Association analysis of 528 intra-genic SNPs in a region of chromosome 10 linked to late onset Alzheimer's disease.код для вставкиСкачать
American Journal of Medical Genetics Part B (Neuropsychiatric Genetics) 147B:727 –731 (2008) Association Analysis of 528 Intra-Genic SNPs in a Region of Chromosome 10 Linked to Late Onset Alzheimer’s Disease A.R. Morgan,1* G. Hamilton,2 D. Turic,1 L. Jehu,1 D. Harold,1 R. Abraham,1 P. Hollingworth,1 V. Moskvina,3 C. Brayne,4 D.C. Rubinsztein,4 A. Lynch,5 B. Lawlor,5 M. Gill,5 M. O’Donovan,1 J. Powell,2 S. Lovestone,2 J. Williams,1,3 and M.J. Owen1 1 Department of Psychological Medicine, School of Medicine, Cardiff University, Cardiff, UK Department of Neuroscience, Institute of Psychiatry, Kings College, London, UK 3 Biostatistics and Bioinformatics Unit, Cardiff University, Cardiff, UK 4 Cambridge Institute for Medical Research, University of Cambridge, Cambridge, UK 5 Mercer’s Institute for Research on Aging, St James Hospital and Trinity College, Dublin, Ireland 2 Late-onset Alzheimer’s disease (LOAD) is a genetically complex neurodegenerative disorder. Currently, only the e4 allele of the Apolipoprotein E gene has been identified unequivocally as a genetic susceptibility factor for LOAD. Others remain to be found. In 2002 we observed genomewide significant evidence of linkage to a region on chromosome 10q11.23–q21.3 [Myers et al. (2002) Am J Med Genet 114:235–244]. Our objective in this study was to test every gene within the maximum LOD-1 linkage region, for association with LOAD. We obtained results for 528 SNPs from 67 genes, with an average density of 1 SNP every 10 kb within the genes. We demonstrated nominally significant association with LOAD for 4 SNPs: rs1881747 near DKK1 (P ¼ 0.011, OR ¼ 1.24), rs2279420 in ANK3 (P ¼ 0.022, OR ¼ 0.79), rs2306402 in CTNNA3 (P ¼ 0.024, OR ¼ 1.18), and rs5030882 in CXXC6 (P ¼ 0.046, OR ¼ 1.29) in 1,160 cases and 1,389 controls. These results would not survive correction for multiple testing but warrant attempts at confirmation in independent samples. ß 2007 Wiley-Liss, Inc. KEY WORDS: late-onset Alzheimer’s disease; chromosome 10; SNP; association Please cite this article as follows: Morgan AR, Hamilton G, Turic D, Jehu L, Harold D, Abraham R, Hollingworth P, Moskvina V, Brayne C, Rubinsztein DC, Lynch A, Lawlor B, Gill M, O’Donovan M, Powell J, Lovestone S, Williams J, Owen MJ. 2008. Association Analysis of 528 Intra-Genic SNPs in a Region of Chromosome 10 Linked to Late Onset Alzheimer’s Disease. Am J Med Genet Part B 147B:727–731. This article contains supplementary material, which may be viewed at the American Journal of Medical Genetics website at http://www.interscience.wiley.com/jpages/1552-4841/suppmat/ index.html. Grant sponsor: Medical Research Council, UK; Grant sponsor: Alzheimer’s Research Trust, UK; Grant sponsor: Welsh Assembly Government, UK. *Correspondence to: Dr. A.R. Morgan, Department of Psychological Medicine, Cardiff University, Heath Park, Cardiff CF14 4XN, UK. E-mail: firstname.lastname@example.org Received 10 July 2007; Accepted 18 October 2007 DOI 10.1002/ajmg.b.30670 ß 2007 Wiley-Liss, Inc. INTRODUCTION Alzheimer’s disease (AD) is a common disease affecting 1% of people up to the age of 60 and as much as 40% over the age of 90 [Bachman et al., 1992]. Rare autosomal dominant forms of familial, early-onset AD can be caused by mutations in betaamyloid precursor protein (APP) [Goate et al., 1991], presenilin 1 [Campion et al., 1995] and presenilin 2 [Sherrington et al., 1996]. However, the more common form of the disease, lateonset AD (LOAD), accounts for the majority of cases (95%) and has a complex pattern of inheritance. The e4 allele of apolipoprotein E (APOE) [Corder et al., 1993] is the only wellestablished genetic risk factor widely accepted to increase the likelihood of developing LOAD [Farrer et al., 1997]. Although