Changing Aging — 3

Genome Instability

From the genome variants that are associated with longevity we can learn what genes and pathways are implicated in the aging process. That’s especially true when it comes to rare variants present in less than 1 percent of the population, and with a high odds ratio or magnitude of effect, that track with human aging. They are found by genome sequencing, either by the exome (the roughly twenty thousand protein coding genes) or the whole genome of three billion letters. Common variants, occurring in more than 1 percent of the population, can be detected through genome chips (known as arrays). These chips have identified more than a hundred letter changes (single nucleotide variants) associated with lifespan, which can be aggregated to provide a polygenic risk score. Combining all the rare and common variants linked to longevity gives us a sense of which traits can be passed from parent to offspring. The estimates for heritability of longevity vary considerably: 12 percent in one of the most rigorous assessments in thirteen million people and well below that in others. That context is important, telling us that genetics is important in the inherited trait of aging, but its contribution is limited.

On the other hand, variations in genes consistently identified through sequencing studies provide insights into the underpinning of human aging. The same genes (with their principal function in parentheses) keep popping up from studies of people with extreme longevity or populations studies: APOE (lipid metabolism); FOXO3 (nutrient sensing); IGF-1R (insulin growth factor-1 receptor); CETP (lipid metabolism); CDKN2A/2B (cell cycle and senescence); certain HLA types (immune system); BRCA1 and BRCA2 (DNA repair genes); ATM (DNA repair); TET2 (telomeres); SH2B3 (inflammation); CHRNA3, CHRA5, CELSR2, and PSRC1 (intercellular communication); NF–kB (inflammation); and USP42, TMTC2, and CLU (protein housekeeping). Those are all nuclear DNA variants. There are also many in mitochondrial DNA that are associated with human lifespan, such as the mechanistic target of rapamycin (mTOR), CPS1, MFN2, and LRPPRC that are involved with mitochondrial function and metabolism.

All that’s interesting, but what’s actionable? An important exome sequencing study in nearly fifty-eight thousand Icelanders brought these investigations to a practical realm. This unique study tied the genomic data—from just fifty-three actionable genes—to the country’s death registry. Four percent carried a disease-causing mutation that changed their lifespan, such as the LDL receptor (6.5 years reduced lifespan), and various cancer mutations (BRCA1, BRCA2, MSH2) were associated with three years of reduced lifespan. The main point is that genomic data at the individual level for one in twenty-five people provided actionable information affecting their lifespan. That 4 percent level of finding important disease-causing mutations has been replicated in several other studies. As pointed out in chapter 6: