Genome-Wide Survey of Runs of Homozygosity

Research

Original Investigation Evidence of Recessive Alzheimer Disease Loci in a Caribbean Hispanic Data Set Genome-wide Survey of Runs of Homozygosity

Mahdi Ghani, PhD; Christine Sato, MSc; Joseph H. Lee, PhD; Christiane Reitz, PhD; Danielle Moreno, BSc; Richard Mayeux, MD; Peter St George-Hyslop, MD; Ekaterina Rogaeva, PhD

Supplemental content at IMPORTANCE The search for novel Alzheimer disease (AD) genes or pathologic mutations jamaneurology.com within known AD loci is ongoing. The development of array technologies has helped to identify rare recessive mutations among long runs of homozygosity (ROHs), in which both parental alleles are identical. Caribbean Hispanics are known to have an elevated risk for AD and tend to have large families with evidence of inbreeding.

OBJECTIVE To test the hypothesis that the late-onset AD in a Caribbean Hispanic population might be explained in part by the homozygosity of unknown loci that could harbor recessive AD risk haplotypes or pathologic mutations.

DESIGN We used genome-wide array data to identify ROHs (>1 megabase) and conducted global burden and locus-specific ROH analyses.

SETTING A whole-genome case-control ROH study.

PARTICIPANTS A Caribbean Hispanic data set of 547 unrelated cases (48.8% with familial AD) and 542 controls collected from a population known to have a 3-fold higher risk of AD vs non-Hispanics in the same community. Based on a Structure program analysis, our data set consisted of African Hispanic (207 cases and 192 controls) and European Hispanic (329 cases and 326 controls) participants.

EXPOSURE Alzheimer disease risk genes.

MAIN OUTCOMES AND MEASURES We calculated the total and mean lengths of the ROHs per sample. Global burden measurements among autosomal chromosomes were investigated in cases vs controls. Pools of overlapping ROH segments (consensus regions) were identified, and the case to control ratio was calculated for each consensus region. We formulated the tested hypothesis before data collection.

RESULTS In total, we identified 17 137 autosomal regions with ROHs. The mean length of the ROH per person was significantly greater in cases vs controls (P = .0039), and this association was stronger with familial AD (P = .0005). Among the European Hispanics, a consensus region at the EXOC4 locus was significantly associated with AD even after correction for multiple testing (empirical P value 1 [EMP1], .0001; EMP2, .002; 21 AD cases vs 2 controls). Among the African Hispanic subset, the most significant but nominal association was observed for CTNNA3, a well-known AD gene candidate (EMP1, .002; 10 AD cases vs 0 controls). Author Affiliations: Author affiliations are listed at the end of this article. CONCLUSIONS AND RELEVANCE Our results show that ROHs could significantly contribute to Corresponding Authors: Ekaterina the etiology of AD. Future studies would require the analysis of larger, relatively inbred data Rogaeva, PhD, and Peter St sets that might reveal novel recessive AD genes. The next step is to conduct sequencing of George-Hyslop, MD, Tanz Centre for top significant loci in a subset of samples with overlapping ROHs. Research in Neurodegenerative Diseases, University of Toronto, Tanz Neuroscience Building, 6 Queen’s Park Crescent W, Room 227, Toronto, ON M5S 3H2, Canada JAMA Neurol. 2013;70(10):1261-1267. doi:10.1001/jamaneurol.2013.3545 ([email protected] and Published online August 26, 2013. [email protected]).

1261

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Research Original Investigation Recessive AD Loci

lzheimer disease (AD) is the most common form of de- In the present study, we investigated a late-onset AD data mentia, characterized by neurotoxic aggregates of tau set of a Caribbean Hispanic population for global burden mea- A and β-amyloid peptides in the brain.1 Known genetic surements of ROHs and the specific loci that could harbor re- factors account for only up to 40% of the risk for AD, half of cessive AD mutations or risk haplotypes. This data set was pre- which is attributable to APOE and approximately 3% to domi- viously analyzed in an SNP-based GWAS that replicated several nant mutations in APP, PSEN1, and PSEN2.2,3 Genome-wide AD genes (APOE, CLU, PICALM, and BIN1).33 We also evalu- association studies (GWASs) have identified associations be- ated the data set for the incidence of rare (>100 kilobase [kb]) tween late-onset AD (age >65 years) and single-nucleotide poly- copy number variations, but only a nominal association with morphisms (SNPs) in 9 loci (CLU, PICALM, BIN1, MS4A4/ AD was detected for a duplication on chr15q11.34 MS4A6E, CR1, CD2AP, CD33, EPHA1, and ABCA7)4-8 in which The Caribbean Hispanic population in our study origi- the pathogenic variants have not yet been revealed apart from nates mainly from the Dominican Republic or from Puerto Rico, CR1.9 In addition, a candidate gene approach discovered a re- which share a complex heritage.35 The native populations of producible association between AD and SORL1,10 and a re- both islands dwindled owing to disease, warfare, and forced cent exome study revealed AD risk variants in TREM2.11 Fi- labor after the Spanish conquered them in the 15th century,

