Unravelling the complexity of kidney renal clear cell carcinoma prognosis: integrating chromatin regulators, gene signatures and associated immune landscapes

doi:10.21203/rs.3.rs-3612712/v1

Download PDF

Research Article

Unravelling the complexity of kidney renal clear cell carcinoma prognosis: integrating chromatin regulators, gene signatures and associated immune landscapes

https://doi.org/10.21203/rs.3.rs-3612712/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Kidney-renal clear cell carcinoma (KIRC) is the main subtype of renal cell carcinoma. KIRC exhibits significant resistance to conventional treatments, highlighting the need for pioneering therapeutic approaches. Within this framework, chromatin regulators (CRs) - proteins important for managing gene expression and orchestrating key biological processes - have been recognized as key players in the initiation and development of KIRC.

Methods

Utilizing the TCGA-KIRC dataset, we conducted differential gene analysis pertaining to chromatin regulators through application of the "limma" R package. We proceeded to establish and validate a prognostic model via LASSO Cox regression, with a particular emphasis on genes exerting substantial influence on KIRC prognosis. Our investigation was further extended to investigate the interrelationship between gene attributes, clinical parameters, the tumor microenvironment, and drug responsiveness. To enhance the predictive efficacy of our models, we harnessed advanced bioinformatics methodologies and techniques for visualizing protein interaction networks.

Results

Through the related studies, we found that the risk score obtained for CRs constituted an autonomous prognostic determinant in KIRC. Subsequently, a Nomogram prediction model was crafted that amalgamated clinical attributes with their corresponding risk evaluations. Ultimately, Polymerase chain reaction (PCR) was applied for comparative analysis of BRD9 expression levels in normal tissues and tumor specimens. Notably, the utilization of BRD9 marker-based constructs yielded significant predictive results.

Conclusion

We have introduced a novel prognostic framework for KIRC that seamlessly incorporates Chromatin Regulators. This innovative model exhibits substantial promise in enhancing the precision of prognostic forecasts for individuals afflicted with KIRC, thereby establishing a foundational platform for the refinement of therapeutic approaches.

KIRC. Chromatin regulators. Tumor microenvironment. Drug sensitivity. Tumor prognostic model. Survival analysis of core genes

Kidney clear cell carcinoma (KIRC) constitutes the predominant subtype among renal cell carcinomas, accounting for approximately 75%-80% of cases. This subtype is associated with a distinct cancer-specific mortality rate of 24%, as evidenced in studies [1–3]. Kidney renal clear cell carcinoma exhibits a conspicuous inclination toward the male demographic [4]. The pathogenesis of this malignancy has been linked to dysregulated growth factors resulting from von Hippel Lindau (VHL) gene perturbations [5]. Additionally, external factors such as smoking[6], obesity [7], and diabetes [8] have been implicated in modulating the KIRC pathogenesis. Despite the diversification of therapeutic strategies, KIRC exhibits remarkable resistance to conventional chemotherapy and radiotherapy. Thus, there is a pressing need to explore novel treatment modalities, leading to the integration of immunotherapeutic approaches in the therapeutic arsenal for KIRC [9, 10]. Given the complex clinical landscape, the identification of novel prognostic markers becomes imperative. These markers not only provide accurate prognostic insights into tumor progression but also have the potential to facilitate the implementation of personalized molecular-targeted therapeutic regimens, thus advancing the field of personalized medicine.

Chromatin Regulators (CRs) represent a class of proteins that play a pivotal role in governing gene expression. Their primary function revolves around the precise modulation of chromatin architecture and status, exerting profound influence within the cellular milieu. CRs exert authoritative control over a spectrum of biological processes, including but not limited to cell proliferation, differentiation, DNA repair, and stress response [11–13]. Remarkably, these regulatory entities emerge as central figures in the genesis and progression of kidney renal clear cell carcinoma, where their perturbations serve as recurrent hallmarks in various human maladies [14]. While many facets concerning the precise roles of chromatin regulators in the context of kidney renal clear cell carcinoma remain enigmatic, extant investigations have delineated their significant contributions to the oncogenic pathways governing the initiation and progression of this malignancy. A comprehensive exploration of these regulators holds the potential to seed novel paradigms and strategies in the domain of therapeutic interventions and preventive measures against kidney renal clear cell carcinoma.

The burgeoning field of bioinformatics has gained substantial recognition for its contemporary utilization in the exploration of various tumor types. The primary aim of this study was to employ a comprehensive array of bioinformatics methodologies to unravel the intricate interrelationship between gene expression profiles associated with chromatin regulators and the prognostic landscape of Kidney Clear Cell Carcinoma (KIRC).

In this investigation, we harnessed machine learning algorithms to meticulously analyze the TCGA-KIRC dataset, resulting in the identification of eight distinctive signature genes. Subsequently, we conducted an independent assessment of their prognostic significance by integrating these signature genes with relevant clinical attributes to construct a robust prognostic model. Concurrently, our research encompassed a multifaceted spectrum of analyses. We conducted clinically relevant correlation assessments to enhance our understanding of intricate interrelationships. Additionally, we conducted enrichment analyses to gain insights into molecular pathways. Differential assessments of drug sensitivity were undertaken, shedding light on potential therapeutic avenues. We also investigated immune infiltration dynamics and interactions within the tumor microenvironment. Notably, our study culminated in a correlative predictive analysis of the tumor microenvironment, further enriching our insights. The overarching goal of our research was to elucidate the prognostic significance of chromatin regulators-related genes in KIRC. Through a comprehensive examination of genomic cohorts, our endeavor aimed to unveil their potential as pivotal prognostic determinants with significant implications for the advancement of innovative paradigms in the field of immunotherapeutic modalities.

