Very-Low-Calorie Hormone Secretion
Katsuaki Tanaka,
Diet-Induced Weight Reduction Reverses Impaired Growth Response to Growth Hormone-Releasing Hormone, Arginine, and L-Dopa in Obesity
Shuji Inoue, Kazushi Numata,
Hiroshi Okazaki,
Saburo Nakamura,
and Yutaro Takamura
To determine whether impaired growth hormone (GH) secretion in obese subjects is a consequence of obesity or a pre-existing pituitary-hypothalamic disorder, we measured (1) plasma GH response to growth hormone-releasing hormone (GRH: 1 Mg/kg body weight [BW]), arginine (0.5 g/kg BW), and L-dopa (500 mgj: and (2) plasma glucose, insulin, and free fatty acids (FFA) in obese subjects before and after weight reduction due to very-low-calorie diet therapy using Optifast (240 kcal/d for 8 to 12 weeks). Body weight and body mass index (BMI) values before and after weight reduction were 87.2 f 4.1 kg and 34.5 -r- 0.9 kg/m’. and 67.6 * 2.7 kg and 27.0 + 0.4 kg/m2. respectively. OH response to GRH, arginine, and L-dopa in obese subjects was markedly impaired before weight reduction, whereas significantly increased responses were noted after weight reduction (P < .Ol). Impaired integrated GH response to GRH, arginine, and L-dopa in obese subjects was significantly restored after weight reduction (P c .Ol 1. Plasma glucose levels did not change, while plasma insulin and FFA levels decreased significantly after weight reduction (P < .Ol, P < .05). There was no significant correlation between integrated GH response to these three stimuli and plasma levels of glucose, insulin, and FFA, respectively. The reversibility of GH response to all three stimuli after weight reduction suggests that impaired GH secretion is a consequence of obesity rather than a pre-existing pituitary-hypothalamic disorder. 0 1990 by W. B. Saunders Company.
I
T IS WELL KNOWN that obese subjects exhibit diminished growth hormone (GH) secretion in response to a wide variety of GH secretion stimulators.’ Impaired GH secretion in response to growth hormone-releasing hormone (GRH),’ arginine,3 and L-dopa has been reported in obese adults, and obese children show impaired GH secretion to GRH’ and arginine.6 Causes have not yet been fully elucidated. Impaired GH output in obese subjects appears directly related to the obesity itself. Reports indicate obese subjects show partial or complete restoration of normal GH response to both GRH’ and arginine’ after successful weight reduction. However, no reports have investigated the effect of weight reduction on L-dopa-stimulated GH response. GRH acts directly on the pituitary to cause GH release,’ while arginine and L-dopa act within the central nervous system (CNS) (hypothalamus). Plasma GRH increases after but does not increase after intravenous L-dopa testing,“,” (IV) arginine testing.‘* Mechanisms of GH response to these three stimuli are therefore different. It is known that glucose and FFA suppress GH secretion.’ The administration of glucose IV is immediately followed by a decrease in plasma GH, usually to undetectable levels.‘3,‘4 Furthermore, IV or orally administered glucose inhibits the FFA elevation inhibits the GH GH response to GRH.“.” response to GRH,” arginine,“,*’ and r-dopa.*’ It is possible that impaired responses seen in obesity could be due, at least in part, to obesity-induced metabolic alterations of these substrates.’ The present study was designed to elucidate the etiology of
From the Third Department of Internal Medicine, Yokohama City University School of Medicine, Urafune-cho. Yokohama, Japan. Address reprint requests to Shuji Inoue, MD. Third Department of Internal Medicine. Yokohama City University School of Medicine, 3-46 Urafune-cho. Minami-ku. Yokohama 232, Japan. @ 1990 by W.B. Saunders Company. 0026-0495/90/3909-0002$03.00/0
892
impaired GH response to these three different stimuli using very-low-calorie diet therapy in obese subjects. MATERIALS AND METHODS Subjects
Informed consent was obtained from 15 obese subjects (four men, age 20 to 28 years, and 11 women, age 16 to 29 years) studied at the Yokohama City University Hospital. The degree of obesity was estimated by the body mass index (BMI = kg/m’). Body weight and BMI values of obese subjects ranged from 6 1.5 to 124 kg (mean, 87.2 + 4.1 kg) and 30.2 to 41.2 kg/m* (mean, 34.5 + 0.9 kg/m’), respectively. All subjects showed normal glucose tolerance and took no medications during the study. After admission to the Yokohama City University Hospital, subjects were fed a low-calorie diet (1.200 kcal/d) for at least 1 week and underwent three separate experimental evaluations (described later). Examinations were performed at minimum 2-day intervals. After evaluation, subjects received a very-low-calorie diet (Optifast-70, Sandoz Pharmaceuticals, Tokyo, Japan, 240 kcal/d) containing 18 g of carbohydrate, 42 g of protein, and 1.2 g essential fatty acids for 8 to 12 weeks. The formula diet is a powdered preparation reconstituted to a liquid with approximately 700 mL water per serving. After therapy. subjects received a low-calorie diet (1,200 kcal/d) for 1 week and received three further evaluations.
Experimental Design All subjects fasted 12 hours the morning of the test. Thirty minutes before injection of the stimulus, an indwelling venous catheter for blood sampling was placed in a forearm vein. Subjects remained awake and supine during the experiments. Synthetic human pancreatic GRH(l-44)NH, (1 rg/kg body weight [BW], Sumitomo Pharmaceuticals, Osaka, Japan) was rapidly injected IV between 8:00 AM and 9:00 AM. Blood samples were obtained immediately before and 15, 30, 45, 60, 90, and 120 minutes after GRH injection. Arginine testing consisted of IV infusion of L-arginine monohydrochloride (0.5 g/kg BW) for 30 minutes. Blood samples were obtained immediately before and 30, 60,90, and 120 minutes after injection. L-dopa testing was achieved by oral administration of L-dopa (0.5 g). Blood samples were obtained immediately before and 30, 60, 90, and 120 minutes after ingestion. Collected blood samples were placed in heparinized tubes
Metabolism.
