Glutathione S-Transferase Deletion Polymorphisms in Early-Onset Psychotic and Bipolar Disorders: A Case-Control Study

Abstract

Objective: To examine glutathione S-transferase (GST) deletion polymorphisms in development of early-onset severe mental disorders, with the hypothesis that patients with GSTM1-null and GSTT1-null genotypes will develop psychotic disorders at a younger age.

Methods: We identified GSTM1 and GSTT1 deletion polymorphisms by multiplex polymerase chain reaction (PCR) in 93 patients with early onset severe mental disorders and 278 control individuals. The diagnoses were confirmed by Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version and Schedule for Affective Disorders and Schizophrenia–Life-Time Version (K-SADS-PL) interviews.

Results: Individuals with the GSTM1-null genotype were at 3.36-fold higher risk of developing early-onset severe mental disorders than carriers of a corresponding active genotype. The risk of those disorders was increased by 6.59-fold in patients with GSTM1-null/GSTT1-active genotype. Patients with the GSTM1-null genotype were at approximately 2-fold increased risk for developing early-onset schizophrenia-spectrum disorder (EOS), early-onset bipolar disorder (EOBD) with psychotic symptoms, or early-onset first-episode psychosis (EOFEP), compared with patients with the GSTM1-active genotype.

Conclusion: The GSTM1-null genotype might be associated with higher risk for early onset severe mental disorders.

Identification and evaluation of early-onset psychotic and bipolar disorders is an important health concern. Because these disorders may be transient, intermittent, and short-term, or part of a long-term psychiatric condition, recognition of risk factors, early detection, and intervention might improve the outcome in patients with these conditions.

Schizophrenia is a chronic, degenerative disorder of the brain characterized by positive (hallucinations, delusions) and/or negative symptoms (decreased motivation, socialization, cognitive dysfunction). The high heritability of schizophrenia has been well established in twin and adoption studies.¹ The onset of schizophrenia typically occurs in early adulthood. Childhood onset is relatively rare, with approximately 1 child in 10,000 affected in the general population; the prevalence of schizophrenia increases in adolescence and peaks in early adulthood. The definition of early-onset schizophrenia (EOS), or adolescent-onset schizophrenia, varies among studies and most commonly refers to schizophrenia with onset before the age of 17 years to 21 years. EOS is associated with higher familial risk, a higher ratio of males affected, and fewer offspring per proband.^2-4 Another important early-onset entity, which is often the first of multiple episodes in chronic schizophrenia, is early-onset first-episode psychosis (EOFEP). Further, early-onset bipolar disorders (EOBPs) also share common features with schizophrenia, including cognitive characteristics, environmental factors, premorbid developmental impairments, neurological signs and heritability, and changes in antioxidant status.^5,6

Several lines of evidence suggest increased oxidative stress in patients with schizophrenia.^7,8 The findings of some genetic association and gene expression studies suggest that, due to polymorphisms in genes encoding antioxidant enzymes, patients with schizophrenia might have altered ability to enact antioxidant mechanisms.^9-12 Several promising candidate genes, including those that encode enzymes involved in glutathione (GSH) metabolism, have been suggested.^10-14 GSH is the key element of antioxidant defense mechanisms in the brain.¹⁵ Gawryluk and associates found decreased levels of GSH in postmortem samples of prefrontal cortex tissue in patients with bipolar disorder and schizophrenia.¹⁶ It has been suggested that these decreased levels may occur due to the reduction in glutathione S-transferase (GST) activity in these brain regions.¹⁷ GST family members play a dual role because they are able to detoxify numerous xenobiotics and possess strong antioxidant activity towards reactive oxygen species (ROSs).¹⁸ Human GSTs consist of 3 families: cytosolic, mitochondrial, and membrane-associated proteins in eicosanoid and glutathione metabolism (membrane-associated proteins in eicosanoid and glutathione metabolism [MAPEG] family). Cytosolic GSTs are further categorized into 7 major classes according to their amino-acid sequence: alpha (5 members), mu (5 members), pi (1 member), theta (2 members), zeta (1 member), omega (2 members), and sigma (1 member) subfamilies. Although members of the same class possess more than 40% amino-acid sequence identity, less than 25% of sequence identity exists among classes.¹⁹ Almost all members of the GST family exhibit genetic polymorphism, resulting in the complete lack or lowering of enzyme activity.²⁰ Due to the homozygous deletion of the GSTM1 gene (NM_000561; GSTM1-null genotype) or GSTT1 gene (NM_0008553.3; GSTT1-null genotype), approximately 50% or 20% of white individuals lack GSTM1 or GSTT1 enzyme activity, respectively.²¹ Specifically, the null genotypes have both alleles missing, causing the gene to become nonfunctional.

