- Email: [email protected]
Increased insulin secretion in puberty: A compensatory response to reductions in insulin sensitivity
Increased insulin secretion in puberty: A compensatory response to reductions in insulin sensitivity Sonia C a p r i o , MD, G e r d Plewe, MD, M i c h a e l P. D i a m o n d , MD, D o n a l d C. Simonson, MD,* Susan D. Boulware, MD, Robert S. Sherwin, MD, a n d William V. T a m b o r l a n e , MD From the Departments of Pediatrics, Obstetrics and Gynecology, and Internal Medicine and the Clinical Research Centers, Yale University School of Medicine, New Haven, Connecticut
Recent studies have suggested that insulin action is reduced during puberty in normal children. To determine whether such resistance leads to excessive insulin secretion, we used the hyperglycemic clamp technique to produce a standard hyperglycemic stimulus (125 mg/dl a b o v e fasting levels for 120 minutes) in 9 preadolescent and 14 adolescent healthy children and in 14 normal adults. Fasting plasma insulin and C-peptide concentrations were higher in adolescents than in preadolescents and adults (p _<0.02). Despite identical glucose increments during the glucose clamp procedure, both firstand second-phase plasma insulin and C-peptide responses were also markedly greater in adolescents than in preadolescents or adults (p <0.01 vs. other groups). Despite sharply increased insulin responses in adolescents, the amount of exogenous glucose required to maintain hyperglycemia was similar in all three groups. Insulin responses in the children were directly correlated with fasting plasma levels of insulin-like growth factor I (r = 0.60 to 0.70, p <0.01). We conclude that glucose-stimulated insulin secretion is normally increased during puberty, a response that may compensate for pubertyinduced defects in insulin sensitivity. (J PEDIATR1989;114:963-7)
Although studies with the euglycemic insulin clamp technique suggest that insulin sensitivity is reduced in normal children during puberty, L,z the effect of this apparent insulin resistance on insulin secretion has not been established. Elevated plasma insulin levels after glucose inges-
Supported by grants No. RR00125 and No. AM20495 from the National Institutes of Health and by grants from the Juvenile Diabetes Foundation International and Squibb-Novo Laboratories, Princeton, N.J. Dr. Caprio is the recipient of a postdoctoral fellowship from the Juvenile Diabetes Foundation International, and Dr. Diamond is the recipient of a Clinical Associate Physician Award from the General Clinical Research Center. Submitted for publication Oct. 10, 1988; accepted Dec. 21, 1988. Reprint requests: Sonia Caprio, MD, Department of Pediatrics, Yale University School of Medicine, 333 Cedar St., New Haven, CT 06510. *Now at Joslin Diabetes Center, Boston.
tion have been reported in normal adolescents?. 3 However, a nu~pber of factors in addition to changes in insulin sensltl~lty can influence the insulin response during oral glucose tolerance testing. For example, individual variations in plasma glucose profiles, differences in the relative size or rate of absorption of the glucose load, variable stimulation of secretion of gut hormones, 4-6 loss of early DHEA-S IGF-I
I
Dehydroepiandrosterone sulfate Insulin-like growth factor I
[
[
(first-phase) insulin release, 7 and reduced hepatic clearance of insulin 8 may alter the insulin response to orally administered glucose, even in the presence of normal glucose tolerance. We used the glucose clamp technique 9 to assess the secretory response of pancreatic beta cells to changes in plasma glucose in normal preadolescents and adolescents
963
964
Caprio et al.
The Journal o f Pediatrics June 1989
~, ,,~/f,i AdO4..... ts
100
J ;;v'
80
PLASMA INSULIN pU/ml
2_50
J~-"
60
III Adu,s
40
200 20 PLASMA GLUCOSE (mg/dl)
150 100
0
A~J ~':~
A.----A adults o--.-tl adolescents e ~ o preadolescents
50 0
I
I
I
I
I
0
30
60
90
120
TIME (min)
i
t
i
and healthy adults. With this procedure a standardized rapid elevation of plasma glucose above fasting levels is achieved and maintained by intravenous administration of a variable-rate infusion of exogenous glucose.9 Because the glucose stimulus is uniform, a reliable comparison of the early and late insulin responses to glucose of different subject populations is possible. Moreover, alterations in hepatic clearance of insulin can be estimated by simultaneous plasma insulin and C-peptide measurements, and the amount of exogenous glucose required to maintain a constant hyperglycemic plateau can be used to assess insulin action. METHODS Subjects. The study population consisted of 23 healthy, nonobese Children and adolescents (age < 18 years) and 14 fully mature young adults (Table). All were normally active, had no recent or chronic illness, were not taking medications, and had no recent change in body weight or dietary habits. The children were grouped according to stage of pubertal development,~~as assessed by a physical examination on the day of admission to the Yale Children's Clinical Research Center, New Haven, Conn. Only two of the 14 adolescents had achieved full sexual maturity (Tanner stage V). Clinical data are summarized in the Table. As expected, circulating levels of dehydroepiandros-
t
i
5.0
/k"
PLASMA 2.0 C-PEPTIDE pmol/ml
_L.--i
1.0 0
i
i
-20 0 Fig. t. Plasma glucose levels during glucose clamp procedure in preadolescents, adolescents, and adults. Shaded area at top of figure illustrates variable-rate infusion of 20% dextrose employed during procedure to produce square-wave hyperglycemicplateau. Data concerning differences in glucose infusion rate required to maintain hyperglycemia in the three groups are given in results section of article. To convert plasma glucose values to milimoles per liter, divide by 18.
