Analysis of association of clinical correlates and 5-HTTLPR polymorphism with suicidal behavior among Chinese methamphetamine abusers
Chih-Ken Chen, MD, PhD, Department of Psychiatry, Chang Gung Memorial Hospital, 222 Mai-Chin Road, Keelung, Taiwan. Email: firstname.lastname@example.org
Abstract Substance use disorders are familial, and genetic factors explain a substantial degree of their familial aggregation. Methamphetamine (MAP) abusers are commonly noted as having psychosis, depression and suicidal behavior. The goals of the present study were (i) to investigate relations of clinical correlates, such as gender, drug use behavior, psychiatric comorbidity and psychiatry family history, with suicidal behavior among Chinese MAP abusers; and (ii) to investigate whether there is an association between a polymorphism in the promotor region of the serotonin transporter gene (5-HTTLPR) and suicidal behavior among Chinese MAP abusers. A total of 439 MAP abusers from a hospital and detention center in Taipei were interviewed with the Diagnostic Interview for Genetic Study and the Family Interview for Genetic Study. The 5-HTTLPR polymorphism was compared between 94 MAP abusers with suicide attempts and 294 MAP abusers without suicide attempts, for whom DNA data were available. The results of the present study indicate that among MAP abusers in Taiwan, suicide attempts were significantly related to female gender, history of MAP-induced psychotic disorder, history of MAP-induced depressive disorder, and family history of psychotic disorders. Among suicide attempters, the attempters with moderate to severe lethality used higher MAP doses than those with minimal to mild lethality. In the present sample the triallelic 5-HTTLPR polymorphism (S, LG, LA) was not associated with MAP-induced depressive disorder, MAP-induced psychotic disorder or suicidal behavior, but studies with larger sample sizes are warranted before excluding the role of the 5-HTTLPR polymorphisms in suicidal behavior among MAP abusers.
The United Nations International Drug Control Program estimated that some 34 million people consume amphetamine-type stimulants worldwide, more than heroin or cocaine.1 Substance abuse is a complex behavior that results from a complex interaction of genetic and environmental risk factors.2,3 There is evidence that substance use disorders are familial and that genetic factors explain a significant degree of their familial aggregation.4–7 Tsuang et al. reported that genetic factors account for 44% of the variance in the individual's risk of developing a stimulant abuse disorder.7 Methamphetamine (MAP) is in general more potent than other amphetamine-type stimulants, and thus cause more serious health problems.1 In addition to addiction, long-term use of MAP can cause psychosis and depression. Whether an individual develops psychosis after MAP abuse may be determined by various factors,8 such as drug-use patterns, premorbid and comorbid mental health9 and familial predisposition.10 Depression is also a common complaint of amphetamine addicts.11 Despite the high prevalence and socioeconomic impact of depressive disorders, their etiology is not totally clear. Heritability estimates demonstrate up to a 50% genetic component based on family aggregation and contrasting monozygotic and dizygotic twin studies. Gene–environment interaction has been recognized for a long time in the pathophysiology of depression. It is well established that suicidal behavior is commonly seen among patients with psychotic or depressive symptoms. We have found that 19% of incarcerated MAP abusers in Taiwan have a history of suicide attempts.12 It is interesting to study whether suicidal behavior among MAP abusers is also related to drug use patterns, comorbid mental health and familial predisposition.
