Atavistic Genetic Expression Dissociation (AGED) During Aging: Meta-Phylostratigraphic Evidence of Cellular and Tissue-Level Phylogenetic Dissociation
The AgeMeta aging signature is dominated by downregulated genes in the “All living organisms” strata. Upregulated genes were more important in the most ancient genes in immune and ovarian cells, but not in HPC, senescent or skin cells where downregulated genes dominated in the most ancient genes. For the other types of cells, it is almost equal between up‐and downregulated genes in the most ancient genes category.
Therefore, our analyses suggest that aging involves not merely a stochastic decay but a heterogeneous cellular phylogenetic dissociation over time (see Appendix A for the significant differences using a hypergeometric test between cumulative percentage distributions of gene evolutionary ages for aging‐related gene sets compared to the baseline genome distribution).
We tested predictions of the Atavistic Gene Expression Dissociation (AGED) hypothesis during aging using a meta‐phylostratigraphic analysis. We applied it to RNA‐seq and scRNA‐seq data from 8 different studies two meta‐analyses including different tissues: RNA‐seq during aging (Palmer et al. 2021; Tikhonov et al. 2024), scRNA‐seq data from skin cells (Zou et al. 2021), an aging meta‐analysis of the scRNA‐seq of senescent cells (Avelar et al. 2020), and brain (cortex, hippocampus and cerebellum cells) (González‐Velasco et al. 2020), immune (Lu et al. 2022), ovarian (Jin et al. 2024) and stem cells RNA‐seq (Wagner et al. 2009) (see Table 1). We found: (1) An atavistic over‐representation of differential expression in the most ancient genes and under‐representation in the evolutionary youngest genes for two multi‐tissue aging databases, and tissues covering skin, ovarian, immune, senescent and mesenchymal‐senescent cells; (2) No significant atavistic over‐representation of the differential gene expression during aging of brain cells and mesenchymal stem cells; (3) an overall age‐dependent increase of heterogeneity in the direction of the phylogenetic position of tissues' transcriptional profiles; and (4) and an overall negative evolutionary age mean shift toward the most ancient genes. Our analyses suggest that aging involves uncoordinated and tissue‐specific phylogenetic changes in gene expression, revealing a cellular dissociation in phylogenetic space during aging resembling the atavistic theory of cancer for aged cells and some tissue and loss of cellular identity (Izgi et al. 2022; Soto and Sonnenschein 2004). For the last study, they focus on the loss of cellular identity and transcriptomic changes during aging but lack the phylostratigraphic analysis we did in this work.
We also found some genetic heterogeneity in the directions of the genetic changes as sometimes it is more upregulation of the most ancient genes that predominates and sometimes it is the opposite, and some cells seem not affected by these atavistic changes (brain and mesenchymal stem cells).
Numerous properties (biochemical, bioelectrical, and biomechanical) are likely affected by cells' phylogenetic status; we focused on transcriptomic analysis as expressed genes provide the only known effective method to estimate the effective phylogenetic age of a given tissue. However, it will be interesting in the future to see whether for example bioelectrical states can also be used to estimate phylogenetic position (i.e., it's not all about the genes).
To analyze the DEGs in the aging cells under various conditions through phylostratigraphic analysis, we used the evolutionary ages of 19,660 human protein‐coding genes as determined by Litman and Stein (2019). These genes were categorized into 19 major phylostrata, as outlined by Domazet‐Lošo and Tautz (2010). Phylostratigraphic analysis, while providing valuable insights into the evolutionary age of genes, is subject to several limitations that impact the accuracy and consistency of its findings. One challenge is ensuring consistency in ortholog identification across different databases (Litman and Stein 2019). Litman and Stein (2019) found a consensus across several databases but for some genes, uncertainty may remain, leading to the possibility of both false positives and negatives. Another limitation is analyzing noncoding genes. Phylostratigraphic analysis on these genes is more challenging as a significantly lower fraction of noncoding genes can be assigned a reliable evolutionary age (Litman and Stein 2019). This discrepancy may come from the rapid divergence and lineage‐specific nature of many noncoding sequences, which results in a lower likelihood of detecting homologs across distant taxa. The absence of a consistent modal value for many noncoding genes further underscores the challenge of dating their origins, necessitating reliance on statistical estimates such as the median, which may still introduce uncertainties. In this paper, we didn't analyze the evolutionary age of the noncoding genes and focused on the protein‐coding ones. The majority of noncoding genes analyzed in prior studies appear prominently from the Euteleostomi phylostratigraphic age, with the largest expansion observed during the Primata age (Litman and Stein 2019). This suggests that noncoding genes likely play critical roles in lineage‐specific regulatory innovations.
