Major depressive disorder (MDD) is a highly disabling mental illness involving an imbalance in brain chemicals, and it majorly contributes to the global burden of disease (1). According to the World Health Organization, an estimated 350 million people of all ages suffer from depression disorder globally (2). In a systematic review, the summary estimate of the prevalence of depression or depressive symptoms among medical students was 27.2%, and that of suicidal ideation was 11.1% (3). A psychiatric disorder is not a sign of personal weakness or a character flaw, but it reveals an opposite result with a highly prevalent heritability that accounts for major psychological (4), physical (5), and social impairments (6). At present, the criteria for MDD diagnosis and treatment are based on various signs and symptoms that do not always fit into strict diagnostic categories, such as the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) (7). All of the possible causes for a set of past experiences have to be examined, including personal private medical information and confidential material, which increases stigma and makes diagnosis even more difficult (8). Recent studies implicate that functional magnetic resonance imaging (fMRI) may provide successful diagnostic information in depression disorder classification (9, 10); however, objective criteria and gold standards in early diagnosis for patients with MDD remain to be elucidated (7, 11). Previously, the microarray technique was used for life science research purposes. Bioinformatics data-mining of gene and microarray technologies has widely been used for differential expression analysis to identify novel diagnostic and therapeutic biomarkers of diseases (12, 13).

Over the past decades, several biomarkers have been proposed for MDD (13–15), but at the moment none of these biomarkers reaches sufficient sensitivity and specificity to be implemented in clinical practice (14). Recently, many potential mechanism studies have demonstrated that multiple genes and cellular pathways participate in the occurrence and development of MDD (15) and other mental illnesses (16). Numerous researchers have found that the pathophysiology of depression results from changes in oxidative stress (17), immune system effects (18), and neuroinflammation (19) in the central nervous system (CNS) through cytokines, which regulate brain activities and emotions. To understand the molecular processes that control neuronal activity and arrive at an objective diagnosis, we tried to obtain novel indicators of possible molecular mechanisms and predict peripheral blood molecular biomarkers in MDD patients and attempted to provide potential therapeutic targets for this challenging disease.

In the present study, two mRNA microarray datasets with MDD and control groups were downloaded from Gene Expression Omnibus (GEO) and screened for differentially expressed genes (DEGs). Gene Ontology (GO) functional annotation analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis in the online Database of Enrichr were performed for the screened DEGs. Then, we established a protein–protein interaction (PPI) network based on the Search Tool for the Retrieval of Interacting Genes (STRING) database and Cytoscape software to identify hub genes related to MDD. Subsequently, the hub gene and miRNA-mRNA pair interactions were identified. This work will provide further insight into the pathophysiology of MDD development at the molecular level and explore the potential molecular targets for new interventional strategies.

Depression is a major human blight that has become a pervasive public health problem (3, 46). Despite the rising prevalence of MDD, we lack an understanding of the distinctive pathophysiology in contrast to many other brain disorders. At present, laboratory blood tests to support MDD diagnosis are not available, so diagnosing this disorder is more challenging than measuring height (47). Recently, however, genetic insights transformed a featureless landscape into one with real scientific toeholds (48). The rapidly developing and wide use of microarray technology has revealed thousands of genetic alterations during the progression of diseases, which may provide promising targets for the early diagnosis of mental illness (49). Thus, there is a great need to identify biomarkers and provide proof of principle for a translational approach to prioritize blood biomarkers of mood state in MDD samples. In the present study, we explored the crucial genes of blood biomarkers and pathways associated with MDD by bioinformatics methods. To achieve this, two mRNA microarray datasets were analyzed to obtain DEGs and hub genes between peripheral blood from patients with MDD and that from the control group. A total of 123 DEGs (25 co-upregulated genes and 98 co-downregulated genes) and 14 hub genes were identified between the two datasets. Then, the DEGs were subjected to functional and pathway enrichment analysis, and a PPI network was constructed and integrated network analysis of miRNA-mRNA interactions performed to enhance our understanding of the molecular mechanisms of MDD.

To analyze the functional and pathway enrichment of DEGs between the subject groups, significant GO BP terms and pathways were obtained, including corticosteroid and glucocorticoid receptor signaling pathways, regulation of cell-matrix adhesion, and mitochondrial membrane permeability involved in apoptotic processes in upregulated genes. Downregulated genes were associated with neutrophil activation involved in the immune response, degranulation and mediated immunity, positive regulation of immune response and interleukin-8 production. Because the neutrophil activation involved in the immune response appeared in the downregulated genes in the pathway analysis results, it could be involved in an important part of MDD. Surprisingly, we observed in the literature that depression is mostly correlated with both peripheral inflammatory processes and alterations in cellular immunity, mainly for cell-mediated immunity. To date, a large number of studies have demonstrated that depression has been associated with positive regulation of interleukin-8 production and immune response (18, 50, 51). Euteneuer et al. (52) revealed that patients with MDD exhibited higher neutrophil and monocyte counts and an increased neutrophil to lymphocyte ratio (NLR) than controls. They also found that lower anti-inflammatory activity was related to more severe somatic depressive symptoms. Although there have been few studies on the immune response and MDD, and it is still unknown how the immune response regulates the pathology of depression. According to our analysis results, we speculate that the cellular immunity system might take part in the progression of MDD.

