Genetic Variants in PVRL2-TOMM40-APOE Region Are Associated with Human Longevity in a Han Chinese Population

Purpose Human longevity results from a number of factors, including genetic background, favorable environmental, social factors and chance. In this study, we aimed to elucidate the association of human longevity with genetic variations in several major candidate genes in a Han Chinese population. Methods A case-control association study of 1015 long-lived individuals (aged 90 years or older) and 1725 younger controls (30–70 years old) was undertaken. Rs2075650 in TOMM40 was firstly genotyped using the ABI SNaPshot method in an initial cohort consisted of 597 unrelated long-lived individuals and 1275 younger controls enrolled from Sichuan. Secondly, eighteen tag single-nucleotide polymorphisms (SNPs) in the PVRL2-TOMM40-APOE locus were genotyped for extensive study in the same cohort. Finally, 5 associated SNPs were genotyped in a replication cohort including 418 older individuals and 450 younger controls. The genotype and allele frequencies were evaluated using the χ2 tests. The linkage disequilibrium (LD) block structure was examined using the program Haploview. Results The case-control study of rs2075650 in TOMM40 showed significant difference in allele frequencies between cases and controls (P = 0.006) in an initial study. Of the 18 SNPs genotyped, rs405509 in APOE and another three SNPs (rs12978931, rs519825 and rs395908) in the PVRL2 gene also showed significant association with human longevity in extensive study in the same cohort. Rs2075650 in TOMM40, rs405509 in APOE and rs519825 in PVRL2 showed a significant association with human longevity in a replication cohort. Conclusion These results suggested that PVRL2, TOMM40 and APOE might be associated with human longevity. However, further research is needed to identify the causal variants and determine which of these genes are involved in the progress of human longevity.


Introduction
Human longevity is considered a multi-factorial phenotype [1]. Worldwide human populations have shown an increase in mean life expectancy in the past two centuries [2,3]. This is mainly due to environmental factors such as improved hygiene, nutrition and health care. The large variation in healthy lifespan among the elderly has prompted research into the determinants of aging and lifespan regulation. Longevity and healthy aging associated genetic research may provide further insights into the mechanisms of aging [4][5][6][7]. The genetic component has been shown to become stronger with increasing age of the individuals [8,9]. The genetic contribution to human lifespan variation was estimated at 25-30% in twin studies [8,10,11]. The most prominent genetic influence is observed in families in which the capacity to attain a long lifespan clusters [12][13][14]. Epidemiological data indicate the presence of a strong familiar component of longevity that is largely determined by genetics and progeric syndromes of accelerated aging have known genetic causes [15]. Very long life, to beyond the age of 90 years, appears to have an even stronger genetic basis [16], which explains why centenarians and near-centenarians tend to cluster in families. Exceptional longevity can be reached with a low degree of age-related disability [17,18], raising the question whether protective mechanisms against disease exist in long-lived subjects.
People who survived with long life (centenarians, octogenarians and nonagenarians) are characterized by marked delay or escape from age-related diseases, therefore analysis of genes that modulate susceptibility to age related diseases in these populations may provide insights into the human longevity [4]. Candidate genes for longevity encode proteins engaged in different biological processes including lipoprotein metabolism and inflammatory processes [1]. Several genes by far have repeatedly been associated with human longevity, such as APOE, FOXO3A and AKT1 [19][20][21][22][23][24][25][26].
Therefore, the present study was carried out to replicate and extend previous findings by testing the association of candidate genes previously reported to contain lifespan associated polymorphisms (including rs2075650 in TOMM40 and rs405509 in APOE), so that we could investigate the association of human longevity with genetic variations in major candidate genes in the Han Chinese population.

Extraction and quantification of genomic DNA
Venous blood from each subject was drawn and collected in an EDTA tube. Genomic DNA was extracted from peripheral blood with a QIAamp DNA Blood Maxi Kit (Qiagen), and was fluorometrically quantified with Quant-iT PicoGreen reagent (Invitrogen) according to the manufacturer's protocol.

