Abstract

Background

   Kinesin family member 14 (KIF14) is a significant multifunctional
   protein that has been linked to several malignancies. However, the
   varied expression profiles of KIF14 and its prognostic relevance in
   Clear cell renal cell carcinoma (ccRCC) have yet to be elucidated.

Methods

   Patients with ccRCC were obtained from The Cancer Genome Atlas (TCGA),
   Gene Expression Omnibus (GEO) and ArrayExpress databases. A comparison
   of KIF14 expression levels between ccRCC and normal tissues was
   conducted using the Wilcoxon rank sum test. Logistic regression
   analysis was subsequently employed to evaluate the relationship between
   KIF14 expression and clinicopathological features. Furthermore, Gene
   Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) term
   analysis, gene set enrichment analysis (GSEA) and single sample gene
   set enrichment analysis (ssGSEA), as well as CIBERSORT, were utilized
   to elucidate the enriched pathways and functions linked to KIF14 and to
   quantify the level of immune cell infiltration. Kaplan-Meier analysis
   was conducted to assess the correlation between KIF14 expression and
   survival. Additionally, KIF14 expression was downregulated in A498
   ccRCC cells, and their proliferation, expansion capacity, cell cycle,
   and apoptosis were assessed through CCK-8 assays, colony formation
   assays, 7-AAD staining, and Annexin V/PI staining, respectively.

Results

   The findings of this study demonstrate that KIF14 mRNA levels are
   notably increased in ccRCC. Furthermore, a positive association was
   observed between KIF14 expression and cancer stage, nodal metastasis,
   and the infiltration of various immune cells in ccRCC. High levels of
   KIF14 were also found to be indicative of poor survival outcomes among
   ccRCC patients. Knockdown of KIF14 in A498 cells resulted in reduced
   proliferation, diminished colony formation capacity, cell cycle arrest,
   increased apoptosis, and downregulation of CyclinD1 and CDK4.

Conclusions

   KIF14 down-regulates cell cycle proteins CyclinD1 and CDK4 to
   facilitate the proliferation of ccRCC cells, suggesting its potential
   as a therapeutic target and prognostic biomarker in ccRCC.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12894-025-01732-8.

   Keywords: KIF14, Clear cell renal cell carcinoma, Immune infiltration,
   Proliferation, Cell cycle

Introduction

   Renal cell carcinoma (RCC), representing 85% of renal cancers and 2–3%
   of adult malignancies [[34]1], saw 431,288 new cases and 179,368 deaths
   globally in 2020 (GLOBOCAN data) [[35]2, [36]3]. Clear cell RCC
   (ccRCC), the predominant (70%) and most aggressive subtype [[37]4],
   often remains asymptomatic until advanced stages. While surgical
   resection remains primary treatment for localized cases, ~ 30%
   experience recurrence and high mortality upon metastasis [[38]5,
   [39]6]. Despite recent survival improvements through targeted therapies
   and immunotherapy, challenges persist in understanding tumor
   heterogeneity, immune evasion mechanisms, and personalizing treatment
   strategies [[40]7, [41]8]. This underscores the critical need to
   identify novel biomarkers and therapeutic targets for ccRCC.

   Kinesin superfamily proteins (KIFs) are ATP-dependent molecular motors
   that transport a variety of cargoes along microtubules in cells and are
   fundamental to cellular function and morphogenesis. They are divided
   into 14 different families, each with distinct structural and
   functional properties [[42]9]. These molecular motors are ubiquitously
   expressed and play key roles in organelle and vesicle trafficking,
   mitosis, spindle assembly and chromosome aggregation [[43]10]. However,
   there is increasing evidence that KIFs are involved in cancer
   development as potential biomarkers. Kinesin family member 14 (KIF14),
   a member of the KIFs family, is a protein-coding gene. This gene
   encodes a member of the kinesin-3 superfamily of microtubule-moving
   proteins, which are involved in many processes including vesicle
   transport, chromosome segregation, mitotic spindle formation and
   cytokinesis [[44]11]. Overexpression of KIF14 can lead to rapid and
   error-prone mitosis, thereby inducing DNA aneuploidy during
   tumorigenesis [[45]12]. In recent years, a growing body of evidence has
   shown that KIF14 is closely associated with the pathogenesis of many
   malignancies, including clear cell renal cell carcinoma [[46]13],
   thyroid carcinoma [[47]14], and colorectal cancer [[48]15]. Fang’s
   research revealed that the deletion of KIF14 induces apoptosis,
   inhibits cell growth, migration, and invasion via the PI3K/AKT pathway
   in bladder cancer [[49]16]. By suppressing KIF14 expression, resistant
   prostate cancer cells can regain sensitivity to both docetaxel and
   cabazitaxel, while also exhibiting reduced proliferation and increased
   apoptosis [[50]17]. In cervical cancer, KIF14 overexpression activates
   the cell cycle and cell viability by blocking p27 degradation [[51]18].
   Similarly, in hepatocellular carcinoma, KIF14 knockdown disrupts
   cytokinesis by impairing the ubiquitination-dependent degradation of
   Kip1 [[52]19]. Despite these advances, the role of KIF14 in ccRCC
   remains poorly defined. Current studies predominantly focus on KIF14’s
   mitotic functions, yet its prognostic significance, immune-modulatory
   roles, and interaction with key cell cycle regulators (e.g.,
   CyclinD1/CDK4) in ccRCC are unexplored. Notably, while KIF14
   overexpression correlates with poor survival in cancer patients
   [[53]20], no study has systematically linked its expression to advanced
   clinicopathological features (e.g., metastasis, immune evasion) in
   ccRCC. This gap is critical because ccRCC exhibits immune
   microenvironment characteristics [[54]21], which may render KIF14 a
   context-dependent oncogene. Addressing these questions could uncover
   novel biomarkers to stratify high-risk patients or synergize with
   immunotherapy.

