During the last decade, the role of bone as an endocrine organ has emerged and caused the conceptualization of a novel hormonal bone-testicular axis [34]. This axis moves from the established knowledge of the positive effects of androgens on both bone mineralization and resorption processes, thereby contributing to the achievement of peak bone mass and determining bone mineral density during the lifespan. Novel evidence has appeared concerning a feedback loop from the bone to the HPG axis through OCN, which is a small polypeptide (49 amino acids) well known as a noncollagenous bone matrix constituent. OCN is produced by osteoblasts and released into circulation (both directly and through active bone resorption) and has been convincingly shown in original experiments on both loss-of-function and gain-of-function OCN mouse models to be involved in peripheral tissue physiology, including muscle, adipose tissue, and pancreas, among others, by binding to the Gpcr6a receptor. Gpcr6a has been demonstrated in Leydig cells in the testis, and its stimulation by OCN influences testosterone production, testicular development, spermatogenesis, and breeding [3, 9]. Two separate research groups have recently investigated these results, which independently developed OCN-deficient mice and did not recapitulate the original endocrine phenotype [35, 36]. Subsequent debates focused on the possible influence of different genetic and environmental backgrounds [37,38,39]. However, in a recent paper by the Karsenty group, the authors were able to replicate metabolic, steroidogenic, and brain effects in mice of various genotypes and origins, in collaboration with an independent Chinese research laboratory, maintained on different genetic backgrounds, expanding on the differential role of embryonic vs. maternal OCN origin in developing and maintaining long-term organismal homeostasis [12].

Little is known about the human physiology and male hypogonadism. We decided to study subjects affected by KS as a model of hypergonadotropic hypogonadism and, as such, exhaustion of the LH-Te feedback loop. Bone metabolism is often affected in subjects with KS, as expected in a condition characterized by frequent overt hypogonadism. Specifically, fractures and reduced bone mass (osteoporosis) are frequently reported in KS [40,41,42], and bone mineral density (BMD) is frequently reduced. Bone quality assessments conducted using novel techniques, such as peripheral quantitative computed tomography (pQCT), have demonstrated low volumetric BMD in KS and reduced trabecular density [43]. Conversely, bone turnover markers, comprising both bone formation and resorption markers, are comparable to controls [44]. However, the role of TRT in preventing or reverting these changes in KS remains unclear. Hence, we aimed to evaluate the bone-testicular axis, focusing on the relationship between tOCN, hypothalamic–pituitary–gonadal (HPG) axis, and testicular endocrine function in the hypergonadotropic hypogonadal milieu of KS.

This large longitudinal cohort revealed increased tOCN levels from prepuberty into puberty, followed by a significant and marked decrease in adulthood. We then investigated the directionality of the supposed association between gonadal status and OCN, i.e., whether OCN promotes Te production by Leydig cells or if Te may be responsible for increased bone formation, thereby increasing serum tOCN levels. First, we observed that tOCN levels were lower in males with hypogonadism (hypotestosteronemic) compared to eugonadism (eutestosteronemic). This result can be explained by the known decline in tOCN concentrations with age and the significantly younger age of the eugonadal cohort compared to subjects with hypogonadism and undergoing TRT [30]. Moreover, not only was tOCN significantly reduced in subjects undergoing TRT compared to both the eugonadal and hypogonadal cohorts, but TRT significantly reduced circulating levels in our pre- and post-TRT analysis, thereby reinforcing the idea that increasing serum Te levels in males do not acutely increase serum tOCN levels.

We then studied the association between serum tOCN levels and HPG-axis hormones. We found no significant association in prepubertal children and pubertal boys with KS, which could be caused by the presence of an already evident, early HPG-axis impairment during mini puberty and prepuberty in this population [45]. Conversely, a small, although significant, positive association was present between tOCN and both LH and FSH levels in the entire adult cohort, which remained despite the exclusion of subjects undergoing TRT. Interestingly, adjusting for age and BMI, which significantly differed among groups according to gonadal status, confirmed the association between tOCN and gonadotropins and further revealed significant inverse correlations with Leydig cell function, specifically Te and cfTe concentrations, as well as with the Te/LH and cfTe/LH ratios. Hence, higher tOCN concentrations are associated with poorer testicular function and Leydig cell sensitivity and with a higher degree of HPG-axis stimulation in adults with KS.