present in a large proportion of LOAD cases possession of an APOE e4 allele is neither necessary nor sufficient for development of the disease, supporting the hypothesis for the existence of other genetic risk factors. Previously, we performed a two-stage genome-wide screen for linkage with LOAD in Caucasian affected sibling pairs (ASPs), which provided evidence for susceptibility loci on chromosomes 9 (MLS ¼ 1.8), 10 (MLS ¼ 3.9), 12 (MLS ¼ 1.3), and 19 (MLS ¼ 1.3) [Myers et al., 2000, 2002]. The locus on chromosome 10, 10q11.23–q21.3, was the only one to achieve genome-wide significance, and other studies have also shown evidence of linkage to chromosome 10 for AD and associated phenotypes [Bertram et al., 2000; Ertekin-Taner et al., 2000; Blacker et al., 2003; Farrer et al., 2003]. Thus, it is reasonable to hypothesize that a genetic risk factor (or factors) for LOAD may lie on chromosome 10. Our linkage peak on chromosome 10 covers a large (44 cM) region and was defined by microsatellite markers D10S1426 (59 cM) and D10S2327 (103 cM). The LOD-1 linkage region is commonly taken as a crude estimate of the 95% confidence interval for the location of a disease causing gene, assuming that the linkage reflects the presence of a single susceptibility locus [Ott, 1991]. In this study we set out to test every gene within the LOD-1 linkage region (defined as the area encompassed by all marker’s producing LOD scores between 2.83 and 3.83, flanked by markers D10S1220 (70 cM/50 Mb) and D10S1670 (87 cM/70 Mb)) for association with LOAD. We investigated SNPs located within these genes at an average density of one SNP every 10 kb. SNPs were first tested for association in pooled DNA samples, and SNPs showing evidence of association in these pools were then individually genotyped. METHODS SNP Selection All known and predicted genes in the LOD-1 region (between D10S1220 (70 cM/50 Mb) and D10S1670 (87 cM/70 Mb)) were 728 Morgan et al. downloaded from the April 2003 and July 2004 freeze of the Ensembl database using the ENSMART data-mining tool, resulting in 67 positional candidates (see supplementary Table S1). SNPs were identified within the genes using dbSNP (http://www.ncbi.nih.gov) and SNPper (http://snpper. chip.org/bio/). Validated SNPs were chosen preferentially and SNPs with a frequency of less than 5% were discarded. We initially chose SNPs to cover all the genes plus 1,000 bp of 50 flanking sequence with a SNP density of 1 SNP/8 kb. Towards the end of the study we re-assessed our coverage taking into account PCR failures and some chosen markers proving non-polymorphic. We genotyped alternative SNPs where possible and obtained genotypes for 528 SNPs at a final average density of 1 SNP/10 kb in each of the 67 candidates (with the exception of the larger genes CTNNA3 and ANK3 (as these are very large genes we did not cover large introns. However, exons of the genes were covered with 1 SNP/10 kb)). Samples DNA samples, and clinical data were collected from 1,160 individuals (70.1% females) with late-onset Alzheimer’s disease and 1,389 control subjects (62.4% females). Age at onset ranged from 59 to 95 years (mean ¼ 75.6 years, SD ¼ 6.84). All cases were Caucasian, of UK origin (parents born in the UK) and diagnosed with probable AD in accordance with the National Institute of Neurological and Communication Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Associations (NINCDS–ADRDA) clinical diagnostic criteria for AD [McKhann et al., 1984]. Controls were matched for age (mean ¼ 76.26 years, SD ¼ 6.23), gender and ethnicity. The sample consisted of individuals ascertained from both community and hospital settings in the UK collected as part of the Medical Research Council genetic resource for LOAD. For further information and full descriptive data on the sample see Morgan et al. . DNA Pooling The concentrations of all DNA samples used