Abstract:
Alzheimer's disease as a neurodegenerative disorder is the commonest type of dementia. A growing number of genes have been reported as the risk factors which increase the susceptibility to Alzheimer's disease. Apolipoprotein E, which its ε4 allele has been reported as a risk factor in late onset Alzheimer's disease (AD), is the main cholesterol carrier in the brain. In this study, 154 AD cases and 162 control subjects from Iranian population were genotyped for APOE. The frequency of ε2ε3 genotype was significantly higher in control subjects than AD patients (13.5% versus 5.2%, p<0.05) and ε3ε4 genotype frequency was significantly higher in AD cases compared with control subjects. APOE -ε2 allele frequency in cases was lower than control subjects but this difference was not significant (4.2% versus 7.7%, p>0.05). It seems that individuals carrying ε4 allele, develop AD 6.5 times more than non-carriers (OR=6.566, 95% CI= 2.89-14.92). It had been reported that ε4 allele acts in dose- and age-dependent manner but we have shown that the risk of developing AD in male APOE -ε4 allele carriers is higher than female ε4 carriers.
Introduction:
Alzheimer's disease (AD), which presents progressive cognitive defects such as memory loss, apraxia and personality changes, is the commonest cause of dementia in the mid and late ages (1, 2). Two neuropathophysiological hallmarks of AD are intracellular neurofibrillary tangles (NFT) and beta amyloid plaques in brain blood vessels. As hundreds genes have been known as the risk factors for late onset AD, the well-known one is apolipoprotein E gene (APOE) which has been recognized as the most important risk factor in 65% of sporadic cases (3).
Apolipoprotein E is the main part of very low density lipoproteins (VLDL), Intermediate density lipoproteins (IDL), chilomicrons and the main cholesterol carrier in the brain and its synthesis is independent in central nervous system (CNS) and lung. As APOE expression is stimulated by any CNS damages or diseases, it seems that apolipoprotein E regulates cholesterol metabolism and distribution in the brain to repair and stabilize neurons' membrane and myelin (4, 5 and 6). Apolipoprotein E isoforms, which coded by ε2, ε3 and ε4 alleles, are only different in two amino acids of 112th and 158th ; ε 2 (Cys112, Cys158), ε3 (Cys112, Arg158) and ε4 (Arg112, Arg158).
APOE-ε4 allele has been identified as a genetic susceptibility factor for AD in various populations. This allele increases the risk of late onset AD and lowers the onset age in a dose-dependent manner (7-11).
This study has been focused on distribution of APOE genotypes and allele frequencies and the association of APOE alleles with AD in Iranian population. It was found that APOE- ε4 allele is a risk factor in developing AD in Iranian population but APOE- ε2 is not protective against AD in this population.
Material and method:
This case and control study involved 154 AD cases (with mean age of 78.55± 7.80) and 162 control subjects (with mean age of 77.14±6.95) in which AD cases and control subjects were included if they were older than 65 years old and the informed consent was signed by them or their legal guardians. The criterion for inclusion as a case was the diagnosis of Alzheimer's disease diagnosed by the expert psychiatrist and lacking any neurologic or psychiatric disorders for control group. Subjects were excluded if they had any family history of dementia or neurologic diseases. AD and control subjects were recruited from Alzheimer's society of Iran and Geriatric centers Farzanegan, Mehrvarzan, Shayestegan, Kahrizak, Hashemi nejhad and Rheumatism Center in Tehran, Iran from 2007 to 2008.
AD subjects were diagnosed by DSM-IV criteria and age, sex, job and genetic background were registered for both case and control groups.
Genomic DNA was extracted from peripheral blood leukocytes by salting-out method. APOE was genotyped by PCR-RFLP method which had been described by Wenham et al (12). DNA was amplified by Polymerase chain reaction (PCR) using forward primer: 5´-TCC AAG GAG CTG CAG GCG GCG CA-3´; and Reverse primer: 5´-ACA GAA TCC GCC CCG GCC TGG TAC ACT GCC A-3´. The 227 bp PCR products were digested by Hha I (10 U/µl, Fermentas) and loaded on a 12% polyacryl amide gel for electrophoresis; finally the gels were stained by silver staining method.
