SEARCH

SEARCH BY CITATION

Keywords:

  • APOE gene;
  • association analysis;
  • genome-wide linkage analysis;
  • Human familial longevity;
  • nonagenarian sibling pairs

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Clear evidence exists for heritability of human longevity, and much interest is focused on identifying genes associated with longer lives. To identify such longevity alleles, we performed the largest genome-wide linkage scan thus far reported. Linkage analyses included 2118 nonagenarian Caucasian sibling pairs that have been enrolled in 15 study centers of 11 European countries as part of the Genetics of Healthy Aging (GEHA) project. In the joint linkage analyses, we observed four regions that show linkage with longevity; chromosome 14q11.2 (LOD = 3.47), chromosome 17q12-q22 (LOD = 2.95), chromosome 19p13.3-p13.11 (LOD = 3.76), and chromosome 19q13.11-q13.32 (LOD = 3.57). To fine map these regions linked to longevity, we performed association analysis using GWAS data in a subgroup of 1228 unrelated nonagenarian and 1907 geographically matched controls. Using a fixed-effect meta-analysis approach, rs4420638 at the TOMM40/APOE/APOC1 gene locus showed significant association with longevity (P-value = 9.6 × 10−8). By combined modeling of linkage and association, we showed that association of longevity with APOEε4 and APOEε2 alleles explain the linkage at 19q13.11-q13.32 with P-value = 0.02 and P-value = 1.0 × 10−5, respectively. In the largest linkage scan thus far performed for human familial longevity, we confirm that the APOE locus is a longevity gene and that additional longevity loci may be identified at 14q11.2, 17q12-q22, and 19p13.3-p13.11. As the latter linkage results are not explained by common variants, we suggest that rare variants play an important role in human familial longevity.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Nonagenarians, centenarians, and their first degree family members have a life-long survival advantage (Perls et al., 2002; Schoenmaker et al., 2006) that can be attributed to a lower risk of coronary artery disease, cancer and type-2 diabetes (Terry et al., 2003; Westendorp et al., 2009). In middle age, members of long-lived families display characteristic of metabolic health such as low glucose levels and preserved insulin sensitivity (Rozing et al., 2010; Slagboom et al., 2011). The clustering of longevity in families suggests a heritable component (Gudmundsson et al., 2000; Perls et al., 2000) which has been estimated at approximately 25% in the general population (Skytthe et al., 2003) with increasing importance at the highest ages (Hjelmborg et al., 2006). However, the genetic basis of longevity, first clearly identified as a research priority by Schächter et al. (Schachter et al., 1993), still remains to be elucidated.

Previously, it has been demonstrated that long-lived families carry as many GWAS-identified disease susceptibility alleles as the general population (Beekman et al., 2010). It can therefore be hypothesized that long-lived families carry gene variants that promote healthy aging and protect from disease. The genomic location of such protective variants can be identified in a genome-wide linkage scan among long-lived siblings. In the past, by genotyping 400 microsatellite markers in 137 long-lived sibships, Puca et al. (Puca et al., 2001) detected linkage at 4q25 (LOD = 3.65). In a subsequent study, modest support was provided for this locus by a genome-wide linkage scan in 95 male sibling pairs concordant for healthy aging (Reed et al., 2004). However, in a targeted study of 4q25 in 164 nonagenarian sibships no linkage was observed (Beekman et al., 2006). By association of the positional candidate genes at 4q25, MTP was suggested to explain the linkage (Geesaman et al., 2003), but this association could not be replicated in other studies (Bathum et al., 2005; Nebel et al., 2005; Beekman et al., 2006). Extension of the original group of long-lived siblings as investigated by Puca et al. to 279 pairs also resulted in lack of evidence for linkage at 4q25, although novel linkage regions for longevity were observed at chromosome 3p24-22 and chromosome 9q31-24 (Boyden & Kunkel, 2010).

A number of studies have performed genome-wide association (GWA) analysis for longevity to uncover common genetic variation involved in this trait. A rather large association study form the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium of 1836 individuals aged over 90 years and 1955 controls between 55 and 80 years did not reveal genome-wide significant loci (Newman et al., 2010) and neither did the analyses of all-cause mortality and survival free of major disease in this cohort (Walter et al., 2011). A smaller Dutch study of 403 nonagenarians and 1670 controls younger than 65 years identified the APOE gene as a mortality locus (Deelen et al., 2011), which was confirmed in a German study of 763 long-lived individuals and 1085 younger controls (Nebel et al., 2011) and a longitudinal study of 1606 Danes showed that the effect size of this association increases at the highest ages (Jacobsen et al., 2010). Apparently, the influence of the common genetic variation on longevity is small which requires large meta-GWA studies for identification. Alternatively, rare genetic variants may play a more important role in longevity. As the previous linkage studies showed contradictory results potentially due to heterogeneity in the longevity phenotype, it is expected that longevity is influenced by many private rare variants.

In this study, we performed genome-wide linkage analysis among 2118 European nonagenarian sibships of the Genetics of Healthy Aging (GEHA) Study (Franceschi et al., 2007; Skytthe et al., 2011). This integrated European project was initiated in 2004 with the aim of identifying genes involved in healthy aging and longevity. The GEHA selection criterion of nonagenarian sibling pairs has been shown to result in families enriched for genetic influences on longevity by a smaller previous study of the same design (the Leiden Longevity Study) (Schoenmaker et al., 2006). This study is characterized by a survival benefit in multiple generations, low prevalence of disease, and beneficial metabolic phenotypes, which is comparable to the observations reported for families of centenarian singletons such as the Longevity Study at Albert Einstein College of Medicine (Barzilai et al., 2010). Sibships over 90 years of age and controls between 55 and 75 years of age have been recruited in 10 European countries and their genomic DNA has been isolated and genotyped centrally. This logistic achievement resulted in the largest linkage study for human longevity.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

In the Genetics of Healthy Aging Study (GEHA), nonagenarian sibling pairs have been recruited in 11 countries among 15 study centers (Skytthe et al., 2011). Genotypings of 5734 SNPs (Illumina HumanLinkage12 set) were available after quality control in 4445 persons belonging to 2118 full sibships with a mean sibship size of 2.1 (Table 1).

