Successful aging (SA) is a multidimensional phenotype involving preservation of cognitive ability, physical function, and social engagement throughout life. Multiple components of SA are heritable, supporting a genetic component. The Amish are genetically and socially isolated with homogeneous lifestyles, making them a suitable population for studying the genetics of SA. DNA and measures of SA were collected on 214 cognitively intact Amish individuals over age 80. Individuals were grouped into a 13-generation pedigree using the Anabaptist Genealogy Database. A linkage screen of 5944 single nucleotide polymorphisms (SNPs) was performed using 12 informative subpedigrees with an affected-only 2-point and multipoint linkage analysis. Eleven SNPs produced 2-point LOD scores >2, suggestive of linkage. Multipoint linkage analyses, allowing for heterogeneity, detected significant LOD scores on chromosomes 6 (HLOD = 4.50), 7 (LOD*= 3.11), and 14 (HLOD = 4.17), suggesting multiple new loci underlying SA.
The most rapidly growing segment of the population in the developed world is individuals over age 65 (Vincent & Velkoff, 2010). According to the U.S. Census, the proportion of the population over age 65 will increase from 13% in 2010 to 20.2% in 2050 and the number of persons over age 65 is expected to increase from approximately 40.2 million in 2010 to an estimated 88.5 million in 2050 (Vincent & Velkoff, 2010). The number of persons over age 85 is expected to increase from 5.8 million in 2000 to 19.0 million in 2050. By 2050, the average life span worldwide is predicted to extend another 10 years.
There are multiple reasons why this increase in individuals over the age of 80 has been observed. These reasons include but are not limited to: advances in medical care by which individuals with significant chronic diseases are treated with more effective therapies and survive longer with their illnesses, an increase in awareness of good health habits and dietary nutrition, and finally an increase in those surviving past 80 due to more effective screening and treatment of diseases like hypertension and diabetes. However, increased longevity with significant disability will lead to increased health care costs. Thus, it is important to identify factors that promote not just longevity, but also full function toward the end of life. A recent report from the National Institute on Aging Biology of Aging Summit (Sierra, 2009) reinforced the need to expand phenotypic measures beyond measures of life span by adding measures of health quality and to consider multiple parameters underlying health span in future studies.
Several lines of evidence suggest that successful aging (SA) is influenced by both genetic factors and their interactions with behavioral and environmental influences. One key component of SA, increased longevity, is heritable (Carmelli, 1982; Swedlund et al., 1983; Farrer et al., 1984). In a study of the Lancaster County Pennsylvania Amish, age at death was estimated to have 25% heritability (Mitchell et al., 2001). Similarly, studies of extreme longevity in samples of centenarians have determined that siblings of centenarians live longer than siblings of individuals who died in their early seventies (Perls et al., 1998). Recent studies of the concordance rates in identical (MZ) and fraternal (DZ) twin pairs have suggested that longevity has about 20–25% heritability with little evidence for shared environmental factors (Herskind et al., 1996; Gurland et al., 2004).
In addition to chronological age, other twin studies have demonstrated that measures of physical function (another domain of SA) are also heritable, perhaps more strongly than age at death or longevity. These studies have shown that several indicators of upper and lower extremity function are about 50% heritable: hand-grip strength (40–50%) (Carmelli & Reed, 2000; Frederiksen et al., 2002; Tiainen et al., 2004), lower extremity function (57%) (Carmelli et al., 2000), balance performance (46%) (El Haber et al., 2006). These results suggest that using these measures to define a composite phenotype, “successful aging,” might facilitate identification of SA genes.
The value of conducting genetic studies in isolated populations such as the Amish has long been recognized (Lander & Botstein, 1987). This value derives from the recent expansion of the population (since the 1820s) from a small number of founders and isolation that restricts the introduction of additional genetic variation (Hastbacka et al., 1992). These factors often lead to large and stable pedigrees for genetic linkage analysis, while the increased level of linkage disequilibrium around trait-associated alleles potentially enhances detection of allelic association. In this report, we conducted a single nucleotide polymorphisms (SNP)-based genome-wide linkage screen (GWLS) in a sample of older Amish individuals to examine linkage and association with a composite SA phenotype.
Materials and Methods
This study is a part of the larger ongoing Collaborative Aging and Memory Project (CAMP), a multi-institutional prospective population-based study of aging and dementia in the Amish communities of Adams, Elkhart, and LaGrange Counties in Indiana and Holmes County in Ohio, conducted from 2002 to present. These communities were formed in several waves of migration in the 18th and 19th centuries (van der Walt et al., 2005). The Amish originally emigrated from Europe to Pennsylvania in the 1700s and a further westward expansion of the Amish population occurred in the early 1800s, when a subset of the Pennsylvania population migrated to Ohio and Indiana. A second wave of migration from Switzerland arrived in the 19th century, eventually settling in Adams County, Indiana, in the 1850s. Present-day Amish in Holmes County, Ohio and Elkhart and LaGrange Counties, Indiana, are largely descended from the first wave of westward immigration from Pennsylvania, while the Adams County, Indiana settlements are largely descended from the second wave of immigration that passed through Pennsylvania (van der Walt et al., 2005). Both sets of communities, therefore, have a degree of shared history and ancestry with the Pennsylvania Amish communities. Written and informed consent was obtained for all participants or their legal guardians.