nally, the G4C2-repeat expansion in C9orf72 has been recently and large numbers of slaves were later imported from West reported in a few patients with clinical AD12,13; however, this Africa.35-37 At present, the population of the Dominican Re- association remains to be validated.14 public is 9.8 million; that of Puerto Rico, 3.7 million.38 Intriguingly, a recent study on the concordance of AD Caribbean Hispanics are known to have a 3-fold-higher risk among parents and offspring suggested that approxi- for AD compared with non-Hispanics in the same community39 mately 90% of early-onset AD cases are likely the result of au- and tend to have large families (≥6 siblings) with evidence of tosomal recessive inheritance15; however, the p.A673V sub- inbreeding, which makes this population very appropriate for stitution in APP is the only known recessive AD mutation.16 an ROH study. A previous study found a single-founder PSEN1 Recessive inheritance has not been widely investigated for mutation (p.G206A) that was unique to Caribbean Hispanics complex traits. The lack of inbred families in most North Ameri- and responsible for 42% of early-onset AD families in this can or European data sets has made mapping of recessive loci population.35 However, the high incidence of late-onset AD in challenging; however, the development of array technolo- this population is not explained by PSEN1.35,39 Hence, we tested gies has recently helped to identify rare recessive mutations the hypothesis that the late-onset AD in the Caribbean His- among long runs of homozygosity (ROHs), in which both pa- panic population might be in part explained by the homozy- rental alleles are identical.17,18 In addition, ROHs can harbor gosity of unknown loci. imprinted chromosomes19,20 or risk haplotypes that predis- pose to a disorder in a homozygous state.21,22 Runs of homozygosity could be inherited from a common ancestor many gen- Methods erations back,23 and longer ROHs are expected in closely related individuals (identical by descent) or inbred populations.24,25 Sample Collection and Genotyping Runs of homozygosity greater than 1 megabase (Mb) are rela- This study was approved by the institutional review boards of tively frequent in the general population and could arise with- Columbia University and the University of Toronto. We exam- out inbreeding as a result of common extended haplotypes at ined a Caribbean Hispanic case-control data set.33,34 Briefly, the loci with rare recombination events.26 Small ROHs (<1 Mb) are data set was composed of 542 unrelated control subjects (68.5% too frequent (especially in inbred populations) to search for rare female) and 547 unrelated AD cases (70.7% female), includ- recessive loci, and therefore most ROH studies use a 1-Mb ing 267 cases who had at least 1 relative with AD, representing cutoff.27,28 familial AD (FAD). Family history of AD was also recorded for At present, ROHs have been associated with a risk for rheu- 47 controls (8.7%); however, none had any AD symptoms at matoid arthritis,21 Parkinson disease,29 and schizophrenia.30 the time of examination (mean age, 82 years). The mean (SD) Recently, genome-wide measurements of ROHs (>1 Mb) were age at onset of AD was 80 (8) years, and the mean (SD) age at studied in 2 outbred AD data sets of North American and Eu- the last examination of the controls was 79 (6) years. Analysis ropean origin.27,28 In both studies, the global burden analysis with the STRUCTURE program (http://pritch.bsd.uchicago.edu of ROHs did not reveal a significant association with AD. The /software.html) had previously determined that 207 AD cases only significant result in locus-specific analyses was ob- were of African ancestry and 329 cases were of European an- tained for a consensus ROH region on chr8p12 in the North cestry; 192 controls were of African ancestry and 326 controls American data set (P = .017; 40 AD cases vs 9 controls),27 but were of European ancestry (11 cases and 24 controls were of no loci survived correction for multiple testing in the Euro- unknown ancestry).33 The proportion of both subgroups was pean data set.28 Surprisingly, homozygosity mapping of a small similar in cases and controls: African Hispanics made up 37.8% data set from an isolated Arab community in Israel (Wadi Ara) of cases and 35.4% of controls, whereas European Hispanics detected that the controls were more inbred than the AD made up 60.1% of cases and 60.1% of controls. cases.31 In addition, a whole-genome study of 2 affected sib- All DNA samples were isolated from whole blood and geno- lings from a consanguineous AD family revealed several shared typed on the Illumina HumanHap 650Y array (Illumina, Inc). ROHs; however, lack of data from unaffected family mem- The genotyping success rate was greater than 99% for all bers complicated the result interpretation.32 samples, providing a reliable source to estimate the ROHs. We

1262 JAMA Neurology October 2013 Volume 70, Number 10 jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Recessive AD Loci Original Investigation Research