Apart from Gawryluk et al, who report that GSTM1 activity was significantly reduced in patients with schizophrenia compared with control individuals, several other population studies have implicated GST gene polymorphisms in the pathogenesis of schizophrenia.^22‐27 The contradictory findings that researchers have obtained so far indicate that other factors, such as ethnicity, should also be considered when assessing risk factors in schizophrenia.²⁸

Apart from patients with SCH, the data on GST polymorphism in patients with psychotic or bipolar disorder in whites are scarce. Thus, in this study, we examined the role of deletion polymorphisms in GSTM1 and GSTT1 genes in the development of early-onset psychotic and bipolar disorders. To our knowledge, this study is among the first to examine GST polymorphisms in early-onset mental disorders. We hypothesize that patients with GSTM1-null and GSTT1-null genotypes will be more susceptible to oxidative stress and prone to developing psychotic disorders at a younger age, compared with their peers who possess other genotypes.

Materials and Methods

Study Subjects

The study was conducted with 93 patients (31 female, 62 male; average age, 21.18 [6.68] years) diagnosed with early-onset severe mental disorders. Our sample included 3 different entities: EOS-spectrum disorder, EOBD with psychotic symptoms, and EOFEP. In the EOS and EOBD groups, we included adult patients with diagnoses verified before the age of 18 years. For the EOFEP group, we included patients younger than 18 years whose disorder had manifested for the first time.

Subjects were recruited as consecutive referrals from the Institute of mental health, University psychiatry clinic, Belgrade, Serbia between June 1, 2009 and December 31, 2011. The inclusion criterion was the presence of schizophrenia, bipolar disorder with psychotic symptoms, and first psychotic episode that had started before the age of 18 years. The diagnosis was verified by the International Classification of Diseases, Revision 10 (ICD-10)²⁹ and the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV)³⁰ criteria confirmed by a referring pediatric/adolescent psychiatrist or adult psychiatrist, through clinical interview with the parent and examination of the child or adolescent, or interviewing only the patient if he or she was older than 18 years.

The control group consisted of 278 healthy control subjects (95 female, 183 male; average age, 31.60 [4.46] years), matched for sex, age, ethnicity, and geographic origin. Although the average age in the control group seems to be higher compared with the study group, statistical significance was not reached (P> .05); this lack of statistical significance, however, is unimportant because EOS, EOBP, and EOFEP are all early-onset disorders. On a practical level, this means that controls who did not have a psychiatric diagnosis up to the moment of their inclusion in the study did not develop any of the examined entities. After informed consent was obtained, subjects were interviewed using a standard questionnaire to collect information, including demographic characteristics, presence of mental disorders, previous hospitalizations, and pharmacotherapy undergone. Only subjects without any previous mental disorders were included in the control group.

The Institutional Ethics Committee of Institute of Mental Health approved this study, and the research was carried out in accordance with the International Code of Medical Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. Before participating, all participants gave written informed consent. For participants younger than 18 years, written informed consent was obtained from their parent(s) or legal guardian(s).

