| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Molecular Neuroendocrinology, National Institute for Medical Research, London NW7 1AA, United Kingdom; Beth Israel Deaconess Medical Center, Division of Endocrinology, Harvard Medical School (N.B.), Boston, Massachusetts 02215; Department of Neurosurgery, Barts, and The London School of Medicine and Dentistry, Queen Mary, University of London (C.M.), London E1 4NS, United Kingdom; and Imperial College London, Department of Neuroendocrinology, Hammersmith Hospital (P.H.), London W12 ONN, United Kingdom
Address all correspondence and requests for reprints to: Prof. Iain C. A. F. Robinson, Division of Molecular Neuroendocrinology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, United Kingdom. E-mail: irobins{at}nimr.mrc.ac.uk.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although ghrelin and GHS can release GH directly from pituitary GH cells, their major effects are exerted in the hypothalamus, in part via the release of GHRH, as the levels of GHRH in hypophysial portal blood increase acutely after GHS injections (8). The full effects of GHSs on GH secretion require an intact GHRH axis (9, 10, 11). Although some GHRH neurons express GHSR, most GHSR+ cells in ARC express neuropeptide Y (NPY) and agouti-related peptide (AGRP) (12, 13, 14), which are more likely targets for the effects of ghrelin on food intake and metabolism (15, 16, 17, 18).
Although ghrelin and GHSs are powerful pharmacological agents for stimulating GH release, the physiological importance of the ghrelin/GHSR system for regulating GH remains unclear. Chronic GHSR activation leads to a paradoxical increase in fat accumulation despite increased GH release, and ghrelin-mediated increases in adiposity occur in GH-deficient animals (2, 19), suggesting that ghrelin plays a GH-independent role in regulating food intake and body composition, and deletion of the genes for ghrelin or the GHSR do not lead to noticeable changes in growth (20, 21).
To study the physiological role of the GHSR in activating GH release, we have generated transgenic mice with overexpression of GHSR in GHRH neurons in an attempt to increase GHSR signaling selectively in the GHRH/GH axis. Stable lines of GH-producing cells overexpressing human GHSR type 1A (hGHSR 1A) were generated, which showed enhanced basal and GHS-stimulated GHSR signaling. We then used a 38-kb rat GHRH cosmid promoter, previously shown to specifically target hypothalamic GHRH neurons (22, 23), to increase GHSR expression in these neurons in transgenic mice, and tested the effects on GHRH expression, GH production and release, growth, food intake, and fat accumulation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Measurements of phosphoinositol (PI) hydrolysis
PI hydrolysis was measured as described by Adams et al. (24). Briefly, 80% confluent GC cell cultures were incubated overnight in DMEM (Life Technologies Inc., Paisley, UK), containing 0.5% fetal calf serum and 5 µCi [3H]inositol (Amersham Pharmacia Biotech, Little Chalfont, UK). Cells were then washed in serum-free DMEM containing 10 mM LiCl and 10 mM inositol and incubated in triplicate wells with test substances for 2 h at 37 C, after which the medium was removed, and the cells were extracted in 3.3% perchloric acid. After the addition of 10 M KOH, the supernatants were applied to anion exchange columns (Dowex AG1-X8, Bio-Rad Laboratories, Hemel Hempstead, UK), and the PIs were eluted in 1 M ammonium formate. Membrane-bound PI was determined similarly by dissolving the cell remnants with 1 M NaOH and 1 M HCl. PI hydrolysis was expressed as: % free PI ÷ (% free + % bound PI) x 100.
Construction of a GHRH-GHSR transgene
We used a rat genomic GHRH cosmid vector with a unique MluI restriction site created in the 5'-untranslated region (5'UTR) of the GHRH hypothalamic exon 1 into which an MluI-linked hGH fragment had been cloned (22). To replace the hGH-coding sequences with GHS-R sequences, an MluI-linked fragment was generated with the hGHS-1A cDNA sequences directly flanked by short hGH 5' and 3'UTR sequences that give efficient expression and processing of transgene RNAs (22) and could be used subsequently to distinguish transgene hGHSR and endogenous mouse GHSR transcripts (23). This fragment was then inserted into the MluI site of the GHRH cosmid and packaged (Gigapack III XL, Stratagene, Amsterdam, The Netherlands) as previously described (22). The final cosmid insert was a 38-kb NotI fragment containing 16 kb of 5' and 14 kb of 3' rat GHRH genomic sequences driving expression of the hGHSR cDNA flanked by short 3'- and 5'UTR hGH sequences.