t
i
i
i
30 60 90 TIME (min)
i
120
Fig. 2. Biphasic plasma insulin and C-peptide responses during glucose clamp procedure. Plasma insulin and C-peptide levels were significantlygreater in adolescents than in preadolescents or adults at all time points (p --<0.02). To convert plasma insulin values to picomoles per liter, multiply by 7.2.
terone sulfate and insulin-like growth factor I were significantly lower in prepubertal than in pubertal children. All of the subjects and the parents of the children gave written consent before taking part in the study, and the protocol was approved by the human investigation committee of the Yale University School of Medicine. Hyperglycemic clamp proeedure. Each subject was studied on the morning after an overnight fast (10 to 12 hours). Two intravenous catheters were inserted: one in an antecubital vein for administration of test substances and the other in a vein of the dorsal part of the contralateral hand for blood sampling. The hand for blood sampling was kept in a heated box (60 ~ to 65 ~ C) to "arterialize" venous blood samplesY After a~rest period of 30 to 60 minutes, baseline fasting blood samples were obtained for the measurement of glucose, insulin, and C-peptide. Fasting IGF-I and DHEA-S levels in the children were also determined. The hyperglycemic clamp technique has been described in detail elsewhere. 9 As illustrated in Fig. 1, plasma glucose concentration was rapidly raised by 125 mg/dl (6.9 mmol/L) above fasting values by infusing, in a decreasing logarithmic manner, a priming dose of glucose to achieve the desired hyperglycemic plateau quickly. Subsequently, plasma glucose (measured at 5-minute intervals) was held constant at this hyperglycemic level for 120 minutes by appropriate adjustment of a variable-rate 20% glucose infusion. Blood samples were also obtained at
Volume 114 Number 6
Insulin secretion in puberty
965
Table. Clinical characteristics of subjects Preadolescents* (n = 9)
Age (yr) Range Sex (female/male) Weight (kg) Surface area (m2) Percent ideal body weight DHEA-S (~g/dl) IGF-I (U/ml)
9.6 __+0.4 8-12 2/7 30 + 2 1.06 + 0.04 97 _+ 2 40 _+ 8:~ 0.74 _+ 0.12:~
Adolescents1" (n = 14)
15.0 ___0.6 12-17 6/8 60 _+ 5 1.68 + 0.09 99 _+ 4 142 + 17 2.29 +_ 0.28
Adults (n = 14)
20.0 _+ 0.6 18-23 8/6 71 _ 3 1.86 + 0.06 103 _+ 5
Data from HamillPVV, Drizd TA, JohnsonCL, Reed RB, Roche AF, Moore WM: Am J ClinNutr 1979;32:607-29. *Tannerstage I. tTanner stages II to V. :~Significantlydifferentfrom valuesin adolescents,p <0.01.
2, 4, 6, 8, and 10 minutes, and every 10 minutes thereafter for 120 minutes, for measurements of plasma insulin and C-peptide. Urine was collected at the beginning and end of the procedure for determination of glucose content. Analyses. Plasma and urine glucose levels were measured by the glucose oxidase method with a Beckman glucose analyzer (Beckman Instruments Inc., Brea, Calif). Plasma insulin and C-peptide were measured by a doubleantibody radioimmunoassay?2,~3 Plasma DHEA-S and IGF-I were measured by Nichols Institute, San Juan Capistrano, Calif. Calculations. During the hyperglycemic clamp procedure the plasma insulin response is biphasic; responses in the first phase (0 to 10 minutes) and second phase (10 to 120 minutes) were calculated as the mean hormone concentration during the respective time periods. The rate of glucose metabolism during the hyperglycemic clamp procedure (expressed in milligrams per square meter of surface area per minute) was calculated at 20-minute intervals according to the equation M = I N F - UC - SC, where M is the glucose metabolism rate, I N F is the glucose infusion rate, UC is the correction for urinary loss of glucose, and SC is the correction for changes in the glucose space. 9 The average rate of glucose metabolism during the last 60 minutes of each clamp study was used to compare insulin action between study groups. All data are expressed as mean +_ SEM. Multiple-group comparisons were made by analysis of variance, and two-group comparisons were done with two-tailed Student unpaired t tests. Correlations between variables were examined by least-squares linear regression analysis. RESULTS As shown in Fig. l, basal fasting plasma glucose concentrations were similar and were raised to the same hyperglycemic plateau (approximately 125 mg/dl [6.9 mmol/L] above basal levels) in all three groups of subjects.
