0021-972X/90/7105-1344$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 5 Printed in U.S.A.
Sex Differences in Serum Luteinizing Hormone and Testosterone in the Human Neonate during the First Few Hours after Birth* P. CORBIER, L. DEHENNIN, M. CASTANIER, A. MEBAZAA, D. A. EDWARDS, AND J. ROFFI Laboratoire d'Endocrinologie, Uniuersite de Paris-Sud, 91405 Orsay, France; Fondation de Recherche en Hormonologie, 94268 Fresnes Cedex, France; Laboratoire de Chimie Clinique, Hopital La Rabta, Tunis, Tunisia; and Department of Psychology, Emory University, Atlanta, Georgia 30322
ABSTRACT. Blood was obtained from human male and female neonates within a few minutes after birth, and at intervals thereafter for up to 21 h. Serum LH was substantially higher at birth for boys than girls. For most boys, serum LH fell precipitously during the next hour; serum LH remained low for the remainder of the period sampled in both boys and girls. In girls, serum testosterone was low at birth and remained low for at least 21 h. At birth, serum testosterone in boys was higher than
I
N the newborn male rat, a surge in serum gonadotropin occurs at birth (1) and is followed by a rapid rise in serum testosterone which peaks at about 2 h; testosterone then falls rapidly by 3-4 h to remain relatively low for the remainder of the 24 h after birth (2). This perinatal peak in serum testosterone has powerful effects on the development of mechanisms controlling positive feedback in gonadotropin secretion (3), the anatomical and functional differentiation of the accessory sex glands (4), and the development of mechanisms for adult sexual behavior (5, 6). A similar increase in serum testosterone has been documented for newborn male mice (7), ferrets (8), and horses (9). The testes of the human infant are active on the day of birth (10), and the cause of this activity has been attributed to the presence of hCG of placental origin (11, 12). In the present study we sampled blood from human male and female infants just after birth, and at regular intervals after that for 21 h, so that we could follow changes in hCG, LH, and testosterone. We show that serum hCG in both sexes declines gradually during the first 6 h after birth, but LH is elevated at birth in human Received December 26,1989. Address requests for reprints to: Philippe Corbier, Laboratoire d'Endocrinologie, batiment 491, Universite de Paris-Sud, 91405 Orsay Cedex, France. * This work was supported by a Grant from the Fondation pour la Recherche Medicale.
for girls, increased dramatically during the first 3 h after birth, and remained elevated (2 to 3 times higher than for girls) between 3 and 12 h after birth. In newborn human males, a sudden discharge of hypophyseal LH appears to stimulate neonatal secretion of testosterone by the testes. The functional significance of this phenomenon remains to be determined. (J Clin Endocrinol Metab 7 1 : 1344-1348, 1990)
male newborns relative to female newborns. Serum testosterone is relatively low at birth for both sexes, but rises dramatically for males during the first several hours after birth. Taken together, these facts suggest that the hypophyseal-gonadal axis is active in human male newborns, something which has important comparative and developmental implications. Materials and Methods Subjects and sampling procedure Blood was obtained from human neonates in Hospital La Rabta (Tunis, Tunisia). Maternal consent was obtained before blood collections. Twenty-seven human neonates (14 boys and 13 girls) served as subjects. No twins were included in the study. Infants were vaginally delivered at term (39-42 weeks as determined by clinical examination of the mothers) and judged to be clinically normal. Blood samples (~1 ml) were obtained by venous puncture from the arm of each infant. The first sample was taken within a few minutes after birth (0 h). For some of the infants we also took samples at 1, 3, 6, 12, and 21 h after birth, although not every infant was sampled at every interval. We also drained venous umbilical cord blood from 7 of the infants (4 girls, 3 boys) into individual tubes for assay. Each blood sample was placed into an individual tube containing heparin as an anticoagulant, and the tube was cooled on dry ice until centrifuging. Each sample was centrifuged for 5 min, after which the serum
1344
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
SEX DIFFERENCES IN SERUM LH AND TESTOSTERONE IN NEONATES was frozen at -20 C. Frozen samples were packed in ice and transported by air from Tunis to Paris for hormone assay.
1345
Results Serum hCG
Determination of hCG, LH, and testosterone hCG was determined by the principle of sandwich assay using two monoclonal antibodies and j0hCG (kit RIA Gnost Behring). RIA Gnost hCG has been standardized against the first IRP (WHO 75-537), in milli International units per ml. LH was determined according to the bio-Merieux one-time RIA procedure using two monoclonal antibodies (Kit 125I-hLH COATRIA). The procedure was slightly modified to increase its sensitivity: specifically, samples were incubated overnight (instead of 1 h) at 20 C, and we doubled the number of points on the curve at the weakest concentrations. Results are expressed in terms of milliunits of standard LH 6840 (first IRP) per ml. The limit of detection for the assay is at 0.25 mlU/ml. Cross-reactivity of hLH against hFSH, hTSH, and hCG is less than 0.1, less than 0.1, and less than 10~5, respectively. RIA of testosterone were first performed in our samples under conditions commonly used for plasma determinations in adult women or children (with adequate extraction and chromatographic separation). However, RIA results were consistently higher than those obtained by a reference method based on isotope dilution mass-spectrometry. The latter technique was therefore adopted. Testosterone assay was performed by gas chromatographymass spectrometry associated with stable isotope dilution as described previously (13) with minor modifications as listed below. Fifty microliters of plasma were diluted to 200 ^1 with twice-distilled water, and 100 pg isotope labeled internal standard [3,4-13C2]testosterone were added (10 ^1 ethanolic solution containing 10 ng/ml). After equilibration at room temperature for 1 h, extraction was performed with 3 ml mixture of nhexane-diethyl ether (4:1, by volume). The organic phase was evaporated and the residue was purified by chromatography on a column (120 X 4 mm) of Sephadex LH-20 swollen in a mixture of n-hexane-ethanol-acetic acid (80:20:1, by volume). The first 1.5 ml eluent were discarded and testosterone was eluted in the next 3 ml. The chromatographic fraction containing testosterone was evaporated, and the diheptafluorobutyrate was formed. The final residue was taken up in 30 n\ n-heptane, and 6 n\ were deposited on the solid injection needle. The ions at nominal masses 680 and 682 were used for calculation of concentrations. Quantitative determination of testosterone in the plasma samples was possible down to a concentration of 300 pg/ml with a sample size of 50 pi serum. On the limited sample volumes which were available for testosterone determination (50 fi\, or less) quantification by selected ion monitoring was made on around 10 pg testosterone derivative injected into the gas chromatograph.
