alpha 2 macroglobulin and the risk of Alzheimer's disease

Citation
Rc. Dodel et al., alpha 2 macroglobulin and the risk of Alzheimer's disease, NEUROLOGY, 54(2), 2000, pp. 438-442
Citations number
35
Categorie Soggetti
Neurology,"Neurosciences & Behavoir
Journal title
NEUROLOGY
ISSN journal
00283878 → ACNP
Volume
54
Issue
2
Year of publication
2000
Pages
438 - 442
Database
ISI
SICI code
0028-3878(20000125)54:2<438:A2MATR>2.0.ZU;2-A
Abstract
Background: alpha 2 Macroglobulin is a panproteinase inhibitor that is foun d immunohistochemically in neuritic plaques, a requisite neuropathologic fe ature of AD. Recently, a pentanucleotide deletion near the 5' end of the "b ait region" of the alpha 2 macroglobulin (A2M) gene was reported to be asso ciated with AD in a large cohort of sibpairs, in which the mutation conferr ed a similar odds ratio with AD as the APOE-epsilon 4 allele for carriers o f at least one copy of the A2M gene (Mantel-Haenszel odds ratio, 3.56). Met hods: We studied three independent association samples of AD patients (n = 309) with an age range of 50 to 94 years and representative controls (n = 2 81) to characterize the allele frequency of the pentanucleotide deletion in this cohort. We detected the mutation near the 5' splice site of exon 18 u sing standard PCR and restriction fragment length polymorphism methods. The results were adjusted for age, gender, education, and APOE polymorphism. R esults: We found that the A2M gene polymorphism conferred an increased risk for AD, with an estimated Mantel-Haenszel ratio of 1.5 (95% CI 1.1 to 2.2; p = 0.025). There was no age- or gender-dependent increase in A2M gene all ele frequencies in AD patients compared with controls. The combined sample showed the expected association between AD and APOE-epsilon 4. In one of ou r three samples there was an interaction between the A2M and APOE-epsilon 4 genes, but the other two samples showed no interaction between the two ris k factors. Conclusions: Our data support an association between the A2M gen e and AD. This association is less pronounced, however, in our cohort than in the previously reported sample of sibpairs.