Identification of AKI signatures and classification patterns in ccRCC based on machine learning

Wang, Li; Peng, Fei; Li, Zhen Hua; Deng, Yu Fei; Ruan, Meng Na; Mao, Zhi Guo; Li, Lin

doi:10.3389/fmed.2023.1195678

ORIGINAL RESEARCH article

Front. Med., 24 May 2023

Sec. Precision Medicine

Volume 10 - 2023 | https://doi.org/10.3389/fmed.2023.1195678

This article is part of the Research Topic Novel molecular mechanisms and clinical strategies in solid tumor recurrence and metastasis: from bench to bedside View all 11 articles

Identification of AKI signatures and classification patterns in ccRCC based on machine learning

Li Wang¹^†

Fei Peng²^†

Zhen Hua Li³

Yu Fei Deng¹

Meng Na Ruan¹

Zhi Guo Mao¹^*

Lin Li¹^*

¹Department of Nephrology, Changzheng Hospital, Naval Medical University, Shanghai, China
²Department of Cardiology, Jinshan Hospital of Fudan University, Shanghai, China
³Department of Cardiology, Changzheng Hospital, Naval Medical University, Shanghai, China

Background: Acute kidney injury can be mitigated if detected early. There are limited biomarkers for predicting acute kidney injury (AKI). In this study, we used public databases with machine learning algorithms to identify novel biomarkers to predict AKI. In addition, the interaction between AKI and clear cell renal cell carcinoma (ccRCC) remain elusive.

Methods: Four public AKI datasets (GSE126805, GSE139061, GSE30718, and GSE90861) treated as discovery datasets and one (GSE43974) treated as a validation dataset were downloaded from the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) between AKI and normal kidney tissues were identified using the R package limma. Four machine learning algorithms were used to identify the novel AKI biomarkers. The correlations between the seven biomarkers and immune cells or their components were calculated using the R package ggcor. Furthermore, two distinct ccRCC subtypes with different prognoses and immune characteristics were identified and verified using seven novel biomarkers.

Results: Seven robust AKI signatures were identified using the four machine learning methods. The immune infiltration analysis revealed that the numbers of activated CD4 T cells, CD56^dim natural killer cells, eosinophils, mast cells, memory B cells, natural killer T cells, neutrophils, T follicular helper cells, and type 1 T helper cells were significantly higher in the AKI cluster. The nomogram for prediction of AKI risk demonstrated satisfactory discrimination with an Area Under the Curve (AUC) of 0.919 in the training set and 0.945 in the testing set. In addition, the calibration plot demonstrated few errors between the predicted and actual values. In a separate analysis, the immune components and cellular differences between the two ccRCC subtypes based on their AKI signatures were compared. Patients in the CS1 had better overall survival, progression-free survival, drug sensitivity, and survival probability.

Conclusion: Our study identified seven distinct AKI-related biomarkers based on four machine learning methods and proposed a nomogram for stratified AKI risk prediction. We also confirmed that AKI signatures were valuable for predicting ccRCC prognosis. The current work not only sheds light on the early prediction of AKI, but also provides new insights into the correlation between AKI and ccRCC.

1. Introduction

Acute kidney injury (AKI) is a complex clinical disorder that manifests as a rapid decrease in glomerular filtration rate within three months and an increase in serum creatinine levels (1). Several stimuli may contribute to the development of AKI. Ischemia–reperfusion injury is one of the major clinical challenges faced by clinicians, especially during the perioperative period of renal transplantation. In AKI patients, persistent renal dysfunction and irreversible nephron loss can lead to chronic kidney failure. In addition, the risk of cardiovascular problems and other complications increases over time in renal failure patients (2), and critically ill patients have even worse prognoses (3). Therefore, the prevention and early detection of AKI are of paramount importance, and it is essential to identify specific signatures in patients with AKI. However, although biomarkers for the detection of AKI such as cystatin C, liver-type fatty acid-binding protein, interleukin-18, and neutrophil gelatinase-associated lipocalin have begun to emerge, these have some limitations (4). A useful tool related to this type of research is machine learning, a field of artificial intelligence that uses computer systems to process data using complex mathematical algorithms. With its powerful algorithms, machine learning has recently been applied in the field of medicine (5, 6). However, few studies have used machine learning to identify novel AKI biomarkers.

