Introduction: Vesicular transport (VT) has a complex relationship with tumor progression and immunity. But prognostic significance of VT in clear cell renal cell carcinoma (ccRCC) is unclear. Thus, we aimed to establish a prognostic model according to VT to predict overall survival of ccRCC patients. Methods: We used patient data from TCGA database and built a prognostic model with 13 VT-related genes (VTRGs) by differential expression analysis, LASSO regression, and univariate/multivariate Cox analysis. The model was validated internally and externally, and survival analysis and ROC curves depicted excellent predictive ability. Furthermore, higher modeled riskscores corresponded to more advanced tumor progression. To further understand the potential reasons for different prognoses in patients, we did enrichment analysis on differentially expressed genes identified by the model in risk groups. The expression levels and roles of SAA1 and KIF18B in ccRCC were verified by qRT-PCR and cell function experiments. Results: Humoral immune response and cAMP signaling pathway may be the biological processes and pathways leading to poor prognosis. Further analysis of immune microenvironment presented that ccRCC patients with poor prognoses had highly immune-infiltrated characteristics. We compared the drug response data of samples from different prognostic patients in the GDSC database and identified drug sensitivity differences associated with prognosis. Finally, we demonstrated that SAA1 and KIF18B could increase the proliferation, migration, and invasion ability of ccRCC cells using cellular experiments. Conclusion: In summary, we further revealed the importance of VTRGs in ccRCC prognosis.

Clear cell renal cell carcinoma (ccRCC) is a malignancy originating from the kidney, mainly from the clear cells in the renal parenchyma [1]. ccRCC accounts for 83–88% of all histological subtypes [2]. Incidence of ccRCC elevated significantly with age, with a higher incidence in middle-aged and elderly people aged 60–70 years and above, and slightly higher in men than in women [3]. Epidemiological studies have shown that factors such as smoking, obesity, high blood pressure, genetic factors, and long-term exposure to certain chemicals are linked to elevated risk of ccRCC [4]. However, the exact cause is still not fully understood. Although some progress has been made in the treatment of ccRCC, its prognosis still faces many challenges. First, although partial nephrectomy or radical nephrectomy surgery has significant therapeutic effects in the treatment of early-stage localized ccRCC, over 30% of patients are with advanced disease, and about 25% of patients eventually experience disease recurrence [5]. Second, ccRCC is a highly heterogeneous tumor with different pathological types, grades, and stages, leading to differences in prognosis among different patients [6]. Some patients have poor response to traditional radiotherapy and chemotherapy, which increases the uncertainty of the prognosis [2]. Given the current challenges in prognosis, we need to work hard to seek better solutions.

Extracellular vesicles are small vesicular bodies with a double-layered membrane structure secreted by various types of cells. Currently, they are mainly divided into exosomes, microvesicles, and apoptotic bodies [7]. The vesicles carry the material to be transported and bud from one organelle, detach, and then confluence with the membrane of another organelle, releasing the material transported to that organelle, a process known as vesicular transport (VT) [8]. VT and dynamics are closely related to various aspects of cell behavior related to cell differentiation, polarity loss, cell homeostasis, invasion, and metastasis [9, 10]. VT also has potential application value in cancer progression and treatment. Wang et al. [11] disclosed that in cancer stem cells of ccRCC patients, exosomes play an important role. They found that exosomes can transport miR-19b-3p to cancer cells, promote epithelial-mesenchymal transition, and accelerate tumor growth and metastasis. Kamerkar et al. [12] targeted the key regulatory factors of macrophage polarization through VT and reprogrammed the immunosuppressive M2 macrophages into anti-tumor M1 macrophages, leading to establishment of adaptive anti-tumor immune response and elimination of tumors. In summary, VT affects occurrence, metastasis, immune escape, and treatment of tumors and may have important implications for the prognosis of ccRCC.

The aim was to comprehensively investigate prognostic value of VT-related genes (VTRGs) in ccRCC and their correlation with immunity. These results may provide a reference for forecasting survival outcomes and immune therapeutic effects of ccRCC.

Data Acquisition

To conduct our study, we obtained a detailed dataset of ccRCC patients from public databases. Our training dataset source was The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/), which provided extensive tumor sample data and reliable basis for our study on ccRCC. The dataset contained 72 normal samples and 542 ccRCC samples (downloaded on April 10, 2023). We obtained the ccRCC validation set (RECA-EU) from International Cancer Genome Consortium (ICGC) (https://dcc.icgc.org/). This validation set included 91 ccRCC patient samples from different regions with recorded survival time (downloaded on January 17, 2023). We acquired 724 VTRGs from relevant literature [13] (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000539095).

Differential Gene Analysis of ccRCC

We used the R package “edgeR” to conduct differential analysis of normal and tumor tissues in the ccRCC data, with a threshold of |logFC|>1 and FDR <0.05 for screening. The screened ccRCC differentially expressed genes (DEGs) were used for subsequent analysis.

Establishment of Prognostic Model

We used univariate Cox analysis in the R package “survival” to screen VTRGs linked with overall survival (OS) of patients in TCGA training set (threshold: p < 0.05). Genes that simultaneously exhibited significant survival and expression differences were selected as candidate genes for prognosis. We used LASSO regression analysis in R packages “glmnet” and “survival” to narrow down the range of candidate genes. LASSO regression was a feature selection and model compression technique that compressed the coefficients of some irrelevant or redundant features to zero by constraining the sum of the absolute values of the coefficients. In our study, LASSO regression helped us identify the most valuable VTRGs associated with prognosis, thereby improving the accuracy and interpretability of the model. We used the multivariate Cox regression function in the R package “survival” to build a risk model and used expression of each VTRG as input variables. By analyzing the regression coefficients and corresponding p values of each gene, we evaluated the prognostic contribution of each gene to ccRCC patients’ OS. Risk model was calculated by the formula:
Riskscore=i=1nXi×βi,
where n, Xi, and βi represented the total number of genes, FPKM values, and regression coefficients of the gene, respectively.

