HLHS is characterized by LV hypoplasia and increased biomechanical pressure on RV by single ventricular physiology [23]. By analyzing HLHS gene RNA-seq profile data, we obtained 14889 genes for HLHS. The expression profiles of these genes were used as data sources to perform WGCNA on HLHS and a total of 16 co-expressed gene modules were identified in this study. Among them, black module and pink module were the two modules most relevant to HLHS. Based on the genes in these two modules, we constructed PPI networks. Finally, 10 hub genes were confirmed for HLHS, which were differentially expressed in LV and RV tissues between controls and HLHS. By verifying these hub genes in the HLHS expression profile data of an independent cohort, we found that most of the genes were consistent in the HLHS cohort of different data sources, indicating that our analysis method was accurate and reproducible.
We identified five hub genes for HLHS in the black module, including Fbn1, Itga8, Itga11, Itgb5 and Thbs2. Fbn1 (Fibrillin-1) has been found to be associated with heart development [24]. Its mutation could increase genetic susceptibility to thoracic aortic aneurysms [25]. Furthermore, its mutation leads to Marfan syndrome (MFS) that is the most common hereditary connective tissue disease [26]. In this study, Fbn1 expression was distinctly up-regulated in LV/RV HLHS compared to controls. Furthermore, there was a significant difference in Fbn1 expression between LV and RV for control mice. Itga8 (Integrin Subunit Alpha 8) inhibits NFκB and JAK-STAT signaling and cardiac injury in myocardium without stress [27]. Itga8 expression was significantly higher in LV/RV HLHS in comparison to controls both in the GSE77798 and GSE23959 datasets. Itga11 (Integrin Subunit Alpha 11) expression has been detected to be increased in methylglyoxal-induced collagen-treated human cardiac fibroblasts and streptozotocin-treated Sprague-Dawley rat cardiac fibroblasts, which may promote the formation of pre-fibrotic fibroblasts and fibrotic stroma in diabetic cardiomyopathy [28]. Itgb5 (Integrin Subunit Beta 5) has been identified to be in significant correlation with coronary artery disease and age-dependent organ fibrosis [29, 30]. We found that Fbn1 had distinctly higher expression in LV/RV HLHS in comparison to controls. Moreover, a significant difference in Fbn1 expression was detected between LV and RV for control mice and neonates. Thbs2 (Thrombospondin 2) mediates cell-matrix interactions, vascular integrity and thrombosis [31]. In our study, its expression was lower in LV HLHS than controls, which was higher in RV HLHS compared ton controls.
Five hub genes including Cblb, Ccl2, Edn1, Itgb3 and Map2k1 were screened for HLHS in the pink module. Cblb (Casitas B-cell lymphoma-B) is lowly expressed in plaques for human atherosclerosis, thereby leading to CD8+ T cell-induced macrophage death and accelerating atherosclerosis [32]. Our findings revealed that Cblb expression was significantly down-regulated in RV HLHS not LV HLHS in comparison to controls. It has been reported that targeting Ccl2 (C-C Motif Chemokine Ligand 2) could ameliorate atherosclerosis [33]. Moreover, Dectin-2-mediated Ccl2 in resident tissue macrophages can facilitate cardiac arteritis [34]. In mice, Ccl2 expression was up-regulated in LV HLHS than controls. However, no significant difference in Ccl2 expression was detected between HLHS and controls. Edn1 (Endothelin 1) genetic locus is correlated to spontaneous coronary artery dissection [35]. It was highly expressed in LV/RV HLHS compared to controls in mice, and was highly expressed in RV HLHS than control neonates. Itgb3 (Integrin Subunit Beta 3) is related to myocardial infarction risk [36]. Map2k1 (Mitogen-Activated Protein Kinase Kinase 1) mutation is often in association with the clinical phenotype of the cardiovascular system skin syndrome [37]. Our study found that Map2k1 was highly expressed in LV HLHS and lowly expressed in RV HLHS compared to controls in mice. Furthermore, its low expression was found in human RV HLHS in comparison to controls.
Complex life phenomena are the result of the interaction of a large number of biological components. Biological research has shifted from collecting gene and protein information to systematically using this information to clarify the synergy between them. In this study, we tried to probe into the molecular mechanism of HLHS through functional enrichment analysis of DEGs-related HLHS. DEGs in LV HLHS were distinctly involved in heart development, apoptotic signaling pathway and ECM receptor interaction. Abnormally expressed genes related to heart development could contribute to the progression of HLHS. Furthermore, imbalance of apoptotic signaling pathway in cardiomyocytes may be an important factor of HLHS. As a previous study, RV tissues in HLHS exhibit immature ECM and increased cardiomyocyte apoptosis [38]. Thus, the roles of these DEGs in HLHS need further exploration. DEGs in RV HLHS were mainly enriched in BMP signaling pathway, regulation of cell development and regulation of blood pressure. Dysregulation of the BMP pathway is the basis of many diseases of different organ systems in humans [39]. As a previous study, changes in gene expression in the BMP pathway has been found in RV tissues for neonates with HLHS [23]. Cardiomyocytes from neonates with HLHS exhibit multiple expression and function differences [40], which could be mediated by a variety of DEGs at a transcriptional level [41]. Our study found that DEGs in RV HLHS were involved in the regulation of blood pressure, as previous studies [42].
Facing the increasing amount of high-throughput data, it is a difficult problem about how to effectively extract useful information to obtain the regulatory relationship between genes in the research of systems biology. The regulatory relationship between genes has spatiotemporal specificity. In different organs, different physiological conditions and pathological states, and at different time points, this regulatory relationship will change accordingly. It is these changes that determine cell proliferation, differentiation, as well as occurrence, development of HLHS. The modularity of the biological network is the result of living organisms to achieve specific biological functions. Modularity provides us with a simple and effective method to understand the regulatory relationship between genes, which is an indispensable method in the research of systems biology. This study is the first to analyze HLHS data by WGCNA. Our results showed that WGCNA can discover biologically meaningful gene modules, and the hub genes related to the clinical information found are consistent with literature reports, which also proves the accuracy and effectiveness of WGCNA of gene expression data. Further excavation of information on gene modules will help us better understand the role and significance of hub genes, key signaling pathways, as well as the regulatory mechanisms between genes on the development of HLHS.