Table 1. Number of sibships and sibship sizes for the GEHA sample sets
ClusterCountryCenterSibship sizeSibshipsMean age (years)Male sibpairsFemale sibpairs
2345
1FinlandTampere124241115092.611102
2UkraineKiev461004793.2138
2PolandWarsaw13140013593.01187
2GermanyKiel904009492.81253
2DenmarkOdense384455043492.350215
2The NetherlandsLeiden150142016693.02082
2UKNewcastle981009992.8755
2UKBelfast524105792.9440
2BelgiumLouvain761007792.91340
2FranceMontpellier238273126993.330157
3ItalyBologna178236120893.414126
3ItalyRome512005392.5632
3ItalyCalabria18171018993.02479
3ItalySassari432004593.4324
3GreeceAthens923009592.95715
  Total1934162193211892.92631145

To investigate potential substructure, identical-by-state (IBS) estimates for all pairs of individuals in the data set were computed using EIGENSTRAT (Price et al., 2006). The first two resulting principal components (C1 and C2) were plotted against each other (Fig. 1), which gives a representation of the data in two dimensions. In the resulting scatter plot, each point represents an individual and each recruitment center is marked by its own symbol. After visual inspection of component 1 (C1) and component 2 (C2), C1 appeared to reflect the North-south localization and C2 dissociated the Fins from the rest of Europe. On the basis of this genetic heterogeneity among the study samples, the linkage analyses were performed in three geographical clusters: (i) Finland with Nsibships = 150; (ii) Northern Europe with Nsibships = 1378, consisting of Ukraine, Poland, Germany, Denmark, The Netherlands, United Kingdom, Belgium, and France; (iii) Southern Europe with Nsibships = 590, consisting of Italy and Greece.

image

Figure 1. Population structure among GEHA study and the grouping of countries into three clusters. Cluster 1: Finland; Cluster 2: Northern Europe consisting of Ukraine, Poland, Germany, Denmark, The Netherlands, United Kingdom, Belgium, and France; Cluster 3: Southern Europe consisting of Italy and Greece.

Download figure to PowerPoint

We calculated nonparametric linkage scores over the whole genome for the three clusters combined (Fig. 2, blue line; Data S2). Besides the inflated LOD score probably due to linkage disequilibrium between markers at the beginning of chromosome 3, the combined linkage score for the GEHA study exceeded the genome-wide significant LOD score of 3 at chromosome 14q11.2 (8 cM), 19p13.3-p13.11 (36 cM), and 19q13.11-q13.32 (68 cM) (Fig. 3). For centers for which population-based mortality rates were available, i.e., Italy, UK, Denmark, France, Netherlands, Germany, and Finland, a weighted NPL statistic was computed. The weights are the products of the age- and sex-specific cumulative hazard for the two siblings (Houwing-Duistermaat et al., 2009) that are based on country-specific mortality rates. In this weighted NPL analysis extremely old sibling pairs obtain more weight (Fig. 2, yellow line) and the LOD score at chromosome 17q12-q22 (82 cM) increased from 2.95 to 3.71 (Fig. 3; Table 2). This increase has been observed in each of the three clusters Northern Europe, Southern Europe, and Finland. The northern European cluster contributes most to the linkage at 14q11.2 (8 cM), 19p13.3-p13.11 (36 cM), and 19q13.11-q13.32 (68 cM), whereas all three clusters contribute to linkage at 17q12-q22 (Fig. S1).

Table 2. Four linkage regions with LOD score above 3 in combined GEHA samples for NPL and age-weighted linkage analyses
 14q11.217q12-q2219p13.3-p13.1119q13.11-q13.32
  1. a

    According to genome build 35.

  2. b

    Population mortality rates were available in: Cluster 1: Finland; Cluster 2: Denmark, UK, The Netherlands, France; Cluster 3: Italy.

  3. c

    weighted with ages of the siblings.

1LOD-drop interval0–12 cMa68–95 cMa20–42 cMa58–85 cMa
Start SNP Rs10484218 Rs2429990 Rs432001 Rs7250748
End SNP Rs977870 Rs12947910 Rs919333 Rs10403760
NPL3.472.953.763.57
NPL_subsetsb3.153.263.853.80
Weighted_NPL_subsetsc2.693.712.961.85
image

Figure 2. Genome-wide linkage graphs of among 2118 sibships of the GEHA study. The blue line displays the NPL LOD scores and the yellow line the age-weighted LOD scores.

Download figure to PowerPoint

image

Figure 3. Chromosomal regions linked to longevity with LOD score above 3. The blue line displays the NPL LOD scores and the yellow line the age-weighted LOD scores.

Download figure to PowerPoint

As gender-specific effects play a role in human longevity, linkage analysis was also performed in male-only (Npairs = 263) and female-only (Npairs = 1145) sibling pairs. Although NPL analysis did not reveal any linkage for male-only pairs, significant linkage (Weighted-NPL 3.61) was observed at 8p11.21-q13.1 (70 cM) among the male-only pairs in the age-weighted analysis (Fig. 4, lower panel; Table 3). Among the female-only sibpairs (Fig. 4, upper panel) significant linkage was observed at chromosome 15q12-q14 (22 cM) (Weighted-NPL 3.16) and at 19q13.33-q13.41 (80 cM) (NPL = 4.97) (Table 3). Further investigation of these linkage areas revealed that the Northern Europe cluster and the Finland cluster provide evidence for the 8p11.21-q13.1 and 15q12-q14 loci, and that the 19q13.33-q13.41 locus seems only to link with female longevity in the Northern Europe cluster (Fig. S2).