Vol39,
No 9 (September), 1990: pp 892-896
GH SECRETION IN OBESITY
893
and immediately chilled and centrifuged at 4“C; plasma was stored at - 20°C until assay.
‘ny’m’l 20-
GRH(1 +
/ly/ky, *
Assay Plasma GH concentration was determined with a radioimmunoassay kit (Eiken Immunochemical Laboratory, Tokyo, Japan). Plasma glucose was determined by the glucose oxidase method. Plasma insulin was determined with a radioimmunoassay kit (Amersham, Tokyo, Japan). Plasma FFA was determined by an enzyme method.” Statistical Analysis Data are expressed as mean c SE. Integrated GH values were calculated using the trapezoidal rule. For statistical comparison, differences between and within groups were tested using two-way ANOVA and Student’s f test for paired data.
15I . .. . . .. .. . . ..
so-
5-
O-
I 0
o Obese pre-VLCLI
**
1 15
..**
l
0 Obese post-VLCO
1 . ._
I 30
.
**
**
SI . . .
1 45
l.
l.
‘.
1 60
I 90
1 120 (minutes)
TIME
RESULTS
Characteristics of Obese Subjects Before and After Weight Reduction Clinical characteristics of obese subjects before and after a very-low-calorie diet therapy are shown in Table 1. Body weight and BMI values ranged from 61.5 to 124 kg (mean, 87.2 -t 4.1 kg) and 30.2 to 41.2 kg/m2 (mean, 34.5 * 0.9 kg/m*) before, and from 47 to 85 kg (mean, 67.8 + 2.7 kg) and 20.4 to 32.6 kg/m2 (mean, 27.0 * 0.4 kg/m2) after weight reduction. Plasma glucose levels did not change, while plasma insulin and free fatty acid (FFA) levels decreased significantly after weight reduction (P < .Ol, P < .05). Plasma GH Response to GRH Plasma GH response to GRH in obese subjects before and after very-low-calorie diet therapy is shown in Fig 1. Before reduction, subjects showed a markedly diminished GH response to GRH. After reduction, significantly increased responses in all samples were noted (P < .05 to P < .Ol). Table 2 shows time-sampled response increases over a period of 2 hours. After weight reduction, integrated GH secretion in response to GRH was significantly raised in both first- and second-hour samples. Figure 2 shows the changes of integrated GH secretion of individuals in response to GRH before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to GRH; however, after reduction, increased responses were noted in all samples.
Fig 1. Mean I+SE) plasma GH responses to GRH in obese subjects before and after weight reduction due to very-lowcalorie diet (VLCDI therapy. l*P < .Ol, lP < .06, differences before and after weight reduction.
Plasma GH Response to Arginine Plasma GH response to IV arginine injection for 30 minutes before and after very-low-calorie diet therapy is shown in Fig 3. Before reduction, obese subjects showed diminished GH response to IV arginine. After reduction, significantly increased responses were noted at 60, 90, and 120 minutes post-injection (P < .05 to P < .Ol). Table 2 shows that, after weight reduction, integrated GH secretion in response to arginine was significantly raised in both firstand second-hour samples. Figure 4 shows the changes of integrated GH secretion of individuals in response to arginine before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to arginine; however, after reduction, increased responses were noted in all samples. Plasma GH Response to L-dopa Plasma GH response to L-dopa before and after very-lowcalorie diet therapy is shown in Fig 5. Before reduction, obese Table 2. Integrated GH Secretion in Response to GRH, Arginine. and L-dopa in Obese Subjects Before and After Weight Reduction Due to Very-Low-Calorie
Diet Therapy
lntmated GH Secretion(ng/mL . min) Table 1. Characteristics of Obese Subjects Before and After Weight Reduction Due to Very-Low-Calorie
O-60 min
Diet Therapy
60-120 min
O-120
min
GRH Before
Before
242.1
f. 26.7
200.7
+ 24.9
442.8
16-29
After
700.5
t 136.4.
422.5
f 85.8t
1123.0
? 197.0.
4 men, 11 women
Arginine Before
302.4
* 65.2
409.6
+ 52.4
712.2
” 105.6
After
574.9
+ 73.2*
842.1
+ 80.8.
1417.0
Age lyr) Sex
Aftef
67.2 + 4.1
67.6 + 2.7*
BMI (kg/m’)
34.5
+ 0.9
27.0
+ 0.4*
Fasting plasma glucose (mg/dL)
97.4
f 3.8
88.9
+ 4.7
Body weight
(kg1
Fasting plasma insulin (pU/mL) Fasting plasma FFAs (pEq/L)
16.4 + 1.2 765.3
f 99.6
9.0 + 1.8. 303.7
* 34.q
NOTE. Values given are the mean + SE. lP -z .Ol,
tP -z .05,
differences before and after weight reduction.
t 42.0
+ 136.1.
L-dopa Before After
54.8 192.7
f 9.2
135.8
+ 14.2
190.6
+ 20.2
f 33.0*
463.4
+ 75.6t
656.1
? 101.3’
NOTE. Values given are the mean + SE. lf < .Ol, tP < .05, differences before and after weight reduction.
894
TANAKA
0
hg/ml*minl 2500
z* S % g % s
(ng/ml*minl
Obese pre-VLCO
1
2500
0 Obese post-VLCO
1
l
0
Obesepre-VLCD
2000-
1500-
= m s E
IOOO-
g
50001 20
25
I
I
30 35 Body Mass Index (kg/m?
40
o Obese pre-VLCD
. Obesepost-VLCO
Arginine(O.Sg/kg) (iiiiijiiiiijijiiiiiiiiiiiil
= :
:
:
:
:
:
Fig 4. Changes of integrated GH secretion of individuals in response to arginine before and after weight reduction due to very-low-calorie diet (VLCD) therapy.
Correlations Among GH Response to GH Secretion Stimulators, Degree of Weight Loss, and Metabolic Substrates A direct correlation was found between the percent increase of the integrated GH response to GRH after weight reduction and to L-dopa (r = ,660, P < .05) and between the percent increase to arginine and to r-dopa (r = ,652, P < .05). However, no correlation was observed between the degree of weight loss and the enhancement of plasma GH secretion in response to GRH, arginine, and L-dopa, respectively. There was no significant correlation between the integrated GH response to GRH, arginine, L-dopa and plasma levels of glucose, insulin, and FFA, respectively (data not shown). Ing/mH 2ol
.*** +. ..
