Taiwanese Journal of Psychiatry

: 2019  |  Volume : 33  |  Issue : 2  |  Page : 83--91

Relations of genetic variants in superoxide dismutase 2 and dystrobrevin-binding protein 1 to methamphetamine psychosis among methamphetamine dependents in Taiwan

Ying-Ching Lai1, Shih-Ku Lin2, Ming-Chyi Huang2, Hsing-Cheng Liu2, Yen-Lung Chiou3, Chun-Hsin Chen4, Wei J Chen5,  
1 Institute of Epidemiology and Preventive Medicine, College of Public Health, National Taiwan University, Taipei; Department of Psychiatry, Cardinal Tien Hospital Yonghe Branch, New Taipei City, Taiwan
2 Department of Psychiatry, Taipei City Psychiatric Center, Taipei City Hospital; Department of Psychiatry, School of Medicine, Taipei Medical University, Taipei, Taiwan
3 Department of Psychiatry, Taipei City Psychiatric Center, Taipei City Hospital, Taipei, Taiwan
4 Department of Psychiatry, School of Medicine, Taipei Medical University; Department of Psychiatry, Taipei Medical University-Wan-Fang Hospital, Taipei, Taiwan
5 Institute of Epidemiology and Preventive Medicine, College of Public Health; Centers of Genomic Medicine and Precision Medicine; Department of Psychiatry, College of Medicine and National Taiwan University Hospital, National Taiwan University, Taipei, Taiwan

Correspondence Address:
Wei J Chen
No. 17, Xu-Zhou Road, Taipei 100


Objectives: Both superoxide dismutase 2 gene (SOD2), encoding a free radical scavenger, and dystrobrevin-binding protein 1 gene (DTNBP1), a candidate gene for schizophrenia, have been implicated in methamphetamine (METH) psychosis. In this study, we intended to compare the distribution of those two genes between METH users with and without psychosis and to evaluate whether the length of time of METH exposure influenced such relationship. Methods: The study sample consisted of 84 cases of patients with METH-induced psychosis and 187 controls (METH users without psychosis). Five single-nucleotide polymorphisms (SNPs) in SOD2 and three SNPs in DTNBP1 were genotyped. We did both single-locus and haplotype analyses, and adjusted for multiple comparisons with an effective number of markers, and multivariable logistic regression analyses for adjusting for age, sex, and duration of METH use. Results: None of the individual SNPs were associated with METH-induced psychosis after adjustment for multiple comparisons. In the subgroup analysis, both rs4880 and rs2855116 in SOD2 were significantly associated with prolonged METH-induced psychosis (p < 0.01 and p < 0.01, respectively). Under the assumption of a codominant model, the CC genotype of rs4880 was significantly associated with METH-induced psychosis after adjustments for age, sex, and the duration of METH use (p < 0.01). This association was not supported by the haplotype analyses or gene–gene interactions between SOD2 and DTNBP1. Conclusion: Functional C-allele of rs4880 in SOD2 was associated with the prolonged subtype of METH-induced psychosis in Taiwanese population. Oxidative stress mechanisms show a rôle in the development of METH-induced psychosis.

How to cite this article:
Lai YC, Lin SK, Huang MC, Liu HC, Chiou YL, Chen CH, Chen WJ. Relations of genetic variants in superoxide dismutase 2 and dystrobrevin-binding protein 1 to methamphetamine psychosis among methamphetamine dependents in Taiwan.Taiwan J Psychiatry 2019;33:83-91

How to cite this URL:
Lai YC, Lin SK, Huang MC, Liu HC, Chiou YL, Chen CH, Chen WJ. Relations of genetic variants in superoxide dismutase 2 and dystrobrevin-binding protein 1 to methamphetamine psychosis among methamphetamine dependents in Taiwan. Taiwan J Psychiatry [serial online] 2019 [cited 2019 Sep 19 ];33:83-91
Available from: http://www.e-tjp.org/text.asp?2019/33/2/83/261752

Full Text


Methamphetamine (METH) is a central nervous system stimulant that acts on several neurotransmitter systems and exerts neurotoxic effects on dopaminergic and other neurons [1],[2]. METH is highly addictive, and its use continues to be one of the most worrying threats of drug use in East and South-East Asia [3]. With the capture-recapture method on information collected from both judicial and medical systems, the prevalence of male METH use in the northern Taiwan from 1999 to 2002 has been estimated to be ranged from 1.24% to 2.47% [4]. According to the 2014 national survey in Taiwan, the lifetime use prevalence of METH is 0.6%, remaining the most frequently used illicit drug since the previous two waves of national surveys in 2005 and 2009 [5].