   In this study, we investigated the expression of KIF14 in renal clear
   cell carcinoma using a bioinformatics approach by obtaining
   transcriptomic data from The Cancer Genome Atlas (TCGA), Gene
   Expression Omnibus (GEO) and ArrayExpress databases. Specifically, our
   analyses focused on KIF14 expression in ccRCC and its correlation with
   clinicopathological features, prognosis and immune infiltration status
   of ccRCC, as well as functional analyses. We also performed functional
   studies of KIF14 in A498 ccRCC cells. Our data showed that KIF14 was
   overexpressed in ccRCC and exhibited oncogenic function in A498 cells.
   These findings contribute to our understanding of the role of KIF14 in
   the development of ccRCC.

Materials and methods

TCGA data source and preprocessing

   RNA-sequencing expression profiles (level 3 HTSeq-FPKM) profiles and
   miRNA data (level HTSeq-RPM) and corresponding clinical information
   [[55]22] for KIRC were downloaded from the TCGA dataset
   ([56]https://portal.gdc.com). Filtered clinical information: Normal
   KIRC cohort, duplicate samples and overall survival less than 30 days
   were removed. Next, RNAseq data in FPKM format were converted to TPM
   (transcripts per million reads) and then log2 transformed.
   Simultaneously, miRNAseq data in RPM format were log2 transformed. The
   final TCGA data, which included 72 normal tissues and 530 tumor tissues
   (TPM data) and 71 normal tissues and 516 tumor tissues (RPM data) were
   used for further analyses. The R software (version 4.1.0) was used to
   process the for TCGA data processing. The study followed the TCGA
   guidelines.

GEO database and the arrayexpress database validation

   Three microarray expression datasets, [57]GSE15641, [58]GSE40435,
   [59]GSE16441 were downloaded from the GEO database
   ([60]http://www.ncbi.nih.gov/geo). All the raw data were downloaded as
   series matrix files. [61]GSE15641 included 32 ccRCC patients and 23
   normal patients; [62]GSE40435 included 101 pairs of ccRCC tumours and
   adjacent non-tumour renal tissue; [63]GSE16441 consisted of 17 ccRCC
   tumors and 17 corresponding non-tumor samples. Filtered data: For genes
   with multiple probes, the probe with the highest expression value (on
   average) was retained. For small-sample datasets, non-parametric tests
   (Wilcoxon rank-sum for unpaired data; Wilcoxon signed-rank for paired
   data) were applied to mitigate normality assumptions. Statistical
   significance was defined as p < 0.05 after false discovery rate (FDR)
   correction. The microarray expression of E-MTAB-1805 was downloaded
   from the ArrayExpress database
   ([64]https://www.ebi.ac.uk/arrayexpress/) and included 15 normal
   tissues and 105 ccRCC tissues.

KIF14 differential expression in ccRCC tissues

   Boxplots and scatter plots, using the ggplot2 package for plotting,
   were used to represent the differential expression of KIF14 in ccRCC.
   For TCGA KIRC specimens, Wilcoxon rank sum test for unpaired samples
   and Wilcoxon signed rank test for paired samples were performed;
   Wilcoxon rank sum test was performed for [65]GSE15641 dataset
   specimens; Wilcoxon signed rank test were used for [66]GSE40435 dataset
   specimens; paired sample t-Test was performed for [67]GSE16441 dataset
   specimens. Receiver Operating Characteristic (ROC) curves were plotted
   using the pROC and ggplot2 packages to investigate the diagnostic
   validity of KIF14.

Functional enrichment analysis

   To clarify the potential biological functions of KIF14 in ccRCC, the
   guilt by association analysis was conducted. We performed a Spearman
   correlation analysis of KIF14 by gene in TCGA KIRC expression
   profiling. Results with absolute correlation coefficient (cor) > 0.5 &
   p-values < 0.001 were considered as statistically significant
   correlation. Next, we conducted the Gene Ontology (GO) and Kyoto
   Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses
   were then performed for KIF14-related genes. The enrichment analysis
   was performed using the clusterProfiler (version4.4.4) R package
   [[68]23] and Metascape [[69]24]. The threshold conditions of enrichment
   results included: adjusted P-value (p.adjust) < 0.05 & false discovery
   rate (FDR) < 0.2 was considered statistically significant. Gene set
   enrichment analysis (GSEA) [[70]25]was used to elucidate the
   significant functional and pathway differences between the high and low
   KIF14 expression groups. Hallmark gene sets was called and each
   analysis procedure was repeated 1000 times, with each gene set
   containing at least 10 genes. Finally, A function or pathway term with
   p.adjust < 0.05 & FDR < 0.25 was considered statistically significant.

Analysis of immune cell infiltration and tumor immune escape

   We undertook immune infiltration analysis for TCGA KIRC sample genes
   using single-sample GSEA (ssGSEA) and CIBERSORT [[71]26] methods to
   assess the relative levels tumor infiltration. ssGSEA analysis was
   performed using the GSVA R package [[72]27] for 24 immunocytes types in
   published signature gene lists [[73]28]; the proportions of the 22
   infiltrating immune cells present in the LM22 datasets from each sample
   were calculated by using the CIBERSORT R package. To better assess the
   distribution of stromal and immune cells in the tumor microenvironment,
   the total StromalScore, ImmuneScore, and ESTIMATEScore were determined
   from each sample using the estimate R package [[74]29]. To explore the
   relationship between KIF14 and the infiltration levels of each immune
   cell subset, Spearman correlations were applied. Moreover, the
   association of immunocytes infiltration with low and high KIF14
   expression groups was investigated using the Wilcoxon rank sum test.

   Understanding the mechanism of tumor immune has important implications
   for tumor treatment, especially implications for cancer immunotherapy.
   Based on the Tumor Immune Dysfunction and Exclusion (TIDE) algorithm,
   we estimate the immunotherapy response with low and high KIF14
   expression groups in the TCGA-KIRC dataset. The TIDE score of patients
   with TCGA KIRC was obtained from the TIDE website
   ([75]http://tide.dfci.harvard.edu) after uploading the RNA-sequencing
   expression profiles. Wilcoxon rank sum test was performed to
   investigate the association of TIDE score with low and high KIF14
   expression groups. Furthermore, the relationship between the different
   expression groups of KIF14 and cancer associated fibroblast-TIDE was
   evaluated by utilizing TIMER2 ([76]http://timer.cistrome.org) database.

