Aging, body composition, and GH
In man, both aging and GH deficiency are associated with reduced protein synthesis, decreases in lean body mass and bone mass, and increases in body fat.
Caloric intake and energy expenditure decline with age, and older persons tend to weigh about 25% more than do younger adults.
Total body protein synthesis, as measured by the labeled-nitrogen glycine tracer technique, is lower in older adults.
Lean body mass, as estimated by potassium-40 counting, decreases from 25 to 70 yr of age by approximately 12 kg (27%) in men and 5 kg (15%) in women.
These longitudinal data agree with cross-sectional results derived from more than 12,000 subjects and presumably reflect atrophic processes in skeletal muscle, liver, kidney, spleen, skin, and bone.
Conversely, total body fat increases from 18-85 yr of age by approximately 18% in men and 12% in women.
Decreased physical activity and/or increased carbohydrate intake may contribute to the observed increase in body fat mass, but would not explain the reduction of lean body mass of up to 50% observed from 20-80 yr of age.
Of note is that in older persons, expansion of fat mass is greater in the intraabdominal region than in the gluteal and thigh regions, and abdominal fat deposition is greater in the visceral compartment than in the SC adipose tissue.
The hypothesis that age-associated increases in adiposity may result from reduced GH secretion is counterbalanced by the findings of significant, age-independent, inverse correlations of various measures of adiposity with basal GH concentrations and spontaneous GH release.
Recently, use of the statistical technique of deconvolution to analyze rhythmic GH secretion has shown that increased adiposity in men leads to decreased frequency of GH secretory episodes and a significant shortening of the circulatory half-life of GH, both of which reduce integrated daily plasma GH concentrations.
GHRH-stimulated GH secretion is also reduced with increased adiposity in both men and women.
Thus, it is possible that there is a reciprocal relationship between adiposity and GH secretion, with more body fat leading to less GH secretion and reduced GH secretion leading to increased body fat.
In addition, age-related GH hyposecretion in men appears to be more strongly related to the proportion of abdominal body fat, as estimated from waist/hip ratios, than to total body fat, as assessed by body mass index.
The observation that GH secretion is inversely related to upper body fat deposition is important, particularly in light of the known influence of excess abdominal fat to increase insulin resistance, perturb lipid homeostasis, and enhance the risks of diabetes, hypertension, and coronary atherosclerosis.
Effects of aging on GH secretion
The effects of human aging on GH secretion have been evaluated by a number of researchers.
Early studies reported unchanged baseline plasma GH levels.
However, interpulse baseline GH concentrations are usually below the levels detectable with polyclonal RIAs and GH secretion varies diurnally, so that random single samples do not adequately characterize GH secretory dynamics.
Studies of 24 h secretion of GH, assessed by sampling every 20 min (Fig. 1), have shown variable reductions (15%-70%) in most 24 h GH secretory parameters in middle-aged (40-65 yr) and in older (>60 yr) men and women.
Mean (±1SD) 24 h GH release in nine young (mean age 26±4 yr) and 10 old (68±6 yr) men sampled at 20 min intervals, demonstrating the normal diurnal pattern of GH secretion and the reduced nocturnal peak amplitude in the older men.
In a recent study performed in healthy nonobese men, aged 21-71, deconvolution analysis has revealed that, with each advancing decade, the GH production rate decreases by 14% and the GH half-life falls by 6%.
Reductions in the number and amplitude of GH spontaneous pulses during sleep have been reported in adults over 50 yr of age, and in two different studies, were found to be confined to the first 3-4 nighttime hours.
Although nocturnal GH secretion is reduced with age in men, old men, like younger men, still have 2- to 4-fold greater secretion at night than during the day.
Responses of GH to indirect stimuli have also been studied in older people.
For example, the GH response to insulin-induced hypoglycemia has been reported to show no change or a decrease with age, a discrepancy that may be partly explained by the inclusion of subjects up to 50 yr of age in the young control group in one of the studies.
The acute GH response to an iv arginine infusion has been found to be similar in young and old men, while an arginine infusion potentiates GH responsivity to GHRH in old men.
Scientists have found that administration of an oral arginine/lysine preparation (6 g of each amino acid /day) for 2 weeks does not increase either spontaneous GH release (as assessed by overnight frequent sampling), plasma IGF-I levels, or GHRH-stimulated GH secretion in old men.
The acute GH response to exercise is reduced with age.
In one study, exercise training for 12 weeks improved the acute GH response after an hour of exercise in older men, although trained older men still had a lower GH response to exercise than did trained or even sedentary younger men.
In another study, older men, unlike younger men, failed to demonstrate an increase in resting or 1 h post-exercise plasma GH levels after 12 weeks of resistance strength training.
Stimulation with GHRH has been reported to be present, but significantly reduced, in healthy old women, and in some, but not other, studies of old men.
It is possible that variations in published findings in old men may have been due, in part, to differences in race and/or life-style (e.g. exercise and diet), to the amount and/or regional distribution of body fat, or to other as yet unidentified factors.
Changes of GH-BPS with age
Most, but not all, studies have shown that serum concentrations of GH-BPS vary with age.
GH binding activity is virtually undetectable in cord sera of premature infants.
In normal-statured subjects, serum GH-BP activity rises progressively from early childhood, reaching peak levels in middle adulthood.
In one study of a large group of subjects aged 20 to 80 yr, GH-BP levels were observed to decrease after the fifth decade in healthy men, but not in women.
In another investigation, GH-BP levels were reduced in a combined group of healthy men and women beyond the age of 60 yr.
