Association Between Clusterin Gene Polymorphisms and Epilepsy in a Han Chinese Population

Introduction

Epilepsy is a common neurological disease affecting ∼1% of the worldwide population. A large group of epilepsies have unknown etiology (Cavazos et al., 2004), and about 20-30% of epileptic patients develop a drug-resistant phenotype after rational antiepileptic drug therapy. Animal studies have shown that inflammation and apoptosis are important pathophysiological processes in the development of epilepsy (Emerit et al., 2004; Engel and Henshall, 2009; Engel et al., 2013; Henshall and Engel, 2013). Clusterin (CLU), also known as apolipoprotein J, is a multifunctional chaperone protein that is involved in the development, cell death, and lipid transportation in a variety of tissues and bodily fluids, including the brain, spinal cord, liver, and cerebrospinal fluid (CSF) (Garden et al., 1991; Polihronis et al., 1993; Bursch et al., 1995). Studies have shown that CLU is implicated in the apoptosis and inflammatory mechanisms of epilepsy. CLU mRNA and protein levels have been shown to be increased in dying hippocampal neurons after status epilepticus in rats (Dragunow et al., 1995). In addition, an increased nuclear CLU level in hippocampal neurons has been demonstrated to be accompanied by extensive hippocampal cell death and apoptosis after kainic acid-induced seizures (Kim et al., 2012). Furthermore, a lower level of CLU in the CSF has been found in drug-resistant epileptic patients than in drug-responsive epileptic patients (Yu et al., 2014). Thus, CLU is probably associated with the pathogenesis of epilepsy.

The CLU gene is located on chromosome 8p21-p12 and contains nine exons. Single nucleotide polymorphisms (SNPs) may influence susceptibility to disease by altering the expression of related genes. CLU SNPs have been reported to be associated with neurological disorders such as Alzheimer's disease (AD), Parkinson's disease, multiple sclerosis, and pseudoexfoliation (Burdon et al., 2008; Stoop et al., 2008; Gao et al., 2011; Lin et al., 2012). Seven SNPs (rs9331908, rs11136000, rs867231, rs867230, rs9331888, rs9314349, and rs484377) have been identified that capture more than 90% of the variation in the CLU gene (Madhav et al., 2010), and the CCG haplotype (rs2279590-rs11136000-rs9331888) has been successfully identified as a genetic susceptibility factor for AD (Lambert et al., 2009). The rs9331949 C allele has been shown to be significantly associated with an increased risk of late-onset AD in Han Chinese (Yu et al., 2013). However, the relationship between CLU SNPs and epilepsy is still undetermined. In this study, we explored whether CLU SNPs (rs11136000, rs9314349, and rs9331949) are associated with epilepsy in a Han Chinese population.

Methods

Results

This study included 249 epileptic cases and 289 healthy controls. The characteristics of the participants are listed in Table 1. The epileptic cases and the healthy controls did not differ significantly in terms of age or gender (p > 0.05).

Table 3 shows the distributions of the CLU SNP genotypes and allelic frequencies in the epileptic and control subjects. There were no significant differences in the distributions of the CLU SNP genotypes or allelic frequencies between the epileptic and control subjects (p > 0.05). The distributions of the genotypes and alleles of the CLU SNPs in early- and late-onset epilepsy, TLE, and DRE groups are shown in Tables 4-7. No deviation from HWE was observed. We did not detect significant differences between the case and the control groups regarding the alleles, genotypes, or haplotypes for the three SNPs (all p > 0.05, data not shown). Our results suggest that CLU SNPs are not associated with epilepsy.

Discussion

Our study is the first to investigate the genetic relationship between CLU SNPs and epilepsy in a Han Chinese population. We failed to find any association between the CLU SNPs and epilepsy, indicating that these CLU polymorphisms may not be a risk factor for epilepsy.

CLU is associated with a number of neurological disorders (Charnay et al., 2012). It promotes amyloid deposition and increases neuritic toxicity in an AD mouse model (PDAPP mouse) (DeMattos et al., 2002). CLU knockout PDAPP mice had significantly fewer fibrillar A (amyloid) deposits and reduced associated neuritic dystrophy than PDAPP mice expressing CLU (DeMattos et al., 2002). Compared to the wild-type mice, CLU knockout mice had slower tissue remodeling during the healing process after permanent middle cerebral artery occlusion (Imhof et al., 2006). This means that the CLU gene is implicated in the apoptosis and inflammatory mechanisms.