Umbilical cord blood for four girls and three boys was assayed for hCG. Values range from 8 to 116 IU/L and variability appears to be independent of the infant's sex (Fig. 1). Serum hCG for 6 girls and 5 boys was assayed from blood samples taken between 0 and 6 h after birth. Values for each individual are shown in Fig. 1. hCG falls gradually during the first 6 h after birth. Serum hCG values vary considerably between individuals, but there is no clear sex-related difference in serum hCG at any point for which we have data. Serum LH Umbilical cord blood was assayed for LH in five girls and five boys. Values for boys and girls were quite low and near the limit of sensitivity for the assay (Fig. 2). In girls, serum LH at birth is relatively low, and levels at 1, 3, and 6 h after birth are not appreciably different from LH levels at birth. In contrast, serum LH in boys is quite high at birth (mean = 1.27 UI/L ± 0.44 at 0 h), and is significantly higher than in girls (P < 0.05). For most of the boys sampled, LH level falls precipitously during the next hour to reach a level (mean = 0.055 UI/L ± 0.044), approaching the limit of the assay's sensitivity. From that point, serum LH in boys remains relatively low (Fig. 2), and differences between the sexes are not significant at 1, 3, or 6 h after birth. 125
100-
D 75-
O O 50 25
Statistical analysis For purposes of calculation and analysis, any time a hormone value fell below the limit of the sensitivity of the assay the value was considered equal to the limit. Group means are reported (± SEM). Differences between group means were analyzed using the t test for independent groups (two-tailed).
1
Hours
FIG. 1. Serum hCG concentration in umbilical cord (C) and peripheral circulation for human neonates taken at different intervals after birth (0,1, 3, or 6 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
CORBIER ET AL.
1346
JCE & M • 1990 Vol 71 • No 5
between 3 and 12 h after birth, testosterone values for neonatal boys are 2 to 3 times higher than for girls (P < 0.001 at 3 and 12 h). There is no overlap in the distribution of serum testosterone values for boys and girls at 0, 1, 3, and 12 h. Even at 21 h, serum testosterone for most boys is quite high and there is little overlap in the distribution of values for boys and girls. Discussion
Hours FlG. 2. Serum LH concentration in umbilical cord (C) and peripheral circulation for human neonates taken at different intervals after birth (0,1, 3, or 6 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls.
•—•boys 0 ogirls
15
10
0 1 3
12
21 Hours
FIG. 3. Serum testosterone concentration for human neonates taken at different intervals after birth (0, 1, 3, 12, or 21 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls. Differences between the sexes are significant (P < 0.05) for each period that samples were taken.
Serum testosterone We took blood samples from 13 girls and 14 boys at birth (0 h) and analyzed each for testosterone. We also obtained blood samples from some of these infants at 1, 3, 12, and 21 h after birth. The individual serum testosterone levels are shown in Fig. 3. In girls, serum testosterone is relatively low at birth and remains low for at least 21 h. In boys, serum testosterone at birth exceeds that for girls (P < 0.0001), and increases dramatically during the first 3 h after, starting from an average of 3.79 ± 0.31 n mol/L at 0 h to 6.69 ± 0.76 n mol/L at 1 h (P < 0.01). At their highest levels,
hCG in cord blood varies considerably between individuals, as does hCG in blood taken from peripheral venipuncture of infants at birth, and sex differences in hCG are not apparent at any time sampled. Serum hCG declines gradually during the first 6 h after birth in boys and girls. This gonadotropin is presumably of placental origin, and its gradual decrease likely reflects the metabolism of the hormone by the infant (14-17). During the end of gestation, plasma LH secretion in fetal human males and females is virtually identical (18). In accordance with this, and other studies of cord serum (19, 20), we find that the concentration of LH in umbilical cord blood is relatively low, and not significantly different for males and females. In contrast, when measured at birth, LH in peripheral blood for newborn males is substantially higher than that measured in cord blood, and we find levels in males that average 10 times higher than serum LH levels in newborn females. This result parallels our previous observations in the rat where we found in the males an LH surge at birth, which was lacking in the females (21). Although sample size is small, with the exception of only one case, there is no overlap in the distribution of serum LH values for boys and girls at birth, and there is no doubt that this sex difference is real. To our knowledge this is the first time a sex difference in serum LH has been reported for the human neonate, probably because no other study has sampled infants at birth. Serum LH for most of the males decreased precipitously during the first hour after birth, and in accord with other studies (22-24), we find no clear evidence of a sex difference in serum LH at any other time during the next 6 h. The ability of the human fetal testis to synthesize and secrete testosterone appears early in fetal development, and by the eighth week of pregnancy, and testes are synthesizing a measurable amount of testosterone (25). In males, serum testosterone increases between 8-17 weeks of pregnancy (25), and it is during this period that anatomical masculinization of the fetus is completed (26, 27). This period of intense endocrine activity is followed by a period during which the testes show signs of involution (29), and testosterone secretion decreases; by the end of gestation serum testosterone levels in fetal males approach those found in fetal females (18, 27).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
SEX DIFFERENCES IN SERUM LH AND TESTOSTERONE IN NEONATES