Clinical statistical and survival analysis

   The association of clinicalpathologic features with KIF14 was analyzed
   using Chi-squared test and logistic regression. Cox regression analysis
   was performed to explore the relevance between different KIF14 groups
   expression and overall survival (OS), progression-free interval (PFI),
   and disease-specific survival (DSS) in patient prognosis using the
   survival R package. A cut-off value of 50% was set as the KIF14
   expression threshold to separating high and low expression cohorts. The
   Kaplan-Meier survival curve was plotted by using the survminer R
   package. P-value, hazard ratio (HR), and 95% confidence intervals were
   investigated.

Cell culture and SiRNA transfection

   All cell lines (ACHN, A498 and 786-O) were sourced from IMMOCELL
   (Xiamen, Fujian, China). ACHN and A498 cells were cultured in DMEM with
   10% fetal bovine serum (Gibco), while 786-O cell were cultured in RPMI
   1640 medium with 10% fetal bovine serum. These cells were maintained in
   a humidified incubator at 37 °C with 5% CO2. For siRNA transfection in
   A498 cells, we utilized both control (siCtrl) and KIF14-targeting
   siRNAs (siKIF14), and the transfection was performed using X-tremeGENE
   siRNA Transfection Reagent (Roche). The siRNA sequences were list as
   Table [77]1.

Table 1.

   The SiRNA sequences of KIF14
   Serial number siRNA name  siRNA sequences
   1             siKIF14-1-F 5’-GGAUGAGUUGAGACAAGAATT-3’
   2             siKIF14-1-R 5’-UUCUUGUCUCAACUCAUCCTT-3’
   3             siKIF14-2-F 5’-GACUGAAGUUGUUAGAUCATT-3’
   4             siKIF14-2-R 5’-UGAUCUAACAACUUCAGUCTT-3’
   5             siKIF14-3-F 5’-GCAAUGAUUGCUACGAUUATT-3’
   6             siKIF14-3-R 5’-UAAUCGUAGCAAUCAUUGCTT-3’
   7             SiCtrl-F    5’-UUUGUACUACACAAAAGUACUG-3’
   8             SiCtrl-R    5’-CAGUACUUUUGUGUAGUACAAA-3’
   [78]Open in a new tab

Western blotting assay

   The A498 cells were subjected to lysis in RIPA buffer supplemented with
   protease inhibitors, followed by complete homogenization. Following
   centrifugation, the supernatant was collected, and the overall protein
   concentration was quantified using a BCA assay. Subsequently, the
   protein samples were denatured and separated by sodium dodecyl
   sulfate-polyacrylamide gel electrophoresis. The separated proteins were
   then transferred onto PVDF membranes, which were subsequently blocked
   using a 5% non-fat milk/TBST solution. After blocking, the membranes
   were treated with primary antibody solutions for one night at 4 °C.
   They were then washed three times with TBST. Next, the membranes were
   incubated with a secondary antibody solution. Finally, the membranes
   were washed with TBST three additional times and incubated with an ECL
   working solution. A Las-3000 Luminescent Image Analyzer (Fujifilm,
   Japan) was used to visualize the signals. The antibodies used for
   Western blotting are as follows: KIF14 antibody (1:1000, 26000-1-AP,
   Proteintech, China), GAPDH antibody (1:10000, 10494-1-AP, Proteintech,
   China), CyclinD1 antibody (1:5000, 60186-1-lg, Proteintech, China),
   CDK4 antibody (1:2000, 11026-1-AP, Proteintech, China), β-actin
   (1:10000, 11224-1-AP, Proteintech, China), and HRP-conjugated goat
   anti-rabbit IgG (1:10000, 511203/511103, ZENBIO, China).

qPCR assay

   Following the protocol provided by the manufacturer, total RNA was
   isolated from A498 cells using TRIzol reagent (Invitrogen, USA). Then,
   cDNA was synthesized using the PrimeScript™ RT Master Mix (TAKARA,
   Japan). For qPCR, the TB Green Premix Ex Taq II (TAKARA, Japan) was
   employed, and the relative expression levels of KIF14 was calculated
   using the 2 − ΔΔCq method. The primers used for qPCR analysis were list
   as Table [79]2.

Table 2.

   The gene primers sequences for qPCR analysis
   Serial number Primer name Primer sequence (5′ to 3′)
   1             GAPDH-F     5′-AGAAGGCTGGGGCTCATTTG-3′
   2             GAPDH-R     5′-AGGGGCCATCCACAGTCTTC-3′
   3             KIF14-F     5′-CCTCACCCACAGTAGCCGA-3′
   4             KIF14-R     5′-AAGTGCCAATCTACCTACAGGA-3′
   5             CyclinD1-F  5′-GCTGCGAAGTGGAAACCATC-3′
   6             CyclinD1-R  5′-CCTCCTTCTGCACACATTTGAA-3
   7             CDK4-F      5′-ATGGCTACCTCTCGATATGAGC-3′
   8             CDK4-R      5′-CATTGGGGACTCTCACACTCT-3′
   [80]Open in a new tab

Colony formation assay

   Cells transfected with siCtrl or siKIF14 were seeded at a density of
   600 cells per well in 6-well plates and grown for 11 days. Following
   that, the cells were washed with PBS and fixed for 30 min in 4%
   paraformaldehyde. After fixation, they were stained with 0.5% crystal
   violet for 15 min. After that, the cells were washed, and an IX73
   inverted microscope (OLYMPUS, Japan) was used to take pictures of the
   colonies, which were then quantified using ImageJ software (NIH).

Cell proliferation assay

   Cell proliferation was measured using the CCK-8 assay. Briefly, A498
   cells were planted at an average density of 2,000 cells per well in
   96-well plates. After 24 h, either control or KIF14-targeting siRNA was
   transfected into the cells. Following an additional 24-hour incubation,
   CCK-8 solution (C0037, Beyotime, China) was added to the 96-well plate
   on the 1st, 2nd, 3rd, 4th, and 5th day to measure cell proliferation.
   The OD450 values were determined using an Infinite F50 microplate
   reader (Tecan).

