Slide Slam N5
The effect of COMT on reading is mediated by top-down fronto-striatal activation.
Martina Villa1,2, Sara Mascheretti2,3, Meaghan Perdue1,2, Bei Feng4, Valentina Lampis3, Denis Peruzzo3, Ginette Dionne4, Pugh Kenneth1,2, Elena Grigorenko2,5, Nicole Landi1,2,4,5; 1Department of Psychological Sciences, University of Connecticut, Storrs, CT, USA, 2Haskins Laboratories, New Haven, CT, USA, 3Child Psychopathology Unit, Scientiﬁc Institute, IRCCS E. Medea, Bosisio Parini, LC, Italy, 4School of Psychology, Université Laval, Québec, Canada, 5Texas Institute for Measurement, Evaluation, and Statistics, University of Houston, Houston, TX, USA, 6Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA, 7St. Petersburg State University, Russia
Introduction: The COMT gene codes for the Catechol-O-methyltransferase enzyme, which metabolizes released dopamine in the prefrontal cortex and is a strong regulator of prefrontal dopamine levels. Variation at codon 158 of the COMT gene results in a valine (Val)-to-methionine (Met) substitution, which has been associated with functional alterations within the prefrontal cortex, and impacting several prefrontally mediated cognitive functions and neurodevelopmental disorder susceptibility. Recently, our group found that the COMT Val158Met polymorphism was associated with reading-related skills, specifically phonological awareness, spelling and reading comprehension . We also showed that the polymorphism was related to functional neural activation during reading in a number of reading-related regions. These findings implicated COMT as a possible risk gene for reading disability (RD). Here, we hypothesize the pathway from gene to reading is mediated by intermediate phenotypes (IPs), specifically brain activation and PA. Methods: We conducted a serial multiple mediation model in a sample of 97 children (age 5–13) using genotype groups (i.e. Val/Val, Val/Met, Met/Met) as predictors, brain activation and phonological processing (PP) as serial multiple mediators, and reading as outcome. The model was tested on a set of 6 ROIs where we previously saw activation differences during reading as a function of genotype group, these include: the right middle frontal gyrus (rMFG), left temporal pole, right precentral gyrus, left inferior frontal gyrus, left parahippocampal gyrus, and left inferior frontal gyrus/insula. PP and reading were obtained by a PCA with letter-word identification, pseudoword reading, spelling, passage comprehension, blending words, memory for digits, non-word repetition, blending non-words, reading sight words, and decoding non-words, as variables. Indirect effects from COMT Val158Met polymorphism to reading via brain activation and PP were tested by using Structural Equation Modelling. We controlled for the effect of age on both behavioral scores (i.e., PP and reading) and brain activation. Results: There was no signiﬁcant combined indirect effect from COMT Val158Met polymorphism to reading via brain activation and PP jointly. However, there was a specific indirect effect of the COMT-Val158Met polymorphism to reading via brain activation in the rMFG (Met/Met vs. Val/Met: β=-0.167, SE=0.088, 95% CI=-0.367/-0.014; and Met/Met vs. Val/Val: β=-0.177, SE=0.097, 95% CI=-0.390/-0.013). Specifically, both the Val/Met and Val/Val genotype groups had lower activation relative to the Met/Met group in rMFG, and brain activation in this ROI was positively related to reading. The combined indirect effect of Val/Met and Val/Val genotype groups to reading via activation in rMFG was significant (β=-0.344, SE=0.178, 95% CI=-0.732/-0.029). The mediation model for the rMFG explained on average 25% of the reading variance. Conclusion: These findings suggest that the COMT Val158Met polymorphism is related to reading via top-down cognitive influences connected to the activity in fronto-striatal networks rather than via PP. We note that these findings are preliminary due to relatively low power (small sample size for genetic and mediation analyses), and require replication. Other relevant behavioral/cognitive abilities (e.g. working memory and executive functions) remain to be explicitly tested in future studies. References  doi: 10.1111/j.1467-7687.2012.01180.x.