Associations between APOE polymorphisms and seven diseases with cognitive impairment including Alzheimer's disease, frontotemporal dementia, and dementia with Lewy bodies in southeast China.

OBJECTIVE: To explore the effect of APOE polymorphisms on patients with cognitive impairments in The Chinese Han population.
MATERIALS AND METHODS: A total of 1027 cases with Alzheimer's disease (AD), 40 cases with vascular dementia (VaD), 28 cases with behavioral variant frontotemporal dementia (bvFTD), 54 cases with semantic dementia (SD), 44 cases with dementia with Lewy bodies (DLB), 583 cases with mild cognitive impairment (MCI), and 32 cases with vascular cognitive impairment no dementia (VCIND) were recruited consecutively from memory disorders clinics in Huashan Hospital between January 2010 and December 2014. The 1149 cognitively normal controls were recruited from the community epidemiologic investigations. The APOE genotypes were determined using the TaqMan assay.
RESULTS: The distribution of genotype and allele frequencies of APOE differed significantly between control and AD or MCI, with ε4 increasing the risk of AD and MCI in a dose-dependent pattern and ε2 decreasing the risk of AD, but not the risk of MCI. As for VaD, significant differences in the APOE genotype distribution were found compared with the controls. E4/4 increased the risk of VaD and ε4 increased the risk of VCIND in women. The allele distribution differed between bvFTD and controls, but genotype and allele frequencies of APOE did not affect the risk of bvFTD, SD, and DLB.
CONCLUSION: In The Chinese Han population, APOE ε4 increased the risk of AD and MCI in a dose-dependent manner and ε2 decreased the risk of AD as reported previously. APOE ε4 might increase risk in VaD and female patients with VCIND, but no effects of APOE on bvFTD, DLB, and SD were found.

Introduction

Dementia is an acquired cognitive impairment syndrome with slow progression. Usually, it starts in mid-adulthood or later and is characterized by dysfunctions and decreases in memory, visual spatial ability, orientation, and calculation as well as alterations in personality, emotion, and behavior. According to the pathogenesis, dementia can be divided into those that are nondegenerative, such as vascular dementia (VaD), and those that are degenerative such as Alzheimer’s disease (AD), frontotemporal dementia (FTD), and dementia with Lewy bodies (DLB). On the basis of an epidemic investigation, in individuals aged older than 65 years, 5.14% had dementia, among whom 62% had AD and 29% had VaD (Jia ). The prevalence of other dementias was low, with 0.36% in DLB and 15–22/10000 in FTD (Onyike and Diehl-Schmid, 2013; Vann Jones and O’Brien, 2014).

The pathogenesis of dementia is obscure and it is considered a complicated disease. The genetic aspects have been indicated to play an important role in the development of these diseases (Farlow and Foroud, 2013; Bruni ; Loy ; Ferencz and Gerritsen, 2015). In many genetic loci related to sporadic dementia, APOE is the only one that has been confirmed to be associated with the risk of AD. APOE, located in 19q13.2, has three common isoforms termed ε2, ε3, and ε4, which derive six genotypes. Its encoding protein, Apolipoprotein E, consists of 299 amino acids and is a cholesterol carrier involving in lipid transportation and injury repair in the brain. APOE ε4 increased the risk of AD in a dose-dependent manner, in contrast to APOE ε2, which plays a protective role (Bertram ; Sando ; Bonner-Jackson ; Wu ).

As its genetic effect was so evident in AD, the APOE polymorphisms have deservedly been studied in other dementias, such as VaD, FTD, and DLB, but the results have been controversial (Fei and Jianhua, 2013; Berge ; Rohn, 2014). A number of studies reported an increased APOE ε4 frequency in VaD, similar to that found in AD (Treves ; Souza ), whereas some other findings did not replicate this association (Frank ; Huang ; Orsitto ; Kim ). As for FTD, the effect of the APOE polymorphisms on the development of the disease was still controversial (Verpillat ; Bernardi ), so was in DLB (Lovati ; Kobayashi ; Boot ; Berge ; Bras ). Most of the studies were carried out in White populations and rarely in the Chinese Han population, mostly in VaD (Huang ; Liu ). Ji found that the APOΕ ε4 prevalence of AD and FTD was similar in the Chinese Han population and suggested that the APOE ε4 allele is a risk factor for both disorders. The effect of APOE polymorphisms on DLB and semantic dementia (SD), the subtype of FTD [the other two subtypes were behavioral variant frontotemporal dementia (bvFTD), and progressive nonfluent aphasia (PNFA)] (Neary ) has not been investigated in the Chinese Han population as yet.