Generation of GHRH-GHSR transgenic mice
All animal experiments were carried out in accordance with the relevant institutional and national guidelines. The 38-kb cosmid insert was released by NotI digestion, purified, and microinjected into fertilized (CBa/CaxC57BL/10)F1 mouse oocytes, which were transferred into the oviducts of pseudopregnant recipients. Tail-tip DNA from the offspring was tested for the presence of the GHRH-GHSR transgene using PCR and Southern blotting.
PCR and Southern blotting
For PCR genotyping, three primers were used. Primers 1 (5'-AAC CAC TCA GGG TCC TGT GGA CA-3') and 2 (5'-CCG AGA ACT TTC ATC TTT CAG-3') amplified a 506-bp hybrid hGH/hGHSR fragment only present in the transgene, whereas primer 1 and a third hGH primer (5'-CCT CTT GAA GCC AGG GCA GGC A-3') amplified an endogenous 300-bp mouse GH product as an internal control. For Southern blotting, DNA was digested with BglII and probed with a full-length hGHSR probe after random-prime 32P labeling using standard procedures.
RT-PCR
RNA was extracted using TRIzol reagent (Life Technologies, Inc.), and 500 ng were transcribed with 200 U Moloney murine leukemia virus reverse transcriptase (Roche Diagnostics, Lewes UK) in 1x Moloney murine leukemia virus reverse transcriptase buffer (Roche Diagnostics) supplemented with 1 µg random primers (Life Technologies, Inc.), deoxy-NTPs (Amersham Pharmacia Biotech; 0.3 mM), 40 U ribonuclease (RNase) inhibitor (Promega Corp., Southampton, UK), and 5 mM dithiothreitol. The mixture was incubated at 37 C for 2 h, and cDNAs were amplified by PCR using appropriate primer pairs. For mouse GHRH these were: forward, TGTTGAGCCCGTTACCGACC; and reverse, TGTCAGCACCTTTGCCGC. For hGHSR, the primer pairs were: forward, TTCGTCAGTGAGAGCTGCACCTAC; and reverse, AAATATCGCCCTACGTGGAAGG. For controls, mouse ß-actin transcripts were amplified using the following primers: forward, TGTAACCAACTGGGACGATATGG; and reverse, GATCTTGATCTTCATGGTGCTAGG.
RNase protection assays (RPAs)
RPAs were performed using the RPA III kit (Ambion, Inc., Huntingdon, UK). [32P]UTP-labeled RNA probes were purified by gel electrophoresis and incubated (1 x 105cpm) with 10 µg hypothalamic RNA samples at 42 C overnight. After hybridization, samples were treated with RNase, and protected fragments were separated on 5% acrylamide gels. Gels were analyzed using ImageQuant (Molecular Dynamics, Sunnyvale, CA), and the amount of protected sample RNA was normalized to ß-actin RNA, measured by RPA in the same samples. Mouse GHRH probes were generated from IMAGE clone 1496474 (HGMP Resource Center, Cambridge, UK) as previously described (23).
In situ hybridization
Coronal frozen brain sections (12 µm) were thaw-mounted onto gelatin- and chrome alum-coated slides and stored at -70 C until use. Sections throughout the ARC were hybridized with full-length [35S]UTP-labeled antisense or sense riboprobes and exposed to x-ray films, all as previously described (25). Because rodent and hGHSR sequences are highly homologous, two sets of probes were used. To compare total GHSR expression between transgenic and nontransgenic brains, a riboprobe corresponding to a full-length rat GHSR receptor cDNA was used. To identify transgene transcripts specifically, we used an oligonucleotide probe corresponding to the 5'UTR sequence of hGH uniquely present in the transgene transcript.
