| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Endocrine Service, Hospital Carlos III and 2Endocrine Service, Hospital Ramón y Cajal, Madrid, Spain
Submitted 17 April 2007 ; accepted in final form 2 August 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
growth hormone; ghrelin; growth hormone-releasing peptide-6; growth hormone secretagogue; aging; growth hormone decline
Early studies in elderly humans showed an age-associated attenuation of GH pulse amplitude and GH secretion in response to several stimuli, including insulin-induced hypoglycemia and arginine administration (32, 43). Subsequent studies revealed a decrease in plasma IGF-I parallel to the decline in GH pulses (51). The studies in humans have been confirmed in experimental animals, and it is now evident that the decline in high-amplitude pulsatile GH release and plasma IGF-I concentrations is one of the best characterized events that occur with aging in humans and rats (61). Thus the aging rat is a model of relative or partial GH deficiency and consequently is suitable for studying the mechanisms of this alteration.
GH has a broad range of actions and therefore its diminished secretion rate has clinical significance and may be responsible for the generalized catabolic state, functional alterations, cognitive impairment, and decreased neuroprotection associated with normal aging. In elderly adults, with or without GH defect, GH increases IGF-I, the lean-to-fat ratio, muscle strength, and skin thickness and reduces total body fat content (7, 42).
In previous studies, we observed that the decrease of GH gene expression and secretion, as well as the expression of GH target genes such as IGF-I and IGF-binding protein (IGFBP)-3, are the result of aging and not the increase in body weight that occurs with aging (59). Subsequently, we showed that aging is not a physiological situation of GH resistance because the peripheral alterations of the GH-IGF-I axis are reversible by repetitive exogenous GH administration (60).
The secretion of GH is regulated by the opposite actions of at least two specific neuropeptides, growth hormone-releasing hormone (GHRH) and somatostatin (SS); GHRH stimulates GH secretion, whereas SS inhibits it (33, 56). GHRH neurons are predominantly located in the arcuate nucleus and the SS neurons in the periventricular nucleus (18, 35).
Reverse pharmacology allowed the development of GH secretagogues (GHS). The GHS receptor (GHS-R) and the endogenous agonist ghrelin were subsequently identified (23, 26, 46). Ghrelin, synthesized in the stomach, anterior pituitary gland, and the hypothalamus, drives GH secretion via a novel mechanism that includes increasing GHRH release, amplifying GHRH signaling in somatotropes, reducing SS release, and antagonizing SS receptor signaling (46).
The fact that ghrelin-producing neurons are present in the arcuate nucleus and that GHS-R is expressed in the pituitary (23, 27) could be determinant for hypothalamic ghrelin playing an important role in the physiological regulation of GH via a novel cascade of the hypothalamic-hypophysial mechanism. There is in vivo evidence that GHS requires GHRH to fully stimulate GH release (33). A direct action of ghrelin on cultured somatotropes has also been shown. However, particularly in humans, the ability of ghrelin to directly influence GH secretion remains controversial.
It is very likely that changes occurring in the specific hypothalamic regulatory hormones are the primary mechanism in the age-related decline of pituitary GH secretion and gene expression (9, 31) that have been reported in the rat. Although a reduced hypothalamic SS gene expression has been found in old rats (47), most studies have shown increased secretion and therefore a higher somatostatinergic tone or action as the main mechanism of GH decline with age (49, 54). Several studies regarding hypothalamic GHRH have suggested a deficiency of the peptide in aging rats and humans (10). It has recently been reported that, in transgenic GHRH-green fluorescent protein mice, aging does not decrease GHRH but causes an enlargement of GHRH-positive axons suggestive of neuropeptide accumulation (2). It has been suggested, and confirmed in our laboratory, that GHRH-receptor (GHRH-R) mRNA levels in the pituitary decline with age (24), which is in agreement with the age-related decline of pituitary GHRH sensitivity, although there is no general consensus on this point (10, 50, 61). Because GHRH-R is upregulated by the hypothalamic GHRH (21), our previous findings suggest that a major event in the decline of GH might be the decrease of hypothalamic GHRH secretion.
Studies to determine whether ghrelin levels change during aging have been inconclusive. Both an age-related decline of plasma ghrelin concentrations in humans (40) and an increase of stomach ghrelin production and secretion in aging rats have been described (13). Plasma ghrelin slightly increased with age, and ghrelin mRNA levels were similar in brains from C57BL/6J male mice aged 2 and 28 mo (55). There are no data on the expression and function of hypothalamic ghrelin in aging.
Circulating IGF-I, a modulator of the GH response, also participates in the negative feedback regulation of GH, acting at the hypothalamus and the pituitary levels through an endocrine long loop feedback (3). The presence of IGF-I mRNA in the hypothalamus (36) suggests that hypothalamic IGF-I may be involved in GH feedback regulation and thus may function as a hypothalamic modulator of GH as previously shown (63). Some studies have shown alterations in hypothalamic IGF-I in GH deficiency situations such as aging (30) and hypophysectomy (63); however, the implication of hypothalamic IGF-I and its receptor in the mechanism of GH decline in aging is not clearly established.