Several lines of evidence suggest that AKI induces or promotes the occurrence and progression of renal cancer. For instance, kidney injury may trigger DNA damage and promote mutated cell clonal proliferation in different kidney compartments (7). In one study, Peired et al. demonstrated a correlation between AKI and the subsequent development of renal cancer by analyzing patients with AKI from multiple independent cohorts (8). In addition, previous studies have shown that renal cancer progression is based on the increased systemic production of various chemokines, cytokines, and inflammatory immune cells (9–11). In another study, Zhou et al. demonstrated that AKI induces a systemic inflammatory response through modulation by inhibitors of the CXCL1/CXCR2 axis, thus increasing the risk of clear cell renal cell carcinoma (ccRCC) formation (12). Moreover, several studies have suggested that renal carcinoma occurs after an AKI episode or after years of chronic kidney disease, as kidney injury is an initial factor in renal cancer (8, 13). Indeed, ccRCCs account for approximately 75% of renal carcinoma cases (14). However, there is no satisfactory AKI-based prognostic model for accurate risk stratification in these patients. Therefore, it is essential to identify specific classifications for ccRCC prognostic prediction at the AKI level.

In this study, public datasets on AKI and ccRCC were downloaded and analyzed. We identified AKI-related biomarkers using functional pathway analysis, immune analysis, and AKI risk prediction. In addition, we carried out an AKI-related signature study by classifying ccRCC patients and integrating biological function enrichment, immune infiltration, chemical drug sensitivity data, and prognostic analyses. Furthermore, reliable signature subtyping was performed to predict the prognosis of patients with ccRCC based on the crucial roles of novel AKI biomarkers. Notably, this is the first study to elucidate novel AKI-related biomarkers with prognostic predictive value in ccRCC.

2. Materials and methods

2.1. Data collection and processing

Five public AKI datasets from the Gene Expression Omnibus (GEO) were downloaded and processed: GSE126805 (n = 83; normal tissues = 41, AKI tissues = 42), GSE139061 (n = 48; normal tissues = 9, AKI tissues = 39), GSE30718 (n = 47; normal tissues = 19, AKI tissues = 28), GSE90861 (n = 46; normal tissues = 23, AKI tissues = 23), and GSE43974 (n = 391; normal tissues = 188, AKI tissues = 203). The former four datasets were integrated and the batch was removed to construct a training cohort, whereas GSE43974 was treated as an independent test cohort. GSE126805, GSE139061, GSE30718, and GSE90861 were used to test whether these biomarkers could distinguish AKI tissues from normal tissues. Approval and informed consent from the institutional review board were not required for the AKI cohorts from the public databases. To reveal the connection between AKI and ccRCC, ccRCC multi-omics information was retrieved from the GDC TCGA portal.

2.2. Batch effect removal

To remove batch effects derived from the study design, sequence platform, and technological replication, we filtered only normal and AKI tissue expression matrices and clinical characteristics from four cohorts: GSE126805, GSE139061, GSE30718, and GSE90861. The batch effect was removed using the default function in the sva package. A principal component analysis (PCA) was used to visualize the efficacy of the batch removal.

2.3. Differential expression and enrichment analysis

We used the limma package to identify differentially expressed genes (DEGs) between normal and AKI tissues in the merged expression matrix. DEGs were identified using a significance threshold of p < 0.05 and an absolute log-fold change >1.5. An enrichment analysis of the DEGs was performed using the clusterProfiler and ggpplot2 packages, and these were visualized using an enrichment plot. Additionally, Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and gene set enrichment analyses (GSEA) were performed to better understand the biological roles of these DEGs. The significantly differentially expressed terms or pathways were identified by a threshold value of p <0.05 and q-value <0.05.

2.4. Identification and verification of AKI-related biomarkers

After identifying the DEGs from the combined expression profile of the combined expression matrix, we sought to identify the most relevant AKI-related biomarkers. We adopted four machine-learning algorithms, including Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression, Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and support vector machine (SVM), to select reliable biomarkers to distinguish AKI from normal tissues. The detailed parameters adopted for these were as recommended in each of the R packages. The detailed parameters of LASSO logistics were as follow: alpha = 1, maximum number of iterations = 8,000, tol = 1e-4). The detailed parameters of Random Forest were as follow: n_estimators = 80, criterion = gini, min_samples_split = 2, min_samples_leaf = 1. And for XGBoost, the detailed hyperparameters were as follow: n_estimators = 500, max_depth = 6, reg_alpha = 0, colsample_bytree = 1. Finally, the detailed parameters of SVM were as follow: tol = 1e-3, max_iter = −1. Furthermore, the receiver operating characteristic (ROC) curve was adopted to further evaluate the accuracy of the filtered biomarkers in the training and testing cohorts using the R package pROC. The GSE90861 dataset was used as a validation cohort to verify the specificity and sensitivity of the biomarkers retrieved from the test cohort.

