Abstract
Nearly 3% of the world ’s population is affected by some form of autoimmune disease, with current estimates of rheumatoid arthritis (RA) prevalence at 0.7-1% of Caucasians. This degenerative disorder is characterized by chronic inflammation of the joints resulting in irreversible joint damage and disability. RA has a complex aetiology with both environmental and genetic factors having a role. Numerous studies have identified genes that predispose individuals to RA and a large proportion of these have been focused on genes playing a role in the immune system. Mental illness is a common form of neurological disease, with estimates of 1 in 4 people experiencing some form of the disease during their lifetime. Schizophrenia (SZ) is one form of mental illness that affects 0.5-1% of the population. An inverse relationship between RA and SZ is well documented. Recent evidence suggests that there may be a common disease pathway for these two disorders. Recently, genes traditionally associated with RA have been implicated in SZ disease pathways and vice versa. AKT1 belongs to a pathway of genes involved in cell survival and function. A number of genes in this pathway have been implicated in autoimmune diseases. DISC1 is known to play a role in the development of SZ. Recent findings suggest this gene is implicated in the same pathway as AKT1.
The first aim of this study was to investigate AKT1 and genes involved in the same pathway for association with RA. Three candidate genes, DISC1, NFATC1 and NFATC2 were identified from RA genome wide association scan study results. Meta-analysis of genotyping results with Wellcome Trust Case Control Consortium (WTCCC RA data for AKT1 SNPs resulted significant associations for AKT1_rs2494731 (p = 0.033, OR = 1.08[1.0-1.2]) and AKT1_rs4983386 (p = 0.006, OR = 1.10[1.0-1.2]). Haplotype analysis of these two SNPs was significant (haplotype 21 p = 0.0009, OR = 1.473[1.2-1.8]) for a combined genotyping dataset (NZRA and UKRA). Conditional analysis provided evidence for independent effects at these loci (p = 0.415) It was therefore concluded that there is evidence for two independent effects of AKT1 in development of RA. Meta-analysis of genotyping and WTCCC RA data for NFATC2_rs8119787 was significant for association (p = 0.032, OR = 0.92[0.86-0.99]) whereas separate analyses for each sample set was not (WTCCC p = 0.057, NZRA p = .0.314 Three SNPs within DISC1 (DISC1_rs4658966p = 0.001 OR = 0.81[0.71-0.91], rs321577 p =0.018 OR = 0.90[0.83-0.98] and DISC1_rs872624p = 0.017 OR = 0.85[0.71-097]) were significantly associated with RA in a meta-analysis of genotyping and WTCCC RA data. Haplotype analysis for three DISC1 SNPs (DISC1_rs872625, DISC1_rs4658966, and DISC1_rs872624) provided evidence of association in three haplotypes (111 p = 0.013 OR = 1.30[1.06-1.61], 211 p = 0.002 OR = 0.60[0.44-0.83] and 21 p = 0.005 OR = 0.65[0.49- 0.88]).
The SNPs were investigated for association with SZ and a broader sample for mental illness (two SZ cohorts and bipolar disease, BD). None of the SNPs within AKT1, NFATC1, NFATC2 or DISC1 were significant for this analysis. All SNPs were also investigated for sex bias within each sample set. No significant associations were found when analyzing the sexes separately for AKT1 or NFATC2. For NFATC1_rs2002311 the results although significant were inconclusive. DISC1 SNP rs9431714 was significant for females in the WTCCC RA (p = 0.004) and Genetic Association Information Network (GAIN) SZ (p = 0.006) sample sets. For this SNP the OR of WTCCC conveyed susceptibility to RA (OR = 1.26[1.07-1.47]) and the OR of GAIN conveyed protection against SZ (OR = 0.88[0.72-1.08]). A similar, although borderline, effect was seen for the WTCCC RA and psychosis datasets for DISC1_rs872624 in females (WTCCC p = 0.056 OR = 0.82[0.67-1.01], psychosis p = 0.041 OR = 1.18[0.92- 1.52]). This analysis provides evidence for a shared disease pathway for RA and SZ in females at least.
This thesis also investigated SNPs from within the genes of interest (AKT1, NFAT and DISC1) in the SZ literature. A meta-analysis was carried out for the common SNPs in the literature for AKT1 and DISC1. For ATK1 none of SNPs examined were significant for association with SZ. This was the result when all information was analysed together and when it was separated for ethnicity. In DISC1 one SNP, rs821597 was significantly associated with SZ (p = 0.025) when the two studies from the literature were analysed in combination with GAIN SZ. This analysis provides additional support for the involvement of this DISC1 SNP in the development of SZ.
The last section of this thesis was focused on the involvement of the GRIK1 and GRIK2 genes in the development of inflammatory arthritis (IA) in a mouse model of the human disease. Recent data suggests a common pathway involving GRIK2 may link RA and SZ. A microsatellite assay, containing 57 microsatellite markers, was prepared to identify breeding animals for involvement in the serum transfer model of IA induction. Following the failure of this method the assay was modified for use in identifying animals for the spontaneous method of IA induction. The spontaneous model was yet to be tested at completion of this thesis.