DRD4 genotype predicts longevity in mouse and human

Abstract

Longevity is influenced by genetic and environmental factors. The brain's dopamine system may be particularly relevant, since it modulates traits (e.g., sensitivity to reward, incentive motivation, sustained effort) that impact behavioral responses to the environment. In particular, the dopamine D4 receptor (DRD4) has been shown to moderate the impact of environments on behavior and health. We tested the hypothesis that the DRD4 gene influences longevity and that its impact is mediated through environmental effects. Surviving participants of a 30-year-old population-based health survey (N = 310; age range, 90-109 years; the 90+ Study) were genotyped/resequenced at the DRD4 gene and compared with a European ancestry-matched younger population (N = 2902; age range, 7-45 years). We found that the oldest-old population had a 66% increase in individuals carrying the DRD4 7R allele relative to the younger sample (p = 3.5 × 10(-9)), and that this genotype was strongly correlated with increased levels of physical activity. Consistent with these results, DRD4 knock-out mice, when compared with wild-type and heterozygous mice, displayed a 7-9.7% decrease in lifespan, reduced spontaneous locomotor activity, and no lifespan increase when reared in an enriched environment. These results support the hypothesis that DRD4 gene variants contribute to longevity in humans and in mice, and suggest that this effect is mediated by shaping behavioral responses to the environment.

Figure 1.

Survival curves of the oldest-old cohort. The known male (blue) and female (pink) survival curves of European ancestry individuals born in 1910 (the average birth year of participants) are shown (source: Berkeley Mortality Database, http://www.demog.berkeley.edu/∼bmd/; UK and USA data). Individuals older than 90 years of age represent 1.4% (male) to 2.9% (female) of their birth cohort. The projected curve of the 1995 birth cohort (orange) suggests that 17% of this cohort (males and females) will reach 90+ years of age (Social Security Administration, Alternative II Forecast, 1998, http://www.demog.berkeley.edu/∼bmd/). Inset, Diagrammatic representation of the human DRD4 gene region. Exon positions are indicated by blocks (yellow = noncoding; orange = coding), and the positions of alu repetitive sequences are represented by pointed blue blocks. The position of a coding 48 bp VNTR in exon 3 is indicated by a green triangle. The 2R to 11R variants of this repeat are indicated below exon 3, along with their frequencies in European ancestry populations (Ding et al., 2002; Grady et al., 2003; Wang et al., 2004; Grady et al., 2005; this study).

Figure 2.

Comparison of expected to observed DRD4 genotypes. Oldest-old individuals (N = 310) were genotyped at the exon 3 DRD4 polymorphism (Fig. 1, inset) and compared with population-based European ancestry control sample DRD4 genotypes (Expected, N = 2902). The DRD4 7R/x category (black bars) contains individuals with at least one 7R (or derived 7R, i.e., 5R, 6R, 8R, 9R, 10R, and 11R) allele, and the non-7R/x category (striped bars) contains the remaining individuals. The expected and observed fraction for each category is shown. Expected 7R/x (N = 635), Non 7R/x (N = 2267); 90+ observed 7R/x (N = 113), Non 7R/x (N = 197); Female 7R/x (N = 86), Non 7R/x (N = 133); Male 7R/x (N = 27), Non 7R/x (N = 64). Asterisks denote observed differences significant at p = 3.5 × 10⁻⁹ (90+ years), p = 1.7 × 10⁻⁹ (Female), and p = 0.034 (Male) (see text).

Figure 3.

DRD4 genotype and self-reported activity. The number of hours per day of activity reported in 1981 is plotted versus participants with a DRD4 7R/x genotype (black bars) or non-DRD4 7R/x genotype (striped bars). Comparisons are shown at the 5th, 25th, 50th, 75th, and 95th percentiles. Including DRD4 2R/x and 3R/x individuals with the DRD4 7R/x individuals instead of the non-7R/x individuals (Ding et al., 2002; Grady et al., 2003; Wang et al., 2004; Grady et al., 2005) gave comparable activity level differences (i.e., an ancestral 4R/4R vs non-4R/4R division).

Figure 4.

Kaplan–Meier survival curves of mice as a function of genotype and exposure to an EE or DE. Insert, Lifespan per group (mean and SE). WT and HT mice reared in an EE lived significantly longer than when reared in a DE. In contrast, survival of DRD4 KO mice was not increased when reared in an EE compared with a DE and was shorter than that of WT and HT mice.

Figure 5.

Spontaneous locomotor activity for different ages in WT, HT, and KO mice reared in a DE and an EE. Values correspond to means (±SE) and correspond to average activity over 90 min. The genotype-by-age interaction on locomotor activity was significant (F = 3.61, p < 0.001, η_p² = 0.034) with overall significantly lower activity levels for KO mice than for WT or HT mice (p < 0.05). The interaction of environment by age with locomotor activity was also significant (F = 2.77, p = 0.017, η_p² = 0.013); overall, DE mice were significantly more active than EE mice (p < 0.05). However, since animals in an EE decrease in spontaneous locomotor activity (Garland et al., 2011) and we did not record 24 h activity levels, we cannot compare activity levels across the two environmental conditions.