Sonia De Zutter

Sequencing of DISC1 pathway genes reveals increased burden of rare missense variants in schizophrenia patients from a northern Swedish population.

In recent years, DISC1 has emerged as one of the most credible and best supported candidate genes for schizophrenia and related neuropsychiatric disorders. Furthermore, increasing evidence – both genetic and functional – indicates that many of its protein interaction partners are also involved in the development of these diseases. In this study, we applied a pooled sample 454 sequencing strategy, to explore the contribution of genetic variation in DISC1 and 10 of its interaction partners (ATF5, Grb2, FEZ1, LIS-1, PDE4B, NDE1, NDEL1, TRAF3IP1, YWHAE, and ZNF365) to schizophrenia susceptibility in an isolated northern Swedish population. Mutation burden analysis of the identified variants in a population of 486 SZ patients and 514 control individuals, revealed that non-synonymous rare variants with a MAF<0.01 were significantly more present in patients compared to controls (8.64% versus 4.7%, P = 0.018), providing further evidence for the involvement of DISC1 and some of its interaction partners in psychiatric disorders. This increased burden of rare missense variants was even more striking in a subgroup of early onset patients (12.9% versus 4.7%, P = 0.0004), highlighting the importance of studying subgroups of patients and identifying endophenotypes. Upon investigation of the potential functional effects associated with the identified missense variants, we found that ∼90% of these variants reside in intrinsically disordered protein regions. The observed increase in mutation burden in patients provides further support for the role of the DISC1 pathway in schizophrenia. Furthermore, this study presents the first evidence supporting the involvement of mutations within intrinsically disordered protein regions in the pathogenesis of psychiatric disorders. As many important biological functions depend directly on the disordered state, alteration of this disorder in key pathways may represent an intriguing new disease mechanism for schizophrenia and related neuropsychiatric diseases. Further research into this unexplored domain will be required to elucidate the role of the identified variants in schizophrenia etiology.

Rare copy number variants in neuropsychiatric disorders : Specific phenotype or not?

From a number of genome‐wide association studies it was shown that de novo and/or rare copy number variants (CNVs) are found at an increased frequency in neuropsychiatric diseases. In this study we examined the prevalence of CNVs in six genomic regions (1q21.1, 2p16.3, 3q29, 15q11.2, 15q13.3, and 16p11.2) previously implicated in neuropsychiatric diseases. Hereto, a cohort of four neuropsychiatric disorders (schizophrenia, bipolar disorder, major depressive disorder, and intellectual disability) and control individuals from three different populations was used in combination with Multilpex Amplicon Quantifiaction (MAQ) assays, capable of high resolution (kb range) and custom‐tailored CNV detection. Our results confirm the etiological candidacy of the six selected CNV regions for neuropsychiatric diseases. It is possible that CNVs in these regions can result in disturbed brain development and in this way lead to an increased susceptibility for different neuropsychiatric disorders, dependent on additional genetic and environmental factors. Our results also suggest that the neurodevelopmental component is larger in the etiology of schizophrenia and intellectual disability than in mood disorders. Finally, our data suggest that deletions are in general more pathogenic than duplications. Given the high frequency of the examined CNVs (1–2%) in patients of different neuropsychiatric disorders, screening of large cohorts with an affordable and feasible method like the MAQ assays used in this study is likely to result in important progress in unraveling the genetic factors leading to an increased susceptibility for several psychiatric disorders.

Response to Zhang et al. (2005) loss-of-function mutation in tryptophan hydroxylase-2 identified in unipolar major depression. Neuron 45, 11-16 [2] (multiple letters)

