Genetics of pubertal delay

Abstract The timing of pubertal development is strongly influenced by the genetic background, and clinical presentations of delayed puberty are often found within families with clear patterns of inheritance. The discovery of the underlying genetic regulators of such conditions, in recent years through next generation sequencing, has advanced the understanding of the pathogenesis of disorders of pubertal timing and the potential for genetic testing to assist diagnosis for patients with these conditions. This review covers the significant advances in the understanding of the biological mechanisms of delayed puberty that have occurred in the last two decades.


| INTRODUCTION
Delayed puberty (DP) is a common problem within the paediatric endocrinology clinic, affecting over 2% of adolescents. It is broadly defined as puberty commencing more than two standard deviations later than the mean age for the population. 1 Due to the trend towards a decreasing age of puberty onset and the diversity in pubertal timing between populations (Figure 1), some experts have argued for adopting age cut-offs for particular ethnic groups. 1 Despite this, consensus remains that absence of breast development (Tanner stage B1) by the age of 13 years in girls and testicular volume remaining less than 4 ml (Tanner stage G1) in boys by the age of 14 years is consistent with a diagnosis of DP. 3 Those patients with faltering progression through puberty, as identified by the use of puberty normograms, also need to be reviewed for conditions associated with DP. 4 DP can be caused by a variety of aetiologies, including selflimited DP (also known as constitutional delay in growth and puberty, CDGP, when associated with short stature for parental height), hypogonadotropic hypogonadism caused by a permanent or functional gonadotropin-releasing hormone (GnRH) deficit, and hypergonadotropic hypogonadism due to gonadal insufficiency. 3,5 Selflimited DP is the most common cause of DP, accounting for 63%-82% of boys and 30%-56% of girls presenting with DP. [4][5][6] This condition is associated with a normal progression through puberty but at a timing later than the healthy population. Although selflimited DP was thought to be a benign pubertal variant, many studies have found that this condition is associated with negative outcomes for adult health. 7 Self-limited DP is often seen in multiple generations of the same family, and 50%-75% of patients with self-limited DP have a positive family history, suggesting a strong genetic basis. 8 The inheritance pattern is complex, but autosomal dominant inheritance is predominant (with or without complete penetrance) and sporadic cases are also reported. 8,9 2 | GENETIC REGULATION OF PUBERTAL

TIMING IN THE GENERAL POPULATION
Evidence from twin studies has demonstrated that the timing of puberty is strongly heritable, and that genetic regulation is an important element in determining when healthy individuals enter puberty. 10,11 While environmental factors such as nutrition, emotional well-being and geographical location influence pubertal timing, estimates from epidemiological data are that 50%-80% of the variation in age of pubertal onset is under genetic regulation. [12][13][14] More recently, results of progressively larger genome wide association studies (GWAS) of age of menarche in women suggest that a large number of different genetic signals play a role in the range of pubertal timing that is observed in the general population. 15 The signals identified to date explain ∼7.4% of the population variance in age at menarche, corresponding to ∼25% of the estimated heritability, and many have concordant effects on the age at voice breaking, a corresponding milestone in males.

| GENETICS OF CENTRAL DP
3.1 | Clinical phenotypes of congenital hypogonadotropic hypogonadism (CHH) and self-limited DP Self-limited DP and CHH can present with the same phenotype, that is, delay entering puberty; however, these two conditions are different in clinical course and requirement for treatment. CHH, or Kallmann syndrome (CHH with anosmia), are pathological conditions with failure to progress through puberty which usually need intensive hormonal therapy, whereas self-limited DP is generally a more benign condition once puberty is established, either after a period of monitoring or a short treatment course of sex steroids. 16 'Red flag' signs, such as micropenis or cryptorchidism in males, or other associated signs, can be a clue to the diagnosis of CHH; however, the majority of DP patients do not have an associated red flag feature at presentation. Moreover, both conditions may present with the same hormonal profile of hypogonadism with low gonadotropin concentrations. 1 While traditionally, CHH and self-limited DP were considered as two separate conditions, it is now apparent that there is a wide spectrum of phenotypes seen in clinical practice, ranging from complete CHH with lack of pubertal development, to partial hypogonadism with an arrest of pubertal development, reversible HH in some patients post treatment, 17,18 to isolated DP. Thus, accurate diagnosis for an individual presenting with central DP in adolescence is frequently challenging. Moreover, 2.5%-15% of CHH patients have been reported to carry multiple deleterious variants in an oligogenic inheritance pattern; 35,48 with increasing discoveries the significant complexity of the inheritance of CHH is becoming more apparent. There are also numerous syndromic conditions associated with hypogonadotropic  (Table 1), including IGSF1 deficiency, which results in a syndrome of X-linked central hypothyroidism with DP and macroorchidism in male patients. 65

| Insights from CHH into the genetic basis of self-limited DP
Although self-limited DP is the most common cause of DP, the underlying genetic basis of this condition remains incompletely understood. The first information about the genetic inheritance of isolated DP was from patients with CHH or Kallmann syndrome (Figure 2), whose relatives were seen to have isolated DP, despite carrying the same genetic mutation as the proband with GnRH deficiency. Analysis of further CHH families suggested that self-limited DP and CHH may share some overlap of their pathophysiology, with homozygous mutations in genes such as GNRHR 67,68 and TAC3 and its receptor 68 causing CHH, while heterozygous carriage of the same variants was associated with the milder phenotype of self-limited DP. [67][68][69] Recently, a heterozygous mutation in a gene previously reported to cause CHH, HS6ST1, has been identified in a family segregating with pure self-limited DP. 69 In addition, analysis of a cohort of self-limited DP (n = 72) identified rare and predicted deleterious variants in CHH genes including AXL, FGFR1, HS6ST1, PROKR2, FEZF1 and TAC3, in patients with self-limited DP 48 (Figure 3). The mechanism by which these variants might contribute to a phenotype of isolated DP has not yet been fully elucidated, but may involve a reduction in the number of adult hypothalamic neurons or an impaired functionality of the GnRH neuroendocrine network, leading to a network that is less responsive to stimulation by upstream signals at pubertal onset with resultant delay.

