2.1 Subjects
A sample of patients (N = 100) of East Slavic origin who had attempted suicide at least two times and were monitored by a psychotherapist was used for this study. All blood samples were collected with the informed consent of the investigated persons after a participant's personal statement signature. The Ethics Committee of the Institute of Molecular Genetics (Institute of Molecular Genetics, Russian Academy of Sciences, Kurchatov sq. 2, Moscow, Russia) approved the study (protocol 03\19, 19 February 2019).
A survey of patients and blood samples was carried out at the Moscow Research and Clinical Center for Neuropsychiatry of the Healthcare Department of Moscow. Depressive symptoms were evaluated with help of Hamilton’s Depression Rating Scale (HAMD) and Beck's Depression Inventory (BDI). Spielberger’s test was used for evaluation of the degree of situational and personal anxiety. The patients were diagnosed with the following: depressive episode (F 32.1), N=22; recurrent depressive disorder (F 33.1), N=22; mixed anxiety and depressive disorder (F 41.2), N=13; bipolar disorder (F 31.3), N=10; schizopathic disorder (F 21.8), N=6; emotionally unstable personality disorder (F 60.31), N=5; other, N=24. Demographic and clinical characteristics of patients are shown in Table 1.
Table 1. Demographic and clinical characteristics of the patients with suicidal behavior
n
|
total
|
100
|
female
|
80
|
male
|
20
|
Age, years
|
min
|
18
|
max
|
77
|
mean, SD
|
31.54±11.13
|
BDI
|
min
|
8
|
max
|
53
|
mean, SD
|
28.85±9.58
|
HAMD
|
min
|
7
|
max
|
35
|
mean, SD
|
21.19±5.22
|
Situational anxiety
|
min
|
25
|
max
|
80
|
mean, SD
|
55.52±12.93
|
Personal anxiety
|
min
|
25
|
max
|
80
|
mean, SD
|
56.44±10.05
|
BDI — Beck's Depression Inventory
HAMD — Hamilton’s Depression Rating Scale
The control group (N = 154) was a sample of the East Slavic population of the city of Moscow and the regions of Central Russia (59 men (38.3%) and 95 women (61.7%); mean age: 62.19 ± 9.45 years). The control group included people over the age of 40 to reduce the likelihood of the presence of a genetic predisposition to suicide, which has not yet manifested. Details of patients and controls characteristics and genotyping results are shown in Supplementary File 1.
2.2 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.
The SNP and VNTR genotyping was carried out using locus-specific PCR as described previously 42-47.
2.3 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 (HWE). Logistic regression approach was applied to establish associations between gene polymorphisms and risk of suicidal attempts. Negative binomial regression was used to analyze association between genotypes and count variables: severity of depressive symptoms (HAMD and BDI scales), personal anxiety and situational anxiety. The statistical significance of polymorphisms was established with a likelihood-ratio test. Akaike information criterion (AIC) was used to identify the model that best fits the data. All calculations were performed in the R statistical environment.
The following genetics models were tested:
- Co-dominant. This model assumes that each genotype can influence risk independently of the others.
- Common allele homozygotes were tested against rare allele homo- and heterozygotes.
- Rare allele homozygotes were tested against common allele homo- and heterozygotes.
- Over-dominant. Heterozygotes were tested against both homozygotes.
- Log-additive. A trend test for the genotypes, according to this model, each allele changes the risk in an additive manner (i.e. the presence of two alleles double the risk compared to the presence of only one allele). The test was based on a logistic regression model and genotypes were coded as 0, 1, or 2 depending on the amount of minor alleles.
The strength of associations between allelic variants of studied polymorphic loci and suicidal behavior was estimated using odds ratios (ORs), with the corresponding 95% confidence intervals (95% CIs). All tests were conducted at a level of significance p < 0.05.
Given the data suggesting the effect of length of SLC6A3 40 bp VNTR locus on gene expression and high activity of the 10R allele compared with the 9R allele, we classified cases and controls as carriers of the long (≥10) and short (<10) alleles.
Because of the complexity of SLC6A4 organization, we analyzed 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 (LA) and low (S and LG) alleles with high and low expression activity, respectively, and high/high (LA/LA), high/low (LA/SA, LA/LG) and low/low (SA/SA, SA/LG, LG/LG) genotypes. The alleles of the DRD4 48 bp VNTR were grouped into long (≥7) and short (<7) allelic variants.