The cost of obtaining transcriptomic and proteomic data, and then using machine learning techniques to develop insights based on that data, has fallen dramatically over the past decade. As a result there is a proliferation of signatures of aging and longevity, as many different research groups analyze many different large transcriptomic and proteomic databases. The example here is one of a number of such signatures created with the idea of finding potential targets for therapy. It is far from clear that one can alter any of the various protein levels related to aging and longevity and obtain meaningful benefits, however. A change can be a side-effect of aging, and end-stage consequence that causes few downstream consequences in and of itself, and will achieve little if reversed.
The identification of protein targets that exhibit anti-aging clinical potential could inform interventions to lengthen the human health span. Most previous proteomics research has been focused on chronological age instead of longevity. We leveraged two large population-based prospective cohorts with long follow-ups to evaluate the proteomic signature of longevity defined by survival to 90 years of age. Plasma proteomics was measured using a SOMAscan assay in 3,067 participants from the Cardiovascular Health Study (CHS) and 4,690 participants from the Age Gene/Environment Susceptibility-Reykjavik Study (AGES-Reykjavik). Logistic regression identified 211 significant proteins in the CHS cohort using a Bonferroni-adjusted threshold, of which 168 were available in the AGES-Reykjavik replication cohort and 105 were replicated.
The strongest associations in CHS that were replicated in AGES-Reykjavik were for GDF-15, NT-pro-BNP, b2-microglobulin, RNase 1, and HE4, providing confidence in such previously identified proteins in aging research. Less-established markers of mortality in the general population, such as angiopoietin-2, and PXDN, also had support in both cohorts. Our study design leveraging a longevity outcome, as opposed to overall survival only, paired with long follow-up time revealed that nearly half (269 out of 471) of proteins associated with overall survival were not associated with exceptional longevity in the CHS, though the strongest associations remained consistent between the two outcomes.
A larger share of significant proteins was associated with both overall survival and longevity in AGES-Reykjavik, which may have occurred due to increased power to detect significant associations in AGES-Reykjavik. This observation suggests that extrapolating findings from associations with overall survival to longevity might be inappropriate. Moreover, we demonstrate for the first time in proteomics studies of longevity that physical and cognitive function may partially mediate associations between proteins and longevity, and that the amount of mediation may depend in part on which particular functional measures are used in the analysis.
Link: https://doi.org/10.1111/acel.14136
View the full article at FightAging