Immunocytochemical detection of Fos protein
Ninety minutes after injection of GHS or saline, mice were terminally anesthetized with pentobarbitone (60 mg/kg, ip) and perfused transcardially with heparinized isotonic saline, followed by 4% paraformaldehyde in 0.1 M phosphate buffer (PB). Brains were incubated in the same fixative containing 15% sucrose, transferred to a 30% sucrose solution in PB overnight, and then stored at -70 C. Coronal sections (30 µm) were cut through the ARC, and every third section was collected into PB. Endogenous peroxidases were inactivated by incubating in PB containing 20% methanol, 0.2% Triton X-100, and 1.5% hydrogen peroxide for 15 min. Sections were then incubated with a rabbit polyclonal anti-Fos antibody (PC38; Merck Biosciences Ltd., Nottingham UK; 1:40,000 in 1% normal sheep serum/0.3% Triton X-100/0.1 M PB) for 24 h at 4 C. After washing, bound antibody was localized using a peroxidase-labeled antirabbit IgG (Vector Laboratories, Inc., Peterborough, UK; 1:200 for 2 h at room temperature) and visualized using a nickel-intensified diaminobenzidine reaction (26), giving a purple/black precipitate. For each brain the number of Fos-positive nuclei was counted blind and bilaterally on each section (15–20 sections/brain) for each region (arcuate, suprachiasmatic, retrochiasmatic, paraventricular, dorsomedial, and ventromedial nuclei; medial and ventromedial preoptic areas; and lateral and anterior hypothalamic area). The number of nuclei per section was averaged for each region of every brain, and the data for each treatment group were pooled and presented as nuclei per section per mouse.
Physiological studies in GHRH-GHSR transgenic mice
Plasma GH responses to GHSs were tested in groups of 5-month-old male or female GHRH-GHSR mice. Under anesthesia (60 mg/kg; Sagatal, Rhone Merieux, Harlow, UK), a jugular vein was catheterized, and blood samples (50 µl) were collected into heparinized tubes before and 5 min after iv injection of either 10–50 ng GHRH [hGHRH-(1–29)NH2] or 50–250 ng GH-releasing peptide-6 (GHRP-6; Ferring AB, Malmo, Sweden) in 50 µl PBS containing 0.05% BSA. After a 90-min recovery period, a second sampling/injection/sampling procedure was carried out. Samples were centrifuged, and the plasma was stored frozen for GH measurements.
High fat feeding
Groups of GHRH-GHSR transgenic and nontransgenic mice (3.5-month-old females; n = 6) were housed in groups and fed a normal (<4% fat) chow diet (3.4% fat, 18.8% protein, 3.7% fiber, 3.8% ash, and 60.3% carbohydrate; 15.6 MJ/kg; Special Diet Services, Witham, UK) or a fat-enriched (30%) diet (protein content maintained at 18.8%; gross energy content, 21.7 MJ/kg) for 2 months. Body weight and daily food intake were measured for 24 d, after which bilateral inguinal, ovarian, and renal fat depots and mesenteric fat were all dissected and weighed.
Chronic treatment with GHRP-6
Two groups of individually housed, 3- to 4-month-old female GHRH-GHSR and wild-type (WT) mice were injected sc twice daily with GHRP-6 (0.5 mg/kg·d in 100 µl saline) or saline vehicle for 3 wk. All mice were fed the 30% high fat diet, and body weight and food intake were recorded. Measurements of food intake in individual mice were also obtained at 1, 2, and 4 h after one of the injections of GHRP-6. At the end of the study, fat pad, muscle, and heart weights were recorded, and right tibial lengths were measured with calipers.
Anxiety and activity analysis of GHRH-GHSR mice
Male and female GHRH-GHSR and WT mice were tested on elevated plus maze (27, 28). The maze has two open arms and two closed arms, and the amount of time spent in the open arm is negatively correlated with anxious behavior. Mice were placed in the maze for 5 min, and the amount of time spent in the open arms was recorded. In a separate study activity was measured in an open field test by recording the number of quadrants entered within a 5-min period.
