| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
a Department of Internal Medicine, Division of Endocrinology and Metabolism, University of Michigan Health System and Department of Veterans Affairs, Ann Arbor
Ariel L. Barkan, Division of Endocrinology and Metabolism, 3920 Taubman Center, University of Michigan Medical Center, Ann Arbor, MI 48109-0354 E-mail: abarkan{at}umich.edu.
Decision Editor: William B. Ershler, MD
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. To study the contribution of aging per se, we evaluated discrete parameters of GH pulsatility in young (n = 8 women, n = 8 men) and elderly (n = 11 women, n = 10 men) subjects closely matched for body mass index. Blood samples for GH were obtained every 10 minutes for 24 hours. Plasma GH was measured by a sensitive chemiluminescent assay. GH pulsatility was assessed using cluster analysis.
Results. The elderly subjects had plasma IGF-I levels and integrated GH concentrations that were 32% to –56% of their sex-matched younger counterparts. The age-associated attenuation in GH was due to a decrease in GH pulse amplitude, whereas pulse frequency and nadir levels were unchanged. The majority of the young subjects (81%) reached their peak GH during the "lights off" period, whereas the majority of the elderly subjects (62%) peaked during the "lights on" period (p = .01).
Conclusions. We conclude that aging in both sexes is accompanied by profound decreases in GH output and in plasma IGF-I concentrations. This effect is separate from the alterations in body mass index that accompany the normal aging process. Attenuation of GH output associated with aging is related solely to the lower GH and, by inference, GH-releasing hormone (GHRH) pulse amplitude.
HUMAN aging is associated with a decline in growth hormone (GH) secretion (1)(2)(3). However, the specific contribution of aging per se to this phenomenon is still uncertain. Many aspects of human aging, such as an increase in relative adiposity (2)(4), decrease in gonadal steroid concentrations (3), alterations in sleep architecture (5)(6), and diminished physical fitness (4) have been shown to contribute independently to the low activity of the somatotropic axis.
Previous studies have assessed parameters of GH pulsatility in elderly and young subjects over wide ranges of body composition and gonadal steroid statuses, and the contribution of aging per se was derived from multiple regression analysis (4)(7). However, because body composition and gonadal steroid milieu are closely interrelated and interdependent, the specific contribution of each is difficult to discern. Moreover, accurate analysis of age-specific GH profiles has been limited by both GH assay sensitivity (3) and infrequent blood sampling (1)(3). To assess the contribution of aging per se, we studied parameters of GH pulsatility in selected groups of healthy and physically active young and elderly individuals who were closely matched for body mass index (BMI). Our earlier data (8) have shown that the lower GH output in elderly persons is due to the attenuated hypothalamic GH-releasing hormone (GHRH) release. This may be due either to less frequent occurrence of GHRH pulses (which should be manifest as decreased GH pulse frequency) or to lower quanta of GHRH released per secretory pulse (decreased GH pulse amplitude). Thus, accurate delineation of the aging-related alterations in GH pulsatility may shed light on the nature of the neuroendocrine effects of senescence in humans.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The study subjects consisted of four groups: young men (n = 8), young women (n = 8), elderly men (n = 10), and elderly women (n = 11). Detailed demographic information is shown in Table 1 . Special care was taken to select healthy, physically active subjects, with similar BMIs. Laboratory tests, medical history, and physical examination were unremarkable in all subjects. Young individuals were healthy and physically active, but did not participate in competitive sports. Elderly individuals were generally healthy, without evidence of cardiovascular, pulmonary, renal, gastrointestinal disease, or diabetes. None took medications known to affect the activity of somatotrophic axis. Some elderly subjects took multivitamins, low-dose aspirin, or cholesterol-lowering agents. All elderly subjects were physically active, and most engaged in regular physical activities (running, swimming, bicycling, rollerblading, aerobics, etc.). Subjects were all night sleepers. Young women were studied in the early follicular phase of the menstrual cycle as confirmed by menstrual diaries and by serum estradiol and progesterone levels. Elderly women (n = 4) who were on hormone replacement therapy stopped taking exogenous steroids for at least 2 weeks prior to testing.
|
Standard hospital meals were served at 7:00 AM, 12:00 PM, and 6:00 PM. No snacking was allowed between meals. Lights were turned off at 11:00 PM and on at 6:30 to 7:00 AM ("lights off " period), during which time the subject was required to sleep or rest in bed, except for trips to the bathroom. Napping was not allowed between 7:00 AM to 11:00 PM ("lights on" period).