Based on the KEGG pathway analysis, downregulated DEGs were enriched for the Toll-like receptor signaling pathway, complement and coagulation cascades, NOD-like receptor signaling pathway, hepatitis B, measles, and influenza A. Recent data have demonstrated that NOD-like receptor pyrin containing 3 (NLRP3) activation appears to bridge the gap between immune activation and metabolic danger signals or stress exposure, which are key factors in the pathogenesis of MDD and other psychiatric disorders. TLRs also seem to be present in humans, and recent studies showed that the mRNA expression of TLR3 and TLR4 was significantly increased in the dorsolateral prefrontal cortex (DLPFC) of depressed individuals compared with controls (53, 54). Further experiments at the transcription and protein expression levels suggest that TLR3 and TLR4 appear to be unique and important in brain functions (55). There is mostly evidence for Toll-like receptors (TLRs) in the brain that are associated with depression and suicide (53). Interestingly, the Clinical Practice Research Datalink (CPRD) study from the UK-based primary care database suggests that influenza A infections are associated with a moderately increased risk of developing depression (56).

A PPI network was constructed to investigate the interrelationship of the DEGs, and 14 hub genes were identified, including RHOQ, TLR8, TLR4, FADD, BCL2A1, ENTPD1, CANT1, LRG1, MMP25, SLCO4C1, ADAM8, MOSPD2, PLAUR, and PLCG1. In addition to PLCG1, all other genes were downregulated in the PPI network. Inflammation is not the only cause of depression and cannot explain its entire pathophysiology, but it is an important pathogenic factor that explains one possible mechanism of depression. The subnetwork of PLCG1, BCL2A1, TLR8, FADD, and TLR4 screened out from our study has been shown to play a role in inflammation (36–39). FADD and BCL2A1 were implicated in nonapoptotic cellular processes and emerged as new actors in innate immunity and inflammation. According to a previous study, the anti-inflammatory effects and TLR profiles are predictors of the response to antidepressant treatment in patients with MDD (36, 55). As a pattern recognition receptor, TLR4 has been shown to play a vital role in neuroinflammation. The TLR4-specific inhibitor Cli-095 markedly inhibited the upregulation of TLR4 in the hippocampus and prefrontal cortex, and improved chronic unpredictable mild stress-induced depression-like behaviors in mice (57). Another study showed that stress significantly increased the expression of TLR4 and NF-κB in the hippocampus, and this phenomenon could be attenuated in TLR4 knockout mice (58). A growing body of research indicates that inflammation plays a critical role in the etiology and pathophysiology of depression.

miRNAs are a group of endogenous non-coding RNA molecules that likely regulate ~30% of human protein coding genes (59). According to the miRNA-mRNA binding data from online prediction tools, we identified genuine human miRNA-mRNA target pairs of MDD. In the present study, PLCG1 was predicted to be a potential target of 20 differentially expressed miRNAs and was upregulated in MDD. Research has shown that BDNF-mediated PLCG1 signaling is required for the formation and function of inhibitory synapses, whereby the disruption of PLCG1 signaling in the hippocampus leads to such dysfunctions. Interestingly, a clinical study showed that 5-HT1A signaling through tyrosine kinase receptors activates PLC/protein kinase C (PKC) signaling, mediating the synaptogenesis and behavioral actions of anti-depressants (60). Furthermore, previous studies have demonstrated that PLAUR plays a role in inflammation, tissue regeneration and axonal regeneration within the central nervous system (CNS) (45). In the brain, the PLAUR/Rho system seems to promote axonal recovery following a synapse function injury (61), which may be a potential target for the development of therapeutic strategies. The binding of recombinant PLAUR activation of β1 integrin via low-density lipoprotein receptor-related protein-1 (LRP1) leads to activation of the Rho family small GTPase Rac1 and Rac1-induced axonal regeneration (62). Furthermore, the miRNA-mRNA target pair network identified that an integral membrane protein ecto-ATPase enzyme, belonging to the nucleoside triphosphate diphosphohydrolase family (ENTPD1), was potentially targeted by various differentially expressed miRNAs. There is evidence that long-term depression might be modulated by ATP and/or its dephosphorylated product adenosine, such as E-NTPDases (41), which might contribute to the neural basis for learning and memory mechanisms. Molecular and cellular studies have demonstrated that the expression of neurotrophic factors, particularly brain-derived neurotrophic factors, is important for synapse function and development (63, 64).

The results of this study should be interpreted within the context of important limitations. First, our study utilized public data, but after screening of mRNA with clinical diagnostic and prognostic predictive value, it should be further explore the function of mRNA by in vitro and in vivo experiments. Second, the samples are from the peripheral blood cells of patients, so the associated analysis of miRNA/mRNAs in the brain regions with depression-related dysfunction may validate the data and strengthen the conclusion. Third, further validation studies could lead to additional insights into the disease process as well as the validation and identification of additional functional biomarker candidates for improved clinical diagnostic of MDD patients.

In summary, this comprehensive bioinformatic analysis has identified numerous useful molecular targets for the future investigation of the mechanisms and selection of biomarkers for MDD. Some important biological processes and pathways, including the corticosteroid and glucocorticoid receptor signaling pathways, the Toll-like receptor signaling pathway, the NOD-like receptor signaling pathway, the neutrophil activation involved in the immune response, as well as the hub genes working in these processes, may provide novel insights into the development and progression of MDD. Furthermore, the potential molecular mechanisms that have been identified simultaneously include innate immunity, neuroinflammation, and neurotrophic factors for synapse function and development. In addition, further molecular biological experiments will be performed by our team to confirm the function of the identified genes in MDD.

The datasets generated for this study can be found in the Gene Expression Omnibus database (GEO, www.ncbi.nlm.nih.gov/geo/): GSE76826 and GSE98793.

GZ and SX conceptualized and designed the article. ZZ, YZ, YW, JL, and JA analysed and interpreted the data. GZ drafted of the article. ZY, LS, and TS were responsible for critical revision of the article for important intellectual content. TS finally approved the article.