Zhang et al. reported a naturally occurring Arg441His missense variant of the human tryptophan hydroxylase-2 (TPH2) gene. The His441 allele was reported to be more abundant in a cohort of 87 depressed patients compared to 219 controls (Zhang et al., 2005). The frequency of His441 was higher in the depressed patients (0.06), among whom there were two His/His homozygotes and seven heterozygotes. His441 was also observed among the 219 controls (allele frequency 0.009), among whom one His/His homozygote and two Arg/His heterozygotes were detected. This reported association with depression is of note in the context of the effect of this substitution in reducing serotonin synthesis by approximately 80% in a heterologous expression assay in a rat cell line (Zhang et al., 2005) and through the observation of the role of TPH2 variants as genetic predictors of depression (Zill et al., 2004) and response to antidepressants (Peters et al., 2004). The authors of this letter represent three independent groups of investigators who have resequenced the relevant region of TPH2 in some 779 unrelated individuals (Table 1), including 403 with major depression (ages 19–74, n = 21 > 60 years). In addition, another 1740 individuals with major depression (from the STAR*D study, ages 18–75, n = 121 > 60 years) were genotyped (Table 1). The sequenced and genotyped individuals represent five ethnic populations. Psychiatric assessment was accomplished using semistructured psychiatric interviews: NIAAA, SCID or SADS-L; NIMH and UCSF, SCID-I/P. Major depression was diagnosed by DSMIII-R or DSM-IV criteria, and a DSM-IV checklist was used for the STAR*D samples. Additional descriptions of individual data sets have been reported (Nielsen et al., 1998; Robin et al., 1997; Roy, 2003; Peters et al., 2004; Rush et al., 2004). All data were collected following informed consent and under human research protocols approved by IRBs of the respective institutions. For direct sequencing, genomic DNA was amplified by PCR with primers encompassing the Arg441His variant, sequenced using the BigDye Terminator V3.1 (Applied Biosystems Inc., Foster City, CA) and analyzed on ABI 3100 or 3730 sequencers. For genotyping, assays were performed using 50-nuclease assay (TaqMan, ID # PMT06-55) and analyzed on an LJL plate reader (Molecular Devices, Sunnyvale, CA). 20% of the

Detailed analysis of the serotonin transporter gene (SLC6A4) shows no association with bipolar disorder in the Northern Swedish population.

Through active reuptake of serotonin into presynaptic neurons, the serotonin transporter (5‐HTT) plays an important role in regulating serotonin concentrations in the brain, and it is the site of binding for tricyclic antidepressants and selective serotonin reuptake inhibitors (SSRIs). Therefore it has been hypothesized that this transporter is involved in the etiology of bipolar (BP) disorder. Inconsistent association study results for the SLC6A4 gene encoding 5‐HTT reported in literature emphasize the need for more systematic and detailed analyses of this candidate gene. We performed an extensive analysis of SLC6A4 on DNA of 254 BPI patients and 364 control individuals from a Northern Swedish isolated population. This analysis consisted of a HapMap LD‐based association study including three widely investigated polymorphisms (5‐HTTVNTR, 5‐HTTLPR, and rs3813034), a copy‐number variation (CNV) analysis and a mutation analysis of the complete coding sequence and the 3′‐UTR of SLC6A4. No single marker showed statistically significant association with BPI, nor did any of the haplotypes. In the mutation analysis 13 novel variants were detected, including 2 amino acid substitutions M389V and I587L, but these are probably not implicated in risk for BP. No deletions or duplications were detected in the CNV analysis. We conclude that variation in the SLC6A4 gene or its regulatory regions does not contribute to the susceptibility for BP disorder in the Northern Swedish population.

Identification of rare copy number variants in high burden schizophrenia families

Over the last years, genome‐wide studies consistently showed an increased burden of rare copy number variants (CNVs) in schizophrenia patients, supporting the “common disease, rare variant” hypothesis in at least a subset of patients. We hypothesize that in families with a high burden of disease, and thus probably a high genetic load influencing disease susceptibility, rare CNVs might be involved in the etiology of schizophrenia. We performed a genome‐wide CNV analysis in the index patients of eight families with multiple schizophrenia affected members, and consecutively performed a detailed family analysis for the most relevant CNVs. One index patient showed a DRD5 containing duplication. A second index patient presented with an NRXN1 containing deletion and two adjacent duplications containing MYT1L and SNTG2. Detailed analysis in the subsequent families showed segregation of the identified CNVs. With this study we show the importance of screening high burden families for rare CNVs, which will not only broaden our knowledge concerning the molecular genetic mechanisms involved in schizophrenia but also allow the use of the obtained genetic data to provide better clinical care to these families in general and to non‐symptomatic causal CNV carriers in particular.