:
45 Body Mass Index (kg/m21
subjects showed a markedly diminished GH response to L-dopa. After reduction, significantly increased responses were noted at 60.90, and 120 minutes post-ingestion (P < .05 to P < .Ol). Table 2 shows that, after weight reduction, integrated GH secretion in response to L-dopa was significantly raised in both first- and second-hour samples. Figure 6 shows the changes of integrated GH secretion of individuals in response to L-dopa before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to L-dopa; however, after reduction, increased responses were noted in all samples.
**
O2X
45
Fig 2. Changes of integrated GH secretion of individuals in response to GRH before and after weight reduction due to very-low-calorie diet (VLCD) therapy.
z
ET AL
o
Obese pre-VLCO
0
Obese post-VLCO
** ..
..
J,
l.
l.
l.
10 ;-I/+
I 0
I 30
I 60 TIME
I 90
I 120 (minutes)
Fig 3. Mean (+ SE) plasma GH responses to arginine in obese subjects before and after weight reduction due to very-lowcalorie diet WLCD) therapy. l*P < .Ol, lP -c .05. differences before and after weight reduction.
0
.*. . . -. . . . . J? I
0
30
I
60
1
90
I
120 (minutes)
Fig 5. Mean (*SE) plasma GH responses to L-dopa in obese subjects before and after weight reduction due to very-lowcalorie diet (VLCD) therapy. l*P < .Ol. lP -c .05, differences before and after weight reduction.
GH SECRETION IN OBESITY
895
hg/ml*min) 0 Obese pm-VLCD 0 Obese post-VLCIJ
2000
1
i5
$0
i5
40
Body Mass Index (kg/m21 Changes of integrated GH secretion of individuals in Fig 6. response to L-dope before and after weight reduction due to very-low-calorie diet WLCD) therapy.
DISCUSSION Secretion of GH from the pituitary results from the net effect of CNS inputs by stimulatory and inhibitory factors (GRH and somatostatin C[SRIF]) and the responsiveness of the pituitary to these factors. GRH, characterized and synthetized by Guillemin et a123and Rivier et a1,24stimulates GH release at the pituitary level, while both arginine and L-dopa stimulate release at the suprapituitary level (hypothalamus). It has been assumed that GH secretion to orally administered L-dopa is mediated by release of endogenous GRH from the hypothalamus.“.” In the present study, GH secretion in response to GRH and L-dopa is markedly attenuated in obese subjects and the percent increase of the integrated GH response to GRH was correlated with that to L-dopa, suggesting impaired GH response to endogenous GRH is a likely explanation for the inadequate response to L-dopa. On the other hand, GH secretion to arginine is not mediated by release of endogenous GRH,” and there is a possibility that suppression of endogenous SRIF secretion may influence GH secretion in response to arginine. GH secretion in response to arginine was as markedly attenuated as that to GRH and L-dopa in obese subjects in the present study. This suggests that modulation of endogenous SRIF release may be involved in obese subjects.’ Reversibility of GH response to GRH, arginine, and L-dopa after weight reduction may be explained by the obesity-related increase in endogenous SRIF release into the portal capillary delivery system. Berelowitz et a125 reported that SRIF release from the hypothalamus is increased and that pituitary GH response to GRH is impaired in genetically obese Zucker rats
in vitro. Therefore, we speculate that the present results may be explained by increased release of SRIF into the portal capillary delivery system, resulting in impaired GH response to these three stimuli. Other possibilities include impaired pituitary response to GRH, arginine, and L-dopa due to (1) decreased GH stores, and (2) elevated levels of metabolic substrates associated with obesity, eg, glucose and FFA. The former explanation is based on the fact that GH stores decrease in rats with ventromedial hypothalamic lesioninduced obesity26-28 and monosodium glutamate treatment.29 However, in obese humans, it is not clear whether the GH pool within the pituitary or the hypothalamic GRH content decreases. Since impaired GH response to L-dopa or arginine in obese subjects can be augmented by propranolol,30 this possibility may be less plausible. The latter explanation depends on the fact that glucose and FFA inhibit GH response.‘3-2’ In the present study, plasma glucose did not decrease significantly after weight reduction, while plasma FFA levels decreased significantly. Therefore, it may be postulated that elevated FFA levels before weight reduction may account for the suppression of GH. However, we found no significant correlation between plasma GH response and plasma levels of glucose and FFA. Irie et a13’ reported that reduction of FFA levels in obese subjects by administration of nicotinic acid fails to increase basal GH levels. Thus, altered metabolic FFA found in obese subjects is not the major cause of impaired GH response to multiple stimuli. No correlation between the degree of weight loss and the enhancement of plasma GH secretion in response to GRH, arginine, and L-dopa was observed in spite of the reversibility of GH response to these three stimuli after weight reduction. Although the reason is so far unexplained, one possibility is that the pre-very-low-calorie diet group and post-very-lowcalorie diet group contain various stages of the degree of obesity. The other possibility would be due to the effect of the very-low-calorie diet. Kelijman and Frohman3* reported that acute nutrient deprivation enhanced GH responses to GRH in obesity independently of weight loss. In order to abolish the effect of the previous very-low-calorie diet, GH stimulation tests were performed following a l-week period on a 1,200-kcal diet after a very-low-calorie diet therapy in the present study. However, we can not deny the possibility that the effect of the very-low-calorie diet on plasma GH responsiveness was not eliminated. In summary, the present study has shown a reversible defect in growth hormone secretion in response to GRH, arginine, and L-dopa in obese subjects. Although the specific mechanisms responsible for this alteration await full elucidation, reversibility of the secretory defect with weight loss suggests an acquired defect secondary to a hormonalmetabolic alteration, rather than a primary defect involved in the pathogenesis of obesity.
