Carepath for overcoming psychosis early (COPE): first 5 years of clinical operation and prospective research in the Cavan-Monaghan early intervention service.

OBJECTIVES: As Ireland confronts the many challenges of broadening the introduction of early intervention services (EIS) for first episode psychosis (FEP) as national policy, this article describes Carepath for Overcoming Psychosis Early (COPE), the EIS of Cavan-Monaghan Mental Health Service, and presents prospective research findings during its first 5 years of operation. METHODS: COPE was launched as a rural EIS with an embedded research protocol in early 2012, following an education programme for general practitioners (GPs). Here, operational activities are documented and research findings presented through to late 2016. RESULTS: During this period, 115 instances of FEP were incepted into COPE, 70.4% via their GP and 29.6% via the Emergency Department. The annual rate of inception was 24.8/100,000 of population aged > 15 years and was 2.1-fold more common among men than women. Mean duration of untreated psychosis was 5.7 months and median time from first psychotic presentation to initiation of antipsychotic treatment was zero days. Assessments of psychopathology, neuropsychology, neurology, premorbid functioning, quality of life, insight, and functionality compared across 10 DSM-IV psychotic diagnoses made at six months following presentation indicated minimal differences between them, other than more prominent negative symptoms in schizophrenia and more prominent mania in bipolar disorder. CONCLUSIONS: COPE illustrates the actuality of introducing and the challenges of operating a rural EIS for FEP. Prospective follow-up studies of the 5-year COPE cohort should inform on the effectiveness of this EIS model in relation to long-term outcome in psychotic illness across what appear to be arbitrary diagnostic boundaries at FEP.

Forging successful partnerships in psychosis research: lessons from the Cavan-Monaghan First Episode Psychosis Study.

Embedding psychosis research within community mental services is highly desirable from several perspectives but can be difficult to establish and sustain, especially when the clinical service has a rural location at a distance from academic settings with established research expertise. In this article, we share the experience of a successful partnership in psychosis research between a rural Irish mental health service and the academic department of a Dublin medical school that has lasted over 30 years. We describe the origins and evolution of this relationship, the benefits that accrued and the challenges encountered, from the overlapping perspectives of the academic department, the mental health service and psychiatric training. We discuss the potential learning that arose from the initiative, particularly for national programme planning for early intervention in psychosis, and we explore the opportunities for enhanced training, career development and professional reward that can emerge from this type of partnership.

Phenotypic effects of maternal immune activation and early postnatal milieu in mice mutant for the schizophrenia risk gene neuregulin-1.

Risk of schizophrenia is likely to involve gene × environment (G × E) interactions. Neuregulin 1 (NRG1) is a schizophrenia risk gene, hence any interaction with environmental adversity, such as maternal infection, may provide further insights into the basis of the disease. This study examined the individual and combined effects of prenatal immune activation with polyriboinosinic-polyribocytidilic acid (Poly I:C) and disruption of the schizophrenia risk gene NRG1 on the expression of behavioral phenotypes related to schizophrenia. NRG1 heterozygous (NRG1 HET) mutant breeding pairs were time-mated. Pregnant dams received a single injection (5mg/kg i.p.) of Poly I:C or vehicle on gestation day 9 (GD9). Offspring were then cross-fostered to vehicle-treated or Poly I:C-treated dams. Expression of schizophrenia-related behavioral endophenotypes was assessed at adolescence and in adulthood. Combining NRG1 disruption and prenatal environmental insult (Poly I:C) caused developmental stage-specific deficits in social behavior, spatial working memory and prepulse inhibition (PPI). However, combining Poly I:C and cross-fostering produced a number of behavioral deficits in the open field, social behavior and PPI. This became more complex by combining NRG1 deletion with both Poly I:C exposure and cross-fostering, which had a robust effect on PPI. These findings suggest that concepts of G × E interaction in risk of schizophrenia should be elaborated to multiple interactions that involve individual genes interacting with diverse biological and psychosocial environmental factors over early life, to differentially influence particular domains of psychopathology, sometimes over specific stages of development.