2.1 Data sources

We collected data related to TCGA-KIRC patients from the TCGA database (https://portal.gdc.cancer.gov/), including transcriptomic data, clinical data, and mutation data. The set of chromatin regulatory factor-related genes [15] was obtained through a comprehensive review of the relevant literature. Subsequently, we identified genes associated with chromatin regulatory factors by analyzing TCGA-KIRC patient data and performed subsequent analyses. To process data related to TCGA-KIRC patients, including transcriptomic data, clinical data, and mutation data, we used Strawberry Perl software. In this investigation, we purposefully curated mRNA-related data for subsequent analysis. The clinical dataset comprised patient demographics, encompassing age and gender, alongside tumor characteristics, including grade, stage, and TNM staging information.

2.2 Screening of chromatin regulators-related differential genes and prognostic genes

We applied the R package "limma" to process the expression data files and chromatin gene list files so as to extract the expression profiles of chromatin regulatory factor genes. Subsequently, we performed differential analysis and gene expression analysis using the "limma" and "sva" packages. Notably, we set the logFCfilter to 0.585 and applied an fdr filter of 0.05 to identify and characterize differential genes. These findings are visualized by volcano maps. We used the "pheatmap" package to generate heatmaps to visualize the differential genes further. To select prognostic genes, we fully integrated expression data and survival data from the TCGA dataset and then performed survival analysis. Forest plots were then used to visualize the analysis results.

2.3 Development and validation of prognostic models

A prognostic model was developed by integrating the TCGA unidirectional dominant gene expression dataset. Prognostic features were determined by applying the least absolute shrinkage and selection operator (LASSO) for Cox regression analysis using the R packages "glmnet" and "survival". The risk score formula used was defined as follows: risk level = Σ(prognostic gene expression × Coef prognostic gene), where Coefi represents the correlation coefficient. In this study, KIRC patients were categorized into low-risk and high-risk groups based on the median risk score. To assess the predictive performance of the model for patient prognosis, the researchers plotted ROC curves, comparing the modeled risk with various clinical parameters. The impact of high- and low-risk features on patient prognosis was assessed by box plots, which showed multiple clinical indicators. Moreover, the contrast in prognosis among patients categorized as the elevated-risk and reduced-risk groups was confirmed through principal component analysis (PCA) and progression-free survival analysis (PFS). Notably, the survival curves of the experimental dataset performed robustly.

2.4 Nomogram construction

To assess the potential of risk scores as independent predictors and construct the corresponding Nomogram, we conducted both univariate and multivariate Cox regression analyses. In the KIRC cohort, the "rms" was employed to create a column chart that integrated risk scores with clinicopathological characteristics for the prediction of 1-, 3-, and 5-year survival outcomes. The robustness of the Nomogram in prognostic prediction was assessed by constructing ROC curves comparing the Nomogram's predictive performance with that of the modeled risk and various clinical parameters.

2.5 Functional enrichment analysis

Functional enrichment analysis of differentially expressed genes within the KIRC domain was undertaken to facilitate annotation and pathway enrichment investigations. Employing clusterProfiler, we performed an analysis of gene ontology (GO) and KEGG pathways to elucidate the functional roles and associations within the implicated genes, thereby establishing a groundwork for subsequent research pursuits.

2.6 Risk characterization for immunological analysis

For risk characterization through immuno-infiltration analysis, we harnessed a diverse set of algorithms, including XCELL [16], TIMER [17], QUANTISEQ [18], MCPCOUNT [19], EPIC [20], CIBERSORT [21], and CIBERSORT-ABS [22]. Furthermore, we conducted a TIDE analysis within both high and low-risk groups.

2.7 Screening and survival analysis of network core genes

We employed protein-protein interaction network (PPI) analysis to elucidate variations among different genes within the the elevated and reduced-hazard groups, aiming to identify interplays between these genes. Subsequently, the protein interactions between different genes were visualized through Cytoscape, with red indicating up-regulated genes and green indicating down-regulated genes. Further analysis involved the identification of the top 5 core genes within the network using the cytoHubba plugin. Survival analyses of these top 5 core genes were executed using the "limma," "survival," and "survminer" packages.

2.8.Prediction of targeted drug sensitivity and immunotherapy response

To forecast the responsiveness of KIRC patients to immunotherapy and the efficacy of frequently employed chemotherapeutic agents, we employed a variety of techniques. Notably, the R package "oncoPredict" facilitated the prediction of divergences in drug sensitivity among KIRC patients when exposed to common chemotherapeutic agents. This approach relied on the analysis of half-maximal inhibitory concentrations (IC50) derived from KIRC patients, sourced from the Genomics of Drug Sensitivity in Cancer (GDSC) database [23].

2.9.Cell Culture and Transient Transfection

Human kidney renal clear cell carcinoma (KIRC) cell lines, including 786-O, were cultured alongside the human renal proximal tubular epithelial cell line HK-2. The culture medium employed was Dulbecco's modified Eagle's medium (DMEM, GIBCO), supplemented with 10% fetal bovine serum (FBS; Hyclone), 100 U/L penicillin, and 100 mg/L streptomycin (Thermo Fisher). These cell cultures underwent incubation under conventional conditions at 37°C in the presence of 5% CO2. For transient transfection purposes, Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA) was utilized in strict accordance with the manufacturer's instructions. This transfection protocol involved the introduction of both Negative Control (NC) and BRD9 siRNA (RiboBio, Guangzhou, China) into colorectal carcinoma (CRC) cells.

2.10.qPCR

Total RNA extraction was performed employing the RNA Eazy Fast Tissue/Cell Kit (TIANGEN Biotech Co., Beijing), in strict adherence to the manufacturer's protocols. Subsequently, cDNA synthesis was conducted utilizing the FastKing RT Kit (TIANGEN Biotech Co., Beijing) according to the prescribed procedure. Real-time PCR analysis was executed employing the SuperReal PreMix Plus (TIANGEN Biotech Co., Beijing) reagent in conjunction with the StepOnePlus Real-Time PCR System. The PCR protocol commenced with an initial pre-denaturation phase at 95°C for 15 minutes, followed by a series of 40 cycles involving denaturation at 95°C for 10 seconds, annealing at 72°C for 20 seconds, and extension at 60°C for 20 seconds. Detailed primer sequences can be found in Supplementary Table 1.