REFERENCES
I. Glass AR, Burman KD, Dahms WT. et al: Endocrine function in human obesity. Metabolism 30:89-104, 1981 2. Kopelman PG, Noonan K, Goulton R, et al: Impaired growth
hormone response to growth hormone releasing factor and insulinhypoglycaemia in obesity. Clin Endocrinol23:87-94, 1985 3. Copinschi G, Wegienka LC, Hane S, et al: Effect of arginine on
TANAKA
896
serum levels of insulin and growth Metabolism 16:485-491, 1967
hormone
in obese
subjects.
4. Laurian L, Oberman Z, Ayalon D, et al: Under-responsiveness of growth hormone secretion after L-DOPA and deep sleep stimulation in obese subjects. Isr J Med Sci 11:482-487, 1975 5. Rosskamp R, Becker M, Soetadji S: Circulating somatomedin-C levels and the effect of growth hormone-releasing factor on plasma levels of growth hormone and somatostatin-like immunoreactivity in obese children. Eur J Pediatr 146:48-SO,1987 6. Josefsberg Z, Kauli R, Keret R, et al: Growth hormone response to insulin tolerance test and arginine stimulation in obese children and adolescents. Pediatr Adolesc Endocrinol 1: 146- 152, 1976 7. Williams T, Berelowitz M, Joffe SN, et al: Impaired growth hormone responses to growth hormone-releasing factor in obesity: A pituitary defect reversed with weight reduction. N Engl J Med 311:1403-1407, 1984 8. El-Khodary AZ, Ball MF, Stein B, et al: Effect of weight loss on the growth hormone response to arginine infusion in obesity. J Clin Endocrinol32:42-5 1, 197 1 9. Frohman LA, Jansson J-O: Growth mone. Endocrine Rev 7:223-253, 1986
hormone-releasing
hor-
10. Saito H, Saito S, Yamazaki R, et al: Clinical value of radioimmunoassay of plasma growth-hormone-releasing factor. Lancet 18:401-402, 1984 11. Chihara K, Kashio Y, Kita T, et al: L-dopa stimulates release of hypothalamic growth hormone-releasing hormone in humans. J Clin Endocrinol Metab 62:466-473, 1986 12. Chihara K, Kashio Y, Okimura Y, et al: Pathophysiology of growth hormone-releasing hormone (GHRH) secretion in humans. Endocrinol Japon 34:9-30, 1987 (suppl 1) 13. Roth J, Glick SM, Yalow RS, et al: Hypoglycemia: A potent stimulus to secretion of growth hormone. Science 140:987-988, 1963 14. Jansz A, Doorenbos H, Reitsma WD: Effect of food intake of growth-hormone level. Lancet 1:250-251, 1963 15. Press M, Tamborlane WV, Thorner MO, et al: Pituitary response to growth hormone-releasing factor in diabetes. Failure of glucose-mediated suppression. Diabetes 33:804-806, 1984 16. Sharp PS, Foley K, Chahal P, et al: The effect of plasma glucose on the growth hormone response to human pancreatic growth hormone releasing factor in normal subjects. Clin Endocrinol (Oxf) 20:497-501, 1984 17. Masuda A, Shibasaki T, Nakahara M, et al: The effect of glucose on growth hormone (GH)-releasing hormone-mediated GH secretion in man. J Clin Endocrinol Metab 60:523-526, 1985 18. Imaki
T, Shibasaki
T, Shizume
K, et al: The effect of free
ET AL
fatty acids on growth hormone (GH)-releasing hormone-mediated GH secretion in man. J Clin Endocrinol Metab 60:290-293, 1985 19. Blackard WG, Hull EW, Lopez-S A: Effect of lipids on growth hormone secretion in humans. J Clin Invest 50:1439-1443, 1971 20. Fineberg SE, Horland AA, Merimee TJ: Free fatty acid concentrations and growth hormone secretion in man. Metabolism 21:491-497, 1972 21. Fraser WM, Blackard WG: The effect of lipid on prolactin and growth hormone secretion. Horm Metab Res 9:389-394, 1977 22. Mizuno K, Toyosato M, Yabumoto S, et al: A new enzymatic method for calorimetric determination of free fatty acids. Ann Biochem 108:6-10, 1980 23. Guillemin R, Brazeau P, Bohlen P, et al: Growth hormonereleasing factor from a human pancreatic tumor that caused acromegaly. Science 218:585-587, 1982 24. Rivier J, Spiess J, Thorner M, et al: Characterization of a growth hormone-releasing factor from a human pancreatic islet tumour. Nature 300:276-278, 1982 25. Berelowitz M, Finkelstein JA, Thominet JL, et al: Impaired growth hormone secretion in spontaneous experimental obesity: A result of increased somatostatin (SRIF) tone. Presented at the Annual Meeting of the Endocrine Society, San Antonio, TX, June 8-10, 1983 26. Frohman LA, Bernardis LL: Growth hormone and insulin levels in weanling rats with ventromedial hypothalamic lesions. Endocrinology 82: 1125- 1132, 1968 27. Wakabayashi I, Inoue S, Satoh S, et al: Effect of hypothalamic ventromedial lesions on plasma growth hormone response to growth hormone-releasing factor in rats. Brain Res 346:70-74, 1985 28. Eikelboom R, Tannenbaum GS: Effects of obesity-inducing ventromedial hypothalamic lesions on pulsatile growth hormone and insulin secretion: Evidence for the existence of a growth hormonereleasing factor. Endocrinology 112:212-219, 1983 29. Millard WJ, Martin JB Jr, Audet J, et al: Evidence that reduced growth hormone secretion observed in monosodium glutamate-treated rats is the result of a deficiency in growth hormonereleasing factor. Endocrinology 110:540-550, 1982 30. Barbarino A, De Marinis L, Troncone L: Growth hormone response to propranolol and L-dopa in obese subjects. Metabolism 271275-278, 1978 31. Irie M, Tsushima T, Sakuma M: Effect of nicotinic acid administration on plasma HGH, FFA and glucose in obese subjects and in hypopituitary patients. Metabolism 19:972-979, 1970 32. Kelijman M, Frohman LA: Enhanced growth hormone (GH) responsiveness to GH-releasing hormone after dietary manipulation in obese and nonobese subjects. J Clin Endocrinol Metab 66:489494, 1988
Katsuaki Tanaka,
Diet-Induced Weight Reduction Reverses Impaired Growth Response to Growth Hormone-Releasing Hormone, Arginine, and L-Dopa in Obesity
Shuji Inoue, Kazushi Numata,
Hiroshi Okazaki,
Saburo Nakamura,
and Yutaro Takamura
To determine whether impaired growth hormone (GH) secretion in obese subjects is a consequence of obesity or a pre-existing pituitary-hypothalamic disorder, we measured (1) plasma GH response to growth hormone-releasing hormone (GRH: 1 Mg/kg body weight [BW]), arginine (0.5 g/kg BW), and L-dopa (500 mgj: and (2) plasma glucose, insulin, and free fatty acids (FFA) in obese subjects before and after weight reduction due to very-low-calorie diet therapy using Optifast (240 kcal/d for 8 to 12 weeks). Body weight and body mass index (BMI) values before and after weight reduction were 87.2 f 4.1 kg and 34.5 -r- 0.9 kg/m’. and 67.6 * 2.7 kg and 27.0 + 0.4 kg/m2. respectively. OH response to GRH, arginine, and L-dopa in obese subjects was markedly impaired before weight reduction, whereas significantly increased responses were noted after weight reduction (P < .Ol). Impaired integrated GH response to GRH, arginine, and L-dopa in obese subjects was significantly restored after weight reduction (P c .Ol 1. Plasma glucose levels did not change, while plasma insulin and FFA levels decreased significantly after weight reduction (P < .Ol, P < .05). There was no significant correlation between integrated GH response to these three stimuli and plasma levels of glucose, insulin, and FFA, respectively. The reversibility of GH response to all three stimuli after weight reduction suggests that impaired GH secretion is a consequence of obesity rather than a pre-existing pituitary-hypothalamic disorder. 0 1990 by W. B. Saunders Company.