Model Performance Evaluation and Validation

Using the prognostic model we established, we were able to calculate riskscore of all ccRCC patients in training and validation sets. We classified patients into high-/low-risk groups based on median riskscore as cutoff. To visually display distribution of riskscore, we utilized R package “ggplot2” to plot riskscore distribution and survival distribution graphs. R package “pheatmap” was utilized to generate a heatmap of gene expression levels to show expression patterns of prognostic genes in varying patients. K-M survival curve could evaluate whether the model could accurately predict patients’ survival by observing the survival rate within a certain time interval. R package “survminer” was utilized to do K-M analysis to evaluate difference in OS. Receiver operating characteristic (ROC) curve evaluated classification ability of the model by balancing true-positive rate and the false-positive rate. Diagnostic performance of the model was evaluated by plotting the ROC curves for 1, 3, and 5 years using R packages “timeROC” and “survival.”

Association Analysis between Riskscore and Tumor Progression

We assessed the intergroup differences in riskscores for different T, M, N, and stage classifications using the Wilcoxon test. We employed ggplot2 to generate boxplots for visualization. Furthermore, we obtained genes related to angiogenesis activity and mesenchymal-epithelial transition from literature [14]. Using the GSVA package, we scored these two indicators for ccRCC patients. Subsequently, we employed the Wilcoxon test to evaluate the differences in these two indicators between high- and low-risk groups.

KEGG and GO

Using the R package “edgeR,” we calculated the DEGs (FDR <0.05 and |log(FC)|>1) between high- and low-risk groups and conducted GO and KEGG analyses by R package “clusterProfiler” to investigate biological functions and signaling pathways of DEGs.

Independent Prognostic Analysis

We combined clinical information and used “survival” and “forestplot” tools in the R package to perform univariate/multivariate Cox regression analyses to confirm whether prognostic model could be an independent prognostic factor for ccRCC and constructed a nomogram. We used R packages “regplot” and “rms” to build a prognostic nomogram and 1-, 3-, and 5-year calibration curves to assess deviation between nomogram and ideal model.

Immune Landscape of High- and Low-Risk Groups

Using ssGSEA method, we computed ssGSEA scores for immune-related cells and immune-related functions in high-/low-risk groups. This method quantified activity level of immune-related cells and functions in samples by comparing gene expression data with predefined immune-related gene sets. We calculated the ssGSEA scores for immune-related cells and functions separately for high- and low-risk groups. Using R package “estimate,” we computed immune, stromal, and ESTIMATE score and tumor purity score for each sample, performed Wilcoxon test, and plotted violin plots for high-/low-risk groups.

Drug Prediction

Using data from the GDSC database, we obtained information on the sensitivity of various anticancer drugs. This information, based on the relationship between cell lines and drugs, could be used to predict the response of samples to specific drugs. Using the R package “oncoPredict,” we compared transcriptome data of samples with drug response prediction models in GDSC to calculate degree of response of each sample to drugs. We performed statistical analysis using Wilcoxon test on log2 IC50 values. This drug sensitivity analysis provided important information for personalized treatment and drug selection.

Cell Culture

Human ccRCC cell lines 786-O (CTCC-002-0016) and A-498 (CTCC-003-0009) were obtained from MEISEN CELL. The 786-O cell line was cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), while the A-498 cell line was cultured in DMEM-H medium supplemented with 10% FBS. Cells were maintained in a 37°C, 5% CO2 humidified incubator.

Cell Transfection

The si-SAA1, si-KIF18B, and corresponding negative control were all purchased from Ribobio (China). Transfection of si-SAA1, si-KIF18B, and the corresponding negative control into ccRCC cell lines was performed using the Lipofectamine 2000 reagent kit (Thermo Fisher, USA) following the instructions. After 48 h of transfection, the transfected cells were used for subsequent experiments.

qRT-PCR

Total RNA was extracted from cells using TRIzol reagent (Life Technologies, USA), and the RNA concentration was determined using the NanoDrop 2000 system (Thermo Fisher Scientific, Inc., USA). According to the kit instructions, total RNA was reverse transcribed into cDNA using the PrimeScript RT Master Mix (Takara, P.R., Japan). The mRNA expression levels were measured using the QuantiTect SYBR® Green PCR Kits (Qiagen, Germany). qRT-PCR was performed on an ABI 7500 real-time PCR system (Thermo Fisher, USA) to detect the expression levels of SAA1 and KIF18B. β-Actin was used as the standardized internal control. The results were compared between the control and experimental groups using the 2−ΔΔCt method to determine the relative expression levels of SAA1 and KIF18B. The primer sequences are provided in Table 1.

Table 1.

Primer set for qPCR

GeneForward primer (5′-3′)Reverse primer (5′-3′)
SAA1 TTT​TCT​GCT​CCT​TGG​TCC​TG TGG​AAG​TAT​TTG​TCT​GAG​CCG 
KIF18B TTC​ACA​CCA​TTC​TCC​TGC​CT ATA​AAT​GAG​AGG​GGA​GGC​CG 
β-Actin TCC​GGC​ACT​ACC​GAG​TTA​TC GAT​CCG​GTG​TAG​CAG​ATC​GC 
GeneForward primer (5′-3′)Reverse primer (5′-3′)
SAA1 TTT​TCT​GCT​CCT​TGG​TCC​TG TGG​AAG​TAT​TTG​TCT​GAG​CCG 
KIF18B TTC​ACA​CCA​TTC​TCC​TGC​CT ATA​AAT​GAG​AGG​GGA​GGC​CG 
β-Actin TCC​GGC​ACT​ACC​GAG​TTA​TC GAT​CCG​GTG​TAG​CAG​ATC​GC 

CCK-8 Assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay kit from MedChemExpress (USA). Transfected cells were collected and seeded in a 96-well plate at a density of 2,000 cells per well. After incubation for 0, 24, 48, 72, and 96 h, 10 μL of CCK-8 detection reagent was added to each well. The cells were then incubated for an additional 2 h in a cell culture incubator. The absorbance was measured at a wavelength of 450 nm using a microplate reader.

