Association of NOS1AP variants and depression phenotypes in schizophrenia

Journal of Affective Disorders 188 (2015) 263–269

Contents lists available at ScienceDirect

Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad

Research report

Association of NOS1AP variants and depression phenotypes in schizophrenia Sern-Yih Cheah a, Bruce R. Lawford a,b, Ross McD. Young a, C. Phillip Morris a, Joanne Voisey a,n a b

Institute of Health and Biomedical Innovation, Queensland University of Technology, 60 Musk Avenue, Kelvin Grove, Queensland 4059, Australia Discipline of Psychiatry, Royal Brisbane and Women’s Hospital, Herston, Queensland 4006, Australia

art ic l e i nf o

a b s t r a c t

Article history: Received 18 June 2015 Received in revised form 27 August 2015 Accepted 29 August 2015 Available online 8 September 2015

Background: The nitric oxide synthase 1 adaptor protein gene (NOS1AP) has previously been recognised as a schizophrenia susceptibility gene due to its role in glutamate neurotransmission. The gene is believed to inhibit nitric oxide (NO) production activated by the N-methyl-D-aspartate (NMDA) receptor and reduced NO levels have been observed in schizophrenia patients. However, association studies investigating NOS1AP and schizophrenia have produced inconsistent results, most likely because schizophrenia is a clinically heterogeneous disorder. This study aims to investigate the association between NOS1AP variants and defined depression phenotypes of schizophrenia. Methods: Nine NOS1AP SNPs, rs1415259, rs1415263, rs1858232, rs386231, rs4531275, rs4656355, rs4657178, rs6683968 and rs6704393 were genotyped in 235 schizophrenia subjects screened for various phenotypes of depression. Result: One NOS1AP SNP (rs1858232) was associated with the broad diagnosis of schizophrenia and eight SNPs were associated with depression related phenotypes within schizophrenia. The rs1415259 SNP showed strong association with sleep dysregulation phenotypes of depression. Conclusion: Results suggest that NOS1AP variants are associated with various forms of depression in schizophrenia and are more prevalent in males. Limitation: Schizophrenia is a clinically heterogeneous disease that can vary greatly between different ethnic and geographic populations so our observations should be viewed with caution until they are independently replicated, particularly in larger patient cohorts. Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.

Keywords: Schizophrenia NOS1AP Polymorphism SNP Psychiatric phenotypes Depression

1. Introduction Schizophrenia is a heterogeneous disorder that was classically identified under Diagnostic and Statistical Manual of Mental Disorders Fifth Edition (DSM-V) criteria by screening patients for hallucinations, delusions, neuromuscular dysfunction and negative symptoms. In many cases however, the symptomatology of schizophrenia usually extends to include various comorbid disorders including but not limited to depression, substance abuse, social phobia, posttraumatic stress disorder and obsessive compulsive disorder. Depression has been known to have high prevalence not just in schizophrenia but in other psychiatric disorders. In fact, the Australian Department of Health reported a 54.5% rate of comorbid depression in people with psychosis (Morgan et al., 2011). The high prevalence of depression within psychosis suggests an n

Corresponding author. Fax: þ 61 7 31386030. E-mail address: [email protected] (J. Voisey).

http://dx.doi.org/10.1016/j.jad.2015.08.069 0165-0327/Crown Copyright & 2015 Published by Elsevier B.V. All rights reserved.

overlapping molecular pathophysiology that underlines both conditions. The NOS1AP gene encodes nitric oxide synthase 1 adaptor protein that binds to signalling molecule neuronal nitric oxide synthase (nNOS) which is involved in downstream glutamatergic signalling in the postsynaptic density. Activation of the NMDA receptor causes a cascade that recruits nNOS and PSD-95 (postsynaptic density protein-95) serving as the anchor between nNOS and the NMDA receptor. Sustained activation of the NMDA receptor allows calcium influx which activates the nNOS for NO production (Brenman et al., 1996; Doucet et al., 2012; Nedvetsky et al., 2002). NOS1AP competes with PSD-95 for nNOS binding hence reducing the interaction between PSD-95 and nNOS (Jaffrey et al., 1998) and changes the NO production activity. Changes in cellular NO levels have also been reported to be associated with various psychiatric conditions such as depression, schizophrenia, bipolar disorder and cognitive deficits, see review (Freudenberg et al., 2015). This places NOS1AP as a component within the

264

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

downstream neuropathology of glutamate neurotransmission as glutamate signalling has long been implicated in schizophrenia and various psychiatric symptoms (Coyle et al., 2003; Lang et al., 2007). Animal studies suggest that NOS1AP acts as a facilitator (instead of inhibitor) for NMDAR-mediated cellular cascades (Cheah et al., 2006; Fang et al., 2000; Li et al., 2013). Antidepressant effects were reported in animal models when glutamate activity was reduced via small-molecule inhibition of nNOS–PSD95–NMDAR complex (Doucet et al., 2013). Another similar animal study also reported antidepressant effects by glutamate activity reduction via smallmolecule dissociation of NOS1AP–nNOS complex (Zhu et al., 2014). These studies suggest a positive relationship between NOS1AP and nNOS activity that increases NO and glutamate activity, promoting depressive effects. Although the number of association studies between NOS1AP and schizophrenia are limited, there is still sufficient evidence to support the role of NOS1AP in schizophrenia. Brzustowicz and colleagues previously identified linkage in chromosome region 1q21-22 (encompassing the NOS1AP gene) in two Canadian family studies (Brzustowicz et al., 2002, 2000). A subsequent follow-up study identified a functional SNP (rs12742393) in the promoter that exhibited strong linkage disequilibrium (LD) with schizophrenia and was associated with high levels of NOS1AP expression in prefrontal cortex of post-mortem schizophrenia brain samples (Wratten et al., 2009). The functional SNP is believed to alter affinity for binding of nuclear proteins. This suggests that differential transcription factor binding may be the mechanism responsible for differential expression between alleles. In a South American study, one SNP (rs6680461) was found associated with schizophrenia and further transmission disequilibrium test analysis revealed that the SNP was over-transmitted in schizophrenia patients with negative symptoms (including avolition and social isolation) (Kremeyer et al., 2009). A Chinese Han population study reported an association between rs348624 and schizophrenia (Zheng et al., 2005). SNP rs348624 is a synonymous polymorphism located in exon 9 that is in strong LD with a nearby splice site SNP (rs11422090) as well as other neighbouring functional SNPs that alter transcription and splicing (Zheng et al., 2005). However, several studies failed to detect association between NOS1AP and schizophrenia (Fang et al., 2008; Nicodemus et al., 2008; Puri et al., 2006). A recent meta-analysis of three pooled European samples (German, Swedish and Spanish) could not find association between NOS1AP SNPs and schizophrenia; but when the three sample sets were analysed independently association between NOS1AP SNPs and schizophrenia was found (Weber et al., 2014). The nitric oxide synthase 1 adaptor protein (NOS1AP) gene is a schizophrenia susceptibility gene that is implicated in other psychiatric conditions including depression, autism spectrum disorders, impulsivity and cognitive deficits, see review (Freudenberg et al., 2015). Since we previously identified an association between NOS1AP and depression in a cohort of posttraumatic stress disorder patients (Lawford et al., 2013), in the present study we have examined association with depression in a cohort of schizophrenia patients to further understand the role of NOS1AP in various depression phenotypes in schizophrenia.

Table 1 Participants' demographic information.

2. Methods

2.2. Ethics

2.1. Subjects

Each participant gave written informed consent before commencement of data collection. Ethics approval for the project was obtained from the Human Research Ethics Committee of the Queensland University of Technology. This study was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki).

2.1.1. Schizophrenia cases The participants were obtained from the Australian Schizophrenia Research Bank (ASRB: http://www.schizophreniaresearch. org.au). They consisted of 235 schizophrenia patients of European

Demographic information

Participants Schizophrenia Control

Male Female Age in years: mean ( 7 SD) Age of onset ( 7SD) Family with schizophrenia history Patients on medication Mean duration of first medication used in months (7 SD)

165 70 43.9 (10.7) 23.2 (6.3) 72 229 41.03 (42.96)a

104 121 45 (13.2)

a Lowest duration recorded is one month and highest duration recorded is more than eight years.

descent with a mean age of 43.9 years (sd 710.7 years). There were 70 females and 165 males with a confirmed diagnosis of schizophrenia according to DSM-IV/ICD-10 diagnostic criteria with a mean onset age of 23.18 years (sd 76.31 years). The demographic history of participants is detailed in Table 1. All participants underwent a clinical and neuropsychological assessment and provided a blood sample at the time of assessment. The participants were assessed using a clinical assessment battery that consists of the Diagnostic Interview for Psychosis (DIP) (Castle et al., 2006) to collect socio-demographic, family and medical history data, and clinical measures for lifetime occurrence of various depression phenotypes (clinical measure details can be found in Supplementary Table 1). The depression phenotype measures are defined by the DIP which covers various forms of depression that can be expressed by patients. All schizophrenia patients and controls were recruited from several sources across five Australian States and Territories (New South Wales, Australian Capital Territory, Queensland, Western Australia and Victoria) using media advertisements, inpatient, outpatient and community mental health service providers, non-government organisations and rehabilitation services. All participants identified themselves as having European decent. In this case the definition of European covered a very large and diverse region spanning from as far north and south as Norway and southern Italy and as far west as Ireland and east as far as Turkey but excluding Middle Eastern Europe. Australians of Aboriginal or Torres Strait Island decent were excluded. 2.1.2. Controls The control group was also obtained from the ASRB. The controls consisted of 121 females and 104 males of European descent, with a mean age of 45.00 years (sd 713.2 years). Healthy controls were screened for a family history of, or treatment for, a psychiatric illness at the time of registration. The controls were assessed using a clinical assessment battery that consists of the Diagnostic Interview for Psychosis (DIP) (Castle et al., 2006) to collect socio-demographic, family and medical history data. They provided a blood sample at time of assessment.