SNP genotyping
To determine if any polymorphic variant is associated with human longevity between the long-lived individuals and younger controls, we firstly carried out the case-control study of rs2075650 in TOMM40 in an initial cohort from Sichuan ( Table 1) using the ABI SNaPshot method. Secondly, 18 tag SNPs in and around the TOMM40 and APOE locus covering the common genetic variation were followed to be genotyped for extensive study in the initial cohort. Finally, associated SNPs from extensive study were genotyped in a replication cohort.
SNP analysis was performed on the ABI 3130 Genetic Analyzer (Applied Biosystems, CA, USA). In brief, the polymerase chain reactions (10 mL final volume) contained 50 ng of genomic DNA, 1 mL of each primer (10 pmol/mL), 1 mL of 10 buffer (Takara Bio Inc., Shiga, Japan), 0.8 mL of deoxyribonucleotide triphosphates (2 mmol/L; Takara Bio Inc.), 0.4 mL MgCl2 (2.5 mmol/L; Takara Bio Inc.), and 0.1 mL of ExTaq polymerase (5 U/mL; Takara Bio Inc.). The product was then processed according to the ABI SNaPshot protocol using primers designed for fluorescent dideoxy nucleotide termination. The SNPs reported in this manuscript have a genotyping success rate of .98% and accuracy as judged by random re-genotyping of 5% of the samples in the cohort by sequencing analysis.

Statistical analysis
A standard x 2 test with a 1-degree-of-freedom (df) was used to assess the Hardy-Weinberg equilibrium (HWE) and the differences of allele frequencies for each SNP between the case and control group. Odds ratios (ORs) with 95 percent confidence intervals (CIs) were assessed for the risk allele of each SNP based on a multiplicative model. For the genotypes, we tested a series of genetic models including additive, dominant/recessive for the SNPs with a p value of ,0.05 of allelic, trend test by using unconditional logistic regression with adjustment for gender. For multiple correction, the P values were corrected as combined P*5. All statistical analyses were performed by using the software SPSS 15.0 (SPSS Inc., Chicago, IL). The linkage disequilibrium (LD) block structure was examined using the program Haploview (version 4.2, Broad Institute, Cambridge, MA) [27]. The D' values and r 2 values for all pairs of SNPs were calculated, and the haplotype blocks were estimated using the program haploview. Conditional Analysis of PVRL2-TOMM40-APOE locus Conditional analyses of the PVRL2-TOMM40-APOE locus were completed at rs12978931 (PVRL2), rs519825 (PVRL2), rs395908 (PVRL2), rs2075650 (TOMM40) and rs405509 (APOE) in longlived cases and younger controls. The allelic dosage of each SNP was individually included as an additional covariate in the logistic regression model. Starting with the major allele of one conditional SNP, we performed the likelihood ratio test for independence of the other SNPs to determine the significance of the difference between the alternated SNP model versus unalternated model and see the independent effect on risk of human longevity.

Results
Case-control study of rs2075650 in TOMM40 in an initial cohort Rs2075650 in TOMM40 was firstly genotyped in an initial cohort consisted of 597 unrelated long-lived individuals and 1275 younger controls enrolled from Sichuan. The case-control study of rs2075650 in TOMM40 showed significant difference in allele frequencies (The MAF = 0.070 for cases and MAF = 0.098 for controls, OR = 1.430(95% CI 1.10-1.85), P = 0.006) between cases and controls in initial study. SNP rs2075650 is located in the intron of the TOMM40 gene at chromosome 19q13.32 but it is a strong proxy of the SNPs that define the APOE alleles [21], and has been shown a significant association longevity [14,21,23,28].
Allelic association study of SNPs in and around TOMM40 in the initial cohort Eeighteen tag SNPs in and around the TOMM40 and APOE locus covering the common genetic variation recurrently regarded as candidates for human longevity were genotyped in the initial cohort for extensive study. The TOMM40 gene located next to APOE locus, which is a very important genetic factor influencing longevity [1,23,25], as reported as a recent GWAS in Dutch Leiden Longevity Study independently confirmed the APOE longevity association [21]. Therefore, we conducted SNP genotyping in and around the TOMM40 gene to identify novel SNPs associated with human longevity in Chinese population. 18 SNPs, spanning about 45 kb (45.365-45.410 Mb), were selected from a genomic region including four genes ( Figure 1A) and genotyped for extensive study in the same cohort ( Table 2). Among them, rs405509 in APOE and three SNPs (rs12978931, rs519825 and rs395908) in PVRL2 showed significant association with human longevity (OR = 1.180(95% CI 1.02-1.37), P = 0.027 for rs405509; OR = 1.152 (95% CI 1.00-1.32), P = 0.0445 for rs12978931; OR = 0.805(95% CI 0.65-1.00), P = 0.048 for rs395908; OR = 1.283(95% CI 1.05-1.57), P = 0.01577 for rs519825), whereas another 14 SNPs did not show significant differences between cases and controls in the cohort ( Table 2). In order to not miss the real association SNPs, we have chosen these five SNPs with P value less than 0.05 but not 0.0026 (multiple correction, 0.05/19 = 0.0026) to the replication study.  Replication study of the five associated SNPs in a replication cohort Furthermore, the five associated SNPs based on extensive study including three SNPs in the PVRL2 gene, rs2075650 in TOMM40 and rs405509 in APOE were genotyped in a replication cohort ( Table 3). Our results showed the three SNPs (rs519825 in PVRL2, rs2075650 in TOMM40 and rs405509 in APOE) are significantly associated with human longevity in the replication cohort (allelic P = 0.022 for rs519825, P = 0.042 for rs405509 and P = 0.0030 for rs2075650) ( Table 3). Combined P values of the three SNPs are 0.00082, 0.0006 and 0.002, respectively; corrected P values of them are 0.04, 0.03 and 0.01.
We also tested the association of these 5 SNPs by using recessive and dominant models. Since 3 statistical models were used in the replication study, the p-value for significant observation should be 0.5/(5 SNP*3 models) = 0.0033. Only the SNP rs2075650 in the TOMM40 had a significant association with human longevity of dominant model (P = 0.0000398, OR = 0.63 (0.51-0.79), Table 4). Another two SNPs (rs12978931 and rs395908) in the PVRL2 gene showed a trend of association with human longevity of recessive model (P = 0.015 and 0.012, respectively; OR = 1.26 (95% CI 1.05-1.51) and 2.27 (95% CI 1.18-4.38), respectively; Table 4).

