Abstract
Objective
Ultra-sensitive hormone assays have detected slight sex differences in blood estradiol (E2) levels in young children before adrenarche. However, the origin of circulating E2 in these individuals remains unknown. This study aimed to clarify how E2 is produced in young girls before adrenarche.
Design
This is a satellite project of the Japan Environment and Children’s Study organized by the National Institute for Environmental Studies.
Methods
We collected blood samples from healthy 6-year-old Japanese children (79 boys and 71 girls). Hormone measurements and data analysis were performed in the National Institute for Environmental Studies and the Medical Support Center of the Japan Environment and Children’s Study, respectively.
Results
E2 and follicle stimulating hormone (FSH) levels were significantly higher in girls than in boys, while dehydroepiandrosterone sulfate (DHEA-S) and testosterone levels were comparable between the two groups. Girls showed significantly higher E2/testosterone ratios than boys. In children of both sexes, a correlation was observed between E2 and testosterone levels and between testosterone and DHEA-S levels. Moreover, E2 levels were correlated with FSH levels only in girls.
Conclusions
The results indicate that in 6-year-old girls, circulating E2 is produced primarily in the ovary from adrenal steroids through FSH-induced aromatase upregulation. This study provides evidence that female-dominant E2 production starts several months or years before adrenarche. The biological significance of E2 biosynthesis in these young children needs to be clarified in future studies.
Introduction
The first physical signs of puberty in boys and girls are testicular enlargement and breast budding, respectively (1). Typically, these signs appear at 11–12 years of age in boys and at 9–10 years of age in girls (1). Several years before developing physical signs, children undergo conspicuous changes in the blood levels of sex hormones. Usually, an increase in circulating adrenal steroids (‘adrenarche’) occurs at 7 or 8 years of age in children of both sexes (1, 2, 3). Subsequently, an increase in circulating gonadal steroids (‘gonadarche’) takes place in a sex-specific manner. In girls, blood levels of estradiol (E2) start to increase around 8 years of age, while in boys, testosterone levels typically increase at 9 years of age or later (1, 4, 5, 6). Thus, sexual dimorphism in blood sex hormone levels is evident in children above 8 years of age. Furthermore, ultra-sensitive hormone assays using liquid chromatography-tandem mass spectrometry (LC-MS/MS) have revealed slight sex differences in circulating E2 levels in children below 7 years of age (6). These results indicate that sex-specific hormone production begins before adrenarche. However, given the limited number of previous reports, these data require further validation. Moreover, this notion raises a fundamental question of how E2 is produced in young girls before adrenarche. To clarify the origin of circulating E2 in young girls, we measured serum levels of sex hormones and gonadotropins in 150 healthy children at 6 years of age.
Materials and methods
Subjects
This project was conducted in the framework of a pilot study of the Japan Environment and Children’s Study (JECS). JECS is a large-scale birth cohort study of the Japan Ministry of the Environment, which is organized by the National Institute for Environmental Studies. The participants of the present study were recruited during a JECS pilot study targeting 6-year-old children. The present study was approved by the ethics committee of the National Institute for Environmental Studies and that of four JECS pilot study-centers. This study was carried out in accordance with the Declaration of Helsinki. Written informed consent was obtained from the participants’ parents or guardians.
The participants of this study consisted of 79 boys and 71 girls aged at 5.75–6.58 years (Supplementary Table 1, see section on supplementary materials given at the end of this article). These children were recruited from four JECS pilot study-centers. All children were apparently healthy and exhibited no physical signs of puberty. The age, height, and body weight were comparable between the boy and girl groups (Table 1). Blood samples were obtained either in the morning (42 boys and 36 girls) or afternoon (30 boys and 30 girls), while information about the sampling time was unavailable for seven boys and five girls.