One severe consequence of METH consumption is METH-induced psychosis [6]. More than half of the METH users experience psychotic symptoms [7]. About 13% of METH users have been found to meet the diagnostic criteria for METH-induced psychosis [8],[9], which increases health service use, poor outcomes, the incidence of psychiatric symptoms across many domains over time, and premature death [10]. Genetic factors have accounted for 43% of the variance in METH use in a large twin study [11], but little is known about the contribution of genes to METH-induced psychosis. Nevertheless, circumstantial evidence suggests that a genetic component may be involved in the susceptibility to METH-induced psychosis. Patients with METH-induced psychosis and their relatives have more schizoid/schizotypal personality traits, which are indicators of genetic vulnerability to schizophrenia, than their counterpart METH users without psychosis [12]. In addition, METH-induced psychosis tends to occur following only a short period of METH consumption [12].

Several studies examined genetic factors and were found that they are associated with susceptibility to METH-induced psychosis [13]. Superoxide dismutase (SOD), as a free radical scavenger and may lessen the damage from METH-induced neurotoxicity, has been examined for its association with METH-induced psychosis [14]. A common SOD2 missense variant (rs4880: C to T, Ala 16 Val) decreases SOD2 enzyme activity [15]. The frequency of allele Ala is high in patients with METH-induced psychosis than healthy controls in Japanese and Taiwanese samples [16]. A second gene investigated for this association was dystrobrevin-binding protein 1 gene (DTNBP1) [17], which has been reported to be associated with schizophrenia and bipolar psychosis [18]. Because of the close resemblance between METH-induced psychosis and these endogenous psychoses, DTNBP1 was chosen in a Japanese study, and the frequencies of two single-nucleotide polymorphisms (SNPs), allele G in rs3213207 and allele T in rs2619538, in DTNBP1 have been found to be elevated in patients with METH-induced psychosis compared with healthy controls [18].

The association of genetic variants in SOD2 or DTNBP1 with the increased risk for METH use per se or with METH-induced psychosis remains uncertain. A more appropriate comparison group should consist of individuals who have used METH but do not have psychosis. A related issue is whether the relations of genetic variants are limited to certain forms of METH-induced psychosis, such as prolonged or transient psychosis. Furthermore, the influence of some known correlates of METH-induced psychosis, such as the initiation age of METH use or the total duration of consumption [12], on these associations has not been explored in previous studies.

We did a case–control study that contrasted patients with METH-induced psychosis with METH users without psychosis in Taiwan and found that METH is one of the most frequently consumed illegal drugs [4]. In this study, we intended to compare the distribution of genetic variants in SOD2 and DTNBP1 between METH users with and without psychosis after adjusting for other known correlates of METH-induced psychosis and to evaluate whether the relation of these genetic variants to METH-induced psychosis is influenced by the length of psychosis.


Study participants

We recruited both cases and controls from the Taipei City Psychiatric Center and Taipei Detention Center from October 2005 to April 2010. The inclusion criteria for cases were those with (a) fulfilled criteria of METH dependence and METH-induced psychosis according to the Diagnostic and Statistical Manual, Fourth Edition (DSM-IV), (b) being Han Chinese in ethnicity, and (c) age being older than 17 years. Individuals were excluded from this study if they had experienced perceptual changes only briefly following METH intoxication. The following inclusion criteria were used for the controls: (a) fulfillment of the DSM-IV criteria of METH dependence but without experiencing psychotic symptoms, (b) Han Chinese in ethnicity, and (c) age being older than 17 years. Individuals were excluded from being controls in this study it they were (a) METH users with comorbid schizophrenia and (b) opportunistic users who failed to meet the criteria for METH dependence.

Written informed consents were obtained from all the study participants after a complete description of the study. Because using METH in Taiwan is illegal, all participants were informed that their legal status would not be influenced by their participation or nonparticipation in the study. This study protocol was approved by the institutional review board of Taipei City Hospital, with the requirement of obtaining informed consents from the study participants.