Colony Formation Assay

Transfected cells from different groups were digested with 0.25% trypsin and seeded in a 12-well plate at a density of 4 × 102 cells per well. The cells were then cultured in fresh medium for 2 weeks. When visible cell colonies were observed, the culture medium was discarded. Cell colonies were fixed with 4% paraformaldehyde, stained with 0.1% crystal violet for 10 min, and rinsed with PBS. Finally, the cell colonies were counted.

Wound Healing Assay

ccRCC cells were seeded in a 6-well plate at a density of 1 × 105 cells per well. When the cells reached 80% confluence, a scratch was made using a 200 μL pipette tip. The cells were then washed with PBS to remove cell debris and further incubated for 24 h. The migration ability of cells was assessed by measuring the change in the width of the scratch gap before and after healing. The imaging of the scratch was observed under a microscope, and analysis was performed using Image J software.

Transwell

The upper chamber of a Transwell apparatus (Corning, USA) was coated with 50 μL of matrix gel (BD Biosciences, USA). ccRCC cells (1 × 105 cells per well) were suspended in serum-free culture medium and seeded in the upper chamber at a volume of 200 μL per well. Simultaneously, 600 μL of cell culture medium containing 10% FBS (Thermo Fisher Scientific, USA) was added to the lower chamber. After 24 h of incubation, the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. The invasive cells were observed under a microscope and counted.

Data Analysis

All statistical analyses were performed using GraphPad Prism 8.2.1 software (GraphPad Software, Inc., La Jolla, USA). All experiments were performed at least three times, and the results were presented as mean ± standard deviation. Differences between groups were compared using t tests or one-way analysis of variance. A p value <0.05 was considered statistically significant, indicated by * in the figures.

Establishment of Prognostic Model

We performed differential analysis on normal and tumor tissue in ccRCC training set. By setting threshold as |logFC|>1 and FDR <0.05, we identified 5907 DEGs, including 1,839 downregulated genes and 4,068 upregulated genes (Fig. 1a) (online suppl. Table 2). Subsequently, we performed univariate Cox regression analysis on VTRGs and selected 371 genes that were noticeably linked to survival (p < 0.05), of which 82 genes were also differentially expressed in ccRCC (online suppl. Table 3). LASSO was a regression analysis method with variable selection ability that could select the most predictive variables from a large group of correlated variables. We used LASSO to analyze the 82 genes to narrow down the selection range (Fig. 1b, c). Finally, we determined 13 prognostic factors through multivariate Cox regression analysis, which were utilized to establish prognostic model (Fig. 1d). Riskscore formula for model was as follows:
Riskscore=0.1232*TUBB2B+0.0004*GJB10.1758*ANK3+0.2223*STX16+0.0495*GJD4+0.307*TUBB6+0.1084*GJB7+0.2996*KIF18A0.09*TGFA+0.0821*CNIH2+0.0535*CLVS10.0754*KIF18B+0.0566*SAA1
Fig. 1.

Gene screening and construction of ccRCC prognostic model. a Volcano plot of DEGs in ccRCC (red represents upregulated genes; green represents downregulated genes). b Coefficient distribution diagram generated for logarithmic (λ) sequences in the LASSO model. c LASSO coefficient spectrum for LASSO Cox analysis. d Forest plot for multivariate Cox analysis.

Fig. 1.

Gene screening and construction of ccRCC prognostic model. a Volcano plot of DEGs in ccRCC (red represents upregulated genes; green represents downregulated genes). b Coefficient distribution diagram generated for logarithmic (λ) sequences in the LASSO model. c LASSO coefficient spectrum for LASSO Cox analysis. d Forest plot for multivariate Cox analysis.

Close modal

Performance Evaluation and Validation of Prognostic Model

We evaluated and externally validated the model performance. We used the prognostic model formula to score ccRCC patients and divided them into high-/low-risk groups with median riskscore as cutoff. We plotted distribution of riskscore and survival status of patients in training set, which presented that riskscore of high-risk group patients was higher, and there were more dead patients in the high-risk group and more alive patients in the low-risk group (Fig. 2a, b). K-M survival analysis presented that survival of low-risk group was better (Fig. 2c). The heatmap of prognostic gene expression showed that ANK3, GJB1, and TGFA were highly expressed in low-risk group, while TUBB2B, GJB1, GJB7, TUBB6, SAA1, KIF18A, KIF18B, CLVS1, GJD4, STX16, and CNIH2 were highly expressed in high-risk group (Fig. 2d). AUC values for 1, 3, and 5 years were 0.77, 0.76, and 0.79, respectively (Fig. 2e). We performed same validation on RECA-EU validation set. Prognostic model had good stratification and predictive ability, with AUC values of 0.63, 0.72, and 0.72 for 1, 3, and 5 years, respectively (Fig. 3a–e). This indicated that prognostic model we constructed had good predictive performance.

Fig. 2.

Internal validation of the prognostic model. a Riskscore distribution for high-/low-risk patients. b Survival status distribution of high-/low-risk patients. c Kaplan-Meier curve for OS of high-/low-risk patients. d Heatmap displaying the expression levels of 13 ccRCC prognostic genes. e ROC curve plotted based on the training set.

Fig. 2.

