Visceral Adipose Tissue Is Associated with Circulating High Affinity ...

0021-972X/97/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1997 by The Endocrine Society

Vol. 82, No. 3 Printed in U.S.A.

Visceral Adipose Tissue Is Associated with Circulating High Affinity Growth Hormone-Binding Protein ´ A. M. ROELEN, HANS P. F. KOPPESCHAAR, WOUTER R. DE VRIES, CORNE YVONNE E. M. SNEL, MANORATH E. DOERGA, PIERRE M. J. ZELISSEN, JOS H. H. THIJSSEN, AND MARINUS A. BLANKENSTEIN Department of Endocrinology, University Hospital Utrecht, and the Department of Medical Physiology and Sports Medicine, Utrecht University (W.R.d.V.), Utrecht, The Netherlands ABSTRACT Recent data show that body fat distribution, specifically visceral fat accumulation, is associated with the regulation of GH secretion. To our knowledge no studies have been performed with regard to the relationship between plasma high affinity GH-binding protein (GHBP) levels and fat distribution in humans. To address this question, we measured plasma GHBP and insulin-like growth factor I levels as well as visceral, sc abdominal, and hip adipose tissue (AT) areas by using magnetic resonance imaging scanning in 12 patients with GH deficiency (GHD) and in 12 age- and sex-matched healthy subjects. The GHD patients were subsequently treated with GH replacement therapy. Regardless of the GH status of the subjects, body mass index and visceral AT area were positively correlated to plasma GHBP (r 5 0.70; P , 0.01 and r 5 0.73; P , 0.01, respectively),

whereas the sc AT areas at the abdominal level tended to correlate positively with GHBP levels, but did not reach significance (r 5 0.44; P 5 0.07). The sc AT areas at the hip level were not correlated with plasma GHBP levels. In the GHD patients the pretreatment visceral and abdominal sc AT areas were positively correlated with the change in GHBP levels after GH replacement (r 5 0.82; P , 0.01 and r 5 0.75; P , 0.01, respectively). The pretreatment sc AT area at the hip level was not associated with the therapy-induced changes in plasma GHBP (r 5 0.28; P . 0.10). In summary, this study shows that visceral fat is associated with circulating GHBP levels, suggesting that visceral fat mass may be involved in the regulation of the plasma GHBP level. Further, the amount of abdominal fat in GHD patients may partially determine the plasma GHBP response to GH replacement therapy. (J Clin Endocrinol Metab 82: 760 –764, 1997)

G

aging (MRI) scanning, in patients with GHD both before and after GH replacement therapy and in age- and sex-matched healthy subjects. Furthermore, we investigated the contribution of body fat distribution to the interindividual variability in plasma GHBP responses to GH replacement therapy in GHD patients.

H SECRETION is influenced by physical characteristics, such as age, sex, and adiposity (1, 2). Furthermore, GH plays a role in regulating fat mass by its lipolytic effect (3, 4). Studies from several research groups have emphasized the importance of regional adipose tissue distribution as a correlate of the metabolic complications that can be observed in obesity. For example, in visceral obesity, there is a high prevalence of insulin resistance, hyperinsulinemia, glucose intolerance, dyslipidemia, and hypertension (5– 8). It is also known that in obesity pulsatile GH secretion is markedly diminished (9). GH-binding protein (GHBP) levels are high in patients with obesity (10, 11). As circulating GHBP may be regarded as an intrinsic part of the GH/insulin-like growth factor I (IGF-I) axis, an effect of body composition on circulating GHBP levels may be expected. In general, positive correlations between body mass index (BMI) and plasma GHBP levels were observed (11–16). Recently, we reported that GH-deficient (GHD) adults have an increased visceral fat mass compared to healthy subjects matched for age and sex (17). To our knowledge no studies have been reported that investigated the association between fat distribution and circulating GHBP in humans. The aim of this study was to investigate the relationship between plasma GHBP and visceral adipose tissue (AT) areas, sc AT areas at the level of the abdomen and hip, as assessed by magnetic resonance im-

