| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
COMMENT |
Endocrinology Service, Garrahan Pediatric Hospital, Buenos Aires, C1245AAM, Argentina
Address all correspondence and requests for reprints to: Alicia Belgorosky, Laboratorio de Investigación, Hospital de Pediatría Garrahan, Combate de los Pozos 1881, Buenos Aires, C1245AAM, Argentina. E-mail: abelgo{at}elsitio.net.
Abstract
The aim of this study was to analyze the possible implication of changes in the GH/IGF-I axis and in insulin sensitivity for the regulation of adrenal androgen secretion of normal prepubertal and adolescent girls. A total of 61 normal girls were evaluated in prepuberty [Group (Gr)1, n = 33; early (Gr1A, n = 16) and late (Gr1B, n = 17)]; puberty (Gr3, n = 28), early (Gr3A, n = 9) and late (Gr3B, n = 19); and during the transition between prepuberty and puberty (Gr2, n = 26). Insulin sensitivity was estimated by the fasting glucose/insulin ratio (G/I).
In Gr1, G/I was significantly higher, and the mean serum IGF-I and serum dehydroepiandrosterone sulfate (DHEAS) were significantly lower than in Gr3 (P < 0.0001). Mean G/I in Gr1A and Gr3A was significantly higher than in Gr1B (P < 0.01) and Gr3B (P < 0.02), respectively, and ratios in Gr1B were also significantly higher than in Gr3A (P < 0.02). However, body mass index (BMI) in Gr1A, Gr1B, and Gr3A was not significantly different, although a significant increment was observed between late prepuberty (Gr1B) and late puberty (Gr3B; P < 0.0001). On the other hand, serum IGF-I levels in Gr1A and Gr3A were significantly lower than those in Gr1B (P < 0.01) and Gr3B (P < 0.02), respectively. The mean serum DHEAS level in Gr1A and Gr3A was significantly lower than in Gr1B (P < 0.01) and Gr3B (P < 0.02), respectively, and the level in Gr1B was also significantly lower than in Gr3A (P < 0.02).
Correlation studies within Gr1, Gr2, and Gr3 were also performed. There was a significant positive correlation between serum DHEAS and age and a significant negative correlation between serum DHEAS and G/I in the three groups. However, a significant positive correlation between serum DHEAS and serum IGF-I was only found in Gr1. Furthermore, a significant negative correlation between BMI and the G/I was found in Gr2 and Gr3. Therefore, changes in insulin sensitivity might be involved in adrenal androgen synthesis both in prepuberty and in puberty, as well as during the transition from prepuberty to puberty. Changes in BMI suggest that adiposity might be a mediator of this effect, particularly during late puberty. On the other hand, the GH/IGF axis might be an important metabolic signal involved in the maturational changes of human adrenal androgens during prepuberty, at the time of adrenarche. Indeed, a significant negative correlation between G/I and serum IGF-I was found in Gr1, as well as in Gr2.
In conclusion, the findings of this study indicate that the GH/IGF-I axis and insulin resistance might be involved in the mechanism of adrenarche during prepuberty in normal girls. Because these relationships had not been seen in boys, we proposed that prepubertal ovarian estrogens might be responsible for the sex difference. The relationship between insulin resistance and adrenal androgens persists during the transition from prepuberty to puberty, as well as during puberty.
ADRENARCHE TAKES PLACE at about 6–8 yr of age in humans (1, 2, 3). Although several hypotheses have been suggested, the mechanism of this phenomenon remains unknown, suggesting that adrenarche might be a multifactorial event (4, 5, 6, 7, 8). It has also been proposed that IGF-I might be a trigger to the commencement of the adrenarche that may not persist after the onset (9).
