Most genetic roots of fungal and animal aging are hundreds of millions of years old according to phylostratigraphy analyses of aging networks

Abstract

Few studies have systematically analyzed how old aging is. Gaining a more accurate knowledge about the natural history of aging could however have several payoffs. This knowledge could unveil lineages with dated genetic hardware, possibly maladapted to current environmental challenges, and also uncover "phylogenetic modules of aging," i.e., naturally evolved pathways associated with aging or longevity from a single ancestry, with translational interest for anti-aging therapies. Here, we approximated the natural history of the genetic hardware of aging for five model fungal and animal species. We propose a lower-bound estimate of the phylogenetic age of origination for their protein-encoding gene families and protein-protein interactions. Most aging-associated gene families are hundreds of million years old, older than the other gene families from these genomes. Moreover, we observed a form of punctuated evolution of the aging hardware in all species, as aging-associated families born at specific phylogenetic times accumulate preferentially in genomes. Most protein-protein interactions between aging genes are also old, and old aging-associated proteins showed a reduced potential to contribute to novel interactions associated with aging, suggesting that aging networks are at risk of losing in evolvability over long evolutionary periods. Finally, due to reshuffling events, aging networks presented a very limited phylogenetic structure that challenges the detection of "maladaptive" or "adaptative" phylogenetic modules of aging in present-day genomes.

Keywords: Aging; Antagonistic pleiotropy hypothesis; Evolution; Geroscience; Network.

Fig. 1

Cumulative plot representing the phylostratification of orthologous protein coding genes in S. cerevisiae and M. musculus. The X-axis reports a reference dating by phylostrata corresponding to ancestral nodes within the phylogenetic lineages leading to yeast (left plots) and mouse (right plots). Phylogenetic origins of orthogroups encoding proteins with known interactions are indicated for aging genes (blue), including the subcategories “pro-longevity genes” (orange), “anti-longevity genes” (burgundy red), and “uncharacterized genes” (yellow), and for non-aging genes (black). The pie chart summarizes the number of genes in each category. A star (*) indicates when inferences of gene origination were impossible for a particular phylostratum due to limited phylogenetic coverage of the OMA database. The Y-axis represents the cumulative proportion of orthogroups from each category that appeared before a given time point. The “wave plots” illustrate when aging genes from a given ancestry accumulated proportionally less (ratio < 1) or more (ratio > 1) than non-aging genes from the same ancestry

Fig. 2

Cumulative plot representing the phylostratification of orthologous protein coding genes in humans. Cumulative plots corresponding to the phylostratigraphy of A genes associated and non-associated with aging (genes recorded in the GenAge database) and genes associated and non-associated with cellular senescence (genes recorded in the CellAge database), and B orthologous hallmark genes in humans (an age of origination was assigned to each protein coding gene from the Ensembl database, assigned to an OMA group, using the OpenGenes database to identify hallmark genes orthologs). For each plot, the X-axis reports a reference dating by phylostrata corresponding to ancestral nodes within the phylogenetic lineage leading to humans. Phylogenetic origins of orthogroups encoding proteins with known interactions are indicated for aging genes (blue) and for non-aging genes (black). The pie chart summarizes the number of genes in each category. A star (*) indicates when inferences of gene origination were impossible for a particular phylostratum due to limited phylogenetic coverage of the OMA database. The Y-axis represents the cumulative proportion of orthogroups from each category that appeared before a given time point. The “wave plots” illustrate when aging genes from a given ancestry accumulated proportionally less (ratio < 1) or more (ratio > 1) than non-aging genes from the same ancestry

Fig. 3

Cumulative plot representing the phylostratification of orthologous protein coding genes in D. melanogaster and C. elegans. The X-axis reports a reference dating by phylostrata corresponding to ancestral nodes within the phylogenetic lineages leading to this drosophila (left plots) and nematode (right plots). Phylogenetic origins of orthogroups encoding proteins with known interactions are indicated for aging genes (blue), including the subcategories “pro-longevity genes” (orange) and “anti-longevity genes” (burgundy red), and for non-aging genes (black). The pie chart summarizes the number of genes in each category. A star (*) indicates when inferences of gene origination were impossible for a particular phylostratum due to limited phylogenetic coverage of the OMA database. The Y-axis represents the cumulative proportion of orthogroups from each category that appeared before a given time point. The “wave plots” illustrate when aging genes from a given ancestry accumulated proportionally less (ratio < 1) or more (ratio > 1) than non-aging genes from the same ancestry

Fig. 4

Cumulative plot representing the phylostratification of protein interactions in M. musculus. The X-axis reports a reference dating by phylostrata corresponding to ancestral nodes within the phylogenetic lineage leading to this mouse. Phylogenetic origins of genes are reported similarly to Fig. 1. Phylogenetic origins of edges representing protein interactions are indicated for aging edges (blue) and for non-aging edges (black). The Y-axis represents the cumulative proportion of edges from each category that appeared before a given time point. The “wave plots” illustrate, from top to bottom, (i) when aging edges that were associated to a given ancestry accumulated proportionally less (ratio < 1) or more (ratio > 1) than non-aging edges from the same ancestry; (ii) the ratio of aging edges induced by aging nodes; (iii) the ratio of non-aging edges induced by non-aging nodes; (iv) the Rcn (aging edges/aging nodes)/(non-aging edges/non-aging nodes)