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

2.3. SNP Genotyping DNA was provided by the ASRB. NOS1AP tag-SNPs were selected using HapMap project phase 2 (International HapMap Consortium, 2003) with a minor allele frequency cut-off of 0.15 and were identified using pair-wise option of Tagger with r2 40.8 threshold. NOS1AP tag-SNPs included rs1415259, rs1415263, rs1858232, rs386231, rs4531275, rs4656355, rs4657178, rs6683968 and rs6704393. Samples were genotyped for the NOS1AP SNPs using a homogeneous MassEXTEND (hME) Sequenom assay performed by the Australian Genome Research Facility (AGRF). The hME assay is based on the annealing of a hME primer adjacent to the targeted SNP. Addition of DNA polymerase and terminator nucleotide mixture allows the extension of the hME primer through the SNP site, thus generating allele-specific extension products with a unique molecular mass. The mass of the extension products were analysed with matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDITOF MS) to determine a genotype. 2.4. Statistical analysis SNP-schizophrenia association was analysed by likelihood ratio chi-square using Statistical Package for the Social Sciences (SPSS) version 21 to compare genotype frequencies between schizophrenia and control groups. SNP-phenotype association was analysed by ANOVA (analysis of variance) using SPSS version 21 to detect measure mean score differences between genotypes in schizophrenia patients. A higher measure score means a higher clinical severity for a clinical measure. Hardy–Weinberg equilibrium was tested using Utility Programmes for Analysis of Genetic Linkage (Ott, 1988). Power calculations for SNP-schizophrenia association was performed using Mann–Whitney power calculator from the Compare2 software package (Abramson, 2004) that screens power from genotype frequencies in a pool of schizophrenia and controls at a desired odds ratio. Power calculations for SNP-phenotype association was performed using the Power and Sample Size.com ANOVA Power Calculator (http://power andsamplesize.com/Calculators/Compare-k-Means/1-Way-ANOVA-Pairwise) that screens power from pairwise comparisons for three genotypes per SNP and calculates the genotype counts required to achieve 80% power to detect a significant variance between two genotypes for each phenotype. Post-hoc Tukey's test was performed on SNPs found associated with depression phenotypes to identify significant differences between genotype pairs for all SNP-depression phenotype associations. Correction for multiple testing of SNP-schizophrenia association and SNP-phenotype associations for each phenotype was performed using the Benjamini–Hochberg method (Benjamini and Hochberg, 1995). SNPs found associated with a depression phenotypes were followed by dominance model analysis to determine the mode of inheritance of alleles. Males and females were also analysed separately to identify SNP-phenotype associations influenced by sex.

3. Results All SNPs were found to be in Hardy–Weinberg equilibrium (P 40.05). Power calculations for SNP-schizophrenia association revealed that all nine SNPs in the study have an 80% chance to detect an association with schizophrenia at 1.8 odds ratio. Power calculations for SNP-phenotype association revealed that all nine SNPs have an 80% chance to detect an association with between one to four depression phenotypes for each SNP.

265

Table 2 Genotype counts and chi-square tests for association between rs1858232 and schizophrenia. Genotype Schizophrenia Control Likelihood ratio

Trend test in a given directiona

AA AG GG

P ¼0.010

a

125 87 22

135 75 9

P¼ 0.041

Extended Mantel–Haenszel test for trend in a given direction.

3.1. Case control association study Genotypes for nine NOS1AP SNPs were determined in 225 controls and 235 schizophrenia cases. Out of the nine SNPs tested only one (rs1858232) was associated with schizophrenia with a Pvalue of 0.041 (Table 2), which was only marginally significant and was not significant after correction for multiple testing (P ¼0.041, q¼ 0.006). 3.2. Phenotype-SNP association study To test for association between NOS1AP and depression in a cohort of schizophrenia patients, ANOVAs between DIP depression measures and nine NOS1AP SNPs were performed. SNPs rs1415259, rs1415263, rs1858232, rs4531275, rs4656355, rs4657178, rs6683968 and rs6704393 all showed association with at least one depression related phenotype (Table 3). SNP rs386231 was not associated with any depression phenotypes. Tukey's test for each SNP found all but one SNP survived posthoc analysis. The association between depressive poor appetite and rs4531275 did not survive post-hoc analysis. Correction for multiple testing found that the association between rs4657178 and “depressive delusions of poverty” (P ¼0.002, q¼ 0.006) survived multiple testing correction. No other associations survived correction. Additionally, analysis to determine the mode of inheritance for each SNP was performed for SNPs that survived Tukey's post-hoc test with a P-value o 0.05 (Table 3). Interestingly, rs1415259 is associated with sleep-related depression phenotypes while other SNPs are associated with either a single depression phenotype or multiple but unrelated depression phenotypes. 3.3. Sex effects on Phenotype-SNP associations Depression often shows strong sex effects so sex association tests were performed for all SNPs to determine if the depressive phenotypes have sex-specific effects. SNPs that were associated with depressive phenotypes in males are: rs1415259 (P¼ 0.028 for interrupted sleep form depressive insomnia, P ¼0.020 for early waking form depressive insomnia); rs1415263 (P ¼0.044 for depressive slowed activity); rs4531275 (P ¼0.038 for depressive slowed activity, P ¼0.007 for depressive poor appetite, P ¼0.007 for depressive delusions of poverty); rs4657178 (P ¼0.002 for depressive delusions of poverty); rs6683968 (P ¼0.011 for depressive delusions of poverty) and rs6704393 (P¼ 0.025 in depressive excessive self-guilt). SNPs that were associated with depressive phenotypes in females are rs1858232 (P ¼0.008 for depressive weight gain) and rs4656355 (P ¼0.047 for interrupted sleep form depressive insomnia).