2.5. Immune components and cell differences between AKI and normal tissues

Single-sample gene set enrichment analysis (ssGSEA) and bulk sequence-based deconvolution algorithms from the GSVA and ESTIMATE packages were used to detect different immune cells and components between normal and AKI tissues. The gene sets used for ssGSEA consisted of 28 types of immune cells, including activated B cells, activated CD4 T cells, activated CD8 T cells, activated dendritic cells, CD56^bright natural killer cells, CD56^dim natural killer cells, central memory CD4 T cells, central memory CD8 T cells, effector memory CD4 T cells, effector memory CD8 T cells, eosinophils, gamma delta T cells, immature B cells, immature dendritic cells, macrophages, mast cells, MDSCs, memory B cells, monocytes, natural killer cells, natural killer T cells, neutrophils, plasmacytoid dendritic cells, regulatory T cells, T follicular helper cells, type 1 T helper cells, type 17 T helper cells, and type 2 T helper cells. The estimated algorithms contained three scores: ESTIMATE, immune, and stromal scores. The correlations between the seven biomarkers and immune cells or their components were calculated using the R package ggcor.

2.6. Identification of different subtypes of AKI and ccRCC

To further reveal the heterogeneity of AKI and the potential relationship between AKI and ccRCC, we performed an unsupervised cluster analysis based on seven novel biomarkers (CCNL1, NFKBIZ, HBB, TRIB1, SOCS3, HSPA6, and EGR1) using the Consensus Cluster Plus package. The optimal cluster number was identified based on the PCA algorithm and cumulative distribution function curve. Differences in estimates, immune scores, and immune components were also qualified and compared between the ccRCC subtypes. A detailed description of the parameters employed are described in previous studies (15, 16).

3. Results

3.1. Batch effects removal and data integration

Figure 1A shows a flowchart of our study. Figure 1B shows the raw PCA results for the four microarray databases (GSE126805, GSE139061, GSE30718, and GSE90861). The four different colors represent different datasets. Every dataset was discrete without any intersection. Figure 1C shows the combat PCA results following batch removal.

FIGURE 1

Figure 1. Differentially expressed genes between AKI and normal tissues. (A) Workflow of our study. (B) PCA plot before batch effect removal. (C) PCA after batch effect removal. PCA, principal component analysis.

3.2. Functional pathway enrichment analysis

A volcano map (Figure 2A) was created to visually display the expression changes of the DEGs between normal and AKI tissues. A total of 108 genes were identified as DEGs according to adjusted p-values <0.05 and logFC cutoffs >1.5; among these, 101 genes were upregulated and 7 genes were downregulated. Next, we performed GO and KEGG enrichment analyses. As depicted in Figures 2B–D, in terms of biological processes (BP), the DEGs were mostly associated with lipopolysaccharides and molecules of bacterial origin. Regarding cellular components (CC), these were mainly involved in the RNA polymerase II transcription regulator complex and transcription regulator complex. In terms of molecular functions (MF), the DEGs were largely associated with tyrosine proteins, threonine phosphatase activity, and MAP kinases. The 20 most significant KEGG pathway terms are shown in Figure 2E. These genes were primarily associated with the TNF signaling pathway, the IL-17 signaling pathway, lipid and atherosclerosis pathways, the AGE − RAGE signaling pathway in those with diabetic complications, and the MAPK signaling pathway.

FIGURE 2

Figure 2. GO/KEGG/GSEA analysis of common differentially expressed genes (DEGs). (A) Volcano plot of the DEGs. (B) BP identified through GO analysis of the DEGs. (C) CC identified through GO analysis of the DEGs. (D) MF identified through GO analysis of the DEGs. (E) KEGG analysis of the differentially expressed signatures. (F–K) GSEA plot showing the most enriched gene sets based on KEGG analysis of all detected genes. BP, biological process; CC, cellular component; MF, molecular function; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; GSEA, gene set enrichment analysis; FC, fold change.