Assessment Tools

We assessed the participants younger than 18 years by using the Schedule for Affective Disorders and Schizophrenia for School-Age Children–Present and Lifetime version (K-SADS–PL), which generates reliable and valid psychiatric diagnoses for children and adolescents. This tool is a semistructured diagnostic interview designed to assess current and past episodes of psychopathologic manifestations in children and adolescents, according to the Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition–Revised (DSM-III-R) and DSM-IV criteria.³¹

For participants older than 18 years but for whom there was no clear evidence in their medical records that their illness had started before the age of 18 years, we assessed them using the Schedule for Affective Disorders and Schizophrenia Lifetime (SADS-L). This tool is a semistructured diagnostic interview that assesses current and lifetime history of Axis I disorders.^25,31

DNA Extraction and Genotyping

We extracted DNA from blood leucocytes using the QIAGEN QIAamp Mini Kit (Qiagen, Inc.). The DNA purification is performed in 4 steps, using spin columns in a standard microcentrifuge. First, the lysate buffering conditions are adjusted to allow optimal binding of the DNA to the spin-column membrane. After loading the specimen, DNA is adsorbed onto the silica membrane during a brief centrifugation, followed by 2 washing steps undertaken to remove residual contaminants. In the final step, purified DNA in a concentrated form is eluted from the spin column using an elution buffer; the resulting product is suitable for direct use in polymerase chain reaction (PCR).

We performed GSTM1 and GSTT1 genotyping via PCR (Image 1), as described by Garcia-Closas³² and Pemble and Taylor,³³ respectively. GSTM1 genotyping was performed by multiplex PCR; the primers we used were as follows: GSTM1 forward: 5′-CTGCCCTACTTGATTGATGGG-3′ and GSTM1 reverse: 5′-CTGGATTGTAGCAGATCATGC-3 (isolated DNA (100-150 ng) was amplified in a total volume of 25-μl reaction mixture containing 20 pmol of primers, commercial Master Mix, and water (Thermo Fisher Scientific, Inc) and subjected to initial denaturation at 94 °C for 4 minutes, followed by 33 cycles at 94° C for 2 minutes, 59 °C for 1 minute, and 72 °C for 1 minute. We performed the final extension at 72 °C for 10 minutes. The PCR products were analyzed in 2% agarose gels, electrophoresed for approximately 20 minutes (125V constant, 0.27A, 50W) at room temperature, and visualized using ethidium bromide staining (Image 2). The assay does not distinguish heterozygous or homozygous wild type genotypes and therefore detects the presence (at least 1 allele present, homozygote or heterozygote) or the absence (complete deletion of both alleles, homozygote) of the genotype. As a result, the GSTM1-active genotype was detected by the band at 215 bp, and the absence of this particular band was indicative of the GSTM1-null genotype. GSTT1 genotyping also was performed by multiplex PCR, using GSTT1-forward: 5′-TTCCTTACTGGTCCTCACATCTC-3′ and GSTT1-reverse: 5′-TCACCGGATCATGGCCAGCA-3′ primers (Figure 1) under the same thermal cycler and gel conditions as for GSTM1 genotyping. Similarly, because the assay does not distinguish between heterozygous or homozygous wild-type genotypes, the presence of 480-bp bands was indicative for the GSTT1-active genotype and the absence of the GSTT1-null genotype (Figure 2). Exon 7 of the CYP1A1 housekeeping gene was coamplified in GSTM1 and GSTT1 genotyping and was used as an internal control for the presence of the amplifiable DNA. We used the following primers: CYP1A1 forward: 5′-GAACTGCCACTTCAGCTGTCT-3′ and CYP1A1 reverse: 5′-CAGCTGCATTTGGAAGTGCTC-3′; the CYP1A1 PCR product corresponded to 312 bp.

Image 1.

A 2% agarose gel electrophoretogram. PCR products of the GSTM1 gene. Lanes 1, 2, 9, 11 and 12 are patients with the GSTM1 active genotype (215 bp band) and lanes 3 through 8 with lane 10 represent the GSTM- null genotype; 312bp band represents the CYP1A1 housekeeping gene, used as internal control for amplifiable DNA; M, DNA Q2 marker; N, negative control without a DNA content.

Image 2.

A 2% agarose gel electrophoretogram. PCR products of the GSTT1 gene. Lanes 1, 3, 4 and 5 are patients with the GSTT1 active genotype (480bp band) and lane 2 represents the GSTT1- null genotype; 312bp band represents the CYP1A1 housekeeping gene, used as internal control for amplifiable DNA; M, DNA Q2 marker; N, negative control without DNA content.