Internal validation of the prognostic model. a Riskscore distribution for high-/low-risk patients. b Survival status distribution of high-/low-risk patients. c Kaplan-Meier curve for OS of high-/low-risk patients. d Heatmap displaying the expression levels of 13 ccRCC prognostic genes. e ROC curve plotted based on the training set.

Close modal
Fig. 3.

Validation of the prognostic model in the RECA-EU validation set. a Riskscore distribution for high-/low-risk patients in the RECA-EU validation set. b Survival status distribution of high-/low-risk patients in the RECA-EU validation set. c Kaplan-Meier curve for OS of patients in the RECA-EU validation set. d Heatmap displaying the expression levels of 13 ccRCC prognostic genes. e ROC curve plotted based on the RECA-EU validation set.

Fig. 3.

Validation of the prognostic model in the RECA-EU validation set. a Riskscore distribution for high-/low-risk patients in the RECA-EU validation set. b Survival status distribution of high-/low-risk patients in the RECA-EU validation set. c Kaplan-Meier curve for OS of patients in the RECA-EU validation set. d Heatmap displaying the expression levels of 13 ccRCC prognostic genes. e ROC curve plotted based on the RECA-EU validation set.

Close modal

Clinical Significance and Nomogram of Prognostic Features

We performed univariate and multivariate Cox regression analyses on riskscore, age, pathological stage, and other features, combined with clinical information of ccRCC samples. In the univariate regression analysis, we observed that age, grade, stage, TNM staging, and riskscore were significant in predicting patient survival (Fig. 4a). In the multivariate Cox regression analysis, age, stage, and riskscore were significant (Fig. 4b). This meant that riskscore could serve as an independent prognostic indicator for the survival of ccRCC patients. To determine effectiveness of the prognostic model in clinical applications, we plotted a nomogram according to riskscore and clinical pathological factors (Fig. 4c) and calibrated curves for 1, 3, and 5 years (Fig. 4d–f). Calibration curves presented a high consistency between predicted and actual OS rates for 1, 3, and 5 years. Nomogram could accurately predict survival rates of patients.

Fig. 4.

Independent prognostic analysis of riskscore and evaluation using nomogram. a Results of univariate Cox regression analysis for different features. b Results of multivariate Cox regression analysis for different features. c Nomogram of prognostic model scores combined with clinical information. d–f Calibration curves for 1-year, 3-year, and 5-year risk prediction.

Fig. 4.

Independent prognostic analysis of riskscore and evaluation using nomogram. a Results of univariate Cox regression analysis for different features. b Results of multivariate Cox regression analysis for different features. c Nomogram of prognostic model scores combined with clinical information. d–f Calibration curves for 1-year, 3-year, and 5-year risk prediction.

Close modal

Association Analysis between Riskscore and Tumor Progression

To further elucidate the prognostic role of the riskscore, we conducted an analysis of the association between the riskscore and tumor progression in patients from TCGA. The results showed that in the group of patients with higher tumor progression, the riskscore was significantly higher (p < 0.05) (Fig. 5a). Furthermore, we analyzed the tumor progression-related scores. The results revealed that patients in the high-risk group had significantly higher scores for angiogenesis activity and epithelial-mesenchymal transition compared to the low-risk group (p < 0.05) (Fig. 5b). These findings indicate that higher riskscores from the model correspond to a higher degree of tumor progression, thus affirming the model’s good predictive performance.

Fig. 5.

Association analysis between riskscore and tumor progression in patients. a Comparison of riskscores among different stages of tumor progression. b Angiogenesis activity score, epithelial-mesenchymal transition score in high-risk and low-risk groups.

Fig. 5.

Association analysis between riskscore and tumor progression in patients. a Comparison of riskscores among different stages of tumor progression. b Angiogenesis activity score, epithelial-mesenchymal transition score in high-risk and low-risk groups.

Close modal

KEGG and GO Analyses

To understand reasons for prognostic differences, we performed differential analysis and selected 1,915 DEGs between two groups (online suppl. Table 4) and conducted GO and KEGG analyses on these genes. GO analysis presented that DEGs were mainly concentrated in biological processes including negative regulation of proteolysis and humoral immune response, the cellular component including collagen-containing extracellular matrix, and the molecular functions including receptor ligand activity, signaling receptor activator activity, passive transmembrane transporter activity, and channel activity (Fig. 6a). KEGG analysis presented that DEGs were mainly enriched in cAMP signaling pathway, neuroactive ligand-receptor interaction, and hypertrophic cardiomyopathy pathways (Fig. 6b).

Fig. 6.

GO and KEGG analyses of DEGs in high-/low-risk groups. a GO annotation of DEGs in high-/low-risk groups. b KEGG enrichment analysis of DEGs in high-/low-risk groups.

Fig. 6.

GO and KEGG analyses of DEGs in high-/low-risk groups. a GO annotation of DEGs in high-/low-risk groups. b KEGG enrichment analysis of DEGs in high-/low-risk groups.

Close modal

Immune Landscape Analysis

The immune landscape had an important impact on the development and treatment of tumors. Evaluating immune microenvironment could reveal more comprehensive clinical information about patients and further assess prognosis. Infiltration levels of T helper cells, CD8+ T cells, macrophages, and other immune cells were significantly higher in high-risk group than in low-risk group (p < 0.05) (Fig. 7a). In terms of immune function, high-risk group showed significantly higher levels of parainflammation, check activity, inflammation promoting, and T-cell co-stimulation immune functions than low-risk group (p < 0.05) (Fig. 7b). Subsequently, we computed immune, stromal, ESTIMATE, and tumor purity scores, which were all higher in high-risk group than in low-risk group (Fig. 7c–e). Tumor purity score was significantly lower in high-risk group than in low-risk group (Fig. 7f).

Fig. 7.

Immune landscape analysis of high-/low-risk groups. a Differential analysis of immune cell infiltration levels in high-/low-risk groups. b Differential analysis of immune function in high-/low-risk groups. c–f Differential analysis of immune score, stromal score, ESTIMATE score, and tumor purity score.

