Association of Single Nucleotide Polymorphisms of Cytokine Genes with Depression, Schizophrenia and Bipolar Disorder

Immune gene variants are known to be associated with the risk of psychiatric disorders, their clinical manifestations, and their response to therapy. This narrative review summarizes the current literature over the past decade on the association of polymorphic variants of cytokine genes with risk, severity, and response to treatment for severe mental disorders such as bipolar disorder, depression, and schizophrenia. A search of literature in databases was carried out using keywords related to depressive disorder, bipolar disorder, schizophrenia, inflammation, and cytokines. Gene lists were extracted from publications to identify common genes and pathways for these mental disorders. Associations between polymorphic variants of the IL1B, IL6, and TNFA genes were the most replicated and relevant in depression. Polymorphic variants of the IL1B, IL6, IL6R, IL10, IL17A, and TNFA genes have been associated with schizophrenia. Bipolar disorder has mainly been associated with polymorphic variants of the IL1B gene. Interestingly, the IL6R gene polymorphism (rs2228145) was associated with all three diseases. Some cytokine genes have also been associated with clinical presentation and response to pharmacotherapy. There is also evidence that some specific polymorphic variants may affect the expression of cytokine genes. Thus, the data from this review indicate a link between neuroinflammation and severe mental disorders.


Introduction
Mental disorders are considered multifactorial diseases. They manifest as clinically significant impairments in cognition, emotion regulation, and behavior and can range from mild to severe [1]. Serious mental illnesses or severe mental illnesses limit one or more major life activities, interfere with quality of life, and can lead to disability. Mental disorders involve general neurobiological processes, such as disruption of the nervous tissue and the blood-brain barrier, autoimmune processes, and neurodegenerative processes. Glial and neuronal changes, metabolic changes in cellular processes, and molecular mechanisms lead to impaired expression of neurospecific proteins. The mechanism of immune response is manifested in changes in levels of hormones and neuromodulators, disruption of psychoendocrine processes, and neurotransmitter systems. Studies of the pathophysiology of severe mental disorders have traditionally emphasized dysregulation of the glutamatergic and monoaminergic systems. However, the mechanisms that cause these neurotransmitter abnormalities are still not clear. Accumulating evidence suggests the interaction of neuroinflammation with the serotonergic, dopaminergic, and glutamatergic systems and the pathogenic role of neuroinflammation in mental disorders [2][3][4][5].
The results of bidirectional Mendelian randomization analysis of genome-wide association studies (GWAS) indicate the causative relationship between inflammatory regulators and the risk of mental disorders including MDD, SCZ, and BD [16]. However, the most replicated and relevant genetic variants of cytokines include polymorphisms in the IL1B, IL6, and IL10 genes [17,18]. Many studies have revealed dysregulation of the concentration of genetically associated cytokines in the blood of patients with MDD [19][20][21], BD [22], and SCZ [23,24]. There is evidence that genetic polymorphisms affect the expression of pro-inflammatory cytokine genes [25]. Dysregulation of cytokine gene expression has been identified in mental disorders [26]. Nevertheless, the results of studies of the associations of cytokine gene variants with severe mental disorders are heterogeneous and require systematization.
This review synthesizes the current literature on polymorphic variants of cytokine genes associated with risk, severity, and response to treatment of mental disorders. The review focused on severe psychiatric disorders such as depression, SCZ, and BD. Severe mental disorders are characterized by a high degree of polygenicity and the involvement of many genes associated with the risk of developing these diseases. The shared genetic basis of immune activation and psychiatric disorders is of significant scientific and clinical interest because it may reveal new links between the immune system and psychiatric disorders. Although there is no clearly responsible single-nucleotide polymorphism (SNP), research proposes that polymorphic variants play an essential role in susceptibility to psychiatric disorders.
The aim of the review is to analyze studies of cytokine genes and establish the genetic overlapping or specificity of cytokine networks in severe mental disorders. Understanding the role of neuroinflammation in the etiology of mental disorders is essential to developing targeted and effective treatments and interventions. Knowledge of these interactions can help identify pathogenic clues and develop new preventive and symptomatic treatments. Advances in this field of science will enable the discovery of early biomarkers of mental disorders that will improve diagnostic and treatment outcomes and, consequently, the quality of life of patients.