Figure 1.

The GSTM1 gene is situated at chromosome 1 (1p13.3), composed of 8 exons, spanning a region of 21,244 bases. GSTM1 null allele arose from a recombination event resulting in deletion of a 20-kb segment. This deletion produces a novel 7.4-kb HindIII fragment with the loss of 10.3- and 11.4-kb HindIII fragments, hence homozygotes for GSTM1 null allele produce no GSTM1 protein. The end points of the polymorphic GSTM1 deletion are: the left repeated region 5 kb downstream from the 3’-end of the GSTM2 gene and 5 kb upstream from the beginning of the GSTM1 gene; the right repeated region 5 kb downstream from the 3’-end of the GSTM1 and 10 kb upstream from the 5’-end of the GSTM5 gene.

Figure 2.

The GSTT1 gene is situated at chromosome 22 (22q11.23), and composed of 5 exons, spanning a region of 8,179 bases. PCR mapping and sequencing revealed a 54251 bp fragment including GSTT1 to be deleted from chromosome 22, most likely by a homologous recombination event between 2 highly homologous sequence stretches that flank GSTT1.

Statistical Analysis

We performed statistical analysis using Statistical Package for the Social Sciences software, version 20.0 (SPSS Inc). In descriptive statistics, we summarized continuous variables by mean (SD). The relative associations between the studied genotypes and early-onset severe mental disorders were evaluated by multinomial logistic regression to calculate odds ratios (ORs) and 95% confidence intervals (CIs). We compared the distributions of GSTM1 and GSTT1 genotypes between case individuals and controls, as well as different types of early onset severe mental disorders, using the χ² test.

Results

Baseline characteristics of controls and patients with early-onset severe mental disorders are presented in Table 1. There was no statistically significant difference between case and control subjects regarding age and sex distribution (P> .05 for each). Among 93 patients included in the study, 49 had EOS (53%), 12 had EOBP (13%), and 32 EOFEP (34%). To avoid diagnostic instability, the diagnostic spectra were confirmed after 1 year of follow-up. Most of the patients were hospitalized and treated by pharmacotherapy (96% and 98%, respectively).

Table 1.

Baseline Characteristics of Control Individuals and Patients With Early Onset Severe Mental Disorders

Characteristic	Casesa	Controlsb
Age (y), mean (SD)	21.18 (6.68)	31.60 (4.46)
Sex, no. (%)
Male	62 (66)	183 (66)
Female	31 (33)	95 (34)
Age at diagnosis (y), mean (SD)	16.07 (4.73)	NA
Disorder type, no. (%)
EOS	49 (53)	NA
EOBD	12 (13)	NA
EOFEP	32 (34)	NA
Hospitalized
Yes	89 (96)	0
No	4 (4)	278 (100)
Pharmacotherapy
Yes	91 (98)	0
No	2 (2)	278 (100)
Epilepsy
Yes	1 (1)	0
No	92 (99)	278 (100)

EOS, early-onset schizophrenia spectrum disorder; EOBD, early-onset bipolar disorder; EOFEP, early-onset first-episode psychosis; NA, not applicable.

^an = 93.

^bn = 278.

We conducted genotyping all recruted patients and controls. Distribution of individual GSTM1 and GSTT1 genotypes, as well as combined GST genotypes in controls and patients with early-onset psychotic mental disorders, are shown in Table 2. The GSTM1-null genotype was more frequent among patients (57%) compared with controls (49%), with an adjusted OR of 3.36 (95% CI, 0.99-11.33; P> .046). However, no significant difference was observed in the distribution of GSTT1 gene variants between patients and controls (P> .05).

Table 2.