Neurodevelopmental disorders: mechanisms and boundary definitions from genomes, interactomes and proteomes.

Neurodevelopmental disorders such as intellectual disability, autism spectrum disorder and schizophrenia lack precise boundaries in their clinical definitions, epidemiology, genetics and protein-protein interactomes. This calls into question the appropriateness of current categorical disease concepts. Recently, there has been a rising tide to reformulate neurodevelopmental nosological entities from biology upward. To facilitate this developing trend, we propose that identification of unique proteomic signatures that can be strongly associated with patient's risk alleles and proteome-interactome-guided exploration of patient genomes could define biological mechanisms necessary to reformulate disorder definitions.

Phenotypic characterization of orofacial movement topography in mutants with disruption of amino acid mechanisms: glutamate N2A/B/D [GluRε1/2/4] subtypes and the GABA synthesizing enzyme GAD65.

To investigate the role of glutamate receptor subtypes and GABA in orofacial function, six individual topographies of orofacial movement, both spontaneous and induced by the dopamine D1-like receptor agonist [R/S]-3-methyl-6-chloro-7,8-dihydroxy-1-[3-methyl-phenyl]-2,3,4,5-tetrahydro-1H-3-benzazepine (SKF 83959), were quantified in mutant mice with deletion of (a) GluN2A, B or D receptors, and (b) the GABA synthesizing enzyme, 65-kD isoform of glutamate decarboxylase (GAD65). In GluN2A mutants, habituation of head movements was disrupted and vibrissae movements were reduced, with an overall increase in locomotion; responsivity to SKF 83959 was unaltered. In GluN2B mutants, vertical and horizontal jaw movements and incisor chattering were increased, with an overall decrease in locomotion; under challenge with SKF 83959, head and vibrissae movements were reduced. In GluN2D mutants, horizontal jaw movements, incisor chattering and vibrissae movements were increased, with reduced tongue protrusions and no overall change in locomotion; under challenge with SKF 83959, horizontal jaw movements were increased. In GAD65 mutants, vertical jaw movements were increased, with disruption to habituation of locomotion; under challenge with SKF 83959, vertical and horizontal jaw movements and incisor chattering were decreased. Effects on orofacial movements differed from their effects on regulation of overall locomotor behavior. These findings (a) indicate novel, differential roles for GluN2A, B and D receptors and for GAD65-mediated GABA in the regulation of individual topographies of orofacial movement and (b) reveal how these roles differ from and/or interact with the established role of D1-like receptors in pattern generators and effectors for such movements.

Diagnostic trajectory, interplay and convergence/divergence across all 12 DSM-IV psychotic diagnoses: 6-year follow-up of the Cavan-Monaghan First Episode Psychosis Study (CAMFEPS).

BACKGROUND: The boundaries of psychotic illness and the extent to which operational diagnostic categories are distinct in the long term remain poorly understood. Clarification of these issues requires prospective evaluation of diagnostic trajectory, interplay and convergence/divergence across psychotic illness, without a priori diagnostic or other restrictions. METHOD: The Cavan-Monaghan First Episode Psychosis Study (CAMFEPS), conducted using methods to attain the closest approximation to epidemiological completeness, incepts all 12 DSM-IV psychotic diagnoses. In this study we applied methodologies to achieve diagnostic reassessments on follow-up, at a mean of 6.4 years after first presentation, for 196 (97%) of the first 202 cases, with quantification of prospective and retrospective consistency. RESULTS: Over 6 years, the 12 initial psychotic diagnoses were characterized by numerous transitions but only limited convergence towards a smaller number of more stable diagnostic nodes. In particular, for initial brief psychotic disorder (BrP), in 85% of cases this was the harbinger of long-term evolution to serious psychotic illness of diagnostic diversity; for initial major depressive disorder with psychotic features (MDDP), in 18% of cases this was associated with mortality of diverse causality; and for initial psychotic disorder not otherwise specified (PNOS), 31% of cases continued to defy DSM-IV criteria. CONCLUSIONS: CAMFEPS methodology revealed, on an individual case basis, a diversity of stabilities in, and transitions between, all 12 DSM-IV psychotic diagnoses over 6 years; thus, psychotic illness showed longitudinal disrespect to current nosology and may be better accommodated by a dimensional model. In particular, a first episode of BrP or MDDP may benefit from more vigorous, sustained interventions.