I
T IS WELL KNOWN that obese subjects exhibit diminished growth hormone (GH) secretion in response to a wide variety of GH secretion stimulators.’ Impaired GH secretion in response to growth hormone-releasing hormone (GRH),’ arginine,3 and L-dopa has been reported in obese adults, and obese children show impaired GH secretion to GRH’ and arginine.6 Causes have not yet been fully elucidated. Impaired GH output in obese subjects appears directly related to the obesity itself. Reports indicate obese subjects show partial or complete restoration of normal GH response to both GRH’ and arginine’ after successful weight reduction. However, no reports have investigated the effect of weight reduction on L-dopa-stimulated GH response. GRH acts directly on the pituitary to cause GH release,’ while arginine and L-dopa act within the central nervous system (CNS) (hypothalamus). Plasma GRH increases after but does not increase after intravenous L-dopa testing,“,” (IV) arginine testing.‘* Mechanisms of GH response to these three stimuli are therefore different. It is known that glucose and FFA suppress GH secretion.’ The administration of glucose IV is immediately followed by a decrease in plasma GH, usually to undetectable levels.‘3,‘4 Furthermore, IV or orally administered glucose inhibits the FFA elevation inhibits the GH GH response to GRH.“.” response to GRH,” arginine,“,*’ and r-dopa.*’ It is possible that impaired responses seen in obesity could be due, at least in part, to obesity-induced metabolic alterations of these substrates.’ The present study was designed to elucidate the etiology of
From the Third Department of Internal Medicine, Yokohama City University School of Medicine, Urafune-cho. Yokohama, Japan. Address reprint requests to Shuji Inoue, MD. Third Department of Internal Medicine. Yokohama City University School of Medicine, 3-46 Urafune-cho. Minami-ku. Yokohama 232, Japan. @ 1990 by W.B. Saunders Company. 0026-0495/90/3909-0002$03.00/0
892
impaired GH response to these three different stimuli using very-low-calorie diet therapy in obese subjects. MATERIALS AND METHODS Subjects
Informed consent was obtained from 15 obese subjects (four men, age 20 to 28 years, and 11 women, age 16 to 29 years) studied at the Yokohama City University Hospital. The degree of obesity was estimated by the body mass index (BMI = kg/m’). Body weight and BMI values of obese subjects ranged from 6 1.5 to 124 kg (mean, 87.2 + 4.1 kg) and 30.2 to 41.2 kg/m* (mean, 34.5 + 0.9 kg/m’), respectively. All subjects showed normal glucose tolerance and took no medications during the study. After admission to the Yokohama City University Hospital, subjects were fed a low-calorie diet (1.200 kcal/d) for at least 1 week and underwent three separate experimental evaluations (described later). Examinations were performed at minimum 2-day intervals. After evaluation, subjects received a very-low-calorie diet (Optifast-70, Sandoz Pharmaceuticals, Tokyo, Japan, 240 kcal/d) containing 18 g of carbohydrate, 42 g of protein, and 1.2 g essential fatty acids for 8 to 12 weeks. The formula diet is a powdered preparation reconstituted to a liquid with approximately 700 mL water per serving. After therapy. subjects received a low-calorie diet (1,200 kcal/d) for 1 week and received three further evaluations.
Experimental Design All subjects fasted 12 hours the morning of the test. Thirty minutes before injection of the stimulus, an indwelling venous catheter for blood sampling was placed in a forearm vein. Subjects remained awake and supine during the experiments. Synthetic human pancreatic GRH(l-44)NH, (1 rg/kg body weight [BW], Sumitomo Pharmaceuticals, Osaka, Japan) was rapidly injected IV between 8:00 AM and 9:00 AM. Blood samples were obtained immediately before and 15, 30, 45, 60, 90, and 120 minutes after GRH injection. Arginine testing consisted of IV infusion of L-arginine monohydrochloride (0.5 g/kg BW) for 30 minutes. Blood samples were obtained immediately before and 30, 60,90, and 120 minutes after injection. L-dopa testing was achieved by oral administration of L-dopa (0.5 g). Blood samples were obtained immediately before and 30, 60, 90, and 120 minutes after ingestion. Collected blood samples were placed in heparinized tubes
Metabolism.
Vol39,
No 9 (September), 1990: pp 892-896
GH SECRETION IN OBESITY
893
and immediately chilled and centrifuged at 4“C; plasma was stored at - 20°C until assay.