3.3. GSEA analysis of the AKI and normal groups

GSEA was performed to determine which KEGG pathways were differentially enriched between the AKI and normal groups according to the NES and value of p (p < 0.01) criteria. As presented in Figures 2F–K, the top three significantly upregulated KEGG pathways enriched in the AKI group were pathways related to cytokine-cytokine receptor interactions, IL-17 signaling, and TNF signaling. In addition, oxidative phosphorylation, fatty acid degradation, and peroxisomes were significantly downregulated in the AKI tissues. These data reveal that numerous pathways may be directly or indirectly involved in the regulation of AKI.

3.4. AKI signature identification and validation

Four machine learning algorithms were applied to identify feature genes: LASSO regression (Figures 3A,B), SVM (Figure 3C), and Random Forest (Figures 3D,E). These were combined with a feature selection analysis to determine the top 30 genes with relative importance, and an XGBoost analysis was then performed using these 30 relatively important genes (Figure 3F). A Venn diagram was used to display the overlapping signatures of the four methods (Figure 4G). After applying the four machine learning algorithms, eight signatures were selected, as shown in Figures 3A–G: tribbles homolog 1 (TRIB1), ST6GALNAC3, suppressors of cytokine signaling 3 (SOCS3), NF-κB inhibitor ζ (NFKBIZ), heat shock protein family A 6 (HSPA6), hemoglobin subunit beta (HBB), early growth response 1 (EGR1), and cyclin L1 (CCNL1).

FIGURE 3

Figure 3. AKI signature identification and verification. (A,B) LASSO. (C) SVM. (D,E) RF 2. (F) XGBoost. (G) Venn diagram for the four machine learning algorithms. (H) ROC curves for the eight genes in the training cohorts. (I) Testing cohorts. LASSO, Least Absolute Shrinkage and Selection Operator; SVM, support vector machine; RF, Random Forest; XGBoost, Xtreme Gradient Boosting; ROC, receiver operating characteristic.

FIGURE 4

Figure 4. Immune cell infiltration rates and the correlations among these and the signatures. (A–C) ESTIMATE, immune, and stromal score differences between the ccRCC subtypes. (D) Correlations between ESTIMATE scores and significant genes. (E) Enriched signatures in the immune cells. (F) Correlations among the seven signatures according to Pearson and Spearman analyses. (G) Boxplot of immune cell infiltration between the normal and AKI clusters. ns > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ns, not significant.

As shown in Figures 3H,I, to further test the diagnostic efficacy of the above genes, ROC analysis was performed to evaluate the accuracy of each gene. The ROC curve showed that the seven biomarkers identified based on the logistic regression were reliable, with the following AUCs in the training set: CCNL1, 0.972; NFKBIZ, 0.937; HBB, 0.726; TRIB1, 0.908; SOCS3, 0.928; HSPA6, 0.905; and EGR1, 0.981. In the validation set, the AUCs were as follows: CCNL1, 0.825; NFKBIZ, 0.868; HBB, 0.821; TRIB1, 0.817; SOCS3, 0.826; HSPA6, 0.781; and EGR1, 0.814. ST6GALNAC3 (AUC = 0.517 in the test set) was not analyzed further in the next workflow.

3.5. AKI marker genes and immune cell infiltration

We analyzed the heterogeneous compositions of the tumor microenvironments (TMEs) in the AKI and normal groups. As depicted in Figures 4A–C, all estimated scores, including the immune, stromal, and estimated scores, were lower in the AKI group compared to the normal group. We then examined the specific correlations between the estimated scores and each of the signatures using Spearman’s correlation analyses, which revealed that TRIB1, EGR1, and CCNL1 were related to the estimate and stromal scores (Figure 4D). As shown in Figure 4E, almost all of these genes were positively or negatively correlated with immune cell infiltration. Of these, EGR1 and eosinophil, HSPA6 and mast cells were significantly and positively relevant. In contrast, TRIB1 and activated B cells, CCNL1 and central memory CD8 T cells were negatively correlated (p < 0.01).