CLU has been reported to accumulate in dying neurons and is related to caspase-3-activated apoptosis in the hippocampus after status epilepticus (Dragunow et al., 1995; Kim et al., 2012). Recent data suggest that CSF-CLU levels are lower in epileptic patients than in healthy individuals, and the CSF-CLU levels in patients with DRE are lower than those of drug-responsive epilepsy, suggesting that CLU may be used as a potential biomarker for DRE (Yu et al., 2014). The CLU levels in rats suffering seizures during their youth and adulthood were significantly higher than those only suffering seizures during adulthood, implicating that inflammation following early-life seizures as well as CLU may contribute to inflammation and neurotoxic mechanisms by activating microglia (Xie et al., 2005; Somera-Molina et al., 2007). Moreover, valproic and kainic acid-stimulated CLU expression in human astrocytes and a ketogenic diet can change CLU- and autophagy-related gene expression, thus preventing CLU accumulation in the hippocampus of mice with seizures (Noh et al., 2005; Kauppinen et al., 2010; Ni et al., 2015). Therefore, it is reasonable that CLU participates in epilepsy through the mechanisms of apoptosis and inflammation. However, we were unable to support the association between epilepsy and the three SNPs in this study.

One explanation for this discrepancy might be attributed to the relatively small sample size compared with genome-wide association studies (GWAS). In our study, the MAFs of CLU rs11136000 (T allele) in the case and control groups were 0.215 and 0.177, respectively, whereas in Caucasians the MAF, it is reported to be 0.349. Due to the low MAF in the Chinese population, a substantially larger sample size is required to detect small or moderate genetic effects of SNPs. It is possible that a positive result would be obtained with a larger sample size or meta-analysis in future studies.

Another reason for this negative finding might be ethnic background differences. For example, a large GWAS showed that the CLU rs11136000 polymorphism is associated with AD risk in Caucasians. However, no positive result has been obtained in other ethnic groups (Jun et al., 2010). In addition, conflicting results have been obtained in China (Liu et al., 2014; Lu et al., 2014), suggesting that the results might be inconsistent despite analysis of the same ethnic population (Chinese). Therefore, we may draw the conclusion that the genetic variants may vary between different ethnic groups.

The location of the SNPs could also contribute to this negative result. The three SNPs in our study are located in the noncoding regions, which seem unlikely to affect CLU expression directly. rs11136000 is located in the intron and is associated with the onset of AD, although with mixed results. One study has suggested that rs11136000 might increase the risk of AD by affecting alternative splicing, mRNA structure stability, and subsequent protein production (Lin et al., 2012). rs9331949 is located in the 3′UTR and participates in promoting tumor generation and creating new microRNA binding sites, resulting in translational repression or messenger RNA degradation, thus affecting the half-life of mRNA (Makhanova et al., 2008; Manikandan et al., 2012). Studies have shown that CLU promoter mutations may affect promoter methylation in breast tumor tissues (Park et al., 2011). rs9314349 is located in the promoter and may regulate CLU protein expression by promoter methylation. Whether these three SNPs affect protein expression through the mechanisms described above needs to be further explored.

Last, the pathogenesis of epilepsy is very complicated. In addition to inflammation and apoptosis mechanisms, it also includes genetic factors, ion channels, neurotransmitters, astrocyte dysfunction, and so on. (Sitnikova et al., 1989; Hirose et al., 2000; Chan et al., 2006; Hildebrand et al., 2013). The cause of epilepsy is also different for different age ranges. The pathogenesis of epilepsy in each case of our study population was difficult to elucidate. This heterogeneity of epilepsy pathogenesis will lead to a certain degree of bias. Different results may be obtained in the future with an increased understanding of the pathogenesis of epilepsy.

In conclusion, we did not find an association between the CLU polymorphisms and epilepsy in this study population. The limitations of this study include the small sample size, ethnic background, and pathophysiological heterogeneity. Furthermore, experimental studies with a larger sample size or meta-analysis in different ethnic groups are needed to clarify the genetic association between CLU polymorphisms and epilepsy.