Table. Global Burden Analyses of ROHs

Entire Data Set AD Cases Controls (n = 547) (n = 542) P Value No. of ROHs 8225 8912 Rate, ROHs per person 15.04 16.44 .99 Total distance spanned by ROHs per person, kb 36 200 34 300 .23 Mean ROH size per person, kb 2133 1934 .0039 Familial AD Subgroup AD Cases Controls P Value (n = 267) (n = 542) No. of ROHs 4041 8912 Rate, ROHs per person 15.13 16.44 .94 Total distance spanned by ROHs per person, kb 39 600 34 300 .05 Mean ROH size per person, kb 2260 1934 .0005 African Hispanic Subgroup AD Cases Controls P Value (n = 207) (n = 192) No. of ROHs 1858 1700 Rate, ROHs per person 8.98 8.85 .39 Total distance spanned by ROHs per person, kb 21 180 17 640 .07 Mean ROH size per person, kb 2014 1875 .12 European Hispanic Subgroup AD Cases Controls P Value (n = 329) (n = 326) No. of ROHs 5949 5966 Rate, ROHs per person 18.08 18.30 .65 Total distance spanned by ROHs per person, kb 44 550 40 190 .11 Abbreviations: AD, Alzheimer disease; kb, kilobase; ROHs, runs of Mean ROH size per person, kb 2221 1994 .013 homozygosity.

included in this study only samples that previously passed of homozygosity were checked globally among the entire data stringent SNP-based quality control.33 The raw genotype data set and each subgroup using the PLINK regional test against have been submitted to the Gene Expression Omnibus, a pub- the consensus regions. The PLINK grouping function was also lic functional genomics data repository at the National Cen- applied to cluster the ROH segments if they were at least 99% ter for Biotechnology Information (GSE33528). identical by genotype. The standardized, pairwise Lewontin disequilibrium co- Identification and Analysis of ROHs efficient was used to estimate the strength of linkage disequi- Runs of homozygosity were identified using a whole-genome librium between SNPs using the Haploview software package.41 association analysis toolset (PLINK; http://pngu.mgh.harvard In addition, we estimated inbreeding (F) using the SNPs from .edu/~purcell/plink/), which has a reliable algorithm for ROH the Illumina HumanHap 650Y array to determine identity by detection.40 As previously described, we used a sliding win- descent as estimated by the genetic relationship matrix, which dow of 50 SNPs across the genome with no more than 1 het- was implemented in the Genome-wide Complex Trait erozygous SNP allowed in the window, and the minimal ROH Analysis.42,43 size was defined as 1 Mb with an SNP density of at least 1 SNP per 50 kb.27,28 We calculated the total and mean lengths of the ROHs per sample and the total number of ROHs for each Results sample. Global burden measurements among autosomal chromosomes were investigated in cases vs controls with a 1-tailed Global Burden Analysis of ROHs test (10 000 permutations) for the number of ROHs and for the In total, we identified 17 137 autosomal regions with ROHs total and mean ROH lengths per individual (as in published greater than 1 Mb (8912 in 542 controls and 8225 in 547 AD ROH studies27,28). We used a 1-tailed test because the Carib- cases), including 4041 ROHs in 267 FAD cases (Table). The bean Hispanic population has a high incidence of AD39 and is global burden measurements of rate and total size of ROHs were more suitable for the detection of risk and not of protective not associated with AD. However, AD was significantly asso- alleles. ciated with a larger mean ROH size per person: 2133 kb in cases Pools of overlapping ROH segments were identified for the vs 1934 kb in controls (P = .0039). This association was stron- entire data set, and we calculated the ratio of cases to con- ger with FAD (P = .0005). Also, the total size of ROHs per per- trols for each consensus region, defined as a shared segment son was marginally longer in FAD cases (40 Mb) vs controls (34 of greater than 100 kb (>3 SNPs) among cases and controls. Runs Mb) (P = .05), which could be the result of consanguineous mar-

jamaneurology.com JAMA Neurology October 2013 Volume 70, Number 10 1263

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Research Original Investigation Recessive AD Loci