Our recent observations seem to indicate that the GH decline is not because of a pituitary defect itself since the capacity to respond to some secretagogues is maintained in somatotropes from old rats, although the low magnitude of the response is proportional to the intracellular GH content (44). To further confirm this hypothesis, we have studied the alterations of hypothalamic GH regulators and the possible reversibility of pituitary age-related GH decline with secretogogues in vitro.
It is known that peripheral alterations of the GH-IGF-I axis are reversible by repetitive exogenous GH administration (60), and the capacity to respond to some secretagogues is maintained in somatotropes from old rats (44). The purpose of this study was to determine the underlying mechanisms responsible for the age-related decline in the GH/IGF-I system, specifically addressing the role played by hypothalamic GHRH and/or SS as well as ghrelin and IGF-I. To better understand the role played by the hypothalamic regulators, we performed in vitro studies where the response of somatotropes to GHS was analyzed in the presence and absence of hypothalamic cells, as well as in vivo experiments.
This study shows that pituitary aging-related GH decline may be reversed in vitro and in vivo by exposing the somatotropes to an appropriate secretagogues stimulus and suggests that alterations in GHRH may be a primary event in this decline.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
For the in vivo studies, 24-mo-old rats were treated with 20 µg/kg growth hormone-releasing peptide (GHRP)-6 every 12 h subcutaneously for 15 consecutive days. Control groups received the vehicle. The animals were killed by decapitation, and blood was collected and stored at –20°C until assayed for immunoreactive (IR) GH and IR-IGF-I. The hypothalami, pituitaries, and livers were removed under sterile conditions, rapidly frozen on dry ice, and stored at –80°C until extraction for radioimmunoassay and mRNA measurements. Five to six animals per group were used.
For the in vitro studies, the pituitary glands from 3-mo-old and 24-mo-old rats were removed under sterile conditions; the neurohypophyses were discarded, and the anterior pituitaries were collected and mechanicoenzymatically dispersed for preparation of primary pituitary cell cultures. Five experiments were conducted with two to three dishes per condition.
For tissue extract studies, tissue were collected and processed as indicated above. Experiments were done four times with three rats in each group. Experimental procedures were approved by the Institutional Committee of Research Ethics.
Reagents. Rat ghrelin and GHRP-6 were from Bachem (Torrance, CA). 125I-[32P]deoxy-CTP and [32P]ribo-UTP were from Amersham Pharmacia Biotech (Bucks, UK). Poly-L-ornithine and DNase I were from Sigma (St. Louis, MO). Papain was from Merck (Darmstadt, Germany). Neutral protease was from Roche Diagnostics (Barcelona, Spain).
Buffers and media. DMEM, FCS, horse serum, Hanks' balanced salt solution (HBSS), and PBS were from GIBCO (Invitrogen Life Technology, Paisley, UK). Penicillin-streptomycin and L-glutamine were from Bio Whittaker (Walkersville, MD). Pituitary defined medium consisted of DMEM (1 g/l glucose) supplemented with 1% BSA, 15 mM HEPES, 0.1 µM hydrocortisone, 0.5 nM triiodothyronine, 10 µM transferrin, 10 nM glucagon, and 0.2 nM parathyroid hormone from Sigma, 4 mM L-glutamine, and 100 U/ml penicillin-streptomycin. Hypothalamic defined medium consisted of DMEM supplemented with 15% FCS, 4 mM L-glutamine, 100 U/ml penicillin-streptomycin. Neurobasal medium consisted of 1x neurobasal medium (GIBCO) supplemented with 2% B27 (GIBCO), 0.5 mM L-glutamine, and 100 U/ml penicillin-streptomycin.
Cell culture and experimental design. Coculture of pituitary cells from the two groups of rats and fetal hypothalamic cells was made through a dual-chamber cell coculture system using the Transwell cell culture insert with pore size of 0.4 µm (Costar, Cambridge, MA), which allows free diffusion of macromolecules.
Pituitary cells from the two groups of rats were cultured as previously described (4). Anterior pituitaries were collected and mechanicoenzymatically dispersed with 0.1% papain, 0.1% neutral protease, and 0.1% DNase for 1 h at 37°C (Sigma). The dispersed pituitary cells were resuspended in defined medium containing 5% FCS, plated on poly-L-ornithine-coated culture plates (24-well), and seeded at a density of 4 x 105 cells/dish. Cultures were kept in a humidified atmosphere of 5% CO2-95% air at 37°C.