Cell cycle assay

   To examine the cell cycle, A498 cells were seeded in 6-well plates at a
   density of 1 × 10^6 cells per well. After 24 h, the cells were
   transfected with either control or KIF14-targeting siRNA. Following an
   additional 24-hour incubation, the cells were treated with pre-chilled
   ethanol overnight at -20 °C. Subsequently, they were washed with PBS
   and incubated with a 7-AAD/PBS solution in the dark for 30 min at 4 °C.
   Finally, a BD FACSCanto™ II machine was used for flow cytometry
   analysis, and the results were processed with FlowJo software (v10).

Apoptosis assay

   Following the manufacturer’s instructions, the FITC Annexin V Apoptosis
   Detection Kit with propidium iodide (PI) (#640914, Biolegend) was used
   to detect apoptotic cells. In brief, A498 cells were seeded in 6-well
   plates at a density of 1 × 10^6 cells per well and cultured for 24 h.
   Subsequently, the cells were transfected with either siCtrl or siKIF14.
   After 24 h, the cells were washed with PBS, digested with a 0.25%
   trypsin solution, pelleted, resuspended in Annexin V binding buffer,
   and treated with Annexin V-FITC/PI working solution. Finally, flow
   cytometry analysis was performed using a BD FACSCanto™ II machine, and
   the results were analyzed using FlowJo software (v10).

Statistical analysis

   The differential expression of KIF14 was assessed using the Wilcoxon
   rank-sum test, ilcoxon signed rank test or Paired Sample t-Test.
   Overall survival (OS) was analyzed using the Kaplan–Meier method and
   the log-rank test. Correlation analysis was conducted using Spearman
   correlation analysis. Unpaired Student’s t-tests and one-way ANOVA were
   employed for comparisons between two groups and multiple groups,
   respectively, utilizing GraphPad Prism (v8) or R software (version
   4.1.0). All experimental results were presented as the mean ± SEM, with
   statistical significance set at p < 0.05. All experiments in this study
   were conducted in triplicate for the purpose of quantification.

Results

KIF14 expression is upregulated in ccRCC tissues

   To evaluate the involvement of KIF14 in ccRCC, we used R software to
   explore the TCGA KIRC cohort to analyze whether KIF14 exhibits
   different expression levels of mRNA in clear cell renal cell carcinoma.
   The results revealed that the transcriptional levels of KIF14 was
   elevated in kidney cancer tissues compared with normal tissues
   (Figs. [81]1A and B and 530 tumors vs. 72 normals, ***p < 0.001). ROC
   curves were constructed to determine the diagnostic accuracy of KIF14
   expression. As shown in Fig. [82]1C, the KIF14 expression had a high
   sensitivity and specificity for KIRC diagnosis. The area under the
   curve (AUC) was 0.928 (95%CI: 0.884–0.973). Subsequently, we analyzed
   the [83]GSE15641 (Figs. [84]1D and 32 tumors vs. 23 normals,
   ***p < 0.001) and [85]GSE40435 (Figs. [86]1E and 101 paired cohort,
   ***p < 0.001) dataset and E-MTAB-1805 (Figs. [87]1F and 105 tumors vs.
   15 normals, ***p < 0.001) cohort data to confirm the expression
   difference of KIF14 between ccRCC samples and healthy tissues. It can
   be observed that the KIF14 is highly expressed in primary kidney
   tumors. Our preliminary studies demonstrated that KIF14 is a potential
   oncogene in clear cell renal cell carcinoma. Genetic perturbation
   similarity analysis with KIF14 in KIRC from the TCGA dataset were
   investigated. We compared 271 KIRC KIF14-high samples with 270
   KIF14-low controls. A total of 674 differentially expressed genes
   (DEGs), covering 354 upregulated GENEs and 320 downregulated GENEs
   (|LogFC| > 1 & p.adj < 0.05) as shown in Fig. [88]1G. DESeq2 package
   was used to analyze DEGs in HTSeq-Counts. Figure [89]1H shows the
   relative expression levels of the top 10 DEGs between the two cohorts.

Fig. 1.

   [90]Fig. 1
   [91]Open in a new tab

   Expression of KIF14 in kidney cancer. (A, B) Differential expression
   levels of KIF14 in TCGA-KIRC. (C) ROC curve of KIF14. (D, E)
   Differentially expressed KIF14 in normal tissue and ccRCC tissue from
   [92]GSE15641 and [93]GSE40435 database. (F) Differentially expressed
   KIF14 in normal tissue and ccRCC tissue from E-MTAB-1805 database. (G,
   H) Volcano plots of the DEGs and heat map showing the top 10 DEGs.
   *p < 0.05, ***p < 0.001. KIRC, kidney renal clear cell carcinoma; AUC,
   area under the curve

Analysis of the association between KIF14 expression and clinicopathological
characteristics

   In order to understand the role and significance of HTRA3 expression,
   539 KIRC samples from the TCGA were analyzed for KIF14 expression data
   and all patient characteristics. As shown in Fig. [94]2A–E, high KIF14
   was remarkably associated with T stage (T3 & T4 vs. T1 & T2,
   ***P < 0.001), clinical stage (stage III & stage IV vs. stage I & stage
   II, ***P < 0.001), histological grade (grade 3 & grade 4 vs. grade 1 &
   grade 2, **P < 0.01), N stage (N1 vs. N0, *P < 0.05), and M stage (M1
   vs. M0, ***P < 0.001). Nevertheless, KIF14 expression has no relation
   with age (Fig. [95]2F). The univariate analysis using Logistic
   regression revealed that KIF14 expression as a categorical dependent
   variable was related with poor prognostic clinicopathological variables
   (Table [96]3). overexpressed HTTA3 expression in KIRC is positively
   correlated with T stage (OR = 2.27 for T3 & T4 vs. T1 & T2, 95%CI:
   1.586–3.280), N stage (OR = 6.50 for N1 vs. N0, 95%CI: 1.765–41.941), M
   stage (OR = 2.53 for M1 vs. M0, 95%CI: 1.518–4.335), pathologic stage
   (OR = 2.15 for stage III & stage IV vs. stage I & stage II, 95%CI:
   1.509–3.076) and histological grade (OR = 2.16 grade 3 & grade 4 vs.
   grade 1 & grade 2, 95%CI: 1.532–3.069). This turned out to be the KIRC
   with high KIF14 expression were more likely to progress to a more
   advanced stage compared to those with low kif14 expression.