During the study period, we identified 593 METH users and excluded 322 (70 individuals with METH-intoxication and perceptual change, 7 patients with schizophrenia, 3 individuals with an uncertain diagnosis, and 242 opportunistic users). The final sample had 84 cases of patients with METH-induced psychosis (22 from Taipei City Psychiatric Center and 62 from Taipei Detention Center) and 187 controls of METH users without psychosis (4 from Taipei City Psychiatric Center and 183 from Taipei Detention Center).

Clinical assessments

One member of the assessment team (consisting of 2 experienced psychologists and 3 psychiatrists) administered the specific METH session of the Chinese version of Diagnostic Interview for Genetic Studies (DIGS-C) [19]. A DSM-IV diagnosis of METH-induced psychosis was made based on the information obtained from the DIGS-C interview. Notably, we expanded the one month psychosis duration criterion in the DSM-IV to six months. Patients whose psychotic symptoms were longer than 1 month but < 6 months were categorized as having prolonged psychosis, whereas patients with psychotic symptoms being < one month were categorized as having transient psychosis. Out of the 84 cases, 21 had prolonged psychosis and 63 had transient psychosis. Those patients who suffered from psychosis for > six months after discontinuing METH use were diagnosed as having schizophrenia and therefore excluded from this study.

We collected several variables relating to METH use, in addition to diagnostic information, including the initiation age and the total and actual durations of METH use. The total duration of METH use was defined as the period from the first day the respondent used METH to the last date of such use, and the actual duration of METH use was derived by subtracting the period of abstinence from the total duration of METH use. The information on the actual duration of METH use was available for 192 (70.8%) of 271 respondents.

Single-nucleotide polymorphism genotyping

We chose five SNPs (rs 2758357, rs4880, rs2855116, rs2758330, and rs2842980) in SOD2 [16] and three SNPs (rs26119539, rs3213207, and rs2619538) in DTNBP1 [18, 20, 21] in this study. The genotyping was conducted using Taq Man probe assays (Taq Man™ SNP Genotyping Assays; Applied Biosystems, Foster City, California, USA) on a 7900 Fast Real-Time PCR System (Applied Biosystems). The Taq Man primer/probe set designed for each SNP allele was purchased from Applied Biosystems. Allelic discrimination was done automatically using SDS 2.2.2 software (Applied Biosystems).

Statistical analysis

Deviation from Hardy–Weinberg equilibrium was examined using the Chi-square test as implemented in the Procedure ALLELE in Statistical Analytic System/GENETICS, version 9.1 (SAS Institute, Cary, North Carolina, USA) among the pooled sample of cases and controls. We examined marker-trait association analyses for alleles and genotypes separately for individual SNPs and used multivariable logistic regression analyses to adjust for covariates, including age, sex, and duration of METH use.

The coefficient D' of linkage disequilibrium (LD) among the SNPs and haplotype block structure were constructed using the software Haploview version 4.1 [22], in which blocks were merged if they had a multiallelic D' > 0.8 and the cumulative frequency of common haplotypes in the merged block was > 80% [23],[24]. The frequency of individual haplotypes and their association with METH-induced psychosis were estimated using the software THESIAS [25].

To correct for multiple tests, we used the method of Nyholt [26], in which the effective number (Meff) of independent comparisons was derived from the variance of the eigenvalues of the observed marker correlation matrix. The more highly correlated the markers, the higher the variance of the eigenvalues and the lower the effect number (Meff) identified. Using this method, the Meff of independent SNPs was determined to be three for both SOD2 and DTNBP1, and therefore, the adjusted threshold of the p value for a significant association was 0.0167 (i.e., 0.05/3).

Potential interactions among the eight SNPs were examined using the software generalized multifactor dimensionality reduction (GMDR) beta version 0.7 [27]. The GMDR is a nonparametric, generalized linear method that collapses high-dimensional genetic data into a single dimension, which permits interactions to be detected in relatively small sample sizes. As an extension of the traditional multifactor dimensionality reduction method, this method allows for unbalanced case–control sample and adjustment for covariates. We used cross-validation/permutations to minimize false-positive results. An empirical p-value for the prediction accuracy was calculated from 1,000 permutations. The interactive model with the highest cross-validation consistency, followed by better predictive accuracy, and the parsimony of parameter number was chosen as the “best model.”