Supplementary Table 1. Primers used in qRT-PCR analysis.

Gene	Sequences (5’-3’)
H-GAPDH-F	GAACGGGAAGCTCACTGG
H-GAPDH-R	GCCTGCTTCACCACCTTCT
H-BRD9-F	ACAGCCAGCCGAAAATGAGA
H-BRD9-R	TCCATGGGGATCTCTGAAGC

2.11.Transwell Assay

A total of 1×10^5 cells were plated, with either Matrigel-coated (BD Biosciences, San Jose, CA) chambers utilized for the invasion assay or uncoated chambers designated for the migration assay. In both cases, the upper chamber was filled with serum-free medium, while the lower chamber was filled with complete DMEM medium. Following a 24-hour incubation period, cells that had migrated or invaded through the membrane were meticulously fixed using 4% paraformaldehyde and subsequently stained with 0.1% crystal violet. The quantification of migrated or invaded cells was accomplished using a light microscope (Thermo Fisher, Waltham, MA, USA).

2.12.CCK-8 Assay

Cell viability was assessed via the Cell Counting Kit-8 (CCK-8) assay. Cells were seeded at a density of 1500 cells per well in 200 µl of complete medium within 96-well plates and maintained at 37°C. Subsequent to each experimental condition, 20 µl of CCK-8 reagent (Beyotime, Shanghai, China) was introduced into each well. Following an additional 2-hour incubation period, the optical density value (OD450nm) was quantified using a microplate reader.

2.13.Statistical analyses

For our statistical analyses, we harnessed R software version 4.2.3 in conjunction with Strawberry Perl version 5.30.0. Utilizing the LASSO Cox regression model, we crafted correlation profiles and evaluated their predictive efficacy by means of ROC curves. Differences in the proportions of tumor-infiltrating immune cells and immune function across groups were assessed using the Wilcox test. Statistical significance was ascertained for distinctions with p < 0.05 and a false discovery rate (FDR) of less than 0.05. The identification of the top 5 network core genes was followed by survival analysis within Cytoscape, employing the cytoHubba plugin, with a significance threshold established at p < 0.05. Data originating from the CCK-8 assay underwent analysis utilizing GraphPad Prism Software (version 8.3.0). The presentation of results includes means ± standard deviation (SD) from three independent experiments, and statistical significance was assessed using Student’s t-test. A p-value below 0.05 was considered indicative of statistical significance.

3.1.Screening for chromatin regulators-related differential genes and prognostic genes

In our endeavor to investigate chromatin regulator differential genes (CRFDGs) with unique expression profiles and prognostic relevance in KIRC patients, we commenced the study with differential expression analysis, contrasting control and disease cohorts, facilitated by the "limma". This initial phase yielded the identification of 219 CRFDGs meeting pre-established criteria (logFC filter = 0.585, fdr filter = 0.05). The expression profiles of these CRFDGs were further visualized through heatmaps and volcano plots (Fig. 1A, B), with the heatmap featuring the top 50 CRFDGs exhibiting the most substantial up-regulation and down-regulation in expression. Incorporating univariate Cox analysis, we identified CRFDGs that demonstrated a significant association with patients' overall survival (OS), ultimately pinpointing 106 chromatin regulators genes with noteworthy prognostic relevance (Fig. 1C). These findings collectively underscore the potential linkage of chromatin regulators genes to the processes of tumorigenesis and progression within KIRC.

3.2.Independent prognostic analyses using LASSO to select modeling genes and clinical features

To construct the prediction model, TCGA-KIRC patient samples served as the cohort for model development. Risk scores for each sample were computed by multiplying the expression levels of selected genes related to chromatin regulatory factors by their respective regression coefficients, followed by aggregation to derive the risk score for each sample. Subsequently, the samples were stratified into high-risk and low-risk groups based on the median value of the entire dataset, which corresponded to the training and testing sets, respectively. The LASSO machine learning algorithm was employed to discern signature genes associated with KIRC prognosis among chromatin regulatory factor-related genes. Using this algorithm, we conducted cross-validation to ascertain the minimum criteria for constructing the LASSO machine learning model, ultimately identifying 8 feature genes (Fig. 2A, B). Due to the strong association observed between our established risk model and unfavorable prognostic outcomes, we conducted both univariate and multivariate independent prognostic analyses, evaluating risk scores alongside common clinical features for each KIRC patient. These analyses unveiled that age (p < 0.001), grading (p < 0.001), staging (p < 0.001), and risk scores (p < 0.001) exhibited significant associations with KIRC prognosis (Fig. 2C). Subsequent multifactorial Cox analysis identified age (p < 0.001), grading (p = 0.029), staging (p < 0.001), and risk score (p < 0.001) as robust and independent prognostic predictors (p-value < 0.05) (Fig. 2D). Principal component analysis (PCA) depicted more pronounced disparities in model genes between high and low-risk cohorts in contrast to genes associated with chromatin regulators within the same cohorts (Fig. 2E, F). Model genes effectively stratified high- and low-risk patients into two relatively independent clusters, confirming the method's efficacy in discerning between these populations. This emphasizes the stability, robust generalizability, and predictive capabilities of the risk model. The areas under the ROC curves for the risk scores exceeded those for other clinical parameters (age, gender, grading, and staging), underscoring the enhanced reliability of the constructed model in predicting patient survival (Fig. 2G). Notably, one-year survival rates stood at 0.642, three-year survival rates at 0.681, and five-year survival rates at 0.706 for patients under this model (Fig. 2H). Validation using TCGA patient data yielded a p-value < 0.001, reaffirming the stability and reliability of our constructed model (Fig. 2I). Assessment of progression-free survival (PFS) revealed that the low-scoring group exhibited a more favorable prognostic outcome compared to the high-risk group (P < 0.001) (Fig. 2J).