Table 3. Sex-specific linkage regions with LOD score above 3 in combined GEHA samples for NPL and age-weighted linkage analyses
Sex8p11.21-q13.115q12-q1419q13.33-q13.41
MenWomenWomen
  1. a

    According to genome build 35.

  2. b

    Population mortality rates were available in: Cluster 1: Finland; Cluster 2: Denmark, UK, The Netherlands, France; Cluster 3: Italy.

  3. c

    weighted with ages of the siblings.

1LOD-drop interval64–82 cMa14–28 cMa66–90 cMa
Start SNP Rs801100 Rs1871009 Rs1236093
End SNP Rs4368961 Rs580839 Rs1661965
NPL1.140.624.97
NPL_subsetsb1.961.443.92
Weighted_NPL_subsetsc3.613.163.60
image

Figure 4. Genome-wide linkage graphs of among 1145 female-only sibships and 263 male-only sibships of the GEHA study. The blue line displays the NPL LOD scores and the yellow line the age-weighted LOD scores.

Download figure to PowerPoint

When we consider the genes in the 1-LOD drop intervals of the linkage analysis as candidate genes for longevity, we have a list of 1151 unique position candidate genes (Data S3) for which we have performed gene set enrichment analysis (Table S3, Data S4). The 1-LOD drop intervals were investigated for association with longevity by comparison of unrelated nonagenarians to geographically matched younger controls. Of each sibpair of the centers that provided the largest contribution to the linkage results [i.e., Netherlands, Denmark, France, Italy (Bologna), and UK (Newcastle)] (Table S1), the eldest sibs were taken as cases and as controls for France and Bologna, GEHA controls were taken. As controls for UK, Netherlands and Denmark already genotyped available younger population controls were used. We performed a meta-analysis for 57 607 imputed SNPs in the seven candidate regions with longevity. We tested for differences in genotype frequencies between cases and geographically matched controls using a trend test, i.e., assuming additive genetic effects. The test statistic was adjusted for uncertainty in imputations and for the inflation factor before computing the p-values. We used a fixed-effect meta-analysis approach to combine the results across centers. Finally, we used a Bonferroni correction based on the number of SNPs within the region to correct for multiple testing.

No significant association (P-value <8.7 × 10−7) was observed at the loci at chromosome 14q11.2, 17q12-q22, and 19p13.3-p13.11, and neither at the sex-specific loci. At 19q13.11-q13.32 locus, rs4420638 at the TOMM40/APOE/APOC1 gene locus showed significant association with longevity (P-value = 9.6 × 10−8). As the linkage plots per center show that especially the sample from Denmark provides the statistical evidence for an increased LOD score at this position (Fig. S3), we tested in the Danish data set with the software LAMP (Burdick et al., 2006) for residual linkage at the 19q13.11-q13.21 locus if the APOEε2ε3ε4 rs7412 and rs429358 genotypes were taken into account. It appeared that the APOEε4 allele, although with moderate significance, contributed to the linkage peak (P-value = 0.02). Remarkably, with a much higher significance the APOEε2 allele also explained the linkage peak (P-value = 1.0 × 10−5).

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

To identify chromosomal regions involved in longevity, we performed the largest linkage analysis thus far reported. This analysis included 2118 nonagenarian sibling pairs from 11 European countries enrolled as part of the Genetics of Healthy Aging (GEHA) project. We identified four regions linked to human longevity at 14q11.2, 17q12-q22, 19p13.3-p13.11, and 19q13.11-q13.32. In addition to these loci, we identified three loci that were linked to longevity in a sex-specific manner: 8p11.21-q13.1 (men), 15q12-q14 (women), and 19q13.33-q13.41 (women). In these seven linked regions, we performed association analysis using GWAS data that was available in a subset of the cohorts, and we showed that genetic variation at the APOE gene locus was significantly associated with longevity. The absence of the APOEε4 allele among the nonagenarians is obvious (Deelen et al., 2011) and contributed to the linkage, suggesting that APOE is a mortality gene. However, the linkage was mainly explained by the shared presence of the APOEε2 allele, indicating that the APOE gene indeed is a longevity gene.

Affected sibpair analysis is a powerful tool to map strong dominant as well as recessive gene effects, but to map rare genetic variation with relatively small effect a group of at least 600 sibling pairs with one sibling over age 95 years should be examined (Tan et al., 2004). As the GEHA project yielded an impressive number of 2118 genotyped nonagenarian sibling pairs to investigate for linkage (Skytthe et al., 2011), of which 899 sibling pairs contained one sib aged over 95 years of age, this linkage study has sufficient power to detect the loci with relatively small effects.

Previous studies for longevity loci detected linkage at 4q25 (Puca et al., 2001), 3p24-22, 9q31-34, 12q24 (Boyden & Kunkel, 2010), 6p12.1, 7q11.21, 14q22.1 (Edwards et al., 2011). None of these loci showed linkage in the GEHA project. The largest of the previous linkage studies examined 279 long-lived sibling pairs, which is considered a relatively small study and therefore the previous linkage results should be considered with caution, given the complexity of human longevity (Tan et al., 2004). Otherwise, the effect of previously detected loci on longevity could be private for a specific genetic background or specific environment, or even private for a small number of families.

In the chromosomal region linked to familial longevity at 19q13.11-q13.32, we found rs4420638 to be significantly associated with familial longevity. This SNP tags the LD block harboring the TOMM40, APOE, and APOC1 genes. It is known that the APOE gene exhibit three isoforms that are caused by two functional SNPs rs7412 (APOEε2) and rs429358 (APOEε4) which have been associated with longevity (Schachter et al., 1994). The APOEε4 allele frequency in GEHA nonagenarian sibling cases was 6.8%, which was significantly lower than the allele frequency of 12.7% among the geographically matched controls. APOEε4 allele carriers have about 50% lower chance to become a nonagenarian then non-APOEε4 allele carriers (OR = 0.48, 95% CI = 0.42–0.55). Recently, this gene region has been explored for other functional variation and multiple cis-elements are found to influence both APOE and TOMM40 promoter activity and a complex transcriptional regulatory structure has been suggested modulates regional expression (Bekris et al., 2012). Hence, future research should consider exploring these other functional variants for their role in familial longevity.