Fetal hypothalamic cells were cultured as previously described (16) using timed-pregnant Wistar rats raised in our laboratory. On day 17 of fetal life, the embryos were removed from the mother by aseptic surgical procedures, and the hypothalamic regions were dissected in PBS. All dissections of fetal tissue were performed using a binocular dissecting microscope. The diencephalon was first isolated by two transverse cuts: one was made anterior to the optic chiasm and the other through the midbrain. Then vertical cuts were used to remove lateral cortex and a horizontal cut to remove the dorsal thalamus. After being rinsed three times with Hanks' solution, the tissues were placed in DMEM-FCS-HBSS and dissociated mechanically by gently passing them through a Pasteur pipette and then through 20-, 21-, and 22-gauge needles. Dispersed cells were resuspended in defined medium and plated on the poly-L-ornithine-coated culture membrane of the Transwell inserts at a density of 4 x 104 cells/dish. Cultures were kept in a humidified atmosphere of 5% CO2-95% air at 37°C. After 5 h of incubation, defined medium was replaced by Neurobasal medium. Experiments were performed after 15 days in vitro, which is when we have observed that high levels of SS secretion and the peak of GHRH secretion occur (11, 14). Medium was changed every 3–4 days.
After 72 h of incubation in serum-supplemented defined medium (5% FCS), pituitary cells were preincubated in defined medium (serum free) for 2 h. Subsequently, the Transwell inserts with the fetal hypothalamic cells were placed in the 24-well plate containing the pituitary cells. The cocultures were stimulated with ghrelin in the absence or presence of SS antiserum for 4 h. Anti-SS14 IgG equivalent to 3 µl of antiserum/dish (binding capacity 0.2 µg/ml) was used. Controls received the same amount of IgG from normal rabbit serum (NRS) (15). Neither IgG from NRS nor SS antiserum affected pituitary cell cultures. Transwell inserts, containing only the medium, were placed in the control wells. No change was observed in GH secretion between pituitary cells cultured alone and pituitary cells cultured in the presence of a Transwell containing only the neurobasal medium. At the end of the experiment, media were removed, boiled for 5 min, and centrifuged for 30 min at 14,000 rpm. The supernatants were immediately frozen and stored at –20°C until being analyzed for GH concentration by RIA. In the studies of pituitary cell response to secretagogues, cells were preincubated in defined medium for 2 h, stimulated, and processed in the same conditions as the cocultures.
RIAs. IR-GHRH and IR-SS were measured in hypothalamic tissue extracts by previously described RIAs with specific GHRH and SS antisera raised in our laboratory as previously described (14, 15). The tissue peptide content was expressed as a function of the total protein determined by the Bradford method.
IR-GH in the pituitary culture media, blood, and pituitary tissues was measured by RIA using the National Pituitary Hormone Distribution Program rat hormone kit (National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD) with a sensitivity limit of 0.8 µg/l. GH values were expressed as percentage of the untreated 3-mo-old group. Serum IGF-I was measured by a commercial RIA (Nichols Institute, San Juan Capistrano, CA).
Tissue peptides were acid ethanol extracted as previously described (29). All parameters were measured in serum and tissues of individual rats, and all samples from the same experiment were measured in the same RIA to avoid interassay variations.
Western analysis.
Pituitaries were lysed in a buffer containing 50 mM NaCl, 0.01 M Tris·HCl, pH 7.6, 0.001 M EDTA, 0.1% Nonidet P-40, 1 µg/ml aprotinin, and 100 µg/ml phenylmethylsulfonyl fluoride. Total protein extracts (10 µg) were resolved by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. After the membranes were blocked, immunodetection was performed using rabbit anti-IGF-I-R
(1:2,000 dilution; Santa Cruz Biotechnology), followed by incubation with a goat peroxidase-conjugated anti-rabbit secondary antibody (1:1,000 dilution; DAKO, Glostrup, Denmark). Immunoreactive bands were visualized using an enhanced chemiluminescence detecting system (Amersham Biosciences, Little Chalfont, Bucks, UK). Quantification of the intensities of the autoradiography bands was done by densitometric scanning using Adobe-Photoshop 2.0 and NIH-Image 1.47 programs (Macintosh). The membranes were systematically treated with Coomassie brilliant blue R-250 solution (0.125%) for 1 h at room temperature to confirm equal loading.
Northern analysis.
Total hypothalamic and pituitary RNA was extracted using Trizol reagent (Life Technologies, Grand Island, NY) according to the protocol supplied by the manufacturer. Fractions of 2–15 µg were electrophoresed in a 1% agarose-0.66 mol/l formaldehyde gel, followed by electrotransfer to a nylon membrane (Nytran, Schleicher and Schuell, Keene, NH) and ultraviolet cross-linking (Hoefer Scientific Instrument, San Francisco, CA). The cDNA probe for SS and GH was labeled with deoxy-[32P]CTP (3,000 Ci/mmol). Prehybridization was carried out for 16 h at 42°C, and membranes were hybridized for 24 h at 42°C using UltraHyb (Ambion, Austin, TX) with an 
1 x 106 cpm/ml labeled antisense SS probe and 5 x 105 cpm/ml labeled antisense GH probe. Autoradiograms and quantification of intensities were done as described above. Equal loading was confirmed, and data were expressed as arbitrary units after making the correction for 18S.
Ribonuclease protection assay.