3.3.Nomogram construction and validation

To improve the clinical relevance and robustness of our developed risk model, we generated a column-line plot that integrated N-stage, sex, M-stage, age, T-stage, risk score, staging, and grading as predictive factors for assessing 1-, 3-, and 5-year prognostic survival probabilities (Fig. 3A). Among these variables, the risk score exerted the most profound impact on the prediction of overall survival (OS), underlining the heightened accuracy of the risk model in forecasting KIRC prognosis. Calibration curves further demonstrated a remarkable alignment between the predicted and observed values for OS probability at 1-, 3-, and 5-year, thereby confirming the Nomogram plot's robust predictive and stability attributes (Fig. 3B). The area under the ROC curve for the Nomogram plot (AUC = 0.858) significantly outperformed the AUCs associated with individual factors, including risk score, age, gender, grading, and staging. This highlights the superior reliability of the constructed Nomogram plot in predicting patient survival (Fig. 3C). To further evaluate the Nomogram's potential as an independent prognostic factor, we conducted both univariate and multivariate independent prognostic analyses, juxtaposing risk scores with common clinical characteristics for each KIRC patient. In both univariate and multivariate Cox analyses, the Nomogram yielded a p-value of less than 0.001, confirming its reliability and independence in forecasting patient prognosis and survival (Fig. 3D-E).

3.4.Clinical Relevance Analysis of Chromatin Regulator-Related Gene Models

To explore potential variations in risk scores among patient cohorts sharing similar clinical characteristics and assess whether prognosis differed within distinct clinical subgroups, we categorized KIRC patients into seven distinct subgroups based on clinical parameters. These parameters encompassed age, gender, G-stage, M-stage, N-stage, clinical stage, and T-stage. Subsequently, we compared risk scores among patient groups with identical clinical features (Fig. 4A-G). Our analysis revealed that there were no statistically significant differences (P > 0.05) in risk scores based on age and gender (Fig. 4A, B). However, patient risk scores exhibited an increasing trend beyond the G1 stage (excluding the G1 stage) (Fig. 4C). In the context of the M stage, patients with M1 stage displayed higher risk scores than those with M0 stage (Fig. 4D). Similarly, within the N-stage, patients in Stage N1 exhibited elevated risk scores compared to those in Stage N0 (Fig. 4E). Concerning the clinical stage, patients had progressively higher risk scores following Stage 1 (excluding Stage 1) (Fig. 4F). Lastly, in terms of T-stage, patients at T1, T2, T3, and T4 showed sequentially higher risk scores (Fig. 4G). It's important to note that a higher risk score indicates a greater risk for the patient. Consequently, by analyzing the clinical relevance of the chromatin regulators model, risk assessment and prognostic studies can be more effectively conducted based on patients' clinical characteristics.

3.5.Prognostic marker-based enrichment analysis in KIRC patients

We conducted GO functional analysis and KEGG enrichment analysis to evaluate the differential genes distinguishing between the high-risk and low-risk groups, with the aim of elucidating associations between biological activities and signaling pathways with risk scores. We considered items as significantly enriched when meeting predefined thresholds of p-value < 0.05 and q-value < 0.05. This analysis yielded distinct enrichment outcomes between the high-risk and low-risk groups. In the context of GO enrichment analysis, the results revealed that biological processes (BP) encompassed functions such as positive regulation of lymphocyte activation and humoral immune response, among others. Cellular components (CC) predominantly included terms related to the immunoglobulin complex and the external side of the plasma membrane. Molecular function (MF) terms were primarily associated with antigen binding and immunoglobulin receptor binding (Fig. 5A-B). Regarding the KEGG enrichment analysis, the notable pathways included Cytokine-cytokine receptor interaction, Viral protein interaction with cytokine and cytokine receptor, Lipid metabolism, and atherosclerosis, among others (Fig. 5C-D). These findings shed light on the biological processes and pathways linked to risk scores, providing valuable insights into the molecular mechanisms underpinning KIRC prognosis.

3.6.Predictive analysis of chromatin regulators-related gene models associated with immune cell infiltration and TIDE

The tumor microenvironment (TME) intricately interacts with tumor cells, exerting an indirect impact on tumor proliferation, invasion, and metastasis [24]. In the tumor microenvironment (TME), the composition and distribution of tumor-infiltrating immune cells (TIICs) represent a pivotal element that bears substantial relevance to the processes of tumor development [25]. Recognizing the critical role of immune cells in immunotherapy, we prognosticated tumor patient outcomes by scrutinizing variations in immune cell infiltration between cohorts categorized as favorable and unfavorable risk groups. Notably, several immune cell types exhibited statistically significant differences between these risk categories, encompassing T cells CD8, T cells CD4 memory resting, T cells CD4 memory activated, T cells follicular helper, T cells regulatory (Tregs), Monocytes, Macrophages M0, Macrophages M1, Macrophages M2, Dendritic cells resting, Dendritic cells activated, Mast cells resting, and Eosinophils. Specifically, T cells CD8, T cells CD4 memory activated, T cells follicular helper, T cells regulatory (Tregs), and Macrophages M0 displayed heightened expression levels in the favorable-risk category, while T cells CD4 memory resting, Monocytes, Macrophages M1, Macrophages M2, Dendritic cells resting, Dendritic cells activated, Mast cells resting, and Eosinophils exhibited elevated expression in the unfavorable-risk category (Fig. 6A). Given the paramount significance of immune function, we conducted ssGSEA analysis to assess immune function disparities between the distinct risk categories. These analyses unveiled a significant upregulation in numerous immune function scores within the adverse-risk category, encompassing APC_co_stimulation, CCR, Check-point, Cytolytic_activity, HLA, and Inflammation-promotion, among others. However, intriguingly, the sole immune function exhibiting higher expression in the favorable-risk category compared to the adverse-risk category was Type_II_IFN_Response (Fig. 6B). To further explore the relationship between risk scores and immune cell infiltration, we harnessed seven distinct computational algorithms (XCELL, TIMER, QUANTISEQ, MCPCOUNTER, EPIC, CIBERSORT-ABS, and CIBERSORT). The majority of immune cell infiltrates demonstrated positive correlations with risk scores across these computational methods, with particular emphasis on T cell NK, T cell CD4 + Th1, T cell regulatory (Tregs), and T cell follicular helper (Fig. 6C). Immunotherapy stands as a cornerstone in contemporary tumor therapy. Consequently, an in-depth analysis of immunotherapy resistance (TIDE) was performed in both the unfavorable-risk and favorable-risk categories to offer valuable insights into tumor treatment prognosis. Notably, a substantial disparity in TIDE scores emerged between these categories, with the adverse-risk category displaying elevated TIDE scores, indicative of an increased potential for immunotherapy resistance and diminished therapeutic efficacy (Fig. 6D).