| New discoveries in self-limited DP genetics
An increasing number of genes have been implicated in the pathogenesis of self-limited DP over the last 5 years. 16 Several strategies have been used to discover the genetic regulation underlying this condition, including interrogation of large cohorts of patients with isolated DP for potential mutations in genes relevant to the timing of puberty in the general population identified from GWAS studies, 70 and for predicted deleterious variants in genes previously recognized from patients with CHH and Kallmann syndrome. 68 The identification of the genetic basis of self-limited DP has been accelerated by the use of next-generation sequencing technology, 71 although in a recent cohort review only 24% of cases with self-limited DP who underwent whole-exome sequencing had likely causal variants identified. 72 To date, 14 genes have been identified as contributing to self-limited DP, including those identified in relatives of CHH probands and others identified from large cohorts of familial self-limited DP which have been   Syndrome. 75 In some cases, defects in this process have been implicated in the pathogenesis of self-limited DP. The first of these was the gene IGSF10, where two mutations in this gene were firstly identified in six families with self-limited DP, and Igsf10-knockdown in zebrafish embryos was shown to lead to impaired GnRH migration. 76 After that initial discovery, mutations inIGSF10 were identified in 11% of a familial self-limited DP cohort. 48 The postulated mechanism by which impairment of GnRH migration might result in self-limited DP, is that this would impair the foetal development of the hypothalamic neuroendocrine network, due to a reduced number or delayed of arrival of hypothalamic GnRH neurons. In individuals with isolated DP one can hypothesize that this might result in a moderate impairment in function of the GnRH pulse generator, with a reduced but not absent ability to be reactivated during adolescence after the mid-childhood dormancy, as is required for the onset of puberty. 76 Following this, a heterozygous mutation in HS6ST1, a gene important for GnRH development via its interaction with ANOS1 and were identified in self-limited DP patients by whole-exome sequencing (WES) analysis in a study which found that deleterious variants in this gene impaired its transcriptional activity on the GnRH promoter, resulting in reduced GnRH transcription and secretion. 84 Furthermore, mutations in TAC3 and TACR3 have also been reported to cause self-limited DP. These genes code for neurokinin B and its receptor, an important element of the KNDy neuronal complex which controls GnRH pulsatility. 85  which lead to the phenotype of CHH. Additionally, heterozygous mutations of GNRHR have been identified in patients who manifest only with self-limited DP. 67,72 Interestingly, a partial loss-of-function mutation has been described in two brothers, one of whom had selflimited DP followed by normal endocrine profiles and fertility in adult life, while the other required testosterone replacement on into adult life consistent with a diagnosis of CHH. 86

| Energy metabolism
Fat mass and obesity-associated protein (FTO) had been implicated by GWAS to have role in the timing of puberty, and to impact on BMI and risk of obesity. 94 Two rare deleterious variants in FTO were identified in 3 Finnish families from a self-limited DP cohort. 70 Patients who carried the variants had extremely low BMI since early life.
Although the mechanism by which FTO might influence pubertal timing is unclear, this may involve energy homoeostasis. FTO might act directly via the mTORC1 signalling pathway, which has role in energy balance and expression of kisspeptin in the hypothalamus, or it might affect BMI, thus influencing pubertal timing indirectly, or potentially exert an effect via both mechanisms. 70 16 Similarly, as described, over the last 5 years an increasing number of genes have been discovered by next-generation sequencing that underlie self-limited DP. Crucially, while there is some overlap in the genetic background of these conditions, the majority of mutations are distinct between the two diseases. 48 Therefore, genetic analysis can potentially be utilized to assist a clinician in distinguishing those adolescents with severe gonadotropin deficiency from those with isolated DP, allowing delivery of accurate and timely treatment to patients. Moreover, it can be helpful to facilitate appropriate counselling on likelihood of inheritance within families and for individuals undergoing fertility treatment. This is supported by a recent study of the use of targeted exome sequencing in the clinical setting to aid the differential diagnosis between CHH and self-limited DP in a cohort of 46 adolescents presenting with severe pubertal delay. 72 6 | CONCLUSION Accumulation of knowledge relating to the genetic basis of pubertal delay has greatly accelerated over the last two decades with the improvements in sequencing technologies. Identification of genetic defects underlying hypogonadotropic and hypergonadotropic hypogonadism have led to a greater understanding of the pathophysiology of these disorders. The genetic control of self-limited DP is still mostly undiscovered, but from the initial findings it appears that the pathogenic mechanisms are related to GnRH neuronal development and biology, starting from neuronal development in the embryo to transcription and secretion of GnRH in the pubertal brain. The main pathophysiology of this condition is thus likely to be due to changes in responsiveness of the GnRH neuroendocrine system, resulting from defects of development of GnRH neurons and GnRH network functionality. Further developments will led to greater clarity on the biology of these conditions and the factors that determine timing of puberty in the healthy population, and can provide the opportunity for improved diagnostics and therapies for patients with disorders of puberty.