Association of the SLC6A4, SLC6A3, COMT and BDNF Gene Polymorphisms with Suicidal Behavior: A Case Control Study

Background Suicide causes about 1 million deaths per year. Familial transmission of suicide, as well as the involvement of serotonin and dopamine systems in suicidal behavior, was conrmed by previous studies. We investigated an effect of 11 polymorphisms of 9 genes related to dopamine and serotonin transmission (SLC6A4, HTR1A, HTR2A, HTR1B, SLC6A3, DRD4, DRD2, COMT, and BDNF) on the risk of suicide. Methods The study was performed on 100 psychiatric clinic patients with repeated suicide attempts. Genomic DNA was obtained from venous blood. Genotyping was performed using locus-specic PCR. Statistical analysis was carried out using a Pearson Chi-squared test, ORs, with the corresponding 95% CIs and one-way ANOVA.


Background
According to the World Health Organization, more than 800 000 people die from suicide each year [1].
Many family, twin, and adoption studies provide evidence for familial transmission of suicide and suicidal behavior [2][3][4][5][6]. The contribution of genetic risk factors was con rmed even after controlling for hereditary mental disorders [7]. The aggregation of suicide in families also cannot be fully explained by a similar environment [8]. Molecular studies have shown that suicide attempts could be associated with altered serotonin (HT) and dopamine (DA) transmission [9][10][11][12]. Given the above data, the search for an association between suicidal behavior and genes encoding important pathways of serotoninergic and dopaminergic transmission (HT and DA transporters, receptors, ferments for HT and DA biosynthesis) is relevant.
The promoter region of the serotonin transporter gene (SLC6A4/5HTT) contains the variable number of tandem repeats (VNTR-polymorphism). The short (S) allele with 14 repeats is associated with lower expression activity than the long (L) allele with 16 repeats [13,14]. Later, a single nucleotide polymorphism (SNP) A → G (rs25531) was detected within the sixth repeat of the S-and L-alleles. It was shown that the expression level of the L G allele is lower than of the L A allele. A meta-analysis of 39 studies of the SLC6A4 VNTR-polymorphism demonstrated a signi cant association between suicidal behavior and the S-allele. The in uence of the rs25531 polymorphism was not considered in this study [15].
Polymorphism C1019G (rs6295) exists in the promoter region of the HTR1A gene. It might enhance or decrease gene expression depending on the location on the presynaptic or postsynaptic membrane [16,17]. Post-mortem study has discovered that the HTR1B gene polymorphism G861C (rs6296) affects gene expression activity and as a result changes the density of the receptors [18]. In the case of HTR1A gene polymorphism A1438G (rs6311), it was shown that the A-allele increases the promoter activity [19].
Association studies for these genes were also performed. Meta-analyses of these investigations detected a signi cant association between HTR2A (rs6311) and suicidal behavior [20] but failed to nd any associations for HTR1A (rs6295) and HTR1B (rs6296) polymorphisms [21,22]. Several studies have con rmed the effect of the length of this locus on gene expression, although contradictory results have been obtained [23][24][25][26]. It was shown that the 9R-allele may result in a high risk of depression and angry-impulsive personality traits [27,28]. The dopamine receptor gene (DRD4) contains a 48 bp VNTR-polymorphism in exon 3 and 120 bp In/Del polymorphism in the promoter region.
The 7-repeat variant of the DRD4 VNTR-polymorphism was found to be associated with such personal traits as impulsivity, aggression, depression and novelty-seeking behavior [29,30]. DRD4 In/Del polymorphism might affect gene expression activity-the long allele is associated with lower expression level than the short allele [31]. The density of the DRD2 receptors in the striatum depends on the allelic variant of the rs1800497 (C/T) polymorphism. It was shown that T-allele carriers have a lower density of DRD2 receptors [32]. The DRD2 rs1800497 polymorphism is associated with impulsivity and suggested to be related to suicidal attempts [33,34]. The enzyme catechol-O-methyltransferase (COMT) plays an important role in dopamine metabolism. SNP rs4680 (G/A or Val/Met) in the COMT gene leads to reduced enzyme activity [35]. A meta-analysis of six studies has demonstrated that the COMT rs4680 polymorphism has a modestly signi cant association with suicidal behavior [22].
Another gene associated with mental disorders is the brain-derived neurotrophic factor (BDNF) gene [36]. BDNF and serotonin are shown to regulate each other positively: HT increases the BDNF expression and BDNF stimulates HT neurons' growth and survival [37]. The BDNF gene contains a common polymorphism rs6264 which leads to the replacement of methionine with valine (Val66Met). Met-allele is associated with decreased protein activity [38] and possibly with suicide attempts [39].
This study aims to estimate the contribution of 11 polymorphisms in the genes SLC6A4 (5HTT), HTR1A, HTR2A, HTR1B, SLC6A3 (DAT1), DRD4, DRD2, COMT and BDNF to suicidal behavior and severity of symptoms of depression and anxiety in patients. As described above, all polymorphisms under study affect gene expression, protein product activity or were shown to be associated with personality traits.