D-amphetamine and antipsychotic drug effects on latent inhibition in mice lacking dopamine D2 receptors.

Drugs that induce psychosis, such as D-amphetamine (AMP), and those that alleviate it, such as antipsychotics, are suggested to exert behavioral effects via dopamine receptor D2 (D2). All antipsychotic drugs are D2 antagonists, but D2 antagonism underlies the severe and debilitating side effects of these drugs; it is therefore important to know whether D2 is necessary for their behavioral effects. Using D2-null mice (Drd2-/-), we first investigated whether D2 is required for AMP disruption of latent inhibition (LI). LI is a process of learning to ignore irrelevant stimuli. Disruption of LI by AMP models impaired attention and abnormal salience allocation consequent to dysregulated dopamine relevant to schizophrenia. AMP disruption of LI was seen in both wild-type (WT) and Drd2-/-. This was in contrast to AMP-induced locomotor hyperactivity, which was reduced in Drd2-/-. AMP disruption of LI was attenuated in mice lacking dopamine receptor D1 (Drd1-/-), suggesting that D1 may play a role in AMP disruption of LI. Further supporting this possibility, we found that D1 antagonist SKF83566 attenuated AMP disruption of LI in WT. Remarkably, both haloperidol and clozapine attenuated AMP disruption of LI in Drd2-/-. This demonstrates that antipsychotic drugs can attenuate AMP disruption of learning to ignore irrelevant stimuli in the absence of D2 receptors. Data suggest that D2 is not essential either for AMP to disrupt or for antipsychotic drugs to reverse AMP disruption of learning to ignore irrelevant stimuli and further that D1 merits investigation in the mediation of AMP disruption of these processes.

Mood congruent psychotic symptoms and specific cognitive deficits in carriers of the novel schizophrenia risk variant at MIR-137.

OBJECTIVE: The Schizophrenia Psychiatric Genome-wide Association (GWAS) Consortium recently reported on five novel schizophrenia susceptibility loci. The most significant finding mapped to a micro-RNA, MIR-137, which may be involved in regulating the function of other schizophrenia and bipolar disorder susceptibility genes. METHOD: We genotyped 821 patients with confirmed DSM-IV diagnoses of schizophrenia, bipolar affective disorder I and schizoaffective disorder for the risk SNP (rs1625579) and investigated the clinical profiles of risk allele carriers using a within-case design. We also assessed neurocognitive performance in a subset of cases (n=399) and controls (n=171). RESULTS: Carriers of the risk allele had lower scores for an OPCRIT-derived positive symptom factor (p=0.04) and lower scores on a lifetime measure of psychosis incongruity (p=0.017). Risk allele carriers also had more cognitive deficits involving episodic memory and attentional control. CONCLUSION: This is the first evidence that the MIR-137 risk variant may be associated with a specific subgroup of psychosis patients. Although the effect of this single SNP was not clinically relevant, investigation of the impact of carrying multiple risk SNPs in the MIR-137 regulatory network on diagnosis and illness profile may be warranted.

Disruption of exploratory and habituation behavior in mice with mutation of DISC1: an ethologically based analysis.

