Association Of Polymorphisms In CLU And Clusterin In People With Alzheimer’s Disease And Mild Cognitive Impairment

Alzheimer’s disease (AD) is the most common cause of dementia in the elderly. Gene and environment together play important role in the pathogenesis of the disease. Two large independent genome wide association studies (GWAS), which have identified clusterin gene (CLU) as one new genetic factor for late onset AD (LOAD) in 2009, were encouraging. The discoveries spurred us to make exploration about the heritable mechanism of AD beyond ApoE. By means of tag SNPs within the whole CLU and molecular epidemiological methods, our work aimed to study the association of polymorphisms in CLU and clusterin in Han Chinese people with sporadic AD and mild cognitive impairment (MCI).Part 1Association of polymorphisms in CLU with Alzheimer’s disease Objective:Determine the frequency of CLU polymorphisms in people with Alzheimer’s disease, and study the association of these polymorphisms with the disease.Materials and Methods:This is a retrospective case-control study. DNA was extracted from 1ml of whole blood using standard extraction method. According to age at onset, AD group was classified into two subgroups: late onset AD (LOAD) and early onset AD (EOAD); according to age, normal control (NC) group was classified into two subgroups:NC elderly and NC middle-aged. Haploview 4.2 was applied to select the tag SNPs within the whole CLU using the tagger program downloading the information of CHB (Han Chinese in Beijing) in HapMap database. Ligase detection reaction (LDR) was used for the detection of CLU tag SNPs, and amplification refractory mutation system (ARMS) was used for the detection of ApoE polymorphism. By means of Haploview 4.2 using the algorithm of expectation maximization (EM) and logistic regression statistical method, we analyzed the haplotypes and factors related with the disease.Results:Four hundred and thirty (n=430) AD patients and four hundred and sixty two(n=462) NC participants were recruited in the study, among which, there were two hundred and seventy eight (n=278) LOAD patients and one hundred and fifty two (n=152) EOAD patients, while three hundred and fifty eight(n=358) NC elderly and one hundred and four(n=104) NC middle-aged. Five tag SNPs were identified. In polymorphism of rs9331888, the frequency of CG genotype was significantly higher in AD group than that in NC (p=0.008), and the same trend in the comparison of EOAD and NC middle-aged (p=0.000). The frequency of G allele was statistically higher in EOAD group than NC Middle-aged(p=0.011). In polymorphism of rs9331942, the frequency of CT genotype was significantly lower in LOAD group than that in NC elderly group(p=0.047). For the analysis of haplotypes of three SNPs (rs2279590-rs11136000-rs9331888), the frequency of G-C-C haplotype was higher in the NC middle-aged group than that in EOAD group(p=0.019). Compared with NC middle-aged group, the frequency of A-G haplotype of rs3087554-rs9331888 was higher in EOAD group (p=0.020), while G-C haplotype lower (p=0.033). The proportion of AD patients carrying the ApoEε4 positive genotype was much larger than NC subjects(p=0.000), while much smaller for ApoEε2 positive genotype (p<0.01). The similar feature was manifested in the comparison of subgroups. No difference was found for genotype and allele in other three SNPs between patients and controls (p>0.05).With multiple logistic regression analysis, the rs9331888 CG genotype was an independent risk factor for AD. Independent of other risk such as "gender", "age", "education", "smoking", "drinking", "hypertension", "diabetes", "dislipidemia", "ApoEε4", "ApoEε2",the CG genotype was associated with a 2.010-fold increased risk for AD than CC genotype(OR=2.010,95% CI:1.356-2.977, p=0.001), with a 4.192-fold increased risk for EOAD than CC genotype(OR=4.192,95% CI:2.127-8.260, p=0.000). and GG genotype was associated with a 2.496-fold increased risk for EOAD than CC genotype (OR=2.496,95% CI:1.137-5.483, p=0.023).G allele was associated with 1.605-fold increased risk for EOAD than C allele (OR=1.605,95% CI:1.098-2.345,p=0.015).Independent of the same factors, the rs9331942 CT genotype was