Inclusion and Exclusion Criteria
The study includes the current literature on the association of polymorphic variants of cytokine genes with risk, severity, and response to treatment for bipolar disorder, depression, and schizophrenia. The review considered original case-control and prospective studies, GWAS, metaanalyses, Mendelian randomization studies on depressive disorders, SCZ, and BD. In addition, we have included studies of comorbid mental disorders with medical illnesses, such as cardiovascular diseases or cancer. We also included pharmacogenetic studies related to neuroinflammation and psychiatric disorders. Most studies analyzed more than one SNP, so they have been repeatedly cited in the text.
We excluded studies that focused on non-humans (rat or mouse models), as well as other types of diseases not comorbid with depression, SCZ, or BD, such as anxiety, neurosis, or neurological disorders.

Search Strategy
This review comprises the studies published in the last 10 years (between January 2013 and March 2023). We limited our review to these years to best demonstrate current data on the genetic associations between neuroinflammation and mental disorders.
For the literature review, we use sources from PubMed, Scopus, and the Web of Science. We focus on the association between genetic polymorphisms of the most reliable cytokines and severe psychiatric disorders, their severity, and their response to treatment. We selected articles that described studies of single nucleotide polymorphisms of cytokine genes in depression, SCZ, and BD. Keywords included the following: "gene" or "SNPs" or "single nucleotide polymorphisms"; "depression" or "depressive disorder" or "schizophrenia" or "bipolar disorder"; "inflammation" or "cytokine" or "interleukin".

Assessment of Studies
The titles and abstracts of articles from each search query were examined to identify those that showed the associations of single nucleotide polymorphisms with depressive disorder, SCZ, or BD. After analyzing the full texts of the selected articles, a decision was made to include or exclude the articles from the review.
For the synthesis we tried to group data by pathology at first and then by particular cytokines. A total of 44 articles were selected for this review. However, the resulting bibliography contains more sources to discuss the identified associations and expand the context. From the included studies we collected the following data: study identifications, number of participants and prime study conclusion.
A Venn diagram was used to summarize data on the association of SNPs of cytokine genes with depression, SCZ and BD. The online tool (https://bioinformatics.psb.ugent.be/ webtools/Venn/, accessed on 30 May 2023) was used to calculate the intersections of the list of SNPs. Inkscape 1.2.2 (Free Software Foundation, Inc., Boston, MA, USA) has been applied to the graphic design of a Venn diagram.

Results
The genes encoding cytokines are highly polymorphic. Previously, it was described that single nucleotide polymorphisms can be associated with increased or decreased production of cytokines. Polymorphisms in the promoter region of cytokine genes may result in inter-individual variation in transcription and expression of genes [25,27], thereby affecting the pathogenesis of mental disorders, the prognosis of their course, and the response to pharmacotherapy. Next, we consider the available data on associations of cytokine gene variants with mental disorders in more detail.

Depressive Disorders
Depressive disorders are a heterogeneous group of diseases. The etiology, precise pathophysiological mechanisms, response to treatment, and outcome of depressive disorders are still poorly understood. The cytokine hypothesis of depression proposes that pro-inflammatory cytokines acting as neuromodulators are a key factor mediating behavioral, neuroendocrine, and neurochemical changes in this disease [28].
Identifying a common genetic substrate for depression and immune activation will help unravel the link between neuroinflammation and depression. This section describes the association of inflammatory cytokine genes with depression, and polymorphisms that increase or decrease this association. Current research indicates that individual polymorphisms play a role in susceptibility to depression and outcome.
Gene polymorphisms encoding cytokines and their receptors can influence their functional activity. Available data on the genetic association of SNPs in cytokines and their receptor genes with depression is presented in Table 1.  It was shown an association between the T allele and the TT genotype of rs1799964 TNFA and low effectiveness of pharmacotherapy; the C allele and CT genotype were associated with good response to therapy. Carryer of GC and CC genotypes of rs1143623 IL1B showed varying levels of disease severity ccording to the HDRS. The combined genotypes of rs1143627-rs17561, rs1143627-rs1799964 and rs1143623-rs1799964, decreased the risk of depression occurrence, rs1143627-rs1800629 increased the risk. Kang et al. (2017) [34] IL1B: rs16944 (−511C/T), +3953C/T 969 patients at 2 weeks after ACS, 711-at 1 year later Depression during the acute ACS was associated with the −511T allele and the IL-1β levels. There was no association with depression in chronic ACS. There was no association with depression in the acute or chronic phase and the +3953C/T genotype.  The IL6-174G/C was associated with depression comorbid to CHD. The IL4-589C/T was associated with CHD.