When MAP is administered in repeated high doses, neurochemical deficits in both the dopaminergic and serotonergic systems of selected areas of the brain are observed.13 Neurotoxicological studies have also established that MAP treatments evoke degeneration of dopaminergic and serotonergic fibers in the brains of many animal species.14,15 The serotonergic system is damaged by MAP administration in various brain areas.13 These deficits include long-term deficits in tryptophan hydroxylase activity as well as decreases in serotonin concentrations, transporter uptake and ligand bindings sites.16,17 Among the various components of the serotonergic system, serotonin transporter (5-HTT) is critically important in regulating the re-uptake of serotonin into the presynaptic neuron, serving as a target site for certain antidepressants.18 Allelic variations in the 5′ flanking transcriptional region of the 5-HTT gene (5-HTTLPR, which controls 5-HTT expression and function) have been associated with depression.195-HTTLPR has been also reported to be associated with psychosis,20 while some researchers failed to replicate this association.21 S-carriers of the 5-HTTLPR polymorphism have been reported as having more suicide attempts over time, using violent means of suicide attempts more frequently than non-S-carriers.22 Most studies on association of 5-HTTLPR and depression or suicidal behavior genotyped the biallelic polymorphism. However, after the initial reports by Nakamura et al.23 subsequent studies have confirmed the functional significance of novel allelic variants at this site.24,25 The LG and S alleles have comparable levels of serotonin transporter expression and depression, and both are lower than that of the LA allele.25,26 It is therefore suggested that 5-HTTLPR should be regarded as a triallelic functional polymorphism (i.e. S, LG and LA alleles), and that the S allele and the LG can be classified as the same allele group.25,26
The goals of the present study were (i) to investigate relations of clinical correlates, such as gender, drug use behavior, psychiatric comorbidity and psychiatry family history, with suicidal behavior among Chinese MAP abusers; and (ii) to investigate whether there is an association between the 5-HTTLPR triallelic polymorphism and suicidal behavior among Chinese MAP abusers.
Subjects and clinical measurements
The methods for case identification have been described in detail elsewhere.9 In brief, a total of 439 MAP users (277 male, 162 female) were recruited from the Taipei City Psychiatric Center (TCPC) and the Taipei Detaining Center (TDC). The eligibility criteria were (i) age >17 years, (ii) ethnic Chinese origin, and (iii) use of MAP. The participants from the TDC were recruited strictly on a voluntary basis. All the eligible subjects were informed that their legal status would not be influenced by whether or not they participated in the present study. Informed consent was obtained from all participants. The present study was approved by the Ethical Committee, Taipei City Psychiatric Center, Taipei, and the Institutional Review Boards, Chang Gung Memorial Hospital, Taiwan.
The subjects received a diagnostic interview with the Chinese versions27 of the Diagnostic Interview for Genetic Study28 and the Family Interview for Genetic Study.28 Based on the informant's responses to the general screening questions, five symptom checklists (depression, mania, alcohol and other drug abuse, psychosis, paranoid/schizoid/schizotypal personality disorder) were completed for each first-degree relative. Variables of MAP use behavior included routes, doses, frequency, duration of MAP use, age of first MAP use, and other drug abuse. Frequency and dose of MAP were approximate estimations of the average days per week and the average amount in grams of MAP used per week during the periods of MAP abuse (which meant that the periods of abstinence were not included). The subjects were asked about the amount but it was difficult for some of the subjects to describe the amount of their MAP use. In that case, they were asked about the average cost on their MAP use per week. The amount was then calculated approximately assuming 1 g cost around $US50. Psychiatric diagnoses were made according to the DSM-IV criteria.29 The DSM-IV suggests that the psychotic symptoms are better accounted for by a primary psychotic disorder if these symptoms persist for a substantial period of time (approx. 1 month) after the end of substance intoxication or acute substance withdrawal. However, in the present study the MAP users with prolonged psychosis after MAP use were still considered to have MAP-induced psychotic disorder. The diagnosis of depressive disorder was made only when at least one major depressive episode occurred before first MAP use, and was not caused by other organic factors. An episode that met the criteria for major depression but occurred exclusively during MAP use or after MAP withdrawal was considered as MAP-induced depressive disorder. For the diagnosis of antisocial personality disorder to be given, the individual must be at least 18 years of age and must have a history of some symptoms of conduct disorder before age 15 years. Suicidal behavior was assessed with the Suicide section of the Diagnostic Interview for Genetic Study. Using this section of the questionnaire we evaluated the lethality based on the lifetime of suicide attempts, scoring 1–6, with a score of ≥4 as the cut-off point.
