Background

Puberty in boys is initiated when gonadotropin-releasing hormone (GnRH) secretion is amplified, inducing the secretion of gonadotropins. Pulsatile secretion of luteinizing hormone (LH) and follicle stimulating hormone (FSH) is accompanied by enlarged testicular volume, testosterone secretion and initiation of spermatogenesis [1]. The age of puberty initiation is variable across sex, ethnicity, and is influenced by genetic and nutritional factors. Precocious puberty (PP) is generally due to premature sexual steroid secretion [1,2,3,4]. In most cases, puberty is considered precocious in boys if it is initiated before 9 years old. The prevalence of PP is about 1 to 8/10000 with a female/male ratio of 9/1. Central precocious puberty (CPP) is caused by earlier pulsatile secretion of GnRH and gonadotropins [2]. Unlike delayed puberty, CPP remains idiopathic in most of the cases (ICPP) [4]. Many observations suggest a possible link between chromosomal anomalies and PP, especially in Klinefelter syndrome (KS) [5,6,7]. KS is the most frequent dysgonosomia in boys that leads to incomplete puberty and hypergonadotropic hypogonadism [8]. Indeed, most cases of PP in KS arise from germ cell tumors secreting human chorionic gonadotropin (hCG). While many cases of ICPP in KS have been reported in the literature, [9,10,11,12,13,14], the physiopathology remains unclear. Furthermore, the hormonal profile of patients with ICPP and KS is not well characterized. The study of patients with divergent problems such as ICPP and KS might improve the understanding of ICPP pathophysiology and point out factors that influence the timing of puberty initiation.

We describe a case of ICPP in a boy who, a few years later, developed hypogonadism in his adulthood secondary to KS (with 2 maternal X chromosomes). In addition to our patient, we analyze 7 other cases of ICPP in KS patients previously reported in the literature. Among patients with ICPP, those who are also suffering from KS have higher levels of gonadotropins and an exaggerated response to GnRH stimulation compared to those with a normal karyotype. This comparison highlights the possibility of a primitive testicular factor, such as precocious germ cell and Leydig cell dysfunction, participating in the induction of ICPP in KS.

Case presentation

A boy, born in 1994, was evaluated for suspicion of precocious puberty. At 8 years old, his height (150 cm) exceeded the mean for age by 3 standard deviations with a growth rate of 10 cm per year. He developed puberty signs (Tanner staging: P2/G3). Both testes measured 25x35mm, the penis was enlarged and increased in length. Bone age was advanced (10 years according to modified Sempé’s method). His past history was marked by cryptorchidism, operated at age of 1 year. No facial dysmorphia, behavioural abnormalities or mental retardation were noticed.

His hormonal profile confirmed a central origin of precocious puberty with high serum testosterone (4.3 ng/ml) for age, and detectable LH (3.5 UI/l) and FSH (7.7 UI/l) (adult male reference values for FSH: 1.1–9.9 UI/l, and LH: 0.9–6.6 UI/l).

Luteinizing hormone-releasing hormone (LHRH) test amplified LH and FSH secretion to 24 and 14 UI/l respectively (Table 1). Hypothalamic-pituitary MRI was normal. Due to advanced bone age and growth acceleration, he was treated with triptoreline (3 mg, intramuscular injection every 28 days). Under this treatment, LH, FSH and testosterone were attenuated to a prepubertal level (0.2 UI/l, 0.4 UI/l and 0.2 ng/ml respectively) (Table 1). In 2005, the treatment was stopped, and puberty resumed. At age of 20, he consulted for hypogonadism and bilateral gynecomastia. His definitive height was 178 cm (predicted height was 173 cm, with mid-parental height estimated to 166 cm) and his weight was 86 kg (+ 1 SD). The testicular size was reduced (25 × 12 mm). Further evaluation confirmed azoospermia, and low testosterone levels (1.16 ng/ml, reference values for adults > 2.75 ng/ml) with hypergonadotropic hypogonadism (LH and FSH increased to 26 and 40 UI/l respectively) (Table 1). Testicular biopsy revealed a hyalinized tissue with failed spermatozoa retrieval.