Genetic association analysis as an approach is suitable to detect the effect on a trait of common genetic variation (minor allele frequency larger than 1%). We did not observe significant associations in the families that contributed to linkage of common variants in the linked regions with longevity which can be explained by the possibility that the positive linkage signals may be based on allelic and/or locus heterogeneity, i.e., different genetic variants in the same gene or gene region contributed to the trait in different families. The less likely explanation is that the power, which was large enough for detecting linkage, may be too small to detect genetic association among 57K SNPs in the linkage regions in 1228 cases. Thus, common variants may not contribute to the longevity phenotype and by applying next generation sequencing analyses of the nonagenarians contributing to linkage rare, private genetic variation in the linkage regions may be found.

This genome-wide linkage scan did not provide substantial evidence for linkage to FOXO3A on chromosome 6, of which the T allele of rs2802295 predisposes to survive into old age (reviewed in Chung et al., 2010). This may be due to the relatively small effect sizes conferred by SNPs at that locus or it could be that the nonagenarian sibling design is underpowered to detect the association of FOXO3A with longevity because its effect seems the strongest at ages above 95, in which age range the GEHA study has limited numbers of participants. Besides the NPL analyses we performed linkage analyses in which the contribution of the sibling pairs has been weighted for the age of the eldest sibling. Thus, when a locus is involved in survival after an extremely old age, as is suggested for the FOXO3A locus, linkage signals would increase in this weighted analyses. Despite the use of these analysis tools, we did not observe any LOD score above LOD score of 1 at the FOXO3A locus. We therefore conclude that genetic variation at the FOXO3A gene might not contribute to longevity in the GEHA Study.

Previously, it has been suggested that genetic variation in the FOXO1 gene is specifically contributing to human female longevity (reviewed in Chung et al., 2010). However, at chromosome 13q14.11 harboring the FOXO1 gene we found no evidence for linkage with female longevity (LOD < 0.05) and at the gene position of FOXO1 we found no evidence for association in the females-only meta-analysis (P-values >0.042) in the GEHA Study. Potentially, the effect of this locus is not only influenced by gender but also by genetic background.

When we investigated the cluster-specific linkage results, we observed that Northern Europe cluster contributes to all linkage loci, the Southern Europe cluster only contributes to the 17q12-q22 and 19p13.3-p13.11 loci and the Finland cluster only contributes to the 14q11.2 and 17q12-q22 loci (Fig. S3). The Southern Europe cluster and the Finland cluster did not obtain LOD scores above 3 on their own and the Northern Europe cluster obtained LOD scores above 3 only at chromosome 19 at 20, 68, and 82 cM. This illustrates that probably, in addition to the sample size of the cluster, the environment and genetic background play a role in the complex phenotype of human familial longevity.

Besides linkage at the APOE locus, we detected linkage at 14q11.2, 17q12-q22, and 19p13.3-p13.11 considering the total GEHA study. These chromosomal regions linked to human longevity cover many genes that are potential candidate genes involved in neurodegenerative disease, auto-immune disease, cardiovascular disease, and cancer. In the NCBI GWAS catalog (http://www.genome.gov/gwastudies/) genome-wide association has been reported at the 14q locus for the T-cell receptor alpha with narcolepsy. At the 17q12-q22 locus, genome-wide association has been reported for MAPT and NSF with Parkinson's disease, for CA10 with age of onset of menarche, and for STAT3 with Multiple Sclerosis and Crohn's disease. At the chromosome 19p13.3-p13.11 locus, genome-wide association has been reported for TYK2 and ICAM3 with Crohn's disease, for LDLR with myocardial infarction, for SMARCA4 with coronary heart disease, and for ABHD8 and ANKLE1 with breast cancer.

In conclusion, we identified four chromosomal regions linked to human familial longevity and we are the first identifying genetic linkage with familial longevity at the APOE gene locus. As genetic variation at the APOE gene explains the linkage, we confirm that the APOE locus is indeed a longevity gene. Identification of additional longevity genes in the linkage regions will provide biological insight into the pathways underlying human familial longevity, metabolic health, and the resistance to age-related disease.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Study subjects

The individuals investigated in this study participate in Genetics of Healthy Aging (GEHA) Study (Skytthe et al., 2011). Families participating in the GEHA study have at least two siblings meeting four inclusion criteria: (i) participants are at least 90 years old, (ii) participants have at least one living brother or sister who fulfills the first criterion and is willing to participate, (iii) the nonagenarian sibship has an identical mother and father, and (iv) the parents of the nonagenarian sibship are European and Caucasian. In total, 2249 sibships have been recruited.

In accordance with the Declaration of Helsinki, written informed consent was obtained from all participants prior to entering the study. Good clinical practice guidelines were maintained. The study protocol was approved by the local medical ethical committees of the 11 participating countries before the start of the study.

DNA isolation

The National Institute for Health and Welfare (THL, Helsinki), extracted high molecular DNA from 7 ml of whole blood using automated DNA purification, Autopure LS (Qiagen), based on Puregene salting out methodology. DNA concentrations were adjusted to 50 ng/μL, verified using PicoGreen dsDNA Quantitation kit (Molecular Probes, Life Technologies, Paisley, UK). The samples were then subjected to quality control by ABI 3730 DNA analyzer (Applied Biosystems, Life Technologies) using two sex-specific and four autosomal microsatellite markers.

