17227
Heterogeneity in 5-Httlpr Genotype-Phenotype Effects
Initially, overtransmission of the low-expressing short (S) allele of the 5-HTTLPR to probands with autism from their parents was found. Subsequently, the most complete meta-analysis [PMID 16103890] provided replication of this finding but more importantly provided statistical support for the heterogeneity across studies. Subsequent studies have examined potential sources of heterogeneity, including potential relationship to phenotype [PMID 17151167] with an indication that the “protective” allele for ASD, the higher expressing long (L) allele may be a risk allele for aspects of restricted and repetitive behaviors. More recently, evidence has been provided that a SNP (rs25531) within the 5HTTLPR leads to a reclassification of long alleles into high expressing (La) and lower expressing (Lg).
Objectives: To determine if the low expressing alleles (S and Lg) are overtransmitted to subjects with ASD.
To determine if the high expressing allele (La) is associated with higher scores on the Restricted and Repetitive Behavior-Revised Compulsions/Rituals/Insistence on Sameness factor (RBS-R CRS).
Methods:
The study was conducted as part of Project I of the IRB approved study, “Autism Center of Excellence: Translational Studies of Insistence on Sameness in Autism.” Subjects ranging in age from 3.16-44.3 years met ADI-R and ADOS criteria on both instruments for either autism or autism spectrum disorder. They had clinical confirmation of an ASD by both a psychologist and a child and adolescent psychiatrist. Genotyping was performed by PCR amplification and capillary electrophoresis sizing as previously reported [PMID: 11920155] with an additional restriction digest with MspI to genotype rs25531. Genotyping occurred after collection of phenotype measures and was blind to all phenotype measures. Genetic analysis was performed using FBAT for affection (transmission to ASD probands) and the quantitative phenotype RBS-R CRS as the primary quantitative trait hypothesis. Age and sex were tested for association with quantitative phenotypes and were adjusted for when p<0.10.
Results: No Mendelian errors were identified for 5-HTTLPR and the data were consistent with HWE (p=0.54). FBAT revealed overtransmission of the low-expressing alleles (S and Lg) to those affected with ASD (all ancestry groups combined: 153 complete nuclear families, S-E(S) ;14.5, Var(S) 14.5, z 2.29, p = 0.022; European ancestry (EA): 107 complete nuclear families, S-E(S) 11, Var(S) 29.5, z 2.025, p = 0.043). In contrast, RBS-R CRS scores were higher when the La allele was transmitted (all ancestry groups combined: 150 complete nuclear families, S-E(S) ;147.85, Var(S) 4361.70, z 2.24, p = 0.025; EA: 105 complete nuclear families, S-E(S) 150.485, Var(S) 2893.819, z 2.78, p = 0.005).
Conclusions:
The study found a pattern which may partially account for the heterogeneity of studies of 5-HTTLPR. We found overtransmission of the low-expressing S and Lg alleles to probands with ASD yet found that the high expressing La allele was associated with higher compulsion, ritual and insistence on sameness factor scores. This paradox may account for heterogeneous results in samples that differ in severity of CRS symptoms.