As shown in Figure 4F, the hub gene correlation analysis revealed that the expression levels of CCNL1 and NFKBIZ were mostly positively correlated, and CCNL1 was linked to every other gene. Figure 4G shows that 28 immune cells were differentially concentrated between the AKI and control groups. Among these, activated CD4 T cells, CD56^dim natural killer cells, eosinophils, mast cells, memory B cells, natural killer T cells, neutrophils, T follicular helper cells, and type 1 T helper cells were more enriched in the AKI cluster.

3.6. Construction of the nomogram model

In order to accurately predict AKI risk using these signatures, a nomogram was constructed. As shown in Figure 5A, in the nomogram, each biomarker gene had a parallel score within the range of 0–100 points in terms of its association with the risk of AKI. As shown in Figures 5B–E, the AUC value was 0.919 in the training set and 0.945 in the testing set. In both the training and validation sets, the calibration plot demonstrated a good fit of the constructed nomogram, indicating satisfactory model accuracy.

FIGURE 5

Figure 5. Nomogram for predicting the risk of AKI. (A) Nomogram. (B,C) ROC curves for the sensitivity of the nomogram and the calibration curves in the training set. (D,E) ROC curves and calibration curves in the validation set.

3.7. ccRCC subtypes with the AKI-related gene classification

We investigated the expression patterns of AKI-related signatures in ccRCC to comprehensively explore the association between AKI and cancer. TCGA-ccRCC samples were classified into different molecular subtypes according to AKI signature expression levels. For this, we divided the optimal cluster numbers using an unsupervised clustering method. As depicted in Figures 6A–C, considering the relative change in area under cumulative distribution function curve and consensus index, we chose k = 2 as the optimal cluster number. The TCGA ccRCC dataset was classified into two optimal clusters: CS1 and CS2. Compared to CS2 patients, CS1 patients had better overall survival (OS) (p = 0.0026, log-rank test) and progression-free survival (PFS) (p = 0.00033, log-rank test) (Figures 6D,E). In addition, we analyzed the expression levels of AKI genes in normal tissues and in the two ccRCC subtypes using a heatmap. CCNL1, NFKBIZ, SOCS3, and HSPA6 were significantly upregulated in the ccRCC subtype compared to normal tissues, whereas TRIB1 and EGR1 were significantly upregulated in the ccRCC subtypes (p < 0.05; Figure 6F).

FIGURE 6

Figure 6. Establishment of two clusters based on AKI signatures in ccRCC. (A) Consensus cluster matrix of TCGA-ccRCC samples when k = 2. (B) Relative change of delta area under different cluster number. (C) Cumulative distribution function curves, k = 2 to 9. (D,E) Survival analysis for overall survival (OS) and progression-free survival (PFS) among the two subtypes in the TCGA-ccRCC dataset. (F) Expression profiles of AKI biomarker genes among the two subtypes and normal tissues. ns > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ns, not significant.

3.8. Gene expression profiles of the ccRCC subtypes and results of the function enrichment analysis

Considering the varying characteristics of each cluster, we identified the gene expression profiles of CS1 and CS2. Supplementary Figure S1A presents an enhanced volcano map that displays the differential gene expression between CS1 and CS2, with 19,168 total differentially expressed genes. In addition, the GO analysis demonstrated that DEGs between ccRCC subtypes were annotated in receptor antagonist, receptor inhibitor, peptidase inhibitor, peptidase regulator, and enzyme inhibitor activity in biological process (Supplementary Figure S1B).

Supplementary Figure S1C shows the results of the GSVA algorithm analysis, which demonstrates that G2M-checkpoint, TGFβ signaling, and E2F oncogenes were more enriched in CS2 than in CS1. In addition, regulon analysis was used to detect transcriptome differences, which demonstrated that HNF1A, EPAS1, ZEB2, TFE3, and TP53 were downregulated in CS2, whereas FOXE1 and TBX18 were upregulated in CS1 (Supplementary Figure S1D).

3.9. Comparison of immune infiltration between the two ccRCC subtypes

To further elucidate the molecular distinctions between the two ccRCC subtypes, we conducted a comprehensive analysis of the immune-related signatures, signaling pathways, anticancer responses, and other relevant factors in both subtypes. We found that the dysfunction scores and Tumor Immune Dysfunction and Exclusion (TIDE) scores in CS1 were significantly lower than those in CS2 (Figures 7A,B), which indicated lower tumor immune evasion and resistance to checkpoint blockage in this cluster. In addition, the expression levels of nine immune checkpoint inhibitor genes were compared between the CS1 and CS2 cells; among them, CD274 was significantly upregulated in CS1 cells, whereas LAG3 and PDCD1 were significantly downregulated in CS2 cells (Figure 7C).

