Investigating the evolution of menopause through computational simulation

Abstract

Menopause is characterized by prolonged lifespan beyond the point of reproductive cessation. Defined so that at least 25% of adulthood is nonreproductive, humans and some toothed whale species are the only groups that have been found to exhibit menopause. Menopause is a puzzling trait that seems to contradict classical evolutionary theory that equates selection operating on reproduction to selection operating on survival. I created two computational models to gain better understanding of the evolution of menopause. The first model explored why menopause is not observed in elephants despite their being characterized by key features in common with menopausal species, specifically offspring care from older females and longevity. Simulations allowed testing the effects of varying age at reproductive cessation and levels of offspring care, modeled by decreases in interbirth intervals. I found that hypothetical populations with greatest post-reproductive lifespans, characterized by longer interbirth intervals and earlier reproductive cessation, were most likely to be out-competed by contemporary elephants. Conversely, hypothetical populations that were most reproductively competitive, those with shorter interbirth intervals and older ages of reproductive cessation, returned post-reproductive lifespans that failed to meet the 25% post-reproductive lifespan criterion for menopause. I identified a small region in the parameter space where populations that were both menopausal and reproductively competitive evolved, but the majority of that region corresponds to biologically unrealistic scenarios. The scenario that is most feasible involves an interbirth interval of 4 years and an age at reproductive cessation of 40 years. The second model studied how menopause might have evolved in humans through a behavioural strategy of ending reproduction early to avoid risk of aneuploidy later in life and diverting resources toward extant kin. I found that populations that ceased reproduction earlier and exhibited greater post-reproductive lifespan returned lower reproductive success. The model also demonstrated that the aneuploidy avoidance behaviour is most successful when reproduction ends at approximately age 50. These concepts have never been explored computationally before, so these experiments contribute a novel simulation-based perspective to the growing body of knowledge surrounding the origin and evolution of menopause.