated groups during 4 h at the early light phase(Fig. 1A). Treatment with 5,000 ng ghrelin increased cumulative food intake approximately fourfold relative to vehicle-treated groups (5.32 ± 1.28 vs. 1.25 ± 0.32 g, n = 10 each, P < 0.01). In this study, no significant difference in cumulative food intake was noted between ghrelin- and vehicle-treated animals 24 h after the injection (data not shown).

ICV injection of ghrelin in SDRs. To explore whether ghrelin's orexigenic effect is mediated by increased release of GH, we examined the effect of a single ICV injection of ghrelin on food intake in GH-deficient SDRs. A single ICV injection of ghrelin at a dose of 500 ng/rat caused a significant increase in cumulative food intake in SDRs compared with vehicle-treated groups during 4 h after the injection (1.66 ± 0.08 vs. 0.70 ± 0.19 g, n = 8-10 each, P < 0.01)(Fig. 1B).

Effects of ghrelin on hypothalamic NPY mRNA expression. Northern blot analysis revealed that NPY mRNA expression is increased significantly in the hypothalamus obtained from rats that received a single ICV injection of 500 ng ghrelin compared with vehicle-treated rats 4 h after the injection(∼ 1.6-fold, n = 12 each, P < 0.05) (Fig. 2A and B).

ICV co-injection of ghrelin and Y1 receptor antagonist. To assess whether the orexigenic effect of ghrelin is mediated by NPY and, if so, to assess whether Y1 receptor is involved, we examined the effect of ICV co-injection of ghrelin and Y1 receptor antagonist in rats. After a single ICV injection of 500 ng ghrelin, cumulative food intake was increased by ∼400%relative to vehicle-treated groups (5.12 ± 0.38 vs. 1.25 ± 0.32 g, n = 10 each, P < 0.01)(Fig. 3). By co-injection of J-115814 at doses of 10 and 30 μg/rat, increased food intake by ghrelin was reversed in a dose-dependent manner. Treatment with 30 μg J-115814 reversed completely the ghrelin-induced increase in cumulative food intake relative to vehicle-treated groups (1.22 ± 0.71 vs. 1.25 ± 0.32 g, n = 10 each, P > 0.1). In this study, treatment with 30μg J-115814 alone did not affect food intake in rats(Fig. 3).

ICV co-injection of leptin and ghrelin. A single ICV injection of 2μg leptin resulted in a significant reduction of cumulative food intake in rats relative to vehicle-treated groups (1.03 ± 0.21 vs. 3.11 ±0.24 g, n = 12 each, P < 0.01)(Fig. 4). This is consistent with our previous report (24). Co-injection of ghrelin (5-500 ng/rat) reversed dose-dependently the inhibition of food intake by leptin. Treatment with 500 ng ghrelin reversed the leptin-induced inhibition of food intake (2.19 ± 0.23 vs. 1.03± 0.21 g, n = 12 each, P < 0.01). Co-injection of 500 ng ghrelin and 2 μg leptin decreased significantly food intake relative to that in vehicle-treated groups (2.19 ± 0.23 vs. 3.11 ± 0.24 g, n = 12 each, P < 0.05). In this study, a single ICV injection of 500 ng ghrelin increased cumulative food intake (5.07 ±0.37 vs. 3.11 ± 0.24 g, n = 12 each, P < 0.01)during 4 h at the onset of the dark phase.

Effects of ICV co-injection of leptin and ghrelin on hypothalamic NPY mRNA expression. Northern blot analysis revealed a significant reduction of hypothalamic NPY mRNA expression in rats that received a single ICV injection of leptin (2 μg/rat) relative to vehicle-treated groups 4 h after the injection (∼65%, n = 12 each, P < 0.05) (Fig. 5A and B). The leptin-induced decrease in NPY mRNA expression was reversed by ICV co-injection of ghrelin at a dose of 500 ng/rat (n = 12 each, P < 0.01) (Fig. 5A and B). Co-injection of 500 ng ghrelin and 2 μg leptin increased NPY mRNA expression relative to vehicle-treated groups (∼ 130%, n = 12 each, P <0.01). Treatment with 500 ng ghrelin alone increased NPY mRNA expression by 160% relative to vehicle-treated groups (Fig. 5A and B).

This study demonstrates that a single ICV injection of ghrelin, a newly discovered naturally occurring GHS, increases food intake in rats. This is consistent with previous reports that central administration of GHRP-6 and KP-102, both of which are synthetic GHSs, increases food intake and body weight in rats(12,13). Although originally isolated from the stomach, ghrelin also occurs in the hypothalamic arcuate nucleus(11). Furthermore, GHS-R is expressed in several hypothalamic nuclei critical for energy homeostasis, such as the arcuate nucleus, ventromedial hypothalamic nucleus, and paraventricular nucleus(3,18). These findings, taken together, suggest that ghrelin acts centrally as an orexigenic peptide. It has been reported that peripheral administration of GHSs can increase food intake and body weight in rats(15). Expression of GHS-R is also detected in peripheral tissues, such as the pituitary, pancreas, and renal pelvis(2,3,4). Currently, it is unknown that ghrelin, which is released from the stomach into the circulation, can affect feeding behavior centrally, peripherally, or both.

