Background
Schizophrenia spectrum disorders (SSD) are often characterised by a plateau or decline in cognitive abilities early in the prodrome. The cause of developmental alteration remains unknown, and investigation of genetic involvement in cognitive function in these disorders may assist the understanding of the underlying neurobiological mechanisms involved. Variation at two single nucleotide polymorphisms (SNPs) of the catechol-O-methyltransferase (COMT) gene have previously shown an influence on COMT protein levels and cognition; rs4680 and rs4818. Here we investigate the influence of the nonsynonymous “Val/Met” SNP rs4680 and a second functional SNP, rs4818, on tasks of cognitive flexibility and attention.
Methods
The sample comprised 48 healthy controls (HC; age = 31.95 ± 12.80; 25 males, 23 females), and 43 with a diagnosis of SSD (age = 41.64 ± 10.36; 26 males, 17 females). Measures of cognitive flexibility and attention included the Wisconsin Card Sorting Test (WCST), Continuous Performance Test-Identical Pairs version (CPT-IP), Trail Making Test (TMT), and the D-KEFS Colour Word Interference Test (CWIT). Due to small cohort sizes, in our preliminary analyses we chose to compare people who should be most severely affected because of inheriting COMT haplotypes associated with poor cognitive functioning (GG rs4818 / GG rs4680: G-G haplotype) to those with haplotypes associated with better cognitive functioning (CC rs4818 / AA rs4680: C-A haplotype). Multivariate analysis of variance factors included COMT haplotype, diagnosis (HC and SSD), and gender, with Bonferroni correction for multiple comparisons; age was included as a covariate. Analyses were also conducted based on a non-functional SNP of the COMT gene; rs165599, as a negative control.
Results
SSD exhibited reduced cognitive performance compared to HC; F(4, 75) = 8.810, p < .001. Investigation of C-A haplotype revealed an interaction with diagnosis on cognitive performance; F(8, 154) = 2.075, p = .041; SSD had reduced performance compared to HC for the WCST, CPT-IP, and TMT in C-A haplotypes (all p < .05). COMT haplotype also interacted with gender on cognitive performance (C-A haplotype; F(8, 154) = 2.315, p = .023, G-G haplotype; F(8, 154) = 2.706, p = .008). Males who were C-A non-carriers and /or G-G haplotype (high COMT activity groups) performed better on CPT-IP (both p < .05) and worse on CWIT (both p < .05) compared to females. Control SNP rs165599 revealed no main effects or significant interactions (all p > .05).
Discussion
The role of the COMT gene in the cognitive abilities of SSD remains contentious as gene expression does not differ from a healthy population. This preliminary analysis revealed an interaction between diagnosis and COMT haplotype, however, this only reached statistical significance for the C-A haplotype, where SSD with C-A haplotype and C-A non-carriers had reduced performance compared to HC on most tasks except TMT. The different effects found across the tasks, which probed various elements of cognitive flexibility and attention, supports a nuanced role of COMT in cognitive function. Further, high COMT activity was beneficial for males on CPT-IP but not CWIT compared to females. Gender interaction remains a significant consideration in studies of the COMT gene, likely involving the catechol-estrogens which are substrates of COMT. As expected there was no significant results with control SNP rs165599, indicating that findings were due to the influence of SNPs rs4680 and rs4818 on COMT activity.
