Possible association between autism and variants in the brain-expressed tryptophan hydroxylase gene (TPH2)

• Published in: American journal of medical genetics. Part B, Neuropsychiatric genetics Vol. 135B; no. 1; pp. 42 - 46
• Main Authors: Coon, Hilary, Dunn, Diane, Lainhart, Janet, Miller, Judith, Hamil, Cindy, Battaglia, Agatino, Tancredi, Raffaella, Leppert, Mark F., Weiss, Robert, McMahon, William
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 05.05.2005
• Subjects: association; autism; Autistic Disorder - enzymology; Autistic Disorder - genetics; Base Sequence; Brain - enzymology; DNA - chemistry; DNA - genetics; DNA Mutational Analysis; Female; Humans; Male; Nuclear Family; Phenotype; Polymorphism, Single Nucleotide; serotonin; tryptophan; Tryptophan Hydroxylase - genetics
• ISSN: 1552-4841; 1552-485X
• DOI: 10.1002/ajmg.b.30168

We report a possible association between autism in our sample and a recently described brain‐expressed tryptophan hydroxylase gene (TPH2). The well‐replicated involvement of the serotonin neurotransmitter system in autism has stimulated interest in many genes in the serotonin pathway as possible candidates for mutations leading to autism susceptibility. Serotonin synthesis is controlled by the rate‐limiting enzyme tryptophan hydroxylase. A mouse study of the original tryptophan hydroxylase gene (TPH1) and the new isoform (TPH2) showed that while TPH1 is primarily expressed peripherally, TPH2 is found exclusively in brain tissue. We searched for human sequence variants in 6,467 nucleotides covering all 11 exons of TPH2, and also 248 nucleotides upstream of the start codon, and 935 nucleotides downstream of the stop codon. Eighteen variants were characterized in 88 subjects with autism studied at our two centers, and 95 unrelated control subjects. Using a model‐free association method and empirical P value estimation, two variants showed frequency differences between autism and control subjects (P = 0.01 for a T‐G variant in intron 1, and P = 0.02 for a A‐T variant in intron 4). A haplotype including these variants showed slightly increased significance (P = 0.005). Further investigation of clinical phenotypes showed a possible association between presence of the variants at these two SNPs and higher scores on the Autism Diagnostic Interview (ADI) domain describing repetitive and stereotyped behaviors (P = 0.007). We conclude that TPH2 may play a modest role in autism susceptibility, perhaps relating specifically to repetitive behaviors, pending replication of this result. © 2005 Wiley‐Liss, Inc.