Total hypothalamic and liver RNA was extracted as previously described. In the ribonuclease protection assay, total RNA was hybridized overnight with 600,000 cpm of labeled antisense rat IGF-I and IGF-IR at 45°C. The hybridization solution contained 75% (vol/vol) formamide, 80 mM Tris·HCl, pH 7.6, 4 mM EDTA, 1.6 M NaCl, and 0.4% SDS. After hybridization, samples were digested using RNase A (40 µg/ml) and RNase T1 (2 µg/ml) for 1 h at 30°C. Protected hybrids were isolated by ethanol precipitation after phenol-chloroform extraction and separated according to size on an 8% polyacrylamide/8 M urea denaturing gel. Gels were exposed to X-ray film (Kodak, Cambridge, UK) at –80°C for 24–36 h. Quantification of the intensities of the autoradiogram bands corresponding to protected hybrids was done by densitometric scanning as described above. All samples were hybridized at the same time with 18S or cyclophilin to correct for the differences in gel loading.
Semiquantitative RT-PCR. Total RNA isolated from hypothalamic extracts was used as a template for GHRH, GHS-R1a, IGF-I, glyceraldehydes-3-phosphate dehydrogenase (GAPDH), and 18S. Briefly, total RNA (2 µg) was reverse transcribed using random primer and Superscript II RT (Invitrogen, Carlsbad, CA).
LightCycler. Sets of specific primers for rat GHRH, IGF-I, and GAPDH were designed using LightCycler Probe Design Software 1.0 (Roche Diagnostics, Mannheim, Germany) based on the sequence data of the gene available in GeneBank (Table 1). The sets of primers were used for LightCycler real-time PCR using SYBR Green I (Roche Applied Science, Indianapolis, IN). The reaction was performed in a total volume of 20 µl in microcapillary tubes, according to the manufacturer's instructions. Each reaction mixture contained 5 µl of cDNAs, 2 µl FasterStart DNA Master SYBR Green I mix, 2 µl of sense and antisense primers each (0.5 µM), 2.4 µl of 25 mM MgCl2, and 6.6 µl of PCR-grade H2O. The LightCycler programs for each gene were as follows: denaturation (95°C/10 min); PCR amplification and quantification (95°C/15 s, 58–60°C/5 s, 72°C/12 s) with single fluorescence measurement at the end of the elongation step, repeated for 40 cycles. The transition temperature at all steps was 20°C/s. To verify that only the specific product was amplified, a melting point analysis was conducted after the final cycle. Specific amplification was also confirmed by electrophoresis of PCR products on 2% agarose gel.
|
Taqman gene expression assays. To detect amplification of GHS-R1a and 18S, primers and probes from the predeveloped primer and probe lists of Applied Biosystems (GHS-R: Rn00821417_m1 and 18S: Hs99999901-s1) were used. PCR was performed in duplicate for each sample using 1 µl 10-fold diluted cDNA as template for GHS-Ra and 18S, 1x Real Master Mix (Eppendorf, Netheler, Germany), and 1x of Taqman Gene Expression Assays in a 20-µl reaction. The amplification was carried out in an ABI PRISM 7000 Sequence Detection System under the following conditions: 2 min 95°C, 40 cycles (15 s 95°C, 1 min 60°C).
The results are given as percentage of 3-mo rats after normalizing GHRH and IGF-I mRNA to GAPDH expression and GHS-R1a mRNA to 18S ribosomal RNA content. Relative target mRNA expression from old to young rats was calculated by using the 2 
method described by Livak and Schmittgen (28).
Immunocytochemistry. The animals were anesthetized with halothane inhalation and perfused transcardially with 250 ml heparin 0.0125% in 0.1 M phosphate buffer (pH 7.4) followed by 500–750 ml of a fresh fixative solution containing 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Brains were removed and cut into small blocks that were postfixed in the same fixative at room temperature for 4 h. The blocks were then transferred to a 30% sucrose solution in 0.1 M phosphate buffer and stored at 4°C overnight. After fixation and cryoprotection, 40 µm frontal-frozen serial sections from the level of –1.8 to –4.8 mm, relative to the bregma, according to the atlas of Paxinos and Watson (38) were obtained with a microtome (Crycut 1800; Leyca, Bensheim, Germany). Free-floating sections were collected in cryoprotectant (30% glycerol, 30% ethyleneglycol in 0.1 M phosphate buffer) and stored at –80°C until being processed. Specific polyclonal antisera raised in rabbits against rat GHRH and SS were used in this study (14, 15). All antisera were diluted in PBS containing 0.2% Triton X-100 (Sigma). The sections were washed several times in PBS and soaked in PBS containing 0.3% H2O2 to block endogenous peroxidase. The sections were then washed in PBS and preincubated in 5% normal goat serum for 1 h at room temperature before incubation with the primary antisera [anti-GHRH (1:1,000) or anti-SS (1:3,000)] overnight at 4°C. The remaining procedures were performed at room temperature. After several more washes in PBS, the sections were incubated 1:250 with a biotinylated goat antirabbit IgG (Dako) for 1 h, washed again in PBS, and incubated with an avidin-biotin-peroxidase complex (Vectastain ABC Elite kit; Vector Laboratories, Burlingame, CA) for 90 min. The immunoperoxidase activity was revealed by incubating the tissue sections in a histochemical medium that contained 0.03% 3,3'-diaminoenzidine tetrahydrochloride (Sigma) dissolved in PBS containing 0.01% H2O2 and 0.088% nickel(II) nitrate (Sigma) for the SS-stained slices. The sections were mounted on SuperFrost Plus slides (Menzel-Glaser), dehydrated through graded ethanols, cleared in xylene, and covered with DPX (BDH Laboratory Supplies, Poole, UK). Control sections were routinely processed by either omitting the primary antibody or by replacing it with an equivalent concentration of preimmune rabbit IgG. These sections showed no staining. Sections incubated with the primary antiserum preabsorbed with recombinant rat GHRH or SS for 24 h at 4°C showed no staining.