Table 1 Clinical and biological parameters of the patient during and after puberty
Full size table

Chromosome analysis was performed on GTG-banded and RHG-banded metaphases prepared from cultured peripheral blood according to the standard protocols.

The origin of the supernumerary X chromosome was investigated by fluorescent multiplex polymerase chain reaction (PCR) to study short tandem repeats polymorphisms (supplementary methods).

Peripheral blood cytogenetic analysis of our patient revealed 47 chromosomes, with an extra X chromosome in all analyzed metaphases, confirming KS. Array-comparative genomic hybridization (CGH) excluded the presence of any other genomic imbalance, except for copy number variations (CNV) previously described in the normal population according to the Database of Genomic Variants (http://projects.tcag.ca/variation). Microsatellite analysis in X chromosome showed two X chromosomes of maternal origin. Furthermore, no mutation was found in the coding regions of MKRN3 gene.

An informed written consent was obtained from the patient and his parents to perform the genetic analyses and to accept the publication of the data in this article.

Literature analysis and discussion

We collected and analyzed clinical and hormonal data in our case and 7 other cases of ICPP in KS reported in the literature [9,10,11,12,13,14]. The clinical and biological data at definitive puberty in each case were also considered. The characteristics of the 8 cases of ICPP in KS are summarized in Table 2. We also compared the hormonal profile between ICPP patients with KS (8 cases) and those with a normal karyotype reported in the literature (the most comprehensive cohort is published by Pigneur et al. [15]). Compared to patients with a normal karyotype (46,XY), KS patients have significantly higher LH and FSH levels during precocious puberty (KS + ICPP vs normal karyotype+ICPP: LH levels 9.4 ± 12 vs 1.1 ± 0.6 UI/l; FSH levels 23.1 ± 38.5 vs 2.7 ± 1.5 UI/l). Moreover, the response to LHRH in KS patients is more drastic (KS + ICPP vs normal karyotype+ICPP: LH levels post-LHRH 58 ± 48 vs 15.5 ± 0.8 UI/; FSH levels post-LHRH: 49.1 ± 62.1 vs 5.7 ± 3.9 UI/l) (Table 3). Even higher basal- and stimulated- LH and FSH levels during ICPP were noted in a KS patient with 48, XXYY karyotype [12] compared to boys with either normal karyotype [15] or 47,XXY (7 patients) (Table 2).

Table 2 Analysis of characteristics of Klinefelter patients with ICPP
Full size table
Table 3 Gonadotropin levels in patients with ICPP with a normal karyotype (adapted from Pigneur et al. [15]) and ICPP in cases with KS
Full size table

In our patient, despite the precocious puberty, target height was attained. Similarly, in other reported cases of ICPP with KS, the height was attained in all patients, with (3 patients including our case) [9, 10] or without anti-androgen treatment (5 patients). This is explained by the potential role of short-stature homeobox gene (SHOX) overexpression in such patients due to their dysgonosomia compensating for an eventual short stature risk related to PP [16]. However, testosterone levels at mid-puberty (14–15 years) and the testicular volume in these ICPP patients are identical to the classical puberty profile of KS [17,18,19,20].

Indeed, around 450 boys with ICPP under the age of 9 years have been described in the published cohorts to date, accounting for about 8% of all central precocious puberty cases before the age of 9 year and 10% before the age of 10 years (Supplementary Table 1) [21,22,23,24,25,26,27,28,29,30,31,32,33]. The incidence of ICPP in boys is around 0.05/10,000 boys under the age of 9 years and 0.24/10,000 boys under the age of 10 years [32, 34, 35], while the incidence of KS is around 15/10,000 births [36]. Thus, the description of 8 cases of KS in around 450 reported cases of ICPP in the general population suggests an “overrepresentation” of ICPP in Klinefelter patients. The risk is estimated to be 1/370 KS patients by the Health Register of Belgium [13].

