Abstract
Men age and die, while cells in their germline are programmed to be immortal. To elucidate how germ cells maintain viable DNA despite increasing parental age, we analysed DNA from 24 097 parents and their children, from Europe, the Middle East and Africa. We chose repetitive microsatellite DNA that mutates (unlike point mutations) only as a result of cellular replication, providing us with a natural ‘cell-cycle counter’. We observe, as expected, that the overall mutation rate for fathers is seven times higher than for mothers. Also as expected, mothers have a low and lifelong constant DNA mutation rate. Surprisingly, however, we discover that (i) teenage fathers already set out from a much higher mutation rate than teenage mothers (potentially equivalent to 77–196 male germline cell divisions by puberty); and (ii) ageing men maintain sperm DNA quality similar to that of teenagers, presumably by using fresh batches of stem cells known as ‘A-dark spermatogonia’.
Highlights
In 1912, Weinberg [1] examined cases of achondroplasia and observed that last-born children were more likely to be affected than younger siblings. This indicated that the disease was at least partly owing to new mutations in the germline of the parents, and that the mutation rate depended on parental age
The incidence of mutation due to failure to copy a gene at cell division would be unlikely to have any strong relation to maternal age; a marked increase of defects with this origin, would be seen at late paternal ages’ (p. 312)
We expect one STR mutation to have occurred per 147/0.0025 cell divisions, based on the reliably known 22 cell divisions during oogenesis as a calibration benchmark
Summary
In 1912, Weinberg [1] examined cases of achondroplasia (a form of dwarfism) and observed that last-born children were more likely to be affected than younger siblings. From a forensic point of view, a little-discussed problem in such classical family studies is the effect of undisclosed nonpaternities—older men are less likely to be fertile, increasing the probability of an illegitimate child by the potentially younger wife, and any genetic difference between the father and the illegitimate child might be misinterpreted as a de novo mutation. This point is relevant to historical studies, which did not include paternity testing. For the linear regression as a whole the confidence interval of 95% was calculated, which defines the standard errors for intercept and slope
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