riages in this island population. Indeed, in our cohort, we ob- .002; 21 cases vs 2 controls). This consensus region was flanked served a noticeable inbreeding comparable to second-cousin by rs7793621 and rs7792010 located at chr7:132988570- marriages (mean F = 0.020). The mean degree of inbreeding 133681101 (hg19) (Supplement [eFigure 2 and eTable 2]). In the was even higher in the 2 subgroups, which was more evident African Hispanic population, the most significant nominal as- in the European Hispanics (F = 0.055) than in the African His- sociation was observed for the consensus region overlapping panics (F = 0.046). CTNNA3 (EMP1, .004), but no region survived correction for The global burden analysis was extended to the popula- multiple testing (Supplement [eTable 3]). Consensus regions tion subgroups, as defined by the Structure program. Alzhei- of CTNNA3 and EXOC4 have several rare, potentially func- mer disease was significantly associated with larger mean tional coding variations, including missense and nonsense lengths of ROH in the European Hispanic subgroup (P = .013), SNPs, according to dbSNP137 (Supplement [eTable 4]). which also had a 2-fold greater total ROH size per individual In addition, we evaluated the samples with ROHs at EXOC4 compared with the African Hispanic subgroup (Table). for genotypic identity using the PLINK grouping function. The largest subgroup with more than 99% identical genotypes con- Association Analysis of Gene-Intersecting ROHs sisted of 11 cases (including 9 European Hispanics) and 1 con- We estimated which genes were intersected by ROHs more fre- trol of African Hispanic ancestry (Supplement [eTable 5]). quently in cases vs controls. This locus-specific analysis of the Manual inspection of the genotypes of these 12 individuals entire data set did not reveal significant findings after correc- (Supplement [eTable 6]) confirmed the shared haplotype that tion for multiple testing. The most significant nominal asso- extends from the 5′-untranslated region to exon 14 of EXOC4, ciation with AD was observed for the ROHs at chr7q33 inter- including the Sec8-exocyst domain (PF04048). The controls secting EXOC4 (NM_021807.3) (empirical P value 1 [EMP1], of European Hispanic origin (n = 326) were used to study the .0006; 28 cases vs 8 controls). In the European Hispanic popu- linkage disequilibrium structure around EXOC4. The shared lation, we observed a significant association between AD and haplotype observed in the 9 European Hispanics belongs to a EXOC4-intersecting ROHs even after correction for multiple linkage disequilibrium block that contains only EXOC4 (Supple- tests (EMP1, .0001; EMP2, .009; 21 AD cases vs 2 controls) ment [eFigure 2]). (Figure, A). The presence of EXOC4-intersecting ROHs was confirmed by manual inspection of the B-allele frequency and log R ratio plots in GenomeStudio Data Analysis Software (Illu- Discussion mina, Inc) (Supplement [eFigure 1]). Although no locus survived correction for multiple tests in the African Hispanic popu- Our results suggest the existence of recessive AD risk loci in the lation, CTNNA3 (NM_013266.2) with its nested gene (LRRTM3 Caribbean Hispanic population. We detected an association be- [NM_178011.3]) on chr10q21.3 revealed the strongest nominal tween AD and a larger genome-wide mean ROH size (P = .0039), association with AD (EMP1, .002). Runs of homozygosity– which was stronger with FAD (P = .0005). The previous stud- intersecting CTNNA3 were detected in 10 cases but not in con- ies on global burden measurements of ROHs did not identify trols (Figure, B). Based on previous copy number variation any association with AD, likely owing to the outbred nature of analysis, the homozygosity at the EXOC4 and CTNNA3 loci can- the investigated North American and European data sets.27,28 not be explained by the presence of large deletions.34 In contrast, our study had greater power to detect association We did not detect any significant findings for ROHs inter- with ROHs because we studied a more homogeneous popula- secting SORL1, TREM2, APOE, or the 9 AD loci identified by re- tion with noticeable inbreeding. Of note, total ROH size was 2 cent large GWASs.4-8 Among known AD loci, the only nomi- times longer in the European Hispanic subset than in the Afri- nal association with AD was detected in the European Hispanic can Hispanic subset, indicating closer relatedness in the Euro- subgroup for CLU (EMP1, .038; 7 cases vs 1 control). Of note, pean Hispanic population, in which we detected the associa- although AD in our cohort was strongly associated with the tion between AD and a larger mean ROH size (P = .013). APOE ε4/ε4 genotype,33 we found no difference in the fre- The association of ROHs with AD could reflect the cumu- quencies of APOE-intersecting ROHs between cases (5 indi- lative effects of multiple ROHs widely distributed through- viduals, including 3 with 3/3, 1 with 2/2, and 1 with 4/4 geno- out the genome (as in schizophrenia44) or the contribution of types) and controls (4 individuals, all with genotype 3/3). specific loci harboring recessive mutations and/or risk haplo- Hence, in a Caribbean Hispanic population, the frequency of types in a subset of patients with AD. In the gene-based ap- extended APOE haplotypes is low, as in other reported proach, the most significant result (EMP2, .009) was ob- populations.27,28 served in the European Hispanics for EXOC4, encoding a component of the exocyst complex involved in the traffick- Analyses of ROH Consensus Regions ing of the N-methyl-D-aspartate receptor.45-47 In the African In the entire data set, we identified 1415 consensus regions (de- Hispanic population, the strongest but nominal association was fined as the region shared by all individuals carrying an ROH found for CTNNA3, which is a previously reported functional at a given locus), but none of them was associated with AD af- and positional AD gene candidate (also known as VR22 or α-3- ter correction for multiple testing (Supplement [eTable 1]). How- catenin). The CTNNA3 protein is a binding partner of β-catenin, ever, in the European Hispanic population, a homozygous seg- and the complex of α/β-catenin interacts with presenilin 1, pro- ment overlapping EXOC4 was significantly associated with AD moting cadherin processing.48-52 The CTNNA3 gene is mapped even after correction for multiple testing (EMP1, .0001; EMP2, to 10q22.2 within an AD locus (OMIM 605526).53,54 Further-

1264 JAMA Neurology October 2013 Volume 70, Number 10 jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Recessive AD Loci Original Investigation Research

Figure. Runs of Homozygosity (ROHs) in a Subset of Caribbean Hispanic Samples

A RM5862 RM3615 RM4200 RX1200 RM6295 RM6559 RM5449 NY0231 RM5663 NY0768 RM4261 RX0843 NY1541 NY0649 RM4137 NY0021 RX0248 NY1783 NY2423 RM6436 RM4248 RX1130 NY1914 RefSeq Genes PLXNA4/NM_020911 EXOC4/NM_021807 BPGM/NM_199186 PLXNA4/NM_001105543 EXOC4/NM_001037126 LRGUK/NM_144648 CALD1/NM_033140 PLXNA4/NM_181775 SLC35B4/NM_032826 CALD1/NM_033139 CHCHD3/NM_017812 AKR1B10/NM_001628 AKR1B10/NM_020299 AKR1B15/NM_001080538 BPGM/NM_001724 CALD1/NM_033138 CALD1/NM_033157 CALD1/NM_004342