Because a previous study on the relations of SOD2 genetic variants to METH-induced psychosis had a group of Han-Chinese nonusers as controls [16], we compared them with this study's controls of METH users without psychosis to evaluate whether the SOD2 genetic variants were associated with METH use per se.


[Table 1] shows that the distributions in age, gender, and other substance dependence were similar between the cases and controls, whereas the initiation age of METH use was younger and the duration of METH use, regardless of total or actual, was longer for cases than controls. The total and actual durations of METH use were highly correlated in respondents with available information on both, with a Pearson's correlation coefficient of 0.62.{Table 1}

The distributions of the genotypes of the eight SNPs were all within the values expected from Hardy–Weinberg equilibrium [Table 2]. The frequencies of the alleles and genotypes for individual SNPs in the controls and cases and the two subtypes of cases are shown for SOD2 in [Table 3] and for DTNBP1 in [Table 4]. For SOD2, none of the individual SNPs were associated with METH-induced psychosis after adjustment for multiple comparisons [Table 3]. But, after stratification by the length of psychotic symptoms, both rs4880 and rs2855116 were significantly associated with prolonged METH-induced psychosis (p < 0.01 and p < 0.01, respectively). In contrast, neither allelic nor genotypic frequencies were different between the cases and controls for the three SNPs in DTNBP1, and the subtype analyses did not show any associations with the SNPs [Table 4].{Table 2}{Table 3}{Table 4}

[Table 5] shows multivariable logistic regression analyses on factors of age, gender, and total duration of METH use as covariates, to adjust for the influence of potential confounders on the relationships of these SNPs to METH-induced psychosis. Under the assumption of a codominant model of SNPs that used the most common genotype as a reference, the CC genotype of rs4880 was significantly associated with METH-induced psychosis (adjusted odds ratio = 21.31, 95% confidence interval = 2.15–211.77, p < 0.01).{Table 5}

Based on the LD matrix, one haplotype block was identified for SOD2 [Figure 1] that consisted of all five genotyped SNPs, and a three-marker haplotype was identified for DTNBP1 [Figure 2]. None of the haplotypes were significantly associated with METH-induced psychosis [Table 6].{Figure 1}{Figure 2}{Table 6}

For a variety of interactive models examined for gene–gene interactions using the GMDR method, ranging from one way to four way, the accuracy that was adjusted for study design was low, and the cross-validation consistencies were equal to 5/10 or less. None of the models reached a significant effect for the prediction accuracy [Table 7].{Table 7}

We compared the distributions of five SOD2 SNPs in this study's control group of METH users without psychosis versus the nonusers of the Han-Chinese controls in a previous study to evaluate the association of METH use per se with the SNPs examined in this study [16]. No differences in the distributions of the markers between the two groups were observed [Table 8]. Therefore, SOD2 was most likely not associated with the initiation of METH use.{Table 8}


This study compared METH users with psychosis with METH users without psychosis and clarified the association of certain genetic variants in SOD2 or DTNBP1 with METH-induced psychosis rather than METH use. The results [Table 2] showed that only one SNP in SOD2 was significantly associated with the prolonged subtype of METH-induced psychosis after correction for multiple comparison and adjustment for covariates (p < 0.01). But, no such association was found for DTNBP1 [Table 3]. We suggest that the specific association with SOD2 rather than DTNBP1 implies that the etiology of METH-induced psychosis is more likely related to oxidative stress following METH consumption.

Among the five SNPs in SOD2 examined in this study [Table 2], the frequency of the CC genotype of rs4880 was significantly increased from 0.53% in the METH users without psychosis to 5.95% in the patients with METH-induced psychosis and to 9.52% in the patients with prolonged psychosis (p < 0.05). As also shown in [Table 2], the association with the prolonged subtype of METH-induced psychosis was significant after correction for multiple comparisons (p < 0.01). The production of neurotoxicity or oxidative stress after METH use is dose dependent [28], and a dose–response relationship for the development of METH-induced psychosis has been observed [12],[29]. Further adjusting for this potential confounder using the duration of METH use as a proxy, we found that the association of the CC genotype of rs4880 with the prolonged subtype of METH-induced psychosis remained after adjustment for METH use duration.