Genotyping for linkage analysis

The Centre National de Génotypage (CNG, Paris) performed genotyping of 6090 SNPs in the nonagenarian siblings of the 2249 sibships recruited in the GEHA project with the Illumina HumanLinkage-12 Genotyping BeadChip. Genotyping call rate was larger than 95% per sample. We observed 19 duplicates that are either MZ twins or duplicate samples of which two were found in two different families. Five full sibs were identified as unrelated and eight sample pairs that are supposed to be unrelated were identified as potential half-sibs. Of all sibships, 35 appeared to be composed of half-sibs. Finally, two sample pairs that are supposed to be unrelated were found full sibs. After removing the samples with unexpected relations, we obtained reliable genotyping for 4445 individuals from 2118 sibships (Table 1). As the recruitment of the GEHA study has been a major enterprise, it turned out in a later phase of the project that some cases were aged below 90 years of age at the time of interview (Table S2 and Fig. S4).

We applied the EIGENSTRAT program (Price et al., 2006) with default parameters to infer axes of variation with the 6090 SNPs per individual to identify genetic population stratification.

Affected sibpair analysis

Unlikely recombinants were detected using Merlin-0.10.2 (Abecasis et al., 2002) and erroneous genotypes were removed with Pedwipe. In addition, Pearson's r2 was estimated using the total data set. For the nonparametric (NPL) analyses, 275 SNPs were removed to obtain a genome-wide SNP set of 5734 SNPs with pairwise r2 smaller than 0.4. MERLIN-0.10.2 (Abecasis et al., 2002) was used to estimate the information content over the genome (Data S1) and to estimate IBD probabilities in the sibling pairs without parents.

Next, proportion of alleles shared identical by descent was estimated in Merlin (Abecasis et al., 2002) allowing for center-specific allele frequencies. Nonparametric linkage analysis (NPL) (Kruglyak & Lander, 1996) was performed using a score test statistic for affected sibling pairs (Callegaro et al., 2009) that was computed at a grid of 2 cM. NPL statistics per cluster were obtained by summing the NPL statistics per involved center. This statistic compares the actual proportion of alleles shared identical by descent by a sibling pair with its expectation of 0.5 under Mendelian segregation. The variance of the statistic was obtained via simulation in Merlin. By simulation, the variance is adjusted for incomplete marker informativeness and correlation of sibling pairs within families. The statistic assumes that there is no linkage disequilibrium between the markers. When this assumption is violated, the peaks will be inflated.

For centers for which population-based mortality rates were available, i.e., Italy, United Kingdom, Denmark, France, Netherlands, Germany, and Finland, a weighted NPL statistic was computed. The weights are the products of the age- and sex-specific cumulative hazard for the two siblings (Houwing-Duistermaat et al., 2009). These cumulative hazards are based on country-specific mortality rates. In this analysis, extremely old sibling pairs obtain more weight in the analysis.

Subjects for association analysis within linked regions

To fine map the 1-LOD drop linkage regions, association analysis was performed in the largest centers and those most contributing to the linkage results. From each pair the eldest siblings from the United Kingdom, Netherlands, Denmark, France, and Bologna, and GEHA controls from France and Bologna were genotyped. As controls for United Kingdom, Netherlands, and Denmark available geographically matched control data sets were used: TwinUK for UK nonagenarians (Perola et al., 2007), controls from the Leiden Longevity Study (LLS) for Dutch nonagenarians (Beekman et al., 2010), and Danish GOYA controls (Paternoster et al., 2011) for Danish nonagenarians. The characteristics of the samples are displayed in table S1.

Genotyping for association analysis within linked regions

The GEHA nonagenarians from the UK, Netherlands, Denmark, and France (CNG, Paris, France) and the LLS controls (genotyping facility Erasmus MC, Rotterdam, The Netherlands) have been genotyped using Illumina Infinium HD Human660W-Quad BeadChips, the GEHA nonagenarians from Bologna, their controls and the France controls have been genotyped using Illumina HumanOmniExpress BeadChips (genotyping facility Erasmus MC, Rotterdam, The Netherlands) and the Danish controls (GOYA) and UK controls (TwinsUK) have been genotyped using Illumina Infinium HD Human610-Quad BeadChips. Autosomal SNPs were included in analysis if they had less than 5% missing data, Hardy–Weinberg P-values in cases and controls greater than 1 × 10−4 and minor allele frequency (MAF) was larger than 1% (if ncases and ncontrols ≥200) or 5%. SNPs that passed this quality control in both cases and controls were used as input for imputation to HAPMAP2 release 22 using IMPUTE2. SNPs with a R2T value lower than 40 and a MAF lower than 1% (if ncases and ncontrols ≥200) or 5% were excluded from analyses (Uh et al., 2011).

Association analysis within linked regions

Association analysis has been performed using CC-assoc applying score test (Uh et al., 2011). For meta-analysis, a fixed-effect approach was used and adjustment for population stratification per study was performed by multiplying the variances with the genomic inflation factor (λ), which measures overdispersion of test statistics from association tests. Scores and adjusted variances of the five countries were combined to obtain a single metastatistic. We tested 57 607 SNPs within the 1-LOD drop intervals and according to Bonferroni adjustment for multiple testing, P-values below 8.7 × 10−7 were considered genome-wide significant.

Combined linkage and association at APOEε2ε3ε4 locus

To investigate whether the linkage identified can be explained by the APOEε2ε3ε4 polymorphisms, we performed linkage and association modeling in pedigrees (LAMP) (Li et al., 2005) in the Danish sample in which rs7412 and rs429358 have been genotyped by TaqMan using standard protocol of the supplier.

Gene set enrichment analyses

For each linkage region, the 1-LOD drop intervals have been determined. Next, the genes located in these 1-LOD drop intervals have been listed and tested for pathway enrichment using default settings in DAVID (http://david.abcc.ncifcrf.gov/).