B RX1023 NY1579 NY1045 RM6259 NY1102 NY1614 RM0937 NY1316 NY0925 NY1174 RefSeq Genes CTNNA3/NM_013266 CTNNA3/NM_001127384 ATOH7/NM_145178 LRRTM3/NM_178011 DNAJC12/NM_021800 DNAJC12/NM_201262 SIRT1/NM_012238 SIRT1/NM_001142498 HERC4/NM_022079 HERC4/NM_015601 MYPN/NM_032578

Runs of homozygosity intersect the EXOC4 gene (A) and the CTNNA3 gene (B). Light blue bars represent ROHs in cases with Alzheimer disease; dark blue bars, ROHs in controls. The arrows represent the physical ROH consensus.

more, CTNNA3 was suggested as an APOE-dependent AD action Database (http://pid.nci.nih.gov/index.shtml). Also, both gene55-57 and reported to influence the level of β-amyloid 42 proteins are implicated in the “Arf6 trafficking events” path- in late-onset AD families.58 Finally, an association between AD way, involved in endocytosis and vesicular transport by act- and a block of CTNNA3 SNPs in our Caribbean Hispanic data ing on membrane lipid composition and actin organization.60,61 set was recently described.59 Hence, the present study fur- In the analysis of ROH consensus regions, only the EXOC4 ther supports the role of CTNNA3 in AD. locus among European Hispanics survived correction for mul- The EXOC4 and CTNNA3 proteins are part of a pathway re- tiple testing. In total, we observed 61 consensus regions nomi- sponsible for the “stabilization and expansion of the E- nally associated with AD, which could not be explained by the cadherin adherens junction,”according to the Pathway Inter- presence of large copy number variations.34 Some of these re-

jamaneurology.com JAMA Neurology October 2013 Volume 70, Number 10 1265

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Research Original Investigation Recessive AD Loci

gions are mapped near known GWAS-significant SNPs (<1 Mb). reported to be associated with β-amyloid levels in cerebrospi- For instance, the consensus region at chr8p21.1 (11 cases vs 1 nal fluid.65 control) is near the CLU gene.4,5,7,62 Two other consensus re- In summary, we found that ROHs could significantly con- gions are close to the significant SNPs detected in the African tribute to the etiology of AD in a population with noticeable American GWAS: rs10850408 (chr12q24.21) and rs2221154 inbreeding. Future studies would require the analysis of larger (chr3p24.1).63 The regions on chr3p26.1 and chr4p15.1 were data sets, including association tests with regression analysis close to significant SNPs identified in a meta-analysis of AD age incorporating the principal components of a population (eg, at onset (rs271066 and rs10517270, respectively).64 Finally, age and sex). To characterize the molecular defects underly- among the largest consensus regions (>1 Mb) on chr21q21.1, ing AD, the next step is to conduct deep sequencing of top sig- chr4p15.2, and chr8p23.3, the locus on chr21q21.1 (EMP1, .004; nificant loci in a subset of samples with overlapping ROHs, fol- 8 AD cases vs 0 controls) contains the neural cell adhesion lowed by segregation studies in families affected by potential molecule 2 gene (NCAM2). Of note, the SNPs in NCAM2 were pathologic mutations.