During hyperglycemia the coefficient of variation of sequential plasma glucose determinations was less than 5%. As shown in Fig. 2, fasting plasma insulin (10 +-- 1 #U/ml) and C-peptide (0.49 ___ 0.03 pmol/ml) levels were significantly higher in the adolescents than in the preadolescents (5 _+ 1/~U/ml and 0.28 _+ 0.02 pmol/ml, respectively, p <0.02) or the adults (7 ___ 1 # U / m l and 0.25 _+ 0.02 pmol/ml, respectively, p <0.01). The identical hyperglycemic stimulus produced a similar time course and biphasic pattern of insulin and C-peptide responses in the three groups of subjects. However, the magnitude of the response was much greater in adolescents than in preadolescents and adults. As shown in Fig. 3, both first- and second-phase insulin responses were significantly higher in adolescents than in healthy preadolescents and adults (first phase 40 + 4 vs. 19 _+ 3 and 26 _+ 3 #U/ml, respectively, p <0.02; second phase 68 _+ 8 vs. 30-+ 5 and 38 +-4 #U/ml, p <0.01). Among the adolescents, boys and girls had similar first-phase (34 + 5 vs. 46 _+ 7 ~tU/ml, respectively) and second-phase (65 _+ 11 vs. 71 + 11 #U/ml) insulin responses (p not significant). The first- and secondphase increases in C-peptide concentrations were significantly greater in adolescents than in the other two groups (p <0.01, Fig. 3). Again there was no difference in the C-peptide response of adolescent boys and girls (data not shown). Among all the children the magnitude of the firstand second-phase insulin response was directly correlated with fasting plasma IGF-I levels (first phase r = 0.68, p <0.01; second phase r = 0.73, p <0.01 ). A similar correlation between insulin responses and IGF-I was seen in the adolescent group alone (first phase r = 0.55, p <0.02; second phase r = 0.62, p <0.01). The first- and secondphase insulin responses were not correlated with DHEA-S levels among all the children or within the adolescent group alone (r --<0.4, p not significant). Despite higher insulin levels during the clamp procedure
966
Caprio et al.
The Journal of Pediatrics June 1989
r-~
80
pre-adolescents Adolescents Adults
60 PLASMA INSULIN pU/ml
:r
40 ////
20
z///
0
5.0
PLASMA C-PEPTIDE pmol/ml
2.0 Wcr
~
1.0 0
PHASE I
PHASE II
Fig. 3. Mean (+ SEM) first-phase (0 to 10 minutes) and second-phase (10 to 120 minutes) plasma insulin and C-peptide responses during glucose clamp procedure. Responses were significantly greater in adolescents (*p <0.02 and **p <0.01) than in other two groups. To convert plasma insulin values to picomoles per liter, multiply by 7.2. in adolescents, the rate of infusion of exogenous glucose that was required to maintain steady-state hyperglycemia was similar in all three groups (preadolescents 310 _ 66, adolescents 403 _ 46, and adults 407 _+ 56 mg/mg2/min; p not significant). DISCUSSION We have recently demonstrated that the rate of exogenous glucose infusion required to maintain euglycemia during intravenous insulin administration (euglycemic insulin clamp technique) is much lower in normal adolescents than in healthy preadolescents and adults2 Although these data strongly suggest that insulin action is impaired during puberty, this interpretation may have been confounded by age-dependent differences in basal glucose turnover (tracer measurements of glucose kinetics were not performed) or in the relative proportion of muscle tissue, the primary site of glucose disposal during euglycemic hyperinsulinemia. We undertook this study to explore that issue and, in particular, to assess whether healthy pubertal children manifest the biologic correlate of impaired insulin action, namely an enhanced insulin secretory response to
maintain normal glucose homeostasis. Hyperinsulinemia has been reported in pubertal children during oral glucose tolerance testing. 2,3 This procedure is limited, however, because glucose-stimulated insulin secretion is not measured directly and because insulin responses can be altered by a number of factors 4-6 other than changes in insulin sensitivity. Our observation that insulin responses during the glucose clamp were 2 to 3 times greater in adolescents than in preadolescents and adults is particularly noteworthy because the hyperglycemic stimulus was nearly identical in all groups and because the responses were not influenced by gastrointestinal absorption of glucose. Since both early and late insulin responses to hyperglycemia were enhanced in pubertal subjects, our data indicate that the hyperinsulinemia observed in those subjects could not be explained by deficient first-phase release, as has been reported in the early stages of non-insulindependent diabetes. 