Bulk and single-nuclei multiomics study of aging in Nothobranchius furzeri

Aging is characterized by a gradual decline in organismal fitness, indirectly measured by biological age. DNA methylation changes have emerged as reliable biomarkers for estimating biological age, leading to the development of predictive models known as epigenetic clocks. Initially created for humans, these clocks have been extended to various mammalian species. Here, I expanded these tools to the killifish N. furzeri, which, with its remarkably short lifespan and expression of canonical aging hallmarks, offers a unique model for experimental aging studies. I developed highly accurate predictive models based on individual CpG sites and co-methylation modules using brain and caudal fin tissues. Applying my models to other datasets revealed that epigenetic aging is accelerated in short-lived strains and could be decelerated by fin regeneration. Additionally, I used longitudinal data to develop both an epigenetic and a proteomic ‘timer’ for direct prediction of individual lifespan based on fin biopsies. The methylation-based predictor showed far superior performance, underscoring the existence of intrinsic determinants of lifespan established early in life. Combining epigenomics, transcriptomics and proteomics, I found a repressive function of DNA methylation on gene expression in early life, which is lost with age, and a progressive decoupling between mRNA and protein levels. These changes may represent causal events in aging and neurodegeneration. Additionally, I investigated the impact of aging on individual cell types. Single-nuclei RNA sequencing of optic tectum and telencephalon in young and old killifish, which display a spontaneous age-associated loss of neuro-regenerative capacity, revealed progenitors of both glial and non-glial (NGP) nature. Specifically, I identified five radial glial (RG) subtypes with neuroepithelial (NE), astroglial or ependymal signatures in different brain locations, consistent with their roles, and one proliferative NGP group. Lineage inference pointed to NE-RG and NGPs as start populations for glio- and neurogenesis. Upon aging, these two populations were the most functionally perturbed, showing a loss of neuro- and gliogenic potential. This integrated single-cell and bulk approaches across multiple biological levels provides a detailed view of aging processes, highlighting new potential biomarkers and mechanisms regulating longevity in N. furzeri, paving the way for future therapeutic interventions.