ARTICLE INFORMATION REFERENCES 13. Cacace R, Van Cauwenberghe C, Bettens K, Accepted for Publication: May 15, 2013. 1. Glenner GG, Wong CW. Alzheimer’s disease: et al. C9orf72 G4C2 repeat expansions in initial report of the purification and characterization Alzheimer's disease and mild cognitive impairment. Published Online: August 26, 2013. Neurobiol Aging. 2013;34(6):1712.e1-1712.e7. doi:10.1001/jamaneurol.2013.3545. of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun. 1984;120(3): doi:10.1016/j.neurobiolaging.2012.12.019. Author Affiliations: Tanz Centre for Research in 885-890. 14. Xi Z, Zinman L, Grinberg Y, et al. Investigation of Neurodegenerative Diseases, University of Toronto, 2. Younkin SG. The role of Aβ42 in Alzheimer’s c9orf72 in 4 neurodegenerative disorders. Arch Toronto, Ontario, Canada (Ghani, Sato, Moreno, Neurol. 2012;69(12):1583-1590. St George-Hyslop, Rogaeva); Taub Institute for disease. J Physiol Paris. 1998;92(3-4):289-292. Research on Alzheimer’s Disease and the Aging 3. Rogaeva E, Kawarai T, George-Hyslop PS. 15. Wingo TS, Lah JJ, Levey AI, Cutler DJ. Brain, Gertrude H. Sergievsky Center, Departments Genetic complexity of Alzheimer’s disease: Autosomal recessive causes likely in early-onset of Neurology, Psychiatry, and Medicine, College of successes and challenges. J Alzheimers Dis. Alzheimer disease. Arch Neurol. 2012;69(1):59-64. Physicians and Surgeons, Columbia University, New 2006;9(3)(suppl):381-387. 16. Di Fede G, Catania M, Morbin M, et al. A York, New York (Lee, Reitz, Mayeux); Department 4. Harold D, Abraham R, Hollingworth P, et al. recessive mutation in the APP gene with of Epidemiology, School of Public Health, Columbia Genome-wide association study identifies variants dominant-negative effect on amyloidogenesis. University, New York, New York (Lee); Department at CLU and PICALM associated with Alzheimer’s Science. 2009;323(5920):1473-1477. of Medicine, University of Toronto, Toronto, disease. Nat Genet. 2009;41(10):1088-1093. 17. Takata A, Kato M, Nakamura M, et al. Exome Ontario, Canada (St George-Hyslop, Rogaeva); 5. Lambert JC, Heath S, Even G, et al; European sequencing identifies a novel missense variant in Cambridge Institute for Medical Research and RRM2B associated with autosomal recessive Department of Clinical Neurosciences, University of Alzheimer’s Disease Initiative Investigators. Genome-wide association study identifies variants progressive external ophthalmoplegia. Genome Cambridge, Cambridge, England (St Biol. 2011;12(9):R92. doi:10.1186/gb-2011-12-9-r92. George-Hyslop). at CLU and CR1 associated with Alzheimer’s disease. Nat Genet. 2009;41(10):1094-1099. 18. Winters KA, Jiang Z, Xu W, et al. Re-assigned Author Contributions: Study concept and design: 6. Carrasquillo MM, Belbin O, Hunter TA, et al. diagnosis of D4ST1-deficient Ehlers-Danlos St George-Hyslop, Rogaeva. syndrome (adducted thumb-clubfoot syndrome) Acquisition of data: Ghani, Sato, Moreno, Mayeux, Replication of CLU, CR1, and PICALM associations with Alzheimer disease. Arch Neurol. after initial diagnosis of Marden-Walker syndrome. Rogaeva. Am J Med Genet A. 2012;158A(11):2935-2940. Analysis and interpretation of data: Ghani, Lee, 2010;67(8):961-964. Reitz, Rogaeva. 7. Naj AC, Jun G, Beecham GW, et al. Common 19. Kearney HM, Kearney JB, Conlin LK. Diagnostic Drafting of the manuscript: Ghani, Lee, Reitz, variants at MS4A4/MS4A6E, CD2AP, CD33 and implications of excessive homozygosity detected St George-Hyslop, Rogaeva. EPHA1 are associated with late-onset Alzheimer’s by SNP-based microarrays: consanguinity, Critical revision of the manuscript for important disease. Nat Genet. 2011;43(5):436-441. uniparental disomy, and recessive single-gene mutations. Clin Lab Med. 2011;31(4):595-613; ix. intellectual content: Sato, Lee, Reitz, Moreno, 8. Hollingworth P, Harold D, Sims R, et al; Mayeux, St George-Hyslop. Alzheimer’s Disease Neuroimaging Initiative; 20. Papenhausen P, Schwartz S, Risheg H, et al. Statistical analysis: Ghani, Lee, Reitz. CHARGE Consortium; EADI1 Consortium. Common UPD detection using homozygosity profiling with a Obtained funding: Ghani, Sato, Moreno, Mayeux, variants at ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 SNP genotyping microarray. Am J Med Genet A. St George-Hyslop, Rogaeva. and CD2AP are associated with Alzheimer’s disease. 2011;155A(4):757-768. Administrative, technical, and material support: Nat Genet. 2011;43(5):429-435. 21. Yang HC, Chang LC, Liang YJ, Lin CH, Wang PL. Mayeux, Rogaeva. A genome-wide homozygosity association study Study supervision: St George-Hyslop, Rogaeva. 9. Hazrati LN, Van Cauwenberghe C, Brooks PL, et al. Genetic association of CR1 with Alzheimer’s identifies runs of homozygosity associated with Conflict of Interest Disclosures: None reported. disease: a tentative disease mechanism. Neurobiol rheumatoid arthritis in the human major Funding/Support: This work was supported by Aging. 2012;33(12):2949.e5-2949.e12. histocompatibility complex. PLoS One. grants R37-AG15473 (Dr Mayeux) and P01-AG07232 doi:10.1016/j.neurobiolaging.2012.07.001. 2012;7(4):e34840. doi:10.1371 /journal.pone.0034840. from the National Institute on Aging, National 10. Rogaeva E, Meng Y, Lee JH, et al. The neuronal Institutes of Health; by the Blanchett Hooker sortilin-related receptor SORL1 is genetically 22. Casey JP, Magalhaes T, Conroy JM, et al. A novel Rockefeller Foundation; by the Charles S. associated with Alzheimer disease. Nat Genet. approach of homozygous haplotype sharing Robertson Gift from the Banbury Fund 2007;39(2):168-177. identifies candidate genes in autism spectrum (Dr Mayeux); by the W. Garfield Weston Foundation disorder. Hum Genet. 2012;131(4):565-579. 11. Guerreiro R, Wojtas A, Bras J, et al; Alzheimer (Dr Rogaeva); by the Canadian Institutes of Health 23. Li LH, Ho SF, Chen CH, et al. Long contiguous Research; by the Wellcome Trust; by the Medical Genetic Analysis Group. TREM2 variants in Alzheimer’s disease. N Engl J Med. stretches of homozygosity in the human genome. Research Council; by the National Institutes of Hum Mutat. 2006;27(11):1115-1121. Health; by the National Institute of Health 2013;368(2):117-127. Research; by the Ontario Research Fund; and by the 12. Majounie E, Abramzon Y, Renton AE, et al. 24. Gibson J, Morton NE, Collins A. Extended tracts Alzheimer Society of Ontario Repeat expansion in C9ORF72 in Alzheimer’s of homozygosity in outbred human populations. (Dr St George-Hyslop). disease. N Engl J Med. 2012;366(3):283-284. Hum Mol Genet. 2006;15(5):789-795.