Statistical analysis. All data are expressed as means ± SE. Tests for significance between sample groups were performed with a two-tailed t-test. For multiple comparisons, ANOVA was used with the Fisher's test for post hoc comparisons. Differences were considered statistically significant if P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
These findings suggest that the dramatic reduction of GHRH gene expression and IR-GHRH content in the hypothalamus and in the median eminence may account for the decreased GH gene expression and secretion in old rats.
Hypothalamic SS mRNA and immunoreactive peptide, and median eminence IR-SS in aging rats. SS, the GH secretion inhibitor, was quantified in the hypothalamus of young and old rats for the better understanding of the mechanism of the reduction of GH secretion in aging. As shown in Fig. 2A, SS mRNA was decreased significantly in the hypothalamic extracts of old rats; it was 60% lower in the old rat group when compared with the young rat group (P < 0.001). IR-SS in hypothalamic extracts was also significantly decreased by sixfold in the old rats when compared with the young rats (Fig. 2B).
|
These results clearly indicate that a significant decrease of SS mRNA and IR-SS in the hypothalamus as well as a marked diminution of IR-SS in the median eminence occurred in the old rats. Consequently, it is most unlikely that the secretion inhibitor SS may account for the GH secretion decline in aging.
GHS-R expression in the hypothalamus of aging rats. Because the secretagogue ghrelin is mainly expressed and secreted in the stomach, we quantified the expression of the hypothalamic ghrelin receptor instead of the secretagogue itself as a way to determine the activity of ghrelin.
GHS-R mRNA was quantified by real-time RT-PCR in hypothalamic extracts of rats of different ages. In the old rats, a significant decrease in the levels of GHS-R mRNA was observed [78.4 ± 4.1% (P < 0.01)] compared with the young rats, as shown in Fig. 3.
|
IGF-I and IGF-I-R gene expression in the hypothalamus of aging rats. Although IGF-I is not considered a major regulatory signal of GH secretion, hypothalamic IGF-I and IGF-I-R gene expression was studied for the better understanding of the mechanism of the GH decline in aging. As shown in Fig. 4A, hypothalamic IGF-I mRNA levels, measured by real time RT-PCR, were significantly decreased in the old rats [58.8 ± 5.3% (P < 0.05)] compared with young rats. Significantly lower levels of circulating IR-IGF-I were also observed in the group of old rats. The decrease of circulating IR-IGF-I (see Fig. 8B) and hypothalamic IGF-I mRNA (Fig. 4A) suggests that IGF-I does not play a role in the mechanisms underlying the decline of GH in aging.
|
|
mRNA levels and content were similar both in young and old rats. Validation of coculture system as a model for the study of ghrelin action on pituitary cells. To analyze the somatotrope responsiveness to secretagogues in a more physiological environment, we established a model of coculture of pituitary and fetal hypothalamic cells, where interactions between the two cellular populations and certain treatments could be studied. To this end, hypothalamic cells from 17-day-old embryos were cultured for 15 days, the time at which we had observed the peak of GHRH and high levels of SS secretion (11, 14).
To characterize the behavior of the somatotropes in this in vitro model, the cumulative GH secretion of pituitary cells from 3 and 24-mo-old rats, in the absence or presence of fetal hypothalamic cells, was analyzed. As shown in Fig. 5, the presence of fetal hypothalamic cells decreased GH secretion with the same intensity in cells of both groups when compared with their own controls [3 mo: 37.1 ± 7.2, (P < 0.01); 24 mo: 45.1 ± 2.5, (P < 0.001)]. These data indicated that the neuropeptides secreted in the coculture exerted an inhibitory effect on basal GH secretion.
|
Responsiveness of aged pituitary cells to ghrelin in the coculture system. In previous studies, we observed that cultured aged pituitary cells, which have lower intracellular content and secretion of GH, retained their capacity to respond to secretagogues, albeit the magnitude of the response was proportional to the intracellular GH content (44).