‘ny’m’l 20-
GRH(1 +
/ly/ky, *
Assay Plasma GH concentration was determined with a radioimmunoassay kit (Eiken Immunochemical Laboratory, Tokyo, Japan). Plasma glucose was determined by the glucose oxidase method. Plasma insulin was determined with a radioimmunoassay kit (Amersham, Tokyo, Japan). Plasma FFA was determined by an enzyme method.” Statistical Analysis Data are expressed as mean c SE. Integrated GH values were calculated using the trapezoidal rule. For statistical comparison, differences between and within groups were tested using two-way ANOVA and Student’s f test for paired data.
15I . .. . . .. .. . . ..
so-
5-
O-
I 0
o Obese pre-VLCLI
**
1 15
..**
l
0 Obese post-VLCO
1 . ._
I 30
.
**
**
SI . . .
1 45
l.
l.
‘.
1 60
I 90
1 120 (minutes)
TIME
RESULTS
Characteristics of Obese Subjects Before and After Weight Reduction Clinical characteristics of obese subjects before and after a very-low-calorie diet therapy are shown in Table 1. Body weight and BMI values ranged from 61.5 to 124 kg (mean, 87.2 -t 4.1 kg) and 30.2 to 41.2 kg/m2 (mean, 34.5 * 0.9 kg/m*) before, and from 47 to 85 kg (mean, 67.8 + 2.7 kg) and 20.4 to 32.6 kg/m2 (mean, 27.0 * 0.4 kg/m2) after weight reduction. Plasma glucose levels did not change, while plasma insulin and free fatty acid (FFA) levels decreased significantly after weight reduction (P < .Ol, P < .05). Plasma GH Response to GRH Plasma GH response to GRH in obese subjects before and after very-low-calorie diet therapy is shown in Fig 1. Before reduction, subjects showed a markedly diminished GH response to GRH. After reduction, significantly increased responses in all samples were noted (P < .05 to P < .Ol). Table 2 shows time-sampled response increases over a period of 2 hours. After weight reduction, integrated GH secretion in response to GRH was significantly raised in both first- and second-hour samples. Figure 2 shows the changes of integrated GH secretion of individuals in response to GRH before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to GRH; however, after reduction, increased responses were noted in all samples.
Fig 1. Mean I+SE) plasma GH responses to GRH in obese subjects before and after weight reduction due to very-lowcalorie diet (VLCDI therapy. l*P < .Ol, lP < .06, differences before and after weight reduction.
Plasma GH Response to Arginine Plasma GH response to IV arginine injection for 30 minutes before and after very-low-calorie diet therapy is shown in Fig 3. Before reduction, obese subjects showed diminished GH response to IV arginine. After reduction, significantly increased responses were noted at 60, 90, and 120 minutes post-injection (P < .05 to P < .Ol). Table 2 shows that, after weight reduction, integrated GH secretion in response to arginine was significantly raised in both firstand second-hour samples. Figure 4 shows the changes of integrated GH secretion of individuals in response to arginine before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to arginine; however, after reduction, increased responses were noted in all samples. Plasma GH Response to L-dopa Plasma GH response to L-dopa before and after very-lowcalorie diet therapy is shown in Fig 5. Before reduction, obese Table 2. Integrated GH Secretion in Response to GRH, Arginine. and L-dopa in Obese Subjects Before and After Weight Reduction Due to Very-Low-Calorie
Diet Therapy
lntmated GH Secretion(ng/mL . min) Table 1. Characteristics of Obese Subjects Before and After Weight Reduction Due to Very-Low-Calorie
O-60 min
Diet Therapy
60-120 min
O-120
min
GRH Before
Before
242.1
f. 26.7
200.7
+ 24.9
442.8
16-29
After
700.5
t 136.4.
422.5
f 85.8t
1123.0
? 197.0.
4 men, 11 women
Arginine Before
302.4
* 65.2
409.6
+ 52.4
712.2
” 105.6
After
574.9
+ 73.2*
842.1
+ 80.8.
1417.0
Age lyr) Sex
Aftef
67.2 + 4.1
67.6 + 2.7*
BMI (kg/m’)
34.5
+ 0.9
27.0
+ 0.4*
Fasting plasma glucose (mg/dL)
97.4
f 3.8
88.9
+ 4.7
Body weight
(kg1
Fasting plasma insulin (pU/mL) Fasting plasma FFAs (pEq/L)
16.4 + 1.2 765.3
f 99.6
9.0 + 1.8. 303.7
* 34.q
NOTE. Values given are the mean + SE. lP -z .Ol,
tP -z .05,
differences before and after weight reduction.
t 42.0
+ 136.1.
L-dopa Before After
54.8 192.7
f 9.2
135.8
+ 14.2
190.6
+ 20.2
f 33.0*
463.4
+ 75.6t
656.1
? 101.3’
NOTE. Values given are the mean + SE. lf < .Ol, tP < .05, differences before and after weight reduction.
894
TANAKA
0
hg/ml*minl 2500
z* S % g % s
(ng/ml*minl
Obese pre-VLCO
1
2500
0 Obese post-VLCO
1
l
0
Obesepre-VLCD
2000-
1500-
= m s E
IOOO-
g
50001 20
25
I
I
30 35 Body Mass Index (kg/m?
40
o Obese pre-VLCD
. Obesepost-VLCO
Arginine(O.Sg/kg) (iiiiijiiiiijijiiiiiiiiiiiil
= :
:
:
:
:
:
Fig 4. Changes of integrated GH secretion of individuals in response to arginine before and after weight reduction due to very-low-calorie diet (VLCD) therapy.
Correlations Among GH Response to GH Secretion Stimulators, Degree of Weight Loss, and Metabolic Substrates A direct correlation was found between the percent increase of the integrated GH response to GRH after weight reduction and to L-dopa (r = ,660, P < .05) and between the percent increase to arginine and to r-dopa (r = ,652, P < .05). However, no correlation was observed between the degree of weight loss and the enhancement of plasma GH secretion in response to GRH, arginine, and L-dopa, respectively. There was no significant correlation between the integrated GH response to GRH, arginine, L-dopa and plasma levels of glucose, insulin, and FFA, respectively (data not shown). Ing/mH 2ol
.*** +. ..