Subjects and Methods Twelve GHD adults, 7 men and 5 women, aged 25– 60 yr, known to have GHD for at least 2 yr (range, 3–20 yr), were included in the study. GHD was defined as a peak plasma GH concentration of 5 mg/L or less after arginine infusion. Twelve healthy sedentary adults (7 men and 5 women), matched for age and sex, also participated in this study, which was approved by the ethical committee of the University Hospital of Utrecht. All subjects gave informed consent. Pertinent physical characteristics of both groups are shown in Table 1. Secondary hypothyroidism (11 patients) was treated with levothyroxine in a daily dose ranging from 75–200 mg (average dose, 134 mg); secondary hypogonadism was treated with 80 mg testosterone undecanoate twice a day orally or with testosterone esters (250 mg) every 3 weeks im (6 men) and with cyclic estrogen (30 mg/day ethinyl estradiol) and progesterone (150 mg/day levonorgestrel; 3 women); secondary adrenal insufficiency (all patients) was treated with cortisone acetate in a daily dose ranging from 10 –37.5 mg (average dose, 28.3 mg). No adjustments were made in replacement dose of any patient during GH treatment. After the pretreatment measurements, all GHD patients were treated during 6 months with daily sc injections of GH (Genotropin, Pharmacia, Uppsala, Sweden). The initial dose was 0.041 mg/kg BWzweek during the first month, followed by 0.082 mg/kgzweek. Blood samples were obtained after an overnight fast. We measured GHBP levels by fast protein liquid chromatography gel chromatography after the addition of a fixed amount (10 ng) of purified, monomeric [125I]GH and different concentrations of radioinert GH (0, 30, and 100 mg/L and 10 mg/L, the latter to determine aspecific binding) to aliquots of serum, as described previously (18). The coefficient of variation (CV)

Received July 29, 1996. Revision received November 20, 1996. Accepted December 5, 1996. Address all correspondence and requests for reprints to: Hans P. F. Koppeschaar, M.D., Ph.D., Department of Endocrinology, University Hospital Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands.

760

BODY FAT DISTRIBUTION AND PLASMA GHBP TABLE 1. Characteristics of the growth hormone deficient patients (n 5 12) before treatment and healthy subjects (n 5 12)

Age (yr) Height (cm) Weight (kg) BMI (kg/m2) Body fat (kg) GHBP (pmol/L) IGF-I (ng/mL) AT areas (cm2) Visceral Subcutaneous Abdominal Hip level

Patients

Healthy subjects

P

47.7 6 10.2 172 6 10 74.9 6 14.7 25.0 6 3.5 18.7 6 4.9 1287 6 290 67 6 36

44.8 6 8.8 175 6 9 70.2 6 11.6 22.9 6 2.2 13.1 6 3.2 1120 6 177 198 6 36

NS NS NS NS 0.05 0.03 0.005

144.0 6 58.1

66.4 6 51.7

0.002

192.6 6 44.6 225.9 6 49.1

136.7 6 54.0 191.5 6 67.9

0.010 NS

Values are mean 6 SD; BMI, body mass index; AT, adipose tissue; NS, not significant. was 4%. Plasma GH concentrations were determined using the RIA of the Oris Industry Co. (Gif-sur-Yvette, France), which had a lower detection limit of 0.5 6 0.04 mg/L (1 mg/L 5 2 mIU/L); the intra- and interassay CVs were 7.7% and 11%, respectively. GHBP levels were obtained by extrapolating the maximal binding capacity from the Scatchard plot (18). After separation of IGFs from IGF-binding proteins by Sep-Pak C18 cartridge chromatography (19), IGF-I was measured by RIA using the antiserum from Underwood and Van Wijk, distributed by the NIDDK (20). At a concentration of 200 ng IGF-I/mL the in-between CV was 5.9%, and the within-assay CV was 7.9%. Total body fat was assessed by bioelectrical impedance measurement, using the tetrapolar BIA-101 analyzer (RJL-Systems, Detroit, MI). Resistance and reactance were measured during application of an alternating electric current of 800 mA at 50 kHz, with the electrodes placed as described by Lukaski et al. (21). The areas of AT were assessed by MRI scanning, performed with a 1.5-T whole body scanner (Gyroscan S15, Philips Medical Systems, Best, The Netherlands) using a multislice inversion-recovery sequence with a 300-ms inversion time, an 820-ms repetition time, and a 20-ms echo time. This sequence highlights AT while effectively suppressing other tissues (22). A transverse scan at the abdominal level, consisting of three slices of 10 mm thickness with a gap of 2 mm, was taken halfway between the lower rib margin and the iliac crest. A similar scan of the hip was taken at the level of the great trochanter. A detailed description of the analysis was published previously (17, 23). All measurements were performed by the same observer and repeated in a second session; the observer was blind to the subjects. The intraobserver CV was 2.9% for the visceral AT areas in controls, and 2.0% and 3.7% in the GHD patients before and after GH replacement therapy, respectively. For the sc abdominal AT areas, the intraobserver CVs were 0.9%, 0.9%, and 1.7%, and for the sc AT areas at the hip level, they were 0.8%, 0.9%, and 1.4%, respectively (17).