For the SA arm of the CAMP study, individuals over age 80 were identified through public directories published by individual Amish communities and referral from individuals already enrolled in the study. Once individuals were identified a door-to-door interview was performed to obtain a baseline examination. The only exclusion criterion for the SA arm of the study was cognitive impairment (individuals screening cognitively impaired were referred to the dementia arm of the study). The experiments reported here were conducted in the sample obtained in the first three and half years of the CAMP study (January 2002–July 2006). Of the 249 cognitively intact individuals over age 80 contacted for the study by July 2006, 233 agreed to participate in the baseline examination (94% participation rate). Of these, 214 of these agreed to provide a blood sample (92% participation in the genetic study among those cognitively intact). The recruitment and ascertainment methods for CAMP have been previously described (Pericak-Vance et al., 1996; Ashley-Koch et al., 2005).
Using an “all common paths” query of the Anabaptist Genealogy Database (Agarwala et al., 2003), all sampled individuals were placed in a 13-generation, 4998 person pedigree with a single common ancestral couple for 46 out of the 48 SA individuals in the earliest generation of the pedigree. There is no evidence of a more recent founder effect among the 48 SA individuals. The first individuals in this pedigree born in Indiana or Ohio were born in the 1820s, 4–5 generations before the oldest sampled individuals in this study.
Definition of SA
SA was defined according to tests and measurements taken at the time of baseline enrollment in the study. SA was defined based on function in all three domains described by Rowe and Kahn (Rowe & Kahn, 1997). This trait was created in order to define a composite trait describing SA rather than examining individual components that are used to define SA; creating a composite trait also allows for the data to be analyzed with binary-trait genetic linkage and association analyses. The specific criteria we used are outlined in Table 1. The first requirement was survival to age 80. All individuals had to be cognitively intact [education-adjusted modified mini-mental state examination (3MS) >86]. SA individuals did not have significant depressive symptoms [geriatric depression scale (GDS) score <6]. Next, we considered whether individuals met standard cutoffs for “high function” on the self-reported measures of physical function: total scores of 0 or 1 on the activities of daily living (ADL) and instrumental ADL (IADL) scales, indicating no assistance or partial assistance on only a single item needed; Nagi score of 0 or 1 indicating no difficulty or difficulty on only one item; Rosow–Breslau score of 3 or 4, indicating limitation on zero or just one item. Lower extremity function was considered by limiting SA to individuals scoring in the top one-third of the sample on the Established Populations for Epidemiologic Studies of the Elderly (EPESE) short physical performance battery summary score (>8 on a 12-point scale). Finally, individuals considered SA were satisfied or very satisfied with life. Of the 214 cognitively intact individuals over age 80 (40% male, mean age 83), 48 met criteria for SA (69% male, mean age 83) (Table 2). The remaining 166 individuals over age 80 (31.3% male and 83.8 mean age, Table 2) were cognitively intact but impaired on at least one other domain of SA.
Table 1. Criteria to identify “successfully aged” individuals
3MS >86 (education adjusted)
No significant depressive symptoms
“High function” self-reported measures of physical function
0 or 1 ADL and IADL scales
0 or 1 Nagi score
3 or 4 Rosow–Breslau score
Lower extremity function
>8 (12-point scale) EPESE
Satisfied or very satisfied with life
Table 2. Demographic and clinical characteristics of study population A.
SA (N= 48)
Non-SA comparison group (N= 166)
*All individuals with 3MS <86 were determined to be cognitively normal in detailed neuropsychological evaluation, and are thus included in the sample.
Gender (% female)
Age (mean (SD1))
0.0% Adams; 43.8% Elkhart; 56.2% Holmes
18.1% Adams; 53.6% Elkhart; 28.3% Holmes
3MS (median [range])
GDS (median [range])
ADL (median [range])
IADL (median [range])
Nagi (median [range])
Rosow–Breslau (median [range])
EPESE (median [range])
Grip strength (dominant hand) (median [range])
Social interaction (median [range])
Social network (median [IQR])
Instrumental support (median [range])
100% very satisfied or satisfied
95.8% very satisfied or satisfied
Table B. .
1Total number of individuals (sampled and unsampled) in the pedigree.
