Pituitary Response to Growth Hormone-Releasing Hormone in IDDM Abnormal Responses to Insulin and Hyperglycemia MARTIN PRESS, SONIA CAPRIO, WILLIAM V. TAMBORLANE, RHAJAT BHUSHAN, MICHAEL THORNER, WYLIE VALE, JEAN RIVER, AND ROBERT S. SHERWIN

In poorly controlled insulin-dependent diabetes mellitus (IDDM), hyperglycemia fails to inhibit the pituitary response to growth hormone-releasing factor (GRF). To evaluate whether this derangement is reversed by a simultaneous elevation of circulating insulin, 0.3 jjtg/kg i.v. GRF 1-40 was administered to nine poorly controlled IDDM subjects (HbAn >11.1%) with and without concomitant infusion of insulin. In the absence of insulin, the poorly controlled IDDM subjects demonstrated a growth hormone response to GRF similar to that of nondiabetic subjects, despite marked hyperglycemia (-16.8 mM). When insulin was infused into these same patients (insulin clamp) to produce combined hyperinsulinemia (528 ± 90 pM) and hyperglycemia (16.5 ± 1.98 mM), the GRF-induced growth hormone rise was markedly exaggerated (65 ± 11 vs. 20 ± 4 |xg/L without insulin infusion, P < 0.001). This enhancement of GRF-stimulated growth hormone release by insulin was strikingly attenuated (22 ± 7 ixg/L) in five well-controlled diabetic subjects studied under conditions of similar hyperinsulinemia (486 ± 84 pM) and hyperglycemia (16.41 ± 0.95 mM)- In contrast, in nondiabetic subjects, acute hyperinsulinemia reduced the growth hormone response to GRF. We conclude that the failure of hyperglycemia to block the pituitary response to GRF in poorly controlled diabetes is not attributable to the lack of a coincident increase in circulating insulin. The paradoxical stimulatory effect of insulin on GRF-induced growth hormone release may contribute to the high spontaneous growth hormone levels characteristically seen in poorly controlled insulin-treated patients, and its attenuation after intensive insulin therapy may

From the Department of Medicine and Pediatrics and General Clinical Research Center, Yale University School of Medicine, New Haven, Connecticut; the Department of Internal Medicine, University of Virginia School of Medicine, Charlottesville, Virginia; and the Peptide Biology Laboratory, Salk Institute, San Diego, California. Address correspondence and reprint requests to Dr. Robert S. Sherwin, Yale University School of Medicine, PO Box 3333, New Haven, CT 06510. Received for publication 14 December 1987 and accepted in revised form 27 September 1991.

DIABETES, VOL. 41, JANUARY 1992

contribute to the reversal of growth hormone hypersecretion in well-controlled diabetic patients. Diabetes 41:17-21,1992

T

he raised circulating growth hormone levels characteristic of poorly controlled insulin-dependent diabetes mellitus (IDDM) actively worsen diabetic control and may play a part in the pathogenesis of diabetic retinopathy (1,2). The mechanism or mechanisms underlying the diabetesinduced alterations in growth hormone secretion observed clinically have not been clarified, however. Previous studies have suggested that growth hormone responses to growth hormone-releasing factor (GRF) are quantitatively similar to nondiabetic subjects (3,4) or increased in some, but not all (5,6) hyperglycemic, conventionally treated patients with long-standing IDDM. On the other hand, when modest hyperglycemia is produced in nondiabetic subjects by intravenous (3,7) or oral (8) glucose administration, there is marked suppression of the pituitary response to GRF. These observations imply that even the "normal" response to GRF seen in chronically hyperglycemic patients with IDDM is inappropriate. This study was designed to investigate whether the failure of hyperglycemia to suppress the pituitary response to GRF in poorly controlled IDDM patients is due to the lack of coincident hyperinsulinemia. For this purpose, the insulin-clamp technique (9) was used to raise insulin levels independently of changes in the plasma glucose concentration, both in patients with IDDM and nondiabetic control subjects. In addition, a second set of studies was performed in diabetic patients treated with intensive insulin regimens to distinguish abnormalities in growth hormone secretory dynamics due to poor metabolic control from those that result from diabetes per se.