4. Discussion The present study tested association between nine NOS1AP SNPs and various phenotypes of depression in schizophrenia

266

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

Table 3 Genotype association of eight NOS1AP SNPs with depression measures in schizophrenia. Depression domain measure scorea,b, 7SE SNP

Genotypes (count)

Slowed activityc

Poor appetitec Weight gainc

Middle insomniad

Early wakingc

Excessive guiltc

Delusions of povertyc

rs1415259

AA(59) AG(70) GG(23)

0.85, 70.17 0.86, 70.16 0.91, 7 0.29 F¼ 0.021 P¼ 0.979

1.10, 7 0.18 0.84, 70.16 0.91, 7 0.29 F¼ 0.608 P¼ 0.546

0.68, 7 0.16 0.59, 7 0.14 1.09, 7 0.28 F ¼ 1.501 P ¼ 0.226

0.34, 7 0.06 0.16, 7 0.04 0.13, 7 0.07 F ¼ 3.841 P ¼ 0.024 A recessive (0.006)

0.59, 7 0.15 0.16, 7 0.08 0.39, 7 0.22 F¼ 3.530 P ¼ 0.032 A recessive (0.016)

0.80, 7 0.17 1.07, 7 0.16 0.65, 7 0.26 F¼ 1.177 P¼ 0.311

0.00, 7 0.00 0.00, 7 0.00 0.13, 7 0.13 F ¼ 2.874 P ¼ 0.060

0.82, 70.18 0.63, 70.13 1.43, 7 0.28 F¼ 3.963 P ¼ 0.021 C dominant (0.008)

1.02, 7 0.19 0.91, 7 0.15 0.93, 70.26 F¼ 0.104 P¼ 0.901

0.66, 7 0.16 0.75, 7 0.14 0.50, 7 0.21 F ¼ 0.440 P ¼ 0.645

0.24, 70.06 0.20, 7 0.05 0.21, 7 0.08 F ¼ 0.125 P ¼ 0.883

0.36, 7 0.14 0.42, 7 0.11 0.14, 7 0.11 F¼ 0.900 P¼ 0.409

1.00, 7 0.19 0.92, 7 0.15 0.56, 7 0.21 F¼ 1.083 P¼ 0.341

0.00, 7 0.00 0.00, 7 0.00 0.11, 7 0.11 F ¼ 2.433 P ¼ 0.091

0.80, 70.15 0.92, 70.17 0.62, 70.33 F¼ 0.330 P¼ 0.719

1.05, 7 0.15 0.90, 70.17 0.54, 70.31 F¼ 0.880 P¼ 0.417

0.67, 7 0.13 0.52, 7 0.13 1.46, 7 0.42 F ¼ 3.444 P ¼ 0.034 G recessive (0.013)

0.26, 7 0.05 0.19, 7 0.05 0.08, 7 0.08 F ¼ 1.217 P ¼ 0.299

0.39, 7 0.11 0.26, 7 0.10 0.54, 7 0.31 F¼ 0.640 P¼ 0.529

0.83, 7 0.14 0.95, 7 0.17 0.92, 7 0.40 F¼ 0.155 P¼ 0.857

0.04, 7 0.04 0.00, 7 0.00 0.00, 7 0.00 F ¼ 0.448 P ¼ 0.640

0.77, 7 0.16 0.72, 7 0.15 1.59, 7 0.37 F¼ 3.274 P ¼ 0.041 T recessive (0.012)

1.12, 7 0.17 0.68, 7 0.14 1.47, 70.37 F¼ 3.486 P ¼ 0.033

0.60, 7 0.14 0.77, 7 0.15 0.53, 7 0.29 F ¼ 0.501 P ¼ 0.607

0.26, 7 0.06 0.17, 7 0.04 0.24, 70.11 F ¼ 0.811 P ¼ 0.446

0.34, 7 0.12 0.39, 7 0.11 0.24, 7 0.18 F¼ 0.189 P¼ 0.828

0.91, 7 0.16 0.92, 7 0.16 0.63, 7 0.30 F¼ 0.349 P¼ 0.706

0.00, 7 0.00 0.00, 7 0.00 0.19, 7 0.19 F ¼ 4.577 P ¼ 0.012 T recessive (0.003)

0.92, 70.28 0.90, 70.16 0.84, 70.17 F¼ 0.044 P¼ 0.957

0.85, 70.26 0.91, 7 0.16 1.05, 7 0.18 F¼ 0.273 P¼ 0.762

0.77, 7 0.24 0.63, 7 0.14 0.68, 7 0.17 F ¼ 0.133 P ¼ 0.875

0.19, 7 0.08 0.15, 7 0.04 0.34, 7 0.06 F ¼ 3.322 P ¼ 0.039 G recessive (0.012)

0.23, 7 0.16 0.21, 7 0.09 0.57, 7 0.15 F¼ 2.631 P¼ 0.075

0.62, 7 0.24 1.16, 7 0.17 0.77, 7 0.17 F¼ 2.220 P¼ 0.112

0.12, 70.12 0.00, 7 0.00 0.00, 7 0.00 F ¼ 2.410 P ¼ 0.093

0.81, 7 0.14 0.80, 70.17 1.15, 7 0.42 F¼ 0.419 P¼ 0.658

1.02, 7 0.15 0.76, 7 0.17 1.23, 7 0.41 F¼ 0.955 P¼ 0.387

0.71, 70.13 0.62, 7 0.16 0.69, 7 0.37 F ¼ 0.095 P ¼ 0.909

0.21, 7 0.04 0.22, 7 0.06 0.23, 7 0.12 F ¼ 0.010 P ¼ 0.990

0.37, 7 0.11 0.33, 7 0.12 0.31, 7 0.24 F¼ 0.051 P¼ 0.950

0.90, 7 0.14 0.84, 7 0.18 1.00, 7 0.37 F¼ 0.088 P¼ 0.916

0.00, 7 0.00 0.00, 7 0.00 0.25, 7 0.25 F ¼ 6.420 P ¼ 0.002 T recessive (0.0004)

