ApoE variant p.V236E is associated with markedly reduced risk of Alzheimer’s disease
- Christopher W Medway†1,
- Samer Abdul-Hay†1,
- Tynickwa Mims1,
- Li Ma1,
- Gina Bisceglio1,
- Fanggeng Zou1,
- Shane Pankratz2,
- Sigrid B Sando3, 4,
- Jan O Aasly3, 4,
- Maria Barcikowska5,
- Joanna Siuda1, 6,
- Zbigniew K Wszolek7,
- Owen A Ross1,
- Minerva Carrasquillo1,
- Dennis W Dickson1,
- Neill Graff-Radford1, 7,
- Ronald C Petersen8, 9,
- Nilüfer Ertekin-Taner1, 7,
- Kevin Morgan10,
- Guojun Bu†1Email author and
- Steven G Younkin†1Email author
© Medway et al.; licensee BioMed Central Ltd. 2014
Received: 21 February 2014
Accepted: 3 March 2014
Published: 10 March 2014
Recent genome-wide association studies (GWAS) of late-onset Alzheimer’s disease (LOAD) have identified single nucleotide polymorphisms (SNPs) which show significant association at the well-known APOE locus and at nineteen additional loci. Among the functional, disease-associated variants at these loci, missense variants are particularly important because they can be readily investigated in model systems to search for novel therapeutic targets. It is now possible to perform a low-cost search for these “actionable” variants by genotyping the missense variants at known LOAD loci already cataloged on the Exome Variant Server (EVS). In this proof-of-principle study designed to explore the efficacy of this approach, we analyzed three rare EVS variants in APOE, p.L28P, p.R145C and p.V236E, in our case control series of 9114 subjects. p.R145C proved to be too rare to analyze effectively. The minor allele of p.L28P, which was in complete linkage disequilibrium (D’ = 1) with the far more common APOE ϵ4 allele, showed no association with LOAD (P = 0.75) independent of the APOE ϵ4 allele. p.V236E was significantly associated with a marked reduction in risk of LOAD (P = 7.5×10−05; OR = 0.10, 0.03 to 0.45). The minor allele of p.V236E, which was in complete linkage disequilibrium (D’ = 1) with the common APOE ϵ3 allele, identifies a novel LOAD-associated haplotype (APOE ϵ3b) which is associated with decreased risk of LOAD independent of the more abundant APOE ϵ2, ϵ3 and ϵ4 haplotypes. Follow-up studies will be important to confirm the significance of this association and to better define its odds ratio. The ApoE p.V236E substitution is the first disease-associated change located in the lipid-binding, C-terminal domain of the protein. Thus our study (i) identifies a novel APOE missense variant which may profitably be studied to better understand how ApoE function may be modified to reduce risk of LOAD and (ii) indicates that analysis of protein-altering variants cataloged on the EVS can be a cost-effective way to identify actionable functional variants at recently discovered LOAD loci.
The international effort to catalog common variants [minor allele frequency (MAF) > 5%] in the human genome (HapMap Project ) paved the way for genome-wide association studies (GWAS), which have proven to be a powerful tool for understanding the genetics of complex diseases. GWAS of late-onset Alzheimer’s disease (LOAD), a genetically complex disease with an estimated 60-80% heritability , have identified common SNPs which reach genome-wide significance at the well-known APOE locus and at nineteen additional loci. The identification of these common GWAS SNPs that replicably associate with LOAD is a significant breakthrough, but it is important to recognize that these SNPs do not identify the functional disease-modifying variant(s) to which they are linked, and they do not fully account for LOAD heritability. It is now clear that at least some of this missing heritability is accounted for by rare variants with large effect size. This is well-illustrated by the recently discovered rare, LOAD-associated missense variants in the TREM2 gene [3, 4]. Importantly, this locus was not detected using the GWAS approach because the TREM2 LOAD-associated variants, which are not included in GWAS genotyping arrays, are too rare to be detected at genome-wide significance by analysis of the common GWAS SNPs to which they are linked.