Disrupted-in-schizophrenia-1 (DISC1) is a gene that has been functionally linked with neurodevelopmental processes and structural plasticity in the brain. Clinical genetic investigations have implicated DISC1 as a genetic risk factor for schizophrenia and related psychoses. Studies using mutant mouse models of DISC1 gene function have demonstrated schizophrenia-related anatomical and behavioral endophenotypes. In the present study, ethologically based assessment of exploratory and habituation behavior in the open field was conducted in DISC1 (L100P), wild-type (WT), heterozygous (HET), and homozygous (HOM) mutant mice of both sexes. Ethological assessment was conducted in an open-field environment to explore specific topographies of murine exploratory behavior across the extended course of interaction from initial exploration through subsequent habituation (the ethogram). During initial exploration, HET and HOM DISC1 mutants evidenced increased levels of locomotion and rearing to wall compared with WT. A HOM-specific increase in total rearing and a HET-specific increase in sifting behavior and reduction in rearing seated were also observed. Over subsequent habituation, locomotion, sniffing, total rearing, rearing to wall, rearing free, and rearing seated were increased in HET and HOM mutants vs. WT. Overall, grooming was increased in HOM relative to other genotypes. HET mice displayed a selective decrease in habituation of sifting behavior. These data demonstrate impairment in both initial exploratory and habituation of exploration in a novel environment in mice with mutation of DISC1. This is discussed in the context of the functional role of the gene vis à vis a schizophrenia phenotype as well as the value of ethologically based approaches to behavioral phenotyping.

The relationship between insight and neurological dysfunction in first-episode psychosis.

PURPOSE: Impaired insight is commonly seen in psychosis and some studies have proposed that is a biologically based deficit. Support for this view comes from the excess of neurological soft signs (NSS) observed in patients with psychoses and their neural correlates which demonstrate a degree of overlap with the regions of interest implicated in neuroimaging studies of insight. The aim was to examine the relationship between NSS and insight in a sample of 241 first-episode psychosis patients. METHOD: Total scores and subscale scores from three insight measures and two NSS scales were correlated in addition to factors representing overall insight and NSS which we created using principal component analysis. RESULTS: There were only four significant associations when we controlled for symptoms. "Softer" condensed neurological evaluation (CNE) signs were associated with our overall insight factor (r = 0.19, P = 0.02), with total Birchwood (r = -0.24, P < 0.01), and the Birchwood subscales; recognition of mental illness (r = -0.24, P < 0.01) and need for treatment (r = -0.18, P = 0.02). Total neurological evaluation scale (NES) and recognition of the achieved effects of medication were also weakly correlated (r = 0.14, P = 0.04). CONCLUSION: This study does not support a direct link between neurological dysfunction and insight in psychosis. Our understanding of insight as a concept remains in its infancy.

Disruption of thermal nociceptive behaviour in mice mutant for the schizophrenia-associated genes NRG1, COMT and DISC1.

Abnormalities in pain perception, especially altered warmth and heat pain sensitivity, have been reported in schizophrenia. Therefore, genes associated with schizophrenia, including neuregulin-1 (NRG1), catechol-O-methyltranferase (COMT) and disrupted-in-schizophrenia-1 (DISC1), may play a role in modulating the physiological and psychological effects of pain stimuli in such patients. Thermal pain sensitivity was assessed in NRG1, COMT and DISC1 mutant mice, and the anti-nociceptive effects of acute Delta(9)-tetrahydrocannabinol (THC) were compared in NRG1 and COMT mutants. At baseline, deletion of NRG1 and DISC1 each reduced thermal pain sensitivity, while deletion of COMT increased pain sensitivity. Neither NRG1 nor COMT deletion altered the anti-nociceptive effects of acute systemic THC (8.0mg/kg). These results indicate a differential contribution of NRG1 and DISC1 vis-à-vis COMT to the processing of thermal nociceptive stimuli and extend their phenotypic relationship to psychotic illness.

Schizophrenia-related endophenotypes in heterozygous neuregulin-1 'knockout' mice.