Altogether, the available evidence supports the notion of the involvement of OCN in testicular function in males, both in conditions of physiological Te levels as well as during hypogonadal states, in a classic endocrine negative-feedback loop. Specifically, in the context of eugonadism, OCN purportedly acts by stimulating Te production both directly at the testis level (increasing Leydig cells steroidogenesis and LH sensitivity) [1, 9, 12, 46,47,48,49,50], as shown by large cohort and population studies [21,22,23,24,25], and indirectly at the central (hypothalamic-pituitary) level, inducing increased LH (and FSH) concentrations, as revealed in the present study. Conversely, Te acts by reducing OCN levels, as evidenced by the reduced circulating tOCN concentrations in subjects pre- and post-TRT (Fig. 1B, C). To the best of our knowledge, this is the first study describing the ‘acute’ effect of Te on tOCN concentrations in hypogonadic men. In fact, the available literature describing bone turnover markers in androgen deficient, non-diabetic men undergoing TRT had their closest time-point at 6 months after TRT start. Nonetheless, the available studies supported either an overall decrease in tOCN concentrations [51], or a differential trend based on baseline Te values above or below ~ 9.2 nmol/L [52], whereas concordant findings point to lower baseline tOCN values, which increase after TRT in type 2 diabetic men, who are however characterised by low bone turnover and insulin-resistance [51, 53].

It is plausible the positive effects of OCN on increasing Te levels are lost in the pathologic condition of HPG-axis ‘exhaustion’, such as the hypergonadotropic hypogonadism of KS, where higher levels of tOCN rather reflect a more severe state of hypogonadism, as shown by significant correlations with HPG-axis hormones after adjusting for confounding factors (Fig. 2A, B). The lack of negative feedback of (reduced) Te concentrations on tOCN production or release from the bone extracellular matrix may have caused this effect. These results are mostly in accordance with the scarce available literature investigating hypogonadic men. Specifically, in two studies investigating the bone-testicular axis in men affected by spinal cord injury (comprising subjects with ‘non-hypergonadotropic hypogonadism’) or idiopathic hypogonadotropic hypogonadism, a positive correlation was evidenced between tOCN and both basal total and cfT concentrations in the former group, which persisted after adjusting for potential confounders [27], whereas a positive correlation was evidenced with peak Te levels after hCG stimulation testing in the latter group [26]. Our results are also in line with a prospective study on obese subjects undergoing bariatric surgery, a condition characterised by functional hypogonadism, where the authors observed significant increases in (cf)Te, as well as both tOCN and uOCN concentrations after 9 months, with the increase in tOCN being the only independent predictor of hypogonadism recovery after multiple adjustments, further strengthening the interdependence between OCN and Te [28].

Lastly, the only available study so far to explore the bone-testicular axis in KS reported a positive association between tOCN values and serum INSL3 concentrations [29], a constitutive biomarker of Leydig cells differentiation status and number, being relatively insensitive to HPG axis control, or other acute factors [54]. Of interest, INSL3 concentrations were significantly reduced in the whole KS cohort, and especially in the TRT-treated subgroup, compared to control men, and were positively associated with tOCN in untreated men with KS only [29]. These results are not in contrast to the present findings, considering the authors did not assess an association between Leydig cells steroidogenic activity or sensitivity and tOCN concentrations, and in consideration of INSL3 being independent of the steroidogenic LH-mediated action [55, 56].

Our study presents some strengths as well as some limitations. This is the first study to thoroughly explore the bone-testicular axis in the context of hypergonadotropic hypogonadism across the lifespan, and it does so using KS as a clinical model. Second, our population was explored using a retrospective longitudinal approach at a single academic referral center for KS, and it evaluated subjects accounting for pubertal stage, gonadal status, and TRT administration. Conversely, we could not assess the specific role of uOCN, which has been proposed by some authors as the ‘metabolically active’ form of OCN, and we could not assess sex steroid levels using the current gold standard methodology of liquid chromatography–tandem mass spectrometry because of the retrospective nature of the study. Furthermore, the present study does not explore the bone health status of the enrolled KS subjects.

In conclusion, the present study is the first to explore the bone-testicular-axis in the context of hypergonadotropic hypogonadism in humans, using a large cohort of children, boys, and adult males with KS as a clinical model. We show how tOCN concentrations peak during pubertal development, demonstrate how Te exerts a negative effect on tOCN concentrations, reveal an inverse correlation between tOCN and Te output and a direct correlation between tOCN and gonadotropin levels, indicating worse testicular function, and propose a negative-feedback loop regulation of the bone-testicular-axis.