Sample demographics for case-control series
Mean AAD ± SD (Years)
ϵ4+ Subjects (%)
In this proof of principle study, we used our large LOAD case-control series (Table 1) to analyze three missense variants in the APOE gene that were mined from the EVS database: rs769452 (T/c, p.L28P), rs769455 (C/t, p.R145C) and rs199768005 (T/c, p.V236E). Comparison of EVS European Americans with the control subjects in our series showed no significant difference in the MAFs for rs769452 (P = 0.27), rs769455 (P = 0.46) or rs199768005 (P = 0.075).
rs769455 (ApoE p.R145C) was successfully genotyped in 3955 AD cases and 4590 controls. With only 4 heterozygotes in the AD cases, 1 in the control group, and no homozygotes, p.R145C was too rare to analyze effectively as expected from its EVS frequency. Analysis by a Fisher’s exact test yielded an odds ratio (OR) and 95% confidence interval (95% CI) of 4.64 (0.52 to 41.56) with a p value of 0.13. In African Americans, the MAF for rs769455 on the EVS is 1.39% as compared to 0.026% in European Americans, so we evaluated this variant in our African American LOAD case control series of 168 LOAD patients and 333 non-demented control subjects. There were 9 heterozygotes in the AD cases compared to 17 in the control group and no homozygotes. A chi-square test showed no evidence of allelic association with LOAD (P = 0.91: OR = 1.05, 0.46 to 2.38), but the small series tested has relatively little statistical power as an OR of approximately 3.3 is required for 80% power to detect association at α = 0.05. Analysis in additional case-control studies is clearly needed to evaluate the association of this rare variant with LOAD.
rs769452 (ApoE p.L28P) was successfully genotyped in 2996 late-onset AD cases and 3951 control samples. There were 36 heterozygotes in the AD cases compared to 20 in the control group and no homozygotes. Analysis of rs769452 by a Fisher’s exact test showed significant (P = 1.6×10−03) association with increased risk of LOAD (OR = 2.39, 1.38 to 4.37). In African Americans (AA), the MAF for rs769452 on the EVS is 0.023% as compared to 0.17% in European Americans, so this variant was not genotyped in our small AA series.
rs199768005 (ApoE p.V236E) was successfully genotyped in 4128 late-onset AD cases and 4986 control samples. There were 2 heterozygotes in the AD cases compared to 23 in the control group and no homozygotes. Confirmatory genotyping using a custom TaqMan assay was 100% concordant. Analysis of rs199768005 by a Fisher’s exact test showed significant (P = 7.5×10−05) association with markedly reduced risk of LOAD (OR = 0.10, 0.03 to 0.45). rs199768005 was not genotyped in our small AA series, as its minor allele was never detected in the much larger set of 2203 EVS AA subjects.
APOE Haplotypes formed by three variants and their association with AD
Minor allele frequency (%)
Our results show that ApoE p.V236E occurs on the APOE ϵ3 backbone creating a rare APOE ϵ3b haplotype, which is significantly associated with LOAD independent of the APOE ϵ2, ϵ3, and ϵ4 alleles. Comparison of the 95% CI for APOE ϵ3b (OR = 0.10, 0.02 to 0.35) with that for APOE ϵ2 (OR = 0.46, 0.38 to 0.54), indicates that, in our series, the ϵ3b allele reduced risk of AD as much or more than the APOE ϵ2 allele (Table 2, Multivariate Logistic Regression). In this regard, it is worth noting that, of the 2 LOAD patients carrying p.V236E, one developed dementia at an advanced age (98 yrs, APOE ϵ3a/ϵ3b genotype) and the other, who was diagnosed at 68, also carried an ϵ4 allele (APOE ϵ3b/ϵ4 genotype), which likely counters the protection afforded by p.V236E. The 23 non-demented control carriers included 7 with ages of 64-88 years with ϵ3b/ϵ4 genotypes, 14 with ages of 68-91 with ϵ3b/ϵ3a genotypes, and 2 with ages of 68 and 92 with ϵ3b /ϵ2 genotypes. To verify the significance of the association observed in our series and to improve the OR estimate for p.V236E, replication in a similarly large series will be important, ideally a series with GWAS genotypes that can be used to adjust for the potentially confounding effect of population stratification. If APOE ϵ2 and ϵ3b act similarly, as seems likely, then analysis of the functional effects of ϵ2 as compared to the novel ϵ3b allele identified here could provide insight into the common or distinct mechanism whereby they reduce risk of LOAD.