7 The resulting persistent hyperinsulinemia in that circumstance also leads to the development of insulin resistanceF4,~5 Elevations in plasma insulin concentration associated with puberty might have resulted from a reduction in hepatic degradation of insulin. However, first- and second-phase C-peptide responses were also increased in the pubertal subjects. The liver does not extract C-peptide, 8 the proinsulin fragment that is secreted by the beta cell in equimolar amounts with insulin, so these data indicate that augmented secretion rather than reduced degradation of insulin was responsible for the hyperinsulinemia observed in the adolescent group. The rate of glucose infusion required to maintain a stable hyperglycemic plateau was similar in all three groups even though plasma insulin levels were much higher in the adolescents. These data indicate that the exaggerated insulin responses in the pubertal subjects were able to compensate for the impairment in insulin action in this age group. The hormonal mechanism(s) underlying these puberty-induced metabolic changes remain open to speculation. However, the insulin and C-peptide responses were similar in adolescent boys and girls, and these responses did not correlate with DHEA-S values, making a direct effect of sex steroids less likely. Among our subjects the insulin responses to hyperglycemia were directly correlated with IGF-1 levels, a finding that is consistent with the observation of Bloch et al? that reduced insulin sensitivity in adolescents was associated with increased IGF-I levels. On the other hand, recent in vivo studies with biosynthetic IGF-1 indicate that this peptide has insulin-like metabolic effectsJ 6,~7 In view of the insulin antagonist effects of growth hormone) 8 the correlation of IGF-I levels with insulin hypersecretion may reflect the puberty-associated
Volume 114 Number 6
increase in circulating growth h o r m o n e concentrations. Thus it is intriguing to speculate t h a t increased secretion of growth h o r m o n e accounts for the insulin resistance, compensatory hyperinsulinemia, and elevations in I G F - I concentration t h a t occur during adolescence. We are indebted to the nurses of the children's and adult units of the General Clinical Research Center (GCRC) for the care of our subjects and to the staff of the core laboratory of the GCRC for performing the insulin and C-peptide assays.
Insulin secretion in puberty
9.
10. 11.
12.
REFERENCES 1. Amiel SA, Sherwin RS, Simonson DC, Lauritano AA, Tamborlane WV. Impaired insulin action in puberty: a contributing factor to poor glycemic control in adolescents with diabetes. N Engl J Med 1986;315:215-9. 2. Bloch CA, Clemmons P, Sperling MA. Puberty reduces insulin sensitivity. J PEDIATR 1987; 110:481-7. 3. Rosenbloom AL, Wheeler L, Bianchi R, Chin FT, Tiwary CM, Gorgic A. Age-adjusted analysis of insulin responses during normal and abnormal glucose tests in normal children and adolescents. Diabetes 1975;24:280-8. 4. Brown J. Gastric inhibitory polipeptide (GIP). Gastroenterology 1973;67:733-4. 5. Said S. Vasoactive intestinal peptide (GIP). Gastroenterology 1974;67:735-7. 6. Sasaki H, Faloona G, Unger R. Enteroglucagon. Gastroenterology 1974;67:746-8. 7. Brunzell JD et al. Relationships between fasting plasma glucose levels and insulin secretion during intravenous glucose tolerance tests. J Clin Endocrinol Metab 1976;42:222-9. 8. Bratusch-Marrain PR, Waldhausl WR, Gasic S, Hofer A.
I3.
14.
15.
16.
17.
18.
967
Hepatic disposal of biosynthetic human insulin and porcine C-peptide in humanS. Metabolism 1984;33:151-7. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214-33. Tanner JM. Growth and adolescence. 2nd ed. Oxford, Eng.: Blackwell, 1966. McGuire EAH, Helderman JH, Tobin JD, Andres R, Berman M. Effects of arterial versus venous sampling on analysis of glucose kinetics in man. J Appl Physiol 1976;41:564-73. Morgan CR, Lararow A. Immunoassay of insulin: plasma insulin levels of normal, subdiabetic and diabetic rats. Diabetes 1963;12:115-26. Kuzuyia H, Blix PN, Itorowitz DL, Steiner DF, Rubenstein AH. Determination of free and total insulin and C-peptide insulin-treated diabetics. Diabetes 1977;26:22-9. Sheehan P, Leonetti F, Rosenthal N. Effect of prolonged hyperinsulinemia on glucose metabolism. Diabetes 1986; 35 (suppl): 16A. Bar SR, Gorden P, Roth J, Kahn CR, DeMeyts P. Fluctuations in the affinity and concentration of insulin receptors on circulating monocytes of obese patients. J Clin Invest t916;58:1123-35. Jacob R, Plewe G, Fagin K, Tamborlane WV, Sherwin RS. Insulin-like gi'owth factor I lowers blood glucose without suppressive hepatic glucose production. Diabetes 1987; 36(suppl. 1):38A. Guler HP, Zapf J, Froesch ER. Short-term metabolic effects of recombinant human insulin-like growth factor I in healthy adults. N Engl J Med 1987;317:137-40. Bratush-Marrain PR, Smith D, DeFronzo RA. The effect of growth hormone on glucose metabolism and insulin secretion in man. J Clin Endocrinol Metab 1982;55:973-82.
