International Meeting for Autism Research: Identification of Molecular Pathways Associated with Autism by Genome Wide Expression Profiling of Lymphoblasts From Autism Patients

Identification of Molecular Pathways Associated with Autism by Genome Wide Expression Profiling of Lymphoblasts From Autism Patients

Thursday, May 12, 2011
Elizabeth Ballroom E-F and Lirenta Foyer Level 2 (Manchester Grand Hyatt)
2:00 PM
R. Luo1, I. Voineagu2, R. A. Mar-Heyming2, J. Ou3 and D. H. Geschwind4, (1)Human Genetics, University of California,Los Angeles, Los Angeles, CA, (2)Program in Neurogenetics, University of California, Los Angeles, Los Angeles, CA, (3)Neurology, UCLA, Los Angeles, CA, (4)Center for Neurobehavioral Genetics, University of California, Los Angeles, Los Angeles, CA
Background: Autism is a disorder of neural development that is characterized by deficits in social interaction and communication and the presence of restricted and repetitive behavior.  The etiology of autism is complex, with both genetic and environmental factors playing a role.  

Objectives: Our aim is to investigate the contribution of common variants to altered gene expression in autism and to identify molecular pathway related to autism disorder. 

Methods: We assayed the gene expression profile of lymphoblast cell lines (LCLs) from the Autism Genetic Resource Exchange (AGRE) by Illumina Human Ref8 version2 and the Simons Foundation Simplex Collection by Illumina Human Ref8 version3 by using two analyses: 1) standard differential expression (DE) analysis and 2) network analysis for analysis at the systems level. 

Results:  We compared gene expression level between 473 affected 274 unaffected siblings in AGRE and identified 1483 DE genes, some of which have been previously reported to be associated with autism. However, DE analysis failed to find consistent signal across AGRE and Simons. For the systems level analysis, we applied Weighted Gene Co-expression Network Analysis (WGCNA) to identify gene co-expression modules in cases and controls separately, in both AGRE and Simon’s data sets. One module that was differential between cases and controls and that was observed in both cohorts was enriched in genes involved in protein folding and the endoplasmic reticulum (ER) stress response. Analysis of genes showing the largest difference in network position between autism and controls (Differentially connectivity; DK) of genes in the ER module highlights CAMK1G and ARMET as two of the most different hub genes between autism and control. We observed CAMK1G differential expression in both lymphoblasts and brain (Voineagu I. et al, unpublished data), which suggests a possible important role in autism. 

Conclusions: Gene expression profiling of lymphoblast cell lines from a large number of autism cases and unaffected siblings identified transcriptional differences associated with autism, but these were non-overlapping between AGRE and Simons cohorts. This is consistent with significant heterogeneity and overall low signal using standard analyses of differential expression. By applying WGCNA, we are able to identify and replicate consistent changes across the two large cohorts, AGRE and Simons. This highlights several differentially connected genes including ARMET and CAMK1G, which are both involved in endoplasmic reticulum response. ARMET is also known to selectively promote the survival of dopaminergic neurons of the ventral mid-brain and modulate GABAergic transmission to the dopaminergic neurons of the substantia nigra. CAMK1G is associated with cytoskeletal re-organization and learning and memory.

See more of: Genetics and Genomics
See more of: Genetics
See more of: Biological Mechanisms
| More