Acknowledgments

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information

Funding: The work described in this study was funded mainly by the EU GEHA Project contract no. LSHM-CT-2004-503-270. The work has additionally been supported by the following programs and agencies: The Competitive Research Funding of the Tampere University Hospital and Academy of Finland (Tampere); United States National Institute of Aging (PO1-AG08761) (Odense); The Innovation Oriented Research Program on Genomics (Senter-Novem IGE05007), the Centre for Medical Systems Biology (CMSB), and the National Institute for Healthy Ageing (NCHA 05060810), all in the framework of the Netherlands Genomics Initiative (NGI)/Netherlands Organisation for Scientific Research (NWO) (Leiden); The Institute for Ageing and Health and the UK NIHR Biomedical Research Centre for Ageing and Age-related disease award to the Newcastle-upon-Tyne Foundation Hospitals NHS Trust (Newcastle); Fondation Caisse d'Epargne Rhône-Alpes Lyon CERAL (2004–2007) (Montpellier); Regione Autonoma della Sardegna (Sassari), European Union's Seventh Framework Programme (FP7/2007-2011) IDEAL-ageing under grant agreement no. 259679.

Access to genotype data from the TwinsUK cohort was kindly provided by the Department of Twin Research (DTR) and Genetic Epidemiology at King's College London. The TwinsUK cohort acknowledges funding from the Wellcome Trust, European Community's FP7 Programme (HEALTH-F2-2008-201865 GEFOS and HEALTH-F4-2007-201413 ENGAGE projects) and the FP5 Programme (GenomEUtwin Project QLG2-CT-2002-01254), BBSRC project grant G20234, and the National Eye Institute via an NIH/CIDR genotyping project (PI: Terri Young). We thank the staff of the Twin Research Unit for their help and support in undertaking this project. We would also like to thank all patients, their families, and control subjects for their voluntary contribution to this research project.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Abecasis GR, Cherny SS, Cookson WO, Cardon LR (2002) Merlin–rapid analysis of dense genetic maps using sparse gene flow trees. Nat. Genet. 30, 97101.
  • Barzilai N, Gabriely I, Atzmon G, Suh Y, Rothenberg D, Bergman A (2010) Genetic studies reveal the role of the endocrine and metabolic systems in aging. J. Clin. Endocrinol. Metab. 95, 44934500.
  • Bathum L, Christiansen L, Tan Q, Vaupel J, Jeune B, Christensen K (2005) No evidence for an association between extreme longevity and Microsomal Transfer Protein polymorphisms in a longitudinal study of 1651 nonagenarians. Eur. J. Hum. Genet. 13, 11541158.
  • Beekman M, Blauw GJ, Houwing-Duistermaat JJ, Brandt BW, Westendorp RG, Slagboom PE (2006) Chromosome 4q25, microsomal transfer protein gene, and human longevity: novel data and a meta-analysis of association studies. J. Gerontol. A Biol. Sci. Med. Sci. 61, 355362.
  • Beekman M, Nederstigt C, Suchiman HE, Kremer D, Van der Breggen R, Lakenberg N, Alemayehu WG, De Craen AJ, Westendorp RG, Boomsma DI, de Geus EJ, Houwing-Duistermaat JJ, Heijmans BT, Slagboom PE (2010) Genome-wide association study (GWAS)-identified disease risk alleles do not compromise human longevity. Proc. Natl Acad. Sci. USA 107, 1804618049.
  • Bekris LM, Lutz F, Yu CE (2012) Functional analysis of APOE locus genetic variation implicates regional enhancers in the regulation of both TOMM40 and APOE. J. Hum. Genet. 57, 1825.
  • Boyden SE, Kunkel LM (2010) High-density genomewide linkage analysis of exceptional human longevity identifies multiple novel loci. PLoS ONE 5, e12432.
  • Burdick JT, Chen WM, Abecasis GR, Cheung VG (2006) In silico method for inferring genotypes in pedigrees. Nat. Genet. 38, 10021004.
  • Callegaro A, van Houwelingen HC, Houwing-Duistermaat JJ (2009) Score test for age at onset genetic linkage analysis in selected sibling pairs. Stat. Med. 28, 19131926.
  • Chung WH, Dao RL, Chen LK, Hung SI (2010) The role of genetic variants in human longevity. Ageing Res. Rev. 9(Suppl 1), S67S78.
  • Deelen J, Beekman M, Uh HW, Helmer Q, Kuningas M, Christiansen L, Kremer D, Van der Breggen R, Suchiman HED, Lakenberg N, Van den Akker EB, Passtoors WM, Tiemeier H, van Heemst D, De CA, Rivadeneira F, De Geus EJ, Perola M, van der Ouderaa FJ, Gunn DA, Boomsma DI, Uitterlinden AG, Christensen K, van Duijn CM, Heijmans BT, Houwing-Duistermaat JJ, Westendorp RG, Slagboom PE (2011) Genome-wide association study identifies a single major locus contributing to survival into old age; the APOE locus revisited. Aging Cell 10, 686698.
  • Edwards DR, Gilbert JR, Jiang L, Gallins PJ, Caywood L, Creason M, Fuzzell D, Knebusch C, Jackson CE, Pericak-Vance MA, Haines JL, Scott WK (2011) Successful aging shows linkage to chromosomes 6, 7, and 14 in the Amish. Ann. Hum. Genet. 75, 516528.
  • Franceschi C, Bezrukov V, Blanche H, Bolund L, Christensen K, De BG, Deiana L, Gonos E, Hervonen A, Yang H, Jeune B, Kirkwood TB, Kristensen P, Leon A, Pelicci PG, Peltonen L, Poulain M, Rea IM, Remacle J, Robine JM, Schreiber S, Sikora E, Slagboom PE, Spazzafumo L, Stazi MA, Toussaint O, Vaupel JW (2007) Genetics of healthy aging in Europe: the EU-integrated project GEHA (GEnetics of Healthy Aging). Ann. N. Y. Acad. Sci. 1100, 2145.