0.80, 70.17 0.69, 70.14 1.50, 7 0.34 F¼ 3.150 P ¼ 0.046 T recessive (0.015)

1.08, 7 0.18 0.77, 7 0.14 1.25, 7 0.33 F¼ 1.533 P¼ 0.219

0.64, 7 0.15 0.72, 7 0.14 0.60, 7 0.28 F ¼ 0.107 P ¼ 0.898

0.29, 7 0.06 0.15, 7 0.04 0.25, 7 0.10 F ¼ 1.870 P ¼ 0.158

0.32, 7 0.12 0.41, 7 0.11 0.20, 7 0.16 F¼ 0.445 P¼ 0.642

0.90, 7 0.17 0.96, 7 0.16 0.53, 7 0.26 F¼ 0.847 P¼ 0.431

0.00, 7 0.00 0.00, 7 0.00 0.16, 7 0.16 F ¼ 3.732 P ¼ 0.026 T recessive (0.007)

0.76, 7 0.16 0.73, 7 0.15 1.19, 7 0.29 F¼ 1.285 P¼ 0.280

1.02, 7 0.17 0.94, 70.17 0.85, 70.26 F¼ 0.154 P¼ 0.857

0.57, 70.14 0.84, 7 0.16 0.54, 7 0.22 F ¼ 1.000 P ¼ 0.370

0.24, 70.05 0.16, 7 0.05 0.27, 7 0.09 F ¼ 0.836 P ¼ 0.435

0.44, 7 0.13 0.31, 7 0.11 0.23, 7 0.16 F¼ 0.580 P¼ 0.561

0.86, 7 0.16 1.15, 7 0.18 0.31, 7 0.16 F¼ 4.045 P ¼ 0.019 C dominant (0.013)

0.00, 7 0.00 0.00, 7 0.00 0.12, 70.12 F ¼ 2.530 P ¼ 0.083

Inheritance mode (P value)

rs1415263

CC(50) CT(79) TT(28)

Inheritance mode (P value)

rs1858232

AA(82) AG(62) GG(13)

Inheritance mode (P value)

rs4531275

CC(65) CT(75) TT(17)

Inheritance mode (P value)

rs4656355

AA(26) AG(67) GG(56)

Inheritance mode (P value)

rs4657178

CC(89) CT(55) TT(13)

Inheritance mode (P value)

rs6683968

GG(59) GT(78) TT(20)

Inheritance mode (P value)

rs6704393

CC(63) CT(67) TT(26)

Inheritance mode (P value) a

Measure scores range from 0 to 1 and 0 to 3. Thirteen other depression measures were not shown as they are not associated with any of the SNPs in this study. Scores ranging from 0 to 3. d Scores ranging from 0 to 1. b c

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

patients. These SNPs are not associated with schizophrenia but do show association with multiple forms of depression in schizophrenia. Our sample consists of a mixed European population which is ethnically diverse and genetically heterogeneous. The use of a genetically heterogeneous population should decrease the possibility of population stratification but decrease the chance of finding genetic associations. Despite this limitation, we were still able to find association between eight NOS1AP SNPs and several depression phenotypes. These depression phenotypes appear to be related to energy drive and motivation, appetite, sleep, and logical process leading to delusions and guilt in these patients, suggesting that these features may share a common pathophysiology linked to depression in schizophrenia patients. In light of our previous finding of association with depression in combat veterans with posttraumatic stress disorder (Lawford et al., 2013), our present results further suggest a role for NOS1AP in depression that is not limited to schizophrenia. A recent genome-wide schizophrenia association study (GWAS) found no association with rs1858232; but found nominal association with another NOS1AP SNP rs1415263 (P¼ 0.023) (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014). The discordant results are possibly due to sample size differences as the association found between rs1858232 and schizophrenia was only marginally significant and did not hold up after multiple correction. The difference could also be due to population make-up as the genome-wide study sample consists of a population mixture of European and East Asian origin. Phenotypes found associated with more than one SNP include depressive slowed activity, interrupted sleep form depressive insomnia and depressive delusions of poverty, suggesting haplotype association. Our results suggest a stronger association between NOS1AP and depression phenotypes in males diagnosed with schizophrenia. However, as there are very few studies published in this area it is difficult to confirm a sex specific association. A crossnational study of sex differences in health covering depression suggested no significant overall sex differences in depression symptoms and they were only age and country-specific (Oksuzyan et al., 2010). In addition to depression, other studies have found association with specific schizophrenia phenotypes including hallucinations, delusions, avolition and social isolation (Kremeyer et al., 2009), obsessive–compulsive disorder and autism spectrum disorder (Delorme et al., 2010). It is not surprising that NOS1AP has such broad effects since NMDA receptor activation is a key component of the glutamate system which is implicated in diverse brain functions (Merritt et al., 2013). It is recognised that certain psychiatric conditions, like hallucinations, may lead to development of secondary phenotypes such as delusions and depression (Shiraishi et al., 2014). This raises the question of whether these non-depression phenotypes are directly due to NOS1AP dysfunction or secondary outcomes due to depression. The mode of action of NOS1AP in schizophrenia is thought to be via the inhibition of the downstream effects of the NMDA receptor on NO activity. It is therefore thought that overexpression of NOS1AP short isoform due to the C-allele of rs1415263 may lead to schizophrenia development (Xu et al., 2005). Other studies have suggested that association between the A allele of rs12742393 and schizophrenia is due to increased transcription factor binding and expression of NOS1AP mRNA (Wratten et al., 2009). Studies of Brodmann's Area 46 (a part of frontal cortex) in schizophrenia patients revealed that the pattern NOS1AP mRNA short isoform levels were increased while decreased levels were present in the cerebellum (Hadzimichalis et al., 2010). Tissue culture studies led to the NOS1AP overexpression hypothesis in schizophrenia (Jaffrey et al., 1998) but animal studies suggested otherwise. In these studies, NOS1AP was acting as a facilitator of NMDAR-mediated

