Screening of immune-related secretory proteins linking chronic kidney disease with calcific aortic valve disease based on comprehensive bioinformatics analysis and machine learning

In recent years, with the widespread applications of microarray and sequencing methods, the molecular landscape and potential mechanisms of miscellaneous diseases can be easily explored [28, 29]. In addition, integrative bioinformatics analysis and machine learning tools are increasingly performed to explore the novel genes, potential diagnostic/prognostic biomarkers, underlying mechanisms, and prospective therapeutic targets based on the big data, which can shed more lights on the diseases [30, 31].

By applying a variety of comprehensive bioinformatics analysis approaches, to our knowledge, the present study is the first to excavate CKD-related pathogenic genes to elucidate the association between CKD and subsequent CAVD. It was surmised that inflammatory and immune processes together with signaling pathways including “cytokine-cytokine receptor interaction”, “PI3K-Akt signaling pathway” and “NF-Kappa B signaling pathway” might be the potential mechanisms underlying CKD-related CAVD. Moreover, two immune-related hub genes, SLPI and MMP9, were employed to develop diagnostic nomogram models to predict the risk of CAVD by machine learning approaches. According to our results, these two hub genes displayed ideal predictive performance for CAVD, as assessed by the ROC curve. At last, through the external validation of our cohort, the upregulated expression patterns of SLPI and MMP9 were confirmed to be consistent with the obtained datasets, and the diagnostic nomogram models based on SLPI and MMP9 levels performed well in significantly differentiating CAVD, particularly CAVD in CKD patients.

Increasing clinical studies have suggested that patients with CKD suffer a significantly increased incidence [32] and accelerated progression of CAVD [33]. As speculated in previous studies, CKD contributes to vascular calcification through calcium deposition, hyperphosphatemia and reactive oxygen species (ROS). Additionally, CKD is assumed to play a significant role in cardiovascular diseases partially by means of excreting secretory proteins, such as pro-inflammatory cytokines, TGF-β and bone-related proteins [34]. However, the potential factors and mechanisms participating in CKD-related CAVD are not fully understood.

CAVD is previously considered as a degenerative disease that occurs with age, however, a growing amount of evidence starts to realize that CAVD is an active pathological change, which is driven by a series of proactive multifactorial processes, including cellular transformation, apoptosis, oxidative stress and immune response [35]. Lately, the roles of inflammation and immunoregulation in the pathogenesis of CAVD have aroused an increasing attention. According to a previous report, the number of leukocytes in the aortic valve increases from 5% at birth to about 12% at 60 days of age [36]. Besides, local macrophages, CD4+ and CD8+ T lymphocytes are found to be activated in the calcified valve, leading to the production of more proinflammatory factors [37]. Furthermore, valvular osteoblast differentiation of valvular interstitial cells (VICs) may be promoted by invading monocytes and macrophages, at the same time, these cells themselves undergo calcification via secreting tumor necrosis factor (TNF) [38]. In this study, the GO-biological process annotation and KEGG enrichment analyses showed that the CKD-related pathogenic genes for CAVD were mostly enriched in the inflammatory and immunological relevant pathways, indicating that the inflammatory-immune pathways might be the potential mechanism in CKD-related CAVD.

Currently, the effective pharmacotherapy for the treatment of CAVD is still lacking, in this regard, it is urgently needed to explore the potential drugs. Numerous important breakthroughs have been made in the past few years in identifying small-molecular compounds with therapeutic potential in a variety of diseases. Small-molecular compounds exhibit several advantages, including high tissue penetration, a tunable half-life and oral bioavailability, making them more effective on treating patients [39]. Quinazoline-4-piperidine sulfamides (QPS) have been depicted as the inhibitors of Ectonucleotide pyrophosphatase/PDE1 (NPP1), which can attenuate the high phosphate-induced mineralization in a cellular model of CAVD [40]. However, no previous studies have disclosed potential small-molecular compounds for therapeutic application of CAVD based on gene expression signatures in the calcified aortic valve via high-throughput screening. Herein, by cMAP analysis, this study provided a novel perspective linking CKD-related pathogenic genes to discover the potential compounds targeting CAVD. The upregulated CKD-related pathogenic genes in the calcified valve were applied to cMAP analysis, and 10 small-molecular compounds (metyrapone, gefitinib, dilazep, aminopentamide, methoxsalen, forskolin, CGP-37157, IKK2-inhibitor, vidarabine and TG-101348) were selected as candidates. Of note, metyrapone, a potent inhibitor of 11-beta hydroxysteriod dehydroger and mineralocorticoid receptor as well as cytochrome P450, showed the highest negative enrichment score in cMAP analysis, implying that it maximally reversed the expression of upregulated CKD-related pathogenic genes in CAVD. Although no direct link is found between metyrapone and calcification, increasing studies have reported that metyrapone can ameliorate numerous cardiovascular disease such as cardiac remodeling [41] and endothelial dysfunction [42] by abrogating corticosterone signaling. Interestingly, the previous studies have established the pathogenic roles of CKD-related corticosterone signaling in vascular calcification [43, 44]. In addition, the metyrapone-mediated corticosterone inhibition also suppresses the production of pro-inflammatory factors, expression of adhesive molecules and accumulation of monocytes in neurovascular disorder [45]. On the basis of the above previous findings, the therapeutic effects of metyrapone make it possible to be a potential agent for the treatment of inflammatory and immunological diseases including CAVD. Thus, it is speculated that early medical intervention with metyrapone in CKD patients may not only improve the kidney function but also inhibit the initiation and progression of CAVD, finally significantly prolong the life span of patients.