Deregulation of the insulin/IGF-I system in the pathophysiology of premature adrenarche has been suggested, as well as progression to polycystic ovary (10, 11, 12, 13). In boys, pronounced adrenarche and precocious pubarche were not found to be associated with insulin resistance, suggesting a sexual dimorphism (14). Along these lines, we have recently assessed the relationship between the GH/IGF-I axis or insulin sensitivity and the regulation of adrenal androgen secretion in 56 normal prepubertal and adolescent boys (15). Our data support the hypothesis that neither the GH/IGF-I axis nor insulin sensitivity is involved in the mechanism of adrenarche. Insulin resistance and body mass index (BMI), however, increase at early puberty rather than at late puberty in boys, and this change could be involved in modulating adrenal androgen steroidogenesis during the transition between late prepuberty and early puberty. However, strong data are not available to support the sexual dimorphism in the regulation of adrenal androgen secretion in normal prepubertal and adolescent girls.
The purpose of this study was to assess the relationship between the GH/IGF-I axis, insulin sensitivity, and adrenal androgens in normal prepubertal and adolescent girls. We have evaluated the relationships among serum IGF-I, BMI, fasting serum glucose/insulin ratio (G/I), as an indirect marker of insulin sensitivity, and dehydroepiandrosterone sulfate (DHEAS), as a marker of adrenal androgen secretion, in 61 normal girls during early and late prepuberty, as well as during pubertal development.
Subjects and Methods
Serum samples were collected from patients attending a general pediatric hospital for ambulatory nonendocrine disorders such as strabismus, inguinal hernia, skeletal congenital defects, etc. All subjects with complaints that might affect metabolic or endocrine function were excluded. Serum samples were taken from blood drawn for routine surgical checkup. A trained pediatric endocrinologist was present during blood sampling to check out exclusion criteria, obtain oral consent from the parents, and perform the clinical examination. Inclusion criteria were as follows: height equal to or higher than -2 SD values; BMI between 15 and 85 percentiles of a population of Hispanic subjects older than 5 yr (16); weight for height between 80% and 110% of normal for the Argentinean population, in subjects less than 5 yr old (17). The study was approved by the Institutional Review Board of Garrahan Pediatric Hospital.
The study included 61 normal girls whose clinical characteristics are listed in Table 1
. Subjects were divided according to Tanner stage for breast (I–V). Group (Gr) 1 included 33 prepubertal subjects, Tanner I. To analyze early and late prepuberty, Gr1 was subdivided into two subgroups according to median chronological age (5.3 yr): subjects below the median, Gr1A (n = 16), and subjects at the median or above, Gr1B (n = 17). Gr3 included 28 subjects, Tanner stages II–V. They were subdivided into two subgroups according to Tanner stage: Tanner II (Gr3A), early puberty (n = 9); and Tanner III–V (Gr3B), late puberty (n = 19). Six girls of Gr3B were postmenarcheal (mean ± SD gynecological age, 3.05 ± 1.61 yr). They had regular menstrual cycles. Sampling was performed in the follicular phase of the cycle. To study hormonal changes during the transition between prepuberty and puberty, an additional group, Gr2 (n = 26), was defined by mixing the late prepuberty and early puberty subgroups.
|
Serum DHEAS was determined with the DPC Immulite Assay (Diagnostic Products Corp., Los Angeles, CA). Sensitivity was 0.05 µmol/liter, and interassay coefficient of variation ranged from 8.1–15%. Serum IGF-I was determined by RIA, after acid-ethanol extraction of serum, as already published (18). Insulin was determined by the Imx sys lens (Abbott Laboratories, Abbott Park, IL), assay sensitivity was 1.0 µU/ml, and interassay coefficient of variation ranged from 3.8–4.5%.
Serum glucose was determined by a Hitachi 911 autoanalyzer (F. Hoffman-La Roche Ltd., Basel, Switzerland). As already reported (19, 20), fasting G/I is a useful measure of insulin sensitivity. Furthermore, because a correlation between the quantitative insulin sensitivity check index (QUICKI; QUICKI = 1/log I + log G) and the euglycemic glucose clamp has been reported, QUICKI has been proposed as a simple and accurate method to assess insulin sensitivity (21). As we have published recently, to validate fasting G/I (milligrams/10-4 units) as an indirect marker of insulin sensitivity in our normal subjects, we studied the correlation of this marker with QUICKI. We found a correlation coefficient of 0.90 (P < 0.0001). Therefore, in this study insulin sensitivity was estimated by the fasting G/I.