RIAs
Plasma samples or pituitary homogenates were assayed for mouse GH and mouse prolactin (PRL) contents by specific RIAs, using reagents supplied by Dr. A. L. Parlow (National Hormone and Peptide Program, NIH, Bethesda, MD). Pituitaries were homogenized in PBS and assayed at several dilutions. Plasma was assayed directly for GH; the limit of detection was 0.2 ng/ml.
Statistical analysis
Unless otherwise stated, results are the mean ± SE. For body weight data, a two-way ANOVA was performed, with time and treatment as independent variables, followed by Bonferroni or Newman-Keuls tests. In vitro data were analyzed by one-way ANOVA and t test. Nonparametric data were analyzed using Kruskal-Wallis and Mann-Whitney tests, with P < 0.05 considered significant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). RT-PCR analysis using primers for the hGHSR readily amplified the expected 495-bp transcript in GHSR+ GC cells (Fig. 1B), but did not detect hGHSR transcripts in untransfected GC cells. The membranes of these cells could be stained with an antibody against the C-terminal domain of hGHSR (not shown), suggesting that they were generating and translocating GHSR protein. To test whether the hGHSR was functionally coupled, two lines (no. 2 and 13) of GHSR+ GC cells and an untransfected GC cell line were incubated with or without GHRP-2 (100 nM) for 2 h, and effects on PI turnover were measured.
|
). In contrast, basal PI turnover was increased in the absence of ligand in both GHSR+ GC cell lines, and both showed a marked increase after GHRP-2 stimulation (Fig. 1C). These experiments were replicated four times with GHSR+ GC cells exposed to different doses of GHRP-2 (0.1–100 mM) and showed a consistently higher basal PI turnover in GHSR+ vs. untransfected GC cells (2.20 ± 0.15% vs. 1.5 0 ± 0.14%; P < 0.01; n = 9), increasing to 5.3 ± 0.34% upon GHRP-2 stimulation (P < 0.001 vs. basal), with a maximal response at 1 mM GHRP-2. Other peptidyl (GHRP-6) and nonpeptidyl (L-163,255) GHS ligands also increased PI turnover (P < 0.01 vs. basal) in these stable GHSR+ GC cell lines (data not shown).
GHRH-GHSR transgenic mice
From fertilized oocytes microinjected with the GHRH-GHSR construct (see Materials and Methods; Fig. 2A) and transferred into pseudopregnant recipients, 42 live pups were obtained, and their tail tip DNA was analyzed by PCR for the presence of the transgene (Fig. 2B). A founder pup with a transgene copy number approximately 8-fold greater than that of WT animals (as estimated by Southern blotting) was used to establish a line of GHRH-GHSR transgenic mice on a CBa/CaxC57BL/10 background. The mice were fully fertile, litter sizes were normal, and the line was maintained hemizygous to obtain equal numbers of WT littermate controls for physiological experiments.
|
). RPA and in situ hybridization with an oligonucleotide probe specific for GHRH-GHSR transgene transcripts confirmed the expression of GHRH-GHSR in transgenic, but not WT, littermates. No transgene expression was observed in a variety of other peripheral tissues (pancreas, stomach, pituitary, gut, spleen, kidney, liver, or heart; data not shown).
Mouse GHRH mRNA levels were measured by RPAs in hypothalamic extracts from WT and GHRH-GHSR mice. There was significantly higher GHRH expression in GHRH-GHSR transgenic mice than in WT littermate controls (4.2 ± 0.3 vs. 2.5 ± 0.5 arbitrary units normalized to actin; n = 4; P < 0.05; Fig. 2D).