Over the past few decades of life, CAVD is usually asymptomatic, but once symptoms occur, CAVD has often stepped into the severe stage. In this case, aortic valve replacements, either by surgical or transcatheter approach, are the only effective treatments, which are associated with the disadvantages of high costs and a high complication rate. Consequently, it is beneficial to diagnose and prevent CAVD in the early stage. It is estimated that one third of the aged population are diagnosed with the early stage of CAVD features, as indicated by the echocardiographic or radiological evidence [46]. Limited by the skills of the echocardiography operator and the quality of the imaging, it is needed to identify more conventional serum biomarkers for the early diagnosis of CKD patients with CAVD. Most noteworthily, a more comprehensive diagnostic nomogram model was established based on two hub genes in this study, which presented a higher diagnostic value for CKD-related CAVD than that of an independent biomarker. Moreover, the nomogram model was efficient in diagnosing patients with sclerotic aortic valve, indicating that this diagnostic nomogram was also potent in predicting the early stage of CAVD. Furthermore, external validation from our cohort revealed the elevated SLPI and MMP9 mRNA levels in aortic valve tissues of CAVD groups compared with control groups. Serum SLPI and MMP9 levels were also increased in patients with CAVD and higher in patients with CAVD and CKD, and our constructed diagnosis nomogram was capable of significantly distinguishing CAVD as well as CAVD in CKD patients.

SLPI belongs to the family of whey acidic proteins [47], which plays an important role in inhibiting human neutrophil-derived serine proteases, such as elastase and cathepsin G [48, 49]. Previous evidence suggests that SLPI may be a novel biomarker and target candidate for acute kidney injury (AKI), indicated by upregulation of SLPI mRNA levels in AKI allografts as well as elevated protein levels of SLPI in plasma and urine of AKI patients [50]. Moreover, it was identified as a novel biomarker for CKD patients with CAVD in our study. SLPI is principally expressed in epithelial cells, but it can also be secreted by endothelial cells, adipocytes and host-defense effector cells [49, 51, 52]. SLPI has been extensively reported to exert its function via several significant biological processes, such as host defense, inflammatory response and cell fate regulation [48]. Upregulation of SLPI increases the levels of osteoblast-related markers including Runx2, Sp7 and Col1a1 in MC3T3-E1 cells (the mouse osteoblast cell line), and promotes the proliferation of MC3T3-E1 cells [53]. Therefore, SLPI activation can strengthen osteoblast differentiation and proliferation. Noteworthily, aortic VIC undergoing osteoblast differentiation is found to have a critical effect on the development process and promote the progression of CAVD [54, 55]. However, the mechanisms regarding of SLPI in CAVD have not been elucidated yet. In this study, SLPI, as an important regulatory factor for inflammation and immunology, showed increased expression in calcified aortic valve in comparison with control aortic valve samples. In this regard, our study indicated that SLPI might provide a potential diagnostic indicator for CKD patients with CAVD.

Besides, MMP9 was identified as the perspective contributor to the diagnosis of CKD patients with CAVD in this study. MMP9, which belongs to the zinc-dependent endopeptidase family, is involved in immunology activation, inflammatory cascade regulation, extracellular matrix (ECM) disassembly and remolding to afford ways for immune cell accumulation in the pathogenesis of different diseases. A few studies have indicated that MMP9 contributes to atherogenesis through facilitating the migration of vascular smooth muscle cells and the invasion of macrophages. In addition, arterial stiffening is credited with the elevated expression of MMP9 since it plays a certain role in elastin degradation, leading to subsequent matrix remodeling. An earlier study has identified MMP9 as a pathogenetic factor for calcified aortic valve stenosis, and inhibition of MMP9 attenuates reactive oxygen species production and calcium deposition by improving the mitochondrial morphology and metabolism in calcified aortic valve interstitial cells [56]. Furthermore, the increased levels of circulating MMP9 is significantly associated with diabetic nephrology progression, and is specially involved in the development of albuminuria in patients with CKD [57]. Interestingly, our data suggested that the expression of MMP9 was significantly upregulated in CKD patients with CAVD. As a result, it was speculated that MMP9 might interrupt the balance between the anabolism and catabolism of ECM, and promote macrophage infiltration to participate in CAVD progression. Conclusively, MMP9 is assumed to be an appropriate biomarker for distinguishing calcification.

In immune cell infiltration analysis, the accumulation of various types of immune cells has been demonstrated to exist in all stages of CAVD, which is significantly related to the severity of aortic stenosis [58,59,60]. Previous studies have demonstrated that calcified aortic valve tissues and peripheral blood harbor diverse kinds of activated T lymphocytes [61, 62], where T cells CD8 exhibit a greater invasion ability than other subpopulations [62]. Moreover, activated T cells CD8 contribute to CAVD via secreting IFN-γ, eventually facilitating the progression of aortic stenosis [63]. Furthermore, macrophages, the heterogeneous innate immune system cells, can be classified as two major phenotypes, including pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages [64]. They can modulate phenotypic switch rapidly in response to the local microenvironment. Both M1 and M2 macrophages are reported to be accumulated in patients with CKD, with a lower proportion of M2 macrophages being detected in calcified aortic valves. In this study, significant differences in the infiltration of immune cells were identified between CAVD and control groups, with higher abundances of Macrophages M0, T cells CD8 and Tregs, whereas lower proportions of B cells naive, Dendritic cells activated, Macrophages M2, Mast cells activated, NK cells activated, Plasma cells and T cells CD4 naive. Furthermore, the hub genes SLPI and MMP9 showed close association with immune cell infiltration in CAVD, implying that the candidate biomarkers might not only distinguish CAVD but also contribute to CAVD by interaction with inflammatory-immune pathways. Thus, it is vital to comprehensively understand the inflammatory-immune pathways related to CAVD in order to develop novel diagnostic or prognostic biomarkers and therapeutic targets for CAVD.

留言 (0)

沒有登入
gif