Abstract
The gene for insulin‐degrading enzyme (IDE), which is located at chromosome 10q24, has been previously proposed as a candidate gene for late‐onset Alzheimer's disease (AD) based on its ability to degrade amyloid β‐protein. Genotyping of single nucleotide polymorphisms (SNPs) in the IDE gene in Finnish patients with AD and controls revealed SNPs rs4646953 and rs4646955 to be associated with AD, conferring an approximately two‐fold increased risk. Single locus findings were corroborated by the results obtained from haplotype analyses. This suggests that genetic alterations in or near the IDE gene may increase the risk for developing AD.
Keywords: Alzheimer's disease, insulin‐degrading enzyme (IDE), risk gene, single nucleotide polymorphism (SNP)
Alzheimer's disease (AD) is characterised by the progressive and severe accumulation of amyloid β‐protein (Aβ) in the brain, and insulin‐degrading enzyme (product of the IDE gene) is a possible candidate enzyme responsible for the degradation and clearance of Aβ.1,2IDE is located at chromosome 10q24, close to a region for which linkage and association with late‐onset AD has been previously described.3 In addition, genetic linkage was simultaneously reported on chromosome 10 in a region roughly 30 Mb from IDE.4,5 Numerous subsequent case–control studies have reported genetic association between IDE and AD, but others have failed to observe significant effects (http://www.alzgene.org).
Because IDE is an excellent candidate on both positional and functional grounds, we investigated whether single nucleotide polymorphisms (SNPs) in the IDE gene or in nearby regions were associated with AD in the eastern Finnish population. Two of the six SNPs tested showed significantly different allele and genotype distributions between AD and control cohorts. These findings were corroborated by the results of haplotype analyses.
Methods
All patients fulfilled the National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer's Disease and Related Disorders Association (NINCDS‐ADRDA) clinical criteria for probable AD without significant vascular contribution for dementia.6 The study group consisted of 370 patients with AD (mean (SD) age at onset 72 (7) years; range 43 to 90 years; 69% women) and 454 healthy control subjects (70 (5) years; range 60 to 87 years; 61% women). In the AD cohort, 12% and 25% of patients had history of diabetes and hypertension, respectively. In total, 46% of patients with AD had a positive family history of AD, but did not show conclusive evidence of autosomal dominant transmission.7 Controls had no signs of dementia as determined by interview and neuropsychological testing. There were 61 patients with AD (16%) and 116 controls (26%) with an age of onset or age at examination of ⩽65 years.
Five non‐coding SNPs in IDE and one SNP in the gene encoding haematopoetically expressed homeobox protein (HHEX) were selected for genotyping, based on previous studies.8,9 Genotyping of SNPs were performed using either a mini‐sequencing technique (SNaPshot Multiplex Ready Reaction Mix; Applied Biosystems, Foster City, California, USA) or by high‐efficiency fluorescence polarisation detected single‐base extension (rs1544210 and rs2251101) (Criterion Analyst AD High‐Throughput Fluorescence Detection System; Molecular Devices, Sunnyvale, California, USA). Apolipoprotein E (APOE) genotyping was performed as previously.7
Statistical analysis
Single locus allele, genotype and logistic regression analyses were carried out using SPSS V.11.0 (SPSS Inc., Chicago, IL, USA). Pairwise linkage disequilibrium (LD), Hardy‐Weinberg equilibrium, haplotype estimation and haplotype block structure analyses were performed with Haploview V. 3.2 (http://www.broad.mit.edu/mpg/haploview/). Statistical significance was set at p<0.05.
Results
The APOE e4 allele was significantly over‐represented among the 370 patients with AD compared with the 454 control subjects (p<0.001; OR 4.8; 95% CI 3.8 to 6.1) (table 1). The minor allele frequency of the IDE intron 20 SNP (12973709) was ⩽0.01 in both the AD and control cohorts, which led us to exclude this SNP from further analysis.
