So far, several datasets containing the human glomeruli mRNA expression profiles from IgAN individuals were produced and subjected to analysis by different strategies. The main objective of such investigations is to clarify the disease’s basic molecular mechanisms, along with introducing key drivers in the progression of the disease. For instance, in one experiment after reanalysis of two IgAN datasets, including GSE73953 and GSE93798, tumor nephrotic factor (TNF) and mitogen-activated protein kinase (MAPK) pathways were introduced as the key involved pathways in the IgAN progression (13). Similarly, in another experiment, after analysis of three datasets, including GSE37460, GSE93798 and GSE104948, Miraji et al, revealed the association of “extracellular matrix receptors interaction pathways”, “extracellular matrix expansion” and “inflammatory mechanisms” with the progression of IgAN (14). By construction of a PPI network among the overlapped DEGs and considering the degree of connectivity between the genes, the authors introduced several hub genes with therapeutic potentials including FN1, ITGB2, FCER1G and PTPRC (14).
Despite these findings, it seems that more excavations, especially by means of potent and comprehensive strategies are needed to cut deeper and reach more exhaustive and robust results, especially in the case of complex diseases like IgAN. In the present study, we applied WGCNA as an advanced and comprehensive algorithm to construct gene co-expression networks, exploring modules, and identifying disease correlated modules and genes in IgAN related samples. At first, the IgAN dataset GSE104948 was subjected to several pre-analysis steps including normalization and outlier removal and after analysis using the ‘limma’ R package, the identified DEGs were subjected to co-expression network construction. After clustering the co-expressed genes and module detection procedure, biological process enrichment analyses and hub-gene identification steps were performed for all the co-expressed modules. Of all the genes in modules that showed a high kME value, as well as a high degree of centrality in the PPI network, 16 genes were identified and after the validation procedure, 11 genes were introduced as true hub genes. Notably, most of the true hub genes (7 out of 11) were coming from the blue module, which also spotted as the top disease correlated module. As far as we know, there is no WGCNA analysis study on this dataset. In addition, for hub-gene identification, the present experiment applied a more comprehensive strategy, when compared to other similar experiments. Researchers usually consider the degree of connectivity in either a PPI network of genes with high kME values in a module or a PPI network including genes of only one module (15, 16). However, to keep the holistic view of systems biology, we mined the hub genes, considering both the list of top genes in the co-expressed modules, as well as a PPI network comprising of the genes in all modules. Moreover, a regulatory network that included the hub gene’s related miRNAs and TFs was constructed to catch a comprehensive view of the disease pathogenesis.
GO and Reactome pathway data are two categories of biological materials essential for understanding mechanisms underlying the disease processes. The immune system signaling, inflammatory responses, cellular communication and the extracellular matrix-related pathways were among the top enriched pathways for genes in the blue module. Such results were in line with previous experiments showing the association of innate immune responses and inflammatory reactions with IgAN (17, 18). Neutrophil degranulation, cytokine signaling and inflammatory responses were other enriched terms that confirmed the strong link of immune-related pathways with the progress of IgAN.
Toll-like receptors (TLRs) are well-known components of the innate immune system. Similar to previous experiments (19, 20), the results of present study also revealed the upregulation of these receptors in the IgAN samples. After binding to their ligands (pathogen-associated molecular patterns), TLRs trigger various immune signal cascades in order to promote immune system activation. Nevertheless, these receptors might cause glomerular damage through the induction of inflammatory cytokines in IgAN patients (21). In the present study, TLR1 and TLR2 were introduced as two hub genes with potential involvement in IgAN progression. Co-expression and interaction of these two receptors finally lead to the activation of NFκB, as well as different immune cells like B cells, dendritic cells, mast cells, NK cells and keratinocytes (22). Accordingly, these receptors might be considered as potential therapeutic targets aimed for the attenuation of immune responses in IgAN. Although the potential role of TLRs has not been extensively investigated in the IgAN progression, they have shown to play an important role in the induction of inflammatory responses in other kidney diseases (22).
CYBB, CSF1R, TYROBP, ITGB2, and CD44 were other identified hub genes in the blue module that function either as regulatory elements or signal transducers in the regulation of immune responses. Despite the CYBB, all of the identified hub genes in the blue module were transmembrane proteins participating in cell-cell communication and/or signal transduction.
The CYBB gene is responsible for coding cytochrome b-245, the key subunit of the NADPH oxidase, which is the membrane-bound oxidase of phagocytes. The key function of NADPH oxidase is host defense through regulation of antigen processing and presentation, as well as regulation of phagocytes and neutrophils (23).
CSF1R is a transmembrane protein acting as a cell-surface receptor for CSF1 and interleukin-34. The crucial role of CSF1R in the regulation of survival, proliferation and differentiation of mononuclear phagocytes like macrophages and monocytes has been shown by previous studies (23). Upon ligand binding, CSF1R enhances the release of proinflammatory chemokines and therefore has a significant role in inflammatory processes (24, 25). Similarly, TYROBP or DAP12 encodes a transmembrane signaling polypeptide, which acts as a signal transduction element. Activation of additional tyrosine kinases, cell activation, integrin-mediated neutrophil activation and formation of inflammatory cytokines are some revealed functions of this transmembrane protein (26, 27).
Cell surface interactions and extracellular matrix organizations were other enriched terms for the blue module genes. These terms are not irrelevant to the immune system-related pathways, since cell-cell and cell-matrix intercommunications are vital in triggering inflammatory responses and activating immune cells. Here, functions of ITGB2 and CD44 as two identified hub genes, as well as Rho GTPases could be of attention, due to their roles in cytoskeletal organization, cellular communication, and immune signaling.