Statistical analysis Student’s t test for unpaired samples was used to assess differences between patients and healthy subjects. To evaluate the effects of GH replacement therapy, Student’s t test for paired samples was used. Plasma GHBP levels were correlated (Pearson correlation coefficient) with BMI, visceral AT, and sc AT at the abdominal and hip levels. Correlations of interest were evaluated with linear regression analysis. Statistical significance was accepted for P , 0.05. Results are expressed as the mean 6 sd.

Results

The characteristics of the GHD patients before treatment and those of the healthy subjects are presented in Table 1. Significant differences between both groups were found for total body fat, plasma GHBP and IGF-I, and all measured AT areas, with the exception of the sc hip AT area. Regardless of

761

the GH status of the subjects, i.e. GHD, GHD plus replacement, or normal, the BMI and visceral AT area were positively correlated to plasma GHBP (r 5 0.70; P , 0.01 and r 5 0.73; P , 0.01; Fig. 1, A and B, respectively), whereas sc AT areas at the abdominal level tended to correlate positively with GHBP levels, but did not reach significance (r 5 0.44; P 5 0.07; Fig. 1C). The sc AT areas at the hip level were not associated with plasma GHBP levels (r 5 0.10; P . 0.10; Fig. 1D). After 6 months of GH replacement therapy, mean plasma IGF-I was significantly increased from 67 6 36 to 226 6 113 ng/mL (P , 0.0001), which is in the range of the mean plasma IGF-I of the healthy subjects (198 6 36 ng/mL). The mean plasma GHBP remained at about the same level (1259 6 312 to 1266 6 302 pmol/L; P 5 0.31; n 5 10 instead of n 5 12, because 2 posttreatment GHBP values are missing). Interindividual variability in plasma GHBP changes ranged from a decrease of 185 pmol/L (214% of the baseline value) to an increase of 230 pmol/L (118% of the baseline value). Body weight changed from 74.9 6 14.7 to 75.5 6 15.6 kg (P 5 0.29), and BMI from 25.0 6 3.5 to 25.2 6 3.8 kg/m2 (P 5 0.31), whereas total body fat decreased from 18.7 6 4.9 to 16.0 6 5.3 kg (P 5 0.004). Expressed as a percentage of the baseline values the decreases in AT areas were 38.2% for visceral AT (P 5 0.002), 15.6% for sc abdominal AT (P 5 0.019), and 12.4% for sc hip AT area (P 5 0.006). The pretreatment visceral AT area and the sc AT area at the abdominal level were positively correlated with the changes in plasma GHBP after GH replacement (r 5 0.82; P , 0.01 and r 5 0.75; P , 0.01; Fig. 2, A and B, respectively), whereas the pretreatment sc AT area at the hip level was not associated with the changes in plasma GHBP (r 5 0.28; P . 0.10; Fig. 2C). After GH replacement therapy, changes in plasma GHBP and those in BMI were significantly negatively correlated (r 5 20.51; P 5 0.02), whereas changes in GHBP and those in visceral AT and the sc AT areas at the abdominal or hip level were not correlated (r 5 0.06, r 5 0.35, and r 5 0.25, respectively). Discussion

To our knowledge, this is the first study in which the AT deposition in different parts of the body, as assessed by MRI scanning, is associated with circulating high affinity GHBP levels in a group of GHD patients and healthy subjects. MRI is a reliable and accurate technique for determining adipose tissue volume (22, 23), whereas circumference and skinfold thickness measurements have been found to be inadequate to determine AT areas in GHD patients (17). Our study shows that adult GHD patients have an increased visceral and sc abdominal AT area compared to healthy subjects matched for age and sex. Six months of GH replacement therapy reduced AT areas, with the most pronounced reduction in visceral AT, as has been previously shown by Bengtsson et al. (24) using computerized tomography. In addition, our data show, regardless of the GH status of the subjects, significant positive correlations between the BMI or the amount of visceral adipose tissue and plasma GHBP, whereas sc adipose deposition at both the abdominal and the hip level was not associated with plasma GHBP levels. Our

762

ROELEN ET AL.

