Session 3: Molecular Genetics of Autism

Edwin H Cook, Jr., M.D.

As a complex genetic disorder with both locus and allelic heterogeneity expected, several genes and variants within genes are expected. Several functional and positional candidates have been studied in our laboratory. Several functional (TPH,HTR1A,HTR1B,HTR2A,HTR2B,HTR2C,HTR5,HTR6,HTR7,PRKCG,OXTR,TNFA) and positional (15q11-q13 based on phenotype of maternal duplication/triplication: UBE3A,ATP10C,GABRA5,GABRG3) candidates have been screened without evidence in favor of transmission disequilibrium (TD). A few candidate genes have had nominal evidence of transmission disequilibrium, AVPR1A (one of two promoter microsatellites p < 0.05), GABRB3 (GABRB3 155CA-2 p .0014), and the serotonin transporter gene (SLC6A4). The GABRB3 155CA-2 marker is in the third intron and was never considered to be functional, but possibly in linkage disequilibrium (LD) with other variants on 15q11-q13. Based on Fisher's approach to meta-analysis (robust to heterogeneity expected in complex genetics), transmission disequilibrium is suggested but not definite when all studies are considered. A length variant with some functional effects in the promoter of the serotonin transporter gene (5-HTTLPR) has been the most well-studied polymorphism in autism. Since approximately 25% of probands with autism have hyperserotonemia and since about 50% of this subset were found to have high serotonin uptake in platelets, increased uptake was expected. Therefore, the original finding of transmission of the short allele was unexpected, given evidence across tissues that ss homozygotes have less serotonin transporter expression and function than ll homozygotes. In addition, in the original study, finding of stronger TD for a haplotype of 5-HTTLPR and an intron 2 VNTR suggested that 5-HTTLPR might only be in linkage disequilibrium (LD) with a nearby variant. Subsequent studies have shown mixed results for 5-HTTLPR with 3 studies showing preferential transmission of the long allele and three studies that showed no significant transmission of 5-HTTLPR. However, in one of these three follow-up studies from our laboratory (in review), a SNP in promoter 1B had p = 0.001. When data from both samples reported from our laboratory are combined, evidence is stronger at the SNP in promoter 1B (p 0.0005) relative to 5-HTTLPR (p 0.007). Furthermore, the most recently published study found heterogeneity of transmission between subjects with different levels of social impairment. In addition, examination of genotype-phenotype relationships at SLC6A4 should include examination of cognitive measures and restricted and repetitive behaviors. The nature of autism makes such studies difficult. Preliminary analyses of genotype-phenotype correlation between variants in SLC6A4 and serotonin transporter expression or function suggests that the intron 2 variants may mark another functional element in regulation of serotonin transporter expression. Furthermore, although most genome-wide linkage studies have not had strong linkage findings overlapping the serotonin transporter gene, recently a single-point lod score of 3.60 was found at the SLC6A4 intron 2 VNTR, although this may be due to a nearby gene (e.g. NF1). In addition to following up current candidates/regions with some evidence for transmission disequilibrium, positional candidates on 2q,7q,16p and 17q will be emphasized, particularly given the strength of recently published linkage findings.

Back to Agenda