| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Studies |
London Centre for Paediatric Endocrinology and Metabolism at University College London (P.C.H., P.J.P., C.G.D.B.), Cobbold Laboratories, Middlesex Hospital, London W1N 8AA; and Medical Research Council Environmental Epidemiology Unit (C.H.D.F., C.O.), University of Southampton, Southampton General Hospital, Southampton, SO16 6YD, United Kingdom.
Address all correspondence and requests for reprints to: Dr. P. C. Hindmarsh, London Centre for Paediatric Endocrinology and Metabolism, Cobbold Laboratories, Middlesex Hospital, Mortimer Street, London W1N 8AA, United Kingdom. E-mail: p.hindmarsh{at}med.ucl.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mean 24-h serum GH concentrations ranged between 0.19 and 2.15 mU/L (1 µg/L = 2.6 mU/L). Pulse periodicity was between 180 and 200 min. There was a positive relationship between peak GH levels and serum IGF-1 and IGFBP-3 levels (r = 0.39; P = 0.009 and r = 0.32; P = 0.03, respectively). GH trough levels were unrelated to these measures of the IGF family. In contrast, GH troughs were related inversely to BMI (r = -0.31; P = 0.04) and waist-hip ratio (r = -0.4; P = 0.006). Peak GH levels were not related to these measures. Factors known to influence these measures, fasting insulin concentration, or cortisol secretion did not alter the trough GH relationship in multiple regression analysis. All GH parameters were related inversely to fasting insulin concentration. Although GH parameters were related inversely to cholesterol and low-density lipoprotein-cholesterol, this effect disappeared when age and fasting insulin levels were introduced into the regression. GH-binding protein levels related most strongly to BMI (r = 0.60; P < 0.001), with no effect of any GH parameter observed in multiple regression analysis. These results suggest that the peak values of a GH concentration profile may influence the IGF axis, whereas trough values may influence body composition and metabolic parameters of GH action.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In animals and man, GH is secreted in a pulsatile manner. The pattern of GH secretion in rodents is an important determinant of the sexually dimorphic pattern of growth (7), liver enzyme function (8), circulating GH-binding protein (GHBP) concentration (9), and IGF-1 mRNA expression in skeletal muscle and liver (10) in these animals. The adult male rat displays a discrete high-amplitude, three-hourly, pulsatile pattern of GH secretion, whereas in the adult female rat, pulsatile secretion is of low amplitude with irregular periodicity.
Little is known about the influence of GH secretory patterns in man on metabolic parameters, although growth in children is known to be GH pulse amplitude modulated (11, 12), and abnormal GH pulsatility has been documented in children with a variety of dysmorphic syndromes whose growth is poor (13). More detailed studies of the role of GH pulsatility on growth and metabolism in man have been limited by the sensitivities of the GH assays currently in use and the lack of algorithms adequate to define the contribution made by the troughs in a GH profile. Finally, several studies of GH replacement in GH-deficient adults, using continuous or pulsatile GH administration, have inferred differential metabolic effects without actually measuring the circulating GH patterns achieved (14, 15, 16).
Using a chemiluminescent GH assay with a sensitivity an order of magnitude lower than conventional immunoradiometric assays, coupled with a technique for determining the trough component of GH profiles (17), we have analyzed 24-h serum GH concentration profiles in 45 men and related different attributes of the GH signal to a number of GH-dependent parameters of body composition and metabolism.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Forty-five adult male volunteers, 59.4–69.9 yr old, were studied. The men were drawn from a cohort of 300 men (who are part of an ongoing study by the Medical Research Council Environmental Epidemiology Unit, Southampton, UK) into the relationship between early growth and the development of cardiovascular risk factors (18). Smoking status (never, ex-, or current smoker) was noted. All men were fit and well and not receiving any medication known to influence GH secretion. The studies were approved by the Ethics Committee of University College London Hospitals and written informed consent obtained.
The men were admitted to the hospital on day 1 and allowed to acclimatize to the hospital environment. Standard meals were provided at 0800, 1230, and 1730 h and exercise permitted within the confines of a hospital admission. Smoking was not permitted. On day 1, an iv cannula was inserted into a forearm vein; and at 0730 h on day 2, a blood sample was drawn for the measurement of serum GH and cortisol concentrations, and sampling was then continued at 20-min intervals for the next 24 h. At 0600 h on day 3, an additional blood sample was drawn to measure serum concentrations of IGF-1, IGFBP-3, insulin, T4, testosterone, estradiol, and GHBP.