Recent studies have suggested that insulin action is reduced during puberty in normal children. To determine whether such resistance leads to excessive insulin secretion, we used the hyperglycemic clamp technique to produce a standard hyperglycemic stimulus (125 mg/dl a b o v e fasting levels for 120 minutes) in 9 preadolescent and 14 adolescent healthy children and in 14 normal adults. Fasting plasma insulin and C-peptide concentrations were higher in adolescents than in preadolescents and adults (p _<0.02). Despite identical glucose increments during the glucose clamp procedure, both firstand second-phase plasma insulin and C-peptide responses were also markedly greater in adolescents than in preadolescents or adults (p <0.01 vs. other groups). Despite sharply increased insulin responses in adolescents, the amount of exogenous glucose required to maintain hyperglycemia was similar in all three groups. Insulin responses in the children were directly correlated with fasting plasma levels of insulin-like growth factor I (r = 0.60 to 0.70, p <0.01). We conclude that glucose-stimulated insulin secretion is normally increased during puberty, a response that may compensate for pubertyinduced defects in insulin sensitivity. (J PEDIATR1989;114:963-7)
Although studies with the euglycemic insulin clamp technique suggest that insulin sensitivity is reduced in normal children during puberty, L,z the effect of this apparent insulin resistance on insulin secretion has not been established. Elevated plasma insulin levels after glucose inges-
Supported by grants No. RR00125 and No. AM20495 from the National Institutes of Health and by grants from the Juvenile Diabetes Foundation International and Squibb-Novo Laboratories, Princeton, N.J. Dr. Caprio is the recipient of a postdoctoral fellowship from the Juvenile Diabetes Foundation International, and Dr. Diamond is the recipient of a Clinical Associate Physician Award from the General Clinical Research Center. Submitted for publication Oct. 10, 1988; accepted Dec. 21, 1988. Reprint requests: Sonia Caprio, MD, Department of Pediatrics, Yale University School of Medicine, 333 Cedar St., New Haven, CT 06510. *Now at Joslin Diabetes Center, Boston.
tion have been reported in normal adolescents?. 3 However, a nu~pber of factors in addition to changes in insulin sensltl~lty can influence the insulin response during oral glucose tolerance testing. For example, individual variations in plasma glucose profiles, differences in the relative size or rate of absorption of the glucose load, variable stimulation of secretion of gut hormones, 4-6 loss of early DHEA-S IGF-I
I
Dehydroepiandrosterone sulfate Insulin-like growth factor I
[
[
(first-phase) insulin release, 7 and reduced hepatic clearance of insulin 8 may alter the insulin response to orally administered glucose, even in the presence of normal glucose tolerance. We used the glucose clamp technique 9 to assess the secretory response of pancreatic beta cells to changes in plasma glucose in normal preadolescents and adolescents
963
964
Caprio et al.
The Journal o f Pediatrics June 1989
~, ,,~/f,i AdO4..... ts
100
J ;;v'
80
PLASMA INSULIN pU/ml
2_50
J~-"
60
III Adu,s
40
200 20 PLASMA GLUCOSE (mg/dl)
150 100
0
A~J ~':~
A.----A adults o--.-tl adolescents e ~ o preadolescents
50 0
I
I
I
I
I
0
30
60
90
120
TIME (min)
i
t
i
and healthy adults. With this procedure a standardized rapid elevation of plasma glucose above fasting levels is achieved and maintained by intravenous administration of a variable-rate infusion of exogenous glucose.9 Because the glucose stimulus is uniform, a reliable comparison of the early and late insulin responses to glucose of different subject populations is possible. Moreover, alterations in hepatic clearance of insulin can be estimated by simultaneous plasma insulin and C-peptide measurements, and the amount of exogenous glucose required to maintain a constant hyperglycemic plateau can be used to assess insulin action. METHODS Subjects. The study population consisted of 23 healthy, nonobese Children and adolescents (age < 18 years) and 14 fully mature young adults (Table). All were normally active, had no recent or chronic illness, were not taking medications, and had no recent change in body weight or dietary habits. The children were grouped according to stage of pubertal development,~~as assessed by a physical examination on the day of admission to the Yale Children's Clinical Research Center, New Haven, Conn. Only two of the 14 adolescents had achieved full sexual maturity (Tanner stage V). Clinical data are summarized in the Table. As expected, circulating levels of dehydroepiandros-
t
i
5.0
/k"
PLASMA 2.0 C-PEPTIDE pmol/ml
_L.--i
1.0 0
i
i
-20 0 Fig. t. Plasma glucose levels during glucose clamp procedure in preadolescents, adolescents, and adults. Shaded area at top of figure illustrates variable-rate infusion of 20% dextrose employed during procedure to produce square-wave hyperglycemicplateau. Data concerning differences in glucose infusion rate required to maintain hyperglycemia in the three groups are given in results section of article. To convert plasma glucose values to milimoles per liter, divide by 18.