for pool construction were determined by the PicoGreen dsDNA Quantitation method (Molecular Probes, Eugene, OR) in a Labsystems Ascent Fluoroskan (Life Sciences International, Basingstoke, UK). Equimolar amounts of DNA were taken from each sample and used to construct DNA pools aiming for a final pool concentration of 5 ng/ml. The DNA pools were validated by genotyping six SNPs in the pools and in the individual samples comprising the pools. Estimation of differences between case and control pools was found to be extremely accurate, with a mean error of 0.007. In the first stage SNPs were typed in DNA pools that were constructed using 366 AD patients (75.9% female, mean AAO ¼ 75.5 years (SD ¼ 6.29)) and 366 age and sexmatched control subjects (mean age ¼ 76.3 years (SD ¼ 6.41), 76.1% female). SNPs showing evidence of association in these pools were then typed in a larger pooled sample consisting of 1,001 AD cases (70.7% female, mean AAO ¼ 75.9 years (SD ¼ 6.43)) and 1,001 age and sex-matched controls (70.7% female, mean age ¼ 76.3 years (SD ¼ 6.17)), and comprising the 366 cases and 366 controls from the first stage. Some SNPs bypassed the first stage and underwent pooled genotyping only in the stage 2 pools. This reflected sample availability as the study progressed and we were able to exploit the growing sample collection. Pooled Genotyping A highly accurate protocol for SNP allele frequency estimation in DNA pools based upon the SNaPshot (Applied Bio- systems, Warrington, UK) chemistry, which is an adaptation of primer extension methods, was applied [Norton et al., 2002]. Forward and Reverse primers were designed using primer 3 software (http://frodo.wi.mit.edu/cgi-bin/primer3/ primer 3_www.cgi) [Rozen and Skaletsky, 2000]. Extension primers were designed using FP PRIMER 1.0.1b software (http://m034.pc.uwcm.ac.uk/fp_primer.html). PCR was performed under standard conditions for SNaPshot, using 15 ng pooled genomic DNA and HotStar Taq DNA polymerase (Qiagen, Crawley, UK). Primer extension products were run on a 3100 DNA sequencer (Applied Biosystems) and the data were processed by using GeneScan Analysis 3.7 (Applied Biosystems). SNP allele frequencies were estimated from peak heights obtained by using Genotyper 2.5 (PE Biosystems, Cheshire, UK). Since small measurement errors in allele frequencies determined from DNA pools can result in false negative as well as false positive findings, we adopted a relaxed significance threshold (P < 0.1) to select markers for further genotyping. Individual Genotyping Polymorphisms showing any evidence of a difference (P < 0.1) in allele distribution between pooled cases and controls, in our enlarged pooled screening sample consisting of 1,001 AD cases and 1,001 age and sex-matched controls, were individually genotyped by either ampliflour or Sequenom technologies. The ampliflour genotyping assay, based on amplification in the presence of tailed allele-specific primers, a common reverse primer and universal fluorescence labeled ampliflour primers (Serologicals: www.serologicals.com), was used for some of the individual genotyping. Primers were designed by ampliflour assay architect software (www.assayarchitect.com), and standard ampliflour PCR conditions were used as described in Myakishev et al. . The amplification signals were analyzed by endpoint measurement using the ABI system (Applied Biosystems). Some SNPs were genotyped using the Sequenom (www. sequenom.de) genotyping platform, which uses the MALDITOF primer extension assay [Jurinke et al., 2002; Storm et al., 2003]. The Iplex genotyping assay was employed, which has increased plexing efficiency and flexibility for the MassARRAY system through single base primer extension with massmodified terminators. Standard conditions were followed as described in Oeth et al. . For both genotyping methods samples were aliquoted and PCR performed in 