No association between the TNFA-308G/A and the CRP-717A/G with depression in CHD.

patients with depression vs. 1HS
The AG genotype of rs2166975 TGFA was associated with an increased risk of depression, the GG genotype reduced the risk. The AG genotype and G allele of the rs2166975 TGFA was associated with increased risk of depression development in men. Genotype rs1800469*AA of TGFB1 was associated with earlier age of onset of the disease, GG genotype increased severity of the depressive episode. The IL1B gene is the most extensively examined in the field of psychiatry among the cytokine gene polymorphisms. Genetic polymorphisms in the IL1B gene have been well studied in depression. IL-1β has been implicated in the pathophysiology of major depression. Recent studies have identified the involvement of the IL1B gene in depression [29,34]. Among polymorphisms in the IL1B gene, more attention is given to the −511C/T (rs16944). The −511C allele in IL1B was associated with higher ILB expression [30] and higher IL-1β levels. The −511C/T polymorphism was associated with a depression [31,34]. The risk variant for depression was the CC genotype (p = 0.001, OR = 1.9 CI 1.3-2.7) [31]. There is much research dedicated to studying the contribution of IL1B-511C/T in the pathogenesis of depression disorders. Thus, it was shown that −511C/T polymorphism is associated with primary depression and post-stroke depression (PSD) at 2 weeks [29], with depression trajectory after acute coronary syndrome [34]. The −511C allele was also associated with more severe depression following chronic interpersonal stress exposure [30], and the T allele was associated with a positive history of major depression [55]. It was shown that depression in acute coronary syndrome was significantly associated with the level of IL-1β and −511T allele [34]. Another study reported that the −511T/T genotype was associated with both depression one week after surgery for breast cancer and persistent depression at one year of follow-up [38].
In a study by McQuaid et al. (2019) [32], it was shown that the severity of depressive symptoms was higher in individuals with the GG genotype of the IL1B rs16944 gene polymorphism. These results are consistent with earlier reports that carriers of the GG genotype have a greater severity of depressive symptoms [30]. The rs16944 IL1B was associated with childhood abuse as a predictor of depression scores. In particular, after childhood abuse, men carrying the rs16944*GG of IL1B showed particularly severe symptoms of depression [32]. In this way, the level of IL-1β and the −511C/T genotype, alone or together, may be biomarkers of depressive disorder. Targeted interventions for people with higher levels of IL-1β and the IL1B-511T allele may reduce the risk of depressive disorder [34].
IL1B rs1143643 was significantly associated with Mausdley Staging Method scores to determine treatment response in MDD [35]. In another study of depressive patients combined with childhood trauma, rs1143643 did not increase depressive symptoms. However, the minor A allele showed a protective effect against depressive symptoms after recent life stress [33]. Patients with GC and CC genotypes of rs1143623 (IL1B-1560G/ C) demonstrated different levels of disease severity as evaluated by the Hamilton Depression Rating Scale [18]. For the +3953C/T polymorphism of IL1B, no associations were found with depression in either the acute or chronic phases [34].
The results of the GWAS, published this year, suggest that specific SNPs, including rs2540315 and rs75746675 in the IL-1 receptor gene IL1R1, were associated with a rapid (within 240 min) antidepressant effect of ketamine infusion in patients with treatmentresistant depression [37]. An important limitation of the study is the small number of people examined (65 patients with treatment-resistant depression divided into 3 groups).
Interleukin-6 (IL-6) is a potent biomarker for depression, as its elevated plasma levels in patients with clinical depression have been confirmed by a range of studies [19,42,45]. Genetically-predicted IL-6 was associated with major depression in a multivariable mendelian randomization study (OR = 1.08; 95% C.I., 1.03-1.12) [56]. It has been shown that IL-6 levels are correlated with the IL6-634C/G polymorphism (rs1800796), and a G to C polymorphism at the −174 position of the IL6 promoter region appears to affect IL6 transcription [27,42]. The −572CC genotype and C allele were significantly associated with depression in the Han Chinese population [40].
