Letter to the Editor Potential involvement of serotonin receptor genes with age of onset and gender in schizophrenia: A preliminary study in a Spanish sample To the Editor: Schizophrenia is a severe disease characterized by psychiatric symptoms, including hallucinations, delusions, impaired information processing and mood disturbances. It is well known that sex and age at onset are essential to understand schizophrenia, because gender differences have etiological value (Salem and Kring, 1998; Seeman, 2004) and the age at onset is related to the course of the disease demonstrating its prognostic value (Peralta and Cuesta, 2007; Vahia et al., 2010). In this study, we investigated the association between these parameters and serotoninergic genes in schizophrenia. Forty-seven Single Nucleotide Polymorphisms (SNPs) from 17 representative genes concerning synthesis, transport and mechanism of serotonin action were genotyped in 196 Caucasian unrelated Spanish patients. Patients met the criteria for schizophrenia outlined in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). Men/women ratio was 2.2:1. Patients were divided into early onset (EO b 21 years) and non-early onset (NEO ≥ 21 years) subgroups following Peralta and Cuesta familial liability classification of schizophrenia (Peralta and Cuesta, 2007). Thirty-nine per cent of the subjects fall into the EO group. All SNPs were in Hardy-Weinberg equilibrium (PN 0.05). The call rate of successful genotypes across all SNPs was 0.95. According to QUANTO program (Gauderman, 2002), the sample allows a statistical power of 80% to detect moderate risks (OR ≥ 1.5) for susceptibility alleles showing a frequency higher than 25%, which occurs in the 84% of the alleles examined. Bonferroni multiple test correction was applied independently in allelic and genotypic analyses, taking into consideration the total number of SNPs studied. In the haplotypic analysis, the P-value obtained for each gene was corrected considering the number of different haplotypes shown by this gene. For the overall haplotype, the number of genes examined was considered. One SNP at the serotonin receptor 1F (HTR1F) gene, rs1503433, was associated to the age of onset at genotypic level assuming an overdominant model (OR: 2.78, CI 95%: 1.51-5.11; corrected P = 0.0456). Concretely there were significant differences in the frequency of the heterozygotes of 32.9% in EO and 57.6% in NEO patients. This result indicates that heterozygosity at this locus could protect patients against early onset of the disease. Because rs1503433 is in a very extensive linkage disequilibrium region at 5’ end of HTR1F, where potential functional polymorphisms could be found, the right level of HTR1F expression is a likely explanation of its heterosis effect. Evidence of molecular heterosis at different neurotransmitter genes has been previously reported (reviewed by Comings and MacMurray, 2000). One haplotype (A-A-G) of the serotonin receptor 2A (HTR2A) gene and one haplotype (G-A-G) of the serotonin receptor 6 (HTR6) gene were both more frequent in EO than in NEO patients (corrected P =0.0427 and 0.014 respectively, Table 1). Both HTR2A and HTR6 haplotypes may affect the gene expression because the corresponding SNPs are located at the promoter and other potential regulatory sequences of these genes. Therefore these haplotypes may contribute to overall imbalance of serotoninergic function affecting other neuro- transmitter systems that interact with serotonin. As a result, these haplotypes could become risk factors for an earlier onset of schizophre- nia. Our findings agree with previous results connecting the age of onset and a silent HTR2A polymorphism (Joober et al., 1999) that is in linkage disequilibrium with another one located at the promoter region. When patients were grouped by gender, we found a haplotype (A-A- A-C) of the serotonin receptor 1B (HTR1B) gene that was significantly more frequent in women than men (corrected P = 0.0382, Table 1). The SNPs encompassed in this haplotype are also located in the promoter and regulatory sequences of HTR1B, therefore with possible influence on gene expression. Because it has been proposed that brain serotoninergic neurotransmission differs between women and men (Jovanovic et al., 2008), differences in gene expression could contribute to sex differences in the prevalence of schizophrenia. Our results support recent findings of gender differences in the molecular genetics of schizophrenia (Hoenicka et al., 2010). The results of this preliminary study support further research to examine the relationship between several serotonin receptor genes and age of onset and gender differences in vulnerability to schizophrenia. To be confirmed, our findings need to be replicated in a larger sample. On the other hand, because most of the SNPs showing significant association are located in the promoter Psychiatry Research 186 (2011) 153–154 Table 1 Significant association results at haplotypic level between several serotonin receptor genes and patients with schizophrenia. Patients grouped by age of onset Gene SNP Haplotype Estimated frequency in EO Estimated frequency in NEO P-value Corrected P-value by Bonferroni HTR2A rs4941573 A-A-G 0.0942 0.0244 0.00854 0.0427 rs1328685 rs4142900 Overall 0.048 0.816 HTR6 rs2314331 G-A-G 0.533 0.443 0.00372 0.014 rs6693503 rs4912138 Overall 0.0289 0.49 Patients grouped by gender Gene SNP Haplotype Estimated frequency in women Estimated frequency in men P-value Corrected P-value by Bonferroni HTR1B rs6297 A-A-A-C 0.511 0.357 0.00822 0.0328 rs130058 rs1213366 rs1213371 Overall 0.0571 0165-1781/$ – see front matter © 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2010.07.005 Contents lists available at ScienceDirect Psychiatry Research journal homepage: www.elsevier.com/locate/psychres