In three previous studies [5–7], rs769452 (ApoE p.L28P) was genotyped in a total of 2630 subjects (1329 AD/1401 Control: 1118/1123 , 117/121 , 93/157 . These studies also found that ApoE p.L28P occurs on the APOE ϵ4 backbone. The risk associated with the minor allele of rs769452, which tags the rare APOE ϵ4b allele, appeared to be greater than the risk of APOE ϵ4 in two of these studies [5, 7] but less in the other study . When the results from these previous series were combined with those presented here, the OR for APOE ϵ4 vs. all other alleles was 4.31 (3.96 to 4.70) as compared to 4.04 (2.74 to 6.00) when APOE ϵ4b was compared to the same referent group. Thus the combined results from all series, like those from our series alone (Table 2), indicate that the minor allele of p.L28P does not substantially modify the risk associated with APOE ϵ4 when it is present on that haplotype. Replication in additional large series will be important to confirm this finding.
ApoE is a 299 amino acid long protein with a highly hydrophobic lipid binding domain in the C-terminal region, and a receptor binding domain in the N-terminal region. Bridged by a protease sensitive hinge region, the N- and C-terminal domains appear to interact when ApoE is delipidated, preventing lipoprotein receptor docking and internalization of unlipidated ApoE . The two missense variants that create the APOE ϵ2 (p.C112R) and APOE ϵ4 (p.R158C) alleles both alter amino acids in the N-terminal region, which may interfere with receptor binding. The missense variant (p.V236E) that creates the APOE ϵ3b allele is the first LOAD-associated variant to alter a C-terminal amino acid . The protein encoded by APOE ϵ3b has previously been described as APOE*2  because upon isoelectric focusing it migrates similarly to the APOE2 protein encoded by the APOE ϵ2 allele. Studies of individuals carrying p.V236E have found no lipoprotein abnormalities . Pathogenicity prediction using SIFT and PolyPhen-2 both suggest p.V236E is damaging, substituting a nonpolar, hydrophobic valine for the negatively charged, hydrophilic glutamic acid. Position 236 is proximal to the lipid binding domain (244-272) and interestingly it is located within a region believed to be important for ApoE oligomerization (230-243) . The substitution of a hydrophobic valine for an ionic glutamic acid is consistent with p.V236E altering the lipid binding property of ApoE, or affecting aggregation. Additionally, in light of the interaction between ApoE N- and C-terminal domains, p.V236E could alter ApoE folding and receptor binding. We are currently investigating these possibilities.
In this proof of principle study, we searched the EVS for protein-altering APOE variants with MAF > 0.1% and found just two, p.L28P (0.17%) and p.V236E (0.12%). Both were tested for association with LOAD in our large case-control series, and one (p.V236E) was significantly associated with markedly decreased risk of LOAD, independent of the APOE ϵ2, ϵ3, and ϵ4 alleles. It will now be important to determine if this same cost-effective approach can be used to identify additional LOAD-associated, protein altering variants in genes at any of the recently discovered LOAD loci that might profitably be investigated to identify novel therapeutic targets for AD.
Materials and methods
Demographic information on the LOAD patients and non-demented control subjects that were analyzed is shown in Table 1. Approval was obtained from the ethics committee or institutional review board of each institution responsible for the ascertainment and collection of samples. Written informed consent was obtained for all individuals who participated in this study.
The Mayo case-control series consists of European Americans ascertained at the Mayo Clinic Jacksonville, Mayo Clinic Rochester, and in the Mayo Clinic autopsy-confirmed samples (Autopsy in Table 1). Additional Caucasian subjects from the United States were obtained through the National Cell Repository for Alzheimer’s Disease (NCRAD in Table 1), and European Caucasian subjects were obtained from Norway  and Poland [14, 15]. All subjects in the Mayo clinical case-control series were diagnosed by a neurologist at the Mayo Clinic in Jacksonville, Florida, or Rochester, Minnesota. The neurologist confirmed a Clinical Dementia Rating score of 0 for all Jacksonville and Rochester subjects enrolled as controls; cases had diagnoses of possible or probable AD made according to NINCDS-ADRDA criteria . Clinical LOAD cases and controls in the NCRAD, Polish, and Norwegian were ascertained similarly. In the autopsy-confirmed series, all brains were evaluated by Dr. Dennis Dickson and came from the brain bank he maintains at the Mayo Clinic in Jacksonville, FL. In the Autopsy series the diagnosis of definite AD was also made according to NINCDS-ADRDA criteria. Only samples with an age-at-diagnosis (AAD) above 60 years, with sex and APOE covariates (ϵ2, ϵ3, ϵ4 alleles) available, were included in this study.