Distribution of Individual and Combined GSTM1 and GSTT1 Genotypes in Patients and Control Individuals

GST Genotype	Patients, No. (%)a	Controls, No. (%)b	OR	CI (95%)	P Value
GSTM1
activec	40 (43)	141 (51)	1.00 [reference]
nulld	53 (57)	137 (49)	3.36	.99-11.33	.046
GSTT1
nulld	22 (24)	65 (23)	1.00 [reference]
activec	71 (76)	213 (77)	2.96	.71-12.22	.13
Combined
GSTM1 active/GSTT1 null	8 (9)	42 (15)	1.00 [reference]
GSTM1 active/GSTT1 active	32 (34)	99 (36)	2.62	.32-11.18	.37
GSTM1 null/GSTT1 null	14 (15)	23 (8)	9.41	.52-24.94	.12
GSTM1 null/GSTT1 activec	39 (42)	114 (41)	6.59	.95-25.65	>.05

GST, glutathione S-transferase; CI, confidence interval; OR, odds ratio, adjusted for age and sex.

^an = 93.

^bn = 278.

^cActive (present) if at least one active allele is present.

^dInactive (null) if no active alleles are present.

When we analyzed the association of combined GSTM1/GSTT1 gene variants with the risk for early-onset severe mental disorders, the risk of developing those disorders was increased 6.59-fold in patients with GSTM1-null/GSTT1-active genotype; this result was found to be marginally statistically insignificant (OR, 6.59; 95% CI, .95-25.65; P> .05). The risk was also increased 9.41-fold in patients with the GSTM1-null/GSTT1-null genotype, although it was statistically insignificant (OR, 9.41; 95% CI, .52-24.94; P = .12) (Table 2).

In an attempt to verify whether GST polymorphisms might be considered a risk factor for any group of disorders individually, we analyzed distribution of individual GSTM1 and GSTT1 genotypes, as well as combined GST genotypes, in patients with EOS, EOBP, and EOFEP compared with controls (Table 3). The results of our analysis showed that patients with the GSTM1-null genotype were at approximately 2-fold increased risk for developing EOS, EOBP, and EOFEP (OR = 1.86, 95% CI = .74-4.66, P =  .18; 2.33, .58-9.37, P = .23; and 2.24, .72-6.95, P = .16, respectively), compared with patients with GSTM1-active genotype, although these results did not reach statistical significance. However, similar effects were not observed in case of GSTT1 polymorphism.

Table 3.

Individual and Combined GSTM1 and GSTT1 Genotypes as Risk Factors for EOS, EOBP, and EOFEP

	Disorder Type
GST Genotype	EOS	EOBP	EOFEP
GSTM1 active a
Ca/Co	22/141	5/141	13/141
OR (95% CI)	1.00 [reference]	1.00 [reference]	1.00 [reference]
P value
GSTM1 nullb
Ca/Co	28/137	7/137	18/137
OR (95% CI)	1.86 (.74-4.66)	2.33 (.58-9.37)	2.24 (.72-6.95)
P value	.18	.23	.16
GSTT1 nullb
Ca/Co	9/65	4/65	9/65
OR (95% CI)	1.00 [reference]	1.00 [reference]	1.00 [reference]
P value
GSTT1 activea
Ca/Co	41/213	8/213	22/213
OR (95% CI)	1.65 (.53-5.09)	.42 (.90-2.05)	1.00 (.27-3.69)
P value	.38	.28	.99
GSTM1 active/GSTT1 null
Ca/Co	4/42	2/42	2/42
OR (95% CI)	1.00 [reference]	1.00 [reference]	1.00 [reference]
P value
GSTM1 active/GSTT1 active
Ca/Co	18/99	3/99	11/99
OR (95% CI)	1.43 (.32-6.43)	.75 (.08-6.08)	2.10 (.33-13.34)
P value	.63	.80	.42
GSTM1 null/GSTT1 null
Ca/Co	5/23	2/23	7/23
OR (95% CI)	2.52 (.29-22.17)	2.36 (.24-24.32)	5.16 (.51-51.53)
P value	.39	.47	.16
GSTM1 null/GSTT1 active
Ca/Co	23/114	5/114	11/114
OR (95% CI)	2.9 (.65-12.90)	.67 (.11-4.05)	4.16 (.60-28.68)
P value	.160	.67	.14

GST, glutathione S-transferase; EOS, early-onset schizophrenia spectrum disorder; EOBD, early-onset bipolar disorder; EOFEP, early-onset first-episode psychosis; Ca/Co, no. of case vs control individuals; OR, odds ratio, adjusted for age and sex; CI, confidence interval

^aActive (present) if at least one active allele is present.