DNA isolation and genotyping
Genomic DNA was obtained from 250 μL of EDTA-anticoagulated venous blood using innuPREP Blood DNA Mini Kit (Analytik Jena AG, Germany), according to the manufacturer's recommendations.

Statistical analysis
Hardy-Weinberg equilibrium calculator software (https://wpcalc.com/en/equilibrium-hardy-weinberg/) was used to calculate the correspondence of the genotype distribution in the population sample to the Hardy-Weinberg equilibrium. Allele and genotype frequencies between cases and control groups were compared using a Pearson Chi-squared test. The strength of associations between allelic variants of studied polymorphic loci and suicidal behavior was estimated using odds ratios (ORs), with the corresponding 95% con dence intervals (95% CIs). One-way ANOVA was performed to determine the effect of studied polymorphisms on symptoms of depression, situational anxiety, and personal anxiety. The normal quantile-quantile (QQ) plot and Shapiro-Wilk test were used to test whether our data were normally distributed. The analysis of frequencies of genotypes and alleles, Chi-squared test and ANOVA were conducted using Statistica (data analysis software system, version 7, http://www.statsoft.com). All tests were conducted at a level of signi cance p < 0.05.

Results
In the control samples, all polymorphisms were in the Hardy-Weinberg equilibrium. The results of the polymorphic loci genotyping in cases and control samples are shown in Table 2.
The strongest association was found for VNTR-polymorphism of the SLC6A3 gene. Given the data suggesting the effect of length of this locus on gene expression and high activity of the 10R allele compared with the 9R allele [23][24][25][26], we classi ed cases and controls as carriers of the long (≥10) and short (<10) alleles. The short allele (S) was more common in cases than it was in the control group (p=0.009). SS and LS genotype frequencies were higher and LL genotype frequency was lower in cases than in controls (p=0.019).
Another gene of the dopaminergic system associated with suicide attempts was the COMT gene. In the group of cases, the G-allele (Val) frequency of COMT rs4680 was signi cantly higher than with the control group (p = 0.028). However, the genotype distributions of COMT rs4680 were not found to be signi cantly different between cases and controls, AA genotype frequency was decreased and GA and GG frequencies were increased in the group of cases compared with the control group (17%, 53%, 30%, respectively, in cases and 29.4%, 47.9%, 22.7%, respectively, in controls; p = 0.064).
We also found a statistically signi cant association between the T-allele of BDNF rs6264 and suicidal behavior. In the group of patients, the T-allele was more common than with controls (p = 0.029). The difference in genotype distribution was not statistically signi cant. However, CC genotype tends to be less common in the group of cases, while frequencies of CT and TT genotypes are higher in cases than with the control group (61%, 32%, 7%, respectively, in cases and 70.7%, 26.6%, 2.7%, respectively, in controls; p = 0.076).
We analysed two loci of the SLC6A4 gene: rs25531 and VNTR 44 bp polymorphism. We found that the rs25531 polymorphism of SLC6A4 showed some statistically signi cant associations. The A-allele frequency of rs25531 was higher in cases than it was in controls (p=0.047). VNTR-polymorphism of SLC6A4 was not found to be associated with suicidal behavior.
Because of the complexity of SLC6A4 organization, we additionally analysed the distribution frequencies of its allelic variants according to their functional characteristics. To verify whether alleles with low or high expression activity are associated with suicidal behavior, we grouped alleles and genotypes according to their expression levels: high (L A ) and low (S and L G ) alleles with high and low expression activity, respectively, and high/high (L A /L A ), high/low (L A /S A , L A /L G ) and low/low (S A /S A , S A /L G , L G /L G ) genotypes.
We found that the distribution of SLC6A4 gene genotypes and alleles showed that alleles with high expression tend to be increased in patients with suicidal behavior compared with the control group, but this association was not statistically signi cant (p = 0.098).
The alleles of the dopamine receptor DRD4 exon 3 were grouped into long (≥7) and short (<7) allelic variants, associations between this locus and suicide were not found. No signi cant associations between suicidal behavior and DRD2 rs1800497, DRD4 In/Del, 5HTR1A rs6295, 5HTR2A rs6311 and 5HTR1B rs6296 alleles and genotypes were discovered. All results of genotyping of cases and controls are shown in Additional les 1 and 2, respectively.
To determine the effect of studied polymorphisms on the patient's clinical characteristics we conducted the one-way ANOVA. This analysis showed the lack of association between loci which were signi cant in the Chi-squared test and severity of the symptoms of depression, situational anxiety, and personal anxiety in patients with suicidal behavior. However, we observed a signi cant effect of SLC6A4 VNTR 44 bp on the situational anxiety and HTR1B loci on the personal anxiety (p = 0,0358 and p = 0,0132, respectively ). The most signi cant results of one-way ANOVA are shown in Table 3 and results for all loci are shown in Additional le 3.