t
i
i
i
30 60 90 TIME (min)
i
120
Fig. 2. Biphasic plasma insulin and C-peptide responses during glucose clamp procedure. Plasma insulin and C-peptide levels were significantlygreater in adolescents than in preadolescents or adults at all time points (p --<0.02). To convert plasma insulin values to picomoles per liter, multiply by 7.2.
terone sulfate and insulin-like growth factor I were significantly lower in prepubertal than in pubertal children. All of the subjects and the parents of the children gave written consent before taking part in the study, and the protocol was approved by the human investigation committee of the Yale University School of Medicine. Hyperglycemic clamp proeedure. Each subject was studied on the morning after an overnight fast (10 to 12 hours). Two intravenous catheters were inserted: one in an antecubital vein for administration of test substances and the other in a vein of the dorsal part of the contralateral hand for blood sampling. The hand for blood sampling was kept in a heated box (60 ~ to 65 ~ C) to "arterialize" venous blood samplesY After a~rest period of 30 to 60 minutes, baseline fasting blood samples were obtained for the measurement of glucose, insulin, and C-peptide. Fasting IGF-I and DHEA-S levels in the children were also determined. The hyperglycemic clamp technique has been described in detail elsewhere. 9 As illustrated in Fig. 1, plasma glucose concentration was rapidly raised by 125 mg/dl (6.9 mmol/L) above fasting values by infusing, in a decreasing logarithmic manner, a priming dose of glucose to achieve the desired hyperglycemic plateau quickly. Subsequently, plasma glucose (measured at 5-minute intervals) was held constant at this hyperglycemic level for 120 minutes by appropriate adjustment of a variable-rate 20% glucose infusion. Blood samples were also obtained at
Volume 114 Number 6
Insulin secretion in puberty
965
Table. Clinical characteristics of subjects Preadolescents* (n = 9)
Age (yr) Range Sex (female/male) Weight (kg) Surface area (m2) Percent ideal body weight DHEA-S (~g/dl) IGF-I (U/ml)
9.6 __+0.4 8-12 2/7 30 + 2 1.06 + 0.04 97 _+ 2 40 _+ 8:~ 0.74 _+ 0.12:~
Adolescents1" (n = 14)
15.0 ___0.6 12-17 6/8 60 _+ 5 1.68 + 0.09 99 _+ 4 142 + 17 2.29 +_ 0.28
Adults (n = 14)
20.0 _+ 0.6 18-23 8/6 71 _ 3 1.86 + 0.06 103 _+ 5
Data from HamillPVV, Drizd TA, JohnsonCL, Reed RB, Roche AF, Moore WM: Am J ClinNutr 1979;32:607-29. *Tannerstage I. tTanner stages II to V. :~Significantlydifferentfrom valuesin adolescents,p <0.01.
2, 4, 6, 8, and 10 minutes, and every 10 minutes thereafter for 120 minutes, for measurements of plasma insulin and C-peptide. Urine was collected at the beginning and end of the procedure for determination of glucose content. Analyses. Plasma and urine glucose levels were measured by the glucose oxidase method with a Beckman glucose analyzer (Beckman Instruments Inc., Brea, Calif). Plasma insulin and C-peptide were measured by a doubleantibody radioimmunoassay?2,~3 Plasma DHEA-S and IGF-I were measured by Nichols Institute, San Juan Capistrano, Calif. Calculations. During the hyperglycemic clamp procedure the plasma insulin response is biphasic; responses in the first phase (0 to 10 minutes) and second phase (10 to 120 minutes) were calculated as the mean hormone concentration during the respective time periods. The rate of glucose metabolism during the hyperglycemic clamp procedure (expressed in milligrams per square meter of surface area per minute) was calculated at 20-minute intervals according to the equation M = I N F - UC - SC, where M is the glucose metabolism rate, I N F is the glucose infusion rate, UC is the correction for urinary loss of glucose, and SC is the correction for changes in the glucose space. 9 The average rate of glucose metabolism during the last 60 minutes of each clamp study was used to compare insulin action between study groups. All data are expressed as mean +_ SEM. Multiple-group comparisons were made by analysis of variance, and two-group comparisons were done with two-tailed Student unpaired t tests. Correlations between variables were examined by least-squares linear regression analysis. RESULTS As shown in Fig. l, basal fasting plasma glucose concentrations were similar and were raised to the same hyperglycemic plateau (approximately 125 mg/dl [6.9 mmol/L] above basal levels) in all three groups of subjects.