Growth and pituitary GH and PRL contents in GHRH-GHSR mice
Male and female GHRH-GHSR transgenic mice were the same size as their WT littermates at weaning, but developed a slight growth acceleration postweaning (Fig. 3A). The difference became significant around 6 wk, but remained small (5–10%) and disappeared as the animals reached adulthood (weights at 230 d: male GHRH-GHSR, 40.9 ± 1.0 g; male WT, 39.7 ± 0.5 g; female GHRH-GHSR, 28.3 ± 0.9 g; female WT, 29.0 ± 1.8 g; P = NS). Pituitary GH and PRL contents were measured in groups of adult male and female GHRH-GHSR and WT mice (Fig. 3B). GH stores (micrograms per pituitary) were significantly higher in male, but not female, transgenic mice, whereas PRL stores (Fig. 3C) were indistinguishable between transgenic and WT animals.
|
. Basal GH levels were similar in GHRH-GHSR transgenic and WT mice, and GHRH injections elicited dose-related GH responses that were equivalent in both GHRH-GHSR transgenic and WT mice. In similar experiments performed with GHRP-6 injections, peak plasma GH responses were lower than in experiments with GHRH injections, but did not differ between GHRH-GHSR and WT mice (Table 1). Similar results were obtained in female mice (data not shown). These results suggested that overexpression of hGHSR in GHRH neurons did not confer increased responsiveness to acute injections of GH secretagogues.
|
). No differences were seen after saline or GHRP-6 injection in GHRH-GHSR vs. WT mice in any other brain region examined (see Materials and Methods).
|
To document these differences and their sensitivity to dietary fat, groups of adult GHRH-GHSR and WT mice were either maintained on their normal low fat (<4%) chow diet or switched to a diet enriched to 30% fat for 2 months, after which their fat pad weights were measured. Over this period, both GHRH-GHSR and WT mice gained weight on normal chow, but the weight gain was significantly less for the GHRH-GHSR mice vs. WT mice (Fig. 5). As expected, both groups of animals switched to the 30% fat diet gained significantly more weight than those remaining on normal chow, but the increase was more variable in the transgenic group, and the difference between the fat-fed groups was not statistically significant (Fig. 5A).
|
These data were confirmed in another experiment in which food intake and fat pad weights were measured in individually housed transgenic and WT littermates. As expected, all fat-fed mice had larger fat pad weights than chow-fed mice, but the difference was only significant for the WT animals (P < 0.05). Regardless of the diet, the GHRH-GHSR transgenic mice tended to have smaller fat pads than their WT littermates (Fig. 5, B–E), but the differences were only statistically significant between the chow-fed GHRH-GHSR and WT mice.
Effects of GHRP-6 treatment on GHRH-GHSR and WT mice fed a high fat diet
We next tested whether chronic treatment with a GHSR ligand would differentially affect food intake and/or fat accumulation in GHRH-GHSR and WT mice fed the same 30% fat diet. Accordingly, groups of 3- to 4-month-old female GHRH-GHSR or WT mice were individually housed, offered the 30% fat diet ad libitum, and injected twice daily with either GHRP-6 (0.5 mg/kg·d, sc) or saline.
All mice gained weight significantly over the course of the study (Fig. 6). Animals receiving GHRP-6 gained more weight than those receiving saline injections (P < 0.05), and the increases were comparable between transgenic and WT mice (Fig. 6). Acute food intake responses after GHRP-6 or saline injection showed no significant differences [food intake (grams) expressed as percent body weight: GHRP-6-injected GHRH-GHSR, 0.40 ± 0.09%; saline-injected GHRH-GHSR, 0.29 ± 0.19%; GHRP-6-injected WT, 0.36 ± 0.17%; saline-injected WT, 0.10 ± 0.06%]. Again, fat pads were significantly smaller in saline-treated GHRH-GHSR transgenic mice compared with WT controls (fat weight expressed as percent body weight: GHRH-GHSR, 5.60 ± 0.91%; WT, 11.61 ± 1.54%; P < 0.01). GHRP-6 treatment had no differential effect on fat pad weight in GHRH-GHSR mice, but fat pads in GHRP-6-treated GHRH-GHSR mice remained significantly smaller than those in GHRP-6-treated WT mice (GHRH-GHSR, 6.42 ± 0.77%; WT, 12.03 ± 1.54%; P < 0.01).