Statistical analysis
All statistical analyses were performed using Statistix 7 (Analytical Software, Tallahassee, FL). Unpaired Student’s t test was used to assess mean value differences. Simple linear regression analysis between several pairs of variables was performed. Furthermore, Pearson’s correlation was calculated to compute a correlation matrix for several variables.
Results
The mean (±SD) G/I, serum IGF-I, and serum DHEAS are shown in Table 1
. In Gr1, G/I in Gr1A was significantly higher than in Gr1B (P < 0.01). G/I also decreased significantly between Gr1B and Gr3A (P < 0.02). In Gr3, G/I was significantly lower than in Gr1 (P < 0.0001), and a significant decrement between Gr3A and Gr3B was also found (P < 0.02).
The mean (±SD) serum IGF-I level in Gr1B was significantly higher than in Gr1A (P < 0.01). In Gr3, serum IGF-I was significantly higher than in Gr1 (P < 0.0001), and a significant increment between Gr3A and Gr3B was also found (P < 0.02).
The mean (±SD) serum DHEAS level in Gr1B was significantly higher than in Gr1A (P < 0.01). Serum DHEAS also increased significantly between Gr1B and Gr3A (P < 0.02). In Gr3, serum DHEAS was significantly higher than in Gr1 (P < 0.000), and a significant increment between Gr3A and Gr3B was also found (P < 0.02).
The mean (±SD) BMI was similar in Gr1A, Gr1B, and Gr3A but increased significantly in Gr3B (late puberty), as shown in Table 1.
Serum DHEAS levels increased significantly with age in Gr1 (r = 0.72; P < 0.0001), Gr2 (r = 0.54; P = 0.0071), and Gr3 (r = 0.632; P = 0.0004). A significant negative correlation between serum DHEAS levels and G/I in Gr1, Gr2, and Gr3 (r = -0.44, P = 0.015; r = -0.47, P = 0.027; and r = -0.39, P = 0.048, respectively) was found. As shown in Fig. 1, a significantly positive correlation between serum DHEAS and serum IGF-I levels was only found in Gr1 (r = 0.53; P = 0.0043). These data suggest that insulin might be involved in the regulation of adrenal androgen steroidogenesis during late prepuberty and also during puberty, but that the GH/IGF-I axis is only involved during late prepuberty when adrenarche takes place.
|
A significant positive correlation between BMI and chronological age in Gr2 and Gr3 was observed (r = 0.44, P = 0.023; and r = 0.52, P = 0.004, respectively). No correlation was found between BMI and serum IGF-I levels. A significant negative correlation between BMI and G/I was observed in Gr2 and Gr3 (r = -0.49, P = 0.011; and r = -0.56, P = 0.0025, respectively).
We have also found a significant negative correlation between G/I and chronological age in Gr1, Gr2, and Gr3 (r = -0.66, P < 0.0001; r = -0.53, P < 0.0061; and r = -0.49, P = 0.008, respectively), as well as between G/I and serum IGF-I in Gr1 and Gr2 (r = -0.42, P = 0.03; and r = -0.51, P = 0.021, respectively).
To sort out the effect of age on simple regression analysis of the different parameters studied, Pearson correlation matrix was performed. It confirmed that there was a significant negative correlation between serum DHEAS levels and G/I (r = -0.44; P < 0.05) and a positive correlation between serum DHEAS and serum IGF-I levels in Gr1 (r = 0.55; P < 0.005). These results suggest that serum IGF-I might be a strong predictive independent variable of serum DHEAS levels during prepuberty.