^bInactive (null) if no active alleles present.

When we analyzed the effect of combined GST genotypes, the highest risk was observed in the group of patients with EOFEP carrying a combination of GSTM1-null/GSTT1-active genotypes (OR, 4.16; 95% CI, .60-28.68; P = .14) and GSTM1-null/GSTT1-null genotypes (OR, 5.16; 95% CI, 0.51-51.53; P = .16), although these results were statistically insignificant. However, patients with EOS who are carriers of GSTM1-null/GSTT1-active and GSTM1-null/GSTT1-null exhibited more than 2.5-fold increased risk for disease development, whereas in patients with EOBP, increased risk was observed only in carriers of both null genotypes (Table 3).

Discussion

The results of this study showed that the GSTM1-null genotype is more frequent among patients compared with controls. Apart from the independent association of GSTM1-null genotype with the risk of early-onset mental disorders, the risk of those disorders was also increased in patients with the combined GSTM1-null/GSTT1-active genotype. Further, patients with GSTM1-null genotype exhibited higher risk for developing EOS, EOBP, and EOFEP, suggesting a possible association of the GSTM1-null genotype with higher risk for developing psychotic disorders at a young age.

It has been suggested that oxidative stress might be implicated in the pathophysiology of schizophrenia.^2,7 Apart from the antioxidant-enzyme defense mechanisms, which seem to be decreased in patients with EOS, glutathione deficit might also be implicated in the pathogenesis of this type of psychotic disorder.^16,17,22,34 GSH is the main nonenzymatic cellular antioxidant and it plays a critical role in cellular protection from ROS-induced damage.³⁵ Further, this tripeptide is involved in the disposal of peroxides by brain cells and protection against ROS, although its content strongly depends on the availability of its precursors.³⁶ The cytosolic GST family catalyzes the conjugation of electrophilic compounds, including products of oxidative stress, with GSH.¹⁸ Although several types of allelic variations have been identified within GST classes, GSTM1 gene has received the most attention in genetic epidemiologic studies in patients with schizophrenia.^{24,28,37‐39} Some study results suggested increased risk associated with the GSTM1-null genotype,^26,37,39 whereas other studies found no association.^12,25 To our knowledge, our results on the independent contribution of the GSTM1 polymorphism towards the risk of early-onset schizophrenia are among the first in the literature because most previous studies on similar topics were conducted in adult subjects. Also, evaluation of the contribution of GSTM1 common copy number variation (CNV) to vulnerability towards schizophrenia also received a lot of attention.^28,40 Although Tam et al have indicated that CNVs confer an increased susceptibility to schizophrenia,⁴¹ obtained results on the contribution of GSTM1 that preceded and followed these findings have been inconsistent.^{12,28,37‐39}

Genetic polymorphism of another widely investigated GST polymorphism, GSTT1, was found to be associated with decreased or increased risk of schizophrenia, or in some cases showed lack of any effect. We have shown that individual GSTT1 polymorphism does not independently contribute to the risk of schizophrenia. These findings are in agreement with those of Matsuzawa et al¹² However, Raffa et al suggested the GSTT1 gene as a candidate gene for susceptibility to schizophrenia in a Tunisian population.²⁵ This finding was supported by the findings of Kashani et al, who suggested that lack of GSTT1 function, meaning the GSTT1-null genotype, might increase the risk of schizophrenia.²⁶ Also, Saadat et al have found that the GSTT1-null genotype was associated with significantly reduced risk of developing schizophrenia, proposing the GSTT1-active genotype as a candidate gene for susceptibility to schizophrenia.⁴²