|
). Heart weight was greater in GHRH-GHSR mice than in WT mice both in absolute terms and when expressed as a percentage of body weight (0.70 ± 0.03% vs. 0.47 ± 0.02%; P < 0.01) and was increased by GHRP-6 treatment in WT, but not GHRH-GHSR, mice (Table 2). Gastrocnemius muscle weight was also increased in GHRH-GHSR mice compared with WT controls, both in absolute terms and as a proportion of body weight (GHRH-GHSR, 0.60 ± 0.01%; WT, 0.50 ± 0.02%; P < 0.01), but GHRP-6 treatment only marginally increased gastrocnemius weight in WT mice (P = 0.06) and not in transgenic mice (Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The hGHSR has been expressed in a number of heterologous cell lines and signals in some hGH-producing adenomas (24), but surprisingly little has been done with expression of GHSR in GH-producing cell lines, which would provide the most appropriate complement of downstream signaling and adaptor molecules. We isolated several stable hGHSR+ GC cell lines in which PI turnover was markedly stimulated by GH secretagogues that had no effect in the untransfected parent cell line. Interestingly, basal PI turnover was significantly enhanced in the absence of ligand in hGHSR+ GC cells. Assuming that this did not reflect autocrine production of some endogenous ligand, this suggested that overexpression of the hGHSR construct per se might be enough to increase basal GHSR signaling.
A recent study with overexpression of GHSR in other heterologous cell systems supports this idea, as it showed that high GHSR expression can induce signaling in the absence of ligand (32). GC cells release GH constitutively in culture, and we did not attempt to measure increased GH in response to GHS in vitro. However, in preliminary experiments these hGHSR+ GC cells were implanted into Wistar-Furth rats and produced a massive GH secretory response after iv challenge with GHRP-6 (our unpublished observations).
Encouraged by this increased basal signaling in GC cells overexpressing GHSR, we turned to an in vivo model. Microinjection of the GHRH-hGHSR transgene construct in oocytes enabled us to establish a line of transgenic mice with approximately 8-fold increased hGHSR copy number vs. mouse GHSR. In situ hybridization confirmed overexpression of the GHSR in the hypothalamic ARC, with many more intensely labeled cells compared with WT littermates. No expression of the hGHSR transgene was detected in any other tissue examined, other than the hypothalamic ARC, and previous studies have shown that this GHRH construct colocalizes transgene expression to GHRH neurons.
Overexpression of GHSR in GHRH neurons doubled ARC GHRH expression in GHRH-GHSR transgenic mice compared with WT littermates. Furthermore, pituitary GH, but not PRL, contents were elevated in male GHRH-GHSR transgenic mice, although not significantly so in female GHRH-GHSR mice. Increased GH should depress GHRH expression by negative feedback (33, 34), so these results strongly suggest that the increase in pituitary GH reflects a steady state up-regulation of GHRH output, which is known to directly stimulate GH synthesis and secretion (35).
An increased activity in the GHRH/GH axis could account for the small, but significant, increase in the postweaning growth rate in transgenic mice, their increased muscle and heart mass (despite chronic high fat feeding), as well as the reduced amount of body fat in adult mice, the latter most prominent in females. Enhanced basal activity in the GHRH/GH axis could be caused by the increased constitutive GHSR signaling we and others (32) observed in the absence of endogenous ligand, but could also reflect a greater responsiveness of GHRH neurons to circulating stomach-derived ghrelin (1) or from a hypothalamic source (7, 36).
However, we found no increased responsiveness to acute administration of GHSs, nor any selective increase in response to homologous vs. heterologous ligands; both GHRP-6 and GHRH elicited large GH responses, but these did not differ between transgenic and WT mice. Furthermore, the ARC cellular Fos responses to peripheral GHS administration (37) were blunted, rather than increased, in GHRH-GHSR transgenic mice.