As part of the Medical Research Council study, details of total, high-density lipoprotein-, and LDL-cholesterol concentrations were available from studies conducted within 1 yr of the GH profiles. Information also was available on height, weight, and waist and hip circumference measured within 6 months of the study, as well as derived values for body mass index [BMI (weight/height2)] and waist-hip ratio (WHR).
Hormone assays
GH. Serum GH concentrations were measured using the Nichols Chemiluminescence assay for human GH [hGH (Nichols Institute Diagnostics, San Juan Capistrano, CA)]. The assay was modified according to published reports (19), and a bottom standard of 0.014 mU/L was included to define the working range more adequately. The within-assay coefficients of variation (CV) were 5.5, 6.8, 10.5, and 10.8% at serum concentrations of 0.4, 10.0, 18.6, and 23.3 mU/L, respectively. Between-assay CV were 12.1, 12.3, and 9.0% at serum concentrations of 3.3, 6.3, and 18.0 mU/L, respectively. The assay standards were calibrated against WHO 1st International Standard 80/505. The sensitivity of the assay [mean of zero tubes (n = 10) 3 SD) was 0.036 mU/L (1 µg/L = 2.6 mU/L), but a limit of detection of 0.05 mU/L was used throughout. Eight of 45 subjects had a GH value less than the detection limit at some point in their profile. Of the total number of samples analyzed (n = 3271), 1.8% were below the detection limit. All samples from each profile were analyzed in a single assay.
IGF-1. Serum IGF-1 was measured using an in-house polyclonal RIA with acid/alcohol extraction. The sensitivity of the assay was 0.07 U/mL. The within-assay CV were 11.3, 6.5, and 4.7% at serum concentrations of 0.23, 1.23, and 3.53 U/mL, respectively. Between-assay CV were 10.5, 12.1, and 5.1% at concentrations of 0.38, 0.99, and 3.53 U/mL, respectively.
IGFBP-3. IGFBP-3 was measured using a RIA kit (Diagnostic Systems Laboratories Inc, Webster, TX). The sensitivity of the assay was 0.9 ng/mL. The within-assay CV were 8.1 and 5.4% at serum concentrations of 2200 and 7800 ng/mL, respectively. Between-assay CV were 9.8 and 4.8% at serum concentrations of 2200 and 8500 ng/mL, respectively.
Serum concentrations of T4, testosterone, cortisol, and estradiol were measured using standard immunoassays (EuroDPC, Llanberis, Gwynedd, UK). Serum GHBP concentrations were measured using a ligand-mediated immunofunctional assay (20) and serum insulin concentration by a double antibody (RIA) procedure.
Analysis of hormone pulsatility
The mean 24-h serum GH concentration was derived from each of the profiles. The periodicity of GH pulsatility was determined by subjecting the profiles to Fourier transformation after the data had been smoothed using a 3-point moving average and stationarized (no systematic change in mean and variance) (21). Fourier transformation is a method that delineates the frequency component of complex signals in a manner analogous to the way a prism splits white light into its constituent colors. The process resolves data into a series of sine and cosine components that represent frequencies in the data. Each frequency has an amplitude attribute that shows the power of that frequency within the data array. If plotted as a histogram, a power spectrum results. The spectral peak at any frequency or period indicates a greater-than-random tendency for peaks and troughs in the time series to recur at that frequency or period.
To determine the effects of target organ exposure to circulating GH concentrations, the distribution method for analysis of 24-h GH profiles of Matthews et al (17) was used. In brief, for each 24-h serum GH concentration profile, a cumulative frequency distribution was calculated. This generates a sigmoidal ogive that, when plotted on probability paper, results in a linear relationship. From this probability-serum GH concentration plot, discrete probabilities (linear probits) can be derived. The threshold concentration at or below which the 24-h profile spends 5% (trough) or 95% (peak) of the time, can then be estimated from the regression equation. These values are termed the observed concentration (OC) at which the profile spends 5% (OC5) or 95% (OC95) of the time. The analysis is analogous to the calculation of ED50 values in pharmacological practice. For the sake of clarity, these derivations will be known as peak for OC95 and trough for OC5.