Collection of clinical and demographic data
The baseline examination included collection of demographic characteristics and a screening of cognitive and physical function.
Participants underwent cognitive screening using a revision of the 3MS examination (Teng & Chui, 1987), developed for use in the Cache County Memory Study (Tschanz et al., 2002). We established the 3MS cutoff point of <87 (after education adjustment) indicating potential cognitive impairment after examining the Receiver Operating Characteristic curves in predicting subsequent dementia and prodromal Alzheimer disease in the Cache County Memory Study. This cut point was selected as having maximal utility in detection of dementia (Khachaturian et al., 2000), optimizing the sensitivity of case detection (98.4%). Depressive symptoms were screened using the 15-item GDS (Yesavage, 1988), and life satisfaction was probed using a single question “Overall how satisfying is your life: very satisfying, satisfying, not so satisfying?”
Self-rated physical function was assessed by several well-established instruments. Basic self-care was assessed using a modified Katz ADL scale consisting of eight items: bathing, dressing, grooming, toileting, continence, transferring, walking, and eating (Katz, 1983). Difficulty with musculoskeletal function was assessed by a five-item scale based on the work of Nagi (1976) used in the EPESE (Seeman et al., 1994): pushing large objects, stooping or kneeling, carrying weights over 10 pounds, extending arms above shoulder level, and writing. Advanced daily living tasks were evaluated by a three-item modified Rosow–Breslau scale (Rosow & Breslau, 1966) assessing difficulty performing heavy work around the house or farm, walking up and down a flight of stairs, and walking a half-mile without help. A final scale assessing more integrative functions, IADL, consisted of six of the eight items suggested by Lawton & Brody (1969): traveling, shopping, preparing meals, doing housework, managing medications, and handling money.
In addition to these self-reported measures of functioning, a short physical performance test developed for use in the EPESE was used to assess lower extremity function (Seeman et al., 1994). The tasks include three measures of balance (side by side, semitandem, and tandem) for 10 s, the time required to sit and stand from a chair five times without using the hands and the time required to walk 10 feet.
In conjunction with the CAMP dementia study, we conducted an SNP-based GWLS of 672 Amish individuals using the 6008-SNP Illumina Human Linkage Panel IVb. After quality control checks [removing SNPs not polymorphic in the sample, with poorly clustered genotyping, or significant departures (P < 0.001) from Hardy–Weinberg equilibrium in the 125 most unrelated individuals], genotypes from 5645 markers were available for analysis. SNP allele frequencies for use in linkage analysis were estimated from all 672 individuals. This analysis focused on the 214 cognitively intact individuals over age 80 enrolled in the SA arm of the CAMP study. In addition to the 214 individuals over age 80, 11 full siblings younger than 80 were used (without phenotypes) to establish linkage phase as described below.
We used the software program PedCut (Liu et al., 2008) to cluster all 48 SA cases and 28 full siblings (17 over age 80, 11 younger) into 12 subpedigrees informative for linkage (e.g., at least two SA individuals in each pedigree). Pedigrees were specified to be no more than 24 “bits,” and were thus small enough to allow Merlin to generate exact 2-point and multipoint LOD scores using the Lander–Green algorithm (Abecasis et al., 2002). A detailed description of the subpedigrees (including number of individuals, number of genotyped SA cases, and number of genotyped non-SA siblings) is provided in Table S1. The subpedigrees ranged in size from 17 to 45 individuals. The pedigrees were largely independent; only two pedigrees shared a total of six individuals—all untyped, connecting ancestors. Multipoint LOD scores allowing for heterogeneity (HLOD) across all pedigrees were calculated using affected-only dominant (trait allele frequency = 0.001) and recessive (trait allele frequency = 0.2) models. Multipoint nonparametric allele-sharing (LOD*) models were also generated. Marker allele frequencies were estimated from the data by the Merlin analysis software. Intermarker distances were determined using the physical and genetic maps provided by Illumina for the Linkage IV panel. The non-SA siblings included in the study were only used to establish linkage phase in the subpedigrees (e.g., to assist in determining sharing of alleles identical-by-descent in SA individuals) but do not contribute to the linkage statistic in the affecteds-only model. Therefore, linkage analysis results only reflect allele sharing between the SA individuals in each subpedigree.
A summary of the clinical and demographic data for our study population is provided in Table 2A for all of the variables examined, stratified by SA status. The SA and non-SA comparison groups have similar mean ages, but more men are SA and a higher proportion of the non-SA comparison group than SA are from Indiana (Table 2A). A summary of the 12 subpedigrees created is provided in Table 2B. All pedigrees represent individuals from Elkhart and Holmes Counties, as only non-SA individuals were available from Adams County. On average, there were four SA individuals per pedigree and two sibs.