Neuregulin-1 (NRG1) has been shown to play a role in glutamatergic neurotransmission and is a risk gene for schizophrenia, in which there is evidence for hypoglutamatergic function. Sensitivity to the behavioural effects of the psychotomimetic N-methyl-D-aspartate receptor antagonists MK-801 and phencyclidine (PCP) was examined in mutant mice with heterozygous deletion of NRG1. Social behaviour (sociability, social novelty preference and dyadic interaction), together with exploratory activity, was assessed following acute or subchronic administration of MK-801 (0.1 and 0.2 mg/kg) or PCP (5 mg/kg). In untreated NRG1 mutants, levels of glutamate, N-acetylaspartate and GABA were determined using high-performance liquid chromatography and regional brain volumes were assessed using magnetic resonance imaging at 7T. NRG1 mutants, particularly males, displayed decreased responsivity to the locomotor-activating effects of acute PCP. Subchronic MK-801 and PCP disrupted sociability and social novelty preference in mutants and wildtypes and reversed the increase in both exploratory activity and social dominance-related behaviours observed in vehicle-treated mutants. No phenotypic differences were demonstrated in N-acetylaspartate, glutamate or GABA levels. The total ventricular and olfactory bulb volume was decreased in mutants. These data indicate a subtle role for NRG1 in modulating several schizophrenia-relevant processes including the effects of psychotomimetic N-methyl-D-aspartate receptor antagonists.

Phenotypic characterization of cognition and social behavior in mice with heterozygous versus homozygous deletion of catechol-O-methyltransferase.

Catechol-O-methyltransferase is an important enzyme in the metabolism of dopamine and an important regulator of aspects of dopamine-dependent working memory in prefrontal cortex that are disturbed in schizophrenia. This study investigated the phenotype of mice with heterozygous deletion vs. homozygous knockout of the catechol-O-methyltransferase gene across paradigms that access processes relevant for psychotic illness. Homozygotes evidenced improved performance in spontaneous alternation, an index of immediate spatial working memory; this effect appeared more substantive in males and was reflected in performance in aspects of the Barnes maze, an index of spatial learning/memory. Heterozygotes evidenced impaired performance in object recognition, an index of recognition memory; this effect was evident for both sexes at a retention interval of 5 min but appeared more enduring in males. There were no material effects for either genotype in relation to sociability or social novelty preference. While homozygous catechol-O-methyltransferase deletion results in improvement in spatial learning/working memory with little effect on social behavior, heterozygous deletion results in impairment of recognition memory. We have reported recently, using similar methods, that mice with deletion of the schizophrenia risk gene neuregulin-1 evidence disruption to social behavior, with little effect on spatial learning/working memory. The data suggest that catechol-O-methyltransferase and neuregulin-1 may influence, respectively, primarily cognitive and social endophenotypes of the overall schizophrenia syndrome.

Role of alpha adrenoceptors in the nucleus accumbens in the control of accumbal noradrenaline efflux: a microdialysis study with freely moving rats.

Microdialysis technique was used to study the effects of the locally applied alpha adrenoceptor agonist phenylephrine and antagonist phentolamine on the basal noradrenaline efflux as well as on the noradrenaline uptake inhibitor desipramine-elicited noradrenaline efflux in the nucleus accumbens (NAc) of freely moving rats. Tetrodotoxin reduced basal noradrenaline efflux by 72%, whereas desipramine increased it by 204%. Phenylephrine reduced the basal noradrenaline efflux by 32% and phentolamine blocked this effect. Phentolamine elevated the basal noradrenaline efflux by 150% and phenylephrine counteracted this effect. The desipramine-elicited noradrenaline efflux was not affected by phenylephrine, but enhanced by phentolamine. Desipramine counteracted the effects of phenylephrine and potentiated those of phentolamine. These results indicate that the accumbal noradrenaline efflux is under inhibitory control of alpha adrenoceptors that are suggested to be presynaptically located on adrenergic nerve terminals in the NAc. Furthermore, this study suggests that the conformational state of alpha adrenoceptors varies across the available amount of noradrenaline. The clinical impact of these data is discussed.

Phenotypic characterization of spatial cognition and social behavior in mice with 'knockout' of the schizophrenia risk gene neuregulin 1.