17

GROWTH HORMONE SECRETION IN DIABETES

RESEARCH DESIGN AND METHODS

We studied 14 (5 men, 9 women) randomly selected, nonobese patients with IDDM. They were 18-42 yr old and had had diabetes for 9-30 yr. None had a detectable C-peptide response to 1 mg i.v. glucagon. Apart from their diabetes, all patients were in good health, and none was taking medications other than insulin. Nine patients were poorly controlled on conventional insulin therapy with a mixture of regular and NPH in divided doses (HbA., 11.1-20.6% [normal 4-8%]), whereas, the other five had received intensified insulin treatment with the insulin pump (n = 2) or with at least three insulin injections (n = 3) for at least 6 mo before study (HbA-, 6.8-8.2%). Two conventionally treated patients had received laser therapy for proliferative retinopathy. None of the patients had impaired renal function. Five healthy male subjects, 26-36 yr old, were also studied. All subjects gave written consent to the study, which was approved by the Human Investigations Committee of the Yale University School of Medicine. All studies were performed after an overnight fast. An indwelling catheter was inserted in an antecubital vein for infusion of test substances. A second catheter in a dorsal vein of the contralateral hand, which was maintained in a box heated to 65°C, was used for blood sampling and for injection of GRF. GRF was synthesized as previously described (10), and no side effects of GRF were observed. The poorly controlled diabetic patients were studied at their elevated fasting glucose concentrations and received an intravenous insulin infusion or a saline control infusion in random order. No subcutaneous insulin was given on the morning of study. On the insulin infusion study day, after the collection of baseline blood samples, semisynthetic human insulin (Squibb-Novo, Princeton, NJ) was given as a primed continuous infusion at a rate of 1 mU • kg" 1 • min~1 for 4 h. Plasma glucose was maintained at baseline levels with a variable glucose infusion (insulin-clamp technique) as previously described (9). The well-controlled diabetic subjects were studied on only one occasion. In these experiments, plasma glucose was raised to a level similar to that of the poorly controlled group and clamped there while a primed continuous infusion of insulin (1 mU • kg" 1 • min~1) was administered for 4 h. In each study, bolus of 0.3 |xg/kg i.v. human GRF 1-40 was given 2 h after the beginning of the study. Serum growth hormone levels were measured at 30-min intervals for the 2-h period before GRF administration and at 10- to 15-min intervals thereafter. The effect of acute hyperinsulinemia on the pituitary response to GRF was also examined in the nondiabetic control subjects and compared with a saline control day. Insulin was again infused at 1 mU • kg" 1 • min~1 while plasma glucose was clamped at the fasting level. Plasma glucose was measured on a Beckman glucose analyzer (Beckman, Fullerton, CA). Plasma growth hormone and insulin were measured by a double-antibody radioimmunoassay (11,12). Free insulin was determined after immediate precipitation of antibody-bound insulin with polyethylene glycol (13). Glycosylated hemoglobin was determined with a minicolumn procedure (IsoLab,

18

GRF

GH (ng/ml)

-I2O

O

60

TIME (min) FIG. 1. Plasma growth hormone (GH) levels before and after administration of growth hormone-releasing factor (GRF) in poorly controlled diabetic subjects receiving insulin (•) or saline (O) infusion.

