Behavioral and Brain Functions BioMed Central

Background: Serotonergic neurotransmission has been implicated in suicidal behavior. Association between suicidal completers and a regulatory C(-1019)G polymorphism (rs6295) in the serotonin 1A receptor (HTR1A) gene was previously reported, whereas a following study showed no association in a sample of suicide attempters. Methods: The involvement of the implicated G-allele of the 5-HTR1A C(-1019)G polymorphism (rs6295) was analyzed with the transmission disequilibrium test (TDT) in a sample of 272 suicide attempter families. Results: No overtransmission of the G-allele was found in the entire sample of suicide attempters (p = 0.1460; n = 272 trios). However, a strong trend for overtransmission of the G-allele was observed in a sub-sample selected for a high level of previous traumatic and/or stressful life events prior to the suicide attempt (p = 0.0630, two-tail; n = 94 trios). Conclusion: The current results show that variation at the rs6295 polymorphism of the HTR1A gene is not associated with suicide attempts generally. However, the results indicate a possible role of the G-allele in suicidal behavior in connection with high exposure to traumatic and/or stressful life events, which is in need of future investigation. Background During the last three decades, evidence indicating that altered serotonergic neurotransmission influences suicidal behavior has accumulated. One of the more consistent findings is the observation of low cerebrospinal fluid (CSF) levels of 5HIAA (5-hydroxy-indole-acetic-acid, a serotonin metabolite) among suicide attempters [1-3]. As a consequence, research regarding specific genes that may influence suicidal behavior has mostly, focused on the serotonergic system [4]. In studies focused on the serotonin 1A receptor (HTR1A) gene, Nishiguchi et al. [5] found no association between suicide and two structural polymorphisms, Pro16Leu and Gly272Asp in this gene. However, a later study [6] observed a relationship between suicide and a functional Published: 20 April 2006 Behavioral and Brain Functions2006, 2:14 doi:10.1186/1744-9081-2-14 Received: 04 April 2006 Accepted: 20 April 2006 This article is available from: http://www.behavioralandbrainfunctions.com/content/2/1/14 © 2006Wasserman et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


Background
Schizophrenia is a complex phenotype with a multifactorial etiology. Reducing heterogeneity by identifying alternative phenotypes or valid subtypes that reflect specific neurobiological processes in the pathophysiology of the disorder is critical to uncovering susceptibility genes [1][2][3]. The concept of the deficit syndrome subtype of schizophrenia arose in an attempt to reduce clinical heterogeneity [4][5][6]. The deficit syndrome is characterized by primary (idiopathic) and enduring negative symptoms that persist through epochs of clinical stability and decompensation. By definition, idiopathic negative symptoms are not attributable to paranoid or psychotic withdrawal, depression, anxiety, or medication side effects. Schizophrenia patients who do not meet the deficit syndrome criteria are classified as nondeficit. Compelling evidence supports the construct validity of the deficit/nondeficit subgrouping on clinical features and prognosis, risk factors, neurocognitive and biological correlates, and pharmacological response profiles, and that dysfunction in the neural circuit that includes the dorsolateral prefrontal cortex (DLPFC) form the neuropathological basis for the deficit syndrome [6,7].
The reformulated dopamine hypothesis of schizophrenia describes interdependent prefrontal hypodopaminergia (associated with negative and cognitive symptoms) and subcortical hyperdopaminergia (associated with positive symptoms) [8]. The COMT gene is implicated in the pathophysiology of schizophrenia by its effect on prefrontal dopamine. A functional single nucleotide polymorphism (SNP) (Val 158 Met) of COMT gene results in two common variants, Valine (Val) and methionine (Met), which correspond to low and high synaptic dopamine respectively [9]. Associations between the Val 158 Met SNP and cognitive phenotypes have been described in healthy and schizophrenia individuals [10][11][12].
The COMT gene maps to chromosome 22q11.2, a region implicated in schizophrenia linkage studies, and in velocardiofacial syndrome (VCFS), which is characterized by psychosis and cognitive deficits [13,14]. COMT is thought to exert a distinctive effect on prefrontal dopamine-related information processing because it is primarily expressed in prefrontal cortical neurons [15,16], and is the principal enzyme in the degradation of dopamine (up to 60%) in the DLPFC; a role supported by data from animal studies [17]. Furthermore, the dopamine transporter, another key dopamine metabolizing enzyme in the brain is less expressed in the prefrontal cortex [18]. Deficit schizophrenia is characterized by impairments in tasks sensitive to frontal and parietal cortical functioning [19]. We have previously demonstrated association between COMT 158 Val allele and schizophrenia in a biracial cohort of patients from the Baltimore Metropolitan area [20]. In the current report, we have extended our previous study to evaluate the association between the variant Val allele and deficit/nondeficit schizophrenia. We hypothesized that the Val allele would occur more frequently in deficit than in nondeficit schizophrenia.