What could explain this unexpected finding? Firstly, the major ARC cell type showing a Fos response to GHS injection is the NPY/AGRP cell line, whereas the GHRH cells targeted by our transgene are a much smaller proportion of Fos-responding cells (12, 38). It is possible that the increased GH release caused by chronic up-regulation of GHSR signaling in GHRH neurons could up-regulate somatostatin or down-regulate endogenous GHSR signaling in the NPY pathway, as NPY and GHRH expression are regulated in an opposite fashion by changes in GH status (39, 40). A reduction in Fos responses in GHRH-GHSR mice could reflect desensitization, because the Fos response to an iv bolus injection of GHRP-6 is lost after a prior continuous exposure to GHRP-6 (41), but such a desensitization should be restricted to GHRH neurons, whereas we found that the Fos responses in all ARC areas were reduced. A more speculative explanation is suggested from the recent study by Holst et al. (32), who have shown that the high basal signaling activity of overexpressed GHSRs is susceptible to silencing by inverse agonists. Whatever the mechanism, some relationship must exist between GHRH and other hypothalamic GHS-responsive neurons, because up-regulation of GHSR in the former leads to a reduction in GHS-induced Fos responses in the latter.
The faster postweaning growth rate, an increase in muscle and heart mass, and a reduction in fat are all consistent with an upward resetting of the GHRH-GH axis in GHRH-GHSR mice (42). The reduced adiposity was particularly notable in the females, which normally develop larger fat depots than males. Interestingly, the lean phenotype of GHSR mice persisted even on a fat-enriched diet, with GHRH-GHSR mice continuing to maintain a lower adiposity than WT mice.
There is abundant evidence linking GH with adiposity. Obesity is associated with reduced GH secretion and responsiveness in rodents (43, 44), GH deficiency promotes the accumulation of fat, which can be reversed by GH treatment, and GH hypersecretion reduces fat mass (45, 46, 47). Other factors, such as increased activity in the hypothalamo-pituitary-adrenal (HPA) axis, augmented by a high fat diet (48), could also contribute to increased body fat. GHSR ligands can transiently increase activity in the HPA axis (49, 50, 51), but this is unlikely to be mediated via GHRH neurons, to which GHSR overexpression is restricted in our mice. Anxiety-related behaviors are affected by changes in the HPA axis, high fat diet, and ghrelin (52), and GH secretagogues have been implicated in states of anxiety and wakefulness (52, 53, 54). However, GHRH-GHSR and WT mice showed no differences in anxiety or exploratory behaviors.
Long-term GHS and ghrelin treatments cause modest increases in body weight in a variety of rodent models (19, 55, 56, 57). Although the effects of GHS on body weight were initially attributed to their GH-releasing effects, it is now clear that a significant proportion of the weight gain is due to increased body fat and reflects GH-independent effects (2, 19). Effects of ghrelin on food intake and fat deposition probably involve changes in the activity of several hypothalamic circuits (58) involving NPY/AGRP-containing and proopiomelanocortin-containing neurons among others (7, 17, 18). We found that chronic GHS treatment increased body weight and some organ weights in both GHRH-GHSR transgenic and WT mice, but had no differential effect on daily food intake or fat accumulation, suggesting that the enhancement of GHSR signaling in GHRH neurons did not alter the overall responses to GHS treatment. This was to be expected because these responses are likely to be mediated by hypothalamic targets other than GHRH neurons.
Our aim was to focus on enhanced GHSR signaling in the GHRH-GH axis. Despite the large number of studies of the ghrelin/GHSR system, there are still many questions about its physiological role in relation to normal GH secretion. Administration of large doses of ghrelin and other GHS have impressive effects on the GH axis (59, 60, 61), but the link between circulating endogenous ghrelin and physiological GH release remains unclear (62, 63, 64). A hypothalamic ghrelin system has been described (7), but it remains to be established whether it activates GHRH neurons to release GHRH into portal blood (8). Preliminary reports from knockout experiments suggest that the GHSR system does not play an essential role in the GH axis (20), certainly not compared with the GHRH receptor or its ligand (65, 66).