Examination of the 2-point LOD scores showed that 11 SNPs produced suggestive evidence (LOD >2) for linkage (Table 3), although none of these SNPs had significant LOD scores >3. Two chromosomes generated multipoint LOD scores allowing for heterogeneity (HLOD) >3 under the dominant HLOD model, and one chromosome generated a nonparametric LOD* >3 (Fig. 1): chromosome 6 with a dominant model [maximum HLOD = 4.49 at rs1409014 (54 Mb) based on human genome build 36/hg18], which had a total of 143 genotyped markers under the linkage peak (Fig. 2A), 1-LOD-down support interval 52–65 Mb, chromosome 7 with a nonparametric model [maximum LOD*= 3.11 at rs517258 (64.18 Mb), 1-LOD-down support interval: 49–75 Mb](Fig. 3C), which included a total of 22 genotyped markers under the linkage peak, and chromosome 14 with a nonparametric model [maximum HLOD = 3.28 and maximum LOD*= 4.17 at rs764602 (48 Mb), 1-LOD-down support interval: 36–56 Mb, based on HLOD; 42–53 Mb based on LOD*], which had a total of 53 genotyped markers under the linkage peak (Fig. 4).
Table 3. Two-point LOD scores >2 for dominant, recessive, and nonparametric models
2-pt LOD score1
1LOD scores are presented for dominant (Dom), recessive (Rec), and nonparametric (Non) 2-point linkage analyses.
These results would indicate that additional fine mapping is necessary to identify high-priority candidate genes for further study.
We performed a GWLS looking for chromosome regions linked to SA in the Amish and observed strong evidence for linkage on chromosomes 6, 7, and 14. The regions on chromosomes 6 and 14 are novel, while the region on chromosome 7 is 6 Mb away from a SNP (rs2371208 position at 81,982,510 base pairs) associated with age at death in the Framingham Study (Lunetta et al., 2007). Further examination of the literature demonstrated that these three linkage peaks are in the vicinity of several potential longevity and aging candidate genes including bone morphogenic protein (BMP) 5 located on 6p12.1, and BMP4 located on chromosome 14q22-q23. These regions will be further examined through fine mapping to determine if any of the linkage peaks can be explained by an association with these candidate genes.
Chromosome 6 notably had a minor linkage peak covering the fragile site (FRA6E) on chromosome 6 that includes the parkin gene (PARK2). The linkage peak had an LOD score ranging between 1.10 and 2.09 within the 161.68 to 170.62 Mb region of the chromosome with a recessive model. PARK2 is a longevity and aging candidate gene and has been associated with several aging-related phenotypes including Parkinson disease, general neuron degeneration, and several forms of cancer. PARK2 has been described as a tumor suppressor gene, consistent with the theory that preserved DNA repair and tumor suppression activity are essential mechanisms for cancer-free longevity and SA (Matheu et al., 2009; Veeriah et al., 2010).
In our study, we observed a higher rate of SA in males relative to females. These results have not been observed consistently in previous literature, and contrast with the prior study of longevity in the Pennsylvania Amish, which reported no effect of sex on age at death or heritability of lifespan (Mitchell et al., 2001). Consistent with our results are studies by Seeman et al. (1994) who reported in the MacArthur Studies of Successful Aging that males were more likely to have better physical performance at baseline in a cohort of 70–79 year olds with high function (Seeman et al., 1994). This would suggest that our observations may reflect that at baseline SA males had better physical function. However, the studies by McLaughlin et al. (2010) of adults aged 65 and older did not observe an inverse association between SA and gender (McLaughlin et al., 2010). One explanation for these differences could be the specific definition of SA used. We attempted to use as comprehensive a definition as possible, integrating factors pertaining to physical and cognitive function, social engagement, and health whereas other studies would more often use only one or two of these components. Another possible explanation for the larger number of SA males relative to females could be the effect of marital status on SA (Li et al., 2006). In our sample, a greater proportion of men were successfully aged (52% of men vs. 17% of women) and currently married (78% of men vs. 45% of women). While more women were widowed (48%) than men (22%), all men in the sample were either currently or previously married, while 7% of the women were never married, However, the relationship between gender, marriage, and SA has not been consistent across studies, with the majority finding no association (Depp & Jeste, 2006). Further work is necessary to fully understand the significance of these findings.
Although these findings are interesting, the SA linkage regions detected are broad with large numbers of candidate genes within each. The regions identified will be further examined through fine mapping to determine if any of the linkage peaks can be explained by segregation of a functional variant in any biologically relevant genes with SA in a larger sample of Amish that includes these pedigrees.
We acknowledge the help of the Vanderbilt CHGR DNA Resources core for processing some of the samples. Some of the samples used in this study were collected while WKS, JRG, and MAP-V were faculty members at Duke University. This work was supported by NIH grants AG019085 to Jonathan L. Haines and AG019726 to William K. Scott.