Genotyping of 5-HTTLRP polymorphisms
Genomic DNA was extracted from peripheral blood using standard methods. Genotyping was accomplished in two stages. Stage 1 distinguished S from L alleles. Stage 2 distinguished LA from LG allele. For stage 1, the sense and antisense primers used for 5-HTTLPR were 5′-GGC GTT GCC GCT CTG AAT GC-3′ and 5′-GAG GGA CTG AGC TGG ACA ACC AC-3′. Polymerase chain reaction (PCR) amplification were performed in a total volume of 20 μL containing 2 μL 10 × PCR buffer, 1 μL dimethylsulfoxide, 2.5 nmol/L dNTP, 3.2 μL 50% glycerol, 5 μmol/L of each primer and 1.5 U Taq DNA polymerase. PCR amplification consisted of 95°C for 5 min, followed by 40 cycles of 95°C for 30 s, 62°C for 30 s and 72°C for 30 s before a final extension step at 72°C for 7 min. PCR products were separated by 3% agarose gel.
Genotypes of stage 2 were determined using DNA sequencing. For stage 2 the primers for sequencing were the same as those used in stage 1 genotyping. Sequencing was performed using the BigDye Terminator Cycle Sequencing Kit v3.1 (Applied Biosystems, Foster City, USA) according to the manufacturer's protocol. The 10-μL sequencing reaction mixture contained 2 μL 5X sequencing buffer, 0.5 μL BigDye v3.1, 4.5 μL dH2O, 1 μL 5 μmol/L primers, and 2 μL purified 5-HTTLPR PCR amplicon. Cycle conditions for sequencing were: 1 min at 96°C, followed by 25 cycles of 10 s at 96°C, 5 s at 50°C and 4 min at 60°C. The sequences were analyzed in an ABI 3730 automated sequencer (Applied Biosystems).
Data were analyzed using SPSS version 12.0 (SPSS, Chicago, IL, USA). Statistical analysis was carried out with independent t-tests for metric variables and χ2 tests for categorical variables. Risk factors for suicidal attempts or severe lethality during univariate analyses were further analyzed using logistic regression analysis. Hardy–Weinberg equilibrium (HWE) was tested for the 5-HTTLPR polymorphisms using χ2 test. For analyses of allele frequencies, the S allele and LG allele were classified as the same group. Extra-long alleles were classified as LA. For analyses of genotype frequencies, the genotypes were classified into three groups: LA/LA; S/LA LA/LG; and S/S, S/LG, LG/LG. Pearson χ2 tests were performed, comparing overall allele and genotype frequencies between suicide attempters and non-attempters. All tests were two-sided, and the level of significance was set at P < 0.05.
A total of 439 unrelated MAP abusers were recruited for analysis of suicidal behavior and other clinical correlates, and 388 of them had 5-HTTLPR genotyping. Of the 439 MAP abusers, 108 had history of suicide attempt. As shown in Table 1, the women had significantly higher rates of suicide attempts than the men (38.3% vs 16.6%, P < 0.001). The age of MAP abusers was 18–60 years (27.3 ± 7.0). Only two of the subjects used MAP via injection, while all the rest used MAP via inhalation. The age of first MAP use ranged from 12 to 48 years, with a mean of 21.7 ± 6.9 years. The average weekly consumption of MAP ranged from 0.1 g to 21 g, with a mean of 2.1 ± 2.9 g. The mean duration of MAP abuse was 30.7 ± 28.1 months. The frequencies of the subjects having misused opiates, cocaine, marijuana, organic solvents, sedatives were 15.3%, 2.3%, 9.3%, 7.4% and 23%, respectively. The suicide attempters used MAP significantly more frequently than non-attempters. Among suicide attempters, those with moderate to severe lethality were older at first MAP use, and used higher MAP doses than those with minimal to mild lethality.