3.7.Screening of network core genes and survival analysis

To elucidate disparities in gene expression patterns between high and low-risk groups and explore their interactions, we conducted a protein interaction network analysis (PPI). The depicted Fig. illustrates nodes representing distinct genes, with connecting lines denoting interactions among these genes. The varied colors of connecting lines signify diverse pieces of evidence for gene interactions (Fig. 7A). Subsequently, we visualized these gene interactions using Cytoscape, with red indicating up-regulated genes and green representing down-regulated genes. Interestingly, all genes appeared to be up-regulated (Fig. 7B). Following this, we employed the cytoHubba plugin to identify the top 5 core genes within the network: CDC20, CDCA8, BUB1, AURKB, and CCNB2 (Fig. 7C). After identifying these top five core genes, we conducted survival analyses. Remarkably, the results of the survival analyses for all five genes indicated a superior prognosis in the low-risk group compared to the high-risk group, with all outcomes reaching statistical significance (P < 0.001) (Fig. 7D-H). These findings underscore the pivotal roles of these five genes in the prognosis of KIRC patients, and further in-depth investigations into these genes will be a focal point in our subsequent research.

3.8.Differential analysis of drug sensitivity related to chromatin regulators genes

Risk scores facilitate precise adjustment of drug dosages, enabling a more comprehensive assessment of the effectiveness of immunotherapy in KIRC patients. Among the nine immunotherapeutic agents employed in KIRC treatment, significant differences in drug sensitivity were evident between the high and low-risk groups (P < 0.05). Specifically, compared to the low-risk group, the high-risk group displayed lower IC50 values for all nine drugs, namely, 5-Fluorouracil, Afuresertib, AZ960, Cisplatin, Cyclophosphamide, Gemcitabine, Pictilisib, Ribociclib, and OSI-027. This observation signifies heightened sensitivity to these nine drugs in the high-risk group (Fig. 8A-I). Utilizing risk scores allows for a more profound understanding of immunotherapy responses in KIRC patients and facilitates the refinement of drug treatments with increased precision.

3.9.BRD9 plays a key role in stimulating the proliferation and migration of kidney renal clear cell carcinoma cells

We observed a significant upregulation of BRD9 expression in kidney renal clear cell carcinoma cells, consistent with our bioinformatics analysis results (Fig. 9A). To gain deeper insights into the potential role of BRD9 in kidney renal clear cell carcinoma, a series of in vitro experiments was conducted. Initially, the CCK-8 assay demonstrated a substantial reduction in cell proliferation following BRD9 silencing (Fig. 9B). Subsequent permeabilization experiments revealed that interference with BRD9 expression significantly impeded cell invasion and migration (Fig. 9C). These collective findings underscore the tumor-promoting properties of BRD9, with its elevated expression playing a pivotal role in enhancing the proliferation, invasion, and migration of kidney renal clear cell carcinoma cells.

Kidney renal clear cell carcinoma, characterized by its elevated incidence, challenging prognosis, and propensity for distant metastasis, particularly to pulmonary, osseous, and cerebral sites, remains a formidable malignancy [26]. The intricate orchestration of gene expression hinges on the dynamics of chromatin, a nucleoprotein complex of paramount significance. It governs interactions between regulatory elements and transcription factors, underscoring its centrality in cellular homeostasis. Notably, disruptions in chromatin regulators have been linked to the initiation of malignancies, emphasizing the pivotal role of chromatin in cancer pathogenesis [12, 27]. Consequently, chromatin regulators are plausible candidates for crucial involvement throughout tumorigenesis, progression, and metastatic dissemination. The current therapeutic arsenal for kidney renal clear cell carcinoma encompasses surgical resection, radiotherapy, and chemotherapy. Unfortunately, this malignancy exhibits robust resistance to these modalities, with recurrence and metastasis persisting even after the radical resection of the primary mass [28–30]. The limited efficacy of these interventions necessitates the exploration of alternative avenues, leading molecularly targeted therapies to emerge as a prominent choice for advanced kidney renal clear cell carcinoma [31]. The central aim of this investigation is to establish a prognostic model rooted in chromatin regulators-related genes. Such a model holds the potential for prognostic insight, unveiling tumorigenic mechanisms and laying the groundwork for novel therapeutic approaches, including immunotherapy, molecularly-targeted interventions, and pharmacotherapeutic strategies tailored to the nuances of kidney renal clear cell carcinoma.