Fig. 7.

Immune landscape analysis of high-/low-risk groups. a Differential analysis of immune cell infiltration levels in high-/low-risk groups. b Differential analysis of immune function in high-/low-risk groups. c–f Differential analysis of immune score, stromal score, ESTIMATE score, and tumor purity score.

Close modal

Drug Prediction

Finally, we performed drug sensitivity prediction for high- and low-risk groups. IC50 values of afatinib and afuresertib were lower in low-risk group, implying that afatinib and afuresertib were more suitable for low-risk group. Conversely, IC50 values of acetalax and alisertib were lower in high-risk group, indicating that acetalax and alisertib were more suitable for treatment of high-risk group (Fig. 8).

Fig. 8.

Drug sensitivity prediction for high-/low-risk groups.

Fig. 8.

Drug sensitivity prediction for high-/low-risk groups.

Close modal

SAA1 and KIF18B Can Promote the Metastasis of ccRCC

Studies have shown that prognostic genes, SAA1 [15‒17] and KIF18B [18‒20], play promoting roles in various tumors. Of particular interest, SAA1 has been demonstrated to have a promoting function in renal clear cell carcinoma [21]. Therefore, we selected these two prognostic genes to further explore their potential functions in ccRCC.

We established two ccRCC cell lines with silenced expression of SAA1 and KIF18B and confirmed the transfection efficiency using qRT-PCR. The experimental results showed a significant decrease in the expression levels of SAA1 and KIF18B in cells transfected with si-SAA1 and si-KIF18B (Fig. 9a). Subsequently, cell viability, proliferation, migration, and invasion abilities were assessed using CCK-8, colony formation, scratch wound healing, and Transwell assays. The experimental results demonstrated that silencing SAA1 and KIF18B significantly reduced cell viability, proliferation, migration, and invasion abilities of ccRCC cells compared to the control group (Fig. 9b–e).

Fig. 9.

Exploration of the potential functions of SAA1 and KIF18B in ccRCC cells. a Transfection efficiency detection of SAA1 and KIF18B. b Cell viability assay of si-SAA1 and si-KIF18B cell lines. c Cell proliferation assay of si-SAA1 and si-KIF18B cell lines. d Wound healing assay of si-SAA1 and si-KIF18B cell lines to assess cell migration ability. e Transwell assay of si-SAA1 and si-KIF18B cell lines to assess cell invasion ability.

Fig. 9.

Exploration of the potential functions of SAA1 and KIF18B in ccRCC cells. a Transfection efficiency detection of SAA1 and KIF18B. b Cell viability assay of si-SAA1 and si-KIF18B cell lines. c Cell proliferation assay of si-SAA1 and si-KIF18B cell lines. d Wound healing assay of si-SAA1 and si-KIF18B cell lines to assess cell migration ability. e Transwell assay of si-SAA1 and si-KIF18B cell lines to assess cell invasion ability.

Close modal

ccRCC is a complex disease involving multiple pathways. Exploring a new and accurate prognostic biomarker could help improve the diagnosis, prognosis assessment, and treatment selection of the disease, thereby improving survival rate and quality of life of patients. We found that the riskscore model composed of 13 genes (TUBB2B, GJB1, ANK3, STX16, GJD4, TUBB6, GJB7, KIF18A, TGFA, CNIH2, CLVS1, KIF18B, SAA1) was a potential prognostic marker for ccRCC and had stable prognostic ability through other datasets. TUBB2B and TUBB6 are microtubule protein genes, and microtubules are a component of the cell skeleton. Mutations in TUBB2B can alter the normal function and structure of microtubules, leading to an increase in the incidence of kidney disease and ccRCC [22]. TUBB2B and TUBB6 have also been shown in other studies to be implicated in prognosis of ccRCC, bladder cancer, and breast cancer [23‒25]. KIF18A and KIF18B are members of kinesin family involved in regulation of intracellular transport and mitosis [26‒28]. Chen et al. [29] disclosed that upregulation of KIF18A may play a key role in carcinogenesis and progression of kidney cells. Liu et al. [30] found that high KIF18A expression is a prognostic risk factor for ccRCC patients and is significantly linked to progression through bioinformatics analysis. KIF18A can also drive cell proliferation, migration, and invasion in hepatocellular carcinoma, breast cancer, and cervical cancer through Wnt/β-catenin protein signaling pathway [31‒33]. In this study, we demonstrated that KIF18B can enhance the proliferation, migration, and invasion abilities of ccRCC cells, consistent with previous findings in other cancer research studies [19, 20]. Deletion of CLVS1 can lead to defects in VT pathways, influencing integrity of the renal glomerular filtration barrier and causing kidney disease [34]. SAA1 is the earliest discovered member of serum amyloid protein family and is mainly regulated by inflammation-related cytokines. It may participate in tumor occurrence through interaction with the extracellular matrix [35]. This study found that SAA1 can promote the proliferation, migration, and invasion of ccRCC, which is consistent with the findings of Xu et al. [21]. In summary, the model we constructed based on the above genes could reflect the prognosis of ccRCC patients to a certain extent.