Table 1. Suicide and drug use in methamphetamine abusers
|Gender|| || ||<0.001|| || ||0.483|
| Male (%)||46 (16.6)||231 (83.4)|| ||23 (50.0)||23 (50.0)|| |
| Female (%)||62 (38.3)||100 (61.7)|| ||25 (43.1)||33 (56.9)|| |
|Age (years)||27.5 ± 6.7|| 27.2 ± 7.1||0.698||25.7 ± 5.9||28.8 ± 6.8||0.014|
|Education (years)|| 9.3 ± 2.1|| 9.3 ± 2.3||0.870|| 9.4 ± 2.1|| 9.3 ± 2.0||0.679|
|Drug use behavior|
| Age of first use (years)||21.1 ± 6.7|| 21.8 ± 7.0||0.481||19.6 ± 6.2||22.6 ± 6.6||0.026|
| Dose (g/week)|| 2.5 ± 3.0|| 2.0 ± 2.9||0.173|| 1.6 ± 1.5|| 3.1 ± 3.5||0.023|
| Frequency (times/week)|| 4.7 ± 2.2|| 4.1 ± 2.1||0.021|| 4.3 ± 2.1|| 4.9 ± 2.2||0.185|
| Duration (months)|| 34.6 ± 30.4|| 29.4 ± 27.3||0.104|| 38.0 ± 31.4|| 31.8 ± 29.9||0.324|
As shown in Table 2, MAP abusers with suicide attempts had significantly higher prevalence of depressive disorders, MAP-induced depressive disorders, MAP-induced psychotic disorders, alcohol use disorders, and antisocial personality than those without suicide attempt. Suicide attempters were significantly more likely to have a family history of psychotic disorders, depressive disorders among their first-degree relatives than non-attempters. Between these two groups, there were no significant differences in the familial rates of drug use disorders or alcohol use disorders. Among suicide attempters, those with moderate to severe lethality were more likely to have depressive disorders than those with minimal to mild lethality, while suicide lethality was not significantly associated with other psychiatric disorders or psychiatric family history.
Table 2. Psychiatric comorbidity, family history and suicide behavior in methamphetamine abusers
| Depressive disorder||17 (15.7)||18 (5.4)||0.001||3 (6.3)||13 (23.2)||0.017|
| MAP depression||31 (28.7)||31 (9.4)||<0.001||17 (35.4)||12 (21.4)||0.113|
| MAP psychosis||66 (61.1)||103 (31.1)||<0.001||28 (58.3)||34 (60.7)||0.805|
| Alcohol use disorder||36 (33.3)||76 (23.0)||0.032||18 (37.5)||17 (30.4)||0.442|
| Antisocial personality†||30 (27.8)||60 (18.1)||0.031||14 (29.2)||15 (26.8)||0.787|
|Psychiatric family history|
| Drug use disorder||19 (17.6)||35 (10.6)||0.054||8 (16.7)||9 (16.1)||0.935|
| Alcohol use disorder||57 (17.2)||16 (14.8)||0.560||6 (12.5)||8 (14.3)||0.790|
| Depressive disorder||7 (6.5)||7 (2.1)||0.025||4 (8.3)||3 (5.3)||0.546|
| Psychotic disorder||11 (10.2)||7 (2.1)||<0.001||5 (10.4)||5 (8.9)||0.797|
After adjusting on logistic regression (variables entered: gender, frequency of MAP use, five associated psychiatric disorders as shown in Table 1, family history of depressive disorders, and family history of psychotic disorders), the risk factors that determined suicide attempts were female gender (P < 0.001), history of MAP-induced psychotic disorders (P < 0.001), history of MAP-induced depressive disorders (P = 0.013), and family history of psychotic disorders (P = 0.036). Among suicide attempters, after adjusting on logistic regression (variables entered: age, age of first MAP use, MAP dose, history of depressive disorders), the attempters with moderate to severe lethality used higher MAP doses than those with minimal to mild lethality (P = 0.018).