Akron, OH). Data are means ± SE. The peak growth hormone response after GRF administration was used as a measure of response to GRF. Statistical analyses were performed with two-way analysis of variance with a repeated measure designed to compare the responses of different groups over time. RESULTS

In the conventionally treated diabetic patients, fasting plasma glucose was 17.08 ± 2.63 mM on the control day compared to 16.52 ± 1.96 mM on the insulin study day. Free insulin levels averaged 528 ± 90 pM during the insulin clamp compared to 72 ± 36 pM during the saline control study (P< 0.01). On the insulin-clamp day, growth hormone levels rose transiently in five of nine patients (Fig. 1) after the commencement of the insulin infusion but returned to baseline before GRF was administered. This did not occur during the saline control study. After GRF, a markedly greater growth hormone response was observed in the poorly controlled diabetic subjects in the presence of hyperinsulinemia compared to the saline control study (Fig. 1). Peak growth hormone levels reached 65 ± 11 vs. 20 ± 4 |xg/L during the control day (P< 0.001). In contrast, in nondiabetic subjects, acute hyperinsulinemia (396 ± 36 vs. 48.0 ± 1.8 pM on control day) did not potentiate the pituitary response to GRF, and variable suppression (which failed to reach statistical significance) was observed (Fig. 2). During the insulinclamp study, the peak growth hormone rise in nondiabetic subjects was only 15% (P < 0.001) of that seen in the poorly controlled hyperinsulinemic patients (Figs. 2 and 3). In the well-controlled diabetic subjects, plasma glu-

DIABETES, VOL. 41, JANUARY 1992

M. PRESS AND ASSOCIATES

100 r

I c

50

X

T 0

L

T

SALINE CONTROL

INSULIN CLAMP

FIG. 2. Peak growth hormone (GH) response to growth hormone-releasing factor during saline control and insulin-clamp studies in poorly controlled insulin-dependent diabetic patients (hatched bars) and nondiabetic subjects (open bars).

cose (16.41 ± 1.00 mM) and insulin (486 ± 84 pM) were raised to levels not significantly different from values in the poorly controlled diabetic group (see above). When GRF was given, the rise in plasma growth hormone in the well-controlled diabetic patients was significantly less than in their poorly controlled counterparts (Fig. 3). The peak growth hormone response (22 ± 7 pg/ml) was only one-third of that seen in the poorly controlled patients (P < 0.01) and approached the response seen in the nondiabetic group (Fig. 3). DISCUSSION

Although raised growth hormone levels have long been recognized as a feature of poorly controlled IDDM (14-

GRF

60

E

40

c

^

20 OL

-I20

0

60

120

TIME(min) FIG. 3. Plasma growth hormone (GH) levels before and after administration of growth hormone-releasing factor (GRF) (after 120 min) in poorly controlled (A) and (O) well-controlled insulin-dependent diabetic patients and in nondiabetic (shaded area) subjects during insulin-clamp studies (insulin infusion 1 mU kg" 1 min~1).