Discussion
Genes related to neurotransmission and neurotrophic function have been widely studied in various mood and behavior disorders. To con rm possible associations studied previously and nd out new associations, we investigated SLC6A3 (40-45 bp. VNTR), DRD2 (rs1800497), DRD4 (In/Del 120 bp. and 48 bp. VNTR), COMT (rs4680), SLC6A4 (VNTR 44 bp. and rs25531), HTR1A (rs6295), HTR2A (rs6311), HTR1B (rs6296) and BDNF (rs6264) polymorphic loci in patients with suicidal behavior. The study was carried out with a sample of patients who had attempted suicide at least two times. We have supposed that the repeated suicidal attempts are not the result of impulsivity or mood disorders only but show the genetic predisposition to suicidal behavior. We have found statistically signi cant associations between suicidal behavior and SLC6A3 (40 bp. VNTR), COMT (rs4680), SLC6A4 (rs25531), BDNF (rs6264) loci.
Although the association between SLC6A3 40 bp. VNTR-polymorphism and mood disorders, as well as personality traits, was con rmed [27,28], we do not know about research that studied an effect of this locus on suicidal tendencies. We discovered a strong association between short (≤ 9 repeats) allele and suicidal behavior (p = 0.009). Carriers of at least one S-allele (SS and LS genotypes) have a signi cantly higher risk of suicide (p = 0.019). Given the data of low expression activity of the 9R-allele [26], we can suppose that suicidal behavior may be associated with dopamine transporter de ciency and consequently with increased dopamine signalling.
Our research may complement the results of a meta-analysis that demonstrated that the COMT rs4680 polymorphism is associated with suicidal behavior [22]. The meta-analysis was performed on six studies and showed that Met (A) allele with low activity is associated with suicidal attempts. However, the authors draw attention to the dependence of the results on the inclusion of all six studies. When ve of the six studies were individually excluded from the analysis, the relationship between COMT Met-allele and suicide was no longer signi cant. According to our results, suicidal behavior is associated with the COMT G-allele (Val) (56,5% in cases vs. 46,6% in controls, p=0,028). Taking these data into account, we can conclude that the COMT rs4680 polymorphism possibly does not affect suicidal behavior.
We also con rmed the results of a meta-analysis that showed the risk of suicide in carriers of the Metallele of BDNF rs6264 [39]. The Met-allele was more common in the cases than in the control group (23% vs. 16%, p = 0,029).
We did not con rm the results of the meta-analysis that showed that the S-allele of the 44 bp. SLC6A4 VNTR-polymorphism is a risk factor for suicide [15]. As far as we know, the relationship between SLC6A4 rs25531 and suicidal attempts was not veri ed before. We have found that the rare G-allele was even rarer in cases (5% vs. 9.9%, p = 0.047). We suppose this locus may contribute to predisposition to suicidal behavior together with other loci, but its contribution is small. A bigger sample is required to verify this association. We also tried to establish whether the effects of 44 bp VNTR and rs25531 polymorphisms on SLC6A4 expression are associated with suicide. We classi ed alleles and genotypes according to their functional activity and did not nd statistically signi cant associations ( Table 2).
The effect of dopamine receptor genes polymorphism on suicidal behavior is not well understood. Although some studies have shown an association between DRD2 rs1800497 or haplotypes with this SNP and suicidal behavior [33,34] no differences between allelic and genotype frequencies of this locus in our samples of patients and controls were found. Just like our research, previous studies failed to nd an association between DRD4 48 bp VNTR-polymorphism and suicidal attempts [29,46]. We also did not nd DRD4 In/Del 120 bp. to be associated with suicide. In our knowledge, this is the rst study aimed to establish if the DRD4 in/del 120 bp. polymorphism related to suicidal behavior.
Our data on the lack of association between HTR1A (rs6295) and HTR1B (rs6296) polymorphisms and suicide are consistent with previous meta-analyses [21,22]. Contrary to the results of a meta-analysis showing a link between HTR2A (rs6311) and suicidal behavior [20], we did not nd statistically signi cant differences.
It was previously shown that all of these genes are associated with major depression. We tried to verify if these loci affect severity of depressive symptoms. ANOVA showed the lack of association between these loci and depression symptoms, as well as with anxiety symptoms, but revealed an effect of SLC6A4 40 bp VNTR and HTR1B loci on situational and personal anxiety, respectively.