Fig. 2.

   [97]Fig. 2
   [98]Open in a new tab

   Relationship between KIF14 expression and clinicopathological
   characteristics, including (A) T stage histological type, (B)
   pathologic stage, (C) histological grade, (D) N stage, (E) M stage, and
   (F) age in KIRC patients in TCGA cohort. *p < 0.05, **p < 0.01,
   ***p < 0.001

Table 3.

   KIF14 expression association with clinical pathological characteristics
   (logistic regression)
   Characteristics Total(N) Odds Ratio(OR) P value
   T stage (T3&T4 vs. T1&T2) 539 2.274 (1.586–3.280) < 0.001
   N stage (N1 vs. N0) 257 6.496 (1.765–41.941) 0.015
   M stage (M1 vs. M0) 506 2.529 (1.518–4.335) < 0.001
   Pathologic stage (Stage III & Stage IV vs. Stage I & Stage II) 536
   2.149 (1.509–3.076) < 0.001
   Histologic grade (G3 & G4 vs. G1 & G2) 531 2.164 (1.532–3.069) < 0.001
   Primary therapy outcome (PD & SD& PR vs. CR) 147 0.704 (0.248–1.867)
   0.489
   [99]Open in a new tab

Functional enrichment and analysis of KIF14-related signaling pathways

   To delve into the biological pathways associated with KIF14 interactive
   genes in KIRC, we utilized GO and KEGG terms enrichment analysis
   employing the clusterProfiler R package. As depicted in Fig. [100]3A,
   genes related to KIF14 were implicated in various biological processes
   (BPs), cellular compositions (CCs), and molecular functions (MFs),
   including nuclear division, cell cycle checkpoint, chromosomal region,
   tubulin binding, among others. The KEGG analysis showed that the genes
   associated with KIF14 were strongly correlated with the Cell cycle,
   Human T-cell leukemia virus 1 infection, p53 signaling pathway,
   NOD-like receptor signaling pathway, and so on (Fig. [101]3B). We also
   validated the functional enrichment analysis of KIF14-related genes
   using the online analysis tool Metascape, and the results were highly
   consistent (Fig. [102]S1). For the sake of determining KIF14-related
   signaling pathways in KIRC, GSEA was used to identify prominently
   activated signaling pathways between low and high KIF14 expression in
   KIRC, and their notable enrichment differences (p.adjust < 0.05 &
   FDR < 0.25) were validated in the Hallmark gene sets (MSigDB,
   h.all.v7.4.symbols.gmt). Based on their normalized enrichment score
   (HES), the most significantly enriched pathways were selected. This
   turned out to be the pathways that showed differential enrichment in
   the KIF14 high expression phenotype included G2M checkpoint, E2F
   targets, IL6 JAK STAT3 signaling, allograft rejection, interferon gamma
   response, and so on. In addition, pathways in the KIF14 low expression
   phenotype included oxidative phosphorylation, fatty acid metabolism,
   and adipogenesis (Fig. [103]3C).

Fig. 3.

   [104]Fig. 3
   [105]Open in a new tab

   Functional and pathway enrichment analysis of KIF14-related genes in
   KIRC. (A) GO terms enrichment analysis. (B) KEGG pathway enrichment
   analysis. (C) GSEA analysis for KIF14 between low and high expression
   in KIRC. BP, biological processes; CC, cellular compositions; MF,
   molecular functions; NES, normalized enrichment score; p.adj, adjusted
   P value; FDR, false discovery rate

Correlation analysis between KIF14 expression and immune cell infiltration in
ccRCC

   The results of functional enrichment analysis by appeal suggest that
   KIF14 may be involved in immune cell infiltration or immune checkpoints
   in RCC. Therefore, we performed immune infiltration analysis by both
   ssGSEA and CIBERSORT in TCGA KIRC patients. As shown in Fig. [106]4A
   and B, the expression of KIF14 was positively associated with helper T2
   (Th2) cells, T helper cells, Macrophages, Treg, Tcm, B cells, etc. and
   negatively associated with the NK cells, helper T17 (Th17) cells, and
   pDC. Furthermore, CIBERSORT analysis exposed the expression of KIF14
   was negatively associated with Mast cells resting, Dendritic cells
   activated, NK cells activated, etc. and positively associated with T
   cells CD4 memory activated, and Macrophages M1 (Fig. [107]4C and D).
   The tumor microenvironment comprises a complex interplay of stromal
   cells, tumor cells, and immune cells. Stromal cells play a supportive
   role in tumor growth, while immune cells can either promote or inhibit
   tumor growth. To gain a deeper understanding of the relevance and
   underlying mechanism of KIF14 expression in ccRCC, we assessed the
   relationship between immune and stromal scores and KIF14 expression
   using ESTIMATE analysis. KIF14 expression exhibited a positive
   correlation with tumor immune, stromal and ESTIMATE scores (sum of
   immune scores and stromal scores), and the group with a low level of
   KIF14 expression had a lower abundance of stromal cells and immune
   cells than the group with a higher level of KIF14 expression
   (Fig. [108]4E-J, ***p < 0.001). Secondly, we examined the potential
   efficacy of immunotherapy in low and high KIF14 expression groups using
   TIDE analysis (Fig. [109]4K, *p < 0.05, ***p < 0.001). The KIF14-high
   group The KIF14-high group had higher TIDE, T-cell dysfunction scores,
   suggesting that immune escape may be more likely to occur in KIF14-high
   patients. Moreover, KIF14 expression was inversely correlated with
   tumor purity and positively correlated with cancer-associated
   fibroblasts (CAF) (Fig. [110]4L).