We observed that other SNPs on SOD2 did not increase the strength of the association in their haplotypes with rs4880 [Figure 1]. This observation further indicates that the functional polymorphism of rs4880– from the C allele to the Ala variant– allows for a more efficient uptake of the SOD2 enzyme into the mitochondrial matrix [30], and that a more active enzyme [15] is involved in METH-induced psychosis. This hypothesis is further supported by the finding that the genetic variants in SOD2 were not related to METH use because their genotype distributions in METH users without psychosis in this study have been similar to the Han-Chinese normal controls of nonusers in a previous study [16].

The involvement of SOD2 with METH-induced psychosis is supported by cell line and animal studies. First, oxidative stress plays a potent rôle in METH toxicity. The ingestion of METH increases the dopamine levels within nerve terminals, which is associated with producing harmful superoxide radicals and hydrogen peroxide during quinone redox cycling [31]. METH can disturb the balance between the capacity of antioxidant enzyme systems to scavenge reactive oxygen species (ROS) and the production of RO [32]. A thiol antioxidant, N-acetylcysteine, protects against METH-induced cell death [33], and giving ascorbic acid or Vitamin E also protects against METH-induced oxidative stress in animal studies [34]. Second, METH toxicity is attenuated when SOD2 is overexpressed. For example, transgenic mice that overexpressed human SOD2, which results in higher SOD2 enzyme activity, protects against METH toxicity [35].

The frequency of the Ala variant-related C-allele of rs4880 in SOD2 was higher, rather than lower, in patients with METH-induced psychosis in this study [Table 2] and [Table 5]. The Ala variant allows for more efficient uptake of the SOD2 enzyme into the mitochondrial matrix to protect against METH-induced oxidative stress, which is contrary to expectations. But, this result agrees with that from previous studies on METH-induced psychosis [16] and other diseases (such as Alzheimer's disease, breast cancer, gastric cancer, and Parkinson's disease) [36],[37],[38]. Some explanations for this paradox may be offered. First, the SOD2 enzyme may play a dual rôle in ROS exposure. SOD2 is clearly an important scavenger of ROS, but potential harmful effects may result from its production of hydrogen peroxide (H2O2) under certain circumstances, causing secondary oxidative damage, especially in individuals with low catalase or glutathione peroxidase, which induces detoxification by converting H2O2 into water. Second, superoxide may act as a lipid peroxidation terminator. The overscavenging of free radicals increases lipid peroxidation and causes cell injury [39].

In contrast to SOD2, the genetic variants of DTNBP1 did not show any association with METH-induced psychosis in the single locus or haplotype analyses [Table 6]. In contrast, a previous study using healthy nonusers as controls reported an association between the DTNBP1 gene and METH-induced psychosis [18]. In addition to the differences in the phenotype of the controls, another possibility for this discrepancy is an ethnic difference in allele frequencies; the genotype frequency of A/G in patients with METH-induced psychosis was 2.38% in this study and 11.2% in a Japanese counterpart [18]. This hypothesis is supported by a similar lack of association of DTNBP1 with schizophrenia in the Taiwanese population [40]. The low frequency of DTNBP1 SNPs genotyped in this study might also account for the lack of its interactive effect with SOD2. Future studies should evaluate genetic variants with a higher frequency using deep sequencing to clarify the role of DTNBP1 in METH-induced psychosis in the Han-Chinese population.

Study limitations

The readers are warned against overinterpreting our study results because this study has the following three limitations:

Our study participants were recruited primarily from a detention center, and the generalizability of the results to community samples may be limitedSome unknown population stratification may have existed despite the use of cases of Han-Chinese ancestryThe sample size of our study did not have sufficient power to detect a genetic effect of moderate or modest magnitude, such as an odds ratio of < 1.5, or alleles of low frequency, such as < 15%.


This study is the first investigation using METH users without psychosis as controls in the assessment of the relationship between genetic variants of SOD2 and METH-induced psychosis. Our results suggest that only the Ala variant-related C-allele of rs4880 in SOD2 was associated with the prolonged subtype of METH-induced psychosis. This finding supports further investigation of the rôle of oxidative stress mechanisms in the development of METH-induced psychosis.