  • Geesaman BJ, Benson E, Brewster SJ, Kunkel LM, Blanche H, Thomas G, Perls TT, Daly MJ, Puca AA (2003) Haplotype-based identification of a microsomal transfer protein marker associated with the human lifespan. Proc. Natl Acad. Sci. USA 100, 1411514120.
  • Gudmundsson H, Gudbjartsson DF, Frigge M, Gulcher JR, Stefansson K (2000) Inheritance of human longevity in Iceland. Eur. J. Hum. Genet. 8, 743749.
  • Hjelmborg JV, Iachine I, Skytthe A, Vaupel JW, McGue M, Koskenvuo M, Kaprio J, Pedersen NL, Christensen K (2006) Genetic influence on human lifespan and longevity. Hum. Genet. 119, 312321.
  • Houwing-Duistermaat JJ, Callegaro A, Beekman M, Westendorp RG, Slagboom PE, van Houwelingen JC (2009) Weighted statistics for aggregation and linkage analysis of human longevity in selected families: the Leiden Longevity Study. Stat. Med. 28, 140151.
  • Jacobsen R, Martinussen T, Christiansen L, Jeune B, Andersen-Ranberg K, Vaupel JW, Christensen K (2010) Increased effect of the ApoE gene on survival at advanced age in healthy and long-lived Danes: two nationwide cohort studies. Aging Cell 9, 10041009.
  • Kruglyak L, Lander ES (1996) Limits on fine mapping of complex traits. Am. J. Hum. Genet. 58, 10921093.
  • Li M, Boehnke M, Abecasis GR (2005) Joint modeling of linkage and association: identifying SNPs responsible for a linkage signal. Am. J. Hum. Genet. 76, 934949.
  • Nebel A, Croucher PJ, Stiegeler R, Nikolaus S, Krawczak M, Schreiber S (2005) No association between microsomal triglyceride transfer protein (MTP) haplotype and longevity in humans. Proc. Natl Acad. Sci. USA 102, 79067909.
  • Nebel A, Kleindorp R, Caliebe A, Nothnagel M, Blanche H, Junge O, Wittig M, Ellinghaus D, Flachsbart F, Wichmann HE, Meitinger T, Nikolaus S, Franke A, Krawczak M, Lathrop M, Schreiber S (2011) A genome-wide association study confirms APOE as the major gene influencing survival in long-lived individuals. Mech. Ageing Dev. 132, 324330.
  • Newman AB, Walter S, Lunetta KL, Garcia ME, Slagboom PE, Christensen K, Arnold AM, Aspelund T, Aulchenko YS, Benjamin EJ, Christiansen L, D'Agostino RB Sr, Fitzpatrick AL, Franceschini N, Glazer NL, Gudnason V, Hofman A, Kaplan R, Karasik D, Kelly-Hayes M, Kiel DP, Launer LJ, Marciante KD, Massaro JM, Miljkovic I, Nalls MA, Hernandez D, Psaty BM, Rivadeneira F, Rotter J, Seshadri S, Smith AV, Taylor KD, Tiemeier H, Uh HW, Uitterlinden AG, Vaupel JW, Walston J, Westendorp RG, Harris TB, Lumley T, van Duijn CM, Murabito JM (2010) A meta-analysis of four genome-wide association studies of survival to age 90 years or older: the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium. J. Gerontol. A Biol. Sci. Med. Sci. 65, 478487.
  • Paternoster L, Evans DM, Nohr EA, Holst C, Gaborieau V, Brennan P, Gjesing AP, Grarup N, Witte DR, Jorgensen T, Linneberg A, Lauritzen T, Sandbaek A, Hansen T, Pedersen O, Elliott KS, Kemp JP, St PB, McMahon G, Zelenika D, Hager J, Lathrop M, Timpson NJ, Smith GD, Sorensen TI (2011) Genome-wide population-based association study of extremely overweight young adults–the GOYA study. PLoS ONE 6, e24303.
  • Perls T, Shea-Drinkwater M, Bowen-Flynn J, Ridge SB, Kang S, Joyce E, Daly M, Brewster SJ, Kunkel L, Puca AA (2000) Exceptional familial clustering for extreme longevity in humans. J. Am. Geriatr. Soc. 48, 14831485.
  • Perls TT, Wilmoth J, Levenson R, Drinkwater M, Cohen M, Bogan H, Joyce E, Brewster S, Kunkel L, Puca A (2002) Life-long sustained mortality advantage of siblings of centenarians. Proc. Natl Acad. Sci. USA 99, 84428447.
  • Perola M, Sammalisto S, Hiekkalinna T, Martin NG, Visscher PM, Montgomery GW, Benyamin B, Harris JR, Boomsma D, Willemsen G, Hottenga JJ, Christensen K, Kyvik KO, Sorensen TI, Pedersen NL, Magnusson PK, Spector TD, Widen E, Silventoinen K, Kaprio J, Palotie A, Peltonen L (2007) Combined genome scans for body stature in 6,602 European twins: evidence for common Caucasian loci. PLoS Genet. 3, e97.
  • Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D (2006) Principal components analysis corrects for stratification in genome-wide association studies. Nat. Genet. 38, 904909.
  • Puca AA, Daly MJ, Brewster SJ, Matise TC, Barrett J, Shea-Drinkwater M, Kang S, Joyce E, Nicoli J, Benson E, Kunkel LM, Perls T (2001) A genome-wide scan for linkage to human exceptional longevity identifies a locus on chromosome 4. Proc. Natl Acad. Sci. USA 98, 1050510508.
  • Reed T, Dick DM, Uniacke SK, Foroud T, Nichols WC (2004) Genome-wide scan for a healthy aging phenotype provides support for a locus near D4S1564 promoting healthy aging. J. Gerontol. A Biol. Sci. Med. Sci. 59, 227232.
  • Rozing MP, Westendorp RG, De Craen AJ, Frolich M, de Goeij MC, Heijmans BT, Beekman M, Wijsman CA, Mooijaart SP, Blauw GJ, Slagboom PE, van Heemst D (2010) Favorable glucose tolerance and lower prevalence of metabolic syndrome in offspring without diabetes mellitus of nonagenarian siblings: the Leiden longevity study. J. Am. Geriatr. Soc. 58, 564569.
