23022
Common CNTNAP2 Variant Relates to Altered Functional Connectivity of the Striatum

Thursday, May 12, 2016: 11:30 AM-1:30 PM
Hall A (Baltimore Convention Center)
K. E. Lawrence1, L. M. Hernandez2, D. H. Geschwind3, S. Y. Bookheimer1 and M. Dapretto1, (1)University of California, Los Angeles, Los Angeles, CA, (2)University of California Los Angeles, Los Angeles, CA, (3)Department of Neurology, UCLA, Los Angeles, CA
Background: Contactin associated protein-like 2 (CNTNAP2) plays an important role in the development of frontostriatal networks and has previously been associated with language endophenotypes in autism spectrum disorder (ASD) and specific language impairment (Penagarikano & Geschwind, 2012). CNTNAP2 messenger RNA is expressed in frontostriatal networks of the human brain, and common variants in CNTNAP2 have previously been related to brain connectivity (Dennis et al., 2011; Tan et al., 2010), as well as activity during language processing (Koeda et al., 2015; Whalley et al., 2011) and implicit learning (Scott-Van Zeeland et al., 2010a). However, no studies to date have investigated how CNTNAP2 may relate to functional connectivity of brain structures that are involved in language learning. A prior study from our lab demonstrated that, unlike typically developing (TD) controls, individuals with ASD do not display increased striatal activity during an implicit language-learning task (Scott-Van Zeeland et al., 2010b). Therefore, in order to characterize how this neural circuitry may be influenced by genetic risk, we investigated the relationship between CNTNAP2rs2710102 (previously associated with age at first word in ASD) and striatal connectivity in youth with and without ASD.

Objectives: Examine how a common CNTNAP2 variant (rs2710102) relates to functional connectivity of the striatum, an important structure for implicit language learning.

Methods: Thirty-seven children and adolescents with ASD and 30 matched TD controls completed a 6-minute resting-state functional MRI scan. Resting-state functional connectivity analyses were conducted using the left caudate and the left putamen as seeds. As prior imaging genetics studies have used either dominant or recessive modes of inheritance (Dennis et al., 2011; Scott-Van Zeeland et al., 2010a; Whalley et al., 2011), we analyzed the data separately using each model. Initial group comparisons were completed in FSL, prethresholded with a joint mask of the within group results and thresholded at Z>2.3 with a cluster threshold of p=.05.  

Results: Under the dominant and recessive models, the risk and non-risk groups both displayed robust functional connectivity between the left caudate and frontal, temporal, and parietal regions. In the dominant model, the non-risk group displayed significantly greater connectivity than the risk group to the medial prefrontal cortex, a region in which CNTNAP2is known to be expressed (Abrahams et al., 2007). When using a recessive model, the risk group showed significantly greater connectivity to bilateral frontal pole regions than the non-risk group. With regards to left putamen connectivity, both groups displayed robust connectivity to frontal and temporal areas for the dominant and recessive models. There were no significant between-group differences in putamen functional connectivity regardless of model.  

Conclusions: Connectivity of the striatum, an area previously associated with language learning differences between ASD and typical development, is significantly modulated by a common genetic variant that has been linked to language in ASD.