The unexpected “overrepresentation” of ICPP in KS raises two hypotheses. Firstly, while gonadotropin levels in KS patients during ICPP are as high as in 15 years old KS patients without ICPP, the testicular destruction is advanced [17, 37]. Higher gonadotropin levels during ICPP in KS patients compared to those in 46,XY patients suggest that a testicular factor could be a “peripheral initiator of GnRH pulsatility” during precocious puberty that is presumed to be central (ICPP). In KS patients, high estradiol levels is one of the earliest testicular anomalies [38]. Estrogen levels explain the chronological differences in puberty between girls and boys during puberty. Estrogen increases Kiss1 expression (gene coding kisspeptin, a major regulator of GnRH activation) in the hypothalamus during puberty. This effect is exerted through estrogen receptor (Erα). If this process is disrupted, puberty could be delayed [39]. Higher estradiol levels detected early in Klinefelter patients were proposed to precociously activate gonadotropin secretion [28, 31]. Indeed, it has been recently shown that estrogen promotes the expression of genes related to precocious puberty which activate GnRH secretion [40]. Germ cell apoptosis is another early phenomenon in KS [29]. The degree of germ cell apoptosis/seminiferous tissue fibrosis and Leydig cell dysfunction in KS are best reflected by FSH and LH levels respectively. Basal- and stimulated-LH and FSH levels are even higher in the patient with 48,XXYY [12]. This suggests that gonadotropin reserve correlates with the degree of testicular destruction, which is maximal in 48,XXYY patients. This hypothesis is also supported by the description of ICPP cases in patients with Turner syndrome and cryptorchidism [5, 41, 42]. Indeed, the testicular destruction and/or higher estrogen levels in Klinefelter patients would increase the hypothalamic/pituitary gonadotropin reserve and explain why these patients have higher basal and GnRH-stimulated LH and FSH levels during ICPP. These data are interesting and highlight the timing and role of testicular destruction in dysgonosomia to initiate GnRH pulsatility.

Secondly, in our patient, both X chromosomes were of maternal origin. None of the reported cases studied the origin of the supplementary X chromosome. It is worth noting that the supplementary X chromosome is equally inherited from paternal and maternal origin. When the origin of this supplementary X chromosome is maternal, it is hypothesized that some deleterious genes could escape the natural X chromosome inactivation [43, 44]. These deleterious genes could be responsible for an earlier activation of puberty in such patients [17, 45]. In line with this hypothesis, 2 X-linked gene mutations are described in the literature to be associated with ICPP in 2 boys: NR0B1 (responsible for X-linked adrenal hypoplasia congenita) [46] and MECP2 (responsible for Rett syndrome) [47]. On the contrary, paternal X chromosome (pX) is associated with further delayed puberty than those with a maternal X chromosome (mX) [17]. Puberty signs with significant increment in LH levels appeared respectively 1.9 and 1.3 years later in patients with pX. Additionally, androgen receptor (AR) gene is located on the X chromosome. It contains CAG repeats in the N terminal domain of exon 1 that could be highly polymorphic. The length of this repetition is inversely correlated with receptor activity [48], and is found to influence clinical presentation of KS [49]. Indeed, shorter CAG repeats and lower AR methylation were linked to premature pubarche [50]. Nevertheless, no study linked CAG repeats or AR activity to precocious puberty in KS patients.

Despite a number of interesting studies investigating early onset of puberty in patients with KS, the genetic basis of idiopathic central precocious puberty in both sexes remains elusive. Further studies using molecular genetics could help understanding the mechanisms of puberty initiation, timing and progression. Specifically, identifying novel X-linked genes in boys with precocious puberty will be instrumental to shed light on the mechanisms underlying puberty initiation.

Conclusions

ICPP is extremely rare in boys. ICPP patients with KS have higher gonadotropin levels than those with a normal karyotype. This suggests a peripheral or “testicular driver” that amplifies gonadotropin secretion through a positive feedback loop, which in turn triggers the initiation of central PP in such patients. Furthermore, maternal supplementary X chromosome in KS might play a role in the timing of puberty initiation. In boys with ICPP, long term endocrinological follow up until definitive puberty and karyotype analysis are important, specially if gonadotropin levels are significantly elevated or testicular volume is reduced.