GO and KEGG presented substantial differences in biological functions of humoral immune response and collagen-containing extracellular matrix and the cAMP signaling pathway between different prognosis patients. Humoral immunity is an immune response mediated by B cells, which plays a role in immune killing and pathogen clearance [36]. Evidence suggests that humoral responses are associated with immune protection against various human cancers [37]. Eleftheriadis et al. [38] showed that renal cancer patients enhance humoral immune responses after receiving nivolumab treatment, protecting patients from pathogen invasion. The extracellular matrix is a dynamic structure that provides cells with structural and mechanical support, and extracellular vesicles also need to cross the extracellular matrix to achieve long-distance communication [39, 40]. In the process of cancer development, dysregulation of the extracellular matrix is a significant feature, and malignant cells can promote the stiffening of the extracellular matrix, thereby negatively affecting cell growth, differentiation, and movement [41]. High level of extracellular matrix-related genes is a marker of dismal prognosis for ccRCC [42]. cAMP, also known as cyclic adenosine monophosphate, is a protein kinase activator and a second messenger involved in regulating cell function and is a substance inherent in the human body [43]. Mutations in some genes in the cAMP signaling pathway have been identified as cancer drivers [44]. Shi et al. [45] found that the heavy metal cadmium drives epithelial-mesenchymal transition, migration, and invasion of renal cancer cells through cAMP signaling pathway. Thus, the dysregulation of humoral immune response and collagen-containing extracellular matrix biological functions and the cAMP signaling pathway may be the reasons for poor prognosis in ccRCC.

The type, quantity, and function of immune cells in tumors are decisive factors in cancer occurrence and treatment response [46]. We found that the infiltration levels of immune cells such as T helper cells, CD8+ T cells, and macrophages were significantly higher in patients with poor prognosis. T helper cell subtypes are related to pathogenesis and progression of several solid tumors. Peng et al. [47] manifested that the number of Th22 and Th17 cells is markedly positively correlated with late-stage and high WHO/ISUP grading of renal cell carcinoma. CD8+ T cells are a crucial anti-tumor immune cell. Braun et al. [48] pointed out that CD8+ T cells are highly infiltrated in advanced ccRCC tumors. Despite playing a critical role in eliminating tumor cells, CD8+ T cells often exhibit differentiation and functional disorders and cannot control tumor progression in the late stage [49], suggesting that CD8+ T cells may be implicated in unfavorable prognosis in ccRCC. Macrophages usually exert an oncogenic role. In primary tumors, macrophages stimulate angiogenesis, enhance tumor cell invasion, movement, and infiltration [50]. Therefore, we speculated that patients with poor prognosis of ccRCC may have highly immune-infiltrated characteristics.

Overall, we developed a prognostic model that could effectively predict prognosis of ccRCC. This model first revealed close relationships between VT, ccRCC prognosis, and immunity. Furthermore, we found that the prognostic genes SAA1 and KIF18B could promote the migration and invasion of ccRCC cells. However, due to the lack of clinical specimens, we cannot verify the relevant conclusions, which is also a limitation of our study. Therefore, this study provides a new perspective for future relevant research.

Ethical approval is not required for this study in accordance with local or national guidelines. Patient consent was not required as this study was based on publicly available data.

The authors declare that they have no conflicts of interest with the contents of this article.

No funds were received for this work.

Dayong Ye: conceptualization, and methodology. Jingxian Luo: data curation and writing – original draft preparation. Yujie Wang: visualization and investigation. Dengjun Han: supervision. Mingqiang Su: validation. Dayong Zhang: writing – reviewing and editing.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from the corresponding author [D.H.] upon reasonable request.