The genotypic distribution of 5-HTTLPR polymorphism was in Hardy–Weinberg equilibrium in both suicide attempters (χ2 = 0.6, P = 0.423) and non-attempters (χ2 = 3.1, P = 0.076), or in MAP abusers as a whole (χ2 = 3.7, P = 0.056). Of the 388 MAP abusers with 5-HTTLPR genotyping, 48 had history of MAP-induced depressive disorders and 141 had MAP-induced psychotic disorders. As shown in Table 3, there was no significant association between 5-HTTLPR polymorphism with MAP-induced depressive disorders or with MAP-induced psychotic disorder. Among the MAP abusers, as shown in Table 4, the frequency of 5-HTTLPR S and LG allele was 69.7% for the suicide attempters and 72.6% for non-attempters. Among the MAP abusers, there was no significant difference in either genotype frequencies or allele frequencies of 5-HTTLPR polymorphism between the suicide attempters and non-attempters. The 5-HTTLPR polymorphism was not significantly associated with the lethality of suicide attempts. There was no significant association between genotypes of 5-HTTLPR polymorphism and suicidal attempts or lethality after controlling for gender on logistic regression.
Table 3. Genotype and allele frequencies for 5-HTTLPR by depression and psychosis among methamphetamine abusers
| LA/LA||2 (4.2)||21 (6.2)||0.786||6 (4.0)||17 (7.1)||0.442|
| S/LA, LA/LG||23 (47.9)||149 (43.8)|| ||68 (45.3)||104 (43.7)|| |
| S/S, S/LG, LG/LG||23 (47.9)||170 (50.0)|| ||76 (50.7)||117 (49.2)|| |
| LA||27 (28.1)||191 (28.1)||0.994||80 (28.6)||138 (29.0)||0.902|
| S, LG||69 (71.9)||489 (71.9)|| ||200 (71.4)||338 (71.0)|| |
Table 4. Genotype and allele frequencies for 5-HTTLPR by suicidal behavior among methamphetamine abusers
| LA/LA||7 (7.4)||16 (5.4)||0.686||4 (9.3)||3 (6.5)||0.802|
| S/LA, LA/LG||43 (45.7)||129 (43.9)|| ||21 (48.8)||21 (45.7)|| |
| S/S, S/LG, LG/LG||44 (46.8)||149 (50.7)|| ||18 (41.9)||22 (47.8)|| |
| LA||57 (30.3)||161 (27.4)||0.435||29 (33.7)||27 (38.6)||0.530|
| S, LG||131 (69.7)||427 (72.6)|| ||57 (66.3)||43 (61.4)|| |
Given that the frequency of S and LG allele among suicide attempters (n = 94) and non-attempters (n = 294) was approximately 70% and 73%, respectively, and the significance level was 0.05, then the power of allelic test was 0.13 and the power of genotypic test was 0.10 for the present study.
The risk factors that determined suicide attempts was female gender, history of MAP-induced psychotic disorders, history of MAP-induced depressive disorders, and family history of psychotic disorders. Among suicide attempters, the attempters with moderate to severe lethality used higher MAP doses than those with minimal to mild lethality. We did not find the triallelic 5-HTTLPR polymorphism to be associated with suicide behavior of MAP abusers in the present Taiwanese sample, nor was the 5-HTTLPR polymorphism associated with MAP-induced depressive disorders or MAP-induced psychotic disorders.
After adjusting on logistic regression, female MAP abusers were more likely to have suicide attempts than their male counterparts. In critical review of the area of gender differences in comorbidity with substance-use disorders, physicians should keep in mind gender differences in psychiatric disorders in the general population.30 The reported difference in rates of suicide behavior by gender among the present subjects was, in large part, consistent with the results of the general population survey in Taiwan.31 Among drug abusers, female subjects were more widely exposed to unfavorable social factors, including less education, higher divorce rates, higher rates of unemployment and earlier onset of illicit drug use,32–34 and had substantially higher incidence of suicidal behavior than male subjects.12,34 The fact that as many as 38.3% of female MAP abusers in the present study reported suicidal behavior warrants special attention to suicidal risk in the treatment of female MAP abusers.