384 sample plates containing both cases and controls on each plate. Quality control measures included independent double genotyping, genotyping of duplicate samples on our sample plates and genotyping of CEPH samples and subsequent comparison with genotypes in the HapMap where available. We also ensured that all results were in HWE, and a greater than 90% genotyping success. Statistical Analysis For the pooled genotyping, each pool was run at least three times and average allele frequencies calculated for each pool and then for cases and controls. A correction factor (k) was estimated using either a known heterozygote CEPH individual or the mean peak height ratio for that SNP base change [Moskvina et al., 2005] to allow for the fact that systematic unequal representation of primer extension products may occur [Norton et al., 2002]. Estimated allele frequencies were converted to allele counts and were tested for approximate statistical significance by chi-square analysis. Association Analysis of 528 Intra-Genic SNPs Genotype and allele frequencies from individual genotyping for each SNP were analyzed by 2 3 and 2 2 chi-square respectively, to determine if there were differences between cases and controls. Polymorphisms that underwent individual genotyping were tested for deviation from Hardy–Weinberg equilibrium (HWE) in cases and controls. RESULTS Markers were initially tested for association with LOAD using allele counts estimated from pooled genotyping. DNA pooling is a simple yet efficient method that allows a large number of markers to be quickly screened and those that are most likely to show association identified [Bansal et al., 2002; Sham et al., 2002]. Individual genotyping was used to confirm the results of pooled genotyping. Our first sets of markers underwent pooled genotyping in 366 cases and 366 controls. SNPs showing evidence of association in these pools were then typed in our second larger pooled sample consisting of 1,001 AD cases and 1,001 controls. Some SNPs bypassed the first stage and underwent pooled genotyping only in the stage 2 pools (this reflected sample availability as the study progressed and we were able to exploit the growing sample collection). From the pooled genotyping we obtained results on 528 SNPs (for list of the 528 SNPs see supplementary Table S2). Eighteen SNPs were associated with LOAD in the stage 2 pools (see results in Table I). These underwent individual genotyping in our complete dataset consisting of 1,160 LOAD patients and 1,389 control subjects (see results in Table I). All individually genotyped markers were in HWE in both cases and controls. Examining the individual genotyping results, it was observed that for 14 of the markers allele frequencies were not significantly different between cases and controls. However, 4 markers continued to show some evidence of association: rs1881747 (allelic P ¼ 0.0107 OR ¼ 1.24, genotypic P ¼ 0.038), rs2279420 (allelic P ¼ 0.022 OR ¼ 0.79, genotypic P ¼ 0.028), rs2306402 (allelic P ¼ 0.024 OR ¼ 1.18, genotypic P ¼ 0.05), and rs5030882 (allelic P ¼ 0.0455 OR ¼ 1.29, genotypic P ¼ 0.07). These levels of significance do not survive adjustment for multiple testing. Two SNPs showed evidence of association with LOAD in our stage 2 pools, but failed to be successfully genotyped individually: rs524256 P ¼ 0.05494 in rs10509119 P ¼ 0.05799 in stage 2 pools. pools, DISCUSSION Pooled genotyping 2 rs10508926 rs1881747 rs4245599 rs919827 rs12354621 rs7906222 rs10509109 rs2279420 rs1050745 rs10994167 rs2393612 rs7072073 rs2248570 rs3793862 rs1629654 rs2306402 rs7091896 rs5030882 2 We tested every known and predicted gene within our LOD-1 linkage region on chromosome 10 for association with LOAD. We covered the genes at an average density of 1 SNP every 10 kb (larger genes in the region: CTNNA3 and ANK3 have not been so extensively covered. The large introns