Plasma GH was measured in duplicate by a chemiluminometric assay (Nichols Institute Diagnostics, San Juan Capistrano, CA). All samples from each subject were measured in the same assay to avoid interassay variability. The sensitivity of the assay was 0.01 µg/l. The mean intrassay coefficient of variation (CV) was 9% between 0.01 and 0.1 µg/l and 5% between 0.1 and 40 µg/l. The interassay CV was 7% at 9 µg/l.
Plasma insulin-like growth factor-I (IGF-I), IGF binding protein-3 (IGFBP-3), testosterone, and estradiol were measured from a sample created by pooling equal amounts of plasma samples obtained every 4 to 6 hours from each subject. Total IGF-I was measured after acid-ethanol extraction by an immunoradiometric assay (Diagnostics Systems Laboratories, Webster, TX) with sensitivity of 0.03 µg/l. IGFBP-3 was measured using radioimmunoassay kits (Diagnostics Systems Laboratories, Webster, TX) with sensitivity of 0.05 µg/l. Plasma estradiol and testosterone were measured using commercial kits (Coat-A-Count, Diagnostics Products Corp, Los Angeles, CA) with sensitivities of 8 ng/l and 0.04 µg/l, respectively. Samples with measured concentrations below the sensitivity of the assay were assigned the value of the assay sensitivity.
Maximum and minimum GH (µg/l) were defined as the highest and lowest GH concentrations measured during the 24-hour period for each subject. Nadir GH (µg/l) was determined as the average of the lowest 5% (7 points) of the GH values measured during the 24-hour period. Integrated GH concentration (IGHC; µg x min/l) was defined as the area under the GH versus time curve for each individual time period using trapezoidal calculations.
Parameters of GH pulsatility were analyzed by Cluster analysis (version 6.00) (10) utilizing a power function variance model, t scores of 2, and a cluster size of 2 x 2. Only pulses that were greater than 0.03 µg/l in amplitude (nadir to peak) were considered as true pulses, as previously described (11).
Statistics
Data from men and women were analyzed separately. Student's t tests were used to compare young and elderly subjects. Due to multiple comparisons, Bonferroni correction was applied, and a difference was considered significant at the p < .025. Data were logarithmically or square root transformed as appropriate prior to analysis.
To compare whether there was a difference in the timing of the peak (maximum) GH pulse, subjects were divided into those having the maximum pulse during the "lights off" period or during the "lights on" period. Fisher's exact test was then used to determine if there was a difference between the young and elderly subjects.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
In several previous studies addressing the contribution of discrete pulse parameters of GH secretion in aging, both pulse frequency and amplitude were found to be attenuated (2)(12)(13). However, all of the previously described studies employed control and study groups heterogeneous for body composition and gonadal steroid milieu. Although increased adiposity and hypogonadism are frequently associated with aging and may be even consequent upon the aging process, they possess independent, powerful influences upon GH secretion. Thus, the contribution of aging per se to the lowered activity of the somatotropic axis in elderly individuals is not well defined.
In our attempt to isolate the effects of aging, we carefully selected healthy, physically active subjects with comparable BMIs. However, more sophisticated assessment of body composition, such as dual-energy x-ray absorptiometry scanning or underwater weighing would likely demonstrate higher percentages of body fat in elderly persons compared with young persons with identical BMIs (14). Also, plasma estradiol levels were lower in the elderly subjects of both sexes, and elderly women had lower testosterone levels than young women.
Thus, although gross disparities between young and elderly subjects were largely eliminated, the more subtle differences in body composition and gonadal milieu were still present, likely as inevitable biological consequences of the aging process. However, the magnitude of these differences was still lower than in the earlier studies of similar nature in which BMI ranged between 18 and 39 kg/m2 (7), and plasma testosterone concentrations differed twofold between the young and elderly men (2). Thus, in the present study, we have achieved the tightest fit between these intervening variables.
The elderly subjects had IGHC and IGF-I levels that were 32% to 56% of their sex-matched younger counterparts. However, both GH pulse frequency and nadir values (which are reflective of interpulse GH levels) were similar in both age groups. The only discrete parameter of GH pulsatility that clearly separated the young from the elderly subjects was GH pulse amplitude. Animal and human studies have shown that GH pulses are consequent upon GHRH secretory discharges (8)(15). Thus, an unchanged GH pulse frequency in elderly individuals of both sexes suggests that the aging-related GH decline is unlikely to result from a decrease in the frequency of GHRH pulses by the hypothalamus.