To further assess the implication of the hypothalamic regulators in the age-related decline of GH, we studied the capacity of ghrelin to reverse the aged-related decline in GH secretion in the following two in vitro systems: the pituitary cell culture and the coculture model. Cultured pituitary cells from 3-mo- and 24-mo-old rats were treated with ghrelin for 4 h. As shown in Fig. 6A, ghrelin produced a significant increase of cumulative GH secretion in the cells from both groups; there was not any significant difference in the magnitude of the response of the two groups when it was compared with their respective basal values (3 mo: 178.2 ± 15.3; 24 mo: 174.1 ± 15.3).
|
Responsiveness of aged rats to GHRP-6 in vivo. The in vitro data from the coculture studies suggested that ghrelin completely reversed the aging decline in basal GH secretion in the presence of hypothalamic cells. To further investigate the role played by the hypothalamic regulators in the age-related GH decline, we carried out an in vivo study where the response to secretagogues in an aged hypothalamic environment was analyzed. Also the effect of secretagogues on the reversibility on the peripheral and pituitary age-related alterations of the GH/IGF-I axis was assessed. This study was preceded by a previous one where we evaluated the length of GHRP-6 treatment. The highest GH secretion was obtained 15 days after GHRP-6 treatment, which produced a significant increase of serum IR-GH (Fig. 7A) and pituitary GH mRNA (Fig. 7E) levels in young and old rats. There was not any significant difference in the magnitude of the response between the two groups when it was compared with their respective basal values [IR-GH: 3 mo 321.7 ± 112 (P < 0.05), 24 mo 304.1 ± 77.6 (P < 0.001); GH mRNA: 3 mo 149.8 ± 5.4 (P < 0.05), 24 mo 156.2 ± 11.7 (P < 0.05)]. As shown in Fig. 7C, GHRP-6 treatment had a different effect on pituitary IR-GH content in young and old rats. Although in young rats GHRP-6 caused a depletion, in the old rats GHRP-6 increased pituitary IR-GH content [3 mo: 58.6 ± 6.1, (P < 0.05); 24 mo: 205.7 ± 23.6, (P < 0.05)] when compared with their respective basal values. These results indicated a clear effect of GHRP-6 on GH synthesis and release in pituitaries from old rats.
|
To complete the study on the reversibility of the peripheral age-related alterations of the GH/IGF-I axis, we analyzed the effect of chronic treatment with GHRP-6 on serum IR-IGF-I and hepatic IGF-I mRNA levels in old rats. Treatment with GHRP-6 for 15 days did not change serum IR-IGF-I in young animals, whereas it produced a significant increase in old rats when compared with their respective basal values [IR-IGF-I: 3 mo 101.4 ± 2.7; 24 mo 171.9 ± 9.4, (P < 0.01)]; Fig. 8A). As shown in Fig. 8C, GHRP-6 treatment caused a significant increase of both a and b liver IGF-I mRNA transcripts in young and old rats. There was not any significant difference in the magnitude of the response of the two groups when it was compared with their respective basal values [IGF-I a: 3 mo 144.3 ± 20.5 (P < 0.05), 24 mo 139.5 ± 26.3 (P < 0.05); IGF-I b: 3 mo 155.2 ± 23.5 (P < 0.05), 24 mo 167.5 ± 31.2 (P < 0.05)].
When the effects of GHRP-6 treatment in aging rats were analyzed as a percentage of normal untreated young rats, a significant decrease of serum IR-IGF-I (Fig. 8B) and of both a and b liver IGF-I mRNA transcripts was observed in the old rats (Fig. 8D); [50.7 ± 2.1, (P < 0.001); 65.1 ± 7.2, (P < 0.01); 47.9 ± 3.9, (P < 0.01), respectively]. GHRP-6 treatment significantly increased serum IR-IGF-I; even more, for the first time we found that it significantly increased liver IGF-I mRNA in old rats up to levels no different from those observed in the untreated young rats (IR-IGF-I: 85.9 ± 2.1; IGF-I a: 93.8 ± 16.6; IGF-I b: 79.3 ± 13.4 vs. 3 mo). These results indicate that GHRP-6 rejuvenated the age-related alterations of the peripheral IGF-I system.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This study confirms that hypothalamic IR-GHRH levels dramatically decrease in aging rats as do GHRH mRNA levels, which result in lower GHRH accumulation in the median eminence. In agreement with other studies (47, 52), we found a drastic decrease of hypothalamic IR-SS and SS mRNA levels in old rats as well as a decreased SS immunostaining in the median eminence. Early passive immunization studies with SS antiserum (48) and reports showing increased hypothalamic SS levels in the aging rats (19) advocated that an increase in the somatostatinergic tone might be a key mechanism of the GH alterations in the aging rat. As a whole, our results support the hypothesis that a major mechanism of GH alterations in aging is a diminution of GHRH secretion rather than an increase of SS tone, as was suggested in earlier studies (65). Previously, some of these alterations in aging male rats (9, 31) had been partially shown. However, the present systematic study, which includes the analysis of most of the hypothalamic signals involved in the regulation of the GH, strongly suggests that the major event of the somatotropic axis decline in aging might be a decrease of the hypothalamic secretagogue GHRH. However, there is unequivocal evidence that, in aging, major alterations may be located at hypothalamic and suprahypothalamic levels. Changes that occur with aging in brain neurotransmitters, catecholamines, and acetylcholine and deficits in receptor sites, turnover, and interaction and transduction mechanisms are possibly instrumental in incurring changes in the functional activity of GHRH synthesizing neurons.
