Age of Fathers, Mutation, and Reproduction

Age of Fathers, Mutation, and Reproduction. In Evolution and Human Reproduction. Martin Fieder and Susanne Huber. In the Oxford Handbook of Evolution, Biology, and Society, Edited by Rosemary L. Hopcroft. DOI 10.1093/oxfordhb/9780190299323.013.29

Our DNA consists of roughly 3.2 billion base pairs (i.e., 3.2 billion pairs of adenine–thymine and guanine–cytosine covering the genomic information of humans, most of (p. 486) whose functions we do not yet understand) that, together with epigenetic signature, make us different from each other. Currently, we have only a relatively limited understanding of the phenotypical outcomes of our genetic makeup (Jobling, Hurles, & Tyler-Smith, 2013). Clearly, human genetics is extraordinarily complex. Nevertheless, there is no doubt that these variations in the DNA make some of us better adapted than others to certain environments. Those better adapted individuals (in the respective environments) eventually end up with more descendants. Due to the reproductive benefits for those better adapted individuals, the genetic information associated with this beneficial phenotype will spread in a population. Adaptation, however, always refers to the current environment. If the environmental conditions change, then a successful adaptation to the original environment may have no or even negative consequences on fertility. Such a maladaptive condition decreases the reproductive success of its carrier or, in the worst-case scenario, causes that lineage to die out.

Most mutations are thought to be neutral—that is, exerting no or hardly detectable effects on the phenotype—and therefore have no immediate adaptive value. Other mutations are harmful, especially if they occur in protein-encoding DNA sequences leading to an altered protein. A small number of mutations, however, may ultimately lead to a phenotype better adapted than others to its current environment. Such a phenotype will be favored by selection. The actual rate of harmful, neutral, or positive mutations, however, remains difficult to estimate (Keightley, 2012), particularly the rate of mutations that are positively selected for. In two Drosophila populations, Schneider, Charlesworth, Eyre-Walker, and Keightley (2011) estimated the rate of positive selected mutations for amino acid coding sequences (i.e., non-synonymous mutations) to be between 1% and 2% of all occurring mutations.

Where do most of the mutations come from? The very recently discovered answer in humans is impressive—from the age of the father (Kong et al., 2012). According to Kong et al., the father’s age explains nearly all newly occurring (i.e., de novo) mutations in a child. Correspondingly, detrimental parental age effects have been demonstrated for a variety of Mendelian and mental disorders and even for educational attainment (for a review, see D’Onofrio et al., 2014). The reason is that in contrast to women, in whom all cell divisions in the egg are completed before birth, men continue producing sperm throughout their reproductive lives. Consequently, the number of cell divisions and chromosome replications that a sperm cell has gone through increases with the age at which the sperm is produced. This increases the risk that “errors” occur in terms of mutations (Crow, 2000).

Because the mutations induced by male age occur randomly in the human genome, the probability that they directly affect reproductive functioning is relatively low because a detrimental mutation occurring somewhere in our genome does not necessarily affect reproductive functioning. In such cases, an individual could still reproduce normally even if he or she carries a potentially harmful mutation. It would pass those harmful mutations on to the next generation, which may then accumulate over generations. It is thus conceivable that a mechanism may exist that helps avoid excessive mutation loads in future generations. We suggest that mate selection may provide such a mechanism to (p. 487) prevent too high mutation load. This view is supported by our recent findings based on a US sample (Wisconsin Longitudinal Study), in which we demonstrated that children of older fathers are less attractive (Huber & Fieder, 2014). Moreover, offspring of older fathers face a higher risk of remaining unmarried and therefore remaining childless (Fieder & Huber, 2015). Marriage was obligatory in the previously mentioned sample, thereby providing a good indicator for mating success. Comparable findings based on large human data sets have confirmed our results (Hayward, Lummaa, & Bazykin, 2015; Arslan et al., 2016). Similar effects of paternal age have also been reported in animal species ranging from bulb mites (Prokop, Stuglik, Żabińska, & Radwan, 2007) to house sparrows (Schroeder, Nakagawa, Rees, Mannarelli, & Burke, 2015). We therefore suggest that this phenomenon is a more fundamental biological principle: An individual’s mutation load could affect mate selection, thus helping to reduce the mutation load of the progeny.

This view is also in line with the mutation–selection balance theory, proposing that a balance of forces between constantly arising, mildly harmful mutations and selection causes variation in genetic quality and phenotypic condition (Miller, 2000; Keller, 2008). This makes it unlikely that the accumulation of new deleterious mutations leads to a detectable fitness decline in current human populations (Keightley, 2012). The mutation–selection balance is assumed to be particularly important in traits influenced by many genetic loci (multigenic, such as human reproduction), providing a large target size for mutations (Keller, 2008).

Although most of the mutations induced by the age of the father are considered neutral or may be harmful, a small proportion of them are advantageous and provide fitness benefits. This raises an interesting question: Are we able to detect potentially promising mutations in a mate that may be adaptive in the long term? Detecting mutations that in the future may lead to an adaptive phenotype is unlikely. We therefore assume that this is probably a random process. Nevertheless, one can speculate that individuals choose extraordinary traits in potential mates—that is, traits that may be associated with newly induced mutations. The numerous examples include the peacock’s tail (Zahavi & Zahavi, 1999), bower birds (Uy & Borgia, 2000), as well as height (Stulp, Barrett, Tropf, & Mills, 2015) and social status in men (Fieder & Huber, 2007; Nettle & Pollet, 2008; Barthold, Myrskylä, & Jones, 2012; Hopcroft, 2015). If such traits carry adaptive benefits outweighing potentially negative impacts, then selection would favor both the carrier of those mutations and the carrier’s mating partners. Accordingly, mutations induced by a father’s age can also be viewed as a “driving force” of evolution. The reason is that without mutations, evolution would not have taken place at all, and without mutations introduced into the population by male age, evolution would at least have been much slower. The positive mutations induced by age might thus be considered an “engine of evolution,” leading to new phenotypes that could potentially be selected for.

Together with the usually higher status of older men, this positive effect might partially explain women’s preference for somewhat older men (Buss, 1989). Basically, this preference reflects a trade-off between benefits associated with higher status and possible detrimental mutations caused by higher paternal age that may be passed to (p. 488) the offspring. However, because some mutations may be adaptive, overall the benefits may outweigh the costs, at least if the age difference between spouses is not too large. Accordingly, women usually prefer men who are only moderately older than themselves (Buss, 1989; Buunk, Dijkstra, Fetchenhauer, & Kenrick, 2002; Schwarz & Hassebrauck, 2012).

Future studies may aim to measure the impact of mutations directly and not just indirectly via the age of fathers, examining, for instance, if there is any evidence for a potential link between father’s age, mutation rate, marriage fertility, and social status. According to D’Onofrio et al. (2014), higher paternal age is associated with lower educational attainment in the offspring. This finding suggests a possible association between de novo mutation rate and educational attainment, leading to the question whether social status goes beyond being solely culturally determined to also contain an inherited component. At least for educational attainment, this has recently been shown (Rietveld et al., 2013).