Main

Development of a specific immunoassay for circulating inhibin B (1), a glycoprotein hormone composed of two dissimilar subunits (α and βB), has led to the discovery of a novel endocrine regulation loop between the pituitary gland and the testes. In adult men and prepubertal boys, stimulation of Sertoli cells by FSH increases circulating inhibin B levels (2, 3). Inhibin B apparently mediates a negative feedback inhibition on FSH secretion because, in healthy men and in men with variable reproductive disorders, serum inhibin B and FSH levels correlate negatively (4–8). Based on this inverse relationship, establishment of a closed feedback loop between these hormones has been suggested to occur in Tanner genital stage 3 or 4 of puberty (9, 10). However, prepubertal and early pubertal-aged boys with primary hypogonadism have supranormal serum FSH concentrations (11, 12), suggesting that Sertoli cells are able to inhibit secretion of FSH before puberty.

It is well recognized that in healthy men and in men with secondary hypogonadism, hCG stimulates secretion of inhibin in addition to FSH (13, 14). The stimulatory effect of gonadotropins on the secretion of inhibin has also been shown in prepubertal boys (15). The heterologous inhibin assays are, however, unable to differentiate between the biologically active inhibin dimers and the inhibin α-subunits and their precursors. These peptides, in contrast to inhibin B, have no apparent endocrinologic significance in men (2, 4, 16). Nevertheless, the effect of hCG stimulation on the secretion of dimeric inhibin B has not been investigated.

In the present study, we investigated the relationship between serum inhibin B and FSH concentrations in prepubertal boys with cryptorchidism, because the disease is associated with degenerative changes in Sertoli cells (17). In addition, to examine the question of whether a short-term Leydig cell stimulation enhances the secretion of inhibin B and to investigate the corresponding changes in the secretion of FSH, we measured serum inhibin B, FSH, and sex steroid levels before and during the hCG treatment.

Methods

Subjects.

Sixteen boys (age range, 1–8 y), referred to the Hospital for Children and Adolescents, University of Helsinki, for an evaluation of incomplete testicular descent, were enrolled. The boys had no clinical signs of puberty, were normally developed for age, and had no other genital abnormalities (hypospadia, micropenis). The location of the testis was considered normal (scrotal) if it situated spontaneously at or could be brought to the bottom of the scrotum; a high scrotal testis situated spontaneously in a supranormal scrotal position and could not be brought to the bottom of the scrotum; a prescrotal testis could be manipulated toward the neck of the scrotum; an inguinal testis could be palpated in the inguinal canal but could not be manipulated caudally from the external orifice of the inguinal canal. The clinical characteristics of the boys are summarized in Table 1. In four boys, another of the testes was not palpable. All boys had two testes, as confirmed by clinical examination 2 wk after the hCG treatment (subject 15) or surgery performed before the age of 2 y (subjects 4, 8, 9). Of the remaining boys, eight had unilaterally and four had bilaterally undescended palpable testes. From all boys, a single peripheral venous blood sample was taken between 0830–1500 h. In addition to the common laboratory screening tests, serum inhibin B and FSH concentrations were measured in 14 boys referred for pediatric evaluation for reasons other than genital abnormality (such as suspicion of allergy and evaluation of growth). None of the boys had cryptorchidism. All the boys were younger than 2 y of age, their ages (number of boys) being as follows; 0.9 (three), 1 (two), 1.1 (two), 1.3 (two), 1.5 (one), 1.6 (three), and 1.7 (one) y. The blood was allowed to clot, and the serum was separated by centrifugation and stored at −70°C until required for analyses.