Pearson correlation analysis also confirmed that, during transition from late prepuberty to early puberty, as well as during puberty, there was a negative correlation between serum DHEAS levels and G/I (r = -0.55, P < 0.05; and r = -0.40, P < 0.005, respectively), but not between DHEAS and serum IGF-I levels (r = 0.31, P = not significant; and r = 0.21, P = not significant, respectively), suggesting that insulin but not the GH/IGF-I axis is involved in the regulation of adrenal androgen steroidogenesis during puberty. A high association between serum DHEAS levels and BMI in Gr2 (r = 0.70; P < 0.005), but not in Gr1 and Gr3 was also confirmed.
Because serum IGF-I levels decline after menarche, all values of Gr3 and Gr3B were recalculated without the values of the six postmenarcheal girls. Significant differences did not change at all between group mean comparisons and correlations.
Discussion
To follow the course of developmental changes, we have divided both prepubertal and pubertal girls into two study groups according to age or developmental stage, respectively. This approach has the advantage of detecting changes within prepuberty or puberty, as well as comparing the transition from prepuberty to puberty.
Pronounced adrenarche with precocious pubarche in girls has been associated with hyperinsulinism and insulin resistance (10, 11, 12). However, a sexual dimorphism in this association has been proposed (14). In the present study, we have found a significant decrease of insulin sensitivity in normal prepubertal girls. Furthermore, we have also found a significant decrease of insulin sensitivity during the transition between late prepuberty and early puberty, as well within puberty. This pattern was clearly different from that of normal prepubertal and pubertal boys, as we previously reported (15). In that study, no change in insulin sensitivity was observed within prepuberty or puberty, but a decrease of insulin sensitivity took place during the transition from prepuberty to puberty. Sex differences in insulin sensitivity during normal prepubertal and pubertal development have also been reported in a study that included a small number of prepubertal subjects (22). It has been postulated that BMI is the major predictor of insulin sensitivity in early puberty, however, other factors have also been proposed, and the reasons for the sex differences remain unclear (23). Indeed, in the present study, although insulin sensitivity decreased in late prepuberty and during the transition between late prepuberty and early puberty, BMI clearly increased in late puberty, suggesting that other factors were involved. Sex steroids can regulate insulin sensitivity. It has been shown that testosterone does not significantly affect insulin sensitivity (24), whereas progesterone and estradiol increase insulin resistance (25, 26). Serum estradiol levels in normal girls, as measured by ultrasensitive recombinant cell assays (27, 28), are significantly higher than in normal boys, not only during puberty but also during the prepubertal years. In a previous study, we have reported that serum bioavailable estradiol increases progressively from infancy to late prepuberty in girls (29). Although we have not determined serum estradiol in this study, we believe that it is possible that the sex differences in the pattern of the G/I that we have found might be secondary to differences in the estrogen milieu. Moreover, it has been proposed that androgens and estrogens are able to modulate, in vivo, the adipose conversion of rat preadipocytes (30). On the whole, androgens elicit antiadipogenic effects, contrary to estrogens, which behave as proadipogenic factors.
Similar to previous reports (29, 31), in this study we have confirmed that serum DHEAS levels increased during prepuberty, during the transition from late prepuberty to early puberty, and during puberty. We have also found a significant negative correlation between serum DHEAS levels and G/I during prepubertal and pubertal development, and a positive correlation between serum DHEAS levels and serum IGF-I, but only during the prepubertal period. These data suggested that in normal girls, insulin is involved in the regulation of adrenal androgen steroidogenesis during the life span and might be one of the peripheral regulators involved in the mechanism of adrenarche in girls. Our study also suggests that the GH/IGF axis might be an important metabolic signal involved in the maturational changes of human adrenal at the time of adrenarche. Surprisingly, during the transition from late prepuberty to early puberty, as well as during puberty, no direct relationship between serum IGF-I and adrenal steroidogenesis seems to be present. It can be speculated that serum IGF-I is necessary for the development of the reticularis zone of the adrenal, although no specific data are available. In boys, we had not found any significant relationship between serum DHEAS and G/I or serum IGF-I during prepuberty (15). These results further confirm a sexual dimorphism in the regulation of adrenal androgen steroidogenesis at adrenarche, suggesting that although there are intraadrenal maturational changes involved in the regulation of the adrenarche, peripheral metabolic signals might be different in the two sexes. In this regard, it has been proposed recently that there is a relationship between the IGF axis and adrenal androgens in prepubertal girls with premature adrenarche (32).