While other studies have measured serum testosterone in human males and females on the day of birth (10, 30, 31), to our knowledge the present study is the first to measure changes in serum testosterone in human infants from the first few moments after birth. We find that serum testosterone in males increases dramatically during the first 3 h after birth, and remains high until at least 12 h. This testosterone is presumably of testicular origin, since serum testosterone in females is relatively low at all of the time points sampled in this study. Forest and Cathiard (10) first described changes in serum testosterone in the human male neonate. They took samples beginning at 5 h after birth and showed that testosterone in boys is highly variable (0.6-5.7 ng/ ml) between 5 and 24 h, but higher on average than serum testosterone values for girls at the same time, and higher than values obtained from boys 2-4 days after birth. Others (30, 31) have also shown that serum testosterone in boys declines during the first few days after parturition. The present study goes beyond this, however, to clearly document a large increase in testosterone in blood during the first 3 h after birth, suggesting the rapid activation of the neonatal testes during this period. Serum hCG declines gradually during the first 6 h after birth, the time during which serum testosterone is rapidly rising. Thus, it seems unlikely that hCG plays a major role in the presumed perinatal activation of the human testis as has been earlier suggested (11,12). The increase in serum testosterone is preceded by an elevated level of serum LH, and it seems likely that the two phenomena are causally related. More specifically, testicular secretion of testosterone during the first hours after birth is probably triggered by a sudden discharge of hypophyseal LH into the bloodstream, which is apparently itself precipitated by parturition. These phenomena are remarkably similar to ones that we have clearly documented in rats (1). In newborn male rats, the sudden and transient increase in serum gonadotropin observed at birth is followed by a rapid increase in serum testosterone which peaks about 2 h after birth (2). In rats, the perinatal surge of testosterone is clearly of testicular origin since it is not seen in females or males castrated at birth (32). The male rat is not unique in this regard: a similar peak in serum testosterone, with roughly the same temporal characteristics, has been found in male mice (7), ferrets (8), and horses (9). The results of the present study clearly suggests that in human males as well, the testes are particularly active during the first few hours after birth. We assume that the perinatal increase in serum testosterone in human males is causally related to the increase in serum LH, presumably of hypophyseal origin, which precedes it. The perinatal LH surge in male rats is completely absent in the females (21). Exactly why the
1347
pituitary of the male is activated at this time is not known. The pituitary itself does not appear to be sexually differentiated in the rat (33), and it does not seem likely that a sex difference in hypophyseal secretion is related to differences in the birth process per se, since there is no good evidence that the process is different depending on the sex of the child. Wolffian duct and genital differentiation is guided by the action of testicular hormones which are secreted in abundance between 3-4 months of gestation (25, 27). Some fetal androgenic action could also condition the manner which the hypophyseal-gonadal axis will ultimately respond to parturition, and that could form the basis for the sex difference in hypophyseal function apparent at birth. It is possible that stimuli associated with birth activate the hypothalamus of males and females equally, but that differences in LH secretion between males and females are due to the form of the secreted molecules. Kaplan et al. (34) have detected much larger quantities of a-subunits than /3-subunits in the serum and hypothalamus of fetal human males and females. The male could respond to the stimuli of birth by the secretion of native LH, and the female might respond by a secretion of a-subunits not detected by the antibody used in conducting our assay. In this case, the sex difference that we observed would be at the level of the pituitary rather than the hypothalamus. This hypothesis, of course, could be easily tested by measuring the secretion of the different subunits in the newborn. In male rats, perinatally elevated levels of testosterone play an important role in promoting behavioral masculinization (6) and defeminization (5), and also appear to suppress the development of mechanisms for positive feedback (3). In mice, neonatal exposure to testicular androgens has a potent effect on the development of hormone-sensitive mechanisms for aggression (35). In monkeys, prenatal exposure to testicular androgens has important effects on both anatomical and behavioral sexual differentiation (36), and it is usually presumed that significant organizational effects of gonadal hormones are restricted to the prenatal period in primate species. The present study and a preliminary report from our laboratory (37) are the first studies to demonstrate an abrupt elevation in serum testosterone in the first few hours after birth in human male neonate. The extent to which this phenomenon contributes to the sexual differentiation of brain and behavior in the human male remains to be determined. Acknowledgment The authors thank H. Soualmia for her many hours of help in obtaining blood samples.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
1348
CORBIER ET AL.