APOE genotype and alleles frequencies were calculated by statistical software of SPSS.17 and compared between AD case and control subjects by χ2 test and Fisher's exact test. When statistical significance was assumed p < 0.05 level, the odd ratios (OR) were calculated by free online epidemiological software of OpenEpi (2.2.1).
AD patients
(n=154)
Control subjects
(n=162)
p value
Age
78.55 ± 7.80a
77.14 ± 6.95
0.091
Sex (M/F)b
63/91
63/99
0.714
Jobs
Housewife
55.8%
56.2%
0.938
Own business
23.4%
21.0%
Worker
9.2%
8.6%
farmer
3.2%
3.1%
employee
8.4%
11.1%
Education
levels
Illiterate
41.6%
43.2%
0.427
primary school
29.2%
29.6%
secondary school
16.2%
12.3%
Diploma
11.1%
9.3%
Academic
1.9%
5.6%
Genetic background
Fars
61.0%
63.6%
0.490
Turk
25.3%
25.3%
Kurd
3.9%
1.8%
Lor
0.7%
2.5%
Shomali
9.1%
6.8%
Table 1) Comparison of mean age, sex, Jobs, education levels and genetic backgrounds between AD cases and control subjects was made using t-test and χ2 test analysis.
a Mean ± S.D.
b Male/Female
Results:
Distribution of age, sex, jobs, education levels and genetic background was the same in both groups so it did not need to use any methods for matching of cases and controls (Table 1). The mean age and females were slightly higher in patients compared with control subjects. The highest frequency of AD was observed in housewives and the lowest one was among farmers. People with academic education had the lowest frequency among patients and illiterate individuals had the most one. The samples were consisted of 5 Iranian backgrounds in which Fars was the most frequent one.
The frequencies of APOE genotypes and alleles in AD cases and control subjects have been shown in Table 2. The frequency of ε2ε2 genotype in control subjects was higher than that in AD cases but it was not significant (Fisher's exact test p = 0.614). The distribution of ε2ε3 genotype was significantly different in both groups (13.6% in controls versus 5.2% in AD, p = 0.011) and OR was found to be 0.3487 (95% CI = 0.1503-0.8091). The genotype frequency of ε3ε3 was significantly higher in control subjects compared with patients (p = 0.018). The ε3ε4 genotype frequency in AD cases was significantly higher than that in control group (20.8% versus 3.7%, p = 0.000).
The distribution of ε2ε4 genotype was the same in both groups and different distribution of ε4ε4 genotype in the groups was not significant (1.9% versus 0, FET p = 0.115).
The APOE- ε4 allele frequency was significantly higher in AD cases compared with Control subjects (12.7% versus 2.2%, p = 0.000).Comparing allele frequency in APOE- ε4 allele carriers with non-carriers, OR was found to be 6.566 (95% CI= 2.89-14.92).
The frequency of APOE- ε3 allele in patients was significantly lower than that in control group (p = 0.010). Despite of higher APOE- ε2 allele frequency in AD cases compared with control subjects, this difference was not statistically significant (p = 0.065 and OR=0.527, 95% CI=0.2545-1.05) (Table 2).
AD patients
(n=154)
Control subjects
(n=162)
p value
ε2ε2
1.3%
0.6%
0.614
ε2ε3
5.2%
13.6%
0.011
ε2ε4
0.6%
0.6%
1.000
ε3ε3
70.2%
81.5%
0.018
ε3ε4
20.8%
3.7%
0.000
ε4ε4
1.9%
0
0.115
ε2
4.2%
7.7%
0.065
ε3
83.1%
90.1%
0.010
ε4
12.7%
2.2%
0.000
Table 2) The genotype and allele frequencies were compared between AD cases and control subjects using χ2 test and Fisher's exact test.
