Expression of micro-RNAs miR-31, miR-146a, miR-181c and miR-155 and their target gene IL-2 are altered in schizophrenia: a case-control study [version 1; peer review: 1 approved, 1 approved with reservations]

Background: Schizophrenia is a severe psychiatric disorder with a heterogeneous clinical phenotype. The association of interleukins and other cytokines and their receptors with schizophrenia has been previously reported. Additionally, a number of studies have reported altered mico-RNA (miRNA) expression in schizophrenia and other psychiatric disorders. The aim of our study was to explore the possible association of miR-31, miR-146a, miR-181c and miR-155 with schizophrenia pathogenesis, as well as their link to IL2 gene expression in disease. Methods: For this case-control study, 225 patients with paranoid schizophrenia and 225 sexand age-matched controls with no family history of schizophrenia were recruited. The expression of studied miRNAs and the IL2 gene was measured using qPCR. DNA samples of all patients and controls were genotyped for IL2 rs2069778 single nucleotide polymorphism (SNP) using PCR with sequence specific primers (PCR-SSP). Statistical analyses include the Mann-Whitney Utest and Fischer’s exact test. Results: All studied miRNAs were over-expressed in schizophrenic patients IL2 gene expression was down-regulated in schizophrenic patients. The IL2 rs2069778 SNP is not associated with schizophrenia but regulates expression of the IL2 gene. Conclusions: Over-expression of studied miRNAs and downregulation of IL2 gene expression may be considered as genetic risk factors for chronic schizophrenia. Abnormalities in studied miRNA expressions result in the deregulation of the T-cell receptor signaling Open Peer Review


Introduction
Schizophrenia (OMIM code: 181500) is a severe psychiatric disorder with a heterogeneous clinical phenotype 1 . While the etiology of this disorder remains largely unknown, it has become evident that immune-inflammatory processes play an important multilevel role in disease development and progression [2][3][4] . Both cellular and humoral components of the innate and adaptive immune system were shown to be altered at different stages of disease development, both locally within the central nervous system as well as at a systemic level. Cytokines and chemokines are essential signal mediators of the immune system that modulate and guide immune/inflammatory responses as well as perform a wide range of other functions related to cell survival, proliferation, differentiation and migration 5 . Considering their biological importance, cytokines, especially interleukins (IL), have received considerable attention in the context of schizophrenia. The association of interleukins and other cytokines and their receptors with schizophrenia has been previously reported both on the level of genetic variants as well as the levels of gene expression and protein abundance, indicating their essential role in disease predisposition, development, progression and treatment response 6-9 .
IL-2 controls a wide range of biological activities, largely depending on the biological context. It is essential for T lymphocyte proliferation and differentiation, but it is also implicated in the generation and maintenance of regulatory T (Treg) cells 10,11 . The role of the IL2 gene in schizophrenia is still unclear and conflicting. However, previous studies, including our own findings, suggest the involvement of IL-2 in the pathogenesis of schizophrenia. Decreased lymphocyte production of IL-2 and increased IL-2 receptors have been reported previously [12][13][14][15][16] . Interleukin-2 receptor gamma (IL2RG) is an important signaling component of receptors for many cytokines, including IL-2, -4, -7, -9, -15 and -21 13 . Moreover, the IL2RG gene is over-expressed in the blood of schizophrenia patients 13 . On the other hand, other groups reported increased IL-2 serum levels in schizophrenia [17][18][19] . Some IL2 genetic polymorphisms were also reported to be associated with schizophrenia 20,21 . Expression of the IL2 gene is controlled at multiple layers. The IL2 gene contains at least two cis elements for transcript stability regulation, located in both the 3' and 5' untranslated regions (UTRs) 22,23 . Single nucleotide polymorphisms (SNPs) in the promoter region of IL2 influence the expression levels of this cytokine. Finally, expression of IL2 is also regulated by micro-RNAs 24 .
Micro-RNAs (miRNAs) are class of small, non-coding RNAs (comprised of about 22 nucleotides). MiRNAs are found in animals, plants and some viruses. They function in the regulation of gene expression at posttranscriptional level and RNA silencing [25][26][27] . As miRNAs are involved in the normal functioning of eukaryotic cells, deregulation of miRNAs has become associated with disease. There is a manually curated "miR2Disease" database, which aims to provide a comprehensive resource of microRNA deregulation in various human diseases 28 . Also, a number of studies have reported altered miRNA expression in schizophrenia, bipolar disorder and major depression and anxiety disorders 29-32 .
The aim of our study was to explore the possible association of miR-31, miR-146a, miR-181c and miR-155 with schizophrenia pathogenesis, as well as their link to IL2 gene expression in disease. This study is the first to report genetic association between schizophrenia and mentioned above miRNAs; however, several studies have reported a role for the IL2 gene in schizophrenia. All of these four micro-RNAs play a major role in regulating expression of the cytokine network. Particularly, miR-31, miR-146a and miR-181c are regulators of IL2 gene expression 33-35 , while miR-155 expression is greatly enhanced following stimulation of macrophages and dendritic cells by Toll-like receptors 36 . We also studied the possible association of the IL2 rs2069778 SNP genotype with IL2 and miRNA levels. This SNP was chosen due to its high minor allele frequency, clinical significance in autoimmune diseases 37 , as well as its location near the regulatory elements of the IL2 gene.