FIGURE 7

Figure 7. Landscapes of specific immune function scores and immune components. (A,B) Immune function scores (Dysfunction and TIDE) among the two ccRCC subtypes. (C–E) Immune antigens, immune pathways, and anti-cancer steps among the two subtypes. ns > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ns, not significant.

Among the different immune-related signatures expressed between the two ccRCC subtypes in the heatmap (Supplementary Figure S2A), chemokines, chemokine receptors, major histocompatibility complexes, and immunoinhibitory and immunostimulatory signatures were significantly differentially expressed. The CS1 subtype showed lower expression of CXCL11, CCR4, HLA-DOA, CD244, and IL2RA. Moreover, CS1 showed higher immune cell infiltration than CS2 in the TME infiltration cell-type heatmap (Supplementary Figure S2B). Mismatch repair, nucleotide excision repair, and base excision repair signatures were significantly higher in CS1, whereas EMT1, EMT2, signatures were significantly lower this cluster (Figure 7D).

As the anticancer immune response can be interpreted as a series of immune cell combat processes, we evaluated the tumor immunophenotypes in the two clusters. As shown in Figure 7E, the CS2 cluster demonstrated higher activity associated with priming and activation (step3), T cell recruiting (step4), dendritic cell recruiting (step4), macrophage recruiting (step4), B cell recruiting (step4), Th2 cell recruiting (step4) and recognition of cancer cells by T cell (step6). In contrast, CS1 demonstrated higher activity associated with the release of cancer antigen (step1) and cancer antigen presentation activity (step2) (p < 0.05).

3.10. Drug sensitivity and genomic mutation analysis of the two subtypes

To evaluate the drug responses of the ccRCC subtypes, the estimated IC50 data for each drug were collected from the GDSC database to explore the potential chemotherapy sensitivity of ccRCC. We found that the CS1 subtype was more sensitive to sunitinib, pazopanib, crizotinib, erlotinib, temsirolimus, and axitinib, whereas the CS2 subtype was more sensitive to lisitinib and gefitinib (Figure 8A). A significant difference was also observed in other omics. For instance, the tumor mutation burden was higher in CS2 (Figure 8B), which may have led to poor prognoses in CS2. In addition, Figure 8C shows that CS2 had higher amounts of copy number alterations, copy number losses, and copy number-gained genomes compared to CS1 (p < 0.001).

FIGURE 8

Figure 8. Drug sensitivity analysis and other omics of the two ccRCC subtypes. (A) Estimated IC50 of the indicated molecular-targeted drugs among the two ccRCC subtypes. (B) Tumor mutation burden between the two subtypes. (C) Bar plot of the genomic fractions that were altered between the two subtypes. ns > 0.05, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. ns, not significant.

3.11. Subtype classification verification using an external dataset

To further confirm the credibility of the ccRCC subtypes, the nearest template prediction algorithms were applied to reconstruct the subtypes in the TCGA–KIRC cohort and divide this cohort into two different subgroups. As shown in Figures 9A–D, the four Kaplan–Meier curves created using the data from different databases revealed that CS1 had a significantly better survival probability than CS2. In the Cancer Cell database, the value of p was 0.012; in the GSE22541 database, the value of p was 0.038; in the CheckMate OS database, the value of p was 0.015; and in the E-MTAB-3267 database, the value of p was 0.046. The stability and reliability of the subtypes were verified based on these results.

FIGURE 9

Figure 9. Subtype classification verification using an external dataset. (A–D) Kaplan–Meier curve of the NTP (nearest template prediction) by the Cancer Cell, Checkmate OS, GSE22541, and E-MTAB-3267 databases.

4. Discussion

AKI is a global concern and challenge associated with high mortality and healthcare costs worldwide (17). It can be classified into pre-renal, intrinsically renal, or post-renal types. Meanwhile, ischemia-reperfusion, a pre-renal injury, is an inevitable event in kidney transplantation surgery (18). In addition, various components of the innate and adaptive immune systems are involved in AKI pathogenesis and repair. After ischemia and reperfusion, free radical-mediated injury releases pro-inflammatory cytokines and induces innate immunity (18, 19). Following this, immune cells traverse the postischemic kidney and demonstrate changes in the associated signaling pathways, and ischemic kidney injury is also established over time (20).