This decrease with age could reflect either a loss of peripheral GH receptors (the putative source of GHBPS) or the known decrease in GH secretion with aging physiological significance of age-related changes in serum GH-BPS, and in the extramembranous portion of the GH receptors from which they are derived, remains to be established.
Mechanisms of decline in GH secretion with aging
Aging in rats has been associated with a reduced in vivo GH response to GHRH, whereas the corresponding in vitro response has been reported to be either unchanged or reduced.
Controversy exists as to whether there is an intrinsic pituitary defect with age in GHRH-mediated signal transduction in the rat.
Conclusions regarding age-related defects in GH secretory capacity of primary suspensions of somatotropes may not be valid due to artifacts introduced by enzymatic digestion of tissues.
Data from studies of intact rats suggest that age-related reductions in GH secretion result primarily from an increase in somatostatin tone, and perhaps, to a lesser extent, from a decrease in GHRH secretion.
Because neuroendocrine regulatory mechanisms differ between rodents and man, it is not clear whether and to what extent the above mechanisms contribute to decreased GH secretion in old people.
To date, it has not been possible to accurately measure GHRH or hypothalamic somatostatin in peripheral blood of humans, both because current RIAs are insufficiently sensitive to detect the low GHRH concentrations in peripheral blood, and because it is not possible to distinguish somatostatin of hypothalamic origin from that derived from paninhibit somatostatin secretion, the observation that iv infusion of arginine potentiates GH responsivity to GHRH in old men suggests that in humans, as in the rat, somatostatin tone is increased with age.
Conversely, the findings that repetitive administration of GHRH to old men and women reverses, at least in part, their diminished GH responsiveness to GHRH, suggest that there may also be a reduction with age in GHRH input to the pituitary.
In addition, although only nonsignificant changes have been found in GH content of pituitary cells from old persons, a reduction with age in the number and size of somatotropes has been reported in one immunocytochemical study.
Finally, infusion of theophylline, a phosphodiesterase inhibitor that increases intracellular cAMP, improves GH responses to GHRH in old, but not young, men suggesting that there may be an intrinsic alteration in GHRH signal transduction in the aging somatotrope.
Effects of aging on IGF-I
In cross-sectional studies of adults, levels of IGF-I in unextracted serum or plasma decrease with age in both men and women.
In one study of 226 healthy men and women, the mean serum IGF-I concentration in subjects in the seventh decade was approximately one-half that of persons in the third decade.
Rudman et al. reported that the prevalence of low IGF-I serum concentrations increases progressively from 11-55% from the fourth through the ninth decades.
Circulating IGF-I levels obtained from serum or plasma samples extracted with acid-ethanol also diminish with age in both men and women.
Figure 2 shows the decrease with age in total plasma IGF-I levels determined after acid extraction in 131 women and 223 men of different ages in the Baltimore Longitudinal Study of Aging (J. T. Tobin, M. R. Blackman, S. M. Harman, and S. S. Sherman).
Similarly, in two studies IGF-I levels in extracted sera were 30%-40% lower in healthy older vs. younger men.
The strong positive relationship between baseline serum IGF-I levels and spontaneous 24-h GH secretion in young adults has not been observed in older subjects.
Several studies have shown that circulating IGF-I levels increase similarly in young and old men after exogenous administration of either GH or GHRH.
In one cross-sectional study, plasma IGF-I was inversely correlated with adiposity, declining from 1.2 U/ml to 0.6 U/ml as percent of ideal body weight increased from 90-l 10%.
In another study of persons aged 17 to 83 yr, there was an age-independent inverse correlation between body mass index and IGF-I in men, but not in women.
There is also a significant inverse correlation of IGF-I levels with age that is independent of adiposity.
Thus, it appears that increased adiposity exerts an effect, independent of age, to decrease IGF-I levels.
The extent to which the effects of obesity and/or aging on circulating IGF-I levels are mediated by their known influences on GH secretion remains to be clarified.
Because both aging and obesity are associated with elevation of plasma insulin levels, one possible mechanism for the above effects is an increase in insulin action on hypothalamic and/or pituitary IGF-I receptors, resulting in enhanced feedback inhibition of GH, and hence IGF-I, secretion.
IGF-I levels in normal women and men in the Baltimore Longitudinal Study on Aging; the decrease with age in mean (±1 SD) IGF-I is shown by decade.
A decrease in physical activity, as assessed by sub maximal or maximal aerobic capacity (VO2 Max), may also contribute to the reduction in serum IGF-I levels with age.
VO2 Max has been significantly correlated with IGF-I levels, independent of age, in healthy men and women.
However, this association appears to be less strong in old vs. young men.
Changes of IGF-BPs with age
Levels of IGFBP-3, the major plasma IGF-I binding protein, are low at birth, increase rapidly during the first weeks of life, and rise more slowly until puberty, when they reach a maximum.
Circulating levels of IGFBP-3 (measured as 125 kDa complex) decrease between 18 and 65 yr of age in both sexes.
In another study, plasma IGFBP-3 (measured as 53 kDa IGF-I binding subunit) was lower in healthy women aged 64-84 yr than in women aged 22-42 yr.
Recently, we observed an age-related decrease in serum IGFBP-3 in men from age 22 to 79 yr, as well as a strong, age-independent positive correlation of IGFBP-3 with IGF-I levels.
Prior clinical reports have shown that changes in plasma IGF-I are positively correlated with IGFBP-3 binding activity.
Although it remains uncertain as to whether IGFBP-3 levels are directly stimulated by GH alone, by IGF-I alone, or by both GH and IGF-I, it seems likely that the decreases in GH and IGF-I secretion with aging are, at least in part, responsible for the reported decrease in IGFBP-3 levels.