Conclusions
In this study, we show that polymorphism of the SLC6A4, SLC6A3, COMT and BDNF genes might be associated with the risk of suicidal behavior. For the rst time, the effect of the dopamine transporter gene (SLC6A3) 40-45 bp VNTR-polymorphism on suicidal behavior was examined and we have found that a short allele may increase the risk of suicidal attempts. We also con rmed the results of previous studies of the BDNF rs6264 locus, the T-allele of which is associated with suicide. In contrast to the metaanalysis of studies of the rs4680 COMT gene polymorphism, which shows that the G-allele (Val) is associated with suicide, we obtained data on the possible association of suicidal behavior with the Metallele. The association of SLC6A4 rs25531 polymorphism with suicide was rst studied, but because the G-allele is rare in the population, this result requires veri cation on a large sample.
We showed that SLC6A4 40 bp VNTR and HTR1B loci might be related to severity of anxiety symptoms.
Further studies are required to verify if these polymorphic loci are associated with suicidal behavior only or if they might increase the risk of severe mental illness leading to suicide attempts.

Availability of data and materials
All data generated or analysed during this study are included in this published article and its supplementary information les.

Competing interests
The authors declare that they have no con ict of interest.

Funding
The present study was supported by the Russian Foundation for Basic Research (projects no. 19-04-00383, no. 17-29-02203-o -m, no. 19-015-00380). The funding body played no role in the study design, the collection, analysis and interpretation of the data or the preparation and approval of the manuscript.
Authors' contributions ER performed genotyping, statistical analysis, data interpretation, and wrote the original draft. MS performed DNA isolation and genotyping. PS contributed to conceptualization and edited the article. AG collected blood samples and patients data. AR contributed to conceptualization and edited the article. DS made contributions to the methodology and genotyping. VV led the investigation and made contributions to the funding acquisition, methodology, and design of the study. All authors have read and approved the manuscript. The most signi cant results marked + p < 0.1, ++ p < 0.05 *-genotypes classi ed as LL, LS and SS (L≥10, <10) **-genotypes classi ed as LL, LS and SS (L≥7, S<7) *** -genotypes classi ed according to their expressional activity Table 3. One-way ANOVA. Effects of gene polymorphisms on the depression and anxiety symptoms