Do people who live into their late 90s and beyond simply carry no pathogenic gene variants, or are they blessed with variants that stave off age-related disease? It's a little bit of both, but more of the latter. That's according to researchers led by Zhengdong Zhang, Albert Einstein College of Medicine, New York, in the September 13 Nature Aging.

  • Rare protective variants in insulin, AMPK signaling appear to protect against chronic age-related disease.
  • Centenarians are less likely to develop AD, diabetes, heart disease.
  • APOE4 carriers with variants in Wnt signaling genes live longer.

The 2,900 oldest old in this study had only slightly fewer common pathogenic variants known from genome-wide association studies (GWAS). However, they boasted more protective variants in genes involved in insulin and AMP-activating protein kinase (AMPK) signaling. In APOE4 carriers, functional variants in the Wnt signaling pathway extended life, suggesting they may protect against the deleterious mechanistic consequences of having ApoE4. A deeper understanding of the genetics of aging informs studies on age-related diseases and the design of anti-aging drugs.

Gene variants within the insulin/insulin-like growth factor 1 (IGF1) and AMPK signaling pathways are well known to make worms, flies, and mice live longer (Zhang et al., 2020; Dec 2009 news). In people, reduced IGF1 signaling contributes to longevity, and multiple GWAS have linked the insulin signaling gene FOXO3 with lifespan (Willcox et al., 2008; Morris et al., 2015). Could rare variants in other animal model longevity pathways influence human lifespan?

To find out, co-first author Jhih-Rong Lin studied two cohorts of Ashkenazi Jewish participants in a longevity study based at the Albert Einstein College of Medicine. Lin gave banked blood samples from 515 centenarians and 496 controls aged 70 to 95 to the other co-first author, Patrick Sin-Chan, Regeneron Pharmaceuticals, New York. Sin-Chan ran whole-exome sequencing to capture variants in coding regions.

Within 17,561 genes, the scientists found 126,405 single-nucleotide polymorphisms (SNPs) and 3,892 insertions/deletions. Centenarians and controls had the same number of variants overall, and scientists found no single variant significantly tied to longevity. But when they grouped variants within the same gene, or within genes of the same pathway, they saw that centenarians carried more variants in genes of the insulin/insulin receptor and AMPK signaling pathways (see image below).

Protective Pathways.Using two different statistical tests (circles, triangles), geneticists compared variants in established aging-associated pathways (left columns) in centenarians and controls. Rare variants in genes involved with insulin and AMPK signaling (black triangles) were tied to extreme old age. [Courtesy of Lin et al., Nature Aging, 2021.]

Scrounging up more power for their study of rare variants in what are, by definition, rare populations, the scientists used three other samples: a German cohort of 1,265 centenarians and 4,195 middle-aged controls, a cognitively normal subset of Alzheimer’s Disease Sequencing Project (ADSP) participants comprising 1,121 centenarians and 38 people under 75, and a UK Biobank proxy cohort of 104 adult children of centenarians and 23,405 adult children of non-centenarians. In all three cohorts, centenarians carried more insulin-signaling variants; those in the UK Biobank also had more variants in AMPK signaling.

Were the rare variants in this study friend or foe? Among the Ashkenazi, participants carried a similar number of pathogenic variants as controls, without affecting lifespan. A recent study in the same cohort also found similar prevalence of pathogenic coding variants among participants regardless of age (Gutman et al., 2020). The authors concluded that some variants in these oldest old must be protective.

To learn if that is the case, Lin, Sin-Chan, and colleagues first assessed how common GWAS variants factored in by calculating each volunteer’s polygenic risk scores for Alzheimer's disease, Type 2 diabetes, coronary artery disease (CAD), stroke, and three types of cancer. Unsurprisingly, risk of either AD, diabetes, or CAD strongly influenced longevity, with centenarians being less prone to all three. “Extreme longevity can largely be explained by reduced risk for complex age-related diseases,” Rudolph Tanzi and Dmitry Prokopenko, Massachusetts General Hospital, Boston, wrote to Alzforum (see comment below).

But how does this relate to rare variants identified in this study? Insulin resistance increases risk for diabetes, CAD, and AD (as well as all-cause dementia). People with higher polygenic risk scores for these three diseases died younger if they also carried rare pathogenic coding variants found in their whole-exome sequence. “The effects of common diseases related to insulin dysregulation can be intensified by cumulative rare variant burden,” Suzanne Craft, Wake Forest School of Medicine, Winston Salem, North Carolina, wrote to Alzforum (full comment below).