References 1. Roffi J, Corbier P, Kerdelhue B. Stimulation de la secretion de LH et de FSH et augmentation du poids testiculaire a la naissance, chez le Rat. C R Acad Sci (Paris). 1977;284:1313-1316. 2. Corbier P, Kerdelhue B, Picon R, Roffi J. Changes in testicular weight and serum gonadotropin and testosterone levels before, during and after birth in the perinatal rat. Endocrinology 1978;103:1985-1991. 3. Corbier P. Sexual differentiation of positive feedback: effect of hour of castration at birth on estradiol-induced luteinizing hormone secretion in immature male rats. Endocrinology. 1985; 116:142-146. 4. Corbier P, Pestis J, Roffi J. L'hyperactivite testiculaire du rat nouveau-ne controle la differenciation fonctionnelle des glandes sexuelles accessoires. Pathol Biol. 1988;36:971. 5. Corbier P, Roffi J, Rhoda J. Female sexual behavior in male rats: effect of hour of castration at birth. Physiol Behav. 1983;30:613616. 6. Roffi J, Chami F, Corbier P, Edwards DA. Testicular hormones during the first few hours after birth augment the tendency of adult male rats to mount receptive females. Physiol Behav. 1987;39:625-628. 7. Motelica-Heino I, Castanier M, Corbier P, Edwards DA, Roffi J. Testosterone levels in plasma and testes of neonatal mice. J Steroid Biochem. 1988;31:283-286. 8. Erskine MS, Tobet SA, Baum MJ. Effect of birth on plasma testosterone, brain aromatase activity, and hypothalamic estradiol in male and female ferrets. Endocrinology. 1988;122:524-530. 9. Corbier P, Motelica-Heino I, Roffi J. L'hyperactivite testiculaire neonatale: etude comparative. J Physiol (Paris). 1987;82:24A. 10. Forest MG, Cathiard AM. Pattern of plasma testosterone and A4 androstenedione in normal newborns: evidence for testicular activity at birth. J Clin Endocrinol Metab. 1975;41:977-980. 11. Forest MG, Cathiard AM, Bourgeois J, Genoud J. Androgenes plasmatiques chez le nourrisson normal et premature. Relation avec la maturation de l'axe hypothalamo-hypophyso-gonadique. In: International symposium on sexual endocrinology of the perinatal period. Colloque INSERM Lyon. 1974;32:315-335. 12. Forest MG. Activites testiculaires en fonction de l'age. Periode neonatale. In: Medecine de la reproduction masculine, Schaison G, Bouchard J, Mahoudeau J, Labrie F, eds Paris: Flammarion 1984;p 147. 13. Silberzahn P, Dehennin L, Zwain IH, Leymarie P. Identification and measurement of testosterone in plasma and follicular fluid of the mare, using gas-chromatography-mass spectrometry associated with isotope dilution. J Endocrinol. 1983;97:51-56. 14. Ho Yuen B, Mincey EK. Human chorionic gonadotropin, prolactin, estriol, and dehydroepiandrosterone sulfate concentrations in cord blood of premature and term newborn infants: relationship to the sex of the neonate. Am J Obstet Gynecol. 1987;156:396-400. 15. Lauritzen C, Lehmann WD. Levels of chorionic gonadotrophin in the newborn infant and their relationship to adrenal dehydroepiandrosterone. J Endocrinol. 1967;39:173-182. 16. Geiger W. Radioimmunological determination of human chorionic gonadotropin, human placental lactogen, growth hormone and thyrotropin in the serum of mother and child during the early puerperium. Horm Metab Res. 1973;5:342-346. 17. Midgley Jr AR, Jaffe RB. Regulation of human gonadotropins II Disappearance of human chorionic gonadotropin following delivery. J Clin Endocrinol Metab. 1968;28:1712-1718. 18. Takagi S, Yoshida T, Tsubata K, Ozaki H, Fujii TK, Nomura Y, Sawada M. Sex differences in fetal gonadotropins and androgens.
JCE & M • 1990 Vol71-No5
J Steroid Biochem. 1977;8:609-620. 19. Crosignani PG, Nencioni T, Brambati B. Concentration of chorionic gonadotropin and chorionic sommamotrophin in maternal serum, amniotic fluid and cord serum at term. J Obstet Gynaecol. 1972;79:122-126. 20. Penny R, Olambiwonnu 0, Frasier D. Follicle stimulating hormone (FSH) and luteinizing hormone-human chorionic gonadotropin (LH-HCG) concentrations in paired maternal and cord sera. Pediatrics. 1974;53:41-47. 21. Corbier P, Roffi J, Kerdelhue B. Comparaison des taux seriques de LH et de FSH chez les rats males et femelles au cours de la periode perinatale. C R Acad Sci (Paris). 1979;286:1157-1160. 22. Cacciari E, Cicognani A, Pirazoli P, Dallacas P, Mazzaracchio MA, Tassoni P, Bernardi F, Salardi S, Zappulla F. GH, ACTH, LH and FSH behaviour in the first seven days of life. Acta Paediatr Scand. 1976;65:337-341. 23. Hagen C, McNeilly AS. The gonadotropic hormones and their subunits in human maternal and fetal circulation at delivery. Am J Obstet Gynecol. 1975;121:926-930. 24. Winter JSD, Faiman C, Hobson WC, Prasad AV, Reyes FI Pituitary-gonadal relations in infancy: I Patterns of serum gonadotropin concentrations from birth to four years of age in man and chimpanzee. J Clin Endocrinol Metab. 1975;40:545-551. 25. Siiteri PK, Wilson JD. Testosterone formation and metabolism during male sexual differentiation in the human embryo. J Clin Endocrinol Metab. 1974;37:113-125. 26. Reyes FI, Boroditsky RS, Winter JSD, Faiman C. Studies on human sexual development. II Fetal and maternal serum gonadotropin and sex steroid concentrations. J Clin Endocrinol Metab. 1974;38:612-917. 27. Reyes FI, Winter JSD, Faiman C. Studies on human sexual development. I. Fetal gonadal and adrenal sex steroids. J Clin Endocrinol Metab. 1973;37:74-78. 28. Pelliniemi LJ, Niemi M. Fine structure of the human foetal testis. Z Zellforsch. 1969,99:507-522. 29. Abramovich DR, Rowe P. Foetal plasma testosterone levels at midpregnancy and at term: relationship to foetal sex. J Endocrinol. 1973;56:621-622. 30. Stahl F, Gotz F, Poppe I, Amendt P, Dorner G. Pre- and early postnatal testosterone levels in rat and human. In: Dorner G, Kawakami M. eds. Hormones and Brain Development. Amsterdam: Elsevier. 1978; p 99-109. 31. Tapanainen J. Hormonal changes during the perinatal period: serum testosterone, some of its precursors, and FSH and prolactin in preterm and fullterm male infant cord blood and during the first week of life. J Steroid Biochem. 1983;18:13-18. 32. Corbier R, Roffi J, Rhoda J, Kerdelhue B. Increased activity of the hypothalamo-pituitary-testicular axis in the rat at birth: implication in the sexual differentiation of the brain? In: Serio M, ed. Sexual differentiation: basic and clinical aspects. New York: Raven Press. 1984;p 133-148. 33. Harris GW, Jacobsohn D. Functional grafts of the pituitary gland. Proc Roy Soc (Biol). 1952;139:263-273. 34. Kaplan SL, Grumbach MM, Aubert ML. a and /? glycoprotein hormone subunits (hLH, hFSH, hCG) in the serum and pituitary of the human fetus. J Clin Endocrinol Metab. 1976;42:995-998. 35. Edwards DA Mice: fighting by neonatally androgenized females. Science. 1968;161:1027. 36. Young CY, Goy RW, Phoenix CH. Hormones and sexual behavior. Science. 1964;143:212-218. 37. Corbier P, Roffi J, Rhoda J, Lebrun F, Sureau C. Increased androgen levels in the human newborn male at birth. Exp Clin Endocrinol (Life Sci Adv). 1987;6:149-150.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
Vol. 71, No. 5 Printed in U.S.A.