Fig. 4.

   [111]Fig. 4
   [112]Open in a new tab

   KIF14 expression was associated with the immunocytes infiltration and
   response to immunotherapy in ccRCC. (A) The correlation between the
   expression of KIF14 and the relative abundance of 24 immune cells was
   analyzed by ssGSEA. (B) The difference in the infiltration level of 16
   immune cells between the KIF14 high and low groups. (C) The correlation
   between the expression of KIF14 and the relative abundance of 21 immune
   cells was analyzed by CIBERSORT. (D) The difference in the infiltration
   level of 10 immune cells between the KIF14 high and low groups. (F-H)
   The relationship between tumor immune/stromal/ESTIMATE scores and KIF14
   expression levels. (I-K) The difference in tumor
   immune/stromal/ESTIMATE scores between the KIF14 high and low groups.
   (L) TIDE, MSI, T cell dysfunction and Exclusion scores in the KIF14
   high and low groups. (M) The correlation between tumor purity/CAF and
   KIF14 expression levels. TIDE, Tumor Immune Dysfunction and Exclusion;
   MSI, microsatellite instability; CAF, cancer-associated fibroblasts.
   *p < 0.05, **p < 0.01, ***p < 0.001

Elevated KIF14 expression was intimately correlated with poor prognosis of
patients with KIRC

   As shown in Fig. [113]5A, among KIRC patients with low KIF14
   expression, the 10-year OS rate was remarkably higher than those with
   high HTRA3 expression (47.2 vs. 39.7%, HR = 1.422, p = 0.020, 95%CI:
   1.057–1.913). Likewise, the 10-year DSS rates and PFI were appreciably
   higher in the KIF14 low group than in the KIF14 high group (66.9 vs.
   60.2%; p = 0.002, 95%CI: 1.245–2.639; 55.6 vs. 37.5%; p = 0.001, 95%CI:
   1.216–2.261; Fig. [114]5B, C). We then performed subgroup survival
   analyses for OS, DSS and PFI. The results showed a poorer prognosis in
   patients with high KIF14 expression in the stage III-IV, stage IV and
   M1 subgroups for OS and in the grade 3 & 4, T3 & T4 and M1 subgroups
   for DSS (Fig. [115]5D-I). In addition, KIRC patients with high KIF14
   levels in the M1 subgroup had worse OS and DSS rates (15.3 vs. 4.0%;
   p = 0.021; 16.7 vs. 4.3%; p = 0.020, 95%CI: 1.074–2.883), suggesting
   that KIF14 has a greater prognostic role in KIRC patients with distant
   metastases. However, there was no significant difference in survival
   between the PFI subgroups.

Fig. 5.

   [116]Fig. 5
   [117]Open in a new tab

   Kaplan–Meier overall survival curve for high and low KIF14 expression
   in KIRC. (A-C) Survival curves showing OS, DSS and PFI comparing
   patients with high and low levels of KIF14 in KIRC. (D, E, H) Survival
   curves showing OS with stage III & IV, stage IV, and MI subgroups. (F,
   G, I) DSS survival curves of grade3 & 4, T3 & T4, and MI subgroups.
   KIRC, kidney renal clear cell carcinoma; OS, overall survival; DSS,
   disease specific survival; PFI, progression free interval

Dysregulation of KIF14 expression may be mediated by the miR-200b/200a/429
cluster in ccRCC

   Many previous studies have shown the presence of a large number of
   dysregulated microRNAs (miRNAs) in ccRCC, and miRNAs bind to the 3’UTR
   region of mRNAs and inhibit post-transcriptional translation of mRNAs.
   To investigate whether miRNAs are involved in the mechanism of
   dysregulation of KIF14 overexpression in ccRCC, we examined the
   relevant miRNAs leading to KIF14 overexpression using microT-CDS v5.0
   [[118]30]. An interesting finding was that there was a regulatory
   relationship between KIF14 and the miR-200b/200a/429 cluster
   (Fig. [119]6A and D). Next, we found that hsa-miR-429 and
   has-miR-200b-3p were low expressed in tumor compared to the normal
   groups and inversely correlated with KIF14 expression in the TCGA KIRC
   cohort (Fig. [120]6B, C, E and F, ***p < 0.001). The [121]GSE16441
   dataset was analyzed to further validate that hsa-miR-429 and
   has-miR-200b-3p were poorly expressed in tumors (Fig. [122]6G and H).
   Although there were no statistically significant differences in OS
   between the low and high hsa-miR-429 expression groups, we observed
   that patients in the high miR-200b-3p expression group had a better
   prognosis (Fig. [123]6I, p = 0.020, 95%CI: 0.522–0.947).

Fig. 6.

   [124]Fig. 6
   [125]Open in a new tab

   The miR-200b/200a/429 cluster leads to upregulation of KIF14 expression
   in ccRCC. (A, D) Possible hsa-miR-429 and has-miR-200b-3p binding sites
   for KIF14. (B, E) Differentially expressed hsa-miR-429 and
   has-miR-200b-3p in normal tissue and ccRCC tissue from TCGA KIRC
   database. (C, F) Correlation analysis between
   hsa-miR-429/has-miR-200b-3p and KIF14. (G, H) Analysis of paired
   differential expression of hsa-miR-429 and has-miR-200b-3p from
   [126]GSE16441. (I, J) Kaplan-Meier analysis of overall survival was
   analyzed according to hsa-miR-429 and has-miR-200b-3p expression
   levels. *p < 0.05, **p < 0.01, ***p < 0.001

KIF14 knockdown reduced proliferation, attenuated colony-forming ability,
impaired cell cycle and increased apoptosis of A498 ccRCC cells