Evidence for the involvement of the glucocorticoid receptor gene in bipolar disorder in an isolated northern Swedish population

Ceulemans S, De Zutter S, Heyrman L, Norrback K‐F, Nordin A, Nilsson L‐G, Adolfsson R, Del‐Favero J, Claes S. Evidence for the involvement of the glucocorticoid receptor gene in bipolar disorder in an isolated northern Swedish population. Bipolar Disord 2011: 13: 614–623.

A network of synaptic genes associated with schizophrenia and bipolar disorder

Identification of novel candidate genes for schizophrenia (SZ) and bipolar disorder (BP), two psychiatric disorders with large epidemiological impacts, is a key research area in neurosciences and psychiatric genetics. Previous evidence from genome-wide studies suggests an important role for genes involved in synaptic plasticity in the risk for SZ and BP. We used a convergent genomics approach, combining different lines of biological evidence, to identify genes involved in the cAMP/PKA/CREB functional pathway that could be novel candidates for BP and SZ: CREB1, CREM, GRIN2C, NPY2R, NF1, PPP3CB and PRKAR1A. These 7 genes were analyzed in a HapMap based association study comprising 48 common SNPs in 486 SZ, 351 BP patients and 514 control individuals recruited from an isolated population in Northern Sweden. Genetic analysis showed significant allelic associations of SNPs in PRKAR1A with SZ and of PPP3CB and PRKAR1A with BP. Our results highlight the feasibility and the importance of convergent genomic data analysis for the identification of candidate genes and our data provide support for the role of common inherited variants in synaptic genes and their involvement in the etiology of BP and SZ.

Identification of a CACNA2D4 deletion in late onset bipolar disorder patients and implications for the involvement of voltage-dependent calcium channels in psychiatric disorders†

The GWAS‐based association of CACNA1C with bipolar disorder (BPD) is one of the strongest genetic findings to date. CACNA1C belongs to the family of CACN genes encoding voltage‐dependent calcium channels (VDCCs). VDCCs are involved in brain circuits and cognitive processes implicated in BPD and schizophrenia (SZ). Recently, it was shown that rare copy number variations (CNVs) are found at an increased frequency in SZ and to a lesser extent also in BPD, suggesting the involvement of CNVs in the causation of these diseases. We hypothesize that CNVs in CACN genes can influence the susceptibility to BPD, SZ, and/or schizoaffective disorder (SZA). A search for CNVs in eight CACN genes in a patient‐control sample of European decent was performed. A total of 709 BP patients, 645 SZ patients, 189 SZA patients, and 1,470 control individuals were screened using the Multiplex Amplicon Quantification (MAQ) method. We found a rare, partial deletion of 35.7 kb in CACNA2D4 in two unrelated late onset bipolar I patients and in one control individual. All three deletions shared the same breakpoints removing exons 17–26 of CACNA2D4, comprising part of the CACHE domain. Based on the data we cannot claim causality to BPD of the identified CACNA2D4 deletion but nevertheless this deletion can be important in unraveling the underlying processes leading to psychiatric diseases in general and BPD in particular.

PCM1 and schizophrenia: A replication study in the Northern Swedish population†

Previous studies implicated centrosomal dysfunction as a source of various neuropsychiatric disorders, including schizophrenia (SZ). Two recent reports [Gurling et al., 2006 ; Datta et al., 2008 . Mol Psychiatry] described an association between polymorphisms in the PCM1 gene and SZ in a UK/Scottish population. In this study, we aimed to replicate these findings in a Northern Swedish association sample of 486 research subjects with SZ and 512 unrelated control individuals. We genotyped 12 previously described SNP markers and carried out haplotype analyses using the same multi‐marker haplotypes previously reported. Though we could not replicate the association with SNPs rs445422 and rs208747, we did observe a significant protective association with intronic SNP rs13276297. Furthermore, we performed a meta‐analysis comprising 1,794 SZ patients and 1,553 controls, which confirmed the previously reported association with rs445422 and rs208747. These data provide further evidence that PCM1—though certainly not a major risk factor in the Northern Swedish population—cannot be ruled out as a contributor to SZ risk and/or protection, and deserves further replication in larger populations to elucidate its role in disease etiology.