Association analysis of AKT1 and schizophrenia in a UK case control sample
Introduction
Emamian et al. (2004) proposed that alterations in brain protein kinase activity contribute to the aetiology of schizophrenia. In pursuit of this hypothesis, they examined the abundance of seven protein kinases in lymphoblast cell lines. Reduced AKT1 expression was found in cell lines derived from schizophrenic patients compared to controls, a finding subsequently confirmed in post-mortem frontal cortex and hippocampus. Moreover, they also found reduced phosphorylation of GSKβ3, a substrate of AKT1. These data provide a plausible case that AKT1 might be involved in the pathophysiology of schizophrenia, a hypothesis whose aetiological relevance they explored by genetic association using 5 SNPs spanning the gene (see Table 1, Table 2, Table 3 for SNP nomenclature) in 268 US families of North European origin containing one or more individuals with schizophrenia. The initial evidence for association was weak but one marker, (SNP3), yielded evidence for association (p = 0.05, uncorrected) as did a number of haplotypes (minimum p = 0.04, corrected) each of which shared alleles T and C at SNPs 2 and 3 (Emamian et al., 2004), (referenced hereafter as the core haplotype). The core haplotype was also associated with reduced AKT1 protein expression in 20 control lymphoblast cell lines. Follow-up studies in three independent Japanese samples gave mixed results. Two studies consisted of over 500 cases and over 400 controls. The first (Ohtsuki et al., 2004) found no association (allelic or haplotypic) while the second (Ikeda et al., 2004) reported weak evidence for association with a different variant and different haplotype to that of Emamian et al. (2004) (Table 2), with allele C of the core being carried in haplotypes that were both over and underrepresented in cases. A third study in a Japanese sample of 124 families found no association (allelic or haplotypic) (Ide et al., 2006). Schwab et al. (2005) found significant association with 3 of 7 SNPs tested in AKT1 in 79 sib pair families of German origin. The associated SNPs included SNP3, p = 0.027, which was nominally significant in the Emamian study as well as two other SNPs, with the strongest result (SNP2a, rs10149779, p =0.002) remaining significant after correction for multiple testing (p = 0.014). The most significant haplotype from the Emamian study (SNP2/SNP3/SNP4, TCG, Table 2) was also significantly over-represented in cases as was the TTA haplotype (formed by the same SNPs), which had been under-transmitted to cases in the study of Emamian et al. (2004) and which does not carry the core TC haplotype. Several other haplotypes created by various permutations of markers were also significantly over-transmitted with illness, with the strongest evidence coming from a haplotype derived from SNP1/SNP2a/SNP3 (p = 0.0013 corrected for multiple testing), Table 2. For all haplotypes in which SNP2 was included, the over-transmitted haplotype carried Emamian's core T allele at SNP2 but the finding of the earlier study was not precisely recapitulated since haplotypes carrying either C or T at core SNP3 were significantly over-transmitted.
Further studies have been less supportive (Bajestan et al., 2006, Liu et al., 2006). The 5 SNPs genotyped by Emamian et al. (2004) were genotyped in 218 families from Taiwan (Liu et al., 2006) with no significant association from either single markers or haplotypes. The same SNPs were also typed in an Iranian case control sample, (schizophrenia cases n = 321, controls n = 383) (Bajestan et al., 2006). Again, neither the SNPs nor the haplotypes from the Emamian study were associated. However, a novel five marker haplotype comprised of SNPs1-5, showed some evidence for association (global p = 0.05 uncorrected) with haplotype AGCAG being more frequent in cases compared to controls (uncorrected p = 0.004, Bonferroni corrected, p = 0.03, case freq 0.068, control freq 0.034). Given the diverse range of ethnicities studied so far, lack of consistency of the patterns of association between studies is potentially explicable in terms of population differences in LD and modest power to detect weak genetic effects. Moreover, in the light of partial replication of the original findings at the level of a specific haplotype in the only other European origin sample so far reported, AKT1 is clearly worth further investigation in other samples of broadly similar ethnicity.
We set out to investigate AKT1 in schizophrenia using a moderately large UK based case control sample under the following strategies. We genotyped SNPs 1–5 from Emamian et al. (2004), and additional markers reported by others, SNP1a, SNP2a, (Schwab et al., 2005) and SNPA (Ikeda et al., 2004). We specifically tested all significant associated haplotypes reported by Emamian et al. (2004), (n = 7), Ikeda et al. (2004), (n = 9), Schwab et al. (2005), (n = 23) and the Iranian 5 marker haplotype (Bajestan et al., 2006), (a total of 30 tests), although our primary hypothesis concerned the European origin haplotypes (n = 28). Additionally, we derived tagged SNPs across the AKT1 locus after genotyping all the above markers in the CEU panel used by the HapMap project and combining those data with all additional markers available in the HapMap (version 1.65) and performed two and three marker haplotype analyses for all marker combinations.