Sex Differences in Serum Luteinizing Hormone and Testosterone in the Human Neonate during the First Few Hours after Birth* P. CORBIER, L. DEHENNIN, M. CASTANIER, A. MEBAZAA, D. A. EDWARDS, AND J. ROFFI Laboratoire d'Endocrinologie, Uniuersite de Paris-Sud, 91405 Orsay, France; Fondation de Recherche en Hormonologie, 94268 Fresnes Cedex, France; Laboratoire de Chimie Clinique, Hopital La Rabta, Tunis, Tunisia; and Department of Psychology, Emory University, Atlanta, Georgia 30322
ABSTRACT. Blood was obtained from human male and female neonates within a few minutes after birth, and at intervals thereafter for up to 21 h. Serum LH was substantially higher at birth for boys than girls. For most boys, serum LH fell precipitously during the next hour; serum LH remained low for the remainder of the period sampled in both boys and girls. In girls, serum testosterone was low at birth and remained low for at least 21 h. At birth, serum testosterone in boys was higher than
I
N the newborn male rat, a surge in serum gonadotropin occurs at birth (1) and is followed by a rapid rise in serum testosterone which peaks at about 2 h; testosterone then falls rapidly by 3-4 h to remain relatively low for the remainder of the 24 h after birth (2). This perinatal peak in serum testosterone has powerful effects on the development of mechanisms controlling positive feedback in gonadotropin secretion (3), the anatomical and functional differentiation of the accessory sex glands (4), and the development of mechanisms for adult sexual behavior (5, 6). A similar increase in serum testosterone has been documented for newborn male mice (7), ferrets (8), and horses (9). The testes of the human infant are active on the day of birth (10), and the cause of this activity has been attributed to the presence of hCG of placental origin (11, 12). In the present study we sampled blood from human male and female infants just after birth, and at regular intervals after that for 21 h, so that we could follow changes in hCG, LH, and testosterone. We show that serum hCG in both sexes declines gradually during the first 6 h after birth, but LH is elevated at birth in human Received December 26,1989. Address requests for reprints to: Philippe Corbier, Laboratoire d'Endocrinologie, batiment 491, Universite de Paris-Sud, 91405 Orsay Cedex, France. * This work was supported by a Grant from the Fondation pour la Recherche Medicale.
for girls, increased dramatically during the first 3 h after birth, and remained elevated (2 to 3 times higher than for girls) between 3 and 12 h after birth. In newborn human males, a sudden discharge of hypophyseal LH appears to stimulate neonatal secretion of testosterone by the testes. The functional significance of this phenomenon remains to be determined. (J Clin Endocrinol Metab 7 1 : 1344-1348, 1990)
male newborns relative to female newborns. Serum testosterone is relatively low at birth for both sexes, but rises dramatically for males during the first several hours after birth. Taken together, these facts suggest that the hypophyseal-gonadal axis is active in human male newborns, something which has important comparative and developmental implications. Materials and Methods Subjects and sampling procedure Blood was obtained from human neonates in Hospital La Rabta (Tunis, Tunisia). Maternal consent was obtained before blood collections. Twenty-seven human neonates (14 boys and 13 girls) served as subjects. No twins were included in the study. Infants were vaginally delivered at term (39-42 weeks as determined by clinical examination of the mothers) and judged to be clinically normal. Blood samples (~1 ml) were obtained by venous puncture from the arm of each infant. The first sample was taken within a few minutes after birth (0 h). For some of the infants we also took samples at 1, 3, 6, 12, and 21 h after birth, although not every infant was sampled at every interval. We also drained venous umbilical cord blood from 7 of the infants (4 girls, 3 boys) into individual tubes for assay. Each blood sample was placed into an individual tube containing heparin as an anticoagulant, and the tube was cooled on dry ice until centrifuging. Each sample was centrifuged for 5 min, after which the serum
1344
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
SEX DIFFERENCES IN SERUM LH AND TESTOSTERONE IN NEONATES was frozen at -20 C. Frozen samples were packed in ice and transported by air from Tunis to Paris for hormone assay.
1345
Results Serum hCG
Determination of hCG, LH, and testosterone hCG was determined by the principle of sandwich assay using two monoclonal antibodies and j0hCG (kit RIA Gnost Behring). RIA Gnost hCG has been standardized against the first IRP (WHO 75-537), in milli International units per ml. LH was determined according to the bio-Merieux one-time RIA procedure using two monoclonal antibodies (Kit 125I-hLH COATRIA). The procedure was slightly modified to increase its sensitivity: specifically, samples were incubated overnight (instead of 1 h) at 20 C, and we doubled the number of points on the curve at the weakest concentrations. Results are expressed in terms of milliunits of standard LH 6840 (first IRP) per ml. The limit of detection for the assay is at 0.25 mlU/ml. Cross-reactivity of hLH against hFSH, hTSH, and hCG is less than 0.1, less than 0.1, and less than 10~5, respectively. RIA of testosterone were first performed in our samples under conditions commonly used for plasma determinations in adult women or children (with adequate extraction and chromatographic separation). However, RIA results were consistently higher than those obtained by a reference method based on isotope dilution mass-spectrometry. The latter technique was therefore adopted. Testosterone assay was performed by gas chromatographymass spectrometry associated with stable isotope dilution as described previously (13) with minor modifications as listed below. Fifty microliters of plasma were diluted to 200 ^1 with twice-distilled water, and 100 pg isotope labeled internal standard [3,4-13C2]testosterone were added (10 ^1 ethanolic solution containing 10 ng/ml). After equilibration at room temperature for 1 h, extraction was performed with 3 ml mixture of nhexane-diethyl ether (4:1, by volume). The organic phase was evaporated and the residue was purified by chromatography on a column (120 X 4 mm) of Sephadex LH-20 swollen in a mixture of n-hexane-ethanol-acetic acid (80:20:1, by volume). The first 1.5 ml eluent were discarded and testosterone was eluted in the next 3 ml. The chromatographic fraction containing testosterone was evaporated, and the diheptafluorobutyrate was formed. The final residue was taken up in 30 n\ n-heptane, and 6 n\ were deposited on the solid injection needle. The ions at nominal masses 680 and 682 were used for calculation of concentrations. Quantitative determination of testosterone in the plasma samples was possible down to a concentration of 300 pg/ml with a sample size of 50 pi serum. On the limited sample volumes which were available for testosterone determination (50 fi\, or less) quantification by selected ion monitoring was made on around 10 pg testosterone derivative injected into the gas chromatograph.