Ethical statement
Informed written consent was obtained from all study participants. The study has been approved by the Ethical Committee of the Institute of Molecular Biology of the National Academy of Sciences RA (IRB00004079, IORG0003427).

Study population
This case-control study was conducted from January 2016 to February 2017. A total of 225 patients with paranoid schizophrenia (SCZ) and 225 sex-and age-matched controls (CTRL) with no family history of schizophrenia were involved in this study (Table 1). This was the maximum available number of schizophrenia patients in Armenia who agreed to participate in this study. From these subjects, 61 patients and 60 controls were tested for micro-RNA expression and 66 patients and 99 controls for IL2 gene expression. There was no specific criteria for dividing subsets in this study; subjects were divided into groups according to availability of biological material (DNA and RNA). All subjects were genotyped for this study.
Paranoid schizophrenia (OMIM code: 181500, ICD-10-CM code: F20.0, DSM-5 code: 295.90) was diagnosed by two independent psychiatrists. Schizophrenia patients were recruited from the clinics of the Psychiatric Medical Center of the Ministry of Health of the Republic of Armenia (MH RA). Healthy subjects with any psychiatric illness during their lifetime, any serious endocrine or neurological disorder, any treatment or medical condition known to affect the brain or meeting the DSM-5 criteria for intellectual disability 38 were excluded from this study. Exclusion criteria for all study subjects included any treatment with immune-modulating drugs and serious medical disorder. Blood sampling and isolation of genomic DNA and total RNA from peripheral blood mononuclear cells A total of 10 ml of peripheral blood was collected in EDTA containing tubes (5ml for RNA and 5ml for DNA isolation) from each study participant.
Peripheral blood mononuclear cells were isolated from whole blood using the following protocol, as described in 13: 10 ml of Red Cell Lysis Buffer (RCLB) (containing 0.144M ammonium chloride, 1 mM sodium bicarbonate) was added to 5 ml of fresh blood. After 5 minutes, the mixture was centrifuged at 1000g for 10 minutes, the supernatant was discarded and the pellet was gently rinsed with RCLB. The pellet was re-suspended in 5 ml of the RCLB buffer and centrifuged at 1000g for 10 minutes. The primers and fluorescently-labeled Locked Nucleic Acid (LNA) probe from the Universal Probe Library for the IL2 gene were selected using the Probe Finder web-based software as follows:

Probe #50
Measurement of miRNA expression levels Reverse transcription for selected micro-RNAs and measurement of micro-RNA expression by quantitative real-time polymerase chain reaction (RT-PCR) were performed using TaqMan Micro-RNA Assays (see Table 2) and TaqMan Universal PCR Master MIX II (no UNG) (Cat. No. 4440040, Thermo Fisher Scientific, Waltham, USA), according to the manufacturer's instructions. qPCR was performed using Realist-DX IAB real-time PCR system (GeneTiCA, Czech Republic).
qPCR was performed using following thermal cycling conditions: polymerase activation for 10 minutes at 95°C; 40 cycles The presence/absence of allele-specific amplicons were visualized by electrophoresis using 2% agarose gel in 0.5x Tris-Borate-EDTA (TBE) buffer stained with ethidium bromide fluorescent dye. To check the reproducibility of results, randomly selected DNA samples of study subjects (10% of total) were genotyped twice.
Genotyping was carried out at Laboratory of Human Genomics and Immunomics, Institute of Molecular Biology NAS RA (Yerevan, Armenia).