Section snippets
Subjects
All case-control subjects used in this study were unrelated Caucasians born in the UK or Ireland. All cases met DSM-IV criteria for schizophrenia. Consensus diagnoses were made by two raters from all available information following a semi-structured interview, SCAN or PSE (Wing et al., 1974, Wing et al., 1990), and examination of case notes. The cases consisted of 456 males and 217 females, average age at collection 44.5 years ± 14.6, whilst the controls consisted of 482 males and 234 females,
Results
Genotype data for SNP2, rs2494738, rs3803304 and SNPA from our assays in the same 90 CEPH DNA samples used in the International HapMap Project were 100% concordant with HapMap data. 100% concordance was also achieved between genotype data of 46 CEPH DNA samples typed in our initial assay optimisation stage and the same samples contained within our case control sample set for all 10 SNPs.
Genotype data were in Hardy Weinberg equilibrium for both cases and controls for all SNPs. No significant
Discussion
Following the initial report (Emamian et al., 2004) and mixed replication evidence (Ohtsuki et al., 2004, Ikeda et al., 2004, Ide et al., 2006, Schwab et al., 2005, Liu et al., 2006, Bajestan et al., 2006) we sought to provide further evidence for association between schizophrenia and polymorphisms in AKT1. The question of when the evidence for association between disease and gene is convincing is a vexing one for several reasons. Ideally, such evidence would come from repeated demonstration of
Contributors
NN, HW, SD, LC, TP performed laboratory assays. NN performed the data-analysis and drafted the manuscript. VM and RS advised on data-analysis. NW participated in the design of the study. IN was responsible for data-management. MI and NI provided haplotypic data and analysis from an independent sample. MOD and MJO participated in the design of the study, interpretation of the data, and drafting of the manuscript. All authors read and approved the final manuscript.
Role of funding source
This work was supported by the MRC (UK) and the National Institute of Mental Health Centers for the Neuroscience of Mental Disorders (Grant ID MH066392). V.M. was supported by a Research Councils UK Fellowship. The MRC and Research Councils UK had no further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.
Conflicts of interest
The author(s) declare that they have no conflicts of interest.
Acknowledgement
We thank SG Schwab and T Arinami for personal communications regarding the TCGG haplotype in their samples.
References (24)
- et al.
A powerful strategy to account for multiple testing in the context of haplotype analysis
Am. J. Hum. Genet.
(2004) - et al.
Association of AKT1 with schizophrenia confirmed in a Japanese population
Biol. Psychiatry
(2004) - et al.
Candidate-gene association studies of schizophrenia
Am. J. Hum. Genet.
(1999) - et al.
Further evidence for association of variants in the AKT1 gene with schizophrenia in a sample of European sib-pair families
Biol. Psychiatry
(2005) - et al.
GOLD-graphical overview of linkage disequilibrium
Bioinformatics
(2000) - et al.
Association of AKT1 haplotype with the risk of schizophrenia in Iranian population
Am. J. Med. Genet. B Neuropsychiatr. Genet.
(2006) - et al.
Haploview: analysis and visualization of LD and haplotype maps
Bioinformatics
(2005) Pedigree disequilibrium tests for multilocus haplotypes
Genet. Epidemiol.
(2003)- et al.
Convergent evidence for impaired AKT1-GSK3beta signaling in schizophrenia
Nat. Genet.
(2004) - et al.
Convergent evidence that oligodendrocyte lineage transcription factor 2 (OLIG2) and interacting genes influence susceptibility to schizophrenia
Proc. Natl. Acad. Sci. U. S. A.
(2006)
Miniaturized sealed-tubes allele-specific PCR
Human Mutat.
Failure to support a genetic contribution of AKT1 polymorphisms and altered AKT signaling in schizophrenia
J. Neurochem.
Cited by (54)
The Changing Legal Landscape of Cannabis Use and Its Role in Youth-onset Psychosis
2020, Child and Adolescent Psychiatric Clinics of North AmericaCitation Excerpt :Heritable variations in dopamine neurotransmission have been found to moderate individual sensitivity to the psychotogenic effects of cannabis. For example, the AKT1 pathway has been implicated in schizophrenia,79,80 is linked to dopamine signaling,81 and is activated by cannabinoids.82 The interaction between Protein kinase B (AKT1) and cannabis use predicted a 2-fold increase of risk of psychotic disorder with cannabis use,83,84 supporting a gene-environment interaction.
Interaction between childhood adversity and functional polymorphisms in the dopamine pathway on first-episode psychosis
2019, Schizophrenia ResearchCitation Excerpt :Moreover, recent meta-analyses concluded that there was no, or at best only weak, evidence of COMT genotype increasing risk for psychosis (Fan et al., 2005; Munafo et al., 2005; Ripke et al., 2013). Our findings on AKT1 rs2494732 polymorphism and psychotic disorder in this sample are in contrast with some previous studies that found an association between this candidate gene and schizophrenia (Bolog et al., 2012; Karege et al., 2012; Norton et al., 2007; Thiselton et al., 2008). Furthermore, despite the DRD2 gene being identified in a recent genome-wide association study (Ripke et al., 2013) as a risk gene for schizophrenia, our results did not support this finding in a more broadly defined psychotic disorder sample.
Brain, blood, cerebrospinal fluid, and serum biomarkers in schizophrenia
2018, Psychiatry ResearchThe effect of interactions between genetics and cannabis use on neurocognition. A review
2018, Progress in Neuro-Psychopharmacology and Biological PsychiatryAssociation of AKT1 gene polymorphisms with sporadic Parkinson's disease in Chinese Han population
2016, Neuroscience Letters