A stable isotope-mass spectrometric method for measuring human spermatogenesis kinetics in vivo

Purpose: Currently it is thought to take 60 to 70 days to produce and ejaculate human sperm. This estimate is derived mainly from a single older, descriptive, kinetic analysis of spermatogenesis. We developed a noninvasive method to assess germ cell turnover time accurately in vivo using stable isotope labeling and gas chromatography/mass spectrometry analyses. We confirmed the postulated length of a normal cycle of spermatogenesis. Materials and Methods: A total of 11 men with normal sperm concentrations ingested (H2O)-H-2 daily for 3 weeks. Semen samples were collected every 2 weeks for up to 90 days. Label incorporation into sperm DNA was quantified by gas chromatography/mass spectrometry, allowing calculation of the percent of new cells present. A cycle of sperm production was determined as the lag time until labeled sperm appeared in the ejaculate. Results: Labeled sperm were detected after a mean +/- SD of 64 +/- 8 days (range 42 to 76). In 1 subject the time lag was 42 days but it was at least 60 in all other subjects. In most subjects plateau labeling in sperm was not attained. In 2 subjects the rise and fall of the labeling curve was steep and reached greater than 85% new cells, suggesting rapid washout of old sperm in the epididymal reservoir. Conclusions: This direct kinetic assessment confirms a course of spermatogenesis that is on the shorter side of traditional estimates based on prior analyses. In addition, the variability observed in healthy men suggests that characteristics such as the epididymal reservoir effect may influence the modeling of in vivo spermatogenesis.