Statistics
Nonnormal distributed data were natural log (Ln) transformed before analysis. Mean values were compared using Student’s t test or one-way ANOVA. The relationships between continuously distributed parameters were explored using univariate or multivariate linear regression.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Anthropometric details of the 45 men are shown in Table 1
. The mean adult height of 172.5 cm is close to the
Tanner-Whitehouse UK British Standards (mean 174.4 cm) (22), which were
largely constructed from the population of South East England, whose
current age would be similar to that of the 45 men studied. A wide
range of measures for BMI and WHR were noted but, only 3 individuals
had a BMI more than 30 kg/m2 (95th percentile). Of the 45
men, 7 had never smoked, 33 were ex-smokers, and 5 continued to
smoke.
|
Mean 24-h serum GH concentration in this group of subjects was
0.91 mU/L (range 0.19–2.15). The spectral peak in the Fourier spectral
power histogram was highest with a GH pulse periodicity between 180 and
200 min, indicating a greater-than-random tendency for peaks in the
time series to recur at this periodicity. Table 1
summarizes the peak
and trough GH values, along with measures of the serum concentrations
of IGF-1 and IGFBP-3.
There was a slight, but nonsignificant, decrease in 24-h mean serum GH concentration, GH peak, GH trough, and serum IGF-1 concentrations with age. There was no influence of either plasma estradiol concentration or the serum concentrations of testosterone or T4 on mean 24-h serum GH concentrations, GH peak and GH trough values, or serum IGF-1 levels. Serum IGFBP-3 values were influenced by plasma estradiol concentrations (r = 0.33; P = 0.04). None of the endocrine factors were influenced by smoking status.
GH and the insulin-like growth factor axis
Figure 1 shows the relationships between
Ln-transformed GH peak values and serum concentrations of IGF-1 (Fig. 1A) (r = 0.39; P = 0.009) and IGFBP-3 (Fig. 1B)
(r = 0.32; P = 0.03).
|
).
|
Although there was a negative nonsignificant relationship of the GH measures with age, introduction of age into the multiple regression analysis made no difference.
GH and body proportions
Figure 2 shows the relationship between
Ln-transformed trough serum GH concentrations and BMI (Fig. 2A) (r
= -0.31; P = 0.04) and WHR (Fig. 2B) (r = -0.40;
P = 0.006).
|
). WHR and
BMI were unrelated to serum IGF-1 and IGFBP-3 levels. GH and carbohydrate and lipid metabolism
Table 3 shows a regression analysis of peak, mean
24-h, and trough serum GH concentrations with insulin and cholesterol
parameters. Both Ln GH peak and Ln GH trough values were related
inversely to fasting serum insulin concentration, as was mean 24-h
serum GH concentration. Although fasting total and LDL-cholesterol
concentrations were related inversely to the Ln GH peak concentration,
these relationships disappeared when the age of the individual was
introduced into the regression equation. Both chronological age and
fasting serum insulin concentration were the major determinants of
fasting total and LDL-cholesterol concentrations. The relationships
with fasting insulin concentration were uninfluenced by BMI or current
weight. Multiple linear regression analysis of age and Ln GH peak with
fasting cholesterol showed a significant effect of chronological age
(partial P = 0.01), to the exclusion of Ln GH peak
(partial P = 0.17).
|
Serum GHBP concentration was strongly related to current weight (r = 0.58; P = 0.0001) and BMI (r = 0.60; P < 0.001). There was no relationship between either WHR or current height and serum GHBP concentration.
Serum GHBP concentrations were not significantly related to Ln GH peak values (r = 0.19; P = 0.23) or mean 24-h mean serum GH level (r = 0.08; P = 0.63) but were related inversely to Ln GH trough (r = -0.34; P = 0.02).
When Ln GH trough and current BMI were introduced into a multiple regression analysis against GHBP, no effect of the GH parameter was observed (BMI partial P < 0.001; Ln GH OC5 partial P = 0.33).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Serum IGF-1 and IGFBP-3 concentrations were associated largely with peak serum GH concentrations, confirming a previous study of physiological GH secretion in a group of individuals with a wider age range (30). There was no relationship between the peak or mean GH level and current height, weight, or BMI of the individual. This latter observation is rather surprising, considering a number of reports that demonstrate an inverse relationship between BMI and parameters of GH secretion, such as daily secretion rate (31), integrated GH concentrations (32), or (indirectly) IGF-1 levels (33). In contrast, the relationship of GH with parameters of body composition (BMI and WHR) seem to be more related to the level of the trough GH value. These observations cannot be explained by an interrelationship between peak and trough values, because they seem to be independent of each other (r = 0.02; P = 0.89).