It is also reported that the −174G/C polymorphism IL6 in interaction with various stress factors increases the risk of depression and has a greater impact on symptoms measured by the Zung Self-rating Depression Scale [42]. Udina et al. (2013) found that carrying the CC genotype of rs1800795 IL6 is associated with less severe IFN-α-induced depression and anxiety [44]. Russian researchers report that −174G/C polymorphism was associated with depression comorbid to coronary heart disease. The frequency of the allele G in this group was higher compared with controls [41]. The polymorphic variant rs1800795 (IL6-174G/C) is associated with recent stress on current depressive symptoms and is associated with lifetime depression at a nominal significance level [41].
The polymorphic variant rs2228145 (Asp358Ala) of the IL-6 receptor gene (IL6R) was associated with a reduced risk of severe depression and/or psychosis; the adjusted 95% odds ratio for patients with the CC genotype compared with the AA genotype was 0.38 (CI 0.15-0.94). This same polymorphic variant was associated with elevated serum IL-6 levels (P = 5.5 × 10 −22 ) [47].
Several studies suggest that IL-8 has neuroprotective functions [57]. At the same time, an investigation of IL8-251T/A in breast cancer patients found no association between the alleles and depression [38]. Furthermore, no association was found between IL8-251T/A and depression in a study of 732 elderly Koreans [58]. In the examination of a symptom in lung cancer, patients with IL8-251TT were more likely to experience severe depression but less susceptible to pain or fatigue [59].
Genetic studies of the anti-inflammatory cytokine IL-10 have shown that carrying genotypes GA and GG of rs1554286 IL10 is a predictor of anxiety (HR 1.85, p = 0.019) in early-stage breast cancer patients in China, which could help identify patients at high risk for psychological problems [60]. An investigation of 167 oncology patients showed that the rare AA genotype of rs1518111 was associated with subsyndromal depression [48].
A study of 398 breast cancer patients prior to surgery showed that the rs1295686*A of IL13 was associated with a symptom cluster of pain, fatigue, sleep disturbance, and depression [49].
IL-18 is expressed in the brain, and it is increased in patients with depression [50,61] and influences stress-related susceptibility to mood and anxiety symptoms by changing amygdala reactivity [62]. Reported that polymorphisms IL18-607A/C and IL18-137C/G were associated with the effects of antidepressant therapy [50]. Thus, patients carrying CA or AA genotypes of −607A/C and patients carrying GC or CC genotypes of −137C/G were significantly more prone to relapse after therapy and presented a significantly lower time to relapse [50]. Recent research on IL18 polymorphisms (rs187238, rs1946518 and rs1946519) has not found differences between depressive patients and healthy controls [61].
Meta-analysis data showed that the level of tumor necrosis factor-α (TNF-α) increased in MDD [63]. Increased levels of TNF-α may be conditioned by the presence of a range of specific polymorphic variants [64]. Higher TNF-α levels were associated with poststroke depression at 2 weeks in the presence of the −850T allele [29]. Another study showed that the TNFA-857CT genotype was associated with increasing the risk for prenatal depression in a Mexican mestizo population, and the −238GA genotype reduced the risk [51]. Homozygous for the rare allele in rs1799964 TNFA belonged to the subsyndromal depressive symptoms in patients with breast cancer [52]. A GWAS showed that among 57 genes and 92 SNPs identified in MDD patients, only rs769178 TNFA was related to depression, and it remained significant after correcting for multiple testing [14].
A number of studies have reported that the TNFA-308G/A polymorphism (rs1800629) is associated with depression [31] and is a risk factor for suicide attempts in MDD [36]. It was found that the TT genotype and the T allele of rs1799964 (−1031T/C) were associated with low effectiveness of pharmacotherapy, and the CT genotype and C allele were associated with positive responses to the treatment of depressive disorder [18]. The opposite data of the metaanalysis showed that there was no association of the TNFA-308G/A alleles or genotypes with poststroke, late-life, maternal, or major depression [65]. Another recent study, including 83 Polish patients, found no statistically significant association between the genotype/allele frequency of TNFA-308G/A and TNFA-1031T/C and depression [64].