   We detected and compared the differences in mRNA levels of KIF14 in
   normal kidney cells and three different renal clear cell carcinoma cell
   lines by qPCR, respectively, and through the experimental results, we
   found that KIF14 expression was up-regulated in renal clear cell
   carcinoma cells and identified A498 as a renal clear cell carcinoma
   cell line with high expression of KIF14 (Fig. [127]7A). To investigate
   the function of KIF14 in ccRCC, we initially depleted KIF14 in A498
   cells. qPCR and Western blotting confirmed that all three siRNAs
   targeting KIF14 efficiently reduced KIF14 expression in A498 cells,
   with siKIF14-1 showing the highest knockdown efficiency
   (Fig. [128]7B-C). Therefore, we selected siKIF14-1 (hereafter referred
   to as siKIF14) for subsequent experiments. A CCK8 test was performed to
   evaluate the effect of KIF14 knockdown on the proliferation of A498
   cells, which revealed a significant decrease in proliferation in
   KIF14-deficient A498 cells compared to control cells (Fig. [129]7D).
   The data obtained from the colony formation assay provided additional
   evidence that the absence of KIF14 significantly compromised the
   ability of A498 cells to generate colonies (Fig. [130]7E and F).
   Consistent with these findings, KIF14 depletion led to cell cycle
   arrest and an increased level of apoptosis in A498 cells, as indicated
   by 7-AAD and Annexin-V/PI staining results (Fig. [131]7G-J).
   Furthermore, through the pre-functional enrichment analysis, we found
   that KIF14 was strongly associated with the cell cycle as shown in
   Fig. [132]3. Therefore, we examined the expression levels of CDK4 and
   CyclinD1 (CCND1) in KIF14-deficient A498 cells by qPCR, and the results
   were as we expected, the expression levels of both CDK4 and CyclinD1
   were decreased after reduced KIF14 expression in A498 cells
   (Fig. [133]7K-M). Spearman correlation analysis also showed that KIF14
   expression was positively correlated with CDK4 and cyclinD1 in the TCGA
   KIRC cohort (Fig. [134]S2). Taken together, these observations provide
   strong evidence for the oncogenic role of KIF14 in ccRCC.

Fig. 7.

   [135]Fig. 7
   [136]Open in a new tab

   The effect of KIF14 deficiency on A498 cells. (A) qPCR revealing higher
   KIF14 mRNA expression in ACHN, A498 and 786-0 compared to HKC. (B) qPCR
   analysis results showing the downregulation of KIF14 expression in A498
   cells by three different siRNAs targeting KIF14. (C) Western blotting
   data confirming the significant decrease in KIF14 expression in A498
   cells upon siRNA-mediated knockdown. (D) CCK8 assay results indicating
   reduced proliferation of KIF14-deficient A498 cells compared to control
   A498 cells. (E) Colony formation assay showing attenuated colony
   expansion capacity in A498 cells following KIF14 depletion. (F)
   Quantification of the colony formation assay results presented in (E).
   (G) 7-AAD staining data demonstrating cell cycle arrest in A498 cells
   upon KIF14 depletion. (H) Quantification of the cell cycle analysis
   results shown in (G). (I) Annexin V/PI staining outcomes revealing
   increased apoptosis in KIF14-depleted A498 cells. (J) Quantification of
   the apoptosis analysis results presented in (I). (K, L) qPCR analysis
   results demonstrating that CDK4 and CyclinD1 mRNA expression levels
   were downregulated in KIF14-deficient A498 cells compared to normal
   cells. (M) The level of CDK4 and CyclinD1 were detected by Western
   blotting. CyclinD1, CCND1. *p < 0.05, **p < 0.01, ***p < 0.001

Discussion

   Numerous literature reports have indicated that KIF14 demonstrates
   oncogenic properties in various malignant tumors; however, its
   involvement in the pathogenesis of clear cell renal cell carcinoma
   (ccRCC) has been minimally addressed in the literature. Therefore, this
   study sought to investigate the potential role of KIF14 in ccRCC.
   Initial bioinformatic analysis utilizing public databases revealed that
   KIF14 expression is significantly upregulated and serves as a
   diagnostic marker in ccRCC. Subsequent investigations demonstrated that
   elevated KIF14 expression in ccRCC is correlated with advanced
   clinicopathological characteristics, alterations in the cell cycle,
   immune infiltration, immune evasion, reduced survival rates and poorer
   prognosis. Finally, it has been shown experimentally that inhibition of
   KIF14 expression in A498 cells suppresses cell proliferation and
   renders the cells more susceptible to apoptosis. All this suggests that
   KIF14 may be a potential prognostic marker for ccRCC patients.

   KIF14 is a member of the N-type, kinesin 3 superfamily located on
   chromosome 1q32.1 and is required for cellular chromosome segregation
   and cytokinesis processes [[137]11, [138]12, [139]31]. KIF14 is
   overexpressed in a variety of cancers, including gastric [[140]32],
   hepatocellular [[141]33], colorectal [[142]15], and cervical [[143]18],
   and is strongly associated with poor prognosis. Consistent with this,
   our data showed that KIF14 was highly expressed in ccRCC tissues (shown
   in Fig. [144]1A-B and D-F) and kidney cancer cells (shown in
   Fig. [145]7A) and that KIF14 overexpression significantly correlated
   with pathologic stage, histological grade, T stage and M stage of RCC
   (shown in Fig. [146]2). The higher KIF14 expression in patients with
   ccRCC, the worse histological types, lower tumor differentiation, later
   clinicopathological stages, and poorer prognosis (shown in
   Fig. [147]5), further supporting the idea of KIF14 as an oncogene, and
   experimental validation showed that knockdown of KIF14 reduced
   proliferation, decreased colony-forming ability, impaired cell cycle
   and increased apoptosis in A498 ccRCC cells (shown in Fig. [148]7D-J).
   Although METL14 in combination with KIF14 has better diagnostic
   potential [[149]34], our results showed that KIF14 also had better
   diagnostic potential in ccRCC (shown in Fig. [150]1C). FOXI1 acts as a
   biomarker for renal intercalated cell (IC)-associated tumors and can
   partially reprogram the ccRCC cell line 786-O towards the
   transcriptional phenotype of chromophobe renal cell carcinoma (chRCC)
   [[151]35, [152]36]. Interestingly, genetic perturbation similarity
   analysis with KIF14 in KIRC showed a biological link between FOXI1 and
   KIF14 (shown in Fig. [153]1G-H), but further relevant experiments are
   needed to verify this.