Clinical data of the participants. Data are presented as the median (range).
| Boys | Girls | P-value | |
|---|---|---|---|
| Number | 79 | 71 | |
| Age (years) | 6.00 (5.83–6.58) | 6.00 (5.75–6.50) | 0.63 |
| Height (cm) | 114.2 (101.8–124.7) | 113.5 (101.6–127.9) | 0.73 |
| Body weight (kg) | 20.4 (14.4–35.1) | 19.0 (13.9–27.7) | 0.12 |
Hormone measurements
Serum levels of E2, testosterone, and dehydroepiandrosterone sulfate (DHEA-S) were measured using LC-MS/MS (LSI Medience, Tokyo, Japan). The methods were described previously (7). The lower detection limits of E2, testosterone, and DHEA-S were 0.092 pmol/L (0.025 pg/mL), 0.017 nmol/L (5 pg/mL), and 0.14 μmol/L (50 pg/mL), respectively. Serum levels of luteinizing hormone (LH) and follicle stimulating hormone (FSH) were measured using standard chemiluminescent enzyme immune assays (LSI Medience). The lower detection limits of LH and FSH were both 0.1 IU/L. We calculated blood E2/testosterone ratios, which reflect the enzymatic activity of aromatase (8).
Statistical analysis
Sex differences in the hormone values and E2/testosterone ratios were analyzed by the Wilcoxon rank sum test. Differences in hormone levels between samples obtained in the morning and afternoon were analyzed using the Wilcoxon signed rank test. Correlations between two hormone values were analyzed by the Spearman rank correlation test and the relationship between E2 and other sex hormones were examined by linear regression analysis. All statistical analyses were performed using R (version 4.0.0, R Core Team, Vienna, Austria). P-values of less than 0.05 were considered significant.
Results
Blood hormone levels
The results of the 150 children are shown in Fig. 1 and Supplementary Table 1. In most children, serum levels of LH were undetectable, while levels of testosterone, E2, DHEA-S, and FSH were low but above the detection limit. Thus, LH was excluded from statistical analyses.
Blood hormone levels and estradiol/testosterone (E2/T) ratios in boys and girls. The upper and lower borders of each box indicate the 75th and 25th percentiles, respectively. Thick horizontal lines represent the median; upper whisker ends, 75th percentile plus 1.5 interquartile range (IQR); and lower whisker ends, 25th percentile minus 1.5 IQR. The values that exceed indicated ranges are plotted as individual circles.
Citation: Endocrine Connections 10, 10; 10.1530/EC-21-0134
Sex differences in blood hormone levels and E2/testosterone ratios
Blood levels of E2 and FSH were significantly higher in the girl group than in the boy group, whereas levels of DHEA-S and testosterone were comparable between the two groups (Fig. 1). Consequently, E2/testosterone ratios were significantly higher in the girl group than in the boy group.
Blood samples obtained in the morning and afternoon contained almost equal levels of hormones, except for testosterone and E2 (Supplementary Table 2). Testosterone levels in the boy and girl groups and E2 levels in the boy group were relatively high in the samples obtained in the morning. However, the differences between samples obtained in the morning and afternoon were small.
In children of both sexes, serum E2 levels were positively correlated with testosterone levels, and testosterone levels were correlated with DHEA-S levels (Table 2). Furthermore, E2 levels were positively correlated with FSH levels in girls, but not in boys. Likewise, linear regression analysis revealed that E2 levels was associated with testosterone and FSH levels only in girls (Table 3).
Correlation between two hormone values.
| P-value | |||||
|---|---|---|---|---|---|
| E2 | DHEA-S | Testosterone | FSH | ||
| Boys | |||||
| r | E2 | 3.12 × 10−15 | 1.59 × 10−11 | 0.81 | |
| DHEA-S | 0.75 | 1.04 × 10−13 | 0.70 | ||
| Testosterone | 0.67 | 0.72 | 0.14 | ||
| FSH | 0.03 | 0.04 | 0.17 | ||
| Girls | |||||
| r | E2 | 0.19 | 6.12 × 10−4 | 3.09 × 10−3 | |
| DHEA-S | 0.16 | 5.57 × 10−10 | 0.40 | ||
| Testosterone | 0.40 | 0.66 | 0.88 | ||
| FSH | 0.35 | –0.10 | 0.02 | ||
E2, estradiol; DHEA-S, dehydroepiandrosterone sulfate; FSH, follicle stimulating hormone.