The AG genotype of the rs2166975 TGFA was associated with an increased risk of depression development, while the GG genotype of the rs2166975 TGFA reduced the risk. That genotype increased the risk of MDD only in the male population [53].
The TGFB + 869T/C polymorphism predicts low activity of TGF-β expression. The study in the Bulgarian population revealed a significant prevalence of the TT genotype of the +869T/C polymorphism in patients with depression (41.3%) compared with healthy subjects (21.2%) (p = 0.05, OR = 2.62). In addition, the combination of TT-GC genotypes (+869T/C, +915G/C) in the gene is negatively associated with disease recurrence of depression [54]. Genotype AA of rs1800469 TGFB was associated with an earlier age of depression onset, while GG genotype increased the severity of the depressive episode [53]. One study showed an association between the rare A allele of rs2229094 TNFA and subsyndromal depression [48].
Thus, literature data show that depression is associated with polymorphic variants of cytokines genes. These associations also include the severity of the depressive disorder or the response to therapy. This complements and expands the data on immune dysfunction in this disease. Further research is needed to determine the precise impact of these polymorphisms and to find potential predisposing or protective alleles that can be used as biomarkers for the risk of depression.

Schizophrenia
There are many studies showing a link between changes in the inflammatory system and SCZ [12,66]. However, there are much fewer studies devoted to research on cytokine genes in SCZ (Table 2).  The T carriers (CT and TT genotypes) of +869T/C were significantly more frequent in SCZ (especially in females) than in HS. Association of polymorphisms in IL2, IL6 and IFNG genes with SCZ was not found. In the patients group it was observed an increased frequency of the T allele and the TT genotype of rs1126647. The T allele and the TT genotype of rs1126647 was associated with paranoid SCZ, and more specifically in females. The haplotypes TTT, ACT and TCT (rs4073-rs2227306-rs1126647) were associated with increased risk for paranoid SCZ, and only the TCT haplotype howed as a risk factor for SCZ. A number of studies demonstrate the association of polymorphic variants of cytokine genes with SCZ. Particularly, the case-control study reported that the A allele and AA genotype of the IL1A-889G/A (rs1800587) polymorphism were associated with SCZ in a South Indian population (p = 0.026; OR = 1.36; CI = 1.04-1.79) [67]. The analysis of 621 patients with SCZ in the Polish population showed an association between rs4848306 in the IL1B gene and SCZ [70]. It was shown that CC and GC genotypes of rs1800796 (IL6-572G/C), but not rs1800795 (IL6-174G/C), were significantly associated with chronic SCZ in a Han Chinese population. The −572GC genotype may serve as a protective factor for SCZ [40]. On the other side, Srinivas et al. (2016) reported that the G allele of IL6-174G/C (rs1800795) was associated with SCZ (p = 0.037). The GG genotype was observed to be in higher frequency in patients in a recessive model (GG vs. GC + CC); this association with SCZ was significant (p = 0.034) [67].
Moreover, there is a trend toward an association of rs1143627, rs16944, and rs1143623 in the IL1B gene with the risk of SCZ. Alleles rs1143627*G, rs16944*A, and rs1143623*G were significantly more frequently transmitted by parents to children with SCZ [68].
IL6-572G/C (rs1800796) was associated with SCZ at the genotypic level (p = 0.015). In an additive model (GC vs. GG + CC), this association was enhanced further (p = 0.003) [67]. The polymorphisms IL10-1082G/A (rs1800896), IL10-592C/A (rs1800872), and IL10-819T/C (rs1800871) were associated with SCZ, and the ACC haplotype was more prevalent in SCZ in Saudi Arabian patients [74]. One study found a significant association between rs11792633 IL33 and the risk of SCZ in an Iranian population. CT and TT genotypes significantly decreased the risk of SCZ [78]. The T carriers (CT and TT genotypes) of TGFB + 869T/C were significantly more frequent in SCZ (especially in females) than in healthy subjects [72].