Consider APOE4. This allele increases risk of AD and cardiovascular problems; as a group, its carriers also die younger (Deelen et al., 2019). In a given person, however, the pathogenic downstream effects of ApoE can worsen, or soften, under the influence of rare variants. On average in this study, APOE4 carriers died 1.5 years younger than noncarriers; if they also carried rare pathogenic coding variants, their lifespan shrank by a whopping 17 years.

Wnt Versus ApoE. ApoE4 carriers (orange) who inherited few rare variants in the Wnt signaling pathway (middle) live about a decade less than noncarriers (blue). Carriers with those Wnt variants (right) live nearly as long as noncarriers. [Courtesy of Lin et al., Nature Aging, 2021.]

In contrast, insulin and AMPK signaling variants extended longevity of APOE4 noncarriers. For carriers, their lifespans were strongly influenced by 152 rare, functional variants in Wnt signaling genes. APOE4 carriers with many such Wnt variants lived nine years longer than did APOE4 carriers with few (see image below). The scientists saw the same in the U.K., though not German, cohorts. Interestingly, ADSP volunteers who lack these possibly protective Wnt variants were more likely to have AD.

Zhang told Alzforum that this study is part one of a large project funded by the National Institute on Aging to develop anti-aging drugs (2017 press release). “Showing that insulin and AMPK signaling influence aging similarly in humans and animals confirms the value of model organisms for studying aging and drugs for age-related diseases,” Zhang said. That said, not all of the longevity pathways known from animal models showed up in this human study. mTOR, for example, was a notable no-show.

Clinical trials of diabetes-related therapies for AD have largely flopped thus far (Nov 2018 conference news).—Chelsea Weidman Burke


  1. Lin et al. have conducted a rigorous evaluation of three independent cohorts (German, UK Biobank, and ADSP longevity cohorts) to investigate key factors in longevity and resilience to age-related diseases including Alzheimer’s. Their results indicate significant longevity associations of functional rare variants in insulin signaling in all three cohorts, and confirm in humans the results of numerous studies using rodent or invertebrate model systems.

    AMPK also showed similar associations in two of three cohorts, but interestingly no associations were observed for mTOR signaling. Further, lifespan was moderated by an interaction between reduced pathogenic rare variant counts and common polygenic disease risk for AD and Type 2 diabetes. This shows that the effects of common diseases related to dysregulation of insulin can be intensified by cumulative rare variant burden. Other findings of interest include the association of Wnt signaling and APOE4 status.

    The findings provide new insights into the architecture of longevity, and a rationale for ongoing therapeutic efforts targeting insulin signaling to enhance longevity and healthspan, and reduce risk of AD.

  2. This is a carefully prepared study of a homogenous cohort of Ashkenazi Jewish centenarians and controls from the same households. The authors used whole-exome sequencing to study rare variant associations (and enrichment) with longevity. While single variant and region-based tests did not reach the Bonferroni-corrected significance level, they found an enrichment of rare coding variants in centenarians vs controls across insulin and AMPK signaling pathways.

    Particularly interesting was the identified interaction between the rare variant allele count in Wnt signaling and APOE4 status. Those variants showed a positive correlation with lifespan among APOE4+ non-centenarians and a general protective effect of rare exonic variants involved in the Wnt pathway.

    In agreement with these new findings, we have previously reported rare variants in the WNT signaling pathway gene, APC,  to be associated with AD (Prokopenko et al., 2020). 

    It is important to note that in their PRS analysis, the authors also show that centenarians have reduced genetic susceptibility to complex diseases such as AD, CAD, and T2D, showing that extreme longevity can largely be explained by reduced risk for multiple complex age-related diseases.


    . Whole-genome sequencing reveals new Alzheimer's disease-associated rare variants in loci related to synaptic function and neuronal development. medRxiv. 2020 Nov 4; PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Long Life With Tight Plaques—Repressing IGF-1 Protects AD Mice
  2. Trials of Diabetes-Related Therapies: Mainly a Bust

Paper Citations

  1. . Genetics of extreme human longevity to guide drug discovery for healthy ageing. Nat Metab. 2020 Aug;2(8):663-672. Epub 2020 Jul 27 PubMed.
  2. . FOXO3: A Major Gene for Human Longevity--A Mini-Review. Gerontology. 2015;61(6):515-25. Epub 2015 Mar 28 PubMed.
  3. . Similar burden of pathogenic coding variants in exceptionally long-lived individuals and individuals without exceptional longevity. Aging Cell. 2020 Oct;19(10):e13216. Epub 2020 Aug 29 PubMed.

External Citations

  1. 2017 press release

Further Reading

Primary Papers

  1. . Rare genetic coding variants associated with human longevity and protection against age-related diseases. Nature Aging, 1, 2021, pp.783–94. Nat Aging.