Female
Male
All
ApoE genotype
ε3/ε3
Reference Group
Reference Group
Reference Group
ε2/ε3
P=0.5 OR=0.663 (0.22-1.75)
P=0.1 OR=0.23 (0.05-1.13)
P=0.08 OR=0.44 (0.19-1.03)
ε3/ε4
P=0.001 OR=7.86 (2.58-23.9)
P=0.08 OR=4.7 (0.96-22.8 )
P=0.001 OR=6.52 (2.63-16.17)
ε4/ε4
No data
P=0.31 OR=7.3 (0.37-144.8)
P=0.2 OR=8.55 (0.44-167. 3)
ε3/ε4+ ε4/ ε4
P=0.001 OR=7.86 (2.58-23.9)
P=0.02 OR=6.25 (1.33-29.4)
P=0.001 OR=7.1 (2.9-17.6)
APOE allele
ε3
Reference Group
Reference Group
Reference Group
ε4
P=0.001 OR=5.59 (2.07-15.05)
P=0.002 OR=8.3 (1.86-37)
P=0.001 OR=6.3 (62.8-14.45)
ε2
P=0.15 OR=0.46 (.17-1.19)
P=0.87 OR=0.8 (0.29-2.24)
P=0.18 OR=0.59 (0.29-1.18)
 This group assumed as reference group for comparison.
Table 3) APOE genotypes and alleles frequencies distributed by sex groups
Table 3 shows APOE genotype and allele frequencies distributed by sex groups. ε2ε3 genotype frequency in control subjects was significantly higher than AD subjects in men group (p = 0.020) whereas the frequency of ε3ε3 genotype was significantly higher in control subjects compared with AD cases in women group (p = 0.029). The genotype frequency of ε3ε4 in AD cases was higher than control subjects in both male and female groups but it was significant just in women group (p = 0.000).
APOE- ε3 allele frequency was significantly higher in control subjects compared with AD cases in male group (p = 0.044) and the frequency of APOE- ε4 allele in patients was significantly higher than control subjects in both males and females with different OR (Males: p = 0.001, OR=8.421 (1.894-37.44) and Females: p = 0.000, OR=5.846 (2.173-15.73)).
Discussion:
According to this study, APOE- ε4 allele is a risk factor for developing late onset AD in Iranian population like many other populations (13-18). Although ε2ε3 genotype seems to play a protective task against AD but the protective role of APOE- ε2 allele has not demonstrated in this study and it may be proved by a bigger sample size.
The risk of developing AD in individuals with ε2ε3 genotype is about 0.35 (0.3487, 95%CI = 0.1503-0.8091) compared with individuals without this genotype so ε2ε3 genotype seems to be protective against AD whereas protective role of ε2 allele has not demonstrated in Iranian population yet. APOE- ε4 allele carriers develops AD, 6.5 times more than non-carriers (6.566, 95%CI= 2.89-14.92) this allele's risk seems different in males and females. Different OR for ε4 allele in men and women indicates that risk of AD in male APOE- ε4 allele carriers (OR=8.421, CI = 1.894-37.44) is higher than female carriers (OR=5.846, CI = 2.173-15.73) so it seems that despite the age-dependant and dosage dependent manner of this allele which were investigated in Iranian population by A.V Raygani et al (8) it may act in a sex-dependent way as well. As three patients were observed with ε4ε4 genotype, it was not possible to assess the dosage-dependent action of ε4 allele in this study.
As the study groups were similar based on potential confounders (age, sex, genetic background, job and education), it can be assumed that the results are mainly unbiased. There was no reliable history or evidence for the time of Alzheimer disease onset so we couldn't evaluate the effect of different genotypes or alleles on the age of onset in the Alzheimer's disease subjects.
In an autopsy-based study the frequency of ε4 allele and ε4ε4,ε 3ε4 genotypes were 40%, 16.5% and 43.2% in Alzheimer patients and 16%, 2.2 and 20.9% in control group (19).
In a group of African Americans AD patients, a significantly increased risk of AD was associated with two ε4 alleles or one ε4 allele when compared to ε 3ε3 genotype (20). In our study the frequencies of ε4 allele and ε4ε4, ε3ε4 genotypes were lower than results of Vaisi Raygani et al (8) but the proportion of them was the same and the results of two studies in Iranian population are consistent. No significant association was found between ε2 allele or related genotypes and AD but it sounds to work as protective factor for Alzheimer disease; however this finding, should be confirmed in further studies.