Some of the study is dedicated to a paranoid form of SCZ. It was shown that genotype TT and allele T of the IL2-330G/T (rs2069762) polymorphism were significantly associated with the paranoid form of SCZ, as well as allele A of rs1800629 TNFA polymorphism [69]. A recent study showed that the rs1126647 of IL8 was a significant risk for SCZ. The TT genotype and T allele at rs1126647 were also associated with a paranoid form of SCZ. Haplotypes TTT, ACT, and TCT at rs4073-rs2227306-rs1126647 in IL8 were associated with increased risk for paranoid SCZ [57].
A number of researchers reported on association polymorphisms of cytokine genes and clinical characteristics of SCZ. In a study aimed at finding associations between interleukin genes and subdomains of negative symptoms in SCZ, calculated based on the Positive and Negative Syndromes Scale, it was found that the association between the IL6-174G/C (rs1800795) polymorphism and AA scores (avolition and apathy) was close to the level of significance. Patients with the IL6-174GG genotype had higher scores compared with the AA genotype [73].
A study published in 2022 found that there is a significant main effect of the IL10-1082G/A polymorphism (F = 5.56, df = 2, p = 0.004) and the IL10-592C/A polymorphism (F = 3.48, df = 2, p = 0.03) on the AA (avolition and apathy) scores of the Positive and Negative Syndromes Scale in patients with SCZ [73]. Post-hoc analysis (Bonferroni corrected) revealed that the mean score on the AA subdomain was higher in the IL10-592AA genotype group compared with the IL10-592CC group (p = 0.002). Differences between IL10-1082G/A genotypes were dose-dependent: the AA score decreased with the number of copies of an A-allele. The lowest score was observed in the group with the IL10-1082GG genotype.
The AA genotype of IL17A-197G/A (rs2275913) was associated with higher total scores of bizarre behavior and apathy in female SCZ patients [76].
Research of 772 inpatients with SCZ and 775 healthy controls in a Han Chinese population has shown that increased IL-18 serum levels and the IL18-607A/C (rs1946518) polymorphism were positively associated with the PANSS (the Positive and Negative Syndrome Scale) general psychopathology subscore and the PANSS total score [77]. Thus, the interaction between increased IL-18 serum levels and the −607A/C polymorphism influenced clinical psychopathological symptoms, and this dependence was present only among patients carrying the C allele.
The polymorphism rs6676671 in IL10 was associated with the early age of SCZ onset [70].
Transcriptomic coexpression analysis in the human brain revealed that genes most significantly co-expressed with IL10 were associated with synaptic vesicle transportation. Moreover, both IL10RA and IL10RB were most significantly co-expressed with genes that regulate inflammation and also with those that participate in synaptic formation [75]. The IL10-592C/A genetic variant was more common in SCZ patients than healthy subjects and was associated with lower serum levels of IL-10 and worse attentional performance in these patients. In the healthy human brain, the IL10 gene and its receptors are involved in the regulation of neuroinflammation and synaptic functions that are important for cognition, and hence their deficiency may contribute to cognitive impairment in SCZ.
Thus, it has been shown that there are associations between different polymorphic variants of cytokine genes and the SCZ, its forms, and its clinical characteristics, but the data obtained is not always unambiguous. These genetic data link immunoinflammation with the pathogenesis of SCZ. Further research is needed to clarify and expand our understanding of the role of the immune system in the pathogenesis of SCZ.

Bipolar Disorder
There are not many studies of cytokine genes in BD. Available data on the genetic association of SNPs in cytokines and their receptor genes with BD are presented in Table 3. Table 3. Association of cytokine and their receptor gene polymorphisms with bipolar disorder.

BD vs. 47 HS; Iranian cohort
The frequency CC and CT genotype of IL1B-511C/T were significantly different between BD patients and healthy controls. The T allele of IL1B-511C/T was significantly more frequent in patients with a positive history of MDD. The T allele of IL1B+3954C/T was significantly more frequent in early onset BD patients. Strenn et al. (2021) [79] IL1B: rs1143623, rs1143627, rs16944, rs1143634 188 BD vs. 54 HS The genotype distribution did not differ between patients with BD and the control group. Most studies in BD are devoted to the IL1B gene. Thus, it is shown that the frequency of CC and CT genotypes of IL1B-511C/T (rs16944) was significantly different between BD patients and healthy controls (p = 0.04 and p = 0.02, respectively) [55]. The T allele of IL1B + 3954C/T (rs1143634) was significantly more frequent in early-onset BD patients [55]. Pu et al. (2021) also showed association −511C/T with risk of BD [71]. Shonibare et al. (2020) showed a link between IL1B-511C/T and brain morphology. The −511C/T was associated with greater lateral occipital cortex surface area and volume in BD adolescents [80].