Materials and methods
Eighty-six unrelated schizophrenia individuals were recruited from the Maryland Psychiatric Research Center (MPRC). Healthy controls (n = 50) were recruited through newspaper advertisements targeting the same geographic region. All participants gave written informed consent, which was approved by the University of Maryland Institutional Review Board. Additional evaluation of the capacity to sign consent was performed on patients to assess their understanding of the study [21]. As in our previous study, there was no COMT genotype or allele association in African-American subjects, so this group was not included in the current study (n = 80). Patients met the Structured Clinical Interview for DSM-IV (SCID-IV) [22] criteria for schizophrenia or schizoaffective disorder. Additional categorization into deficit (n = 21), and nondeficit syndrome (n = 65) was performed with the Schedule for the Deficit Syndrome (SDS). For the diagnosis of deficit schizophrenia, this instrument requires the presence of two or more idiopathic and enduring negative symptoms (Kirkpatrick et al, 1989). Inter-rater reliabilities among the clinical interviewers were above 0.80 (interclass correlation). The Structured Interview for DSM-IV Personality Diagnoses (SIDP) [23], and Family History-Research Diagnostic Criteria (FH-RDC) [24] were administered in controls. Healthy controls had no Axis I diagnoses, or schizophrenia spectrum personality disorders.
Genotyping methods are described elsewhere [20]. Briefly, venous whole blood was collected and DNA isolated by standard means. COMT Val 158 Met SNP was determined as a restriction fragment length polymorphism (RFLP) after PCR amplification according to the method of Egan et al. (2001).
Fisher's exact test was used to evaluate whether the observed distribution of COMT Val 158 Met genotypes was consistent with that expected under Hardy Weinberg equilibrium. Nominal logistic regression (with age and sex as predictor variables) was used to determine the effect of genotype on disease status under an additive genetic model (i.e., by assigning values of -1, 0, and 1 to the three genotype classes). We first evaluated the effect of genotype on the total schizophrenia endpoint, combining the deficit and nondeficit subtypes. We then evaluated genotype effects on the deficit only schizophrenia endpoint.

Results
The sample consists of 136 European-American subjects (21 deficit cases, 65 nondeficit cases, and 50 healthy controls) presented in Table 1. The observed COMT genotype distributions were consistent with that expected under Hardy Weinberg equilibrium. Comparisons of COMT genotype frequencies between cases and controls are presented in Table 1. COMT genotype was significantly associated with schizophrenia (deficit and nondeficit combined; age-and sex-adjusted Odds Ratio = 3.24, p = 0.004 by genotype test), with Val allele frequencies of 0.68 and 0.43, in cases and controls, respectively. In contrast, genotype (and allele) frequencies did not differ between deficit and nondeficit cases (allele frequencies 0.74 and 0.66, respectively; age-and sex-adjusted p = 0.8). Deficit schizophrenia was observed in one female: COMT genetic variation has been previously shown to produce sexual dimorphism in brain dopamine levels; we repeated the analyses comparing COMT genotype between deficit and nondeficit subtypes in males only. We again observed that genotype (and allele) frequencies did not differ between both groups (allele frequencies 0.72 and 0.58, respectively; p = 0. 19). Given the small sample size, we estimated the least detectable difference in allele frequencies that could have been observed in our sample between deficit and nondeficit cases. Given the observed Val allele frequency of 0.66 in nondeficit cases, the frequency in the deficit group would have had to have been less than 0.42 or greater than 0.87 in order to detect a significant difference between groups (at alpha = 0.05) with 80% power.

Discussion
In this report we did not observe a significant difference between deficit and nondeficit schizophrenia subtypes in either COMT allele or genotype frequencies. Our report however provides an extension of our earlier finding of association between COMT 158 Val allele and schizophrenia in a larger sample of European-American subjects with the addition of deficit schizophrenia individuals. Similar to our prior report, we failed to find association between COMT gene and schizophrenia in the African-American subgroup. This suggests that COMT allele frequencies might be different in both ethnic groups in this Baltimore sample. To avoid potential population substratification, we did not include African-Americans in the current report. In our previous study based on 61 European-Americans with schizophrenia, we observed allele frequencies of 0.62 and 0.50 in cases and healthy controls (p = 0.04). The differences in allele frequencies persisted, and indeed increased in this study on a modestly expanded sample (to 0.680 vs. 0.430), which includes 86 European-American schizophrenia cases presented in this report. Evidence from family-based and case-control studies generally suggest that COMT Val 158 Met polymorphism might be associated with schizophrenia [25] with some negative reports [26,27]. The main limitation of this study is the small sample size, which may have limited detecting associations of only modest size between COMT gene and deficit/nondeficit subtypes. The differences are so small that the power needed to detect a significant difference between both subtypes would indeed require a large sample. However, our findings are unlikely due to type II error. Our reported allele and genotype frequencies were similar between the deficit and nondeficit schizophrenia subgroups, but considerably higher than the frequencies observed in healthy controls. In this sample, the patient group is comprised of a significantly higher proportion of males than the control group. Furthermore, there are substantially more males in the deficit group, which is consistent with prior reports on deficit schizophrenia (Kirkpatrick et al, 2001). Prior studies have reported a sexual dimorphism in COMT-associated dopamine levels on several neuropsychiatric phenotypes in a region-specific manner, particularly in the frontal cortex [28,29]. Analyses using a regression model controlled for age and sex effects on our COMT findings in this Baltimore sample.

Conclusion
The results do not support a differential COMT gene effect on the deficit/nondeficit subtypes of schizophrenia. Future studies in larger independent samples investigating joint interactions with other candidate dopamine system genes (e.g. dopamine transporter gene [DAT1]) and COMT gene on schizophrenia endophenotypes are warranted.