Kidney function biomarkers have emerged as new tools for risk assessment and clinical therapy guidance. These biomarkers are of particular interest because several microarray studies have demonstrated differential expression of single molecules or hub genes, which are believed to play a role in the development of certain incentive-induced AKI models (21–25). In this study, we screened early-stage AKI profiles derived from renal transplant and ordinary kidney biopsy tissues from the GEO database, conducted comprehensive analyses of four GEO datasets, and obtained seven genes from four machine learning models for ischemia–reperfusion injury AKI prediction. Moreover, we integrated the identified DEGs to identify potential biomarkers and pathways involved in AKI pathogenesis. In the GO database, protein tyrosine/threonine phosphatase activity, MAP kinase tyrosine/serine/threonine phosphatase activity, and MAP kinase phosphatase activity were significant associated with AKI. Moreover, the GSEA revealed that the cytokine-cytokine receptor interaction, IL-17, and TNF signaling pathways were positively upregulated in AKI.

Numerous studies have shown that tyrosine phosphatase and MAP kinase activities have essential physiological and pathological functions in kidney diseases (26, 27). Previous studies have also suggested that IL-17 mediates renal fibrosis and neutrophil infiltration in ischemia reperfusion (28, 29). In addition, the TNF pathway was reported to participate in renal glomerular endothelial injury (30). To clarify the differential enrichment of immune cells and elucidate their possible mechanisms of action in AKI, we analyzed marker genes associated with renal immunity. Our analysis revealed significant differences in immune, stromal, and estimated scores between the AKI and normal clusters. Additionally, immune cells, including activated CD4 T cells, CD56^dim natural killer cells, eosinophils, mast cells, memory B cells, natural killer T cells, neutrophils, T follicular helper cells, and type 1 T helper cells were significantly upregulated in the AKI group. Consistent with previous studies, T cells are known to release proinflammatory cytokines and chemokines to promote acute renal injury, whereas natural killer T cells can induce renal inflammation and injury or protect against it, depending on the context of their activation (31, 32). Furthermore, to determine the exact AKI risk-prediction efficiency, a highly accurate predictive nomogram was constructed by integrating a seven-gene risk score. Satisfactory agreement was observed in the calibration plot of the nomogram, suggesting that our seven-gene-based risk nomogram may contribute to predicting the risk of AKI and provide early treatment guidance beyond single conventional clinical parameters.

Facilitated by bioinformatics and sequencing technology, seven genes (EGR1, SOCS3, TRIB1, CCNL1, HBB, HSPA6, and NFKBIZ) were identified as precise indicators of AKI. Notably, earlier studies have confirmed that these genes participate in the renal physiology and pathology of AKI. Early growth response 1 (EGR1) is an early transcription factor that is induced by many cellular factors, including growth factors and hypoxia (33). Previous studies have validated that EGR1 is associated with mediating renal epithelial cell regeneration and is significantly upregulated in AKI kidney samples (34, 35). EGR1 also regulates inflammation and fibrosis in kidney tissues (36). And upregulation of SOCS3 in stressed proximal tubules also plays an important role in AKI by modulating the macrophage phenotype and inhibiting reparative proliferation (37). TRIB1 (38) might be a therapeutic target for renal injury through the regulation of renal tubular cell proliferation. Gunther et al. reported an increase in blood-derived CCNL1 expression in kidney transplant recipients with acute rejection compared to recipients with no rejection (39). Other studies have indicated that HBB, HSPA6, and NFKBIZ may play essential roles in kidney disease (40–42). The roles of these genes in ccRCC also have been reported. For instance, EGR1 serves as an independent prognostic factor in ccRCC patients by inhibiting the proliferation, invasion, and metastasis of ccRCC (43). In addition, Tomita et al. found that SOCS3 might be a potential target in IFN-α-resistant RCC treatment (44). Furthermore, studies have reported that NFKBIZ and HSPA6 are associated with ccRCC (42, 45).