In our investigation, we initiated by harnessing transcriptomic data and pertinent clinical information obtained from the TCGA repository, encompassing patients with Kidney Clear Cell Carcinoma (KIRC) alongside healthy controls. Employing a fusion of machine learning methodologies and chromatin regulator-associated genes as variables, we identified a set of eight discriminative genes. This ensemble subsequently underwent independent prognostic evaluation. By employing the median risk score as a threshold, patients were stratified into cohorts characterized by elevated and diminished risk levels. This stratification formed the basis for a multifaceted analytical exploration that included functional enrichment analysis, immunoassay profiling, and drug sensitivity assessments. Significantly, this analysis revealed substantial disparities between the cohorts with heightened and diminished risk, reinforcing our observations and affirming the clinical significance of the model we developed. Simultaneously, our inquiry encompassed the establishment of a Protein-Protein Interaction (PPI) network, leading to the identification of five pivotal core genes. Subsequent survival analyses underscored the pronounced significance of these genes in governing the prognosis of KIRC patients. These revelations provide a robust empirical foundation upon which clinicians can rely for informed decision-making pertaining to diagnosis, prognosis, and therapeutic strategies in the realm of KIRC.

The PPI network analysis in this investigation has unveiled five cardinal genes critically implicated in the prognosis of KIRC, specifically CDC20, CDCA8, BUB1, AURKB, and CCNB2. Insights into CDC20 indicate that its mutation disrupts cell division, hindering progression into later cell cycle phases and compromising chromosome segregation precision, a pivotal genomic locus for potential KIRC therapeutic interventions [32]. Notably, aberrant CDC20 expression is associated with the development and prognosis of various malignancies, including uroepithelial bladder cancer [33], pancreatic cancer [34], gastric cancer [35], lung cancer [36], and hepatocellular carcinoma [37]. CDCA8's functional role in cell division components makes it a crucial Fig. in cell cycle orchestration, exerting a profound influence on kidney renal clear cell carcinoma prognosis [38, 39]. Furthermore, elevated CDCA8 expression correlates with improved overall survival among bladder cancer patients, often serving as a biomarker that encapsulates TNM staging and overall survival in such patients [40]. Elevated CDCA8 expression is also a prominent feature in breast cancer, predicting an unfavorable prognosis and reduced survival prospects among affected patients [41]. Elucidating the role of BUB1, it emerges as a pivotal phosphorylation-dependent player in the spindle checkpoint, notably phosphorylated and activated in the presence of unattached chromosomes. This process safeguards chromosome segregation fidelity. Conversely, aberrant BUB1 phosphorylation coupled with mislocalization leads to dysfunctional mitotic checkpoint operations, as evidenced in Brca2-mutant thymic lymphomas [42, 43]. AURKB, an arbiter of mitotic regulation, holds a central position within cellular dynamics. Depletion of AURKB through dalcetrapib (AZD1152-HQPA) has demonstrated remarkable efficacy in mitigating neuroblastoma cell proliferation [44, 45]. Concurrently, CCNB2, which is relevant for prognosis, is associated with cell proliferation and immune infiltration, revealing its impact on hepatocellular carcinoma and gastric cancer [46, 47].

The intricacies of the tumor microenvironment (TME) wield a commanding influence over tumor progression, impacting vital facets such as proliferation, invasion, and metastasis. This influence is further characterized by a diverse assembly of infiltrating immune cells. A meticulous exploration into the interplay between immune cells and the TME offers a pathway toward unraveling the molecular underpinnings and providing novel dimensions for immunotherapeutic interventions, which are of paramount significance for advancing clinical outcomes [48]. In this study, we conducted a comprehensive evaluation of immune cell expression and their functional attributes within cohorts stratified into high and low-risk categories. The outcomes revealed a distinctive profile: notably, T cell subsets, including CD8-positive T cells, activated memory CD4-positive T cells, follicular helper T cells, and regulatory T cells (Tregs), exhibited heightened expression among individuals in the elevated-risk category. Conversely, individuals in the diminished-risk category displayed an increased prevalence of resting memory CD4-positive T cells. This discernment positions T cell populations as pivotal constituents within the high and low-risk categories, with heightened expression correlating with a more favorable prognostic trajectory [49]. The implications stemming from these meticulous analyses bear substantial potential for translation into clinical applications. Through these insights, the diagnostic, prognostic, and therapeutic landscape for tumor patients can be refined, empowering clinicians to devise judicious, tailored strategies that encapsulate precision and individualization.

While our prognostic model, founded on eight chromatin-regulated related genes, demonstrates commendable predictive efficacy for renal clear cell carcinoma (KIRC) patient prognosis, we must acknowledge certain inherent study limitations. Foremost among these constraints is our exclusive reliance on publicly available data from the TCGA database, which carries the potential for real-world bias. Consequently, a cautious interpretation of our findings is warranted. To mitigate this limitation, we made efforts to explore alternative transcriptomic resources; however, the limited availability of KIRC data compelled us to resort to experimental validation. To enhance the accuracy of our model, the accumulation of a substantial volume of clinical data concerning KIRC patients and additional information pertaining to their involvement in immunotherapeutic interventions is imperative. This expanded dataset will serve as a benchmark for rigorously assessing the accuracy and utility of our model, thereby fortifying its empirical efficacy. In conclusion, our study underscores the significant potential of our prognostic model, while also acknowledging the constraints associated with data availability and experimental validation. Collaborative efforts involving larger clinical cohorts and the integration of more comprehensive treatment data will enhance the credibility and applicability in enhancing the utility of this model.

To conclude, our study has effectively pinpointed distinct genes intricately associated with chromatin regulatory mechanisms. These findings introduce novel prognostic biomarkers and potential therapeutic targets for individuals with KIRC. Moreover, we have developed a predictive model based on these prognostic biomarkers, capable of precise prognosis for KIRC patients while providing insights into their immune status. This model holds the potential to improve the effectiveness of personalized systemic interventions for KIRC, representing a significant step toward optimized therapeutic strategies in this context.

Data Availability Statement

The datasets utilized in our investigation are accessible via the TCGA repository (http://cancergenome.nih.gov/). The unprocessed data can be obtained from jianguoyun at the following link: https://www.jianguoyun.com/p/DXwgQi4Qpd_3CxiytJsFIAA.