of these genes were excluded for practical reasons). Our strategy for this study was to test for associations using pooled genotyping, with progression to individual genotyping for SNPs that showed evidence of association from pooling. From the 528 SNPs investigated we have observed evidence for association with LOAD for 4 SNPs on Chr 10—rs2279420, rs5030882, rs2306402, and rs1881747. rs2279420 is located at Chr 10: 61,451,953 within the 30 UTR of ANK3. ANK3 (ankyrin 3) (MIM: 600465) is a membranecytoskeleton linker. It participates in the maintenance/targeting of ion channels and cell adhesion molecules at the nodes of Ranvier and axonal initial segments. rs5030882 is located at chr10: 70,122,477 and is a synonymous SNP in an exon of CXXC6 (CXXC finger 6) (MIM: 607790). There is not much known about the protein encoded by this gene, except that CXXC6 is fused to MLL in a case of pediatric acute myeloid leukemia containing the t(10;11) (q22;q23) [Lorsbach et al., 2003]. rs2306402 is located at chr10: 68,605,491 and is an intronic SNP in CTNNA3 (catenin, alpha 3) (MIM: 607667). CTNNA3 is predominantly expressed in heart and testis and is expressed at lower levels in brain. CTNNA3 is a strong biological candidate for AD as it inhibits WNT signaling, a pathway implicated in AD not least as it negatively regulates the tau-kinase glycogen synthase-3 [Busby et al., 2004]. Also, Alpha-T-catenin is a binding partner of beta-catenin and in turn, beta-catenin interacts with PSEN1. CTNNA3 has had both positive and negative associations reported with AD previously [Blomqvist et al., 2004; Busby et al., 2004; Cellini et al., 2005; Martin et al., 2005; Bertram et al., 2007]. rs1881747 is located at chr10: 54,003,581 between the genes DKK1 (256,159 bp away) and MBL2 (292,566 bp away). DKK1 (dickkopf homolog 1 precursor) (MIM: 605189) encodes a TABLE I. Results—Significant SNPs From Stage 2 Pooled Genotyping, and Individual Genotyping Results SNP stage 729 Individual genotyping SNP position Cases Controls P Cases Controls P OR 51,790,795 54,003,581 60,035,761 60,555,306 60,979,709 61,000,491 61,028,388 61,451,953 61,457,255 61,480,366 61,551,943 61,565,377 61,571,799 61,615,065 67,341,116 68,605,491 69,343,251 70,122,477 0.84 0.84 0.54 0.86 0.81 0.87 0.67 0.89 0.75 0.79 0.80 0.51 0.86 0.59 0.80 0.84 0.74 0.92 0.86 0.81 0.57 0.84 0.78 0.84 0.70 0.91 0.80 0.81 0.84 0.54 0.90 0.67 0.78 0.80 0.72 0.90 0.13 0.01935 0.033 0.06 0.014 0.00967 0.07 0.00636 0.0035 0.08 0.00034 0.046 0.00028 0.00000148 0.11 0.012 0.0557 0.02597 0.86 0.85 0.55 0.87 0.80 0.88 0.71 0.88 0.77 0.98 0.59 0.46 0.94 0.93 0.77 0.82 0.68 0.95 0.88 0.83 0.55 0.85 0.80 0.88 0.71 0.90 0.77 0.98 0.61 0.46 0.95 0.92 0.75 0.79 0.67 0.94 0.0782 0.0107 0.76 0.24 1 0.96 0.86 0.022 0.75 0.29 0.25 0.88 0.17 0.37 0.19 0.024 0.39 0.0455 1.19 1.24 1.02 0.90 0.99 0.99 0.99 0.79 1.02 0.79 0.89 1.01 1.19 0.91 1.11 1.18 0.93 1.29 730 Morgan et al. protein that is a member of the dickkopf family. It is a secreted protein with two cysteine rich regions and is involved in embryonic development through its inhibition of the WNT signaling pathway. MBL2 (soluble mannose-binding lectin precursor) (MIM: 154545) encodes the soluble mannosebinding protein found in serum. MBL2 recognizes mannose and N-acetylglucosamine on bacterial pathogens, and is capable of activating the classical complement pathway. rs1881747 does not appear to be in LD with SNPs in DKK1 or MBL2 using the results of the HapMap project. In conclusion, we have observed association with SNPs in ANK3, CXXC6, CTNNA3 and in the region near DKK1. However, we recognize that the strength of the effects appears small and these results will not survive adjustment for multiple testing and could thus be false positives. However, it is possible that we may be observing small genetic effects. Alternatively, the associations could reflect underlying LD