Among the seven polymorphisms of the IL1B gene researched by Pu et al. (2021), the minor alleles of five SNPs exhibited significantly increased frequencies in the BD compared with controls ( Table 2). Four of the five SNPs (rs16944, p = 0.00570, OR = 1.199 for the A allele; rs1143627, p = 0.00793, OR = 1.190 for the G allele; rs12621220, p = 0.00746, OR = 1.199 for the T allele; rs1143623, p = 0.0127, OR = 1.184 for the G allele) were significantly associated with the risk of BD [71].
One study researched the IL6 and IL6R genes. The C allele and CC genotype of IL6R rs2228145 were associated with the early onset of BD [81].
Thus, despite the limited literature data, it has been shown that there are associations between polymorphic variants of cytokine genes and the risk of BD. These data are proof of the immunoinflammatory theory of BD pathogenesis. However, given the limited data, new studies on large samples are needed to expand the existing understanding of the role of cytokine gene polymorphisms in BD.
The studied literature sources allowed us to conclude that mental disorders may have common etiological and maintenance processes as well as cognitive-affective, interpersonal, and behavioral features. The general results of the overall sample of studies indicate that three mental illnesses have common genetic associations.

Discussion
This review synthesizes the current literature on the association between genetic variants of cytokines and mental disorders risk, severity, and response to treatment. This review suggests that common variants of cytokine genes are associated with both immune alterations and the pathogenesis of psychiatric disorders, including depression, BD, and SCZ. It is hypothesized that immunity may cause psychiatric symptoms accompanying various disorders, and inflammation may be associated with dysregulation of the glutamatergic and monoaminergic systems and oxidative damage in psychiatric illness. Thus, immunity can be associated with the pathogenesis of psychiatric symptoms in patients with classical psychiatric disorders. Therefore, the data in this review indicate an association between neuroinflammation and depression, BD, and SCZ [82,83].
All data from this review on the association of SNPs in cytokine genes with depression, SCZ, and BD are summarized in Figure 1. The summed results indicate that three mental illnesses have common genetic associations. In particular, depression and BD share many genetic associations. Interestingly, the IL6R gene polymorphism (rs2228145) was associated with all three mental disorders. This may indicate a common immune-related genetic basis for these pathologies.
Accumulated data indicate that mental disorders are accompanied by changes in cytokine levels. For example, the anti-inflammatory cytokine IL-2 is altered in people with schizophrenia [63,84,85]; IL-6 concentrations are associated with depression [42,45,86]; elevated levels of IL-6 are trait markers for BD [87] and schizophrenia [63,88]; elevations of IL-8 in cerebrospinal fluid are seen in patients with schizophrenia [89]; IL-18 is increased in patients with depression [50,61]; the level of TNF-α increased in patients with SCZ, MDD [63] and BD [90]. It is important to note that changes in the expression of cytokine genes may be conditioned by the presence of a range of specific polymorphic variants [64]. Thus, the IL1B-511C allele is associated with higher IL1B expression [30], the IL6-634C/G polymorphism is correlating with IL-6 levels, and the IL6-174G/C is affecting its transcription [27,42]; the ACC haplotype (−1082G/A, −592C/A, −819T/C) of IL10 was associated with intermediate production of IL-10 in SCZ patients [74].