Although increased expression of the GHSR in GHRH neurons appears to leads to an upward resetting of the GH axis in GHRH-GHSR mice, interpretation of their responses to exogenous GHS administration is complicated, because they still have a full complement of their endogenous GHSR in many different cell types, including hypothalamic NPY cells and pituitary GH cells. Although the phenotype of GHSR knockout mice is not dramatic, GH responses to ghrelin are clearly lost (20). It will thus be of interest to cross our GHRH-GHSR mice with GHSR-null mice to be able to evaluate the effects of ghrelin in their progeny, whose only GHSR signaling pathway will be confined to GHRH neurons.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
S.L. and N.B. contributed equally to this work.
Abbreviations: AGRP, Agouti-related peptide; ARC, arcuate nucleus; CMV, cytomegalovirus; GHRP, GH-releasing peptide; GHS, GH secretagogue; GHSR, GH secretagogue receptor; h, human; HPA, hypothalamo-pituitary-adrenal; NPY, neuropeptide Y; PB, phosphate buffer; PI, phosphoinositol; PRL, prolactin; r, rat; RNase, ribonuclease; RPA, ribonuclease protection assay; UTR, untranslated region; WT, wild-type.
Received November 6, 2003.
Accepted for publication December 23, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. A. Reed, S. C. Benoit, P. T. Pfluger, M. H. Tschop, D. A. D'Alessio, and R. J. Seeley Mice with chronically increased circulating ghrelin develop age-related glucose intolerance Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E752 - E760. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Liu, J.-P. Fortin, M. Beinborn, and A. S. Kopin Four Missense Mutations in the Ghrelin Receptor Result in Distinct Pharmacological Abnormalities J. Pharmacol. Exp. Ther., September 1, 2007; 322(3): 1036 - 1043. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. S. Farhy, C. Y. Bowers, and J. D. Veldhuis Model-projected mechanistic bases for sex differences in growth hormone regulation in humans Am J Physiol Regulatory Integrative Comp Physiol, April 1, 2007; 292(4): R1577 - R1593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C Carreira, J. P Camina, E. Diaz-Rodriguez, R. Alvear-Perez, C. Llorens-Cortes, and F. F Casanueva Adenosine does not bind to the growth hormone secretagogue receptor type-1a (GHS-R1a). J. Endocrinol., October 1, 2006; 191(1): 147 - 157. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Mano-Otagiri, T. Nemoto, A. Sekino, N. Yamauchi, Y. Shuto, H. Sugihara, S. Oikawa, and T. Shibasaki Growth Hormone-Releasing Hormone (GHRH) Neurons in the Arcuate Nucleus (Arc) of the Hypothalamus Are Decreased in Transgenic Rats Whose Expression of Ghrelin Receptor Is Attenuated: Evidence that Ghrelin Receptor Is Involved in the Up-Regulation of GHRH Expression in the Arc Endocrinology, September 1, 2006; 147(9): 4093 - 4103. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Veldhuis, J. N. Roemmich, E. J. Richmond, and C. Y. Bowers Somatotropic and Gonadotropic Axes Linkages in Infancy, Childhood, and the Puberty-Adult Transition Endocr. Rev., April 1, 2006; 27(2): 101 - 140. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and inserted into the first hypothalamic exon of the rGHRH gene in a 38-kb cosmid (exons shown as 
, not drawn to scale). B, Genotyping of GHRH-GHSR mice by PCR of tail DNA (F, forward primer; R, reverse primer; see Materials and Methods) generated an endogenous 300-bp product in all mice and an additional 506-bp product in transgenic (T), but not WT, mice. C, In situ hybridization of hypothalamic sections using a rGHSR riboprobe. Upper panel, Brightfield showing GHSR mRNA in the ARC (arrows) of WT and transgenic (T) mice. Lower panel, The same sections were dipped in photographic emulsion and analyzed by darkfield microscopy (magnification, x40) to show more highly labeled neurons (arrowheads) in the ARC of GHRH-GHSR transgenic mice. D, RPA for mouse GHRH mRNA (top panels) and ß-actin (bottom panels) in pooled hypothalamic extracts from WT and GHRH-GHSR transgenic (T) littermates.
) and female (
and 
) GHRH-GHSR transgenic (T) mice (