Multiple linear regression model with estradiol as the outcome variable.
| Explanatory variable | Coefficient | Standard error | T-value | P-value | |
|---|---|---|---|---|---|
| Boys | (intercept) | 0.09 | 0.05 | 1.74 | 0.09 |
| DHEA-S | 7.01 × 10−4 | 2.68 × 10−4 | 2.62 | 0.01 | |
| Testosterone | 6.51 × 10−5 | 2.39 × 10−3 | 0.03 | 0.98 | |
| FSH | –0.02 | 0.03 | –0.60 | 0.55 | |
| Model summary; adjusted R-squared = 0.11, F-statistic P-value = 7.07 × 10−3 | |||||
| Girls | (intercept) | 0.06 | 0.30 | 0.21 | 0.84 |
| DHEA-S | –1.28 × 10−3 | 1.84 × 10−3 | –0.70 | 0.49 | |
| Testosterone | 0.038 | 0.01 | 3.75 | 3.68 × 10−4 | |
| FSH | 0.27 | 0.12 | 2.33 | 0.02 | |
| Model summary; adjusted R-squared = 0.26, F-statistic P-value = 4.01 × 10−5 | |||||
DHEA-S, dehydroepiandrosterone sulfate; FSH, follicle-stimulating hormone.
Discussion
We measured serum levels of sex hormones and gonadotropins in 150 healthy children at 6 years of age. In most children, levels of E2, testosterone, DHEA-S, and FSH were above the detection limit, while LH levels remained extremely low. Blood levels of DHEA-S and testosterone in these children were lower than the levels previously documented for children during adrenarche (9, 10), indicating that our participants had not yet entered adrenarche. These results are consistent with the report by Tung et al. that the average age of adrenarche in the East Asian population is 7.7 years (3). Nevertheless, we found a significant difference in serum E2 levels between boys and girls. This sexual dimorphism in circulating E2 levels cannot be ascribed to the female-dominant E2 production in the adrenal gland, because serum levels of DHEA-S, a marker for adrenal steroidogenesis (2, 11), were comparable between the boy and girl groups. Instead, a significant sex difference was observed in the E2/testosterone ratios. These results indicate that in healthy children, female-dominant aromatase activation precedes adrenarche. This notion provides an answer to the previously raised question of why serum testosterone levels of a girl with congenital aromatase deficiency remained within the normal range until 5 years of age and became abnormally high thereafter (12). Notably, we found a slight difference in testosterone levels between samples obtained in the morning and afternoon. These results are consistent with the previous findings that there is a diurnal rhythm of testosterone secretion in both prepubertal and pubertal children (10, 13).
In children of both sexes, serum levels of E2 were correlated with testosterone levels, which inturn were correlated with DHEA-S levels. These results imply that circulating E2 in these children was synthesized primarily from adrenal steroids through the intermediacy of testosterone. We speculate that the conversion of testosterone to E2 in girls occurs mostly in the ovary, because (i) CYP19A1, the gene encoding aromatase, is strongly expressed in the ovary (8, 14), (ii) blood E2 levels in our female participants were higher than those in previously reported girls and young adults with pure ovarian dysgenesis (2.41 ± 1.67 pmol/L) (15), and (iii) Wilson et al. observed an increase in blood E2 levels in healthy prepubertal girls at 5–12 years of age but not in age-matched girls with ovarian dysfunction due to Turner syndrome (16). However, it is possible that other CYP19A1-expressing tissues, such as the skin, adipose tissues, and bone (8) also contribute to the aromatization of circulating testosterone in young girls.
FSH levels were slightly higher in girls than in boys. Female-dominant FSH secretion in prepubertal children has been documented in previous reports (4, 17) and is likely to reflect sex-specific hypothalamic function that governs gonadotropin secretion (1). Notably, serum FSH levels were positively correlated with E2 levels only in girls. These results indicate that in girls, FSH upregulates CYP19A1 in the ovary to facilitate E2 production. Indeed, FSH is known to stimulate CYP19A1 expression in the ovary (18). Moreover, Francois et al. have shown that in normal infant mice, FSH induces E2 production in the ovary without causing follicular growth (19). Interestingly, the authors proposed that E2 production in the immature murine ovary is indispensable for the programming of future reproductive function. Thus, the relatively small increase in serum E2 levels observed in our female participants may also be essential for ovarian function in later life. In contrast, LH is unlikely to contribute to the female-dominant E2 production in young children, because serum levels of LH remained undetectable in most of our participants.