Table 1 Clinical characteristics of boys with cryptorchidism * S, scrotal; HS, high scrotal; P, prescrotal; I, inguinal; −, not palpable.† Three times at 1-wk intervals. ‡ Second blood sample was not obtained.
Full size table

Eleven boys were given hCG treatment (Pregnyl, Organon). According to the national hCG treatment recommendation, boys <3 y of age were given 3 times 1500 IU, boys aged between 3 and 6 y (subject 14, Table 1) received 3 times 3000 IU, and boys >6 y of age (subjects 15 and 16) received 3 times 5000 IU. The injections were given intramuscularly at 1-wk intervals. On the fourth day after the last (third) hCG injection, a second blood sample was obtained from nine boys (not from subjects 13 and 16). Serum was separated and stored as described above. Informed consent was obtained from the parents. The study protocol was accepted by the ethical committee of the Hospital for Children and Adolescents, University of Helsinki.

Hormone assays.

Serum FSH levels were measured using a time-resolved ultrasensitive fluoroimmunometric assay (Wallac, Finland). The FSH standards were calibrated against the second international reference preparation (IRP) of pituitary FSH/LH (78/549). The sensitivity of the assay was 0.04 IU/L, as defined by a mean + 2 SD of multiple zero sample measurements. The within-assay CV was 2.2 and 1.3% at concentrations of 0.12 and 7.5 IU/L, respectively. The between-assay CV was <5%. Serum LH levels were measured using a time-resolved ultrasensitive fluoroimmunometric assay as described previously (12). Serum inhibin B levels were measured (in duplicate) using a commercially available inhibin B ELISA kit (Serotec, UK). The standard provided by the manufacturer is human follicular fluid extracted inhibin. According to the manufacturer, the sensitivity of this assay is 15.6 pg/mL. At a mean concentration of 115 pg/mL, calculated from the repeated measurement of the same sample, the within-assay CV was <5%. The between-assay CV, determined from 11 consecutive assay runs, was <7%.

Serum testosterone levels were measured (in duplicate) with a commercially available RIA kit (Orion Diagnostica, Finland). According to the manufacturer, the assay has a sensitivity of 0.1 nmol/L and displays low cross-reactions to other steroid hormones (4.5% to 5α-dihydrotestosterone and <0.5% to other steroid hormones, including estradiol and androstenedione). This assay was used because it requires minimal amounts of sera. At a mean concentration of 15.2 nmol/L, the within-assay CV was <5%. The between-assay CV was <7%. Serum estradiol concentrations in the boys receiving hCG treatment were measured (in duplicate) with an RIA kit from Sorin Biomedica, Italy, with ether extraction of both standards and samples. The assay sensitivity, determined as a mean + 2 SD of multiple zero samples, was 24 pmol/L. The within- and between-assay CV were 15%.

Statistical analyses.

The correlations between variables were assessed using Spearman rank correlation analysis (rS = Spearman rank correlation coefficient) because of the nonlinear relationship between the variables investigated. The means and variances of serum inhibin B and FSH levels in boys younger than 2 y of age with and without cryptorchidism were compared by t and F tests, respectively. Paired sign test was used to investigate the similarity of changes in serum hormone levels in the course of the hCG treatment;t test was used to measure the quantitative changes in serum inhibin B levels. Statistical significance was accepted for p value <0.05.

RESULTS

In boys with cryptorchidism, the mean serum inhibin B concentration was 138 pg/mL (range, 62–226 pg/mL), mean FSH concentration was 0.7 IU/L (0.1–1.6 IU/L), and serum LH levels ranged between <0.1 and 0.2 IU/L. In 13 boys, circulating testosterone concentrations were below the assay detection limit (0.1 nmol/L); subjects 3, 9, and 10 (Table 1) had serum testosterone levels of 0.2 nmol/L. Subject 2 clearly differed from the other boys by having the highest gonadotropin (FSH 1.6 IU/L, LH 0.2 IU/L) levels in combination with high serum inhibin B concentration (221 pg/mL). Therefore, this boy was excluded from all correlation analyses. In the remaining 15 boys, serum inhibin B and FSH levels were inversely related (rS = −0.79, p< 0.005;Fig. 1). In addition to FSH, serum inhibin B concentration correlated negatively with age (rS = −0.80, n= 15, p< 0.005). This relationship was nonlinear, the youngest boys having the highest inhibin B levels followed by a nadir after the age of 2 y (Fig. 1).