Here, we carried out a case–control study of five dementias, AD, bvFTD, VaD, SD, and DLB, and two cognitive impairments, mild cognitive impairment (MCI) and vascular cognitive impairment no dementia (VCIND), in Southeast Chinese Han patients to explore whether APOE polymorphisms would affect the risks of these conditions. This is the first report of the relationship of APOE polymorphisms with the risk of SD/DLB in the Chinese Han population.

Materials and methods

Participants

Two thousand and forty-four patients with cognitive impairment were recruited consecutively from memory disorders clinics in Huashan Hospital between January 2010 and December 2014. Among these, 1027 patients with AD, 40 patients with VaD, 28 patients with bvFTD, 54 patients with SD, 44 patients with DLB, 583 patients with MCI, and 32 patients with VCIND were included in the study. Another 55 patients whose dementia was secondary to an explicit diagnosis of general paresis of insane, Parkinson’s disease with dementia, corticobasal degeneration, sequela of brain injury, and PNFA were excluded because of clear causes or small case numbers. The rest of the 159 patients were excluded because of the presence of unspecified dementia. The 1149 cognitively normal controls were recruited from community epidemiologic investigations (Ding ).

The diagnosis of AD was made according to the diagnostic guidelines for AD of the National Institute of Aging and Alzheimer’s Association (Jack ). VaD was diagnosed according to the criteria of the National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherche et 1’Enseignement en Neurosciences (NINDS-AIREN) (Udaka, 2011). FTLD was diagnosed on the basis of the report of the Work Group on Frontotemporal Dementia and Pick’s Disease (McKhann ). DLB was diagnosed according to the DLB Consortium criteria (McKeith ). MCI was diagnosed according to the criteria established by the International Psychogeriatric Association Expert Conference on MCI (Gauthier ). VCIND was diagnosed according to the Vascular Cognitive Impairment Harmonization Standards of the National Institute of Neurological Disorders and Stroke–Canadian Stroke Network (NINDS-CSN) (Hachinski ). The enrollment procedures were as reported previously (Guo ; Sun ). Written consents were obtained from the participants or their legally authorized caregivers. This study was approved by the ethics committee of Huashan Hospital.

Genotyping of APOE

Genomic DNA was extracted from peripheral blood using a Blood Genomic DNA Extraction Kit (Tiangen, Shanghai, China). The APOE genotypes were determined using the TaqMan assay according to the method described previously (Koch ).

Statistical analysis

Statistical analyses were carried out using SPSS, version 19.0 (SPSS Inc., Chicago, Illinois, USA). Hardy–Weinberg equilibrium tests of APOE polymorphisms within the groups were performed using χ2 analysis. The χ2-test or the Student t-test was used to test for the differences between the patients of each group and control participants in the distribution of sex, age at onset (AAO), education level, and mini-mental state examination scores. The χ2-test was used to compare the genotypes and allele frequencies between patients groups and control participants. Odds ratio (OR) and the 95% confidence interval (CI) for testing possible associations between patients groups and control group were determined by binary logistic regression analyses; AAO and sex were used as covariates. The potential effects of each genotype on AAO in patients of each group were calculated by one-way analysis of variance and further analysis by post-hoc least significant difference. P less than 0.05 was considered significant.

Results

General information

General information of the participants is shown in Table 1. The age of the control participants was significantly higher than that of the patients with AAO of AD, MCI, VCIND, bvFTD, and SD. The sex distribution of VaD and DLB was different from that of the controls. The mini-mental state examination score was significantly lower in the patient groups. The distributions of the six common genotypes of APOE were under Hardy–Weinberg equilibrium in all patients and control participants (Supplementary Table, Supplemental digital content 1, ).