267

cellular cascades (Cheah et al., 2006; Fang et al., 2000; Li et al., 2013), which does not support the idea that NOS1AP overexpression could inhibit NMDAR-mediated NO production. Given these inconsistencies, it is possible that the expression pattern of NOS1AP or its effects may vary in different regions of the brain. Excessive glutamate neurotransmission is thought to be associated with depression. Our findings support a role for NOS1AP in depression in schizophrenia. Evidence suggests that drugs that stimulate re-uptake of glutamate or inhibit glutamate activity exert antidepressant effects (Banasr et al., 2010; Gourley et al., 2012; Mineur et al., 2007). A review of the role of NO in depression supports the positive correlation between NO levels and major depression, further strengthening the notion that depression is caused by excessive NO activity due to excessive glutamate neurotransmission (Dhir and Kulkarni, 2011). An interesting observation in our study is that rs1415259 is associated with sleep related depressive phenotypes, which has not previously been reported. Stenberg observed that NO is an important factor for triggering recovery sleep after periods of sleep deprivation and inhibitors of NO synthase decrease spontaneous sleep (Stenberg, 2007). This suggests a role for NOS1AP in the NO pathway which also affects sleeping patterns. Out of the nine SNPs reported in the present study, only rs1415263 has previously been reported to have a functional role while the function of the other eight SNPs are yet to be confirmed. SNP rs1415263 was associated with increased NOS1AP mRNA expression (Xu et al., 2005) and was in strong linkage equilibrium with another functional SNP that increases mRNA expression (Wratten et al., 2009). Multiple previous studies examining association between NOS1AP and schizophrenia revealed inconsistent results that failed to definitively establish a link between NOS1AP and schizophrenia. Our study found only one SNP (rs1858232) that was marginally associated with schizophrenia and this association was not validated in a larger GWAS (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014). However, our positive association results between eight NOS1AP SNPs and multiple depression phenotypes within schizophrenia were consistent with the literature where numerous animal studies, phenotype association studies and molecular and pharmacological models strongly point towards a link between depression and NOS1AP.

5. Limitations A meta-analysis looking at NOS1AP by Weber et al. (2014) found that some NOS1AP SNPs are associated in smaller independent samples but are not associated in a larger pooled sample, possibly suggesting that there is population stratification in schizophrenia cohorts of different ethnic or geographical origin. SNP-phenotype association results corrected for multiple testing suggest that many of our association findings may not pose significant effects and should be viewed with caution. Although correction for multiple testing was performed we do not believe that it is totally appropriate to do so for these analyses as the multiple clinical measures do not represent independent variables, i.e. they are all depression measures that are likely to be related to one another. Further replication studies of larger cohorts screened for appropriate phenotypes should be conducted to validate our association findings. In addition, the function of NOS1AP SNPs, and correlation of NO-NOS1AP mRNA levels should be studied to understand the role of NOS1AP in depression phenotypes of schizophrenia. This should shed light on whether depression phenotypes share a common aetiology with schizophrenia due to NOS1AP function, or whether NOS1AP has a role in depression aetiology independent of its role in schizophrenia.

268

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

Funding statement This work was financially supported by the Queensland State Government, the Nicol Foundation and the Institute of Health and Biomedical Innovation, QUT.

Conflict of interest statement All authors declare that they have no conflicts of interest.

Authors contributions Authors Sern-Yih Cheah, Joanne Voisey and C. Phillip Morris made a substantial contribution to conception and design; the analysis and interpretation of the data and draughted and critically reviewed the manuscript. Ross McD. Young and Bruce R. Lawford made a substantial contribution to the conception and design and reviewed the final manuscript.

Acknowledgements This study was supported by a Post-graduate Research Award scholarship provided by the Queensland University of Technology, and the Australian Schizophrenia Research Bank (ASRB), which is supported by the National Health and Medical Research Council of Australia, the Pratt Foundation, Ramsay Health Care, the Viertel Charitable Foundation, and the Schizophrenia Research Institute.

Appendix A. Supplementary material Supplementary material associated with this article can be found in the online version at doi:10.1016/j.jad.2015.08.069.