Table 1 Allele and genotype frequencies of IDE and HHEX SNPs .
| NCBI rs number and location (kb)* | Allele | Allele frequency | p Value for allele‡ | Genotype | Genotype frequency | p Value for genotype‡ | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| Control (n = 908) | AD (n = 740) | Control (n = 454) | AD (n = 370) | |||||||
| rs1544210† HHEX (0) | A | 0.566 | 0.535 | 0.21 | AA | 0.322 | 0.300 | 0.32 | ||
| G | 0.434 | 0.465 | AG | 0.476 | 0.446 | |||||
| GG | 0.192 | 0.232 | ||||||||
| rs3758505 IDE 5′‐UTR (153) | G | 0.052 | 0.053 | 0.93 | GG | 0.004 | — | 0.48 | ||
| T | 0.948 | 0.947 | GT | 0.095 | 0.105 | |||||
| TT | 0.901 | 0.895 | ||||||||
| rs4646953 IDE 5′‐UTR (154) | C | 0.338 | 0.284 | 0.02 (0.10) | CC | 0.126 | 0.076 | 0.04 (0.20) | ||
| T | 0.662 | 0.716 | CT | 0.425 | 0.416 | |||||
| TT | 0.449 | 0.508 | ||||||||
| rs4646955 IDE Intron 3 (194) | C | 0.380 | 0.327 | 0.03 (0.15) | CC | 0.152 | 0.095 | 0.04 (0.20) | ||
| T | 0.620 | 0.673 | CT | 0.456 | 0.465 | |||||
| TT | 0.392 | 0.441 | ||||||||
| rs2251101 IDE 3′UTR (276) | C | 0.177 | 0.174 | 0.87 | CC | 0.030 | 0.027 | 0.87 | ||
| T | 0.823 | 0.826 | CT | 0.289 | 0.295 | |||||
| TT | 0.678 | 0.678 | ||||||||
*Distance from HHEX SNP rs1544210. All SNPs were in Hardy‐Weinberg equilibrium in both cases and controls (p>0.1).
†Patients with AD (n = 362), controls (n = 449).
‡Allele and genotype frequencies were compared using two‐sided Pearson's χ2 and Fisher's exact tests, respectively. p Values were correct for multiple testing using Bonferroni correction (in parentheses).
AD, Alzheimer's disease; SNP, single nucleotide polymorphism.
SNPs rs3758505, rs4646953 and rs4646955, located within the same 40 kb region in the 5′ half of IDE, were in strong LD (D′ values ⩾0.994), and originated from the same haplotype block. Comparison of the genotype and allele distribution of SNPs between the AD and control cohorts showed that rs4646953 and rs4646955 showed nominal evidence of association with AD (table 1). When the sample was stratified based on onset age using 65 years as the cut‐off point, nominal genotype and allele association of rs4646953 and rs4646955 were observed for patients with late‐onset, but not those with early‐onset AD (data not shown). SNPs rs4646953 and rs4646955 conferred an approximately two‐fold increased risk for AD in logistic regression analyses in both whole and late‐onset cohorts (table 2). Stratification according to gender and APOE status did not reveal any major differences between subgroups for either of these SNPs.
Table 2 Age, gender and APOE status adjusted odds ratios obtained from univariant and multivariant logistic regression for IDE and HHEX SNPs.