We also found a decrease in hypothalamic GHS-R mRNA levels in aging rats. It has been described that ghrelin acts on GH regulation at hypothalamic and pituitary levels, thus increasing GHRH release, amplifying GHRH signaling in the somatotropes, reducing SS release, and antagonizing SS receptor signaling (46). Because of the important role of ghrelin in GH regulation, interacting with both GHRH and SS, deficits in endogenous ghrelin signaling may contribute to the alterations of GH secretion in aging. Recent studies conducted in C57BL/6J male mice show that the levels of GHS-R mRNA in the pituitary and the whole brain do not decline with age. It is possible that GHS-R expression is differentially altered in specific brain centers and is masked by assaying whole brain. Some species differences appear to be involved in the effect of aging on the GH/IGF-I axis. In rats, the effect of aging on the GH/IGF-I axis has been shown to be dependent upon genetic background (55).
We observed a decreased IGF-I gene expression and no alteration of IGF-I-R mRNA levels and IGF-I-R protein content in the hypothalamus of aging rats. These findings in aging rats are in agreement with data from other animal models of GH and IGF-I deficiency, such as the hypophysectomized rat in which a diminished hypothalamic IGF-I gene expression was described (63), and a food-deprived rat in which hypothalamic IGF-I mRNA levels were decreased while IGF-I-R mRNA levels were unaffected (36). Because IGF-I inhibits GH secretion at the hypothalamic level, these results, together with the low circulating IGF-I, clearly exclude that changes in IGF-I may account for the decline of the GH in aging. Given that IGF-I is reduced at all levels, its neuroendocrine inhibitory action and the negative long loop feedback cannot account for the mechanism of the declined GH secretion.
Our results indicate that there is an alteration of the GH regulators in the aging rat. In spite of this, reports in humans have shown that chronic treatment with ghrelin mimetics restores GH secretion and pulsatile GH secretion in the elderly. Also, our previous in vitro studies (44) show the capacity of the old pituitary cells to respond to some GHS. For the better understanding of the role played by the hypothalamic GH regulators in the age-related GH decline, we performed in vitro studies where the response of somatotropes to GHS was analyzed in the presence and absence of hypothalamic cells, as well as in vivo studies where the hypothalamus-pituitary axis was intact. To this end, we set up a coculture system of pituitary cells and fetal hypothalamic cells that allowed us to study the somatotrope responsiveness to secretagogues in a more physiological environment where interactions between the two cellular populations under certain treatments can be studied. In this in vitro model, we observed that the presence of hypothalamic cells containing GH regulatory neuropeptides modified basal GH secretion with the prevalence of an inhibitory action. This inhibition was completely abolished with the blockade of SS with antiserum, thus indicating that the inhibitory effect was exerted by SS. This inhibition could be due to the fact that SS secretion by fetal hypothalamic cells in this in vitro model is 10 times greater than that of GHRH (11, 14). Although the profile on neuropeptides secreted by the fetal hypothalamic cells is probably different from that of young and old animals, the outcome of the in vitro experiments coincides with our in vivo studies.
This study also indicates that the presence of fetal hypothalamic cells decreased GH secretion with the same intensity in cells of both age groups. Early studies showed an increased sensitivity of old pituitary cells to SS (48, 54), and this was postulated as the mechanism responsible for the diminished GH secretion in aging. Neither our previous studies in pituitary cell cultures (44), those of others (12, 37), nor the data shown here support this hypothesis. One problem common to many studies on aging is that different animal strains are used and submitted to different nutritional conditions. It has been reported that moderate caloric restriction altered the subcellular distribution of SS mRNA and increased GH pulse amplitude in aged animals (53).
Regarding the response of GH to ghrelin in aging rats, there is no consensus on in vivo studies (45, 62), and there are no in vitro studies in aging. Our data indicate that the magnitude of the GH response to ghrelin is not significantly different between young and old pituitary cells when compared with their own basal secretion, either in pituitary cultures or in cocultures with fetal hypothalamic cells. When the response of GH to ghrelin was compared with the basal GH secretion of cells from young rats in the pituitary cell cultures, the levels of GH in the old pituitary cultures did not reach the basal levels observed in the young pituitary cultures. However, in the coculture model, the response of GH to ghrelin in old pituitary cells was similar to that found with cells from untreated young rats. These data indicate that the presence of hypothalamic cells is required for full biological activity of GHS as has been suggested (57). Ghrelin has been shown to stimulate GHRH release from hypothalamic explants (64). In addition, these results show an in vitro model where the lower GH secretion of pituitary cells from old rats is stimulated with ghrelin.
Data regarding the increase in circulating GH with chronic GHS treatment in the old rat are conflicting. Our in vivo study demonstrates that GH release in response to exogenous GHRP-6 was not significantly different in rats aged 3 or 24 mo, indicating that old rats maintain the sensitivity to GHS. Although these are single serum sample data, they are supported by the results on pituitary IR-GH and GH mRNA content, and serum IR-IGF-I and liver IGF-I mRNA. These findings agree with previous studies in aged freely moving rats (45) and in aged mice (55). A decline in GH secretion following administration of GHRP-6 during aging in female rats has also been reported (62). The discrepancy with our study could be related to differences in the experimental design since the rat strain, age, dose of GHRP-6, and length of the treatment were different. In rats, the effect of aging on the GH/IGF-I axis has been shown to be dependent upon genetic background (25). A diminished ghrelin-induced GH response in the old rat could be related to the decrease in hypothalamic GHRH, since an inefficient treatment with GHS has been observed when GHRH is absent (34, 39).