We analyzed data within a narrow age range and in a larger population sample than hitherto reported. Also, there is no reason to believe, from the anthropometric standpoint (22, 33), that the sample chosen is anything other than typical of the male population between 59 and 69 yr old. The narrow age range allows the removal of a potential confounding variable in the GH-BMI relationship (30), namely age, which was a more important determinant of BMI than serum IGF-1 level in one study (33). The relationship between GH and WHR stresses the need to consider markers of body composition other than BMI (30). Low trough GH concentrations are likely to be associated with an increased WHR, which may be important because WHR is itself known to be an important predictor of cardiovascular disease (34), so the relationship with low trough GH levels may go some way to explain how GH deficiency is associated with an increased risk of death from heart disease (35). An alternative explanation, which needs to be tested more formally, is that trough concentrations could be related to negative feedback from metabolic factors associated with adiposity.
The trough component of the GH signal also seems to be associated inversely with the circulating concentration of GHBP. This observation might be thought to support the alternative explanation above. Although it has been broadly assumed that GHBP is regulated by GH, the precise regulation of this protein within the physiological range of GH concentrations is unclear (36). Even in GH-insufficient children receiving GH therapy, GHBP poorly reflects the GH replacement status (37). In the rat, the circulating level of GHBP is predominantly regulated in a positive manner by trough concentrations (9, 38). In man, the opposite seems to be the case, but even this relationship was rendered insignificant when the dominant effect of either BMI or current weight was introduced into the multiple regression. These results suggest that GHBP may be more of a marker of body size and that levels may be influenced by factors other than GH secretory status. Similar observations have been made in subjects with obesity (39).
Both peak and trough values seem to be related to the fasting insulin concentration, but neither peak nor trough levels were major determinants of total or LDL-cholesterol, which seemed to be determined simply by age (and independent of GH).
These results demonstrate that in healthy 60-yr-old men, GH secretion takes place in a pulsatile fashion, with a pulse periodicity similar to that observed in younger individuals. The GH-dependent components of the IGF family seem to be observed with the peak GH level, whereas the trough GH concentration seems to be an important factor in determining body composition.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 6, 1996.
Revised December 16, 1996.
Revised March 14, 1997.
Accepted March 14, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  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]
|
|
|
|
 |
|
 R. I. G. Holt, E. Webb, C. Pentecost, and P. H. Sonksen Aging and Physical Fitness Are More Important Than Obesity in Determining Exercise-Induced Generation of GH J. Clin. Endocrinol. Metab., December 1, 2001; 86(12): 5715 - 5720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Waters, C. R. Qualls, R. Dorin, J. D. Veldhuis, and R. N. Baumgartner Increased Pulsatility, Process Irregularity, and Nocturnal Trough Concentrations of Growth Hormone in Amenorrheic Compared to Eumenorrheic Athletes J. Clin. Endocrinol. Metab., March 1, 2001; 86(3): 1013 - 1019. [Abstract] [Full Text]
|
|
|
|
|
|
T. Laursen, C. H. Gravholt, L. Heickendorff, J. Drustrup, A.-M. Kappelgaard, J. O. L. Jørgensen, and J. S. Christiansen Long-Term Effects of Continuous Subcutaneous Infusion Versus Daily Subcutaneous Injections of Growth Hormone (GH) on the Insulin-Like Growth Factor System, Insulin Sensitivity, Body Composition, and Bone and Lipoprotein Metabolism in GH-Deficient Adults J. Clin. Endocrinol. Metab., March 1, 2001; 86(3): 1222 - 1228. [Abstract] [Full Text]
|
|
|
|
|
|
P. C. Hindmarsh, E. Dennison, S. M. Pincus, C. Cooper, C. H. D. Fall, D. R. Matthews, P. J. Pringle, and C. G. D. Brook A Sexually Dimorphic Pattern of Growth Hormone Secretion in the Elderly J. Clin. Endocrinol. Metab., August 1, 1999; 84(8): 2679 - 2685. [Abstract] [Full Text]
|
|
|
|
|
|
E. S. Kurland, F. K. W. Chan, C. J. Rosen, and J. P. Bilezikian Normal Growth Hormone Secretory Reserve in Men with Idiopathic Osteoporosis and Reduced Circulating Levels of Insulin-Like Growth Factor-I J. Clin. Endocrinol. Metab., July 1, 1998; 83(7): 2576 - 2579. [Abstract] [Full Text]
|
|
|
|
|
|
J. N. Roemmich, P. A. Clark, V. Mai, S. S. Berr, A. Weltman, J. D. Veldhuis, and A. D. Rogol Alterations in Growth and Body Composition During Puberty: III. Influence of Maturation, Gender, Body Composition, Fat Distribution, Aerobic Fitness, and Energy Expenditure on Nocturnal Growth Hormone Release J. Clin. Endocrinol. Metab., May 1, 1998; 83(5): 1440 - 1447. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