Future research addressing noncoding genes would be valuable, providing deeper insights into their evolutionary trajectories and functional implications in various lineages, particularly within the “Primata” phylostratographic age, where their abundance is highest.
Senescent behavior is not restricted to proliferative arrest. Indeed, another important characteristic of senescent cells is the senescence‐associated secretory phenotype (SASP), where cells secrete inflammatory cytokines, growth factors, and proteases into their environment (Campisi 2013). From an atavistic point of view, cellular senescence may not be just a passive response to molecular damage but an active, evolutionarily conserved process that occurs when the controls of multicellularity fail and cells pursue their own goals (Khosla et al. 2020), and that plays a role in cellular interaction and adaptation to high‐stress conditions (Thomas et al. 2016). However, in the case of aging, this process is dysregulated and leads to a defect of tissue homeostasis and promotes chronic inflammation (Campisi 2013).
Heterochronic parabiosis, the surgical connection of the circulatory systems of young and old organisms, has demonstrated that aging is not irreversible but is instead highly plastic and responsive to systemic factors (Conboy et al. 2005). From the perspective of the atavistic theory of aging, the rejuvenating effects observed in heterochronic parabiosis experiments may not come from the inhibition or reduction of pro‐aging factors along with the introduction of youthful signals, but rather from the reconstitution of the multicellular integration that has been lost and the reduction of cellular phylogenetic dissociation. If aging is a trajectory of increasing cellular atavism, where cells revert to evolutionarily ancient behaviors, then exposure to young systemic factors may act as a re‐establishment of developmental/morphogenetic constraints, pulling aged cells back into the organizational structure of a more coherent, goal‐directed system.
Cancer can be understood as the failure of the computational boundary that regulates organ‐level integration in the body (Levin 2019). The cells lose multicellular integration into networks that pursue large‐scale anatomical setpoints, and function more as autonomous units that treat the rest of the body as external environment, similarly to their ancestral unicellular behaviors (Levin 2021; Bussey et al. 2017; Davies and Lineweaver 2011; Rubin 1985). Therefore, cancer is believed to occur due to failure of the regulatory networks that usually restrict individual‐level goals and favor morphogenetic goals such as bioelectric signaling, resulting in uncontrolled cellular autonomy and proliferation (Levin 2019). This reduction of the radius of cooperation/coordination among cells, from organ‐level to individual cell level, is consistent with the loss of phylogenetic age we observed across tissues with age.
In the atavistic model of cancer, cancer progression is seen as an expression of the genetic to the ancestral cellular states, which are defined by the reactivation of the genetic programs that have been inherited from the unicellular or early multicellular evolutionary stages (Lineweaver and Davies 2021; Thomas et al. 2017). While the somatic mutation theory (SMT) has been used to explain the development of cancer as a result of sequential mutations and natural selection events within the human body (Soto and Sonnenschein 2004), the atavistic model is based on the idea that ancient functionalities that were essential for early and unicellular life forms are involved in the disease. These functionalities are uncontrolled growth, lack of response to growth inhibitors and increased stress tolerance. Phylostratigraphic analysis has been used to establish that many genes associated with cancer were born during critical transitions in the evolution of life on Earth such as the rise of unicellular, eukaryotes or multicellularity (Lineweaver and Davies 2021).
Also, in the active inference framework (Friston 2012; Kirchhoff et al. 2018; Parr et al. 2022), cells are described as having a model of themselves and the outside world (Kuchling et al. 2020; Pio‐Lopez et al. 2022). Maybe evolutionary stage is part of their self‐model, and the transcriptional changes we see are a readout of the change of this model with age.
Atavistic dysregulation during aging suggests new avenues for longevity therapeutics: strategies should focus on restoring and reinforcing multicellular coherence, re‐establishing a multi‐cellular and evolutionary younger genetic expression that preserves tissue and organ integration (perhaps by stimuli that remind cells of their modern evolutionary status). In the context of aging as a loss of morphostatic information and goal‐directedness (Pio‐Lopez and Levin 2024; Pio‐Lopez et al. 2024), cellular dissociation in morphogenetic space suggests that giving evolutionarily recent information for the organism may counteract aging. We suggest this is the beginning of a roadmap that pursues healthspan and longevity by managing the processing, by cells and tissues, of information across a wide range of temporal and spatial scales.
Conceptualized study: L.P.‐L., M.L.; performed bioinformatics analysis: L.P.‐L; interpreted data: L.P.‐L., M.L.; wrote and edited manuscript: L.P.‐L., M.L.
This work was supported by Templeton World Charity Foundation (TWCF0606) and Astonishing Labs.
The Levin lab has a sponsored research agreement with Astonishing Labs, a company seeking to impact the longevity and biomedicine of the aging field.