Umbilical cord blood for four girls and three boys was assayed for hCG. Values range from 8 to 116 IU/L and variability appears to be independent of the infant's sex (Fig. 1). Serum hCG for 6 girls and 5 boys was assayed from blood samples taken between 0 and 6 h after birth. Values for each individual are shown in Fig. 1. hCG falls gradually during the first 6 h after birth. Serum hCG values vary considerably between individuals, but there is no clear sex-related difference in serum hCG at any point for which we have data. Serum LH Umbilical cord blood was assayed for LH in five girls and five boys. Values for boys and girls were quite low and near the limit of sensitivity for the assay (Fig. 2). In girls, serum LH at birth is relatively low, and levels at 1, 3, and 6 h after birth are not appreciably different from LH levels at birth. In contrast, serum LH in boys is quite high at birth (mean = 1.27 UI/L ± 0.44 at 0 h), and is significantly higher than in girls (P < 0.05). For most of the boys sampled, LH level falls precipitously during the next hour to reach a level (mean = 0.055 UI/L ± 0.044), approaching the limit of the assay's sensitivity. From that point, serum LH in boys remains relatively low (Fig. 2), and differences between the sexes are not significant at 1, 3, or 6 h after birth. 125
100-
D 75-
O O 50 25
Statistical analysis For purposes of calculation and analysis, any time a hormone value fell below the limit of the sensitivity of the assay the value was considered equal to the limit. Group means are reported (± SEM). Differences between group means were analyzed using the t test for independent groups (two-tailed).
1
Hours
FIG. 1. Serum hCG concentration in umbilical cord (C) and peripheral circulation for human neonates taken at different intervals after birth (0,1, 3, or 6 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
CORBIER ET AL.
1346
JCE & M • 1990 Vol 71 • No 5
between 3 and 12 h after birth, testosterone values for neonatal boys are 2 to 3 times higher than for girls (P < 0.001 at 3 and 12 h). There is no overlap in the distribution of serum testosterone values for boys and girls at 0, 1, 3, and 12 h. Even at 21 h, serum testosterone for most boys is quite high and there is little overlap in the distribution of values for boys and girls. Discussion
Hours FlG. 2. Serum LH concentration in umbilical cord (C) and peripheral circulation for human neonates taken at different intervals after birth (0,1, 3, or 6 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls.
•—•boys 0 ogirls
15
10
0 1 3
12
21 Hours
FIG. 3. Serum testosterone concentration for human neonates taken at different intervals after birth (0, 1, 3, 12, or 21 h). Closed circles (•) show values for individual boys; open circles (O) show values for girls. Differences between the sexes are significant (P < 0.05) for each period that samples were taken.
Serum testosterone We took blood samples from 13 girls and 14 boys at birth (0 h) and analyzed each for testosterone. We also obtained blood samples from some of these infants at 1, 3, 12, and 21 h after birth. The individual serum testosterone levels are shown in Fig. 3. In girls, serum testosterone is relatively low at birth and remains low for at least 21 h. In boys, serum testosterone at birth exceeds that for girls (P < 0.0001), and increases dramatically during the first 3 h after, starting from an average of 3.79 ± 0.31 n mol/L at 0 h to 6.69 ± 0.76 n mol/L at 1 h (P < 0.01). At their highest levels,
hCG in cord blood varies considerably between individuals, as does hCG in blood taken from peripheral venipuncture of infants at birth, and sex differences in hCG are not apparent at any time sampled. Serum hCG declines gradually during the first 6 h after birth in boys and girls. This gonadotropin is presumably of placental origin, and its gradual decrease likely reflects the metabolism of the hormone by the infant (14-17). During the end of gestation, plasma LH secretion in fetal human males and females is virtually identical (18). In accordance with this, and other studies of cord serum (19, 20), we find that the concentration of LH in umbilical cord blood is relatively low, and not significantly different for males and females. In contrast, when measured at birth, LH in peripheral blood for newborn males is substantially higher than that measured in cord blood, and we find levels in males that average 10 times higher than serum LH levels in newborn females. This result parallels our previous observations in the rat where we found in the males an LH surge at birth, which was lacking in the females (21). Although sample size is small, with the exception of only one case, there is no overlap in the distribution of serum LH values for boys and girls at birth, and there is no doubt that this sex difference is real. To our knowledge this is the first time a sex difference in serum LH has been reported for the human neonate, probably because no other study has sampled infants at birth. Serum LH for most of the males decreased precipitously during the first hour after birth, and in accord with other studies (22-24), we find no clear evidence of a sex difference in serum LH at any other time during the next 6 h. The ability of the human fetal testis to synthesize and secrete testosterone appears early in fetal development, and by the eighth week of pregnancy, and testes are synthesizing a measurable amount of testosterone (25). In males, serum testosterone increases between 8-17 weeks of pregnancy (25), and it is during this period that anatomical masculinization of the fetus is completed (26, 27). This period of intense endocrine activity is followed by a period during which the testes show signs of involution (29), and testosterone secretion decreases; by the end of gestation serum testosterone levels in fetal males approach those found in fetal females (18, 27).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
SEX DIFFERENCES IN SERUM LH AND TESTOSTERONE IN NEONATES