DIABETES, VOL. 41, JANUARY 1992

16), the underlying pathophysiology is still not well understood. In addition to increased spontaneous growth hormone secretion, patients with diabetes may show an exaggerated response to provocative stimuli (17-19). Growth hormone levels may also fail to suppress normally with glucose, or even rise paradoxically (20,21), and may rise during insulin administration (22,23). Elevated basal levels of growth hormone in poorly controlled IDDM patients commonly are not associated with an exaggerated pituitary response to GRF, implying a hypothalamic mechanism (3,4). Because levels of insulinlike growth factor I (IGF-I), which is believed to be involved in negative feedback regulation of growth hormone secretion, tend to be reduced in diabetes (22-25), this may represent a physiological adaptation whereby the body attempts, by means of raised growth hormone levels, to maintain normal IGF-I levels. However, additional factors appear to contribute to abnormal growth hormone secretion in patients with diabetes. Because the response to GRF is potently suppressed by modest hyperglycemia in nondiabetic subjects (3,7), a quantitatively normal response to GRF in patients with poorly controlled diabetes is inappropriate for the prevailing high glucose levels. It is not clear whether the suppressive effect of hyperglycemia in nondiabetic subjects is mediated directly on the pituitary or via stimulation of hypothalamic somatostatin secretion. Nor is it clear whether this effect is exerted directly by glucose itself or whether it is mediated by the secondary rise in insulin levels. Insulin inhibits the responsiveness of pituitary cells to GRF in vitro, whereas glucose has no such effect (26). In our studies, the fact that acute hyperinsulinemia tended to reduce the pituitary response to GRF in nondiabetic subjects suggests that the suppressive effect of hyperglycemia may be mediated, at least partially, by the secondary rise in circulating insulin levels. This could also explain the inconsistency of glucose in suppressing the pituitary response to GRF in patients with IDDM, even after a period of improved metabolic control. If this hypothesis is correct, the suppressive effect of hyperglycemia should occur only when accompanied by hyperinsulinemia. However, far from causing suppression, hyperinsulinemia actually caused a paradoxical increase in the pituitary response to GRF in conventionally treated hyperglycemic diabetic patients. This effect of insulin on pituitary sensitivity to GRF could explain the rise in growth hormone levels sometimes seen in diabetic patients in ketoacidosis after the initiation of insulin therapy (22,27). More importantly, sensitization of the pituitary by insulin may contribute to the increased spontaneous growth hormone secretion seen in IDDM subjects throughout the day. The elevated systemic insulin levels commonly induced by insulin injections could augment the response to GRF released from the hypothalamus. The markedly exaggerated GRF response induced by combined hyperglycemia and hyperinsulinemia in conventionally treated diabetic subjects was not apparent in patients maintained on intensive insulin therapy for an extended period. Although the growth hormone response to GRF in the well-controlled patients was not signifi-

19

GROWTH HORMONE SECRETION IN DIABETES

cantly higher than that seen in the nondiabetic control subjects receiving insulin, the striking differences in glucose levels between the two studies (only the diabetic subjects were hyperglycemic) may have obscured a small residual abnormality in the well-controlled diabetic group. Our data suggest that the dramatic response to GRF seen in conventionally treated patients in the presence of hyperinsulinemia in some way reflects an effect of poor metabolic control rather than being a feature of diabetes per se. The potentiation of the response to GRF by insulin in poorly controlled diabetes could result from the loss of an inhibitory effect of insulin on the pituitary, or it might reflect aberrant suppression of the counterbalancing effect of hypothalamic somatostatin (or both). Insulin inhibits somatostatin secretion from the pancreas and gut (28,29). However, it is not known whether insulin exerts a similar effect on hypothalamic somatostatin secretion. Moreover, if the potentiation of the response to GRF by insulin in patients with diabetes is due to suppression of somatostatin, it is not clear why a similar effect is not observed in nondiabetic subjects. Note that, in this regard, studies with anticholinergic agents have implied a decrease in hypothalamic somatostatin tone in IDDM patients compared with nondiabetic subjects (6). Alternatively, it has been suggested by some (23) but not all (5) studies that diabetic microangiopathy may predispose to aberrant growth hormone responses, perhaps via changes in capillary permeability (23). Microangiopathy in our patients with long-standing diabetes may therefore have contributed to their altered response to hyperinsulinemia. On the other hand, Sharp et al. (30) reported an increase in basal growth hormone concentrations (albeit of a lesser magnitude than in our studies) in response to insulin in newly diagnosed non-insulindependent diabetic patients without detectable microangiopathy. Finally, it is possible that IGF-I levels were low in the poorly controlled diabetic group and that this contributed to the aberrant GRF response during hyperinsulinemia. Regardless of the underlying mechanisms, the synergistic interaction between insulin and GRF observed in poorly controlled patients with IDDM should contribute significantly to the diurnal growth hormone elevations that adversely affect metabolic control in diabetes (1,31). Conversely, the attenuation of this deranged GRF response after intensive insulin therapy breaks this vicious cycle, contributing to a lowering of growth hormone levels and to the concomitant improvement in glycemic control.