Neuregulin-1 (NRG1) has been identified as a candidate susceptibility gene for schizophrenia. In the present study the functional role of the NRG1 gene, as it relates to cognitive and social processes known to be disrupted in schizophrenia, was assessed in mice with heterozygous deletion of transmembrane (TM)-domain NRG1 in comparison with wildtypes (WT). Social affiliative behavior was assessed using the sociability and preference for social novelty paradigm, in terms of time spent in: (i) a chamber containing an unfamiliar conspecific vs. an empty chamber (sociability), or (ii) a chamber containing an unfamiliar conspecific vs. a chamber containing a familiar conspecific (preference for social novelty). Social dominance and aggressive behavior were examined in the resident-intruder paradigm. Spatial learning and memory were assessed using the Barnes maze paradigm, while spatial working memory was measured using the continuous variant of the spontaneous alternation task. Barnes maze data revealed intact spatial learning in NRG1 mutants, with elevated baseline latency to enter the escape hole in male NRG1 mutants reflecting an increase in activity level. Similarly, although a greater number of overall arm entries were found, spontaneous alternation was unaffected in NRG1 mice. Social affiliation data revealed NRG1 mutants to evidence a specific loss of WT preference for spending time with an unfamiliar as opposed to a familiar conspecific. This suggests that NRG1 mutants show a selective impairment in response to social novelty. While spatial learning and working memory processes appear intact, heterozygous deletion of TM-domain NRG1 was associated with disruption to social novelty behavior. These data inform at a novel phenotypic level on the functional role of this gene in the context of its association with risk for schizophrenia.

Evidence for association and epistasis at the DAOA/G30 and D-amino acid oxidase loci in an Irish schizophrenia sample.

The D-amino acid oxidase (DAO) signaling pathway has been implicated in schizophrenia pathogenesis. This may be mediated through modulation of NMDA function by DAO, which is in turn activated by DAO activator (DAOA, formerly G72). Chumakov et al. (2002); PNAS 99: 13675-13680, identifying the novel schizophrenia susceptibility gene DAOA/G30 and a number of independent studies have since reported evidence of association between the DAOA and DAO genes and schizophrenia. However, at least two studies have failed to replicate the epistatic interaction between these loci described in the original report and there have been differences in the associated alleles/haplotypes reported at each locus. In this study, we performed association and epistasis analyses of the DAOA/G30 and DAO loci in a sample of 373 cases with DSM-IV schizophrenia/schizoaffective disorder and 812 controls from the Republic of Ireland. Corrected for the number of tests performed, we found evidence for association between markers at both genes and schizophrenia: DAOA/G30 (P = 0.005, OR = 1.34 (1.09, 1.65)) and DAO (P = 0.003, OR = 1.43 (1.12, 1.84). The data suggest that evidence for association at DAO (marker rs2111902) is more consistent than previously realized, particularly in Caucasian schizophrenia populations. We identified evidence for epistatic interaction between the associated SNPs at DAOA and DAO genes in contributing to schizophrenia risk (OR = 9.3 (1.4, 60.5). Based on these data, more systematic investigation of genes involved in DAO signaling is required.

Mice lacking the alpha4 nicotinic receptor subunit fail to modulate dopaminergic neuronal arbors and possess impaired dopamine transporter function.

Neuronal nicotinic acetylcholine receptors (nAChRs) at presynaptic sites can modulate dopaminergic synaptic transmission by regulating dopamine (DA) release and uptake. Dopaminergic transmission in nigrostriatal and mesolimbic pathways is vital for the coordination of movement and is associated with learning and behavioral reinforcement. We reported recently that the D2 DA receptor plays a central role in regulating the arbor size of substantia nigra dopaminergic neurons. Given the known effects of nAChRs on dopaminergic neurotransmission, we assessed the ability of the alpha4 nAChR subunit to regulate arbor size of dopaminergic neurons by comparing responses of wild-type and alpha4 nAChR subunit knockout [alpha4(-/-)] mice to long-term exposure to cocaine, amphetamine, nicotine, and haloperidol, and after substantia nigra neurotoxic lesioning. We found that dopaminergic neurons in adult drug-naive alpha4(-/-) mice had significantly larger terminal arbors, and despite normal short-term behavioral responses to drugs acting on pre- and postsynaptic D2 DA receptors, they were unable to modulate their terminal arbor in response to pharmacological manipulation or after lesioning. In addition, although synaptosome DA uptake studies showed that the interaction of the D2 DA receptor and the dopamine transporter (DAT) was preserved in alpha4(-/-) mice, DAT function was found to be impaired. These findings suggest that the alpha4 subunit of the nAChR is an independent regulator of terminal arbor size of nigrostriatal dopaminergic neurons and that reduced functionality of presynaptic DAT may contribute to this effect by impairing DA uptake.