JCE & M • 1997 Vol 82 • No 3

FIG. 1. Relationship between BMI (A), visceral AT area (B), sc AT area at the abdominal level (C), sc AT area at the hip level (D), and circulating GHBP levels. Least squares regression lines are depicted, regardless of the GH status of the subjects (n 5 34 in stead of n 5 36, because 2 posttreatment GHBP values are missing). ç, Male GHD patients; å, female GHD patients; É, male controls; Ç, female controls.

finding of an association of BMI with plasma GHBP (r 5 0.70) is consistent with previously reported studies (11, 12, 14 –16). Recently, Rasmussen et al. (13) showed that elevated GHBP levels in obese subjects were restored to normal by dietinduced weight loss. Next to the relationship between GHBP and BMI, the relationship between plasma GHBP and body fat distribution might be of relevance for the activity of the GH-IGF-I axis. In the study by Rasmussen et al. (13), multiple stepwise regression analysis revealed that changes in waist circumference and abdominal sagittal diameter during weight loss were the major determinants of and accounted for 54% of the fall in GHBP levels. In our study, linear regression analysis revealed that 53% (R2 5 0.53) of the variability in plasma GHBP could be explained by variation in visceral AT mass. Thus, 47% of the variability in GHBP is not explained; however, it is conceivable that other metabolic factors, such as insulin sensitivity, diet composition, and thyroid hormone metabolism, may be operative. Indeed, the positive association that we found between visceral fat and plasma GHBP levels suggests that visceral fat may be involved in the regulation of the plasma GHBP level. This reasoning is based on the reported increased lipolytic rate in visceral fat (25, 26), which, via increased plasma FFA levels, decreases GH release (27, 28), possibly resulting in an in-

creased GH sensitivity and up-regulation of the GH receptor. Based on our results and the assumption that GHBP is derived from proteolytic cleavage of the extracellular domain of the GH receptor (14, 29), it is tempting to speculate that an increase in plasma GHBP is a consequence of up-regulation of the GH receptor-binding sites in hyposomatotropic states such as GHD and obesity. Recent data concerning GH replacement therapy in GHD patients showed an interindividual variability in the GHBP response (30 –33). In this study we also found a variable response of plasma GHBP, ranging from 214% to 118% of the baseline value. Notwithstanding the small number of subjects in our study, pretreatment abdominal fat mass (i.e. visceral and sc AT) and the changes in plasma GHBP levels after GH replacement therapy were significantly positively correlated (r 5 0.82; P , 0.01 and r 5 0.75; P , 0.01), whereas no significant correlations were observed between changes in these AT areas and changes in GHBP. Thus, it seems that the pretreatment amount of abdominal fat mass partly determines the GHBP changes induced by GH replacement therapy. Therefore, it is conceivable that a low pretreatment abdominal fat mass shifts the dose-response curve of GH replacement therapy to the left, i.e. that a lower dose of GH is needed to be effective. However, this does not provide a

BODY FAT DISTRIBUTION AND PLASMA GHBP

763

FIG. 2. Relationship between the change in plasma GHBP levels after 6 months of GH replacement therapy in GHD patients and the pretreatment values of visceral AT area (A), sc AT area at the abdominal level (B), and sc AT area at the hip level (C). Least squares regression lines are depicted. å, Female subjects; ç, male subjects.

satisfactory explanation for why no correlations between the changes in AT areas and the changes in GHBP after GH therapy were found. It could be argued, however, that in a static condition concerning body fat (pretreatment), central fat mass and GHBP are significantly correlated, but that this correlation disappears in a dynamic state (GH treatment period), in which not only central fat mass and GHBP activity change, but other factors, such as insulin activity and catecholamines, change as well. Our observation that not only the visceral fat mass, but also the sc abdominal fat mass, albeit more weakly, were associated with changes in plasma GHBP, suggests that the sc abdominal fat mass is not only an energy depot, but may be metabolic active. Taken together, the data presented in this study clearly indicate a significant role of abdominal fat mass in circulating high affinity GHBP levels. If circulating GHBP levels are indicative of the GH receptor status of target tissues, an increase in abdominal fat may induce an increased density of GH receptor-binding sites, compensating for the decreased GH secretion in hyposomatotropic states.

