5-HTTLPR In Relation to Behavioral and Emotional Self-Regulation In Children with High Functioning Autism

Background: Non-syndrome specific processes (e.g., genetics, temperament, family relationships) are hypothesized to alter the expression of autism and therefore contribute to the heterogeneous behavioral and emotional presentation of this disorder. Serotonergic neurons densely innervate the amygdala and AC regions (areas associated with emotional and behavioral self-regulation), thus implicating neurotransmitter genes (5-HTTLPR) as potential modifiers of behavioral and emotional functioning in children with autism.

Objectives: To examine genetic variability (5-HTTLPR) in relation to behavioral and electrophysiological indices of self-regulation and parent-reported behavior problems in a sample of children and adolescents with HFA and a matched comparison group.

Methods: Participants were 50 HFA and 40 age- and verbal IQ-matched comparison children, between the ages of 8 and 16. Participants completed a modified Eriksen Flanker task and saliva samples were collected for genetic analysis. Two fragments from the 5-HTTLPR polymorphism were extracted: short (S) and long (L). Participants were classified as having the S variant if they carried at least one short allele, and as having the L variant if both alleles were long. EEG was collected continuously during the Flanker task from 15 scalp sites using a Lycra stretch electrocap. 18 HFA and 19 comparison children had analyzable EEG data. Response-locked ERPs were created for error and correct trials separately and the amplitude of the error-related negativity (ERN) was scored at midline sites as the most negative peak occurring in the time window -20 to 150ms on error trials. Dependent measures of interest from the Flanker task were: ERN amplitude and self-correction rates. Parents completed the Behavioral Assessment System for Children (BASC-2).  

Results : A chi-square analysis revealed that 5-HTTLPR group did not differ between the diagnostic groups, χ²(1, N = 90)=.17, p=.812. A series of 2 (diagnostic group) x 2 (5-HTTLPR S vs. L) ANOVAs were conducted to examine associations with (a) ERN amplitude, (b) rates of self-correcting, and (c) parent reported behavior problems. Across both diagnostic groups, 5-HTTLPR allele frequency differences predicted ERN amplitude at FCz, F(1, 37)=4.00, p=.049, and Cz, F(1, 37)=4.43, p=.043, such that the L variant was associated with a smaller ERN amplitude. Additionally, the percent of self-corrected trials was predicted by (a) diagnostic group, (b) 5-HTTLPR group, and (c) the interaction of diagnostic and 5-HTTLPR group. HFA adolescents carrying the S variant self corrected significantly more than carriers of the L variant, t(27)=2.29, p=.030, while allelic frequency did not affect self correction rate for the comparison sample, t(19)=.05, p=.958. Finally, across both diagnostic groups, parent-reported Externalizing Problems were associated with 5-HTTLPR group, F(1, 78)=3.84, p=.052. The L variant was reported as exhibiting significantly more externalizing problems.

Conclusions: Results highlight the importance of genetic markers, such as 5-HTTLPR, in relation to behavior and physiological measures of self-regulation in both HFA and typically developing children. Together these results suggest that across all children, variation in the serotonin transporter gene may be related to emotional and behavioral differences in part through the effects of serotonin on the structure and function of neural systems related to self-regulation.