:
45 Body Mass Index (kg/m21
subjects showed a markedly diminished GH response to L-dopa. After reduction, significantly increased responses were noted at 60.90, and 120 minutes post-ingestion (P < .05 to P < .Ol). Table 2 shows that, after weight reduction, integrated GH secretion in response to L-dopa was significantly raised in both first- and second-hour samples. Figure 6 shows the changes of integrated GH secretion of individuals in response to L-dopa before and after weight reduction. Before reduction, subjects showed a markedly impaired integrated GH response to L-dopa; however, after reduction, increased responses were noted in all samples.
**
O2X
45
Fig 2. Changes of integrated GH secretion of individuals in response to GRH before and after weight reduction due to very-low-calorie diet (VLCD) therapy.
z
ET AL
o
Obese pre-VLCO
0
Obese post-VLCO
** ..
..
J,
l.
l.
l.
10 ;-I/+
I 0
I 30
I 60 TIME
I 90
I 120 (minutes)
Fig 3. Mean (+ SE) plasma GH responses to arginine in obese subjects before and after weight reduction due to very-lowcalorie diet WLCD) therapy. l*P < .Ol, lP -c .05. differences before and after weight reduction.
0
.*. . . -. . . . . J? I
0
30
I
60
1
90
I
120 (minutes)
Fig 5. Mean (*SE) plasma GH responses to L-dopa in obese subjects before and after weight reduction due to very-lowcalorie diet (VLCD) therapy. l*P < .Ol. lP -c .05, differences before and after weight reduction.
GH SECRETION IN OBESITY
895
hg/ml*min) 0 Obese pm-VLCD 0 Obese post-VLCIJ
2000
1
i5
$0
i5
40
Body Mass Index (kg/m21 Changes of integrated GH secretion of individuals in Fig 6. response to L-dope before and after weight reduction due to very-low-calorie diet WLCD) therapy.
DISCUSSION Secretion of GH from the pituitary results from the net effect of CNS inputs by stimulatory and inhibitory factors (GRH and somatostatin C[SRIF]) and the responsiveness of the pituitary to these factors. GRH, characterized and synthetized by Guillemin et a123and Rivier et a1,24stimulates GH release at the pituitary level, while both arginine and L-dopa stimulate release at the suprapituitary level (hypothalamus). It has been assumed that GH secretion to orally administered L-dopa is mediated by release of endogenous GRH from the hypothalamus.“.” In the present study, GH secretion in response to GRH and L-dopa is markedly attenuated in obese subjects and the percent increase of the integrated GH response to GRH was correlated with that to L-dopa, suggesting impaired GH response to endogenous GRH is a likely explanation for the inadequate response to L-dopa. On the other hand, GH secretion to arginine is not mediated by release of endogenous GRH,” and there is a possibility that suppression of endogenous SRIF secretion may influence GH secretion in response to arginine. GH secretion in response to arginine was as markedly attenuated as that to GRH and L-dopa in obese subjects in the present study. This suggests that modulation of endogenous SRIF release may be involved in obese subjects.’ Reversibility of GH response to GRH, arginine, and L-dopa after weight reduction may be explained by the obesity-related increase in endogenous SRIF release into the portal capillary delivery system. Berelowitz et a125 reported that SRIF release from the hypothalamus is increased and that pituitary GH response to GRH is impaired in genetically obese Zucker rats
in vitro. Therefore, we speculate that the present results may be explained by increased release of SRIF into the portal capillary delivery system, resulting in impaired GH response to these three stimuli. Other possibilities include impaired pituitary response to GRH, arginine, and L-dopa due to (1) decreased GH stores, and (2) elevated levels of metabolic substrates associated with obesity, eg, glucose and FFA. The former explanation is based on the fact that GH stores decrease in rats with ventromedial hypothalamic lesioninduced obesity26-28 and monosodium glutamate treatment.29 However, in obese humans, it is not clear whether the GH pool within the pituitary or the hypothalamic GRH content decreases. Since impaired GH response to L-dopa or arginine in obese subjects can be augmented by propranolol,30 this possibility may be less plausible. The latter explanation depends on the fact that glucose and FFA inhibit GH response.‘3-2’ In the present study, plasma glucose did not decrease significantly after weight reduction, while plasma FFA levels decreased significantly. Therefore, it may be postulated that elevated FFA levels before weight reduction may account for the suppression of GH. However, we found no significant correlation between plasma GH response and plasma levels of glucose and FFA. Irie et a13’ reported that reduction of FFA levels in obese subjects by administration of nicotinic acid fails to increase basal GH levels. Thus, altered metabolic FFA found in obese subjects is not the major cause of impaired GH response to multiple stimuli. No correlation between the degree of weight loss and the enhancement of plasma GH secretion in response to GRH, arginine, and L-dopa was observed in spite of the reversibility of GH response to these three stimuli after weight reduction. Although the reason is so far unexplained, one possibility is that the pre-very-low-calorie diet group and post-very-lowcalorie diet group contain various stages of the degree of obesity. The other possibility would be due to the effect of the very-low-calorie diet. Kelijman and Frohman3* reported that acute nutrient deprivation enhanced GH responses to GRH in obesity independently of weight loss. In order to abolish the effect of the previous very-low-calorie diet, GH stimulation tests were performed following a l-week period on a 1,200-kcal diet after a very-low-calorie diet therapy in the present study. However, we can not deny the possibility that the effect of the very-low-calorie diet on plasma GH responsiveness was not eliminated. In summary, the present study has shown a reversible defect in growth hormone secretion in response to GRH, arginine, and L-dopa in obese subjects. Although the specific mechanisms responsible for this alteration await full elucidation, reversibility of the secretory defect with weight loss suggests an acquired defect secondary to a hormonalmetabolic alteration, rather than a primary defect involved in the pathogenesis of obesity.