This study has several limitations. First, this is a cross-sectional study. Thus, we could not determine the onset age of female-dominant E2 production. Since previous studies have shown that blood FSH levels are constantly higher in girls than in boys from infancy through adolescence (4), there may be a sex difference in aromatase activity even in children under 6 years of age. Second, we did not measure levels of sex hormone binding globulin. Blood levels of total fractions of sex hormones may not be parallel with the levels of unbound (free and bioavailable) hormones (20). Third, our conclusions are based solely on the results of blood hormone measurement. Venous sampling is useful to determine the tissue origin of circulating sex hormones in children, although this test is generally not feasible in humans. Lastly, we did not address the functional importance of E2 production in young girls. Further studies are necessary to clarify whether sex hormone production in childhood is associated with sexual maturation during puberty or reproductive function in adulthood.
In conclusion, the results show that sexual dimorphism in sex hormone production is already apparent in 6-year-old children. Excessive circulating E2 in these girls is most likely synthesized in the ovary from adrenal steroids through FSH-induced aromatase upregulation. The biological significance of E2 production in such young children needs to be clarified in future studies.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EC-21-0134.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was funded by the National Institute for Environmental Studies. The findings and conclusions of this article are solely the responsibility of the authors and do not represent the official views of the aforementioned institute. This study was also supported by a grant from the Japan Society for the Promotion of Science (17H06428).
References
- 1↑
Styne DM Physiology and disorders of puberty. In Williams Textbook of Endocrinology, 14th ed., pp. 1023–1194. Eds Melmed S, Auchus RJ, Goldfine AB, Koenig RJ, Rosen CJ. Pennsylvania: Elsevier, 2020.
- 2↑
Auchus RJ, Rainey WE. Adrenarche: physiology, biochemistry and human disease. Clinical Endocrinology 2004 60 288–296. (https://doi.org/10.1046/j.1365-2265.2003.01858.x)
- 3↑
Tung YC, Lee JS, Tsai WY, Hsiao PH. Physiological changes of adrenal androgens in childhood. Journal of the Formosan Medical Association 2004 103 921–924.
- 4↑
Zec I, Kučak I, Begčević I, Simundić AM, Tišlarić-Medenjak D, Megla ZB, Vrkić N. Reference intervals for reproductive hormones in prepubertal children on the automated Roche cobase 411 analyzer. Clinical Biochemistry 2012 45 1206–1212. (https://doi.org/10.1016/j.clinbiochem.2012.05.019)
- 5↑
Kushnir MM, Blamires T, Rockwood AL, Roberts WL, Yue B, Erdogan E, Bunker AM, Meikle AW. Liquid chromatography–tandem mass spectrometry assay for androstenedione, dehydroepiandrosterone, and testosterone with pediatric and adult reference intervals. Clinical Chemistry 2010 56 1138–1147. (https://doi.org/10.1373/clinchem.2010.143222)
- 6↑
Frederiksen H, Johannsen TH, Andersen SE, Albrethsen J, Landersoe SK, Petersen JH, Andersen AN, Vestergaard ET, Schorring ME & Linneberg A et al.Sex-specific estrogen levels and reference intervals from infancy to late adulthood determined by LC-MS/MS. Journal of Clinical Endocrinology and Metabolism 2020 105 754–768. (https://doi.org/10.1210/clinem/dgz196)
- 7↑
