GeneBites

Scientists use computational models to understand how changes to our DNA can impact aging.

“Who is the fairest of them all?” is a phrase many have heard growing up over retellings of Snow White, a tale in part about a Wicked Queen consumed by jealousy over the young beautiful princess. This tale represents a larger obsession that many have with youth—an obsession that has fueled a $17 billion anti-aging industry to pump out products that keep us looking and feeling young. The anxiety fueling this obsession comes from the fact that we know very little about what causes aging and even less about how to stop it.

One of the most burning questions that scientists who study aging postulate is the cause of aging itself. A popular theory suggests that aging comes from increasingly common genetic mutations in the DNA that prevent affected genes from functioning, eventually resulting in a decline of organ function and eventual death. Studies broadly show that aging is associated with epigenetic changes — chemical modifications to DNA and histones that, in turn, change gene expression and chromatin structure without changing the DNA sequence. While many current models for studying aging consist of animals like fruit flies or birds, it’s unclear if the aging mechanisms in these short-lived model organisms translate to long-lived humans. Consequently, more and more researchers are looking toward large-scale genomic data to study aging.

DNA methylation is one of the most commonly studied epigenetic modifications associated with aging and occurs when methyl groups attach to DNA molecules. Epigenetic aging clocks are machine learning models that can quantify age based on the level of methylation at certain sites. However, these models are correlational and not causational and can’t prove that DNA methylation changes cause aging. This paper used Mendelian randomization (MR), a statistical approach that uses genetics to estimate causality similar to how randomization provides a causal estimate in Randomized Control Trials. It used large-scale genetic data to offer a comprehensive map of methylation sites causal to aging traits.

The study concluded that age-related changes in DNA methylation do not necessarily have positive or negative effects on age-related traits. Instead, they can be adaptive or damaging at the epigenetic level. However, identifying specifically the damaging and protective methylation sites can be useful for understanding and quantifying aging. The causal epigenetic clock models developed for this study distinguish these protective changes from damaging events, providing insights into aging mechanisms and testing interventions that delay aging and reverse biological age.

Using these models, the paper identified several DNA methylation sites that improved longevity. Higher methylation at transcription factor binding site BRD4 (a gene that contributes to cell senescence and promotes inflammation) may mitigate the negative effects of BRD4 and promote healthy longevity. Additionally, the study found that higher methylation of CREB1 (a gene related to type II diabetes and neurodegeneration) may support longevity. They also found that HDAC1, a histone deacetylase that binds to methylated sites, plays a damaging role during aging, as increased DNA methylation may inhibit healthy longevity. 

By constructing computational models, these various findings shed light on how genes and aging interact – and in particular whether certain genes influence aging positively or negatively. By pinpointing specific gene markers linked to longevity and vulnerability, we uncover opportunities to understand aging better and develop interventions to slow it down. This pursuit isn’t just about wishful thinking; it reflects our ongoing quest for health and vitality, hinting at a future where aging isn’t just about getting older but about staying strong and living well.

Edited by Caroline Kisielinski, Jayati Sharma, & JP Flores