Identification in 2 independent samples of a novel schizophrenia risk haplotype of the dystrobrevin binding protein gene (DTNBP1).

CONTEXT: Recent research suggests that variation in the gene encoding dystrobrevin binding protein (DTNBP1) confers susceptibility to schizophrenia. Thus far, no specific risk haplotype has been identified in more than 1 study. OBJECTIVES: To confirm DTNBP1 as a schizophrenia susceptibility gene, to identify and replicate specific risk and protective haplotypes, and to explore relationships between DTNBP1 and the phenotype. DESIGN: Genetic association study based on mutation detection and case-control analysis. SETTING: All subjects were unrelated and ascertained from general (secondary care) psychiatric inpatient and outpatient services. PARTICIPANTS: The Cardiff, Wales, sample included 708 white subjects from the United Kingdom and Ireland (221 females) who met DSM-IV criteria for schizophrenia and were individually matched for age, sex, and ethnicity to 711 blood donor controls (233 females). Mean +/- SD age at first psychiatric contact for cases was 23.6 +/- 7.7 years; mean age at ascertainment was 41.8 +/- 13.5 years. The Dublin, Ireland, sample included 219 white subjects from the Republic of Ireland who met DSM-III-R criteria for schizophrenia or schizoaffective disorder and 231 controls. The mean age of the Irish cases was 46.0 +/- 8.5 years; mean age at first psychiatric contact was 25.2 +/- 12.4 years. MAIN OUTCOME MEASURE: Evidence for association between the DTNBP1 locus and schizophrenia. RESULTS: In the Cardiff sample, there was no evidence for association with previously implicated haplotypes but strong evidence for association with multiple novel haplotypes. Maximum evidence was found for a novel 3-marker haplotype (global P<.001), composed of 1 risk haplotype (P =.01) and 2 protective haplotypes, 1 common (P =.006) and 1 rare (P<.001). Specific risk and protective haplotypes were replicated in the Dublin sample (P =.02,.047, and.006, respectively). The only phenotypic variable associated with any haplotype was between the common protective haplotype and higher educational achievement (P =.02, corrected for multiple tests). CONCLUSIONS: DTNBP1 is a susceptibility gene for schizophrenia. Specific risk and protective haplotypes were identified and replicated. Association with educational achievement may suggest protection mediated by IQ, although this needs to be confirmed in an independent data set.

Confirmation and refinement of an 'at-risk' haplotype for schizophrenia suggests the EST cluster, Hs.97362, as a potential susceptibility gene at the Neuregulin-1 locus.

Two recent association studies have implicated the neuregulin-1 gene (NRG1) at chromosome 8p21-22 as a susceptibility gene for schizophrenia. Stefansson et al identified three 'at-risk' haplotypes (HapA, B and C) which spanned the NRG1 locus and shared a common core haplotype. Subsequently, they demonstrated evidence that the core haplotype was associated with schizophrenia in an independent Scottish sample. To confirm and refine this haplotype we investigated the NRG1 locus in an independent Irish case-control sample. We did not find the core haplotype to be associated in our sample. However, we identified a refined 2-marker haplotype (HapB(IRE)) that shared common alleles with one of the Icelandic 'at-risk' haplotypes and is in significant excess in the Irish cases (19.4%) vs controls (12.3%) (P=0.013). This refined 'at-risk' haplotype is also in significant excess in the Scottish case sample (17.0% vs 13.5%; P=0.036). Interestingly, this refined 'at-risk' haplotype is positioned close to an EST cluster of unknown function (Hs.97362) within intron 1 of NRG1.