Statistical analyses
The Shapiro-Wilk test for normality revealed non-parametric distribution of the obtained data. Therefore, the significance of difference in gene expression levels between each study group was analyzed by the Mann-Whitney U test. p-values less than 0.05 were considered as significant. Statistical analysis was performed using GraphPad Prism (version 5) software. Allele and genotype frequencies were checked for Hardy-Weinberg equilibrium and were in equilibrium. No investigation of potential sources of bias was undertaken.

Pathway analysis
For characterization of enriched functions and biological pathways of the studied miRNAs targets, we used miRsystem 43 , which performs enrichment analyses, accounting both for target genes as well as the levels of individual miRNA expression.  (Figure 2). It is worth noting that expression of IL2 in schizophrenic patients carrying the CC genotype was lower than in corresponding controls (p=0.0216), while the difference in expression between carriers of the T minor allele was not significant (p=0.22). Median expression levels of all studied miRNAs were significantly higher in schizophrenic patients as compared to healthy controls (Table 3 and Figure 3) (Figure 4). We have not found any association between IL2 rs2069778 variants and expressions of miR-155 and miR-146a.

Functional annotation of biological pathways containing targets for studied miRNAs
Our analysis demonstrated significant up-regulation of IL2 expression-modulating miRNAs in schizophrenia. However, it is known that miRNAs can affect multiple targets. For characterization of enriched functions and biological pathways of the studied miRNAs targets, we used a freely available online integrated system called miRsystem 43 . The analysis resulted in 30 pathways from 4 databases (KEGG 46 , Biocarta, Reactome, Pathway Interaction Database) significantly enriched with miRNA targets. The majority of these were related to immune/ inflammatory system pathways where IL2 is either an effector or a target gene (Table 4). Figure 5 is an example of miR target enriched pathway 46 .

Discussion
In this study, we observed decreased levels of IL2 expression in peripheral blood mononuclear cells of schizophrenic patients, paralleled with increased expression of IL2-regulating miRNAs (miR-31, miR-146a, miR-155 and miR-181c). In addition, we demonstrated that carriage of the minor allele for IL2 rs2069778 is associated with increased IL2 expression levels which might suggest either a regulatory role for this SNP or a linkage with other SNPs that can modulate gene expression.   Furthermore, there are studies reporting no difference in the IL2 levels of schizophrenic patients as compared to controls 52 . Because expression was studied in different populations, we can speculate that the observed discrepancies could be partially explained by different genetic backgrounds. Moreover, the validity of our results is supported by the detected increase of IL2 expression-modulating miRNAs measured using independent assay techniques. Finally, consistent with IL2's role in maintenance of Treg cells, their low levels were also reported in schizophrenia 53 .
Though the studies of miRNA involvement in schizophrenia are a relatively new direction, there are already results implicating miRNA deregulation in the pathogenesis of schizophrenia. miR-137 is the micro-RNA best known for its role in schizophrenia pathogenesis 54-56 . This micro-RNA is also well known due to a genetic polymorphism (SNP variant) in its gene, which was described as a genetic risk factor for the development of schizophrenia in a European population 57-59 .
There are other recent studies which confirmed the role of distinct micro-RNAs such as miR-195, miR-181b, miR-301a, miR-19, miR-206, miR-30a and miR-219 in the pathogenesis of schizophrenia 60-64 . In this study, we reported four miRNAs (miR-31, miR-146a, miR-155, miR-181c) that were up-regulated in schizophrenic patients. Besides targeting IL2 expression, these molecules have many other targets that are involved in immune/inflammatory pathways, confirming the essential role of immune system disturbances in disease development and progression.
The limitation of the present study is the inability to recruit medication-free patients for assessment of the effect of treatment on IL2 and miRNA expression. However, in many studies cited in this paper 50,51 , regardless the direction of difference in IL2 levels, no differences were observed between treated and untreated patient groups. It should also be noted that we measured IL2 gene and miRNA expression in two different patient groups with little overlap, which prevented us from performing direct correlation analysis between the levels of IL2 and miRNAs.
Overall, our findings further strengthen the role of immune system deregulation in the development and progression of schizophrenia and necessitate further research towards understanding the changes of the Th1/Th2/T-reg response in this disease and in response to antipsychotic treatment.