To conform to most of the literature on ApoE, our numbering of ApoE residues begins with the first amino acid that remains after removal of the 18 amino acid leader sequence. This is different from EVS numbering which begins with the first amino acid in the leader sequence . The protein encoded by the APOE ϵ3b allele, which is created by the minor allele of p.V236E (see Table 2), has previously been described as APOE*2 [10, 11] because upon isoelectric focusing it migrates similarly to the APOE2 protein encoded by APOE ϵ2 allele.
APOE missense variants resulting in p.L28P (rs769452), p.R145C (rs769455) and p.V236E (rs199768005) were genotyped using SEQUENOM’s MassArray iPLEX technology (SEQUENOM Inc, San Diego, CA, USA). SEQUENOM’s Typer Analyzer 4.0 was used to conduct off machine processing and genotype calling. Confirmatory genotyping of p.V236E was carried out using a custom TaqMan assay in an ABI PRISM 7900HT Sequence Detection System with 384-Well Block Module (Applied Biosytems, California, USA). TaqMan assays were also employed to genotype the APOE missense variants resulting in p.R158C (rs7412) and p.C112R (rs429358) in order to identify the well-known APOE ϵ2, ϵ3, and ϵ4 alleles. Cluster calling was carried out using SDS software v2.2.3 (Applied Biosytems, California, USA). All Sequenom and TaqMan probe sequences are available on request.
Analysis of control subjects using PLINK  (http://pngu.mgh.harvard.edu/~purcell/plink/), showed that all variants were in Hardy Weinberg equilibrium (P > 0.80). Allelic association was evaluated using Fisher’s exact method in PLINK. Haplotypic analysis was performed using the haplo.stats package in the R programming language (v2.14.1). Logistic regression was carried out adjusting for sex and age at diagnosis.
We would like to thank the patients and their families who participated in this research. Supported by NIH grants; R01 AG032990 (NET), P50 AG016574 (RP/NET/SGY), R01 NS080820 (NET), U01 AG046139 (NET/SGY), R01 AG18023 (SGY), R01 AG035355 (GB), R01 AG027924 (GB), R01 AG046205 (GB). CWM is funded by an Alzheimer’s Research UK US travel fellowship and the Sir Terry Pratchett Prize. JS is supported by the Stowarzyszenie na Rzecz Rozwoju Neurologii Wieku Podeszlego grant. ZKW is partially supported by the NIH/NINDS P50 NS072187, The Michael J. Fox Foundation for Parkinson’s Research, and the gift from Carl Edward Bolch, Jr., and Susan Bass Bolch.
- Tanaka T: The international HapMap project. Nature. 2003, 426: 789-796. 10.1038/nature02168.View ArticleGoogle Scholar
- Gatz M, Reynolds CA, Fratiglioni L, Johansson B, Mortimer JA, Berg S, Fiske A, Pedersen NL: Role of genes and environments for explaining Alzheimer disease. Arch Gen Psychiatry. 2006, 63: 168-174. 10.1001/archpsyc.63.2.168.View ArticlePubMedGoogle Scholar
- Jonsson T, Stefansson H, Ph.D SS, Jonsdottir I, Jonsson PV, Snaedal J, Bjornsson S, Huttenlocher J, Levey AI, Lah JJ, Rujescu D, Hampel H, Giegling I, Andreassen O a, Engedal K, Ulstein I, Djurovic S, Ibrahim-Verbaas C, Hofman A, Ikram MA, van Duijn CM, Thorsteinsdottir U, Kong A, Stefansson K: Variant of TREM2 associated with the risk of Alzheimer’s disease. N Engl J Med. 2013, 368: 107-116. 10.1056/NEJMoa1211103.PubMed CentralView ArticlePubMedGoogle Scholar
- Guerreiro R, Wojtas A, Bras J, Carrasquillo M, Rogaeva E, Majounie E, Cruchaga C, Sassi C, Kauwe JSK, Younkin S, Hazrati L, Collinge J, Pocock J, Lashley T, Williams J, Lambert J-C, Amouyel P, Goate A, Rademakers R, Morgan K, Powell J, St. George-Hyslop P, Singleton A, Hardy J: TREM2 Variants in Alzheimer’s disease. N Engl J Med. 2013, 368: 117-127. 10.1056/NEJMoa1211851.PubMed CentralView ArticlePubMedGoogle Scholar