2. 3.

4. 5. 6.

7. 8. 9.

10.

11.

References 1. Iranmanesh A, Lizarralde G, Veldhuis JD. 1991 Age and relative adiposity are specific negative determinants of the frequency and amplitude of growth

12.

hormone (GH) secretory bursts and the half-life of endogenous GH in healthy men. J Clin Endocrinol Metab. 73:1081–1088. Veldhuis JD. 1996 Gender differences in secretory activity of the human somatotropic (growth hormone) axis. Eur J Endocrinol. 134:287–295. Underwood LE, van Wijk JJ. 1992 Normal and aberrant growth. In: Wilson JD, Foster DW, eds. Williams textbook of endocrinology, 8th ed. Philadelphia: Saunders; 1079 –1138. Marcus C, Margery V, Kamel A, Bro¨nnegård M. 1994 Effects of growth hormone on lypolysis in humans. Acta Paediatr. 406(Suppl):54 –58. Bjo¨rntorp P. 1991 Metabolic implications of body fat distribution. Diabetes Care. 14:1132–1143. Despre´s JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C. 1990 Regional distribution of body fat, plasma lipoprotein, and cardiovascular disease. Arteriosclerosis. 10:497–511. Frayn KN. 1995 The importance of muscle and fat metabolism for complications of obesity. Int J Obesity. 19(Suppl 2):16 –17. Reaven GM. 1993 Role of insulin resistance in human disease (syndromeX)–an expanded definition. Annu Rev Med. 44:121–131. Veldhuis JD, Iranmanesh A, Ho KKY, Waters MJ, Johnson ML, Lizarralde G. 1991 Dual defects in pulsatile growth hormone secretion and clearance subserve the hyposomatotropism of obesity in man. J Clin Endocrinol Metab. 72:51–59. Jo¨rgenson JOL, Pedersen SB, Frystyk J, et al. 1995 Serum concentrations of insulin-like growth factors (IGFs), IGF binding proteins 1 and 3 and growth hormone binding protein in obese women and the effects of growth hormone administration: a double-blind, placebo-controlled study. Eur J Endocrinol. 133:65–70. Silbergeld A, Lazar L, Erster B, Keret R, Tepper R, Laron Z. 1989 Serum growth hormone binding protein activity in healthy neonates, children and young adults: correlations with age, height and weight. Clin Endocrinol (Oxf). 31:295–303. Martha Jr PM, Rogol AD, Blizzard RM, Shaw MA, Baumann G. 1991 Growth

764

13. 14. 15.

16.

17.

18.

19.

20. 21. 22.

ROELEN ET AL.

