NO is a pleiotropic signaling molecule and has an important role for cognition and emotion. In the brain, NO is produced by neuronal nitric oxide synthase (NOS-I, encoded by NOS1) coupled to the NMDA receptor via PDZ interactions; this protein-protein interaction is disrupted upon binding of NOS1 adaptor protein (encoded by NOS1AP) to NOS-I. As both NOS1 and NOS1AP were associated with schizophrenia, we here investigated these genes in greater detail by genotyping new samples and conducting a meta-analysis of our own and published data. In doing so, we confirmed association of both genes with schizophrenia and found evidence for their interaction in increasing risk towards disease. Our strongest finding was the NOS1 promoter SNP rs41279104, yielding an odds ratios of 1.29 in the metaanalysis. As findings from heterologous cell systems have suggested that the risk allele decreases gene expression, we studied the effect of the variant on NOS1 expression in human post-mortem brain samples and found that the risk allele significantly decreases expression of NOS1 in the prefrontal cortex. Bioinformatic analyses suggest that this might be due the replacement of six transcription factor binding sites by two new binding sites as a consequence of proxy SNPs. Taken together, our data argue that genetic variance in NOS1 resulting in lower prefrontal brain expression of this gene contributes to schizophrenia liability, and that NOS1 interacts with NOS1AP in doing so. The NOS1-NOS1AP PDZ interface may thus well constitute a novel target for small molecules in at least some forms of schizophrenia.

Attention deficit/ hyperactivity disorder (ADHD) is a heritable neurodevelopmental disorder featuring complex genetics with common and rare variants contributing to disease risk. In a high proportion of cases, ADHD does not remit during adolescence but persists into adulthood. Several studies suggest that NOS1, encoding nitric oxide synthase I, producing the gaseous neurotransmitter NO, is a candidate gene for (adult) ADHD. We here extended our analysis by increasing the original sample, adding two further samples from Norway and Spain, and conducted subgroup and co-morbidity analysis. Our previous finding held true in the extended sample, and also meta-analysis demonstrated an association of NOS1 ex1f-VNTR short alleles with adult ADHD (aADHD). Association was restricted to females, as was the case in the discovery sample. Subgroup analysis on the single allele level suggested that the 21-repeat allele caused the association. Regarding subgroups, we found that NOS1 was associated with the hyperactive/impulsive ADHD subtype, but not to pure inattention. In terms of comorbidity, major depression, anxiety 445Neuropsychiatric Genetics disorders, cluster C personality disorders and migraine were associated with short repeats, in particular the 21-repeat allele. Also, short allele carriers had significantly lower IQ. Finally, we again demonstrated an influence of the repeat on gene expression in human post-mortem brain samples. These data validate the role of NOS-I in hyperactive/impulsive phenotypes and call for further studies into the neurobiological underpinnings of this association.

Attention-deficit/hyperactivity disorder (ADHD) is the most frequent psychiatric disorder in children, where it displays a global prevalence of 5 %. In up to 50 % of the cases, ADHD may persist into adulthood (aADHD), where it is often comorbid with personality disorders. Due to a potentially heritable nature of this comorbidity, we hypothesized that their genetic framework may contain common risk-modifying genes. SPOCK3, a poorly characterized, putatively Ca(2+)-binding extracellular heparan/chondroitin sulfate proteoglycan gene encoded by the human chromosomal region 4q32.3, was found to be associated with polymorphisms among the top ranks in a genome-wide association study (GWAS) on ADHD and a pooled GWAS on personality disorder (PD). We therefore genotyped 48 single nucleotide polymorphisms (SNPs) representative of the SPOCK3 gene region in 1,790 individuals (n aADHD = 624, n PD = 630, n controls = 536). In this analysis, we found two SNPs to be nominally associated with aADHD (rs7689440, rs897511) and four PD-associated SNPs (rs7689440, rs897511, rs17052671 and rs1485318); the latter even reached marginal significance after rigorous Bonferroni correction. Bioinformatics tools predicted a possible influence of rs1485318 on transcription factor binding, whereas the other candidate SNPs may have effects on alternative splicing. Our results suggest that SPOCK3 may modify the genetic risk for ADHD and PD; further studies are, however, needed to identify the underlying mechanisms.

Our preliminary results demonstrate MPH treatment differences in ADHD patients and healthy controls in a peripheral primary cell model. Our results need to be replicated in larger samples and also using patient-derived neuronal cell models to validate the contribution of those genes to the pathophysiology of ADHD and mode of action of MPH.