1266 JAMA Neurology October 2013 Volume 70, Number 10 jamaneurology.com

Downloaded From: https://jamanetwork.com/ on 10/01/2021 Recessive AD Loci Original Investigation Research

25. Kirin M, McQuillan R, Franklin CS, Campbell H, and Caucasians in northern Manhattan. Neurology. 53. Myers A, Holmans P, Marshall H, et al. McKeigue PM, Wilson JF. Genomic runs of 2001;56(1):49-56. Susceptibility locus for Alzheimer’s disease on homozygosity record population history and 40. Howrigan DP, Simonson MA, Keller MC. chromosome 10. Science. 2000;290(5500):2304- consanguinity. PLoS One. 2010;5(11):e13996. Detecting autozygosity through runs of 2305. doi:10.1371/journal.pone.0013996. homozygosity: a comparison of three autozygosity 54. Ertekin-Taner N, Graff-Radford N, Younkin LH, 26. Curtis D, Vine AE, Knight J. Study of regions of detection algorithms. BMC Genomics. September et al. Linkage of plasma Aβ42 to a quantitative locus extended homozygosity provides a powerful 23, 2011;12:460. on chromosome 10 in late-onset Alzheimer’s method to explore haplotype structure of human 41. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: disease pedigrees. Science. 2000;290(5500): populations. Ann Hum Genet. 2008;72(pt analysis and visualization of LD and haplotype 2303-2304. 2):261-278. maps. Bioinformatics. 2005;21(2):263-265. 55. Bertram L, Mullin K, Parkinson M, et al. Is α-T 27. Nalls MA, Guerreiro RJ, Simon-Sanchez J, et al. 42. Yang J, Lee SH, Goddard ME, Visscher PM. catenin (VR22) an Alzheimer’s disease risk gene? Extended tracts of homozygosity identify novel GCTA: a tool for genome-wide complex trait J Med Genet. 2007;44(1):e63. doi:10.1136 candidate genes associated with late-onset analysis. Am J Hum Genet. 2011;88(1):76-82. /jmg.2005.039263. Alzheimer’s disease. Neurogenetics. 56. Miyashita A, Arai H, Asada T, et al; Japanese 2009;10(3):183-190. 43. Yang J, Benyamin B, McEvoy BP, et al. Common SNPs explain a large proportion of the heritability Genetic Study Consortium for Alzheimer’s Disease. 28. Sims R, Dwyer S, Harold D, et al. No evidence for human height. Nat Genet. 2010;42(7):565-569. Genetic association of CTNNA3 with late-onset that extended tracts of homozygosity are Alzheimer’s disease in females. Hum Mol Genet. associated with Alzheimer’s disease. Am J Med 44. Keller MC, Simonson MA, Ripke S, et al; 2007;16(23):2854-2869. Schizophrenia Psychiatric Genome-Wide Genet B Neuropsychiatr Genet. 2011;156B(7): 57. Martin ER, Bronson PG, Li YJ, et al. Interaction 764-771. Association Study Consortium. Runs of homozygosity implicate autozygosity as a between the α-T catenin gene (VR22) and APOE in 29. WangS,HaynesC,BaranyF,OttJ. schizophrenia risk factor. PLoS Genet. Alzheimer’s disease. J Med Genet. Genome-wide autozygosity mapping in human 2012;8(4):e1002656. doi:10.1371 2005;42(10):787-792. populations. Genet Epidemiol. 2009;33(2):172-180. /journal.pgen.1002656. 58. Ertekin-Taner N, Ronald J, Asahara H, et al. Fine 30. Kurotaki N, Tasaki S, Mishima H, et al. 45. Inoue M, Chang L, Hwang J, Chiang SH, Saltiel mapping of the α-T catenin gene to a quantitative Identification of novel schizophrenia loci by AR. The exocyst complex is required for targeting of trait locus on chromosome 10 in late-onset homozygosity mapping using DNA microarray Glut4 to the plasma membrane by insulin. Nature. Alzheimer’s disease pedigrees. Hum Mol Genet. analysis. PLoS One. 2011;6(5):e20589. 2003;422(6932):629-633. 2003;12(23):3133-3143. doi:10.1371/journal.pone.0020589. 46. Apodaca G. Opening ahead: early steps in 59. Reitz C, Conrad C, Roszkowski K, Rogers RS, 31. Sherva R, Baldwin CT, Inzelberg R, et al. lumen formation revealed. Nat Cell Biol. Mayeux R. Effect of genetic variation in LRRTM3 on Identification of novel candidate genes for 2010;12(11):1026-1028. risk of Alzheimer disease. Arch Neurol. Alzheimer’s disease by autozygosity mapping using 2012;69(7):894-900. 