Genetic association study between α 1-antichymotrypsin polymorphism and Alzheimer disease in Chinese Han population

INTRODUCTION

Alzheimer disease (AD) is a neurodegenerative disorder characterized by abnormal deposition of extracellular senile amyloid plaques in the brain. At least four genes have been found to be involved in the genetic etiology of AD. Various mutations in the amyloid precursor protein (APP) gene, presenilin 1 (PS1) gene, and presenilin 2 (PS2) gene could segregate with early-onset familial AD, but they account for only a small proportion of AD. The apolipoprotein E (ApoE) ϵ4 allele has been identified as a major susceptibility marker for sporadic AD, responsible for about 50% of AD cases [Corder et al., 1993; Strittmatter et al., 1993], but it is neither necessary nor sufficient for the expression of AD. Therefore, some other environmental or genetic factors must be involved in AD. The gene encoding α 1-antichymotrypsin (ACT) is a good natural candidate gene because ACT is one of several proteins that have been detected in amyloid deposit of AD brains and was found to be able to promote the polymerization of A_β(1–42) peptide into amyloid filaments [Eriksson et al., 1995].

Kamboh et al. [ 1995] have described a polymorphism in the signal peptide of ACT defined by alleles A and T. They found that the ACT A allele is overpresented in AD cases, behaving as a modifier gene altering the AD risk associated with ApoE ϵ4 allele, furthermore, the ACT AA genotype is strongly associated with ApoE ϵ4 in AD cases. This was confirmed in some studies [Deskosky et al., 1997; Muramatsu et al., 1996] but not in others [Hains et al., 1996; Helisalmi et al., 1997; Muller et al., 1996; Murphy et al., 1997].

In the present study, we investigate the ACT polymorphism and ApoE polymorphism in 125 AD cases and 141 healthy controls to verify whether the common polymorphism of ACT gene is associated with AD in Chinese Han population.

MATERIALS AND METHODS

AD patients were recruited from in-patient clinics in the Shanghai Mental Health Center. All underwent complete medical, psychiatric, neurological, and neuropsychologic assessments, and met the criteria of DSM-IIR and NINCDS-ADRDA for probable AD. Of 125 AD cases, 61 were female and 64 were male. Mean age at onset was 70.6 years (SD = 8.6, range = 50–89). One hundred forty normal controls were unrelated persons selected randomly from community volunteers with a wide age range of 30–81 (mean age = 52.0 ± 18.6 years; male = 61, female = 80), all without memory complaints and family history of AD or other mental disorders.

Genomic DNA was extracted from peripheral lymphocytes according to standard procedures and was genotyped by polymerase chain reaction methods and restriction fragment length polymorphism typing as previously described for ApoE [Corder et al., 1993] and ACT [Kamboh et al., 1994]. Statistical analysis was performed using χ² tests.

RESULTS

Table I presents the ACT genotype and allele frequencies in the 125 AD patients and 141 healthy control subjects. The distributions of ACT genotypes in two groups were both in Hardy-Weinberg equilibrium. Unlike Kamboh et al. [1995] we found no differences of three ACT genotype frequencies between AD cases and controls. The odds ratio (OR) for developing AD was 1.08, χ² = 0.09, 95% CI: 0.65–1.79, P > 0.05, 1.02, χ² = 0.08, 95% CI: 0.63–1.66, P > 0.05, and 0.80, χ² = 0.34, 95% CI: 0.38–1.67, P > 0.05 in ACT TT, TA, and AA three genotypes, respectively, and 1.09, χ² = 0.24, 95% CI: 0.77–1.55, P > 0.05, in ACT*T allele, suggesting no obvious association between any allele and genotype of ACT polymorphism and AD.

When the data were split into ApoE* ϵ4-positive and ApoE* ϵ4-negative groups (Table II), the distribution of ACT polymorphism was still similar between AD cases and controls in both subgroups ApoE* ϵ4-positive group: χ² = 0.085, df = 2, P > 0.05; ApoE* ϵ4-negative group: χ² = 0.23, df = 2, P > 0.05. Even among AD cases, there was no difference in ACT genotype frequency between ϵ4 carriers and non-ϵ4 carriers, χ² = 1.29, P > 0.05, df = 2. Thus, the ApoE status did not affect the relationship between ACT polymorphism and AD.

ApoE allele frequencies among three ACT genotypes are shown in Table III. In the three ACT genotypes, the ApoE* ϵ4 allele frequency increased and the ϵ3 allele frequency decreased in AD cases as compared to normal controls. The ApoE* ϵ4 was strongly associated with AD in ACT*AT, OR = 3.16, 95% CI: 1.57–6.34, P < 0.001, and TT genotypes, OR = 3.46, 95% CI: 1.61–7.46, P < 0.001, but not in the ACT*AA genotype, OR = 2.32, 95% CI: 0.59–1.70, P > 0.05. Furthermore, no interactive effects of ApoE allele and ACT genotypes were detected, χ² = 2.24, P > 0.05, df = 4.

DISCUSSION

We failed to detect the association between ACT AA genotype and AD in our samples, and did not find the interaction between ACT polymorphism and ApoE* ϵ4 allele, although the ApoE* ϵ4 allele had been identified as a major risk factor for AD in Chinese [Jiang et al., 1996]. These results were contrary to the findings of Kamboh et al. [1995], but in accordance with other reports. The discrepancy might result for many reasons. First, there was a gender distribution difference between AD group and control subjects. In our study, the age of AD patients also differed slightly from the age of controls. These two differences could have affected our results, lowering the power to detect an association between ACT polymorphism and AD. Second, different ethnic backgrounds and other environmental factors such as geological distribution and diet might also have modified the effects of ACT polymorphism in AD in different populations.

We also failed to detect the association between ApoE* ϵ4 allele and AD in ACT*AA genotype in our samples, but there was still a trend toward an increase of ApoE* ϵ4 allele frequency in AD with AA genotype as compared to control subjects. This negative result might be due to the small sample size after the stratification of AD cases according to the ApoE status, therefore, so that we lacked the power to find the association.

Thus, the association between ACT polymorphism and AD should be studied further when age and gender factors are adjusted. In our study, we had tried to restrict the control individuals to the same age range as patients, but no effects of ACT A alleles and AT genotype on AD risk were detected. Therefore, we conclude that the ACT polymorphism might not be a risk factor for developing sporadic AD in Chinese Han population.