Conditional, Disequilibrium and Haplotype Association Analysis of PVRL2-TOMM40-APOE locus
To verify our results, we took the three significant association SNPs (PVRL2 rs519825, TOMM40 rs2075650 and APOE rs405509) and sequentially conditioned on the minor allele of each variant. Results of the conditional analysis showed that rs2075650 was significant associated SNP with longevity in the PVRL2-TOMM40-APOE locus ( Table 5). The analysis of linkage disequilibrium (LD) and haplotype block structure showed these SNPs were in the same LD block with each other ( Figure 1B). Risk haplotype AACGA generated from these five SNPs proved to be significantly different between the cases and controls (P = 3.75610 29 , Table 6).

Discussion
In the present study, we investigated genetic variations contributing to longevity and observed that rs519825 in PVRL2, a 45 kb-distance with the neighbouring gene APOE, is associated with human longevity. These three SNPs, including rs2075650 in TOMM40 and rs405509 in APOE, are located in at chromosome 19q13.32 in the APOE locus (Figure 1), which has shown consistent evidence for the association with longevity [1,21,23,25]. The strength of this study is that, by case-control study, we have replicated the previously reported association of the APOE locus with longevity [1,25] as the major locus. In addition, we found that rs519825 in PVRL2, around TOMM40 and APOE, and rs405509 in APOE were associated with human longevity in the Han Chinese population.
To date, there is abundant evidence that human longevity is heritable based on twin studies and family studies, and approximately 25% of the overall variation in human lifespan can be attributed to genetic factors [3][4][5], which become more relevant for extreme longevity [6]. The Insulin/IGF-1 pathway has been reported to regulate lifespan extension in organisms ranging from invertebrates to mammals [29]. However, no convincing causal genes have yet been identified and efforts to map loci responsible for variation in human lifespan have limited success. One approach to investigate aging and longevity is to compare frequencies of genetic variants between nonagenarians or centenarians and the general population. In humans, both candidate Table 3. Replication study for the association of the associated SNPs in a replication cohort. The genotype counts are presented as homozygote/heterozygote/wildtype; d Allelic P value has been adjusted for sex; e Data from different study cohorts (initial cohort and replication cohort) were combined using Mantel-Haenszel models with fixed effects; f P values after multiple correction (combined P*5).
doi:10.1371/journal.pone.0099580.t003 Table 4. Genotype Analysis of 5 SNPs by recessive and dominant models. gene and genome-wide genetic association approaches have been applied in an attempt to identify longevity loci. Multiple genes could mediate the aging process but would have their effects through numerous different patho-physiological processes and diseases that act as intermediate factors on the pathway to death [30,31]. Therefore, any common variation in genes associated with aging probably has a small effect. The more consistent evidence obtained by repeated observation in independent cohort studies for association with longevity has so far only been observed for three loci, the APOE locus [1,25,32], the FOXO3A locus [33][34][35][36] and the AKT1 locus [34]. APOE is involved in lipoprotein metabolism and believed to pose an effect on longevity [1]. The genetic origin of the three common variants of the human apolipoprotein E (apoE) protein, known as E2, E3 and E4, has been related to a number of age-related diseases, including Alzheimer disease, as well as to healthy aging and longevity. The APOE e4 haplotype is by far the most validated genetic variation and has repeatedly been associated with human longevity [19][20][21][22][23][24][25]37]. The best known longevity variant rs2075650 in TOMM40/APOE has been showed to be associated with human longevity, it reached irrefutable genome wide significance and replicated in the independent cohorts including German, Dutch and Danish cohorts [21,23,28]. Schupf et al found that the frequency and likelihood of carrying a G allele in rs2075650 of TOMM40 was lower among offspring in a long life family study (LLFS) in US Caucasian, compared with the likelihood in married-in controls [38]. Therefore, replication of this association in other long-lived cohorts will be needed to elucidate these results.