During hyperglycemia the coefficient of variation of sequential plasma glucose determinations was less than 5%. As shown in Fig. 2, fasting plasma insulin (10 +-- 1 #U/ml) and C-peptide (0.49 ___ 0.03 pmol/ml) levels were significantly higher in the adolescents than in the preadolescents (5 _+ 1/~U/ml and 0.28 _+ 0.02 pmol/ml, respectively, p <0.02) or the adults (7 ___ 1 # U / m l and 0.25 _+ 0.02 pmol/ml, respectively, p <0.01). The identical hyperglycemic stimulus produced a similar time course and biphasic pattern of insulin and C-peptide responses in the three groups of subjects. However, the magnitude of the response was much greater in adolescents than in preadolescents and adults. As shown in Fig. 3, both first- and second-phase insulin responses were significantly higher in adolescents than in healthy preadolescents and adults (first phase 40 + 4 vs. 19 _+ 3 and 26 _+ 3 #U/ml, respectively, p <0.02; second phase 68 _+ 8 vs. 30-+ 5 and 38 +-4 #U/ml, p <0.01). Among the adolescents, boys and girls had similar first-phase (34 + 5 vs. 46 _+ 7 ~tU/ml, respectively) and second-phase (65 _+ 11 vs. 71 + 11 #U/ml) insulin responses (p not significant). The first- and secondphase increases in C-peptide concentrations were significantly greater in adolescents than in the other two groups (p <0.01, Fig. 3). Again there was no difference in the C-peptide response of adolescent boys and girls (data not shown). Among all the children the magnitude of the firstand second-phase insulin response was directly correlated with fasting plasma IGF-I levels (first phase r = 0.68, p <0.01; second phase r = 0.73, p <0.01 ). A similar correlation between insulin responses and IGF-I was seen in the adolescent group alone (first phase r = 0.55, p <0.02; second phase r = 0.62, p <0.01). The first- and secondphase insulin responses were not correlated with DHEA-S levels among all the children or within the adolescent group alone (r --<0.4, p not significant). Despite higher insulin levels during the clamp procedure
966
Caprio et al.
The Journal of Pediatrics June 1989
r-~
80
pre-adolescents Adolescents Adults
60 PLASMA INSULIN pU/ml
:r
40 ////
20
z///
0
5.0
PLASMA C-PEPTIDE pmol/ml
2.0 Wcr
~
1.0 0
PHASE I
PHASE II
Fig. 3. Mean (+ SEM) first-phase (0 to 10 minutes) and second-phase (10 to 120 minutes) plasma insulin and C-peptide responses during glucose clamp procedure. Responses were significantly greater in adolescents (*p <0.02 and **p <0.01) than in other two groups. To convert plasma insulin values to picomoles per liter, multiply by 7.2. in adolescents, the rate of infusion of exogenous glucose that was required to maintain steady-state hyperglycemia was similar in all three groups (preadolescents 310 _ 66, adolescents 403 _ 46, and adults 407 _+ 56 mg/mg2/min; p not significant). DISCUSSION We have recently demonstrated that the rate of exogenous glucose infusion required to maintain euglycemia during intravenous insulin administration (euglycemic insulin clamp technique) is much lower in normal adolescents than in healthy preadolescents and adults2 Although these data strongly suggest that insulin action is impaired during puberty, this interpretation may have been confounded by age-dependent differences in basal glucose turnover (tracer measurements of glucose kinetics were not performed) or in the relative proportion of muscle tissue, the primary site of glucose disposal during euglycemic hyperinsulinemia. We undertook this study to explore that issue and, in particular, to assess whether healthy pubertal children manifest the biologic correlate of impaired insulin action, namely an enhanced insulin secretory response to
maintain normal glucose homeostasis. Hyperinsulinemia has been reported in pubertal children during oral glucose tolerance testing. 2,3 This procedure is limited, however, because glucose-stimulated insulin secretion is not measured directly and because insulin responses can be altered by a number of factors 4-6 other than changes in insulin sensitivity. Our observation that insulin responses during the glucose clamp were 2 to 3 times greater in adolescents than in preadolescents and adults is particularly noteworthy because the hyperglycemic stimulus was nearly identical in all groups and because the responses were not influenced by gastrointestinal absorption of glucose. Since both early and late insulin responses to hyperglycemia were enhanced in pubertal subjects, our data indicate that the hyperinsulinemia observed in those subjects could not be explained by deficient first-phase release, as has been reported in the early stages of non-insulindependent diabetes. 