1.
Zhao
J
,
Eyzaguirre
E
.
Clear cell papillary renal cell carcinoma
.
Arch Pathol Lab Med
.
2019
;
143
(
9
):
1154
8
.
2.
Hsieh
JJ
,
Purdue
MP
,
Signoretti
S
,
Swanton
C
,
Albiges
L
,
Schmidinger
M
, et al
.
Renal cell carcinoma
.
Nat Rev Dis Primers
.
2017
;
3
:
17009
.
3.
Siegel
RL
,
Miller
KD
,
Wagle
NS
,
Jemal
A
.
Cancer statistics, 2023
.
CA Cancer J Clin
.
2023
;
73
(
1
):
17
48
.
4.
Capitanio
U
,
Bensalah
K
,
Bex
A
,
Boorjian
SA
,
Bray
F
,
Coleman
J
, et al
.
Epidemiology of renal cell carcinoma
.
Eur Urol
.
2019
;
75
(
1
):
74
84
.
5.
Kotecha
RR
,
Motzer
RJ
,
Voss
MH
.
Towards individualized therapy for metastatic renal cell carcinoma
.
Nat Rev Clin Oncol
.
2019
;
16
(
10
):
621
33
.
6.
Gulati
S
,
Martinez
P
,
Joshi
T
,
Birkbak
NJ
,
Santos
CR
,
Rowan
AJ
, et al
.
Systematic evaluation of the prognostic impact and intratumour heterogeneity of clear cell renal cell carcinoma biomarkers
.
Eur Urol
.
2014
;
66
(
5
):
936
48
.
7.
Yanez-Mo
M
,
Siljander
PRM
,
Andreu
Z
,
Zavec
AB
,
Borràs
FE
,
Buzas
EI
, et al
.
Biological properties of extracellular vesicles and their physiological functions
.
J Extracell Vesicles
.
2015
;
4
:
27066
.
8.
Brown
FC
,
Pfeffer
SR
.
An update on transport vesicle tethering
.
Mol Membr Biol
.
2010
;
27
(
8
):
457
61
.
9.
Goldenring
JR
.
A central role for vesicle trafficking in epithelial neoplasia: intracellular highways to carcinogenesis
.
Nat Rev Cancer
.
2013
;
13
(
11
):
813
20
.
10.
Masgrau-Alsina
S
,
Sperandio
M
,
Rohwedder
I
.
Neutrophil recruitment and intracellular vesicle transport: a short overview
.
Eur J Clin Invest
.
2020
;
50
(
6
):
e13237
.
11.
Wang
L
,
Yang
G
,
Zhao
D
,
Wang
J
,
Bai
Y
,
Peng
Q
, et al
.
CD103-positive CSC exosome promotes EMT of clear cell renal cell carcinoma: role of remote MiR-19b-3p
.
Mol Cancer
.
2019
;
18
(
1
):
86
.
12.
Kamerkar
S
,
Leng
C
,
Burenkova
O
,
Jang
SC
,
McCoy
C
,
Zhang
K
, et al
.
Exosome-mediated genetic reprogramming of tumor-associated macrophages by exoASO-STAT6 leads to potent monotherapy antitumor activity
.
Sci Adv
.
2022
;
8
(
7
):
eabj7002
.
13.
Qian
C
,
Jiang
Z
,
Zhou
T
,
Wu
T
,
Zhang
Y
,
Huang
J
, et al
.
Vesicle-mediated transport-related genes are prognostic predictors and are associated with tumor immunity in lung adenocarcinoma
.
Front Immunol
.
2022
;
13
:
1034992
.
14.
Qiu
C
,
Shi
W
,
Wu
H
,
Zou
S
,
Li
J
,
Wang
D
, et al
.
Identification of molecular subtypes and a prognostic signature based on inflammation-related genes in colon adenocarcinoma
.
Front Immunol
.
2021
;
12
:
769685
.
15.
Takehara
M
,
Sato
Y
,
Kimura
T
,
Noda
K
,
Miyamoto
H
,
Fujino
Y
, et al
.
Cancer-associated adipocytes promote pancreatic cancer progression through SAA1 expression
.
Cancer Sci
.
2020
;
111
(
8
):
2883
94
.
16.
Yasukawa
Y
,
Hattori
N
,
Iida
N
,
Takeshima
H
,
Maeda
M
,
Kiyono
T
, et al
.
SAA1 is upregulated in gastric cancer-associated fibroblasts possibly by its enhancer activation
.
Carcinogenesis
.
2021
;
42
(
2
):
180
9
.
17.
Niu
X
,
Yin
L
,
Yang
X
,
Yang
Y
,
Gu
Y
,
Sun
Y
, et al
.
Serum amyloid A 1 induces suppressive neutrophils through the Toll-like receptor 2-mediated signaling pathway to promote progression of breast cancer
.
Cancer Sci
.
2022
;
113
(
4
):
1140
53
.
18.
Gao
T
,
Yu
L
,
Fang
Z
,
Liu
J
,
Bai
C
,
Li
S
, et al
.
KIF18B promotes tumor progression in osteosarcoma by activating β-catenin
.
Cancer Biol Med
.
2020
;
17
(
2
):
371
86
.
19.
Xie
J
,
Wang
B
,
Luo
W
,
Li
C
,
Jia
X
.
Upregulation of KIF18B facilitates malignant phenotype of esophageal squamous cell carcinoma by activating CDCA8/mTORC1 pathway
.
J Clin Lab Anal
.
2022
;
36
(
10
):
e24633
.
20.
Qiu
MJ
,
Zhang
L
,
Chen
YB
,
Zhu
LS
,
Zhang
B
,
Li
QT
, et al
.
KIF18B as a regulator in tumor microenvironment accelerates tumor progression and triggers poor outcome in hepatocellular carcinoma
.
Int J Biochem Cell Biol
.
2021
;
137
:
106037
.
21.
Xu
Z
,
Wu
Y
,
Fu
G
,
Chen
X
,
Sun
J
,
Tian
J
, et al
.
SAA1 has potential as a prognostic biomarker correlated with cell proliferation, migration, and an indicator for immune infiltration of tumor microenvironment in clear cell renal cell carcinoma
.
Int J Mol Sci
.
2023
;
24
(
8
):
7505
.
22.
Jeruschke
S
,
Jeruschke
K
,
DiStasio
A
,
Karaterzi
S
,
Büscher
AK
,
Nalbant
P
, et al
.
Everolimus stabilizes podocyte microtubules via enhancing TUBB2B and DCDC2 expression
.
PLoS One
.
2015
;
10
(
9
):
e0137043
.
23.
Chen
L
,
Xiang
Z
,
Chen
X
,
Zhu
X
,
Peng
X
.
A seven-gene signature model predicts overall survival in kidney renal clear cell carcinoma
.
Hereditas
.
2020
;
157
(
1
):
38
.
24.
Kim
B
,
Jung
M
,
Moon
KC
,
Han
D
,
Kim
K
,
Kim
H
, et al
.
Quantitative proteomics identifies TUBB6 as a biomarker of muscle-invasion and poor prognosis in bladder cancer
.
Int J Cancer
.
2023
;
152
(
2
):
320
30
.
25.
Nami
B
,
Wang
Z
.
Genetics and expression profile of the tubulin gene superfamily in breast cancer subtypes and its relation to taxane resistance