Rho GTPases are well-known for their regulatory roles in cytoskeleton dynamics, cell movement, cellular signaling, phagocytosis and inflammation (28, 29). Based on some reports, Rac-1 and RhoA, as two key members of Rho GTPases, are listed as two mediators of podocyte dysfunction and therefore, their inhibition might be beneficial for handling chronic kidney diseases (CKDs) (29, 30). Despite such findings, there is a limited number of studies concerning the involvement of Rho GTPases in the IgAN pathogenesis. Considering their roles in signaling transduction and cytoskeleton organization, these molecules could be candidates of more investigations exploring their involvement in IgAN progression.
As part of integrin heterodimers, ITGB2 is participating in both cell adhesions and surface-mediated signaling. According to previous experiments, there is a negative correlation between ITGB2 and eGFR in patients with CKD (31). In addition to cell surface interactions, ITGB2 has also been shown to be involved in the regulation of immune system-related pathways like toll-like receptors cascades, neutrophil degranulation, and interleukin signaling pathways (32, 33). However, so far, a limited number of studies have pointed to the possible role of ITGB2 in IgAN progression and it seems that more investigations are needed to shed a light on this issue (34).
CD44 was another marked hub-gene showing an up-regulated pattern in IgAN patients. CD44 which is a well-known cell-surface glycoprotein is involving in diverse biological pathways like hematopoiesis, cell adhesion, proliferation, migration, and lymphocyte activation (35). Due to various physiological activities of CD44, so far, involvement of this glycoprotein has been shown in a wide range of disorders, including vascular disease, arthritis, infections, and cancers (36). Considering the contribution of CD44 in cell-cell and cell-matrix connections, same as ITGB2, this protein might play a role in immune system signaling and triggering the inflammatory cascades, thus could be considered as a target of more investigations aimed for attenuation of immune responses in IgAN. According to previous experiments, up-regulation of CD44 in glomerular visceral epithelial cells could be a sign of active injury in glomerular and kidney dysfunction in IgAN patients (37). In another experiment, a significant correlation was observed between the expression of CD44 in glomerular and tubulointerstitial and renal damage degree in IgAN individuals (38). Therefore, in addition to its therapeutic potential, CD44 may be regarded as a reliable marker of the IgAN progression.
Other identified hub genes in the present experiment included proteasome 20S subunit beta 9 (PSMB9), G protein subunit gamma 11 (GNG11), platelet and endothelial cell adhesion molecule 1 (PECAM1) and DNA methyltransferase 1 (DNMT1) were coming from the black and red modules. Genes of the black module were mainly enriched in vasculature development, defense response to virus, leukocyte proliferation, and regulation of cell adhesion.
Considering the functions of GNG11, as a member of the heterotrimeric G protein complex and PECAM1, as a receptor on platelets, monocytes, granulocytes, macrophages, lymphocytes and endothelial cells, here again, we can see the contribution of these two hub genes in intercommunication and stimulation of immune cells (39, 40). Likewise, PSMB9 as an essential subunit of the 20S proteasome complex, is playing a key role in antigen processing, generation of class I binding peptides and finally activation of CD8 T cells and NF-κB pathway (41, 42). As far as we know, there is no investigation pointing to the potential role of the above-mentioned proteins in the progression of IgAN. However, their contribution in immune cell intercommunication and activation could imply their association with autoimmune diseases like IgAN.
Another identified hub-gene was DNMT1, which is a well-known epigenetic factor transferring methyl groups to CpG structures in DNA. Based on some findings, genomic factors also could have an impact on the pathogenicity of IgAN (43, 44), and in this context, methylation of DNA by DNMT1 could be of attention. Although, we found no investigation pointing to the role of DNMT1 in the IgAN progression, inhibition of this epigenetic factor in the kidneys of diabetic nephropathy db/db mice model led to podocyte protective effects (45). Hindering the progression of kidney diseases like IgAN by inhibition of DNA methylation could be an innovative therapeutic idea and at this point, DNMT1 could be a potential target.
Taking some more steps towards translational medicine, we also performed miRNA and TF enrichment study and constructed a network comprising of true hub genes, and their related miRNAs and TFs. Up to now, the position and diverse roles of miRNAs have been verified in various diseases, where their mutations or aberrant expressions could trigger or augment a condition. Such connections between the disease phenotype and miRNA dysfunction/dysregulation have raised the idea that miRNA modulation might change the disease progression. In the present study, miR-129-2, miR-34a, and miR-27a were identified as top up-stream regulators of the hub genes. The involvement of miR-27a and miR-34a in regulating inflammatory responses in some kidney diseases like diabetic nephropathy and IgAN have been shown previously (46, 47). However, the present experiment is the first one indicating the potential regulatory role of miR-129 in the progression of IgAN. This miRNA has been shown to orchestrate different genes involving in cell proliferation, cell cycle, apoptosis, DNA methylation, and metastasis. Moreover, its aberrant expression was observed in different cancers, pointing to the potential role of this miRNA in cancer development (48).
In case of TFs, signal transducer and activator of transcription 3 (STAT3) was identified as top TF, affecting the expression of the hub genes. STAT3 is expressed in different cell types, like leukocytes and fibroblasts and following activation by interleukin-6, this TF could target genes that induce the production of growth factors, cytokines, and ECM components (49). Since, these elements are contributing to tissue fibrosis, inhibition of STAT3 activation (phosphorylation) might hinder the process of kidney fibrosis in CKDs (50). It seems that STAT3 and the three enriched miRNAs are other pieces of the big network orchestrating the inflammatory responses in the IgAN disease. Though, further investigations are required to validate this hypothesis.
All in all, it should be mentioned that the main limitation of this study was the absence of an experimental section, examining the expression of the identified hub genes in IgAN samples. However, to compensate for this limitation we performed validation procedure in other IgAN datasets.