Figure 1
figure 1

Serum (S) inhibin B levels plotted against FSH concentrations (left panel) and age (right panel) in 16 prepubertal boys with undescended testes. Arrow, subject 2 (Table 1) differed clearly from the other boys by having high serum levels of both inhibin B and FSH.

Full size image

Because inhibin B levels changed age-dependently, we investigated the relationship between serum inhibin B and FSH concentrations in a subgroup of boys between 1 and 1.7 y of age. In these boys, the inverse relationship between serum FSH and inhibin B levels remained significant (rS = −0.84, n= 11, p= 0.008;Fig. 2). In contrast, the serum levels in the 14 control boys of similar age without cryptorchidism did not display a corresponding negative association (Fig. 2). The means of serum inhibin B and FSH levels between these subgroups did not differ (subject 2 excluded). However, these boys with cryptorchidism had wider distribution of serum FSH levels than those with normally located testes (F= 0.18, p< 0.01;Fig. 2).

Figure 2
figure 2

Association of S-inhibin B with FSH (left panel) and distributions of these hormones (right panel) in 11 boys with cryptorchidism (•) and in 14 subjects with normally descended testes (○). The boys in these subgroups were of similar age (range, 0.9–1.7 y).

Full size image

Individual serum hormone concentrations in the nine boys receiving hCG treatment, measured at baseline and on the fourth day after the third hCG injection, are shown in Figure 3. Testosterone secretion was stimulated in each boy (p< 0.005). Before the treatment, none of the boys had detectable circulating estradiol concentration. hCG stimulation increased estradiol secretion in five boys (mean stimulated concentration, 33.4 pmol/L; range, 26–37 pmol/L); in subjects 3, 11, 14, and 15 (Table 1), estradiol levels remained undetectable. Inhibin B concentrations did not change uniformly (p= NS). In the five youngest boys, hCG stimulation decreased serum inhibin B levels (p< 0.03), and in the four oldest subjects receiving the treatment, the mean level increased from 91 to 135 pg/mL (p< 0.05, Fig. 3); the mean relative increase was 53% (range, 16–101%). Finally, hCG stimulation clearly affected pituitary FSH secretion; in all boys, circulating FSH levels decreased (p< 0.005, Fig. 3). On the fourth day after the last hCG injection, serum FSH, estradiol, or inhibin B levels did not correlate.

Figure 3
figure 3

S-testosterone (Testo), estradiol, inhibin B, and FSH levels in nine boys with undescended testes, measured before (-) and during (+) the hCG treatment. Note the different scale in the S-estradiol axis. In the four oldest boys receiving the treatment (arrows), the S-inhibin B level increased. Dashed line indicates sensitivity of each assay.

Full size image

DISCUSSION

In the cryptorchid testis, Sertoli cells display degenerative changes varying from mild dilatation of the endoplasmic reticulum to total destruction of the cells (17), which probably has allowed us to detect evidence of inhibin B–mediated control of pituitary FSH secretion in prepubertal subjects. In boys with undescended testes, serum levels of these hormones correlated negatively. Although the dynamic alterations in the hypothalamo-pituitary-gonadal axis reported to occur during the first 2 y of life (18) may have confounded the homogeneity of our study group, the reciprocal relationship was clear also in the subjects between 1 and 2 y of age. In addition, these boys had greater variability in serum FSH concentrations than the boys of similar age without cryptorchidism. These findings suggest that the reciprocal regulation between FSH and inhibin B is operational in early childhood.