REFERENCES
I. Glass AR, Burman KD, Dahms WT. et al: Endocrine function in human obesity. Metabolism 30:89-104, 1981 2. Kopelman PG, Noonan K, Goulton R, et al: Impaired growth
hormone response to growth hormone releasing factor and insulinhypoglycaemia in obesity. Clin Endocrinol23:87-94, 1985 3. Copinschi G, Wegienka LC, Hane S, et al: Effect of arginine on
TANAKA
896
serum levels of insulin and growth Metabolism 16:485-491, 1967
hormone
in obese
subjects.
4. Laurian L, Oberman Z, Ayalon D, et al: Under-responsiveness of growth hormone secretion after L-DOPA and deep sleep stimulation in obese subjects. Isr J Med Sci 11:482-487, 1975 5. Rosskamp R, Becker M, Soetadji S: Circulating somatomedin-C levels and the effect of growth hormone-releasing factor on plasma levels of growth hormone and somatostatin-like immunoreactivity in obese children. Eur J Pediatr 146:48-SO,1987 6. Josefsberg Z, Kauli R, Keret R, et al: Growth hormone response to insulin tolerance test and arginine stimulation in obese children and adolescents. Pediatr Adolesc Endocrinol 1: 146- 152, 1976 7. Williams T, Berelowitz M, Joffe SN, et al: Impaired growth hormone responses to growth hormone-releasing factor in obesity: A pituitary defect reversed with weight reduction. N Engl J Med 311:1403-1407, 1984 8. El-Khodary AZ, Ball MF, Stein B, et al: Effect of weight loss on the growth hormone response to arginine infusion in obesity. J Clin Endocrinol32:42-5 1, 197 1 9. Frohman LA, Jansson J-O: Growth mone. Endocrine Rev 7:223-253, 1986
hormone-releasing
hor-
10. Saito H, Saito S, Yamazaki R, et al: Clinical value of radioimmunoassay of plasma growth-hormone-releasing factor. Lancet 18:401-402, 1984 11. Chihara K, Kashio Y, Kita T, et al: L-dopa stimulates release of hypothalamic growth hormone-releasing hormone in humans. J Clin Endocrinol Metab 62:466-473, 1986 12. Chihara K, Kashio Y, Okimura Y, et al: Pathophysiology of growth hormone-releasing hormone (GHRH) secretion in humans. Endocrinol Japon 34:9-30, 1987 (suppl 1) 13. Roth J, Glick SM, Yalow RS, et al: Hypoglycemia: A potent stimulus to secretion of growth hormone. Science 140:987-988, 1963 14. Jansz A, Doorenbos H, Reitsma WD: Effect of food intake of growth-hormone level. Lancet 1:250-251, 1963 15. Press M, Tamborlane WV, Thorner MO, et al: Pituitary response to growth hormone-releasing factor in diabetes. Failure of glucose-mediated suppression. Diabetes 33:804-806, 1984 16. Sharp PS, Foley K, Chahal P, et al: The effect of plasma glucose on the growth hormone response to human pancreatic growth hormone releasing factor in normal subjects. Clin Endocrinol (Oxf) 20:497-501, 1984 17. Masuda A, Shibasaki T, Nakahara M, et al: The effect of glucose on growth hormone (GH)-releasing hormone-mediated GH secretion in man. J Clin Endocrinol Metab 60:523-526, 1985 18. Imaki
T, Shibasaki
T, Shizume
K, et al: The effect of free
ET AL
fatty acids on growth hormone (GH)-releasing hormone-mediated GH secretion in man. J Clin Endocrinol Metab 60:290-293, 1985 19. Blackard WG, Hull EW, Lopez-S A: Effect of lipids on growth hormone secretion in humans. J Clin Invest 50:1439-1443, 1971 20. Fineberg SE, Horland AA, Merimee TJ: Free fatty acid concentrations and growth hormone secretion in man. Metabolism 21:491-497, 1972 21. Fraser WM, Blackard WG: The effect of lipid on prolactin and growth hormone secretion. Horm Metab Res 9:389-394, 1977 22. Mizuno K, Toyosato M, Yabumoto S, et al: A new enzymatic method for calorimetric determination of free fatty acids. Ann Biochem 108:6-10, 1980 23. Guillemin R, Brazeau P, Bohlen P, et al: Growth hormonereleasing factor from a human pancreatic tumor that caused acromegaly. Science 218:585-587, 1982 24. Rivier J, Spiess J, Thorner M, et al: Characterization of a growth hormone-releasing factor from a human pancreatic islet tumour. Nature 300:276-278, 1982 25. Berelowitz M, Finkelstein JA, Thominet JL, et al: Impaired growth hormone secretion in spontaneous experimental obesity: A result of increased somatostatin (SRIF) tone. Presented at the Annual Meeting of the Endocrine Society, San Antonio, TX, June 8-10, 1983 26. Frohman LA, Bernardis LL: Growth hormone and insulin levels in weanling rats with ventromedial hypothalamic lesions. Endocrinology 82: 1125- 1132, 1968 27. Wakabayashi I, Inoue S, Satoh S, et al: Effect of hypothalamic ventromedial lesions on plasma growth hormone response to growth hormone-releasing factor in rats. Brain Res 346:70-74, 1985 28. Eikelboom R, Tannenbaum GS: Effects of obesity-inducing ventromedial hypothalamic lesions on pulsatile growth hormone and insulin secretion: Evidence for the existence of a growth hormonereleasing factor. Endocrinology 112:212-219, 1983 29. Millard WJ, Martin JB Jr, Audet J, et al: Evidence that reduced growth hormone secretion observed in monosodium glutamate-treated rats is the result of a deficiency in growth hormonereleasing factor. Endocrinology 110:540-550, 1982 30. Barbarino A, De Marinis L, Troncone L: Growth hormone response to propranolol and L-dopa in obese subjects. Metabolism 271275-278, 1978 31. Irie M, Tsushima T, Sakuma M: Effect of nicotinic acid administration on plasma HGH, FFA and glucose in obese subjects and in hypopituitary patients. Metabolism 19:972-979, 1970 32. Kelijman M, Frohman LA: Enhanced growth hormone (GH) responsiveness to GH-releasing hormone after dietary manipulation in obese and nonobese subjects. J Clin Endocrinol Metab 66:489494, 1988