Authors' contributions

GW, JH, and GT conceptualized the study. GW, JH, HQC, YL, JP, LL, CL and GT wrote the draft. HQC, GT, and GW undertook the literature review and data collection. Data analysis and visualization were conducted by JH, GW, YL, JP, and GT. In vitro experiments were executed by GW, CL. The manuscript's final revision received contributions from JH, GW, and GT. All authors have reviewed and approved the final manuscript.

Conflict of Interest Statement

The authors declare that they conducted this study free from any current or potential commercial or financial affiliations that could create a conflict of interest.

Funding

This research was underwritten by the Luzhou Science and Technology Department's Applied Basic Research Program (No: 2022-JYJ-145), the foreign (border) high-end talent initiative of Sichuan Province Science and Technology Department (No: 2023JDGD0037), and YiBin Science and Technology Department Social Development Projects (No:2022SF004).Medical Scientific Research Project of Yibin Municipal Health Commission (No:2023YW026)

Acknowledgments:

The authors extend their gratitude to the TCGA project for its invaluable contributions.

Yang Q, Chu W, Yang W, et al. Identification of RNA Transcript Makers Associated With Prognosis of Kidney Renal Clear Cell Carcinoma by a Competing Endogenous RNA Network Analysis. Front Genet. 2020;11:540094.
Li X, Yu Q. PON1 hypermethylation is associated with progression of renal cell carcinoma. J Cell Mol Med. 2019;23(10):6646–57.
Mohd AB, Ghannam RA, Mohd OB, et al. Etiologies, Gross Appearance, Histopathological Patterns, Prognosis, and Best Treatments for Subtypes of Renal Carcinoma: An Educational Review. Cureus. 2022;14(12):e32338.
Liao Z, Wang D, Song N, Xu Y, Ge H, Peng Z. Prognosis of clear cell renal cell carcinoma patients stratified by age: A research relied on SEER database. Front Oncol. 2022;12:975779.
Kase AM, George DJ, Ramalingam S. Clear Cell Renal Cell Carcinoma: From Biology to Treatment. Cancers 2023, 15(3).
Colli LM, Machiela MJ, Myers TA, et al. Burden of Nonsynonymous Mutations among TCGA Cancers and Candidate Immune Checkpoint Inhibitor Responses. Cancer Res. 2016;76(13):3767–72.
Yang J, Wu SJ, Dai WT, et al. The human disease network in terms of dysfunctional regulatory mechanisms. Biol Direct. 2015;10:60.
Song XS, Fan B, Ma C, et al. [Clinical research on the correlations between type 2 diabetes mellitus and renal clear cell carcinoma]. Zhonghua wai ke za zhi [Chinese journal of surgery]. 2013;51(7):627–30.
Dimitrieva S, Schlapbach R, Rehrauer H. Prognostic value of cross-omics screening for kidney clear cell renal cancer survival. Biol Direct. 2016;11(1):68.
Zhang G, Chen X, Fang J, et al. Cuproptosis status affects treatment options about immunotherapy and targeted therapy for patients with kidney renal clear cell carcinoma. Front Immunol. 2022;13:954440.
Gonzalez-Perez A, Jene-Sanz A, Lopez-Bigas N. The mutational landscape of chromatin regulatory factors across 4,623 tumor samples. Genome Biol. 2013;14(9):r106.
Luo M, Ríos G, Sarnowski TJ, et al. Editorial: New Insights Into Mechanisms of Epigenetic Modifiers in Plant Growth and Development. Front Plant Sci. 2019;10:1661.
Malina S, Cizin D, Knowles DA. Deep mendelian randomization: Investigating the causal knowledge of genomic deep learning models. PLoS Comput Biol. 2022;18(10):e1009880.
Mashtalir N, Dao HT, Sankar A, et al. Chromatin landscape signals differentially dictate the activities of mSWI/SNF family complexes. Sci (New York NY). 2021;373(6552):306–15.
Lu J, Xu J, Li J, et al. FACER: comprehensive molecular and functional characterization of epigenetic chromatin regulators. Nucleic Acids Res. 2018;46(19):10019–33.
Aran D, Hu Z, Butte AJ. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 2017;18(1):220.
Li T, Fu J, Zeng Z, et al. TIMER2.0 for analysis of tumor-infiltrating immune cells. Nucleic Acids Res. 2020;48(W1):W509–w514.
Finotello F, Mayer C, Plattner C, et al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med. 2019;11(1):34.
Becht E, Giraldo NA, Lacroix L, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol. 2016;17(1):218.
Racle J, de Jonge K, Baumgaertner P et al. Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data. eLife 2017, 6.
Chen B, Khodadoust MS, Liu CL et al. Profiling Tumor Infiltrating Immune Cells with CIBERSORT. Methods in molecular biology (Clifton, NJ) 2018, 1711:243–259.
Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453–7.
Geeleher P, Cox NJ, Huang RS. Clinical drug response can be predicted using baseline gene expression levels and in vitro drug sensitivity in cell lines. Genome Biol. 2014;15(3):R47.
Chai R, Liao M, Ou L et al. Circadian Clock Genes Act as Diagnostic and Prognostic Biomarkers of Glioma: Clinic Implications for Chronotherapy. BioMed research international 2022, 2022:9774879.
Fan L, Li B, Li Z, et al. Identification of Autophagy Related circRNA-miRNA-mRNA-Subtypes Network With Radiotherapy Responses and Tumor Immune Microenvironment in Non-small Cell Lung Cancer. Front Genet. 2021;12:730003.
Meng T, Huang R, Zeng Z, et al. Identification of Prognostic and Metastatic Alternative Splicing Signatures in Kidney Renal Clear Cell Carcinoma. Front Bioeng Biotechnol. 2019;7:270.
Zullow HJ, Sankar A, Ingram DR, et al. The FUS::DDIT3 fusion oncoprotein inhibits BAF complex targeting and activity in myxoid liposarcoma. Mol Cell. 2022;82(9):1737–1750e1738.
Ni Y, Stingo FC, Ha MJ, et al. Bayesian Hierarchical Varying-sparsity Regression Models with Application to Cancer Proteogenomics. J Am Stat Assoc. 2019;114(525):48–60.
Qu G, Liu L, Yi L, et al. Prognostic prediction of clear cell renal cell carcinoma based on lipid metabolism-related lncRNA risk coefficient model. Front Genet. 2022;13:1040421.
Xiong X, Chen C, Yang J, et al. Characterization of the basement membrane in kidney renal clear cell carcinoma to guide clinical therapy. Front Oncol. 2022;12:1024956.
et al. Retrospective claims analysis of best supportive care costs and survival in a US metastatic renal cell population. ClinicoEconomics and outcomes research: CEOR 2013, 5:347–354.
Hartwell LH, Culotti J, Reid B. Genetic control of the cell-division cycle in yeast. I. Detection of mutants. Proc Natl Acad Sci USA. 1970;66(2):352–9.
Choi JW, Kim Y, Lee JH, et al. High expression of spindle assembly checkpoint proteins CDC20 and MAD2 is associated with poor prognosis in urothelial bladder cancer. Virchows Archiv: an international journal of pathology. 2013;463(5):681–7.
He S, Liu D, Chen Z. REC8 inhibits proliferation, migration and invasion of breast cancer cells by targeting CDC20. Mol Med Rep 2022, 26(1).
Wang Z, Wan L, Zhong J, et al. Cdc20: a potential novel therapeutic target for cancer treatment. Curr Pharm Design. 2013;19(18):3210–4.
Wang K, Zhang M, Wang J, et al. A Systematic Analysis Identifies Key Regulators Involved in Cell Proliferation and Potential Drugs for the Treatment of Human Lung Adenocarcinoma. Front Oncol. 2021;11:737152.
Li J, Gao JZ, Du JL, et al. Increased CDC20 expression is associated with development and progression of hepatocellular carcinoma. Int J Oncol. 2014;45(4):1547–55.
Gu Y, Lu L, Wu L, et al. Identification of prognostic genes in kidney renal clear cell carcinoma by RNAseq data analysis. Mol Med Rep. 2017;15(4):1661–7.
Naiman N, Fujioka K, Fujino M, et al. Repression of Interstitial Identity in Nephron Progenitor Cells by Pax2 Establishes the Nephron-Interstitium Boundary during Kidney Development. Dev Cell. 2017;41(4):349–365e343.
Li S, Liu X, Liu T, et al. Identification of Biomarkers Correlated with the TNM Staging and Overall Survival of Patients with Bladder Cancer. Front Physiol. 2017;8:947.
Phan NN, Wang CY, Li KL, et al. Distinct expression of CDCA3, CDCA5, and CDCA8 leads to shorter relapse free survival in breast cancer patient. Oncotarget. 2018;9(6):6977–92.
Chen RH. Phosphorylation and activation of Bub1 on unattached chromosomes facilitate the spindle checkpoint. EMBO J. 2004;23(15):3113–21.
Lee H. Impaired phosphorylation and mis-localization of Bub1 and BubR1 are responsible for the defective mitotic checkpoint function in Brca2-mutant thymic lymphomas. Exp Mol Med. 2003;35(5):448–53.
Dos Santos EO, Carneiro-Lobo TC, Aoki MN, et al. Aurora kinase targeting in lung cancer reduces KRAS-induced transformation. Mol Cancer. 2016;15:12.
Bogen D, Wei JS, Azorsa DO, et al. Aurora B kinase is a potent and selective target in MYCN-driven neuroblastoma. Oncotarget. 2015;6(34):35247–62.
Zou Y, Ruan S, Jin L, et al. CDK1, CCNB1, and CCNB2 are Prognostic Biomarkers and Correlated with Immune Infiltration in Hepatocellular Carcinoma. Med Sci monitor: Int Med J experimental Clin Res. 2020;26:e925289.
Shi Q, Wang W, Jia Z, et al. ISL1, a novel regulator of CCNB1, CCNB2 and c-MYC genes, promotes gastric cancer cell proliferation and tumor growth. Oncotarget. 2016;7(24):36489–500.
Zhao Z, Ding Y, Tran LJ, et al. Innovative breakthroughs facilitated by single-cell multi-omics: manipulating natural killer cell functionality correlates with a novel subcategory of melanoma cells. Front Immunol. 2023;14:1196892.
Hu B, Meng Y, Qu C, et al. Combining single-cell sequencing data to construct a prognostic signature to predict survival, immune microenvironment, and immunotherapy response in gastric cancer patients. Front Immunol. 2022;13:1018413.