References Abramson, J.H., 2004. WINPEPI (PEPI-for-Windows): computer programs for epidemiologists. Epidemiol. Perspect. Innov. 1, 6. Banasr, M., Chowdhury, G.M., Terwilliger, R., Newton, S.S., Duman, R.S., Behar, K.L., Sanacora, G., 2010. Glial pathology in an animal model of depression: reversal of stress-induced cellular, metabolic and behavioral deficits by the glutamatemodulating drug riluzole. Mol. Psychiatry 15, 501–511. Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B: Stat. Methodol. 57, 289–300. Brenman, J.E., Chao, D.S., Gee, S.H., McGee, A.W., Craven, S.E., Santillano, D.R., Wu, Z., Huang, F., Xia, H., Peters, M.F., Froehner, S.C., Bredt, D.S., 1996. Interaction of nitric oxide synthase with the postsynaptic density protein PSD-95 and alpha1syntrophin mediated by PDZ domains. Cell 84, 757–767. Brzustowicz, L.M., Hayter, J.E., Hodgkinson, K.A., Chow, E.W., Bassett, A.S., 2002. Fine mapping of the schizophrenia susceptibility locus on chromosome 1q22. Hum. Hered. 54, 199–209. Brzustowicz, L.M., Hodgkinson, K.A., Chow, E.W., Honer, W.G., Bassett, A.S., 2000. Location of a major susceptibility locus for familial schizophrenia on chromosome 1q21–q22. Science 288, 678–682. Castle, D.J., Jablensky, A., McGrath, J.J., Carr, V., Morgan, V., Waterreus, A., Valuri, G., Stain, H., McGuffin, P., Farmer, A., 2006. The diagnostic interview for psychoses (DIP): development, reliability and applications. Psychol. Med. 36, 69–80. Cheah, J.H., Kim, S.F., Hester, L.D., Clancy, K.W., Patterson 3rd, S.E., Papadopoulos, V., Snyder, S.H., 2006. NMDA receptor-nitric oxide transmission mediates neuronal iron homeostasis via the GTPase Dexras1. Neuron 51, 431–440. Coyle, J.T., Tsai, G., Goff, D., 2003. Converging evidence of NMDA receptor hypofunction in the pathophysiology of schizophrenia. Ann. NY Acad. Sci. 1003, 318–327. Delorme, R., Betancur, C., Scheid, I., Anckarsater, H., Chaste, P., Jamain, S., Schuroff, F., Nygren, G., Herbrecht, E., Dumaine, A., Mouren, M.C., Rastam, M., Leboyer, M., Gillberg, C., Bourgeron, T., 2010. Mutation screening of NOS1AP gene in a large sample of psychiatric patients and controls. BMC Med. Genet. 11, 108. Dhir, A., Kulkarni, S.K., 2011. Nitric oxide and major depression. Nitric Oxide 24, 125–131. Doucet, M.V., Harkin, A., Dev, K.K., 2012. The PSD–5/nNOS complex: new drugs for