While other studies have measured serum testosterone in human males and females on the day of birth (10, 30, 31), to our knowledge the present study is the first to measure changes in serum testosterone in human infants from the first few moments after birth. We find that serum testosterone in males increases dramatically during the first 3 h after birth, and remains high until at least 12 h. This testosterone is presumably of testicular origin, since serum testosterone in females is relatively low at all of the time points sampled in this study. Forest and Cathiard (10) first described changes in serum testosterone in the human male neonate. They took samples beginning at 5 h after birth and showed that testosterone in boys is highly variable (0.6-5.7 ng/ ml) between 5 and 24 h, but higher on average than serum testosterone values for girls at the same time, and higher than values obtained from boys 2-4 days after birth. Others (30, 31) have also shown that serum testosterone in boys declines during the first few days after parturition. The present study goes beyond this, however, to clearly document a large increase in testosterone in blood during the first 3 h after birth, suggesting the rapid activation of the neonatal testes during this period. Serum hCG declines gradually during the first 6 h after birth, the time during which serum testosterone is rapidly rising. Thus, it seems unlikely that hCG plays a major role in the presumed perinatal activation of the human testis as has been earlier suggested (11,12). The increase in serum testosterone is preceded by an elevated level of serum LH, and it seems likely that the two phenomena are causally related. More specifically, testicular secretion of testosterone during the first hours after birth is probably triggered by a sudden discharge of hypophyseal LH into the bloodstream, which is apparently itself precipitated by parturition. These phenomena are remarkably similar to ones that we have clearly documented in rats (1). In newborn male rats, the sudden and transient increase in serum gonadotropin observed at birth is followed by a rapid increase in serum testosterone which peaks about 2 h after birth (2). In rats, the perinatal surge of testosterone is clearly of testicular origin since it is not seen in females or males castrated at birth (32). The male rat is not unique in this regard: a similar peak in serum testosterone, with roughly the same temporal characteristics, has been found in male mice (7), ferrets (8), and horses (9). The results of the present study clearly suggests that in human males as well, the testes are particularly active during the first few hours after birth. We assume that the perinatal increase in serum testosterone in human males is causally related to the increase in serum LH, presumably of hypophyseal origin, which precedes it. The perinatal LH surge in male rats is completely absent in the females (21). Exactly why the
1347
pituitary of the male is activated at this time is not known. The pituitary itself does not appear to be sexually differentiated in the rat (33), and it does not seem likely that a sex difference in hypophyseal secretion is related to differences in the birth process per se, since there is no good evidence that the process is different depending on the sex of the child. Wolffian duct and genital differentiation is guided by the action of testicular hormones which are secreted in abundance between 3-4 months of gestation (25, 27). Some fetal androgenic action could also condition the manner which the hypophyseal-gonadal axis will ultimately respond to parturition, and that could form the basis for the sex difference in hypophyseal function apparent at birth. It is possible that stimuli associated with birth activate the hypothalamus of males and females equally, but that differences in LH secretion between males and females are due to the form of the secreted molecules. Kaplan et al. (34) have detected much larger quantities of a-subunits than /3-subunits in the serum and hypothalamus of fetal human males and females. The male could respond to the stimuli of birth by the secretion of native LH, and the female might respond by a secretion of a-subunits not detected by the antibody used in conducting our assay. In this case, the sex difference that we observed would be at the level of the pituitary rather than the hypothalamus. This hypothesis, of course, could be easily tested by measuring the secretion of the different subunits in the newborn. In male rats, perinatally elevated levels of testosterone play an important role in promoting behavioral masculinization (6) and defeminization (5), and also appear to suppress the development of mechanisms for positive feedback (3). In mice, neonatal exposure to testicular androgens has a potent effect on the development of hormone-sensitive mechanisms for aggression (35). In monkeys, prenatal exposure to testicular androgens has important effects on both anatomical and behavioral sexual differentiation (36), and it is usually presumed that significant organizational effects of gonadal hormones are restricted to the prenatal period in primate species. The present study and a preliminary report from our laboratory (37) are the first studies to demonstrate an abrupt elevation in serum testosterone in the first few hours after birth in human male neonate. The extent to which this phenomenon contributes to the sexual differentiation of brain and behavior in the human male remains to be determined. Acknowledgment The authors thank H. Soualmia for her many hours of help in obtaining blood samples.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
1348
CORBIER ET AL.