In this study, ghrelin can increase food intake in GH-deficient SDRs,indicating that ghrelin does not require GH for its orexigenic effect. This is consistent with the notion that the orexigenic effect of hexarelin analogs when injected centrally and peripherally does not involve GH release(15). It was reported that systemic administration of GHS can induce c-fos mRNA expression in the hypothalamic arcuate GHRH neurons(14,17),where it can stimulate the release of GHRH(46). Given that GHRH is a potent orexigenic peptide(47), it is interesting to speculate that ghrelin increases food intake via GHRH. However, the orexigenic effect of KP-102 is not blocked by GHRH antagonist(13). Thus, it is likely that ghrelin stimulates feeding behavior via mechanisms different from those of GHRH.

A previous study showed that GHS-R mRNA is expressed in a large population of hypothalamic arcuate neurons containing NPY(18). Indeed, systemic administration of GHS can induce c-fos mRNA expression in some of hypothalamic arcuate NPY neurons(17). These findings suggest that GHS can act through GHS-R that is expressed in hypothalamic arcuate neurons containing NPY. In this study, we demonstrate that a single ICV injection of ghrelin can increase hypothalamic NPY mRNA expression. Furthermore, the orexigenic effect of ghrelin is abolished by ICV co-injection of Y1 receptor antagonist. These observations suggest that ghrelin increases food intake at least partly through the activation of the hypothalamic NPY/Y1 pathway. It is interesting to speculate that other actions of ghrelin might be mediated by the hypothalamic NPY/Y1 pathway.

Evidence has accumulated indicating that NPY plays a role in leptin action(25,26,27,28). Hypothalamic NPY production is increased in leptin-deficient ob/obmice or db/db mice or fa/fa rats with leptin receptor mutation(25,48,49),and Y1 antagonist partially reverses the spontaneous feeding of fa/farats(32,34). It has been shown that leptin treatment can reduce the otherwise increased hypothalamic arcuate NPY mRNA expression during fasting(28,40),when it inhibits significantly fasting-induced refeeding(28,40,41). Furthermore, fasting-induced refeeding is severely affected in Y1-deficient mice (35). Thus, NPY, when induced in response to leptin signaling deficiency, may activate Y1 receptor,thus leading to a marked induction of feeding. It is also conceivable that the satiety effect of leptin is mediated at least partly through the inhibition of hypothalamic NPY/Y1 pathway. In this study, we demonstrate that the satiety effect of leptin is abolished by ICV co-injection of ghrelin, which indicates the antagonism of the satiety effect of leptin by ghrelin. It has been reported that the biologically active longest isoform of leptin receptor mRNA is expressed in ∼50% of hypothalamic arcuate NPY neurons(50), where GHS-R mRNA is mostly expressed (>90% of hypothalamic arcuate NPY neurons)(18). It is likely that ghrelin and leptin act on the same NPY neurons to regulate hypothalamic NPY production. Therefore, we postulate that ghrelin and leptin share the hypothalamic NPY/Y1 pathway as one of the downstream mechanisms; the antagonism of the satiety effect of leptin by ghrelin may be due to the ghrelin-induced activation of the hypothalamic NPY/Y1 pathway. Evidence has suggested that the Y5 receptor is also involved in the control of food intake(51,52). Further studies are needed to elucidate the role of the hypothalamic NPY/Y5 receptor pathway in orexigenic action of ghrelin.

In this study, although the leptin-induced decrease in hypothalamic NPY mRNA expression is completely abolished by ICV co-injection of ghrelin, the satiety effect of leptin is only partially reversed by ghrelin. These observations suggest the involvement of other orexigenic or anorexigenic systems in the antagonism of leptin action by ghrelin. Because Agoutirelated protein (AGRP), an endogenous antagonist of the hypothalamic melanocortin system (24), is expressed in the arcuate NPY neurons (53),it is interesting to see whether ghrelin can increase hypothalamic AGRP mRNA expression. Indeed, we have observed that hypothalamic AGRP mRNA expression is increased significantly in rats that received a single ICV injection of ghrelin compared with vehicle-treated rats (M.S., Y.O., K.N, unpublished data). These observations suggest that the orexigenic effect of ghrelin is mediated at least partly by increased production of AGRP, which should contribute to the inhibition of the hypothalamic melanocortin system. In this regard, it is interesting to know the role of the hypothalamicα-MSH/MC4-R/AGRP pathway in the orexigenic action of ghrelin.

In conclusion, we demonstrate that ghrelin is a novel orexigenic peptide that antagonizes leptin action through the activation of hypothalamic NPY/Y1 receptor pathway. This study suggests that ghrelin may play an important role in the regulation of energy homeostasis.

This work is supported in part by research grants from the Japanese Ministry of Education, Science, Sports, and Culture; the Japanese Ministry of Health and Welfare; the Yamanouchi Foundation for Research on Metabolic Disorders; the Foundation for Total Health Promotion; the Kato Memorial Trust for Nanbyo Research, the Inamori Foundation; the Naito Foundation; the Uehara Memorial Foundation; and “Research for the Future” (RFTF) of the Japanese Society for the Promotion of Science (JSPS-RFTF 96100204 and 98L00801).

We would like to thank S. Yura for discussions and M. Ihara and A. Kanatani for providing J-115814. The authors also acknowledge M. Nagamoto for technical assistance and Y. Isa and Y. Nakajima for secretarial assistance.