Assessment of the contribution of APOE gene variants to metabolic phenotypes associated with familial longevity at middle age

Abstract

Offspring of long-lived families are characterized by beneficial metabolic phenotypes in glucose and lipid metabolism and low 25-hydroxyvitamin D. Although the genetic basis for human longevity remains largely unclear, the contribution of variation at the APOE locus has been repeatedly demonstrated. We aimed to assess whether ApoE isoforms mark the familial longevity status in middle age and subsequently to test to what extend this association is mediated by the metabolic characteristics marking this status. From the Leiden Longevity Study (LLS), we included offspring from nonagenarian siblings and partners as controls. Using the metabolic phenotypes of familial longevity as mediators, we investigated how APOE gene variants associated with LLS offspring/control status (in 1,515 LLS offspring and 715 controls). Within the LLS (mean age = 59.2 years), ApoE ε4 was not associated with a lower likelihood of being an LLS offspring, whereas ApoE ɛ2 was significantly associated with a higher likelihood of being an LLS offspring (odds ratio = 1.43), but this difference was not mediated (p-values>0.05) by any of the investigated metabolic phenotypes (e.g., diabetes and glucose). Therefore, variation at the APOE locus may not influence familial longevity status in middle age significantly through any of the metabolic mechanisms investigated.

Keywords: 25-hydroxyvitamin D; APOE; glucose; human longevity; mediation.

Figure 1. Graphical presentation of the biological routes through which APOE could affect familial longevity

“Intermediate phenotypes” denotes the metabolic differences between offspring and controls in phenotype (e.g., lower level of serum glucose, lower level of serum 25-hydroxyvitamin D, lower frequency of type 2 diabetes, in LLS offspring compared with controls). In this study, we aim to assess whether this route is plausible.

Figure 2. Association between ApoE isoforms and propensity to be an LLS offspring

Carriers of the ApoE ε3 isoform were used as reference. Analyses were adjusted for age and sex, and corrected for familial relationships using robust standard errors. A total of 300 participants was ApoE ɛ2 carrier, 1369 participants carried ApoE ɛ3/ɛ3, and 499 participants carried ApoE ɛ4. For these analyses, 62 participants carrying the ε2/ε4 isoform were excluded.