hormone-binding protein activity is inversely related to 24-hour growth hormone release in normal boys. J Clin Endocrinol Metab. 73:175–181. Rasmussen MH, Ho KKY, Kjems L, Hilsted J. 1996 Serum growth hormonebinding protein in obesity: effect of a short-term, very low calorie diet and diet-induced weight loss. J Clin Endocrinol Metab. 81:1519 –1524. Baumann G, Shaw MA, Amburn K. 1994 Circulating growth hormone binding proteins [Review]. J Endocrinol Invest. 17:67– 81. Counts DR, Gwirtsman H, Carlsson LMS, Lesem M, Butler GB. 1992 The effect of anorexia nervosa and refeeding on growth hormone binding protein, the insulin-like growth factors (IGFs), and the IGF-binding proteins. J Clin Endocrinol Metab. 75:762–767. Schaefer F, Baumann G, Haffner D, et al. 1996 Multifactorial control of the elimination kinetics of unbound (free) growth hormone (GH) in the human: regulation by age, adiposity, renal function, and steady state concentrations of GH in plasma. J Clin Endocrinol Metab. 81:22–31. Snel YEM, Brummer RJM, Doerga ME, et al. 1995 Adipose tissue by magnetic resonance imaging in growth hormone deficient adults: the effect of growth hormone replacement and a comparison with control subjects. Am J Clin Nutr. 61:1290 –1294. Roelen CAM, Donker GH, Thijssen JHH, Blankenstein MA. 1992 A method for measuring the binding affinity and capacity of high affinity growth hormone binding protein using FPLC to separate bound and free ligand. J Liquid Chromatogr. 15:1259 –1275. Davenport ML, Svoboda ME, Koerber KL, Van Wijk JJ, Clemmons DR, Underwood LE. 1988 Serum concentrations of insulin-like growth factor II are not changed by short term fasting and refeeding. J Clin Endocrinol Metab. 67:1231–1236. Furlanetto RW, Underwood LE, Van Wijk JJ, d’Ercole AJ. 1977 Estimation of somatomedin-C levels in normals and patients with pituitary disease by radioimmunoassay. J Clin Invest. 60:648 – 657. Lukaski HC, Johnson PE, Bolonchuk WW, Lykken GI. 1985 Assessment of fat-free mass using bioelectrical impedance measurements of the human body. Am J Clin Nutr. 41:810 – 817. Abate N, Burns D, Peshock RM, Garg A, Grundy SM. 1994 Estimation of adipose tissue mass by magnetic resonance imaging: validation against dissection in human cadavers. J Lipid Res. 35:1490 –1496.

JCE & M • 1997 Vol 82 • No 3

23. Fowler PA, Fuller MF, Glasbey CA, Cameron GG, Foster MA. 1992 Validation of the in vivo measurement of adipose tissue by magnetic resonance imaging of lean and obese pigs. Am J Clin Nutr. 56:7–13. 24. Bengtsson BÅ, Ede´n S, Lo¨nn L, et al. 1993 Treatment of adults with growth hormone (GH) deficiency with recombinant human GH. J Clin Endocrinol Metab. 76:309 –317. 25. Krotkiewski M, Bjo¨rntorp P, Sjo¨stro¨m L, Smith U. 1983 Impact of obesity in metabolism in men and women: importance of regional adipose tissue distribution. J Clin Invest. 72:1150 –1162. 26. Rebuffe´-Scrive M, Andersson B, Olbe L, Bjo¨rntorp P. 1989 Metabolism of adipose tissue in intraabdominal depots of nonobese men and women. Metabolism. 38:453– 458. 27. Imaki T, Shibasaki T, Shizume K, et al. 1985 The effect of free fatty acids on growth hormone-releasing hormone-mediated GH secretion in man. J Clin Endocrinol Metab. 60:290 –293. 28. Casanueva FF, Villanueva L, Dieguez C, et al. 1987 Free fatty acids block growth hormone (GH) releasing hormone-stimulated GH secretion in man directly at the pituitary. J Clin Endocrinol Metab. 65:634 – 642. 29. Leung DW, Spencer SA, Cachianes G, et al. 1987 Growth hormone receptor and serum binding protein: purification, cloning, and expression. Nature. 330:537–543. 30. Davila N, Alcaniz J, Salto L, Estrada J, Barcelo B, Baumann G. 1994 Serum growth hormone-binding protein is unchanged in adult panhypopituitarism. J Clin Endocrinol Metab. 79:1347–1350. 31. Ho KKY, Jo¨rgensen JOL, Valiontis E, Waters MJ, Rajikovic IA, Christiansen JS. 1993 Different modes of growth hormone administration do not change GH binding protein activity in man. Clin Endocrinol (Oxf). 143–148. 32. Martha PM, Reiter EO, Davila N, Shaw MA, Holcombe JH, Baumann G. 1993 Serum growth hormone (GH)-binding protein/receptor: an important determinant of GH responsiveness. J Clin Endocrinol Metab. 75:1464 –1469. 33. Roelen CA, Koppeschaar HPF, de Vries WR, et al. 1996 High affinity growth hormone-binding protein, insulin-like growth factor-I and insulin-like growth factor-binding protein-3 in adults with growth hormone deficiency. Eur J Endocrinol. 135:82– 86.