  • Schachter F, Cohen D, Kirkwood T (1993) Prospects for the genetics of human longevity. Hum. Genet. 91, 519526.
  • Schachter F, Faure-Delanef L, Guenot F, Rouger H, Froguel P, Lesueur-Ginot L, Cohen D (1994) Genetic associations with human longevity at the APOE and ACE loci. Nat. Genet. 6, 2932.
  • Schoenmaker M, De Craen AJ, de Meijer PH, Beekman M, Blauw GJ, Slagboom PE, Westendorp RG (2006) Evidence of genetic enrichment for exceptional survival using a family approach: the Leiden Longevity Study. Eur. J. Hum. Genet. 14, 7984.
  • Skytthe A, Pedersen NL, Kaprio J, Stazi MA, Hjelmborg JV, Iachine I, Vaupel JW, Christensen K (2003) Longevity studies in GenomEUtwin. Twin Res. 6, 448454.
  • Skytthe A, Valensin S, Jeune B, Cevenini E, Balard F, Beekman M, Bezrukov V, Blanche H, Bolund L, Broczek K, Carru C, Christensen K, Christiansen L, Collerton JC, Cotichini R, De Craen AJ, Dato S, Davies K, De BG, Deiana L, Flachsbart F, Gampe J, Gilbault C, Gonos ES, Haimes E, Hervonen A, Hurme MA, Janiszewska D, Jylha M, Kirkwood TB, Kristensen P, Laiho P, Leon A, Marchisio A, Masciulli R, Nebel A, Passarino G, Pelicci G, Peltonen L, Perola M, Poulain M, Rea IM, Remacle J, Robine JM, Schreiber S, Scurti M, Sevini F, Sikora E, Skouteri A, Slagboom PE, Spazzafumo L, Stazi MA, Toccaceli V, Toussaint O, Tornwall O, Vaupel JW, Voutetakis K, Franceschi C (2011) Design, recruitment, logistics, and data management of the GEHA (Genetics of Healthy Ageing) project. Exp. Gerontol. 46, 934945.
  • Slagboom PE, Beekman M, Passtoors WM, Deelen J, Vaarhorst AA, Boer JM, van den Akker EB, van Heemst D, De Craen AJ, Maier AB, Rozing M, Mooijaart SP, Heijmans BT, Westendorp RG (2011) Genomics of human longevity. Philos. Trans. R. Soc. Lond. B Biol. Sci. 366, 3542.
  • Tan Q, Zhao JH, Iachine I, Hjelmborg J, Vach W, Vaupel JW, Christensen K, Kruse TA (2004) Power of non-parametric linkage analysis in mapping genes contributing to human longevity in long-lived sib-pairs. Genet. Epidemiol. 26, 245253.
  • Terry DF, Wilcox M, McCormick MA, Lawler E, Perls TT (2003) Cardiovascular advantages among the offspring of centenarians. J. Gerontol. A Biol. Sci. Med. Sci. 58, M425M431.
  • Uh HW, Deelen J, Beekman M, Helmer Q, Rivadeneira F, Hottenga JJ, Boomsma DI, Hofman A, Uitterlinden AG, Slagboom PE, Bohringer S, Houwing-Duistermaat JJ (2011) How to deal with the early GWAS data when imputing and combining different arrays is necessary. Eur. J. Hum. Genet. 20, 572576.
  • Walter S, Atzmon G, Demerath EW, Garcia ME, Kaplan RC, Kumari M, Lunetta KL, Milaneschi Y, Tanaka T, Tranah GJ, Volker U, Yu L, Arnold A, Benjamin EJ, Biffar R, Buchman AS, Boerwinkle E, Couper D, De Jager PL, Evans DA, Harris TB, Hoffmann W, Hofman A, Karasik D, Kiel DP, Kocher T, Kuningas M, Launer LJ, Lohman KK, Lutsey PL, Mackenbach J, Marciante K, Psaty BM, Reiman EM, Rotter JI, Seshadri S, Shardell MD, Smith AV, van Duijn CM, Walston J, Zillikens MC, Bandinelli S, Baumeister SE, Bennett DA, Ferrucci L, Gudnason V, Kivimaki M, Liu Y, Murabito JM, Newman AB, Tiemeier H, Franceschini N (2011) A genome-wide association study of aging. Neurobiol. Aging 32, 21092128.
  • Westendorp RG, van Heemst D, Rozing MP, Frolich M, Mooijaart SP, Blauw GJ, Beekman M, Heijmans BT, De Craen AJ, Slagboom PE (2009) Nonagenarian siblings and their offspring display lower risk of mortality and morbidity than sporadic nonagenarians: the Leiden longevity study. J. Am. Geriatr. Soc. 57, 16341637.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgments
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
acel12039-sup-0001-FigureS1-S3.pdfapplication/PDF153K

Fig. S1 Linkage regions in the three geographically clusters at chromosome 14 (upper left panel), chromosome 17 (upper right panel) and chromosome 19 (lower panel).

Fig. S2 Sex-specific linkage regions in the three geographically clusters. Light blue line represents Northern Europe cluster, orange line represents Southern Europe cluster, purple line represents Finland.

Fig. S3 Linkage at chromosome 19 for the three largest centers from the Northern European cluster.

acel12039-sup-0002-FigS4-TableS1-S3.docapplication//msword102K

Fig. S4 Histogram of ages in the GEHA study.

Table S1 Nonagenarian cases and controls for association analyses in candidate longevity regions.

Table S2 Mean age and age range of GEHA nonagenarian sibling cases.

Table S3 Gene Ontology term enrichment among the genes resided in regions linked to longevity.

acel12039-sup-0003-DataS1.xlsapplication/msexcel802KData S1 Information content.
acel12039-sup-0004-DataS2.xlsapplication/msexcel757KData S2 Genome wide linkage results.
acel12039-sup-0005-DataS3.xlsapplication/msexcel122KData S3 Positional candidate genes.
acel12039-sup-0006-DataS4.xlsapplication/msexcel821KData S4 Gene set enrichment results from DAVID.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.