Age-related changes in Leydig cell number, daily sperm production, and circulating hormone levels were studied in 15 men between 20 and 48 yr of age and 15 men between 50 and 76 yr of age. Testes and blood samples were obtained at autopsy less than 15 h after death due to trauma or heart attack. Leydig cell number was determined by quantitative histometric estimation of the proportion of glutaraldehyde-perfused, decapsulated testicular parenchyma occupied by Leydig cell nuclei of measured average volume in both testes of each subject. Daily sperm production was determined by phase contrast cytometry of round spermatid nuclei in homogenates of both fixed testes from each individual. LH, FSH, PRL, and testosterone in serum from the heart or large veins were quantified by RIA. No relationship was detected between any of the testicular or hormonal values and postmortem time. The average total number of Leydig cells was reduced by 44% in the older men and was negatively correlated with age (p = -0.62) in all men. Mean serum LH was elevated more than 2-fold in the older men and was positively correlated with age (p = +0.53) in all men. Neither serum testosterone nor serum PRL changed significantly with age. Daily sperm production was more than 50% lower in older men and was negatively correlated with age (p = -0.76) in all men. Serum FSH was more than 3-fold higher in the older men and was positively correlated with age (p = +0.72) in all men. The highest FSH levels were found in men with the lowest rates of sperm production, and FSH and daily sperm production were inversely correlated even after the effects of age were removed. These findings show that the response of the human testis to aging is variable and that the predictive value of age for most testicular characteristics is weak at the level of individual men. Nevertheless, age accounts for more than a third of the variation in Leydig cell number, and it explains more than half the variation in daily sperm production. Hence, age is the largest single contributing factor yet identified in the search for explanations underlying the variation in testicular structure and function found in groups of normal men.

No abstract

The human Sertoli cell population was characterized in 14 men by histometric analysis and by direct counts of nuclei in testicular homogenates. Testes obtained at autopsy were perfused with glutaraldehyde and embedded in Epon. Nucleolar and nuclear volumes were determined by the formula of a sphere given the diameter of the nucleoli or average diameter of nuclei measured at the height and width. Nuclear volume was also estimated by adding volumes of nuclear profiles in 0.5-micron serial sections. Sertoli cell number/g was calculated by the product of the percentage nucleoli or nuclei in the parenchyma, parenchymal volume, and histologic correction factor divided by the volume of a single nucleolus or nucleus. Also, Sertoli cell nuclei were counted directly in homogenates of fixed parenchyma. Number of Sertoli cells/g was similar (P greater than 0.05) whether determined by serial sections or in homogenates, but the estimate based on the nucleolar method was higher (P less than 0.01) and the nuclear measurement method was lower (P less than 0.01) than that for serial sections. A group of 37 men aged 20 to 48 yr had significantly (P less than 0.01) more Sertoli cells than did 34 men aged 50 to 85 yr. It is concluded that: 1) the homogenate method is valid for quantification of the Sertoli cell population, 2) Sertoli cells are evenly distributed in different regions of the testis, 3) the average human Sertoli cell supports relatively few germ cells, 4) the human Sertoli cell population declines with age, and 5) there is a significant relationship between sperm production rates and number of Sertoli cells.