Psychiatric disorders are highly polygenic and likely to be affected by many genetic loci, each of which has little effect [91]. Our study of the literature data from the last 10 years showed that the most replicated and relevant genetic variants of cytokines include polymorphisms in the genes for IL1B, IL6, and IL10. However, as we have found, even for the most replicated findings, there are many inconsistent results in association studies of genetic variants, immune responses, and effects on disease. For example, numerous studies have revealed a relationship between the minor C allele, or CC genotype, of IL1B-511C/T and clinical psychopathological symptoms of different forms of depression [29][30][31]34,38]. On the other side, there is another study demonstrating the association of the TT genotype, not the CC, of IL1B-511C/T with depression [38]. This diversity of results may be explained, at least in part, by the heterogeneity of the depression immunophenotype, by environmental influences, genes and environment interactions, genetic variants, and gene expression variations [14]. Moreover, the combined effects of gene-environment interactions rather than purely genetic mechanisms play an essential role in neuroinflammatory processes. Thus, as we found, the effects of some SNPs may only become evident in the presence of stressors such as stroke, surgery, or cancer. For example, the IL6-174G/C polymorphism in interaction with various stress factors increases the risk of depression [42], and the IL1B-511C/T was associated with more severe depression following chronic interpersonal stress exposure [30]. These findings support the putative hypothesis that pro-inflammatory genetic variation increases the risk of stress-induced depression [30].
The range of polymorphic variants of cytokine genes identified in recent years is quite high. An analysis of the literature data leads to the identification of the involvement of some polymorphisms in various pathologies. For example, rs16944 IL1B showed a significant association with depression in scientific studies [31,32,34], symptom severity [30], as well as with SCZ [68] and BD [55,71]. The polymorphism rs2228145 in IL6R is associated with depression [47] and BD [81]. The IL6-174G/C polymorphism was associated with the severity of symptoms of current depressive disorder [30], with recent stress on current depressive symptoms, with lifetime depression [43], and with mean scores on the AA subdomain of the Positive and Negative Syndromes Scale in patients with schizophrenia [73]. It was also found that rs16944 and 1143623 of the IL1B gene were associated with both depression [18,[29][30][31] and BD [71]. The polymorphic variants of IL6, IL18, and TNFA were associated with depression [39,40,44,50,51] and SCZ [67,69,71,77].
Conversely, there are a number of studies demonstrating the absence of significant associations between polymorphic variants of immunoinflammation genes and different mental disorders [34,38,58,65]. Therefore, these data require replication in other studies.
Perspectives. It is known that the effectiveness of treatment depends on the genetic and clinical heterogeneity of patients and the occurrence of side effects [92]. Only about 30 percent of the patients respond to treatment and experience remission. Future pharmacogenetic testing will help determine the person-specific genetic factors that may predict clinical response and side effects [93].
Limitations of the evidence. Significant limitations of the mentioned studies are that they had different sample sizes (tens to thousands). Clinical samples in some of the studies were generally small (less than 100 samples) [37,55,78] and were underpowered for reliable detection of associations between neuroinflammation and mental disorders. Future studies with larger sample sizes are needed to explore these associations. Another limitation of these studies is the sample's clinical heterogeneity from study to study. The next factor in studies heterogeneity is covariates in analyses. Some of the studies do not include factors influencing inflammatation but not related to mental disorders such as age, gender, medication use, alcohol consumption, obesity, race and ethnicity, menopausal status, and other factors. Including these factors in standardization would strengthen research findings and facilitate comparisons of multiple studies and meta-analyses. Moreover, different researchers studied mental disorders, in particular depression, in combination with organic disorders such as cancer or stroke. Therefore, all these factors can probably introduce heterogeneity into the pooled results.
Limitations of the review. We focus on the studies published only in the last 10 years. There is much data on the topic published earlier that was not included in the review. We included only three pathologies: depressive disorder, SCZ, and BD. Neither addictive disorders, nor anxiety, nor neurosis, nor other mental disorders or animal models were considered. We included only studies on single nucleotide polymorphisms and did not consider duplications, deletions, and other chromosome mutations.

Conclusions
Numerous studies demonstrate an association between different severe psychiatric disorders and inflammation-related biomarkers. The shared genetic basis of immune activation and psychiatric disorders may reveal new links between the immune system and psychiatric disorders. This review synthesizes the current literature on the association between polymorphic variants of cytokine genes and mental disorders such as depression, SCZ, and BD, their severity, and their response to treatment. Understanding the effect of minor genetic variants on the immune system and their contribution to the development of psychiatric disorders is important for the identification of vulnerable individuals, the establishment of genetic biomarkers, and the development of new approaches to therapy based on pharmacogenetics.