47. Sans N, Prybylowski K, Petralia RS, et al. NMDA genome wide SNP data. J Alzheimers Dis. 60. D’Souza-Schorey C, Chavrier P. ARF proteins: 2011;23(2):349-359. receptor trafficking through an interaction between PDZ proteins and the exocyst complex. Nat Cell roles in membrane traffic and beyond. Nat Rev Mol 32. Clarimón J, Djaldetti R, Lleó A, et al. Whole Biol. 2003;5(6):520-530. Cell Biol. 2006;7(5):347-358. genome analysis in a consanguineous family with 48. De Strooper B, Annaert W. Where Notch and 61. Klein S, Franco M, Chardin P, Luton F. Role of early onset Alzheimer’s disease. Neurobiol Aging. the Arf6 GDP/GTP cycle and Arf6 GTPase-activating 2009;30(12):1986-1991. Wnt signaling meet: the presenilin hub. JCellBiol. 2001;152(4):F17-F20. proteins in actin remodeling and intracellular 33. Lee JH, Cheng R, Barral S, et al. Identification of transport. J Biol Chem. 2006;281(18):12352-12361. novel loci for Alzheimer disease and replication of 49. Raurell I, Codina M, Casagolda D, et al. Gamma-secretase-dependent and -independent 62. Antunez C, Boada M, Gonzalez-Perez A, et al. CLU, PICALM, and BIN1 in Caribbean Hispanic The membrane-spanning 4-domains, subfamily A individuals. Arch Neurol. 2011;68(3):320-328. effects of presenilin1 on β-catenin·Tcf-4 transcriptional activity. PLoS One. (MS4A) gene cluster contains a common variant 34. Ghani M, Pinto D, Lee JH, et al. Genome-wide 2008;3(12):e4080. doi:10.1371 associated with Alzheimer’s disease. Genome Med. survey of large rare copy number variants in /journal.pone.0004080. 2011;3(5):33. doi:10.1186/gm249. Alzheimer’s disease among Caribbean Hispanics. G3 63. Logue MW, Schu M, Vardarajan BN, et al; (Bethesda). 2012;2(1):71-78. 50. Boo JH, Song H, Kim JE, Kang DE, Mook-Jung I. Accumulation of phosphorylated β-catenin Multi-institutional Research on Alzheimer Genetic 35. Athan ES, Williamson J, Ciappa A, et al. A enhances ROS-induced cell death in Epidemiology (MIRAGE) Study Group. A founder mutation in presenilin 1 causing early-onset presenilin-deficient cells. PLoS One. comprehensive genetic association study of Alzheimer disease in unrelated Caribbean Hispanic 2009;4(1):e4172. doi:10.1371 Alzheimer disease in African Americans. Arch families. JAMA. 2001;286(18):2257-2263. /journal.pone.0004172. Neurol. 2011;68(12):1569-1579. 36. Tang H, Choudhry S, Mei R, et al. Recent 51. Georgakopoulos A, Marambaud P, 64. Kamboh MI, Barmada MM, Demirci FY, et al. genetic selection in the ancestral admixture of Efthimiopoulos S, et al. Presenilin-1 forms Genome-wide association analysis of age-at-onset Puerto Ricans. Am J Hum Genet. 2007;81(3): complexes with the cadherin/catenin cell-cell in Alzheimer's disease. Mol Psychiatry. 626-633. adhesion system and is recruited to intercellular 2012;17(12):1340-1346. 37. Demographics of Puerto Rico. and synaptic contacts. Mol Cell. 1999;4(6): 65. Han MR, Schellenberg GD, Wang LS; http://en.wikipedia.org/wiki/Demographics_of 893-902. Alzheimer’s Disease Neuroimaging Initiative. _Puerto_Rico. Accessed May 2013. 52. Marambaud P, Shioi J, Serban G, et al. A Genome-wide association reveals genetic effects 38. Dominican people (Dominican Republic). presenilin-1/γ-secretase cleavage releases the on human Aβ42 and τ protein levels in http://en.wikipedia.org/wiki/Dominican_people. E-cadherin intracellular domain and regulates cerebrospinal fluids: a case control study. BMC Accessed May 2013. disassembly of adherens junctions. EMBO J. Neurol. October 2010;10:90. doi:10.1186/1471-2377-10-90. 39. Tang MX, Cross P, Andrews H, et al. Incidence 2002;21(8):1948-1956. of AD in African-Americans, Caribbean Hispanics,

jamaneurology.com JAMA Neurology October 2013 Volume 70, Number 10 1267

Downloaded From: https://jamanetwork.com/ on 10/01/2021