associated with a 0.584-fold increased risk for AD than CC genotype(OR=0.584,95% CI:0.350-0.973, p=0.039). Besides, ApoEε4 genotype was an independent risk factor for AD(OR=3.182,95% CI:2.352-4.304, p=0.000), while ApoEε2 genotype was an independent protective factor(OR=0.415,95% CI:0.263-0.655, p=0.000). For AD, education was a protective factor (OR=0.489,95% CI: 0.358-0.667,p=0.000) and smoking, a risk factor (OR=1.503,95% CI:1.033-2.1888, p=0.033).For LOAD, education was a protective factor (OR=0.439,95% CI:0.302-0.637, p=0.000) and age, a risk factor (OR1=1.110,95% CI:1.076-1.145, p=0.000). For EOAD, male was an independent risk factor. Male was associated with a 2.103-fold increased risk than female(OR=2.103,95% CI:1.196-3.697, p=0.01).CLU rs9331888 genotype may be relevant to severity of EOAD.Conclusions:CLU rs9331888 CG genotype and ApoEε4 positive genotype were independent risk factors for sporadic AD, and the rs9331888 effect was obvious for EOAD and related with its severity. CLU rs9331942 CT genotype and ApoEε2 positive genotype were independent protective factors for sporadic AD. Other three CLU SNPs were found no association with the disease.Part2Association of polymorphisms in CLU with mild cognitive impairment and cognitive functionObjective:Investigate the association of polymorphisms in CLU with mild cognitive impairment and cognitive function.Materials and Methods:This is a retrospective case-control study. Participants included patients with mild cognitive impairment (MCI) and normal controls (NC).MCI patients reached the Peterson criteria. DNA was extracted from lml of whole blood using standard extraction method. Ligase detection reaction (LDR) was used for the detection of CLU tag SNPs, and amplification refractory mutation system (ARMS) was used for the detection of ApoE polymorphism. All participants were administered a battery of Huashan Hospital neuropsychological tests. Evaluating indicators were as follows:Auditory Verbal Learning Test (AVLT), Trail Making Test B (TMTB),Complex Figure Test (CFT),Stroop Colour Words Test (SCWT), and Verbal Fluency Test (VFT). Covariate analyses were applied to evaluate the association of genotype and cognitive function according to CLU or ApoE polymorphism stratification.Results:Sixty patients (n=60) with MCI and sixty one (n=61) NC were recuited. There was no significant difference of frequency for genotype and allele in five CLU tag SNPs (p>0.05). The frequency of rs9331888 CG genotype in MCI group was higher than NC group, and there was nearly statistically difference (p=0.173). Difference was found for ApoE allele distribute between MCI and NC group (p=0.041). As the association of genotype and cognitive function, people with rs9331888 CG genotype finished AVLT long term memory worse than those with non-CG (p=0.026) for NC group in dependent of age and education. People with ApoEε4 allele performed worse in SCWT than those with non-Apo Eε4 allele (P=0.005) in NC group, and worse in AVLT long term memory worse in MCI group(p=0.063). Conclusions:CLU rs9331888 may be related to mild cognitive impairment. The cognitive function may be affected by the CLU and ApoE polymorphism.Part 3Association of serum clusterin and cognitive impairmentObjective:Explore the relationship of serum clusterin and cognitive impairment.Materials and Methods:Three groups as normal control (NC), mi Id cognitive impairment (MCI) and Alzheimer’s disease (AD) were included. ELISA method was used to evaluate the serum clusterin. Covariation analysis was applied to compare the level of serum clusterin among three groups.Results:There were thirty (n=30) NC subjects, thirty two patients (n=32) with MCI and twenty four patients (n=24) with AD. Considering "gender" "age", "hypertension", "dislipdemia", "serum glucose", "heart disease", analysis of covariation for the level of serum clusterin proved no difference anong three groups. Level of serum clusterin was higher in MCI group than NC group and there was nearly statistically difference (p=0.053). No difference was found for serum clusterin between AD of rapid progression and AD of slow progession.Conclusions:To the certain extent, serum clusterin may be related to the cognitive impairment.