.
Cancers
.
2018
;
10
(
8
):
274
.
26.
Du
Y
,
English
CA
,
Ohi
R
.
The kinesin-8 Kif18A dampens microtubule plus-end dynamics
.
Curr Biol
.
2010
;
20
(
4
):
374
80
.
27.
Mayr
MI
,
Storch
M
,
Howard
J
,
Mayer
TU
.
A non-motor microtubule binding site is essential for the high processivity and mitotic function of kinesin-8 Kif18A
.
PLoS One
.
2011
;
6
(
11
):
e27471
.
28.
Hirokawa
N
,
Takemura
R
.
Kinesin superfamily proteins and their various functions and dynamics
.
Exp Cell Res
.
2004
;
301
(
1
):
50
9
.
29.
Chen
QI
,
Cao
B
,
Nan
N
,
Wang
YU
,
Zhai
XU
,
Li
Y
, et al
.
Elevated expression of KIF18A enhances cell proliferation and predicts poor survival in human clear cell renal carcinoma
.
Exp Ther Med
.
2016
;
12
(
1
):
377
83
.
30.
Liu
Q
,
Zhang
X
,
Tang
H
,
Liu
J
,
Fu
C
,
Sun
M
, et al
.
Bioinformatics analysis suggests the combined expression of AURKB and KIF18B being an important event in the development of clear cell renal cell carcinoma
.
Pathol Oncol Res
.
2020
;
26
(
3
):
1583
94
.
31.
Yang
B
,
Wang
S
,
Xie
H
,
Wang
C
,
Gao
X
,
Rong
Y
, et al
.
KIF18B promotes hepatocellular carcinoma progression through activating Wnt/β-catenin-signaling pathway
.
J Cell Physiol
.
2020
;
235
(
10
):
6507
14
.
32.
Liu
L
,
Zhang
Z
,
Xia
X
,
Lei
J
.
KIF18B promotes breast cancer cell proliferation, migration and invasion by targeting TRIP13 and activating the Wnt/β-catenin signaling pathway
.
Oncol Lett
.
2022
;
23
(
4
):
112
.
33.
Wu
Y
,
Wang
A
,
Zhu
B
,
Huang
J
,
Lu
E
,
Xu
H
, et al
.
KIF18B promotes tumor progression through activating the Wnt/β-catenin pathway in cervical cancer
.
Onco Targets Ther
.
2018
;
11
:
1707
20
.
34.
Lane
BM
,
Chryst-Stangl
M
,
Wu
G
,
Shalaby
M
,
El Desoky
S
,
Middleton
CC
, et al
.
Steroid-sensitive nephrotic syndrome candidate gene CLVS1 regulates podocyte oxidative stress and endocytosis
.
JCI Insight
.
2022
;
7
(
2
):
e152102
.
35.
Malle
E
,
Sodin-Semrl
S
,
Kovacevic
A
.
Serum amyloid A: an acute-phase protein involved in tumour pathogenesis
.
Cell Mol Life Sci
.
2009
;
66
(
1
):
9
26
.
36.
Tomar
N
,
De
RK
.
A brief outline of the immune system
.
Methods Mol Biol
.
2014
;
1184
:
3
12
.
37.
Conejo-Garcia
JR
,
Biswas
S
,
Chaurio
R
.
Humoral immune responses: unsung heroes of the war on cancer
.
Semin Immunol
.
2020
;
49
:
101419
.
38.
Eleftheriadis
T
,
Pissas
G
,
Liakopoulos
V
,
Stefanidis
I
.
Anti-PD1 immunotherapy for metastatic renal cancer boosted humoral immunity in a hemodialysis patient
.
J Immunother
.
2021
;
44
(
4
):
164
6
.
39.
Bonnans
C
,
Chou
J
,
Werb
Z
.
Remodelling the extracellular matrix in development and disease
.
Nat Rev Mol Cell Biol
.
2014
;
15
(
12
):
786
801
.
40.
Lenzini
S
,
Bargi
R
,
Chung
G
,
Shin
JW
.
Matrix mechanics and water permeation regulate extracellular vesicle transport
.
Nat Nanotechnol
.
2020
;
15
(
3
):
217
23
.
41.
Liu
T
,
Babaniyi
OA
,
Hall
TJ
,
Barbone
PE
,
Oberai
AA
.
Noninvasive in-vivo quantification of mechanical heterogeneity of invasive breast carcinomas
.
PLoS One
.
2015
;
10
(
7
):
e0130258
.
42.
Chen
X
,
Lin
J
,
Chen
M
,
Chen
Q
,
Cai
Z
,
Tang
A
.
Identification of adhesion-associated extracellular matrix component thrombospondin 3 as a prognostic signature for clear cell renal cell carcinoma
.
Investig Clin Urol
.
2022
;
63
(
1
):
107
17
.
43.
Sutherland
EW
,
Rall
TW
.
Fractionation and characterization of a cyclic adenine ribonucleotide formed by tissue particles
.
J Biol Chem
.
1958
;
232
(
2
):
1077
91
.
44.
Bolger
GB
.
The cAMP-signaling cancers: clinically-divergent disorders with a common central pathway
.
Front Endocrinol
.
2022
;
13
:
1024423
.
45.
Shi
H
,
Sun
X
,
Kong
A
,
Ma
H
,
Xie
Y
,
Cheng
D
, et al
.
Cadmium induces epithelial-mesenchymal transition and migration of renal cancer cells by increasing PGE2 through a cAMP/PKA-COX2 dependent mechanism
.
Ecotoxicol Environ Saf
.
2021
;
207
:
111480
.
46.
Angell
H
,
Galon
J
.
From the immune contexture to the Immunoscore: the role of prognostic and predictive immune markers in cancer
.
Curr Opin Immunol
.
2013
;
25
(
2
):
261
7
.
47.
Peng
Z
,
Hu
Y
,
Ren
J
,
Yu
N
,
Li
Z
,
Xu
Z
.
Circulating Th22 cells, as well as Th17 cells, are elevated in patients with renal cell carcinoma
.
Int J Med Sci
.
2021
;
18
(
1
):
99
108
.
48.
Braun
DA
,
Hou
Y
,
Bakouny
Z
,
Ficial
M
,
Sant’ Angelo
M
,
Forman
J
, et al
.
Interplay of somatic alterations and immune infiltration modulates response to PD-1 blockade in advanced clear cell renal cell carcinoma
.
Nat Med
.
2020
;
26
(
6
):
909
18
.
49.
Philip
M
,
Schietinger
A
.
CD8(+) T cell differentiation and dysfunction in cancer
.
Nat Rev Immunol
.
2022
;
22
(
4
):
209
23
.
50.
Dallavalasa
S
,
Beeraka
NM
,
Basavaraju
CG
,
Tulimilli
SV
,
Sadhu
SP
,
Rajesh
K
, et al
.
The role of tumor associated macrophages (TAMs) in cancer progression, chemoresistance, angiogenesis and metastasis: current status
.
Curr Med Chem
.
2021
;
28
(
39
):
8203
36
.