letter nature genetics volume 19 august 1998 357 Alpha-2 macroglobulin is genetically associated with Alzheimer disease Deborah Blacker 1,4 , Marsha A. Wilcox 4 , Nan M. Laird 5 , Linda Rodes 2,3 , Steven M. Horvath 5 , Rodney C. P. Go 6 , Rodney Perry 6 , Bracie Watson, Jr. 6 , Susan S. Bassett 7 , Melvin G. McInnis 7 , Marilyn S. Albert 1,3 , Bradley T. Hyman 3 & Rudolph E. Tanzi 2,3 1 Department of Psychiatry, Massachusetts General Hospital and Harvard Medical School. 2 Genetics and Aging Unit and the 3 Department of Neurology, Massachusetts General Hospital and Harvard Medical School. 4 Department of Epidemiology, Harvard School of Public Health. 5 Department of Biostatistics, Harvard School of Public Health. 6 Department of Epidemiology, University of Alabama at Birmingham. 7 Department of Psychiatry, Johns Hopkins University Medical Institutions. Correspondence should be addressed to R.E.T. (e-mail: tanzi@helix.mgh.harvard.edu). Alpha-2-macroglobulin (α- 2 M; encoded by the gene A2M ) is a serum pan-protease inhibitor that has been implicated in Alzheimer disease (AD) based on its ability to mediate the clear- ance and degradation of Aβ, the major component of β-amyloid deposits. Analysis of a deletion in the A2M gene at the 5’ splice site of ‘exon II’ of the bait region (exon 18) revealed that inheri- tance of the deletion (A2M-2) confers increased risk for AD (M antel-Haenzel odds ratio= 3.56, P=0.001). The sibship dis- equilibrium test (SDT) also revealed a significant association between A2M and AD (P=0.00009). These values were compa- rable to those obtained for the APOE- ε4 allele in the same sam- ple, but in contrast to APOE- ε4, A2M-2 did not affect age of onset. The observed association of A2M with AD did not appear to account for the previously published linkage of AD to chromosome 12, which we were unable to confirm in this sam- ple. A2M , LRP1 (encoding the α- 2 M receptor) and the genes for two other LRP ligands, APOE and APP (encoding the amyloid β-protein precursor), have now all been genetically linked to AD, suggesting that these proteins may participate in a com- mon neuropathogenic pathw ay leading to AD. AD is a genetically heterogeneous neurodegenerative disorder characterized by global cognitive decline and distinct neu- ropathological hallmarks in the brain. Defects in three different genes, APP, PSEN1 and PSEN2, account for 30-40% of early- onset familial AD (ref. 1). In contrast, late-onset AD (LOAD) has been associated with the risk factor APOE- ε4 on chromosome 19 (refs 2-4) and LRP1, a gene encoding a neuronal receptor for both apoE and APP, the low-density lipoprotein receptor-related protein 5,6 . Family, twin and population data all suggest that other LOAD genes remain to be identified 1 . To identify novel AD genes, we employed a candidate gene strategy focusing on genes for other known LRP ligands. α- 2 M is a major LRP ligand and abundant serum pan-protease inhibitor 7,8 . In brain, α- 2 M is upregulated during injury along with LRP (ref. 9) and has been localized to senile plaques (SP) in AD (ref. 10). α- 2 M binds tightly to Aβ peptide 11,12 , the major compo- nent of β-amyloid, and attenuates fibrillogenesis and neurotoxic- ity of Aβ (refs 12,13). α- 2 M also mediates Aβ degradation 14 and clearance via endocytosis through LRP (ref. 15). In view of these findings, we tested for genetic association between A2M on chro- mosome 12p and AD in the National Institute of Mental Health (NIMH) Genetics Initiative AD sample, a large sample of affected sibpairs and other small families with AD (ref. 3). During the course of this study, chromosome 12 linkage to AD was reported as part of a genome screen 16 . As A2M maps within 30 cM of the implicated chromosome 12 markers, we also attempted to con- firm this result in our sample. We initially determined genotype counts and allele frequencies for the A2M exon 18 splice acceptor deletion 17 ( A2M-2) in probands, oldest unaffected individuals in each family and ‘strin- gent’ unaffecteds (who were at or above the age at which the member of their family with the latest onset of AD was affected; oldest in each family), stratified on individual APOE- ε4 ‘dose’. The combined genotype frequencies for the possession of one or two A2M-2 alleles were higher in the probands (29.5%) than in the oldest unaffecteds (22.9%) and stringent unaffecteds (19.3%). The A2M-2 allele frequency for probands (16.4%) was higher than those observed in unaffected individuals (oldest unaffecteds, 12.9%; stringent unaffecteds, 10.5%). The effect was greatest among the APOE- ε4 zero-dose individuals (probands, 15.0%; oldest unaffecteds, 4.3%; stringent unaffecteds, 3.7%). Based on all siblings in the 104 families with at least one affected and one unaffected sibling with A2M data available, the Mantel-Haenzel odds ratio (Table 1) for being affected as a function of carrying at least one A2M-2 allele was 3.56 (95% CI= {1.80, 7.03}; P= 0.0003). The analogous odds ratio for pos- sessing two copies of APOE- ε4 was similar in this sample (3.54; 95% CI= {1.76, 7.12}; P= 0.0004). Conditional logistic regression analyses adjusted for the effect of APOE- ε4 on risk for AD are shown (Table 1). Comparable and significant magnitudes of risk were conferred by carrying one or more A2M-2 alleles (model 1) or by APOE- ε4 homozygosity (model 2). The magnitude of risk changed little when both genes were included (model 3). There was also no evidence of an interaction between the two genes with regard to conferred risk (model 4). We obtained similar results when we restricted the analyses to stringent unaffecteds. SDT analysis, a family-based association test that does not require parental data (Horvath, S. and Laird, N., manuscript sub- mitted), revealed an association between A2M-2 and AD in the total sample (Z= 4.74; P= 0.00009) that was comparable to the Table 1 • Conditional logistic regression for the effect of A2M-2 and APOE-ε4 on risk for AD Model Variables Estimated 95% CI P -value odds ratio 1 A2M-2 carrier 3.56 (1.80, 7.03) 0.0003 2 APOE-ε4/ε4 3.54 (1.76, 7.12) 0.0004 3 A2M-2 carrier 3.45 (1.71, 6.94) 0.0005 APOE-ε4/ε4 3.45 (1.67, 7.10) 0.0008 4 A2M-2 carrier 3.40 (1.57, 7.35) 0.0018 APOE-ε4/ε4 3.39 (1.51, 7.64) 0.0032 interaction 1.07 (0.25, 4.46) 0.932 © 1998 Nature America Inc. • http://genetics.nature.com © 1998 Nature America Inc. • http://genetics.nature.com