In our study, we applied a three-stage approach to investigate genetic variations contributing to longevity in a Han Chinese population. We found that three variations in PVRL2 showed significant association with human longevity in initial cohort. Furthermore, the association of rs519825 in PVRL2 could be replicated in another cohort, as well as the rs2075650 in TOMM40 and rs405509 in APOE. Haplotype analysis showed that the three SNPs (rs2075650 in TOMM40, rs405509 in APOE and rs519825 in PVRL2) were in the same LD block. Rs7412 and rs429358 are not significant associated with human longevity in Chinese population in this study, this finding was consistent with previous report shown rs429358 and rs7412 were not significantly different between Super-Seniors and controls [37]. The G allele in rs2075650 of TOMM40 was lower in the long-live individuals than that in the younger controls, which is consistent with previous studies [21,23,28,38]. SNP rs405509 is a A/C variation upstream of the APOE gene, which has been reported to be associated with human lifespan [1,21,23,25]. Rs2075650 is close to and in linkage disequilibrium with the rs429358, the APOE e4 allele, and the investigators suggested that rs2075650 might not have an independent effect on longevity [21,23,28]. In this study, there were lower APOEe4 (rs429358) allele frequencies and higher ApoE e2 (rs7412) allele frequencies in older group compared with younger cohort. This result suggested that ApoE e4 carriers have an increased risk for mortality, while ApoE e2 carriers are protected from diseases, which is consistent with previous studies [25,39]. However, Yu et al's study [40] showed that rs429358 and rs7412 are not in high linkage disequilibrium with each other, which suggested that they are not necessarily expected to behave the consistent associations. The controversy regarding the association of rs429358 and rs7412 with longevity is possibly due to differences in the study design and population control selection between the studies. These differences of APOE genotype may caused by different ethnic group and influenced by other genetic and environmental factors. Therefore, further study with an Table 5. Conditional analysis of the PVRL2-TOMM40-APOE locus in a replication cohort. increased sample size would be needed in order to conclusively test for variant associations in APOE. PVRL2 encodes a single-pass type I membrane glycoprotein with two Ig-like C2-type domains and an Ig-like V-type domain. This protein is one of the plasma membrane components of adherents junctions. Expression of PVRL2 has been detected in many organs [41] including brain (http://genecards.ccbb.re.kr/ cgi-bin/carddisp.pl?gene = PVRL2) and in several neuronal cell lines [42,43]. So far, the biological relationship of PVRL2, as well as TOMM40 and APOE with human longevity, have not been well evaluated. Therefore, it is meaningful to explore the role of PVRL2 in influencing human longevity. Further biological studies will be helpful to explore the joint effects of different gene variants in longevity-related pathways and determine which genes are actually involved in the human longevity pathogenesis.
In conclusion, our results replicated the association between human longevity and a known variant in TMM40 near APOE in the mainland Han Chinese population. In addition, we have found another two variants (rs519825 in PVRL2 and rs405509 in APOE) were significantly associated with human longevity. The three SNPs are localized in APOE locus, which is universally recognized as a major disease susceptibility locus for human longevity, suggesting that this locus may contribute to human longevity in the Han Chinese population. Therefore, future exact studies in much larger cohorts to identify the causal variants and functional study are very important to verify whether any of these genes are involved in the development of human longevity.
Author Contributions Table 6. The haplotype association with human longevity in the replication cohort at PVRL2-TOMM40-APOE locus.