Endless paces of degeneration—applying comparative genomics to study evolution's moulding of longevity

Why can mice not live more than five years and dogs not more than 30, yet bats can live over 40 years and humans over a century? Differences in longevity between closely related species are one of the greatest mysteries in biology, and identifying the processes responsible could ultimately presage the development of therapies against a multitude of age-related diseases. The variation in mammalian longevity must have a genomic basis, with recent genome sequencing efforts opening up exciting opportunities to decipher it; some promising results are beginning to emerge. Analysis of two bat genomes revealed that a high proportion of genes in the DNA damage checkpoint–DNA repair pathway, including ATM, TP53, RAD50 and KU80, are under selection in bats [1]. This finding is exciting because these genes have been directly associated with ageing in model systems and, therefore, it points towards a potential role for averting DNA damage in longevity assurance mechanisms; a notion dating back several decades that remains contentious. In addition, the report of a systematic scan for proteins with accelerated evolution in mammalian lineages in which longevity increased over the course of their evolution, hinted that some repair systems, such as the ubiquitin–proteasome pathway and a few proteins related to DNA damage repair, might have been selected for in long-lived lineages [2]. However, much work remains to improve the signal-to-noise ratio of this and similar methods.

With decreasing costs of sequencing, the growing number of genomes of species with diverse lifespans is expected to facilitate studies in this area. As such, we can make an increasing number of comparisons such as those described above. Put simply, if we study long-lived species and find that they share genetic adaptations—for example in DNA damage response pathways—then we might assume that those adaptations are important to increase longevity. There are major intrinsic difficulties with this type of analysis, however, that one must keep in mind. Perhaps the best illustration is that despite the dramatic phenotypic divergence between humans and chimpanzees, only a relatively small number of genetic adaptations that are probably responsible for such divergence have thus far been identified [3]. One difficulty is that the genomic elements underlying species differences remain controversial. Possible processes include mutations in coding and non-coding sequences, gene family expansion and contraction, and copy number variation, all of which we think must be explored in the context of longevity adaptations. Whilst changes in regulatory regions might be important, standard methods are lacking for the detection of selection on functional non-coding sequences on a genome-wide scale and this, we think, is a limitation for progress in this area. Another limitation is that experimental validation of promising candidates is often extremely difficult to obtain.

Applying comparative genomics to study the evolution of longevity also has unique challenges. For one, the force of natural selection weakens with age, indicating that, although under low-hazard conditions selection favours genes and pathways conferring longevity, selective pressure for longevity is significantly less than for other traits. Furthermore, we think that the integration of additional data—for example gene expression and age-related phenotypic data—is crucial to link genotypes to phenotypes and identify physiological adaptations that are required for extended longevity. Unfortunately, such data and even the necessary samples to generate it are as yet only available for a subset of species. In our opinion, another crucial issue is the extent to which common mechanisms underlie the extension of longevity by evolution in different species. Just as rare variants contribute to missing human heritability, taxa-specific adaptations might contribute to longevity. It can be assumed that the environment—for example, diet—of each species will influence the physiological and biochemical pathways that must be optimized to fend off ageing and age-related diseases. However, the ageing process, despite progressing at different rates, is remarkably similar across most mammals studied [4], hinting that retarding ageing might involve adaptations in similar pathways. The degree of overlap between longevity assurance mechanisms is, in our view, a crucial determinant of how much we can expect to learn about species differences in ageing in the foreseeable future. If common pathways do indeed underlie longevity evolution in multiple species, even if involving different genetic elements in different taxa, then it is reasonable to expect that they can be identified by using comparative genomics as more genomes of short- and long-lived species are sequenced. We hope to live long enough to help unravel this age-old problem.