Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams
Structural covariance network identifies FoxP2 gene allele-specific variations and its association to reading and language
Poster A39 in Poster Session A, Thursday, October 6, 10:15 am - 12:00 pm EDT, Millennium Hall
Nabin Koirala1, Anne Hooker2, Martina Villa2, Kelly Mahaffy2, Sara Mascheretti3, Meaghan Perdue5, Elena Grigorenko1,4, Nicole Landi1,2; 1Haskins Laboratories, 2University of Connecticut, 3IRCCS Eugenio Medea, Bosisio Parini, Italy, 4University of Houston, 5University of Calgary
The Forkhead box protein P2 (FOXP2) gene was first linked to language impairment in the KE family. Since then, genetic associations with several aspects of challenged speech and language function and the alterations in neural structure and function have been revealed. A few studies on FOXP2 single nucleotide polymorphisms (SNPs) have also demonstrated associations between specific SNPs and brain function and structure in the general population. However, the impact of allele-specific variation on neural structure and behavior has not been well-established. In this study, we investigated the allele-specific alterations in three different SNPs of FOXP2 gene which were previously linked to speech and language related alterations in brain function and structure. Genetic, neuroimaging and behavioral data from a total of 45 participants (mean age: 9.70 ± 1.31 years, 17 females) were analyzed for the study. Three SNPs, rs4727799, rs6980093, and rs10249531 were considered and, for each SNP, we compared ancestral to derived genotypes. Cortical morphometries (thickness, volume) were obtained using FreeSurfer toolbox from all participants using the T1-weighted MPRAGE anatomical scans. The obtained cortical measures were then analyzed using a Graph theoretical framework to compute different network measures for each SNP. Finally, relations between morphometry for the significant network of regions for each SNP and behavior (measures of reading and language) was evaluated using support vector regression analysis. We observed that path length and related network measures (global efficiency and closeness centrality) were significantly different between the ancestral and derived genotypes for all three SNPs. For rs4727799 and rs6980093, the ancestral genotype group exhibited significantly (p < 0.05, FDR corrected) lower path length and higher global efficiency and closeness centrality in a network of brain regions when compared to the derived genotype group. The network included left (lateral orbital sulcus, anterior insular cortex, precentral gyrus, sylvian fissure) and right (frontomarginal gyrus, pericallosal sulcus and superior parietal gyrus) brain regions. However, for SNP rs10249531, the derived allele carriers showed lower path length and higher global efficiency and closeness centrality in a different network of brain regions. This network included left (inferior prefrontal sulcus, medial occipito-temporal and lingual sulcus) and right (orbital H-shaped sulcus, transverse, and superior temporal cortex) brain regions. Finally, using cortical thickness from these regions we registered an association between language- and reading-related measures (the Woodcock Johnson Word Attack, PPVT, and CTOPP indicators) and genotype. We demonstrate that a structural covariance network approach can reveal FOXP2 allele-specific variations across a specific network of brain regions. Furthermore, the network of these regions was also predictive of reading and language scores. Importantly, the network of brain regions that differentiated ancestral and derived groups was different for the different SNPs indicating that variation across SNPs impacts different networks in the brain.
Topic Areas: History of the Neurobiology of Language, Speech Motor Control