What might be the physiologic importance of the closed regulation loop between FSH and inhibin B in prepubertal boys? It is commonly accepted that before the onset of final sexual maturation, FSH has an important role in stimulating the proliferation of immature Sertoli cells. For example, FSH stimulates Sertoli cell proliferation in rats before (19) and after (20) birth and has a mitogenic effect on immature Sertoli cells of Macaca mulatta, a subhuman primate species (21). Our previous investigation suggests that FSH is capable of inducing proliferation of immature Sertoli cells in prepubertal boys (3); thus, it is conceivable that under physiologic circumstances, the developing Sertoli cell population would exert an inhibition on pituitary FSH secretion. After early puberty, Sertoli cells differentiate and divide no more (22, 23). Thus, the inverse relationship between circulating FSH and inhibin B levels observed in the current study represents the first evidence that immature Sertoli cells in humans can inhibit FSH release. Parenthetically, the results of one of the boys failed to fit the general pattern of inverse relationship between serum inhibin B and FSH levels. This 1-y-old subject had disproportionally high inhibin B concentration together with the highest serum levels of gonadotropins of all the boys with cryptorchidism. We speculate that in this subject, the hypothalamo-pituitary-gonadal axis still displays some postnatal activity and, subsequently, that the inverse relationship between these hormones will become evident later in prepuberty; previously it has been shown that boys have much higher inhibin B levels in infancy than later in prepubertal life (9, 18).

On the other hand, one would expect evidence of reciprocal regulation between FSH and inhibin B also in boys with normally located testes. In these boys, serum inhibin B and FSH levels did not correlate negatively, a finding consistent with the recent report by Andersson et al. (18). The exact cause of this finding remains unclear, especially as these hormones correlate negatively in healthy men. However, the endocrine regulation system between FSH and inhibin B may differ in men and prepubertal boys. In men, for example, both Sertoli cells and germ cells are probably needed for the formation of inhibin B dimer, whereas, in prepubertal boys, only Sertoli cells seem to produce inhibin B (24). In addition, administration of recombinant human FSH to adult men stimulates inhibin B secretion with a maximal response after a time lag of several days (2), whereas the temporal relationship between these hormones in prepuberty has not been elucidated. Therefore, by investigating single serum samples of prepubertal boys with normally descended testes, evidence of reciprocal regulation between inhibin B and FSH may have escaped detection.

Which factors mediated the decrease in pituitary FSH secretion observed in each boy during the hCG treatment? Of gonadal hormones, only testosterone displayed a uniform reciprocal increase. An increase in estradiol levels might also have been detected by using an ultrasensitive estradiol assay (25). In five boys, inhibin B levels decreased. Administration of estrogens to prepubertal children lowers excretion of FSH in urine (26), and, in the CNS of prepubertal subjects, conversion of androgens to estrogens may occur (27). Thus, rather than by inhibin B, the decrease in FSH secretion in the present study was more likely mediated by sex steroids.

The suppressive effect of hCG on circulating FSH levels is consistent with a previous study on M. mulatta (28). In this species, administration of hCG suppressed FSH levels but immunoreactive inhibin concentrations remained unchanged, leading to a speculation that hCG may still have stimulated testicular inhibin secretion. Supporting this idea, four boys in the current study displayed an hCG-induced increase in circulating inhibin B concentration. This finding contrasts the idea of a classic feedback regulation loop between inhibin B and FSH. However, as previously pointed out by Andersson et al. (9), increasing intratesticular testosterone concentration in early puberty facilitates early stages of spermatogenesis, which may have an indirect stimulatory action on the secretion of inhibin B by Sertoli cells. Because hCG stimulates testosterone production, our observation may be attributed to mechanisms similar to those seen in boys in early puberty. Thus, the hCG-mediated increase in inhibin B concentrations is likely to be indirect via Leydig cell-Sertoli cell or Sertoli cell-germ cell interaction(s); however, the exact pattern of events by which hCG stimulates inhibin B secretion remains unclear.

In conclusion, our results show that in prepubertal boys with undescended testes, serum FSH and inhibin B levels correlate negatively, suggesting that the reciprocal regulation between these two hormones is operational already in early childhood. Moreover, stimulation of Leydig cells by hCG suppressed pituitary FSH secretion, a phenomenon apparently not mediated by inhibin B. Overall, stimulation of Leydig cells by hCG did not induce a uniform change in serum inhibin B levels. In the four oldest boys, however, the level increased, suggesting that stimulation of Leydig cells indirectly enhanced the secretion of inhibin B.