ACKNOWLEDGMENTS

This work was supported in part by a grant from SquibbNovo Incorporated and National Institutes of Health Grants DK-20495 (R.S.); RR-125 (General Clinical Research Center); DK-23632 and HD-13197 (M.O.T.); and AM-20917, AA-03504, and DK-32632 (Peptide Biology Laboratory). M.P. was supported by a Clinical Associate Physician Award from the Division of Research Resources, National Institutes of Health (RR-125).

20

We are grateful to the nurses and staff of the General Clinical Research Center, particularly Mary Walesky, RN, Ralph Jacob, Aida Grossman, and Andrea Belous for technical assistance. REFERENCES 1. Press M, Tamborlane WV, Sherwin RS: Importance of raised growth hormone levels in mediating the metabolic derangements of diabetes. N EnglJMed 210:810-15, 1984 2. Lundbaek K, Christensen JJ, Jensen VA, Johansen K, Olsen TS, Hansen AP, Oskov H, Osterby R: Diabetes, diabetic angiopathy, and growth hormone. Lancet 2:131-33, 1970 3. Press M, Tamborlane WV, Thorner MO, Vale W, Rivier J, Gertner JM, Sherwin RS: Pituitary response to growth hormone-releasing factor in diabetes: failure of glucose-mediated suppression. Diabetes 33:804-806, 1984 4. Sharp PS, Foley K, Kohner EH: Evidence for a central abnormality in the regulation of growth hormone secretion in insulin-dependent diabetes. Diabetic Med 1:205-208, 1984 5. Pietschmann P, Schernthaner G, Prskavec F, Gisinger C, Freyler H: No evidence for increased growth hormone responses to growth hormone-releasing hormone in patients with diabetic retinopathy. Diabetes 36:159-62, 1987 6. Giustina A, Bossoni S, Cimino A, Pizzocolo G, Romanelli G, Wehrenberg WB: impaired growth hormone response to pyrisodstignine in type I diabetic patients with exaggerated GH-releasing hormonestimulated GH secretion. J Clin Endocrinol Metab 71:1486-90, 1990 7. Sharp PS, Foley K, Chahal P, Kohner GM: The effect of plasma glucose on the growth hormone response to growth hormone releasing factor in normal subjects. Clin Endocrinol 20:497-502, 1984 8. Davies RR, Turner S, Johnston DG: Oral glucose inhibits growth hormone secretion induced by human pancreatic growth hormone releasing factor 1-44 in normal man. Clin Endocrinol 21:477-81, 1984 9. DeFronzo RA, Tobin J, Andres R: Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237:E2U-23, 1979 10. Thorner MO, Rivier J, Spiess J, Borges JL, Vance ML, Bloom SR, Rogol AD, Cronin MJ, Kaiser DL, Evans WS, Webster JD, Valle W: Human pancreatic growth hormone releasing factor selectively stimulates growth hormone-secretion in man. Lancet 1:24-28,1983 11. Schalch DS, Parker ML: A sensitive double antibody immunoassay for human growth hormone in plasma. Nature (Lond) 203:1141-42, 1964 12. Morgan CR, Lazarow AL: Immunoassay of insulin: two antibody system: plasma insulin levels of normal, subdiabetic and diabetic rats. Diabetes 12:115-26, 1963 13. Kuzuya H, Blix PM, Horwitz DL, Steiner DF, Rubenstein AH: Determination of free and total insulin and C-peptide in insulin-treated diabetics. Diabetes 26:22-29, 1977 14. Hansen AP, Johansen K: Diurnal patterns of blood glucose, serum free fatty acids, insulin, glucagon and growth hormone in normals and juvenile diabetics. Diabetologia 6:27-33, 1970 15. Molnar GD, Taylor WF, Langworthy A, Fatourechi V: Diurnal growth hormone and glucose abnormalities in unstable diabetics. J Clin Endocrinol Metab 34:837-46, 1972 16. Osplin CM, Fariq ACS, Carlsen EC, Vaccaro V, Barr RE, Iranmanesh A, Less MM, Veldhius JD, Evans WS: Alterations in the pulsatile mode of growth hormone release in men and women with insulin dependent diabetes mellitus. J Clin Endocrinol Metab 69:239-45, 1989 17. Tamborlane WV, Sherwin RS, Koivisto V, Hendler R, Genel M, Felig P: Normalization of the growth hormone and catecholamine response to exercise in juvenile-onset diabetic subjects treated with a portable insulin infusion pump. Diabetes 28:785-88, 1979 18. Burday SZ, Fine PH, Schalch DS: Growth hormone secretion in response to arginine infusion in normal and diabetic subjects: relationship to blood glucose levels. J Lab Clin Med 71:897-911, 1968 19. Cremer GM, Molnar GD, Taylor WF, Rosevear JW, Ackerman E: Growth hormone release in unstable diabetes: tests with saline, arginine, glucagon, and epinephrine. Acta Diabetol Lat 10:121621, 1973 20. Yde H: Abnormal growth hormone response to ingestion of glucose in juvenile diabetics. Acta Med Scand 186:449-504, 1969 21. Hansen AP: The effect of intravenous glucose infusion on the exercise-induced serum growth hormone rise in normals and juvenile diabetics. Scand J Clin Lab Invest 28:195-205, 1971 22. Sonksen P, Srivastara NC, Thompkins CV, Nabarro JDN: Growth