We present the first evidence demonstrating that the reduced concentration of pituitary IR-GH and GH mRNA exhibited by the old rats is increased by GHRP-6, indicating that GHRP-6 may induce GH synthesis in the aged somatotrope and consequently reverse the lower GH levels in aging. The response of GH mRNA to secretagogues is controversial. Chronic treatment of 18-mo Sprague-Dawley rats with hexarelin did not restore GH mRNA content in the pituitaries of old animals (5). GHRP-6 and ghrelin have been shown to induce activation of Pit-1 in cultured anterior pituitary cells from infant rats (22). In adult rats, pituitary GH mRNA does not change in response to ghrelin administration (8).
Pituitary IR-GH levels are differently affected by GHRP-6 treatment. Pituitary IR-GH content is decreased by GHRP-6 in young rats, whereas it is increased in old rats. This may be because of the aging alterations observed in the hypothalamic regulators. In fact, the diminution in SS could increase the pituitary GH response to secretagogues, and the diminished GHRH could impair GH secretion, resulting in increased pituitary GH content.
We also demonstrate that GHRP-6 stimulates IGF-I secretion in the old rats up to levels found in the untreated young adult rats, but does not modify it in the young rats.
These results agree with our previous studies which showed that, after GH administration IR-IGF-I, brain IGF-I mRNA and IR-IGF-I increased in aging rats but were not modified in young rats (29, 60). GH also increased the low brain SS levels in the aging rats but was noneffective in adult rats (29). The striking difference in the response of peripheral IR-IGF-I to GHRP-6 between young and old rats is compatible with the aging-related decline in IGFBP-3 (60). In addition, GHRP-6-induced GH could negatively regulate the activity of liver GH receptors.
For the first time we observe that it markedly increases hepatic IGF-I mRNA. Our data are consistent with previous studies in elderly humans where the treatment with secretagogues restored IGF-I levels (6). Others found that, in aging rats, the IGF-I response to GHRP-6 vanished with aging in spite of the presence of a GH response in aging rats induced by the peptide (45). Although no evidence exists on the action of GHS on hepatic IGF-I mRNA levels when adult male rats were treated with GHRP-6, IGF-I mRNA levels increased in the cerebellum, hypothalamus, and hippocampus (20).
In summary, this study confirms that, in the old male rat, the decline in GH secretion rate is because of a reduction of GH gene expression and GH reserve that may result from the reduction of GHRH gene expression and secretion and is not the result of an increased somatostatinergic tone or action. There is reduction of hypothalamic GHS-R gene expression in old rats that might partly be because of a decreased GHRH and that could determine a low stimulatory action of GHS released at this level. The feedback exerted by IGF-I is not altered in aging rats and is ruled out as the primary mechanism responsible for the GH decline of aging. Additionally, we demonstrate that the diminished GH secretion of pituitary cells from old rats could be rejuvenated, at least in our in vitro model of coculture with fetal hypothalamic cells, by pharmacological ghrelin administration. The in vivo study reported in this article indicates that long-term treatment with GHRP-6 rejuvenates the GH-IGF-I axis in the aging rats. The GH-releasing activity of GHRP-6 appears to be dependent on the integrity on the hypothalamic pituitary axis involving GHRH-secreting neurons. The age-related decrements in GH secretion result from inadequate stimulation, rather than from maladaptive changes in the mechanism of GH release/or intrinsic pituitary defect. In this context, GHS could be more appropriate to restoring GH secretion in aged individuals than GHRH with its major limitation of a short half-life. Recent studies with CJC-1295, a long-lasting human GHRH analog, have shown that once-daily administration of the GHRH analog normalizes growth in the GHRH knockout mouse (1). In humans, prolonged stimulation of GH and IGF-I has been shown with a single subcutaneous administration of CJC-1295 in normal subjects (58). This analog may represent a more valid and physiological approach than GHS, particularly when growth hormone deficiency (GHD) is due to hypothalamic dysfunction. The development of stable orally active GHRH analogs and drug delivery systems could offer new therapeutic perspectives for these drugs. However, GHS which are known to activate the endogenous GHRH system, may represent a more physiological therapy in situations where the integrity of the hypothalamus is partially preserved such as some short-stature pathologies, adult GH deficiency, and aging. At present, the oral bioavailability of the GHS makes them particularly convenient for clinical use in a number of potential therapeutic opportunities. Our results give experimental support to the concept of reversibility of the pituitary alteration of the GH/IGF-I system that occurs in aging.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Current address of M. García-San Frutos: Health Science Faculty, University Rey Juan Carlos, Alcorcón 28922, Madrid, Spain.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