The authors thank Che-Chung Wu for his help with the genotyping work.

 Financial Support and Sponsorship

This work was supported in part by grants from the National Science Council (NSC97-2314-B-532-001-MY3), Taipei City Hospital (TCH98001-62-007), and Taipei Institute of Pathology (TIP10104A) to MCH and the “Aim for the Top University Project” of the National Taiwan University, funded by the Ministry of Education, to WJC.

 Conflicts of Interest

There are no conflicts of interest.


1Volz TJ, Fleckenstein AE, Hanson GR: Methamphetamine-induced alterations in monoamine transport: implications for neurotoxicity, neuroprotection and treatment. Addiction 2007; 102 (Suppl 1): 44-8.
2Seger D: Cocaine, metamfetamine, and MDMA abuse: the role and clinical importance of neuroadaptation. Clin Toxicol (Phila) 2010; 48: 695-708.
3United Nations Office on Drugs and Crime. World Drug Report 2018. 2. Global Overview of Drug Demand and Supply: Latest Trends and Cross-Cutting Issues. Vienna: United Nations Office on Drugs and Crime, 2018.
4Chiang SC, Chen CY, Chang YY, et al.: Prevalence of heroin and methamphetamine male users in the Northern Taiwan, 1999-2002: capture-recapture estimates. BMC Public Health 2007; 7: 292.
5Chen WJ, Wu SC, Tsay WI, et al.: Differences in prevalence, socio-behavioral correlates, and psychosocial distress between club drug and hard drug use in Taiwan: results from the 2014 national survey of substance use. Int J Drug Policy 2017; 48: 99-107.
6McKetin R, Hickey K, Devlin K, et al.: The risk of psychotic symptoms associated with recreational methamphetamine use. Drug Alcohol Rev 2010; 29: 358-63.
7Sommers I, Baskin D, Baskin-Sommers A: Methamphetamine use among young adults: health and social consequences. Addict Behav 2006; 31: 1469-76.
8McKetin R, McLaren J, Lubman DI, et al.: The prevalence of psychotic symptoms among methamphetamine users. Addiction 2006; 101: 1473-8.
9Salo R, Flower K, Kielstein A, et al.: Psychiatric comorbidity in methamphetamine dependence. Psychiatry Res 2011; 186: 356-61.
10Kittirattanapaiboon P, Mahatnirunkul S, Booncharoen H, et al.: Long-term outcomes in methamphetamine psychosis patients after first hospitalisation. Drug Alcohol Rev 2010; 29: 456-61.
11Tsuang MT, Bar JL, Harley RM, et al.: The Harvard twin study of substance abuse: what we have learned. Harv Rev Psychiatry 2001; 9: 267-79.
12Chen CK, Lin SK, Sham PC, et al.: Pre-morbid characteristics and co-morbidity of methamphetamine users with and without psychosis. Psychol Med 2003; 33: 1407-14.
13Bousman CA, Glatt SJ, Everall IP, et al.: Genetic association studies of methamphetamine use disorders: a systematic review and synthesis. Am J Med Genet B Neuropsychiatr Genet 2009; 150B: 1025-49.
14Cadet JL, Brannock C: Free radicals and the pathobiology of brain dopamine systems. Neurochem Int 1998; 32: 117-31.
15Sutton A, Khoury H, Prip-Buus C, et al.: The ala16Val genetic dimorphism modulates the import of human manganese superoxide dismutase into rat liver mitochondria. Pharmacogenetics 2003; 13: 145-57.
16Nakamura K, Chen CK, Sekine Y, et al.: Association analysis of SOD2 variants with methamphetamine psychosis in Japanese and Taiwanese populations. Hum Genet 2006; 120: 243-52.
17Harrison PJ, Weinberger DR: Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence. Mol Psychiatry 2005; 10: 40-68.
18Kishimoto M, Ujike H, Motohashi Y, et al.: The dysbindin gene (DTNBP1) is associated with methamphetamine psychosis. Biol Psychiatry 2008; 63: 191-6.
19Chen WJ, Liu SK, Chang CJ, et al.: Sustained attention deficit and schizotypal personality features in nonpsychotic relatives of schizophrenic patients. Am J Psychiatry 1998; 155: 1214-20.