No competing interests reported.

PCRresults.xlsx

Download PDF

Version 1

posted

You are reading this latest preprint version

Unravelling the complexity of kidney renal clear cell carcinoma prognosis: integrating chromatin regulators, gene signatures and associated immune landscapes

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1 Data sources

2.2 Screening of chromatin regulators-related differential genes and prognostic genes

2.3 Development and validation of prognostic models

2.4 Nomogram construction

2.5 Functional enrichment analysis

2.6 Risk characterization for immunological analysis

2.7 Screening and survival analysis of network core genes

2.8.Prediction of targeted drug sensitivity and immunotherapy response

2.9.Cell Culture and Transient Transfection

2.10.qPCR

2.11.Transwell Assay

2.12.CCK-8 Assay

2.13.Statistical analyses

3. Results

3.1.Screening for chromatin regulators-related differential genes and prognostic genes

3.2.Independent prognostic analyses using LASSO to select modeling genes and clinical features

3.3.Nomogram construction and validation

3.4.Clinical Relevance Analysis of Chromatin Regulator-Related Gene Models

3.5.Prognostic marker-based enrichment analysis in KIRC patients

3.6.Predictive analysis of chromatin regulators-related gene models associated with immune cell infiltration and TIDE

3.7.Screening of network core genes and survival analysis

3.8.Differential analysis of drug sensitivity related to chromatin regulators genes

4. Discussion

5. Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1