- Kamboh MI, Aston CE, Perez-Tur J, Kokmen E, Ferrell RE, Hardy J, DeKosky ST: A novel mutation in the apolipoprotein E gene (APOE*4 Pittsburgh) is associated with the risk of late-onset Alzheimer’s disease. Neurosci Lett. 1999, 263: 129-132. 10.1016/S0304-3940(99)00129-9.View ArticlePubMedGoogle Scholar
- Baron M, Jimenez-Escrig A, Orensanz L, Simon J, Perez-Tur J: Apolipoprotein E Pittsburgh variant is not associated with the risk of late-onset Alzheimer’s disease in a Spanish population. Am J Med Genet B Neuropsychiatr Genet. 2003, 120B: 121-124. 10.1002/ajmg.b.20028.View ArticlePubMedGoogle Scholar
- Scacchi R, Gambina G, Ferrari G, Corbo RM: Screening of two mutations at exon 3 of the apolipoprotein E gene (sites 28 and 42) in a sample of patients with sporadic late-onset Alzheimer’s disease. Neurobiol Aging. 2003, 24: 339-343. 10.1016/S0197-4580(02)00089-1.View ArticlePubMedGoogle Scholar
- Chen J, Li Q, Wang J: Topology of human apolipoprotein E3 uniquely regulates its diverse biological functions. Proc Natl Acad Sci USA. 2011, 108: 14813-14818. 10.1073/pnas.1106420108.PubMed CentralView ArticlePubMedGoogle Scholar
- Bu G: Apolipoprotein E and its receptors in Alzheimer’s disease: pathways, pathogenesis and therapy. Nat Rev Neurosci. 2009, 10: 333-344. 10.1038/nrn2620.PubMed CentralView ArticlePubMedGoogle Scholar
- van den Maagdenberg a M, Weng W, de Bruijn IH, de Knijff P, Funke H, Smelt a H, Gevers Leuven J a, van’t Hooft FM, Assmann G, Hofker MH: Characterization of five new mutants in the carboxyl-terminal domain of human apolipoprotein E: no cosegregation with severe hyperlipidemia. Am J Hum Genet. 1993, 52: 937-946.PubMed CentralPubMedGoogle Scholar
- Zhao SP, van den Maagdenberg AM, Vroom TF, van ’t Hooft FM, Gevers Leuvens JA, Havekes LM, Frants RR, Van der Laarse A, Smelt AH: Lipoprotein profiles in a family with two mutants of apolipoprotein E: possible association with hypertriglyceridaemia but not with dysbetalipoproteinaemia. Clin Sci (Lond). 1994, 86: 323-329.View ArticleGoogle Scholar
- Frieden C, Garai K: Structural differences between apoE3 and apoE4 may be useful in developing therapeutic agents for Alzheimer’s disease. Proc Natl Acad Sci USA. 2012, 109: 8913-8918. 10.1073/pnas.1207022109.PubMed CentralView ArticlePubMedGoogle Scholar
- Sando SB, Melquist S, Cannon A, Hutton ML, Sletvold O, Saltvedt I, White LR, Lydersen S, Aasly JO: APOE epsilon 4 lowers age at onset and is a high risk factor for Alzheimer’s disease; a case control study from central Norway. BMC Neurol. 2008, 8: 9-10.1186/1471-2377-8-9.PubMed CentralView ArticlePubMedGoogle Scholar
- Maruszak A, Safranow K, Branicki W, Gawęda-Walerych K, Pośpiech E, Gabryelewicz T, Canter JA, Barcikowska M, Zekanowski C: The impact of mitochondrial and nuclear DNA variants on late-onset Alzheimer’s disease risk. J Alzheimers Dis. 2011, 27: 197-210.PubMedGoogle Scholar
- Klimkowicz-Mrowiec A, Marona M, Wołkow P, Maruszak A, Styczynska M, Barcikowska M, Zekanowski C, Szczudlik A, Slowik A: Interleukin-1 gene -511 CT polymorphism and the risk of Alzheimer’s disease in a Polish population. Dement Geriatr Cogn Disord. 2009, 28: 461-464. 10.1159/000259460.View ArticlePubMedGoogle Scholar
- McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM: Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer’s disease. Neurology. 1984, 34: 939-944. 10.1212/WNL.34.7.939.View ArticlePubMedGoogle Scholar
- Exome variant server, NHLBI GO exome sequencing project (ESP), Seattle, WA. [URL:http://evs.gs.washington.edu/EVS (October 2013)]
- Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, Maller J, Sklar P, de Bakker PIW, Daly MJ, Sham PC: PLINK: a tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet. 2007, 81: 559-575. 10.1086/519795.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.