On the other hand, regulation of GH pulse amplitude is likely to be multifactorial: changes in the pituitary sensitivity to GHRH, in the amount of GHRH per pulse, altered Somatostatin milieu, and the contribution of the endogenous GH secretagogue (GHS) (16) all may potentially play a role. Because the structure and the mechanism of action of endogenous GHS and its interactions with the endogenous GHRH and SRIF are not yet known, one can only speculate about its potential role as a GH pulse generator. We and others have previously shown that both young and elderly men of similar BMI and gonadal milieu respond equally well to standard boluses of GHRH (8)(17)(18). This suggests similar pituitary responsiveness and comparable hypothalamic SRIF milieu. By inference, lower GH pulse amplitude in elderly persons may indicate weaker GHRH stimulus. Recently, we have shown that the overall nocturnal GHRH output is diminished in elderly men versus their younger counterparts, as assessed by the comparison of the percentage of inhibition of spontaneous nocturnal GH output by graded doses of a specific GHRH receptor antagonist (8). This study, however, did not differentiate between the frequency and the amplitude of GHRH pulses. Our current results suggest that the frequency of GHRH pulses is not impaired in elderly persons. Thus, we hypothesize that a lower amount of GHRH released per secretory pulse is the primary mechanism of age-associated attenuation of GH levels. Indeed, exogenous GHRH boluses reliably increased plasma GH and IGF-I levels in elderly persons into the young-normal range (17).
Despite similar declines in daily GH output and the similar changes in GH pulse amplitude in both sexes, we cannot be yet certain that the same conclusions about the amplitude of GHRH pulses can be applied to women. GH regulation in humans is sexually dimorphic (9), and we do not yet have data on semiquantification of hypothalamic GHRH output in women. Thus, more definite conclusions about the potential neuroendocrine mechanisms of the aging-related GH decline in women must await further studies.
Nocturnal augmentation of GH secretion is a well-known and reproducible finding (19). Similar to previous studies (12), the magnitude of the nocturnal GH rise was attenuated by aging. Interestingly, the timing of the maximal GH peak was also different in elderly persons. The nocturnal GH peak has been found to be strongly associated with slow-wave sleep (5)(6)(12)(19). Whether this was true for the young subjects in this study is uncertain, since polysomnography was not performed. However, all of the young men and the majority of the young women had their GH peaks in the 8-hour period when the lights were turned off and when sleeping was permitted. Surprisingly, the majority of the elderly subjects had their GH peak when the lights were still on (i.e., when the subjects were definitely still awake as documented by continuous nursing observation). Van Corvorden and colleagues (20) have shown earlier that the distribution of rapid eye movement stages during sleep is advanced in elderly persons, suggesting that circadian timekeeping may be modified during normal senescence. Since GHRH is a sleep-promoting peptide, and sleep disturbances are frequent in the elderly (21)(22), it would be of interest to reassess the relations between the hypothalamic GHRH release and sleep physiology in elderly persons.
In conclusion, we show that in a group of healthy and physically fit elderly men and women that were closely matched to their younger counterparts for BMI, and to a lesser extent, gonadal steroid concentrations, GH output is markedly diminished. This is due exclusively to the attenuated GH pulse amplitude, suggesting that aging is associated with the diminished quantity of GHRH released per secretory episode.
|
Acknowledgments |
|---|
Received December 4, 1999
Accepted March 13, 2000
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. G.-S. Frutos, L. Cacicedo, C. Fernandez, D. Vicent, B. Velasco, H. Zapatero, and F. Sanchez-Franco Insights into a role of GH secretagogues in reversing the age-related decline in the GH/IGF-I axis Am J Physiol Endocrinol Metab, November 1, 2007; 293(5): E1140 - E1152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Bonkowski, R. W. Pamenter, J. S. Rocha, M. M. Masternak, J. A. Panici, and A. Bartke Long-lived growth hormone receptor knockout mice show a delay in age-related changes of body composition and bone characteristics. J. Gerontol. A Biol. Sci. Med. Sci., June 1, 2006; 61(6): 562 - 567. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
.025; **p 