| SNP | OR (95% CI) univariate | OR (95% CI) adjusted for age, gender and APOE | ||
|---|---|---|---|---|
| Whole population | >65 years | Whole population | >65 years | |
| rs1544210 AA+AG vs GG | 1.09 (0.81 to 1.47) | 1.25 (0.89 to 1.74) | 1.02 (0.72 to 1.44) | 1.19 (0.81 to 1.76) |
| rs3758505 GG+GT vs TT | 0.93 (0.59 to 1.47) | 0.88 (0.53 to 1.45) | 1.01 (0.60 to 1.70) | 0.96 (0.53 to 1.75) |
| rs4646953 CT+TT vs CC | 1.75 (1.09 to 2.82)* | 2.21 (1.29 to 3.80)** | 2.08 (1.20 to 3.60)** | 3.00 (1.58 to 5.69)* |
| rs4646955 CT+TT vs CC | 1.72 (1.11 to 2.64)* | 2.08 (1.29 to 3.36)** | 1.95 (1.18 to 3.23)** | 2.58 (1.45 to 4.57)** |
| rs2251101 CT+CC vs TT | 1.00 (0.75 to 1.34) | 1.00 (0.72 to 1.39) | 1.10 (0.78 to 1.54) | 1.11 (0.75 to 1.63) |
SNP, single nucleotide polymorphism.
*p<0.05; **p<0.01.
Using SNPs rs3758505, rs4646953 and rs4646955 (5′ haplotype block), four haplotypes were identified (table 3). Global association test with TTT, TCC, and pooled (frequencies ⩽0.05) haplotypes revealed a borderline association with AD (p = 0.06). Assessment of individual haplotype distributions showed that the TTT may be significantly over‐represented in the AD cohort (nominal p = 0.03). Conversely, haplotype TCC was under‐represented among patients with AD (nominal p = 0.02). The frequencies of the TCC and TTT haplotypes were similar in the patients with early‐onset and those with late‐onset AD compared with the whole cohort, but the distribution of these haplotypes was significantly different from patients with late‐onset AD compared with age‐matched controls (TCC nominal p = 0.03 and TTT nominal p = 0.03).
Table 3 Haplotype frequencies of IDE SNPs rs3758505, rs4646953 and rs4646955 in patients with AD and controls.
| IDE haplotypes for SNPs rs3758505, rs4646953 and rs4646955 | Control (n = 908 alleles) | AD (n = 740 alleles) | p Value | |
|---|---|---|---|---|
| TCC | 0.34 | 0.28 | 0.02 | |
| TTT | 0.57 | 0.62 | 0.03 | |
| GTT | 0.05 | 0.05 | 0.93 | |
| TTC | 0.04 | 0.04 | 0.94 | |
| Overall | 0.06 | |||
AD, Alzheimer's disease; SNP, single nucleotide polymorphism.
Discussion
Our data from both single locus and haplotype association analyses lend further support to IDE being a susceptibility gene for AD. The approximately two‐fold risk increase observed for rs4646953 and rs4646955 in an eastern Finnish population appeared to be independent of gender, APOE status and age. Several case‐control studies have reported that genetic variants in IDE increase the risk for AD (http://www.alzgene.org). To date, there appear to be more negative than positive findings, and at the time of writing (20 April 2007) none of the continuously updated meta‐analyses conducted as part of the AlzGene database show significant risk effects for any of the IDE SNPs. This situation is not uncommon in genetically heterogeneous and complex diseases, for which study populations often stem from different genetic backgrounds and ethnic groups. The control allele and genotype frequencies of all the SNPs studied here differed from those previously observed in Caucasians, probably owing to the isolated genetic background of the Finnish population. Interestingly though, our study is in line with analyses from a Chinese case–control sample, which also found significant association between rs4646953 and AD risk, albeit with the opposite allele.10 Moreover, we observed the IDE risk effect among the unstratified cohort and not in the APOE e4 carriers, as observed in the Chinese study. Taken together, our findings and those of previous reports suggest the presence of an AD risk‐modifying variant in the IDE gene. More studies are warranted to investigate these associations and their functional implications further, and to evaluate the potential of using IDE as a therapeutic tool in the prevention and treatment of AD.
Acknowledgements
Financial support for this project was provided by the Health Research Council of the Academy of Finland, EVO grant 5772708 of Kuopio University Hospital and the Nordic Centre of Excellence of Neurodegeneration, as well as grants from the NIMH, NIA and Alzheimer Association. Mikko Hiltunen and Lars Bertram were supported by the Extendicare Foundation. Lars Bertram was also supported by the Deutsche Forschungsgemeinschaft, the Harvard Center for Neurodegeneration and Repair, and the National Alliance for Research on Schizophrenia and Depression. We thank Ms Marjo Laitinen and Ms Petra Mäkinen for technical help.