Similar to what we had found in normal boys (15), a significant negative correlation between G/I and serum IGF-I was observed in normal girls within prepuberty and during the transition from late prepuberty to early puberty, but not within puberty. This pattern of IGF-I and insulin levels may be related to changes in GH release (33). On the other hand, the G/I decrease of late puberty might be related to the increment of BMI in late pubertal girls.
We propose that the GH/IGF-I axis would be involved in the prepubertal decrement of insulin sensitivity. Ovarian estrogens, in turn, might also influence the maturation of the GH/IGF-I axis (29). The increase in serum IGF-I and insulin could modulate intraadrenal changes in steroidogenesis, leading to adrenarche.
Finally, changes in insulin sensitivity might also have an impact in the GnRH/gonadotropic/gonadal axis. As reported by Brüning et al. (34), insulin regulates LH synthesis and/or secretion in central nervous system insulin receptor knockout mice. Therefore, the decrement of insulin sensitivity during prepuberty in normal girls, but not in normal boys, might mature the GnRH/gonadotropic/ovarian axis, resulting in an earlier development of puberty in girls.
In summary, the findings of this study indicate that the GH/IGF-I axis and insulin resistance might be involved in the mechanism of adrenarche during prepuberty in girls. Because these relationships had not been seen in boys, we proposed that prepubertal ovarian estrogens might be responsible for the sex difference. The relationship between insulin resistance and adrenal androgens persists during the transition from prepuberty to puberty, as well as during puberty.
Acknowledgments
The contributions of D. Chirico and L. Del Rio are acknowledged. We thank the National Pituitary Agency for the generous supply of IGF-I RIA reagents.
Footnotes
This work was supported by grants from Consejo Nacional de Investigaciones Científicas y Técnicas, Fondo para la Investigación Científica y Tecnológica, Ministerio de Salud (Beca Carrillo-Oñativia) of Argentina, and Pharmacia Endocrine Care International Fund for Research and Education.
Abbreviations: BMI, Body mass index; DHEAS, dehydroepiandrosterone sulfate; G/I, glucose/insulin ratio; Gr, group; QUICKI, quantitative insulin sensitivity check index.
Received June 25, 2002.
Accepted December 9, 2002.
References
, lyase) from guinea pig adrenal microsomes: dual function of a single enzyme and effect of cytochrome b5. Biochim Biophys Acta 833:151–160[Medline]
-hydroxylase and 17, 20 lyase activities of P450c17. Contributions of serine 106 and P450 reductase. Endocrinology 132:2498–2506[Abstract]
2. Endocrinology 141:649–656
This article has been cited by other articles:
 |
|
 |
|
  Z Hochberg Juvenility in the context of life history theory Arch. Dis. Child., June 1, 2008; 93(6): 534 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Baer, G. A. Colditz, W. C. Willett, and J. F. Dorgan Adiposity and Sex Hormones in Girls Cancer Epidemiol. Biomarkers Prev., September 1, 2007; 16(9): 1880 - 1888. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Rosenfield Identifying Children at Risk for Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. S. Hannon, G. Rao, and S. A. Arslanian Childhood Obesity and Type 2 Diabetes Mellitus Pediatrics, August 1, 2005; 116(2): 473 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cappel, D. Mauger, and D. Thiboutot Correlation Between Serum Levels of Insulin-like Growth Factor 1, Dehydroepiandrosterone Sulfate, and Dihydrotestosterone and Acne Lesion Counts in Adult Women Arch Dermatol, March 1, 2005; 141(3): 333 - 338. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
