TB is an infectious disease that seriously harms human health. It is caused by M. tuberculosis that is parasitic in macrophages(41). As the human body’s first line of immune defense, macrophages can kill M. tuberculosis through phagocytosis, oxidative stress, acidification and antigen presentation(42). Due to the lack of specific auxiliary examination indicators, it is difficult to realize the early diagnosis and effective treatment of TB patients with the traditional TB diagnosis scheme.
The method of bioinformatics is helpful to analyze the expression of key genes to reveal the potential molecular mechanism of the biological behavior of TB, and to provide novel views for elucidating the pathogenesis of TB. Microarray technology allows us to explore the host genetic changes and gene expression related to TB, and has proven to be a useful method for identifying new biomarkers in other diseases(43).In this study, four GEO microarray data sets (GSE51029, GSE52819, GSE54992 and GSE65517) were integrated to identify DEGs between PBMC of TB patients and healthy persons to offset the false positive rate in the analysis of independent datasets. According to the criteria of |log FC|> 1 and p values < 0.05, this study used FunRich software to display the intersection of DEGs in the four microarray expression datas, and found a total of 46 overlapping DEGs that can be considered as candidates. The GO function and KEGG pathway enrichment analysis, PPI network analysis and hub gene selection, the construction of miRNA–hub genes network and target TF–hub genes network, and the verification of the expression of hub genes at the mRNA level were carried out successively.
According to the results of GO enrichment analysis, for the BP category, DEGs participate in the defense response to the virus, the regulation of the immune effect process, the cellular response of IFN–γ, the regulation of the innate immune response and the chemokine-mediated signal pathway. Previous studies have shown that during infection, macrophages encounter M. tuberculosis before being stimulated by IFN–γ produced by T-helper 1 (Th1) cells(44). However, IFN–γ stimulation is necessary for the complete activation of antibacterial and antigen presentation functions in macrophages(45). For the MF category, DEGs were found to be associated with cytokine/chemokine receptor binding, cytokine/chemokine activity, G protein coupled receptor binding, receptor ligand activity, and signal receptor activator activity. M. tuberculosis is known to induce host proinflammatory mediators that play an important role in disease control(46), including chemokines, which are small molecular weight proteins involved in immune regulation and inflammation(41). For the CC category, DEGs were found to be significantly enriched in membrane rafts, the mitochondrial outer membrane, cytoplasmic vesicle cavities, endocytic vesicles and nuclear chromatin. KEGG enrichment analysis showed that NOD–like and toll-like receptor signaling pathways were associated with M. tuberculosis infection. Innate immune cells are known to use various pattern recognition receptors, such as toll–like receptors (TLRs), C-type lectin receptors, and NOD–like receptors to respond to pathogen components when performing a variety of biological functions(47, 48). Previous research using eperimental models of TB have emphasized the importance of TLRs in the prevention of M. tuberculosis infection.(49). In addition, antigen recognition of NOD–2 (nucleotide-binding oligomerization domain 2), a member of the NOD–like receptor family, is also crucial in conferring immunity against viruses or bacteria, which may include M. tuberculosis(50, 51). This suggests that the coordinated triggering of TLRs and NOD-2 may lead to a stronger and lasting immune response, thereby limiting the growth of M. tuberculosis(52), which would be consistent with our results showing enrichment of these pathways during M. tuberculosis infection.
By constructing a PPI network and analyzing it with MCODE and Cytohubba in Cytoscape, five hub genes were identified, including STAT1, TLR7, CXCL8, CCR2 and CCL20. Previous studies have reported that in the early stage of TB infection, STAT1 can promote downstream apoptotic factors to activate transcription through phosphorylation(53). At the same time, STAT1 plays an important role in the polarization of macrophages to the M1 type, involved in the immune response to viruses and bacteria, including M. tuberculosis. Polarized M1 macrophages have been shown to eliminate M. tuberculosis infection more effectively than Polarized M2 macrophages(54). It is also reported that after ssRNA upregulates TLR7, the number of M. tuberculosis in macrophages is significantly reduced, and the macrophage viability is significantly increased, indicating that TLR7 can effectively inhibit the growth of M. tuberculosis and increase the viability of macrophages(42). Kane et al. found that fibroblasts have a previously unrecognized role in regulating TB inflammation via a CXCL8-dependent contribution to immune cell recruitment and mycobacterial killing in granulomas(55). In the study of Dunlap et al., the mouse model provided evidence that the CCR2 axis is essential for protective immunity against the emerging M. tuberculosis lineage infection(56). Another report showed that CCL20 is overexpressed in monocytes infected by M. tuberculosis and inhibits the production of reactive oxygen species (ROS)(57).Therefore, prior research provides biologically plausible mechanisms by which these genes may be involved in immune responses to M. tuberculosis infection, supporting our results.
To study the molecular mechanisms of potential hub gene disorders, it is necessary to search for potential miRNAs through bioinformatics methods. The miRNA is an endogenous non–coding RNA molecule with a length of 18–22 nt that targets the 3'UTR region of a gene. It can regulate gene expression at the post-transcriptional level to degrade or inhibit the translation of target genes(58). MiRNAs are known to regulate protein translation inhibition or targeted mRNA cleavage(59). More and more evidences have showed that miRNAs are closely related to the occurrence and development of cancer and other major diseases. In our analysis, the three DEGs most associated with miRNA regulation were CXCL8, TLR7, and STAT1. At the same time, we observed 10 miRNAs, and found that their targeting involves at least 2 hub genes. Among these, hsa-mir-335-5p was found to be associated with the greatest number of genes; however, there are relatively few previous studies on this miRNA. One study found that the gain or loss of hsa-miR-335-3p function can lead to changes in the expression of GATA4 and NKX2–5 markers during the cardiac differentiation of human embryonic stem cells(60). In addition, hsa-miR-335-3p has been identified as an upstream regulator of two modules related to the recurrence of osteosarcoma patients(61). The results of bioinformatics analysis found that hsa-mir-335-5p has high potential value used as a new biomarker. Our study also established a TF–hub genes regulatory network to further explore the molecular regulatory mechanisms underlying TB(62). TFs are the primary regulators of gene expression and are associated with pathogenesis in TB. In our study, we also found several TFs that interact closely with hub genes, including max-like protein X (MLX), transcription factor DP 1 (TFDP1), retinoid X receptor alpha (RXRA), zinc finger protein 197 (ZNF197), glucocorticoid modulatory element binding protein 2 (GMEB2), and tripartite motif containing 22 (TRIM22). The complex interactions between TFs and hub genes have made a huge contribution to the development of the disease. Our analysis, which identifies miRNAs and TFs associated with newly identified hub genes, provides potential candidates for the development of therapeutic targets and exploration of the biological mechanism of TB in future research.