7 The resulting persistent hyperinsulinemia in that circumstance also leads to the development of insulin resistanceF4,~5 Elevations in plasma insulin concentration associated with puberty might have resulted from a reduction in hepatic degradation of insulin. However, first- and second-phase C-peptide responses were also increased in the pubertal subjects. The liver does not extract C-peptide, 8 the proinsulin fragment that is secreted by the beta cell in equimolar amounts with insulin, so these data indicate that augmented secretion rather than reduced degradation of insulin was responsible for the hyperinsulinemia observed in the adolescent group. The rate of glucose infusion required to maintain a stable hyperglycemic plateau was similar in all three groups even though plasma insulin levels were much higher in the adolescents. These data indicate that the exaggerated insulin responses in the pubertal subjects were able to compensate for the impairment in insulin action in this age group. The hormonal mechanism(s) underlying these puberty-induced metabolic changes remain open to speculation. However, the insulin and C-peptide responses were similar in adolescent boys and girls, and these responses did not correlate with DHEA-S values, making a direct effect of sex steroids less likely. Among our subjects the insulin responses to hyperglycemia were directly correlated with IGF-1 levels, a finding that is consistent with the observation of Bloch et al? that reduced insulin sensitivity in adolescents was associated with increased IGF-I levels. On the other hand, recent in vivo studies with biosynthetic IGF-1 indicate that this peptide has insulin-like metabolic effectsJ 6,~7 In view of the insulin antagonist effects of growth hormone) 8 the correlation of IGF-I levels with insulin hypersecretion may reflect the puberty-associated
Volume 114 Number 6
increase in circulating growth h o r m o n e concentrations. Thus it is intriguing to speculate t h a t increased secretion of growth h o r m o n e accounts for the insulin resistance, compensatory hyperinsulinemia, and elevations in I G F - I concentration t h a t occur during adolescence. We are indebted to the nurses of the children's and adult units of the General Clinical Research Center (GCRC) for the care of our subjects and to the staff of the core laboratory of the GCRC for performing the insulin and C-peptide assays.
Insulin secretion in puberty
9.
10. 11.
12.
REFERENCES 1. Amiel SA, Sherwin RS, Simonson DC, Lauritano AA, Tamborlane WV. Impaired insulin action in puberty: a contributing factor to poor glycemic control in adolescents with diabetes. N Engl J Med 1986;315:215-9. 2. Bloch CA, Clemmons P, Sperling MA. Puberty reduces insulin sensitivity. J PEDIATR 1987; 110:481-7. 3. Rosenbloom AL, Wheeler L, Bianchi R, Chin FT, Tiwary CM, Gorgic A. Age-adjusted analysis of insulin responses during normal and abnormal glucose tests in normal children and adolescents. Diabetes 1975;24:280-8. 4. Brown J. Gastric inhibitory polipeptide (GIP). Gastroenterology 1973;67:733-4. 5. Said S. Vasoactive intestinal peptide (GIP). Gastroenterology 1974;67:735-7. 6. Sasaki H, Faloona G, Unger R. Enteroglucagon. Gastroenterology 1974;67:746-8. 7. Brunzell JD et al. Relationships between fasting plasma glucose levels and insulin secretion during intravenous glucose tolerance tests. J Clin Endocrinol Metab 1976;42:222-9. 8. Bratusch-Marrain PR, Waldhausl WR, Gasic S, Hofer A.
I3.
14.
15.
16.
17.
18.
967
Hepatic disposal of biosynthetic human insulin and porcine C-peptide in humanS. Metabolism 1984;33:151-7. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214-33. Tanner JM. Growth and adolescence. 2nd ed. Oxford, Eng.: Blackwell, 1966. McGuire EAH, Helderman JH, Tobin JD, Andres R, Berman M. Effects of arterial versus venous sampling on analysis of glucose kinetics in man. J Appl Physiol 1976;41:564-73. Morgan CR, Lararow A. Immunoassay of insulin: plasma insulin levels of normal, subdiabetic and diabetic rats. Diabetes 1963;12:115-26. Kuzuyia H, Blix PN, Itorowitz DL, Steiner DF, Rubenstein AH. Determination of free and total insulin and C-peptide insulin-treated diabetics. Diabetes 1977;26:22-9. Sheehan P, Leonetti F, Rosenthal N. Effect of prolonged hyperinsulinemia on glucose metabolism. Diabetes 1986; 35 (suppl): 16A. Bar SR, Gorden P, Roth J, Kahn CR, DeMeyts P. Fluctuations in the affinity and concentration of insulin receptors on circulating monocytes of obese patients. J Clin Invest t916;58:1123-35. Jacob R, Plewe G, Fagin K, Tamborlane WV, Sherwin RS. Insulin-like gi'owth factor I lowers blood glucose without suppressive hepatic glucose production. Diabetes 1987; 36(suppl. 1):38A. Guler HP, Zapf J, Froesch ER. Short-term metabolic effects of recombinant human insulin-like growth factor I in healthy adults. N Engl J Med 1987;317:137-40. Bratush-Marrain PR, Smith D, DeFronzo RA. The effect of growth hormone on glucose metabolism and insulin secretion in man. J Clin Endocrinol Metab 1982;55:973-82.