Table 1

Characteristics of the participants in the patient and control groups

The E2 allele reduced the risk and the ε4 allele increased the risk in a dose-dependent manner in AD and MCI

The distribution of genotype and allele frequencies of APOE was compared between each patient and control. The distribution of genotype and allele frequencies of APOE differed significantly between control and AD or MCI. There were more ε4-carrying patients (ε2ε4, ε3ε4, ε4ε4) in the AD and MCI groups than in the control group (Table 2). Among AD and MCI patients, ε4 increased AD the risk in a dose-dependent manner. The E2 allele decreased the risk in AD, but not in MCI. These significances remained after the patients were further stratified by sex (Table 3). We also investigated the effect of ε2 and ε4 alleles on the risk of AD and MCI in different ranges of AAO and found that ε2 played a protective role against AD in patients with AAO of 56–65 and above 76, while MCI with AAO 76–80 (Table 4). The E4 allele increased the risk of AD in patients with AAO above 56, with the highest OR of 4.894 in 76–80. As for MCI, ε4 increased the risk in patients with AAO 56–80.

Table 2

Distribution of allele frequencies and genotypes of APOE in patients and controls

Table 3

Effect of APOE ε2 and ε4 haplotype on the risk of AD, MCI, VaD, VCIND, FTD, DLB, and SD stratified by sex

Table 4

Effect of APOE ε2 and ε4 haplotype on the risk of AD, MCI, VaD, VCIND, FTD, DLB, and SD stratified by age at onset

E4/4 increased the risk of VaD and ε4 increased the risk of VCIND in women

The APOE ε4/4 genotype increased the risk of VaD, whereas the ε3/4 genotype, the ε4 allele, or the ε2 allele statistically didn't increase the risk. (Table 5). When further stratified by sex, ε4 was found to increase the risk of VCIND in female patients (Table 3), but neither ε2 (+) nor ε4 (+) status affected the risk of VaD and VCIND after stratification by AAO (Table 4). We investigated the AAO of VaD and VCIND according to the dosage of ε2 and ε4, but we found no effect of the allele on AAO (Table 6).

Table 5

Logistic regression of APOE genotypes and allele frequencies in patients and controls

Table 6

The APOE ε2 and ε4 allele dosage effect on age at onset in patients with cognitive impairment

Genotype and allele frequencies of APOE did not affect the risk of bvFTD, SD, and DLB

Only the allele distribution was found to be different between bvFTD patients and controls, with more ε4 and less ε2 in bvFTD (Table 2). However, when it was further analyzed by logistic regression, the relationships did not exist (Table 5). The negative results remained after stratification by sex and AAO (Tables 3 and 4).

Moreover, we investigated the AAO according to the dosage of ε2 and ε4 in bvFTD, SD, and DLB and found no effect of ε2 or the ε4 allele on AAO (Table 6).

Discussion

The results of the relationship between APOE and the risk of AD in the current research were identical to those reported in many previous investigations, with ε4 increasing the risk of AD, lowering the AAO, and ε2 exerting a ‘protective’ effect against the risk of AD (Chartier-Harlin ; Pastor ).

We found that ε4 increased the risk of MCI in a dose-dependent manner, which was also in accordance with previous reports in both White and Chinese populations (Borenstein ; Boyle ; Albert ; Wang ). In contrast, the APOE ε2 allele appeared to confer cognitive benefits in the White population (Blacker ; Bonner-Jackson ). Whether ε2 could decrease the risk of MCI in the Chinese Han population still remains controversial, perhaps because of its considerably lower frequency compared with ε4. Borenstein and colleagues investigated 34 MCI patients and 32 controls among Shanghai urban residents, but did not find the ‘protective’ effect of ε2. The population in their study was quite similar to ours, as were the results (Borenstein ). Wang tested APOE polymorphisms in a Han population and different ethnic minority groups in North China and found the ε2 allele protective for MCI only in the Han population (OR=0.48, 95% CI: 0.24–0.96). These differences might be explained by the regional and ethnic diversity.

The protective and risk effects of APOE on AD and MCI were found different among varied age ranges and sex distributions, but the results were controversial (Qiu ; Corrada ; Altmann ). Qiu and colleagues found that the APOE ε4 allele had a stronger risk effect in men than women, and the ε2 allele conferred a protective effect only in younger-old people (<85 years) but not in the oldest old (> 85 years). However, other population-based studies showed that ε4 led to a higher risk of AD in women and ε2 was not related to prevalent dementia in either sex (Corrada ; Altmann ). In our results, the ε4 allele increased the risk of AD and MCI in both men and women, and almost all age ranges older than 56 years. For the ε2 allele, our data showed its protective effect in both sexes and in the 56–65 and 76–80 AAO range in AD. Interestingly, the ε2 allele was found to decrease the risk of MCI in the 76–80 AAO range.