depression? Pharmacol. Ther. 133, 218–229. Doucet, M.V., Levine, H., Dev, K.K., Harkin, A., 2013. Small-molecule inhibitors at the PSD-95/nNOS interface have antidepressant-like properties in mice. Neuropsychopharmacology 38, 1575–1584. Fang, C., Tang, W., Tang, R.Q., Wang, L., Zhou, G.Q., Huang, K., Li, X.W., Feng, G.Y., He, M., Du, L.Z., Zhu, S.M., Xing, Y.L., Sang, H., He, L., Shi, Y.Y., 2008. Family-based association studies of CAPON and schizophrenia in the Chinese Han population. Prog. Neuropsychopharmacol. Biol. Psychiatry 32, 1210–1213. Fang, M., Jaffrey, S.R., Sawa, A., Ye, K., Luo, X., Snyder, S.H., 2000. Dexras1: a G protein specifically coupled to neuronal nitric oxide synthase via CAPON. Neuron 28, 183–193. Freudenberg, F., Alttoa, A., Reif, A., 2015. Neuronal nitric oxide synthase (NOS1) and its adaptor, NOS1AP, as a genetic risk factors for psychiatric disorders. Genes Brain Behav. 14, 46–63. Gourley, S.L., Espitia, J.W., Sanacora, G., Taylor, J.R., 2012. Antidepressant-like properties of oral riluzole and utility of incentive disengagement models of depression in mice. Psychopharmacology 219, 805–814. Hadzimichalis, N.M., Previtera, M.L., Moreau, M.P., Li, B., Lee, G.H., Dulencin, A.M., Matteson, P.G., Buyske, S., Millonig, J.H., Brzustowicz, L.M., Firestein, B.L., 2010. NOS1AP protein levels are altered in BA46 and cerebellum of patients with schizophrenia. Schizophr. Res. 124, 248–250. International HapMap Consortium, 2003. The international HapMap project. Nature 426, 789–796. Jaffrey, S.R., Snowman, A.M., Eliasson, M.J., Cohen, N.A., Snyder, S.H., 1998. CAPON: a protein associated with neuronal nitric oxide synthase that regulates its interactions with PSD95. Neuron 20, 115–124. Kremeyer, B., Garcia, J., Kymalainen, H., Wratten, N., Restrepo, G., Palacio, C., Miranda, A.L., Lopez, C., Restrepo, M., Bedoya, G., Brzustowicz, L.M., Ospina-Duque, J., Arbelaez, M.P., Ruiz-Linares, A., 2009. Evidence for a role of the NOS1AP (CAPON) gene in schizophrenia and its clinical dimensions: an association study in a South American population isolate. Hum. Hered. 67, 163–173. Lang, U.E., Puls, I., Muller, D.J., Strutz-Seebohm, N., Gallinat, J., 2007. Molecular mechanisms of schizophrenia. Cell Physiol. Biochem. 20, 687–702. Lawford, B.R., Morris, C.P., Swagell, C.D., Hughes, I.P., Young, R.M., Voisey, J., 2013. NOS1AP is associated with increased severity of PTSD and depression in untreated combat veterans. J. Affect. Disord. 147, 87–93. Li, L.L., Ginet, V., Liu, X., Vergun, O., Tuittila, M., Mathieu, M., Bonny, C., Puyal, J., Truttmann, A.C., Courtney, M.J., 2013. The nNOS–p38MAPK pathway is mediated by NOS1AP during neuronal death. J. Neurosci. 33, 8185–8201. Merritt, K., McGuire, P., Egerton, A., 2013. Relationship between glutamate dysfunction and symptoms and cognitive function in psychosis. Front. Psychiatry 4, 151. Mineur, Y.S., Picciotto, M.R., Sanacora, G., 2007. Antidepressant-like effects of ceftriaxone in male C57BL/6J mice. Biol. Psychiatry 61, 250–252. Morgan, V.A., Waterreus, A., Jablensky, A., Mackinnon, A., McGrath, J.J., Carr, V., Bush, R., Castle, D., Cohen, M., Harvey, C., Galletly, C., Stain, H.J., Neil, A., McGorry, P., Hocking, B., Shah, S., Saw, S., 2011. People Living with Psychotic Illness 2010: Report on the Second Australian National Survey. Department of Health and Ageing, Canberra. Nedvetsky, P.I., Sessa, W.C., Schmidt, H.H., 2002. There’s NO binding like NOS binding: protein–protein interactions in NO/cGMP signaling. Proc. Natl. Acad. Sci. USA 99, 16510–16512. Nicodemus, K.K., Marenco, S., Batten, A.J., Vakkalanka, R., Egan, M.F., Straub, R.E., Weinberger, D.R., 2008. Serious obstetric complications interact with hypoxiaregulated/vascular-expression genes to influence schizophrenia risk. Mol. Psychiatry 13, 873–877. Oksuzyan, A., Crimmins, E., Saito, Y., O’Rand, A., Vaupel, J.W., Christensen, K., 2010. Cross-national comparison of sex differences in health and mortality in Denmark, Japan and the US. Eur. J. Epidemiol. 25, 471–480. Ott, J., 1988. Utility programs for analysis of genetic linkage, program, hwe version 1.10. Columbia University, New York. Puri, V., McQuillin, A., Thirumalai, S., Lawrence, J., Krasucki, R., Choudhury, K., Datta, S., Kerwin, S., Quested, D., Bass, N., Pimm, J., Lamb, G., Moorey, H., Kandasami, G., Badacsonyi, A., Kelly, K., Morgan, J., Punukollu, B., Nadeem, H., Curtis, D., Gurling, H.M., 2006. Failure to confirm allelic association between markers at the CAPON gene locus and schizophrenia in a British sample. Biol. Psychiatry 59, 195–197. Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014. Biological insights from 108 schizophrenia-associated genetic loci. Nature 511, 421–427. Shiraishi, N., Watanabe, N., Kinoshita, Y., Kaneko, A., Yoshida, S., Furukawa, T.A., Akechi, T., 2014. Brief psychoeducation for schizophrenia primarily intended to change the cognition of auditory hallucinations: an exploratory study. J Nerv. Ment. Dis. 202, 35–39. Stenberg, D., 2007. Neuroanatomy and neurochemistry of sleep. Cell Mol. Life Sci. 64, 1187–1204. Weber, H., Klamer, D., Freudenberg, F., Kittel-Schneider, S., Rivero, O., Scholz, C.J., Volkert, J., Kopf, J., Heupel, J., Herterich, S., Adolfsson, R., Alttoa, A., Post, A., Grussendorf, H., Kramer, A., Gessner, A., Schmidt, B., Hempel, S., Jacob, C.P., Sanjuan, J., Molto, M.D., Lesch, K.P., Freitag, C.M., Kent, L., Reif, A., 2014. The genetic contribution of the NO system at the glutamatergic post-synapse to schizophrenia: further evidence and meta-analysis. Eur. Neuropsychopharmacol. 24, 65–85.

S.-Y. Cheah et al. / Journal of Affective Disorders 188 (2015) 263–269

Wratten, N.S., Memoli, H., Huang, Y., Dulencin, A.M., Matteson, P.G., Cornacchia, M. A., Azaro, M.A., Messenger, J., Hayter, J.E., Bassett, A.S., Buyske, S., Millonig, J.H., Vieland, V.J., Brzustowicz, L.M., 2009. Identification of a schizophrenia-associated functional noncoding variant in NOS1AP. Am. J. Psychiatry 166, 434–441. Xu, B., Wratten, N., Charych, E.I., Buyske, S., Firestein, B.L., Brzustowicz, L.M., 2005. Increased expression in dorsolateral prefrontal cortex of CAPON in schizophrenia and bipolar disorder. PLoS Med. 2, e263. Zheng, Y., Li, H., Qin, W., Chen, W., Duan, Y., Xiao, Y., Li, C., Zhang, J., Li, X., Feng, G.,

269

He, L., 2005. Association of the carboxyl-terminal PDZ ligand of neuronal nitric oxide synthase gene with schizophrenia in the Chinese Han population. Biochem. Biophys. Res. Commun. 328, 809–815. Zhu, L.J., Li, T.Y., Luo, C.X., Jiang, N., Chang, L., Lin, Y.H., Zhou, H.H., Chen, C., Zhang, Y., Lu, W., Gao, L.Y., Ma, Y., Zhou, Q.G., Hu, Q., Hu, X.L., Zhang, J., Wu, H.Y., Zhu, D.Y., 2014. CAPON–nNOS coupling can serve as a target for developing new anxiolytics. Nat. Med. 20, 1050–1054.