DIABETES, VOL. 41, JANUARY 1992

PRESS AND ASSOCIATES

23.

24.

25. 26. 27.

hormone and cortisol responses to insulin infusion in patients with diabetes mellitus. Lancet 2:155-59, 1972 Sharp PS, Foley K, Vitelli F, Maneschi F, Kohner EM: Growth hormone response to hyperinsulinemia in insulin-dependent diabetics: comparison of patients with and without retinopathy. Diabetic Med 1:55-58, 1984 Tamborlane WV, Hintz RL, Bergman M, Genel M, Felig P, Sherwin RS: Insulin infusion pump treatment of diabetes: influence of improved metabolic control on plasma somatomedin levels. N Engl J Med 305:303-307, 1981 Amiel SA, Sherwin RS, Hintz RL, Gertner JM, Press CM, Tamborlane WV: Effect of diabetes and its control on insulin-like growth factors in the young subject with type I diabetes. Diabetes 33:1175-79,1984 Melmed S: Insulin suppresses growth hormone secretion by rat pituitary cells. J Clin Invest 73:1425-33, 1984 Cryer PE, Daughaday WH: Diabetic ketosis: serial plasma hormone

DIABETES, VOL. 41, JANUARY 1992

concentrations during therapy. Diabetes 19:519-23, 1970 28. Patel YC, Wheatley T, Malaisse-Lagae F, Orci L: Elevated portal and peripheral blood concentration of immunoreactive somatostatin in spontaneously diabetic (BBL) Wistar rats: suppression with insulin. Diabetes 29:757-61, 1980 29. Rouiller D, Schusdziarra V, Unger RH: Insulin inhibits somatostatinlike immunoreactivity release stimulated by intragastric HCL. Diabetes 30:735-38, 1981 30. Sharp PS, Mohan V, Maneschi F, Vitelli F, Cloke HR, Burrin JM, Kohner EM: Changes in plasma growth hormone in diabetic and nondiabetic subjects during the glucose clamp. Metabolism 36:7175, 1987 31. Campbell PJ, Bolli GB, Cryer PE, Gerich JE: Pathogenesis of the dawn phenomenon in patients with insulin-dependent diabetes mellitus. N EnglJ Med 312:1473-89, 1985

21

Pituitary response to growth hormone-releasing hormone in IDDM. Abnormal responses to insulin and hyperglycemia.

In poorly controlled insulin-dependent diabetes mellitus (IDDM), hyperglycemia fails to inhibit the pituitary response to growth hormone-releasing fac...
618KB Sizes 0 Downloads 0 Views