Inflammatory microenvironments often develop prior to malignant changes and tumorigenesis (46). A previous study indicated that AKI induces malignant renal cell carcinoma via CXCR2 in the proximal tubular kidney epithelial cells of mice (12). These findings highlight the potential of using these seven genes (CCNL1, NFKBIZ, HBB, TRIB1, SOCS3, HSPA6, EGR1) in guiding clinical decision-making and the early prevention of AKI. In order to determine if certain AKI signatures have predictive value for ccRCC, we divided the ccRCC patient data into two clusters based on the key AKI genes. We found that CS1 had better prognostic value compared to CS2 in terms of OS and PFS. In addition, the expression profile of EGR1 was more enriched in normal tissues than in CS1 and CS2. This suggests that EGR1 may inhibit the proliferation, invasion, and metastasis of ccRCC, which is consistent with previous results (43). CCNL1, SOCS3, and HSPA6, which are highly expressed in patients with ccRCC, may promote tumor generation. Immune-related signatures were highly expressed in CS2 cells, whereas immune-related inhibitors were dysregulated. Another observation was that oncogenes related to G2M-checkpoint, TGFβ signaling, and E2F targets were more enriched in CS2. The higher tumor mutation burden and fraction of genome alerts may also explain the poor prognosis of CS2. Furthermore, the CS1 subtype was more sensitive to drug therapy. Thus, this subtype classification will contribute to ccRCC prognostic analysis, the early prevention of carcinogenesis, and guidelines for clinical immunotherapy.

Machine learning, which benefits from artificial intelligence and computer science, is widely used in nephrology research (47). Chen et al. used a XGBoost model to predict progression at 5 years in patients with biopsy-proven IgA nephropathy (48). For other applications, several studies have utilized machine learning to predict the development of acute kidney injury following surgery and volume responsiveness in patients with oliguric acute kidney injury (49–51). Recently, shared gene signatures and molecular mechanisms have been demonstrated between these two diseases (52, 53). This is relevant because shared genes are important for disease prevention and early treatment. In contrast to other studies, we obtained AKI signatures using machine learning and elaborated on the value of these biomarkers for predicting AKI risk and ccRCC prognosis. However, our study does have some limitations, as most of our results were based on bioinformatics analyses. Therefore, follow-up experiments should be conducted in AKI and ccRCC, and the specific roles of these seven signatures require further investigation in future molecular experiments.

5. Conclusion

In conclusion, a comprehensive bioinformatics analysis of AKI was conducted, and seven genes (CCNL1, NFKBIZ, HBB, TRIB1, SOCS3, HSPA6, and EGR1) that may be involved in the biological processes of AKI were identified. We also explored the immune correlations associated with these seven signatures and evaluated their risk associations. Our analysis showed that CS1, along with the AKI signature cluster, may lead to better outcomes and treatment sensitivity for ccRCC. These findings provide potential targets for risk prediction and offer a new perspective on the relationship between kidney injury and kidney cancer.

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material.

Author contributions

LW designed and wrote this manuscript. FP analyzed the data. ZL, MR, and YD collected the data. ZM and LL supervised the manuscript. All authors contributed to the study and approved the submitted version.

Funding

This study was supported by the Youth Scientific Research Project of Shanghai Municipal Health Commission [grant number: 20214Y0367] and the Youth Startup Fund of Jinshan Hospital of Fudan University [grant number: JYQN-LC-202007].

Acknowledgments

The authors would like to thank the five GEO dataset contributors for generously sharing their precious data.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmed.2023.1195678/full#supplementary-material

SUPPLEMENTARY FIGURE S1 | Function enrichment analysis between the CS1 and CS2 subgroups. (A) Enhanced volcano map of the differentially expressed genes in the two clusters. (B) GO enrichment analysis. (C) GSVA analysis of the differential pathways between the two ccRCC subtypes. (D) Regulon scores of different transcriptional factors. Yellow represents active expression of the transcription factors. Blue represents repressed expression of the transcription factors.

SUPPLEMENTARY FIGURE S2 | Immune landscapes of the two ccRCC subtypes. (A) Heatmap of the immune signatures between CS1 and CS2, including chemokine, chemokine receptor, MHC, immunoinhibitory, and immunostimulatory signatures. (B) Heatmap of tumor-infiltrating immune cells between CS1 and CS2 based on the TIMER, CIBERSORT, CIBERSORT-ABS, quanTIseq, MCP-counter, xCell, and EPIC algorithms. MHC, major histocompatibility complex.

References

1. Levey, AS, and James, MT. Acute kidney injury. Ann Intern Med. (2017) 167:ITC66-ITC80. doi: 10.7326/AITC201711070