In our study, APOE ε4/4 was found to be a risk factor for VaD and ε4-carrying status to be a risk factor in female patients with VCIND. Researches from many groups have verified the relevance of APOE ε4 and the increased risk of VaD (Souza ; Baum ; Bharath ; Liu ) and reported that it also influenced cognitive decline after stroke (Ballard ; Wagle ). A meta-analysis showed that the pooled OR value in VaD patients in a Chinese population with the ε4/4 genotype was 3.34 [95% CI (1.89–5.88)] (Liu ). Significant risk factors for cognitive impairment after stroke are APOE ε4, prestroke cognitive reduction, previous stroke, and neurological impairment (Wagle ), but some other researches did not report any links (Huang ; Orsitto ; Kim ). Kim investigated the association of VaD with the APOE polymorphisms in Koreans and found no association between APOE ε4 or the ε2 allele and the risk of VaD, even after stratification by sex and age. This may be attributed to the complex environmental, compound risk factors of stroke, ethnic backgrounds, and use of different methods among researches (Huang ). There is an association between APOE ε4 and cognitive decline in elderly adults (Packard ), so as to APOE ε4 and hippocampal volume loss (Jak ). When stroke occurs, cognitive impairment may manifest as a result of stroke-related structural and functional changes primarily of the hippocampus and reduced cognitive compensatory potential. A recent research found an amino-terminal fragment of apolipoprotein E within neurofibrillary tangles, blood vessels, and reactive astrocytes in the VaD by immunohistochemistry, supporting a role for the proteolytic cleavage of apolipoprotein E in the VaD and supporting the susceptible role of the APOE polymorphism in this disease (Rohn ).

Whether APOE polymorphisms are correlated to FTD remains unclear. Some researches reported that APOE ε4 increased the risk of FTD (including one study in a Chinese population) (Stevens ; Seripa ; Fei and Jianhua, 2013), whereas others did not (Geschwind ; Verpillat ). One meta-analysis, including 364 FTD patients and 2671 controls, found no significant relationship of ε4 with the risk of FTD, whereas ε2 was likely to be a risk factor for FTD (Verpillat ). However, ε2 was reported to be a protective factor for FTD as well (Bernardi ). In contrast to the above studies, in our study, we stratified bvFTD and SD of FTD. We did not find a relationship between APOE polymorphisms and bvFTD. As for SD, reports related to APOE polymorphisms are uncommon. One report described an increased frequency of the APOE ε4 allele in patients with SD compared with those with bvFTD and PNFA (Short ). In our study, no positive results were found. The controversial association between APOE polymorphisms and FTD might be attributed to the genetic heterogeneity of FTD.

The Ε4 allele has been proven to increase the risk for the development of DLB and decrease its AAO in many studies (Kobayashi ; Boot ; Berge ; Bras ). However, the protective effect of the ε2 allele remains uncertain (Singleton ). One study of 156 DLB patients and 519 controls showed that ε2 reduced the risk for the development of DLB (P=0.004, OR 0.4, 95% CI: 0.3–0.8) and the AAO was delayed by 4 years in ε2-carrying patients (Berge ). In our study, neither ε2 (+) nor ε4 (+) status affected the risk of DLB. The negative results might be attributed to the small sample number and the ethnic background.

This study had a few limitations. The numbers of VaD, VCIND, bvFTD, SD, and DLB groups in our study were small. It is possible that many effects failed to reach significance or even led to false positives. Thus, this should be verified in a much larger sample in the Chinese Han population to better describe the role of APOE in the genetic pictures of these diseases in the future.

Conclusion

In this case–control study, we found that ε4 increased the risk of AD and MCI in a dose-dependent manner and ε2 decreased the risk of AD in the Chinese Han population. APOE ε4 might increase the risk in VaD and female patients with VCIND. The relationship between APOE and DLB and SD was first reported in the Chinese Han population. Further researches should focus on investigations of other major and minor genes affecting these cognitive impairment diseases, especially bvFTD, DLB, and SD. At the same time, our results should be confirmed in a much larger sample of the Chinese Han population to better describe the role of APOE in the genetic pictures of these diseases.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's website ().