Meta-analyses of the association between genetic polymorphisms of neurotrophic factors and schizophrenia

Abstract

Based on the neurodevelopmental hypothesis of schizophrenia, neurotrophic factors (NTFs) may be involved in its pathogenesis. Previous association studies between schizophrenia and neurotrophic factors have shown inconsistent results, which might be due to the heterogeneity and small sample size of the studies. To reach a conclusive understanding of the association, we used a meta-analytic method to study the association of schizophrenia with the polymorphisms in two candidate genes, ciliary neurotrophic factor (CNTF) and neurotrophin 3 (NT3). In our study, two meta-analyses were performed. One included eight studies examining the association of schizophrenia with the A3 allele in a dinucleotide repeat polymorphism of the NT3 gene promoter (N=1938). The other was employed in nine studies examining the association with a null mutation of the CNTF gene (N=2393). Neither of these analyses provided evidence for association. However, our sub-analyses showed a trend of association between the NT3 polymorphism and schizophrenics in Japanese, as well as an association between the CNTF null mutation and schizophrenics without psychiatric family history. These results suggested that the variations at the NT3 and the CNTF genes do not influence the schizophrenia risk, but a role in the susceptibility of subgroups of the patients cannot be excluded.

Introduction

Schizophrenia is a severe and complex mental disorder affecting approximately 1% of the general population. Although the clinical manifestations and disease course of schizophrenia have long been recognized, its pathogenesis and etiologies are largely unclear. Family, twin and adoption studies have suggested both genetic and environmental determinants to schizophrenia (Tsuang, 2000). To account for the risk factors, neuropathological findings, pathogenesis, and clinical course of schizophrenia, a neurodevelopmental hypothesis was proposed that postulated: (1) a disturbance in brain development occurs during pre- or peri-natal life, (2) this results in static brain lesions, and (3) the behavioral consequences remain latent for a substantial period after the action of the primary cause has ceased Weinberger, 1995, Weinberger and Lipska, 1995, Lewis and Levitt, 2002.

Neurotrophic factors (NTFs) play a critical role in neurodevelopment including neuronal survival, differentiation, and regulation of synaptic efficacy in the central nervous system Ip and Yancopoulos, 1996, Schuman, 1999, Huang and Reichardt, 2001. Two groups of NTFs have been identified based on their actions and signal transduction mechanisms (Thome et al., 1998). The first group are neurotrophins, which exhibit a high affinity for the tropomyosin-related kinase (trk) receptors and include nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin 3 (NT3), and neurotrophin 4/5 (NT 4/5). A second heterogeneous group, belonging to the cytokine family, includes ciliary neurotrophic factor (CNTF), leukemia inhibitory factor (LIF), and interleukin 6 (IL6). Based on the neurodevelopmental hypothesis, genes encoding these factors are potential candidate genes for schizophrenia.

The search for susceptibility genes involved in psychiatric diseases has focused mainly on linkage and case control association studies (Craddock and Owen, 1996). In particular, the latter has been used to test whether one potential candidate gene is involved in the disease of interest. Among the NTFs, the association of the NT3 and CNTF genes with schizophrenia has received extensive study.

The human NT3 gene maps to chromosome 12p13 (Maisonpierre et al., 1991). It includes one exon (1029 bp) and 82 single nucleotide polymorphisms (SNPs) (based on the latest updated National Center for Biotechnology Information (NCBI) database, http://www.ncbi.nlm.nih.gov/). In the developing brain, NT3 mRNA is by far the most abundant neurotrophin mRNA and its level decreases with maturation (Maisonpierre et al., 1990). In a post-mortem study, it was found that NT3 concentrations in the frontal and parietal cortices were significantly lower in schizophrenic patients (Durany et al., 2001). NT3-deficient mice displayed severe deficits in sensory and sympathetic neurons Ernfors et al., 1994, Farinas et al., 1994, but the motor neurons and the major anatomical brain regions seem to develop normally (Farinas et al., 1994). Taken together, these suggested that NT3 played an important role in early brain development and may be involved in the pathogenesis of schizophrenia.

An association between one of the alleles in a dinucleotide repeat polymorphism (CA)_n in the promotor region of the NT3 locus and schizophrenia was initially found in a Japanese study (Nanko et al., 1994), in which the A3 allele (carrying 23 CA repeats) frequency in schizophrenic patients was significantly higher than in the controls. This association was only replicated in another Japanese study (Hattori et al., 2002). Most later studies failed to replicate this finding Dawson et al., 1995, Nimgaonkar et al., 1995, Arinami et al., 1996, Gill et al., 1996, Jonsson et al., 1997, Virgos et al., 2001. However, Dawson et al. (1995) reported that there is a higher frequency of the A3 allele in male schizophrenics than male controls. Another NT3 polymorphism, causing a missense mutation with an amino acid substitution (Gly→Glu) at position 63, was found to be associated with a severe subgroup of schizophrenics (onset before the age of 25 years and a duration more than 10 years), but not with all schizophrenics (Hattori and Nanko, 1995). However, this mutation was found to be very rare in Caucasians Jonsson et al., 1997, Thome et al., 1997a, Virgos et al., 2001.

The CNTF gene maps to human chromosome 11q12 (Lev et al., 1993). It includes two exons (195 and 1696 bp) and 249 SNPs (NCBI database). In the developing brain, CNTF enhances neuronal survival in the hippocampus and increases the expression of cholinergic markers in neuronal precursors Ip et al., 1991, Rudge et al., 1996. The CNTF-deficient mice showed progressive atrophy and degeneration of spinal motor neurons (Masu et al., 1993). In addition, CNTF is suggested to be a protective factor for the injured neurons Siegel et al., 2000, Linker et al., 2002.

A null mutation of the CNTF gene (G to A transition close to the boundary between the first intron and the second exon) was found to be associated with endogenous psychosis (Thome et al., 1996b). This result raised special interest in examining whether alteration of the CNTF gene played any role in schizophrenia. However, most subsequent studies could not find any association between this mutant allele and schizophrenia Arinami and Toru, 1996, Li et al., 1996, Nothen et al., 1996, Thome et al., 1996a, Thome et al., 1997a, Thome et al., 1997b, Gelernter et al., 1997, Sakai et al., 1997, Tanaka et al., 1998, Virgos et al., 2001. Nevertheless, in one study with German subjects, the frequency of subjects carrying the mutant allele was significantly higher in patients with schizophrenic psychosis (Thome et al., 1996a). In addition, this null mutation of CNTF was found to be associated with schizoaffective disorder (Tanaka et al., 1998).

The inconsistent results of the association studies of both NT3 and CNTF with schizophrenia probably reflects the small sample sizes in the individual studies to detect small differences between cases and controls, or the heterogeneity in the sample composition. One way to address these issues is to perform meta-analyses. The goal of the present article is to examine the association of schizophrenia with NT3 and with CNTF from the published literature. To fulfill this goal, a systemic review of the literature was carried out using meta-analytic methods to pool results from all the published individual studies.