STUDY QUESTION: What are the consequences of ageing on human Leydig cell number and hormonal function? SUMMARY ANSWER: Leydig cell number significantly decreases in parallel with INSL3 expression and Sertoli cell number in aged men, yet the in vitro Leydig cell androgenic potential does not appear to be compromised by advancing age. WHAT IS KNOWN ALREADY: There is extensive evidence that ageing is accompanied by decline in serum testosterone levels, a general involution of testis morphology and reduced spermatogenic function. A few studies have previously addressed single features of the human aged testis phenotype one at a time, but mostly in tissue from patients with prostate cancer. STUDY DESIGN, SIZE, DURATION: This comprehensive study examined testis morphology, Leydig cell and Sertoli cell number, steroidogenic enzyme expression, INSL3 expression and androgen secretion by testicular fragments in vitro. The majority of these endpoints were concomitantly evaluated in the same individuals that all displayed complete spermatogenesis. PARTICIPANTS/MATERIALS, SETTING, METHODS: Testis biopsies were obtained from 15 heart beating organ donors (age range: 19-85 years) and 24 patients (age range: 19-45 years) with complete spermatogenesis. Leydig cells and Sertoli cells were counted following identification by immunohistochemical staining of specific cell markers. Gene expression analysis of INSL3 and steroidogenic enzymes was carried out by qRT-PCR. Secretion of 17-OH-progesterone, dehydroepiandrosterone, androstenedione and testosterone by in vitro cultured testis fragments was measured by LC-MS/MS. All endpoints were analysed in relation to age. MAIN RESULTS AND THE ROLE OF CHANCE: Increasing age was negatively associated with Leydig cell number (R = -0.49; P < 0.01) and concomitantly with the Sertoli cell population size (R= -0.55; P < 0.001). A positive correlation (R = 0.57; P < 0.001) between Sertoli cell and Leydig cell numbers was detected at all ages, indicating that somatic cell attrition is a relevant cellular manifestation of human testis status during ageing. INSL3 mRNA expression (R= -0.52; P < 0.05) changed in parallel with Leydig cell number and age. Importantly, steroidogenic capacity of Leydig cells in cultured testis tissue fragments from young and old donors did not differ. Consistently, age did not influence the mRNA expression of steroidogenic enzymes. The described changes in Leydig cell phenotype with ageing are strengthened by the fact that the different age-related effects were mostly evaluated in tissue from the same men. LIMITATIONS, REASONS FOR CAUTION: In vitro androgen production analysis could not be correlated with in vivo hormone values of the organ donors. In addition, the number of samples was relatively small and there was scarce information about the concomitant presence of potential confounding variables. WIDER IMPLICATIONS OF THE FINDINGS: This study provides a novel insight into the effects of ageing on human Leydig cell status. The correlation between Leydig cell number and Sertoli cell number at any age implies a connection between these two cell types, which may be of particular relevance in understanding male reproductive disorders in the elderly. However aged Leydig cells do not lose their in vitro ability to produce androgens. Our data have implications in the understanding of the physiological role and regulation of intratesticular sex steroid levels during the complex process of ageing in humans. STUDY FUNDING/COMPETING INTEREST(S): This work was supported by grants from Prin 2010 and 2017. The authors have no conflicts of interest. TRIAL REGISTRATION NUMBER: N/A.