Among the present subjects suicide attempters were more likely to have history of MAP-induced psychotic disorders and MAP-induced depressive disorders than non-attempters. MAP abuse has been reported in several studies to result in psychotic symptoms that sometimes last more than several months or recur after MAP reuse at low doses or a wide variety of stressors.35,36 The fact that depression is a common complaint of MAP abusers may result from primary affective illness, situational aspects of the addictive lifestyle, or drug withdrawal.11 We have previously proposed that genetic liability to psychotic disorders such as schizophrenia accounts, at least in part, for individual differences in risk for developing psychosis after MAP, and that MAP abuse interacts with the genetic liability to produce psychotic symptoms in genetically predisposed individuals.10 It has been well accepted that psychosis and depression increase risks for suicidal behavior. Therefore, it is not surprising to note that suicide attempters were more likely to have a family history of psychotic disorders than non-attempters in the present study.
Many studies have examined the association between the 5-HTTLPR polymorphism and suicide but have yielded inconsistent results. Most of the results of these studies show no statistically significant evidence of linkage disequilibium,37 while some studies demonstrate an association between S-carrier genotypes allele of the 5-HTTLPR and violent suicide attempts in patients who have schizophrenia, mood disorders, alcohol use disorders, or suicide completion.38–44 The meta-analyses of Lin and Tsai found that there was no association between 5-HTTLPR polymorphism and suicidal behavior, although there exists the association of S allele with suicidal behavior in some particular psychiatric populations, and with violent suicide.37 A very recent study by Zalsman et al. found that there was no significant association between the triallelic 5-HTTLPR polymorphism and suicide attempts or lethality of suicide attempts.45
There are several possible explanations for the observed lack of association between the triallelic 5-HTTLPR polymorphism and suicide behavior in the present sample. First, it is possible that the association of the 5-HTTLPR polymorphism with suicide behavior does not exist among MAP abusers. So far, there has been no study reporting the association of 5-HTTLPR polymorphism with suicidal behavior in MAP abusers. Second, it is entirely possible that increased risk to suicide behavior with the 5-HTTLPR polymorphism may be absent in the Taiwanese population, although it might exist in other populations. It is known that the allelic distributions of genes vary among different ethnic groups. In a meta-analysis of 17 previously published studies, Lin and Tsai concluded that the frequencies of the S allele of 5-HTTLPR in normal subjects were 0.43 and 0.74 for Caucasian and Asian populations, respectively.37 This difference might result in different strengths in the association between this polymorphism and the suicide phenotype in these two ethnic groups. Third, it is possible that the association of the 5-HTTLPR polymorphism with suicide behavior is too small to be detected in the present study owing to insufficient statistical power. The sample size was small when comparing suicide attempters (n = 94) and non-attempters (n = 294) among MAP abusers with DNA data. In the present study the power of allelic test was 0.13 and the power of genotypic test was 0.1 for detecting an association between 5-HTTLPR polymorphism with suicide behavior. The present study therefore did not have enough power to detect small to moderate effects of these genes on suicidal behavior among MAP abusers. We have genotyped just a single genetic variant from the serotonin transporter gene. This is not sufficient statistical power to claim a lack of association with a phenotype. In order to adequately make this association, screening the entire gene or identifying a sufficient number of variants from the SNP databases are needed. In the present sample the triallelic 5-HTTLPR polymorphism was not associated with MAP-induced depressive disorders or MAP-induced psychotic disorders. Again, the present study did not have enough power to detect small to moderate effects of these genes on MAP-induced depressive disorders or MAP-induced psychotic disorders among MAP abusers.
Some other study limitations might affect interpretation of the present results. The subjects were recruited from a hospital and a detention center, and did not represent MAP users in the community. Therefore we should be cautious in applying our conclusions to MAP users in the general population. The major weakness of the present study was the retrospective nature of the data. The factors we studied may have been influenced by recall bias. Reporting bias and observer bias could not be ruled out, but were reduced by applying highly structured instruments and using trained researchers.
This study was supported by research grants (NSC91-2314-B-109-002, NSC94-2320-B-182A-018, NSC95-2320-B-182A-017) from the National Science Council, Taiwan.