20Raybould R, Green EK, MacGregor S, et al.: Bipolar disorder and polymorphisms in the dysbindin gene (DTNBP1). Biol Psychiatry 2005; 57: 696-701.
21Williams NM, Preece A, Morris DW, et al.: Identification in 2 independent samples of a novel schizophrenia risk haplotype of the dystrobrevin binding protein gene (DTNBP1). Arch Gen Psychiatry 2004; 61: 336-44.
22Barrett JC, Fry B, Maller J, et al.: Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005; 21: 263-5.
23Gabriel SB, Schaffner SF, Nguyen H, et al.: The structure of haplotype blocks in the human genome. Science 2002; 296: 2225-9.
24Chen YC, Giovannucci E, Lazarus R, et al.: Sequence variants of toll-like receptor 4 and susceptibility to prostate cancer. Cancer Res 2005; 65: 11771-8.
25Tregouet DA, Garelle V. A new JAVA interface implementation of THESIAS: testing haplotype effects in association studies. Bioinformatics 2007; 23:1038-9.
26Nyholt DR: A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. Am J Hum Genet 2004; 74: 765-9.
27Lou XY, Chen GB, Yan L, et al.: A generalized combinatorial approach for detecting gene-by-gene and gene-by-environment interactions with application to nicotine dependence. Am J Hum Genet 2007; 80: 1125-37.
28Choi HJ, Yoo TM, Chung SY, et al.: Methamphetamine-induced apoptosis in a CNS-derived catecholaminergic cell line. Mol Cells 2002; 13: 221-7.
29McKetin R, Lubman DI, Baker AL, et al.: Dose-related psychotic symptoms in chronic methamphetamine users: evidence from a prospective longitudinal study. JAMA Psychiatry 2013; 70: 319-24.
30Wispé JR, Clark JC, Burhans MS, et al.: Synthesis and processing of the precursor for human mangano-superoxide dismutase. Biochim Biophys Acta 1989; 994: 30-6.
31Wu CW, Ping YH, Yen JC, et al.: Enhanced oxidative stress and aberrant mitochondrial biogenesis in human neuroblastoma SH-SY5Y cells during methamphetamine induced apoptosis. Toxicol Appl Pharmacol 2007; 220: 243-51.
32Chen HM, Lee YC, Huang CL, et al.: Methamphetamine downregulates peroxiredoxins in rat pheochromocytoma cells. Biochem Biophys Res Commun 2007; 354: 96-101.
33Chandramani Shivalingappa P, Jin H, Anantharam V, et al.: N-acetyl cysteine protects against methamphetamine-induced dopaminergic neurodegeneration via modulation of redox status and autophagy in dopaminergic cells. Parkinsons Dis 2012; 2012: 424285.
34Pubill D, Chipana C, Camins A, et al.: Free radical production induced by methamphetamine in rat striatal synaptosomes. Toxicol Appl Pharmacol 2005; 204: 57-68.
35Cadet JL, Sheng P, Ali S, et al.: Attenuation of methamphetamine-induced neurotoxicity in copper/zinc superoxide dismutase transgenic mice. J Neurochem 1994; 62: 380-3.
36Wiener HW, Perry RT, Chen Z, et al.: A polymorphism in SOD2 is associated with development of Alzheimer's disease. Genes Brain Behav 2007; 6: 770-5.
37Yao S, Barlow WE, Albain KS, et al.: Manganese superoxide dismutase polymorphism, treatment-related toxicity and disease-free survival in SWOG 8897 clinical trial for breast cancer. Breast Cancer Res Treat 2010; 124: 433-9.
38Xu Z, Zhu H, Luk JM, et al.: Clinical significance of SOD2 and GSTP1 gene polymorphisms in Chinese patients with gastric cancer. Cancer 2012; 118: 5489-96.
39Nelson SK, Bose SK, McCord JM: The toxicity of high-dose superoxide dismutase suggests that superoxide can both initiate and terminate lipid peroxidation in the reperfused heart. Free Radic Biol Med 1994; 16: 195-200.
40Liu CM, Liu YL, Fann CS, et al.: No association evidence between schizophrenia and dystrobrevin-binding protein 1 (DTNBP1) in Taiwanese families. Schizophr Res 2007; 93: 391-8.