Abbreviations
AD - Alzheimer's disease
APOE - apolipoprotein E
IDE - insulin‐degrading enzyme
LD - linkage disequilibrium
SNP - single nucleotide polymorphism
Footnotes
Competing interests: None declared.
Ethics approval: The Ethics Committee of Kuopio University Hospital and Kuopio University approved the study.
References
- 1.Qiu W Q, Walsh D M, Ye Z, Vekrellis K, Zhang J, Podlisny M B, Rosner M R, Safavi A, Hersh L B, Selkoe D J. Insulin‐degrading enzyme regulates extracellular levels of amyloid beta‐protein by degradation. J Biol Chem 199827332730–32738. [DOI] [PubMed] [Google Scholar]
- 2.Farris W, Mansourian S, Leissring M A, Eckman E A, Bertram L, Eckman C B, Tanzi R E, Selkoe D J. Partial loss‐of‐function mutations in insulin‐degrading enzyme that induce diabetes also impair degradation of amyloid beta‐protein. Am J Pathol 20041641425–1434. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Bertram L, Blacker D, Mullin K, Keeney D, Jones J, Basu S, Yhu S, McInnis M G, Go R C, Vekrellis K, Selkoe D J, Saunders A J, Tanzi R E. Evidence for genetic linkage of Alzheimer's disease to chromosome 10q. Science 20002902302–2303. [DOI] [PubMed] [Google Scholar]
- 4.Ertekin‐Taner N, Graff‐Radford N, Younkin L H, Eckman C, Baker M, Adamson J, Ronald J, Blangero J, Hutton M, Younkin S G. Linkage of plasma Abeta42 to a quantitative locus on chromosome 10 in late‐onset Alzheimer's disease pedigrees. Science 20002902303–2304. [DOI] [PubMed] [Google Scholar]
- 5.Myers A, Holmans P, Marshall H, Kwon J, Meyer D, Ramic D, Shears S, Booth J, DeVrieze F W, Crook R, Hamshere M, Abraham R, Tunstall N, Rice F, Carty S, Lillystone S, Kehoe P, Rudrasingham V, Jones L, Lovestone S, Perez‐Tur J, Williams J, Owen M J, Hardy J, Goate A M. Susceptibility locus for Alzheimer's disease on chromosome 10. Science 20002902304–2305. [DOI] [PubMed] [Google Scholar]
- 6.McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan E M. 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 198434939–944. [DOI] [PubMed] [Google Scholar]
- 7.Lehtovirta M, Soininen H, Helisalmi S, Mannermaa A, Helkala E L, Hartikainen P, Hanninen T, Ryynanen M, Riekkinen P J. Clinical and neuropsychological characteristics in familial and sporadic Alzheimer's disease: relation to apolipoprotein E polymorphism. Neurology 199646413–419. [DOI] [PubMed] [Google Scholar]
- 8.Bertram L, Saunders A J, Mullin K, Sampson A, Moscarillo T, Tanzi R E. Further assessment of novel Alzheimer's disease loci on chromosome 10 and 9. Neurobiology of Aging 200223S1206 [Google Scholar]
- 9.Prince J A, Feuk L, Gu H F, Johansson B, Gatz M, Blennow K, Brookes A J. Genetic variation in a haplotype block spanning IDE influences Alzheimer disease. Hum Mutat 200322363–371. [DOI] [PubMed] [Google Scholar]
- 10.Bian L, Yang J D, Guo T W, Sun Y, Duan S W, Chen W Y, Pan Y X, Feng G Y, He L. Insulin‐degrading enzyme and Alzheimer disease: a genetic association study in the Han Chinese. Neurology 200463241–245. [DOI] [PubMed] [Google Scholar]