References 1. Roffi J, Corbier P, Kerdelhue B. Stimulation de la secretion de LH et de FSH et augmentation du poids testiculaire a la naissance, chez le Rat. C R Acad Sci (Paris). 1977;284:1313-1316. 2. Corbier P, Kerdelhue B, Picon R, Roffi J. Changes in testicular weight and serum gonadotropin and testosterone levels before, during and after birth in the perinatal rat. Endocrinology 1978;103:1985-1991. 3. Corbier P. Sexual differentiation of positive feedback: effect of hour of castration at birth on estradiol-induced luteinizing hormone secretion in immature male rats. Endocrinology. 1985; 116:142-146. 4. Corbier P, Pestis J, Roffi J. L'hyperactivite testiculaire du rat nouveau-ne controle la differenciation fonctionnelle des glandes sexuelles accessoires. Pathol Biol. 1988;36:971. 5. Corbier P, Roffi J, Rhoda J. Female sexual behavior in male rats: effect of hour of castration at birth. Physiol Behav. 1983;30:613616. 6. Roffi J, Chami F, Corbier P, Edwards DA. Testicular hormones during the first few hours after birth augment the tendency of adult male rats to mount receptive females. Physiol Behav. 1987;39:625-628. 7. Motelica-Heino I, Castanier M, Corbier P, Edwards DA, Roffi J. Testosterone levels in plasma and testes of neonatal mice. J Steroid Biochem. 1988;31:283-286. 8. Erskine MS, Tobet SA, Baum MJ. Effect of birth on plasma testosterone, brain aromatase activity, and hypothalamic estradiol in male and female ferrets. Endocrinology. 1988;122:524-530. 9. Corbier P, Motelica-Heino I, Roffi J. L'hyperactivite testiculaire neonatale: etude comparative. J Physiol (Paris). 1987;82:24A. 10. Forest MG, Cathiard AM. Pattern of plasma testosterone and A4 androstenedione in normal newborns: evidence for testicular activity at birth. J Clin Endocrinol Metab. 1975;41:977-980. 11. Forest MG, Cathiard AM, Bourgeois J, Genoud J. Androgenes plasmatiques chez le nourrisson normal et premature. Relation avec la maturation de l'axe hypothalamo-hypophyso-gonadique. In: International symposium on sexual endocrinology of the perinatal period. Colloque INSERM Lyon. 1974;32:315-335. 12. Forest MG. Activites testiculaires en fonction de l'age. Periode neonatale. In: Medecine de la reproduction masculine, Schaison G, Bouchard J, Mahoudeau J, Labrie F, eds Paris: Flammarion 1984;p 147. 13. Silberzahn P, Dehennin L, Zwain IH, Leymarie P. Identification and measurement of testosterone in plasma and follicular fluid of the mare, using gas-chromatography-mass spectrometry associated with isotope dilution. J Endocrinol. 1983;97:51-56. 14. Ho Yuen B, Mincey EK. Human chorionic gonadotropin, prolactin, estriol, and dehydroepiandrosterone sulfate concentrations in cord blood of premature and term newborn infants: relationship to the sex of the neonate. Am J Obstet Gynecol. 1987;156:396-400. 15. Lauritzen C, Lehmann WD. Levels of chorionic gonadotrophin in the newborn infant and their relationship to adrenal dehydroepiandrosterone. J Endocrinol. 1967;39:173-182. 16. Geiger W. Radioimmunological determination of human chorionic gonadotropin, human placental lactogen, growth hormone and thyrotropin in the serum of mother and child during the early puerperium. Horm Metab Res. 1973;5:342-346. 17. Midgley Jr AR, Jaffe RB. Regulation of human gonadotropins II Disappearance of human chorionic gonadotropin following delivery. J Clin Endocrinol Metab. 1968;28:1712-1718. 18. Takagi S, Yoshida T, Tsubata K, Ozaki H, Fujii TK, Nomura Y, Sawada M. Sex differences in fetal gonadotropins and androgens.
JCE & M • 1990 Vol71-No5
J Steroid Biochem. 1977;8:609-620. 19. Crosignani PG, Nencioni T, Brambati B. Concentration of chorionic gonadotropin and chorionic sommamotrophin in maternal serum, amniotic fluid and cord serum at term. J Obstet Gynaecol. 1972;79:122-126. 20. Penny R, Olambiwonnu 0, Frasier D. Follicle stimulating hormone (FSH) and luteinizing hormone-human chorionic gonadotropin (LH-HCG) concentrations in paired maternal and cord sera. Pediatrics. 1974;53:41-47. 21. Corbier P, Roffi J, Kerdelhue B. Comparaison des taux seriques de LH et de FSH chez les rats males et femelles au cours de la periode perinatale. C R Acad Sci (Paris). 1979;286:1157-1160. 22. Cacciari E, Cicognani A, Pirazoli P, Dallacas P, Mazzaracchio MA, Tassoni P, Bernardi F, Salardi S, Zappulla F. GH, ACTH, LH and FSH behaviour in the first seven days of life. Acta Paediatr Scand. 1976;65:337-341. 23. Hagen C, McNeilly AS. The gonadotropic hormones and their subunits in human maternal and fetal circulation at delivery. Am J Obstet Gynecol. 1975;121:926-930. 24. Winter JSD, Faiman C, Hobson WC, Prasad AV, Reyes FI Pituitary-gonadal relations in infancy: I Patterns of serum gonadotropin concentrations from birth to four years of age in man and chimpanzee. J Clin Endocrinol Metab. 1975;40:545-551. 25. Siiteri PK, Wilson JD. Testosterone formation and metabolism during male sexual differentiation in the human embryo. J Clin Endocrinol Metab. 1974;37:113-125. 26. Reyes FI, Boroditsky RS, Winter JSD, Faiman C. Studies on human sexual development. II Fetal and maternal serum gonadotropin and sex steroid concentrations. J Clin Endocrinol Metab. 1974;38:612-917. 27. Reyes FI, Winter JSD, Faiman C. Studies on human sexual development. I. Fetal gonadal and adrenal sex steroids. J Clin Endocrinol Metab. 1973;37:74-78. 28. Pelliniemi LJ, Niemi M. Fine structure of the human foetal testis. Z Zellforsch. 1969,99:507-522. 29. Abramovich DR, Rowe P. Foetal plasma testosterone levels at midpregnancy and at term: relationship to foetal sex. J Endocrinol. 1973;56:621-622. 30. Stahl F, Gotz F, Poppe I, Amendt P, Dorner G. Pre- and early postnatal testosterone levels in rat and human. In: Dorner G, Kawakami M. eds. Hormones and Brain Development. Amsterdam: Elsevier. 1978; p 99-109. 31. Tapanainen J. Hormonal changes during the perinatal period: serum testosterone, some of its precursors, and FSH and prolactin in preterm and fullterm male infant cord blood and during the first week of life. J Steroid Biochem. 1983;18:13-18. 32. Corbier R, Roffi J, Rhoda J, Kerdelhue B. Increased activity of the hypothalamo-pituitary-testicular axis in the rat at birth: implication in the sexual differentiation of the brain? In: Serio M, ed. Sexual differentiation: basic and clinical aspects. New York: Raven Press. 1984;p 133-148. 33. Harris GW, Jacobsohn D. Functional grafts of the pituitary gland. Proc Roy Soc (Biol). 1952;139:263-273. 34. Kaplan SL, Grumbach MM, Aubert ML. a and /? glycoprotein hormone subunits (hLH, hFSH, hCG) in the serum and pituitary of the human fetus. J Clin Endocrinol Metab. 1976;42:995-998. 35. Edwards DA Mice: fighting by neonatally androgenized females. Science. 1968;161:1027. 36. Young CY, Goy RW, Phoenix CH. Hormones and sexual behavior. Science. 1964;143:212-218. 37. Corbier P, Roffi J, Rhoda J, Lebrun F, Sureau C. Increased androgen levels in the human newborn male at birth. Exp Clin Endocrinol (Life Sci Adv). 1987;6:149-150.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 18 November 2015. at 14:46 For personal use only. No other uses without permission. . All rights reserved.