Saito K, Matsuzaki T, Iwasa T, Miyado M, Saito H, Hasegawa T, Homma K, Inoue E, Miyashiro Y & Kubota T et al.Steroidogenic pathways involved in androgen biosynthesis in eumenorrheic women and patients with polycystic ovary syndrome. Journal of Steroid Biochemistry and Molecular Biology 2016 158 31–37. (https://doi.org/10.1016/j.jsbmb.2016.02.010)
- 8↑
Miller WL, Auchus RJ. The molecular biology, biochemistry, and physiology of human steroidogenesis and its disorders. Endocrine Reviews 2011 32 81–151. (https://doi.org/10.1210/er.2010-0013)
- 9↑
Ankarberg-Lindgren C, Dahlgren J, Andersson MX. High-sensitivity quantification of serum androstenedione, testosterone, dihydrotestosterone, estrone and estradiol by gas chromatography–tandem mass spectrometry with sex- and puberty-specific reference intervals. Journal of Steroid Biochemistry and Molecular Biology 2018 183 116–124. (https://doi.org/10.1016/j.jsbmb.2018.06.005)
- 10↑
Ankarberg C, Norjavaara E. Diurnal rhythm of testosterone secretion before and throughout puberty in healthy girls: correlation with 17beta-estradiol and dehydroepiandrosterone sulfate. Journal of Clinical Endocrinology and Metabolism 1999 84 975–984. (https://doi.org/10.1210/jcem.84.3.5524)
- 11↑
Witchel SF, Pinto B, Burghard AC, Oberfield SE. Update on adrenarche. Current Opinion in Pediatrics 2020 32 574–581. (https://doi.org/10.1097/MOP.0000000000000928)
- 12↑
Constanzo M, Garcia-Feyling J, Saraco N, Marino R, Pérez Garrido N, Touzon MS, Viterbo G, Lazzati JM, Patiño HC & Mattone C et al.Accelerated pubertal tempo in a 46,XY aromatase-deficient patient. Hormone Research in Paediatrics 2018 90 275–282. (https://doi.org/10.1159/000492128)
- 13↑
Ankarberg-Lindgren C, Norjavaara E. Twenty-four hours secretion pattern of serum estradiol in healthy prepubertal and pubertal boys as determined by a validated ultra-sensitive extraction RIA. BMC Endocrine Disorders 2008 8 10. (https://doi.org/10.1186/1472-6823-8-10)
- 14↑
Stocco C Aromatase expression in the ovary: hormonal and molecular regulation. Steroids 2008 73 473–487. (https://doi.org/10.1016/j.steroids.2008.01.017)
- 15↑
Huang H, Wang CQ, Tian QJ. Clinical features and management of 33 patients with 46,XX pure gonadal dysgenesis. Gynecological Endocrinology 2016 32 995–998. (https://doi.org/10.1080/09513590.2016.1190820)
- 16↑
Wilson CA, Heinrichs C, Larmore KA, Craen M, Brown-Dawson J, Shaywitz S, Ross J, Klein KO. Estradiol levels in girls with Turner’s syndrome compared to normal prepubertal girls as determined by an ultrasensitive assay. Journal of Pediatric Endocrinology and Metabolism 2003 16 91–96. (https://doi.org/10.1515/jpem.2003.16.1.91)
- 17↑
Neely EK, Hintz RL, Wilson DM, Lee PA, Gautier T, Argente J, Stene M. Normal ranges for immunochemiluminometric gonadotropin assays. Journal of Pediatrics 1995 127 40–46. (https://doi.org/10.1016/s0022-3476(9570254-7)
- 18↑
Hobeika E, Armouti M, Kala H, Fierro MA, Winston NJ, Scoccia B, Zamah AM, Stocco C. Oocyte-secreted factors synergize with FSH to promote aromatase expression in primary human cumulus cells. Journal of Clinical Endocrinology and Metabolism 2019 104 1667–1676. (https://doi.org/10.1210/jc.2018-01705)
- 19↑
François CM, Petit F, Giton F, Gougeon A, Ravel C, Magre S, Cohen-Tannoudji J, Guigon CJ. A novel action of follicle-stimulating hormone in the ovary promotes estradiol production without inducing excessive follicular growth before puberty. Scientific Reports 2017 7 46222. (https://doi.org/10.1038/srep46222)
- 20↑
Selby C Sex hormone binding globulin: origin, function and clinical significance. Annals of Clinical Biochemistry 1990 27 532–541. (https://doi.org/10.1177/000456329002700603)