Results obtained in analysis of the combined effect of GSTM1 and GSTT1 genotypes are, again, inconsistent. Our study has shown, however, with marginal statistical significance, that the risk of developing early-onset mental disorders was increased 6.59-fold in patients with the GSTM1-null/GSTT1-active genotype. This result is in agreement with the findings of Kashani et al, who suggested that the impairment in the function of GSTs may increase the risk of schizophrenia.²⁶ The role of GSTs as phase II biotransformation enzymes may provide a possible explanation for this finding—specifically, under physiological conditions, GST Mu-class catalysis conjugation of catechol o-quinones with glutathione.³⁷ Reduced or lacking activity of these enzymes may lead to an excess of neurotoxic compounds of catecholamine o-quinones, which may contribute to the development of certain forms of schizophrenia.³⁷

However, not all reactions catalyzed by GSTs will result in detoxification. In particular, the GST Theta class is responsible for bioactivation of certain toxic and mutagenic compounds and drugs.⁴³ However, the role of GSTs in antioxidant defense should not be forgotten because oxidative stress has been suggested as a potential mechanism underlying the development of schizophrenia. Gravina et al have indicated that combination of different GST polymorphisms has a role in predisposition to schizophrenia, probably due to affecting the capacity of the cell to detoxify the oxidized metabolites of catecholamine. In other words, genetic polymorphism of GSTs results in the complete lack or lowering of the enzyme activity. Hence, Gravina et al found that combination of the GSTM1-null, GSTA1 low activity, and GSTT1-active genotypes represents a risk factor for schizophrenia.²³

Micó et al have suggested that patients with more severe and chronic diseases, such as schizophrenia and bipolar disorder, have higher markers of oxidative stress than patients with first early-onset psychotic episode.⁴⁴ However, our results on the distribution of GST genotypes in patients with EOS, EOBP, and EOFEP showed that the GSTM1-null genotype was more frequent in all 3 types of mental disorders, although the GSTT1-active genotype was more frequent in the EOS group and the GSTT1-null genotype in the EOBP and EOFEP groups of patients. Still, no significant difference regarding the distribution of GST genotypes was observed among different types of early-onset severe mental disorders.

Apart from the possible role of GST polymorphisms in susceptibility for developing various mental disorders, GSTs might also be evaluated from a different perspective. It has recently been suggested that oxidative stress might mediate vascular damage in patients with psychosis, and it has been established that patients with psychosis are at higher risk for developing cardiovascular diseases⁴⁵ and diabetes mellitus.⁴⁶ GSTM1-null and GSTT1-null genotypes have already been recognized as potential determinants of higher risk for coronary artery disease⁴⁷ and even cardiovascular death,⁴⁸ as well as an increased susceptibility to advanced atherosclerosis⁴⁹ in patients without mental psychiatric diseases. As a result, it seems reasonable to assume that these genetic polymorphisms might make certain contributions to the development and progression of mental disorders.

Some limitations in the present study should be considered. First, it is well known that in case-control studies, selection bias might influence the results. Our control group was hospital-based; therefore, the use of population controls may have been more appropriate. Also, future researchers on this topic should strongly consider assessing a larger sample. A larger sample size might verify the difference in the effect of GST polymorphisms among various types of early-onset severe mental disorders. Regarding ethnic specificity, because it has already been suggested that ethnicity might affect antioxidant defense mechanisms, further genetic association studies of GST genes in patients with severe mental disorders from different ethnic populations (eg, whites, ethnic Japanese, and African Americans) might lead to less inconsistent results.

Glossary

Abbreviations

EOS

early-onset schizophrenia

EOFEP

early-onset first-episode psychosis

EOBPs

early-onset bipolar disorders

GSH

glutathione

GST

glutathione S-transferase

ROSs

reactive oxygen species

MAPEG

membrane-associated proteins in eicosanoid and glutathione metabolism

K-SADS-PL

Schedule for Affective Disorders and Schizophrenia for School-Age Children–Present and Lifetime Version

ICD-10

International Classification of Diseases, Revision 10

DSM-IV

Diagnostic and Statistical Manual of Mental Disorders, 4^th Edition

DSM-III-R

Diagnostic and Statistical Manual of Mental Disorders, 3^rd Edition–Revised

SADS-L

Schedule for Affective Disorders and Schizophrenia Lifetime

PCR

polymerase chain reaction

ORs

odds ratios

CIs

confidence intervals

CNV

copy number variation

NA

not applicable