with other variants in this region on Chromosome 10. As no convincing gene association was identified in our analysis of the LOD-1 region of our chromosome 10 linkage peak it is possible that the gene(s) responsible for the linkage may lie outside this region, and within the wider area of our linkage peak. However, despite our lack of highly significant findings, this study cannot fully exclude any of the genes within the LOD-1 region of our linkage peak on Chromosome 10 as potential contributors to the overall risk for the disease. It is possible that the linkage peak may be explained by many weak variants in different genes. Although we have covered the genes with SNPs every 10 kb, current methods employing the results of the HapMap project and high-density SNP genotyping incorporating an LD-based approach to SNP selection in the case of the common disease–common variant hypothesis and sequencing of genes to exclude rare pathogenic variants would sample more of the variance in the genes in the LOD-1 region of our linkage peak. Also, chromosome 10 has become more fully sequenced/annotated during recent years compared to when this study began. We have checked the most recent genome assemblies and can confirm that there are no ‘new’ genes in this region that we have not covered. However, it is possible that the locations of some of the SNPs (and genes in which the SNPs are located) may have changed since the original design of the project and so some of the genes within our LOD-1 region may not have been so extensively covered as originally planned. Recently, we have undertaken a further linkage analysis in which we amalgamated 3 large samples to give a total of 723 affected relative pairs [Hamshere et al., 2007]. Our linkage peak on chromosome 10 remains. However it has shifted slightly to the right, with a LOD-1 region spanning 17 cM from 68 to 85 cM. Our LOD-1 region from Myers et al. , and on which the study presented here is based, is 70.2– 86.2 cM. Thus we now have approximately 2 cM that we have not covered. Further analysis in our linked region is required. It is likely that within the next year detailed genotyping of genes in this region will be undertaken as part of genome wide association studies and we anticipate that such studies will provide further information that may help to identify the elusive chromosome 10 gene/genes for LOAD. ACKNOWLEDGMENTS We would like to thank the many individuals with AD and their families who participated in this study. Our research is supported by grants from the Medical Research Council and the Alzheimer’s Research Trust. The BBU is supported by the Welsh Assembly Government. REFERENCES Bachman DL, Wolf PA, Linn R, et al. 1992. Prevalence of dementia and probable senile dementia of the Alzheimer type in the Framingham Study. Neurology 42(1):115–119. Bansal A, van der Boom D, Kammerer S, et al. 2002. Association testing by DNA pooling and effective initial screen. Proc Natl Acad Sci USA 99(26):16871–16874. Bertram L, Blacker D, Mullin K, et al. 2000. Evidence for genetic linkage of Alzheimer’s Disease to chromosome 10q. Science 290:2302– 2303. Bertram L, Mullin K, Parkinson M, et al. 2007. Is alpha-T catenin (VR22) an Alzheimer’s disease risk gene? J Med Genet 44(1):e63. Blacker D, Bertram L, Saunders AJ, et al. 2003. Results of a high-resolution genome screen of 437 Alzheimer’s disease families. Hum Mol Genet 12:23–32. Blomqvist ME, Andreasen N, Bogdanovic N, et al. 2004. Genetic variation in CTNNA3 encoding alpha-3 catenin and Alzheimer’s disease. Neurosci Lett 358(3):220–222. Busby V, Goossens S, Nowotny P, et al. 2004. Alpha-T-catenin is expressed in human brain and interacts with the Wnt signalling pathway but is not responsible for linkage to chromosome 10 in Alzheimer’s disease. Neuromol Med 5(2):133–146. Campion D, Flaman JM, Brice A, et al. 1995. Mutations of the presenilin I gene in families with early-onset Alzheimer’s disease. Hum Mol Genet 4(12):2373–2377. Cellini E, Bagnoli S, Tedde A, et al. 2005. Insulin degrading enzyme and alpha-3 catenin polymorphisms in Italian patients with Alzheimer disease. Alzheimer Dis Assoc Disord 19(4):246–247. Corder EH, Saunders AM, Strittmatter WJ, et al. 1993. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer’s disease in late onset families. Science 261(5123):921–923. Ertekin-Taner N, Graff-Radford N, Younkin LH, et al. 2000. Linkage of plasma Abeta42 to a quantitative locus on chromosome 10 in late-onset Alzheimer’s disease pedigrees. Science 290(5500):2303–2304. Farrer LA, Cupples LA, Haines JL, et al. 1997. Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium. JAMA 278(16):1349–1356. Farrer LA, Bowirrat A, Friedland RP, et al. 2003. Identification of multiple loci for Alzheimer disease in a consanguineous Israeli-Arab community. Hum Mol Genet 12:415–422. Goate A, Chartier-Harlin MC, Mullan M, et al. 1991. Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature 349(6311):704–706. Hamshere ML, Holmans PA, Avramopoulos D, et al. 2007. Genome-wide linkage analysis of 723 affected relative pairs with late-onset Alzheimer’s Disease. Hum Mol Genet 16(22):2703–2712. Jurinke C, van der Boom D, Cantor CR, et al. 2002. The use of massARRAY technology for high throughput genotyping. Adv Biochem Eng Biotechnol 77:57–74. Lorsbach RB, Moore J, Mathew S, et al. 2003. TET1, a member of a novel protein family, is fused to MLL in acute myeloid leukemia containing the t(10;11)(q22;q23). Leukemia 17(3):637–641. Martin ER, Bronson PG, Li YJ, et al. 2005. Interaction between the alpha-T catenin gene (VR22) and APOE in Alzheimer’s disease. J Med Genet 42(10):787–792. McKhann G, Drachman D, Folstein M, et al. 1984. 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 34(7):939–944. Morgan AR, Turic D, Jehu L, et al. 2007. Association studies of 23 positional/functional candidate genes on chromosome 10 in lateonset Alzheimer’s disease. Am J Med Genet Part B 144B(6):762– 770. Moskvina V, Norton N, Williams N, et al. 2005. Streamlined analysis of pooled genotype data in SNP-based association studies. Genet Epidemiol 28:273–282. Myakishev MV, Khripi nY, Hu S, Hamer DH. 2001. High-throughput SNP genotyping by allele-specific PCR with universal energy-transferlabeled primers. Genome Res 11:163–169. Myers A, Holmans P, Marshall H, et al. 2000. Susceptibility locus for Alzheimer’s Disease on chromosome 10. Science 290:2304–2305. Association Analysis of 528 Intra-Genic SNPs Myers A, Wavrant De-Vrieze F, Holmans P, et al. 2002. Full genome screen for Alzheimer disease: stage II analysis. Am J Med Genet 114(2):235– 244. Norton N, Williams NM, Williams HJ, et al. 2002. Universal, robust, highly quantitative SNP allele frequency measurement in DNA pools. Hum Genet 110(5):471–847. Oeth P, Beaulie M, Park C, et al. 2005. Iplex Assay: Increased plexing efficiency and flexibility for massARRAY system through single base primer extension with mass-modified terminators. Sequenom application note. April 28 2005. pp. 1–12. Ott J. 1991. Analysis of human genetic linkage, Revised Edition. Baltimore: Johns Hopkins University Press. 731 Rozen S, Skaletsky HJ. 2000. Primer3 on the WWW for general users and for biologist programmers. In: Krawetz S, Misener S, editors. Bioinformatics methods and protocols: Methods in molecular biology. Totowa, NJ: Humana Press. pp. 365–386. Sham P, Bader JS, Craig I, et al. 2002. DNA pooling: A tool for large-scale association studies. Nat Rev Genet 3(11):862–871. Sherrington R, Froelich S, Sorbi S, et al. 1996. Alzheimer’s disease associated with mutations in presenilin 2 is rare and variably penetrant. Hum Mol Genet 5(7):985–988. Storm N, Darnhofer-Patel B, van der Boom D, et al. 2003. MALDI-TOF mass spectrometry-based SNP genotyping. Methods Mol Biol 212:241– 262.