   KEGG pathway analysis implicated that KIF14 is involved in cell cycle
   regulation, and low levels of CDK4 and cyclinD1 in KIF14-deficient A498
   cells compared to control cells were confirmed in subsequent
   experimental studies (shown in Fig. [154]7K-L). In addition, we also
   showed that the KIF14-high phenotype was correlated with the G2/M
   checkpoint signaling pathway, p53 signaling pathway, E2F targets,
   IL6/JAK/STAT3 and the KIF14-low phenotype was correlated with the
   oxidative phosphorylation, fatty acid metabolism (shown in
   Fig. [155]3C). Recent studies have shown that maintenance of the G2/M
   checkpoint influences the effects of immunotherapy response in kidney
   cancer [[156]37], and in a study of 103 patients with untreated clear
   cell renal cell carcinoma, downregulation of oxidative
   phosphorylation-related metabolism was among the many pathways
   dysregulated in ccRCC [[157]38]. Here, Du et al. found that abnormal
   fatty acid metabolism in ccRCC leads to activation of lipid storage
   pathways as a necessary step in the development of malignancy
   [[158]39]. This demonstrates that KIF14 may promote ccRCC cell growth,
   metastasis and poor survival through G2/M checkpoint or tumor
   metabolism signaling.

   Li et al. identified the associations between 22 types of tumor
   infiltrated immune cells and clinical outcomes in ccRCC through
   bioinformatics analysis [[159]40]. Our immune infiltration analysis
   revealed that KIF14-high tumors exhibit an immunosuppressive TME
   characterized by enriched Tregs and M1 macrophages, alongside depleted
   cytotoxic NK cells and dendritic cells (DCs) (Fig. [160]4A-D). This
   aligns with prior studies showing that Tregs suppress antitumor
   immunity by inhibiting effector T-cell activation [[161]41]. while
   tumor-associated macrophages (TAMs) promote angiogenesis and
   extracellular matrix remodeling [[162]42]. Notably, DCs are critical
   for antigen presentation and T-cell priming [[163]43]; their reduction
   in KIF14-high tumors may impair immune surveillance, facilitating
   immune escape. Mechanistically, we propose that KIF14-driven cell cycle
   dysregulation (via CyclinD1/CDK4 downregulation upon KIF14 knockdown;
   Fig. [164]7K-M) may indirectly shape the TME. Rapidly proliferating
   tumor cells (due to KIF14 overexpression) could alter metabolic demands
   (e.g., hypoxia, lactate accumulation), which recruit immunosuppressive
   cells like Tregs and M1-like macrophages [[165]44, [166]45]. Supporting
   this, our GSEA indicated that KIF14-high tumors downregulate oxidative
   phosphorylation (Fig. [167]3C), a metabolic pathway linked to immune
   evasion in ccRCC. Our analysis further revealed that KIF14 expression
   positively correlated with ImmuneScore, StromalScore, and
   ESTIMATEScore, while inversely associated with tumor purity
   (Fig. [168]4E-J, L). These findings suggest that KIF14-high tumors
   harbor a more complex tumor microenvironment (TME) enriched with both
   immune and stromal components. The negative correlation between KIF14
   and tumor purity underscores that KIF14-driven tumors are less
   dominated by malignant cells but instead infiltrated by diverse stromal
   and immune populations. This is consistent with studies showing that
   low tumor purity correlates with poor prognosis in cancers like gastric
   carcinoma [[169]46], as non-tumor cells (e.g., cancer-associated
   fibroblasts, immune cells) can facilitate therapeutic resistance
   [[170]47]. Clinically, the elevated TIDE score in KIF14-high patients
   (Fig. [171]4K) suggests heightened T-cell dysfunction, potentially
   explaining their poor response to immunotherapy. This is critical, as
   PD-1/PD-L1 inhibitors rely on pre-existing T-cell infiltration
   [[172]48]. Future studies should explore whether KIF14 inhibition could
   synergize with immune checkpoint blockade by reversing TME suppression.

   While our findings provide novel insights into KIF14’s role in ccRCC,
   several limitations must be acknowledged. First, our analyses primarily
   relied on retrospective public datasets (e.g., TCGA, GEO), which may
   introduce selection bias due to heterogeneous clinical protocols or
   incomplete follow-up data. Second, the lack of in vivo validation
   (e.g., xenograft models) limits our ability to conclusively establish
   KIF14’s oncogenic function in a physiological context. Third, while we
   identified correlations between KIF14 and immune infiltration, the
   causative mechanisms (e.g., whether KIF14 directly recruits Tregs or
   alters cytokine secretion) remain speculative. To address these gaps,
   future studies should: (1) validate KIF14’s role in vivo using
   patient-derived xenografts (PDX) or genetically engineered mouse models
   (GEMMs) to recapitulate ccRCC progression and therapy resistance; (2)
   integrate multi-omics approaches (e.g., single-cell RNA-seq, spatial
   transcriptomics) to dissect how KIF14 reshapes the tumor-stroma-immune
   ecosystem; (3) explore clinical relevance by correlating KIF14
   expression with immunotherapy response in prospective ccRCC cohorts;
   and (4) investigate therapeutic synergy between KIF14 inhibition (e.g.,
   siRNA, small-molecule inhibitors) and immune checkpoint blockade. These
   efforts could bridge the translational gap and position KIF14 as an
   actionable target for precision oncology in ccRCC.

Conclusion

   This study initially identified an upregulation of KIF14 expression in
   patients with ccRCC, with high expression levels correlating with
   disease progression, poor prognosis, and increased immune infiltration
   in ccRCC. Subsequent experiments demonstrated that knockdown of KIF14
   in A498 cells led to reduced proliferation, decreased colony formation
   capacity, cell cycle arrest, increased apoptosis, and downregulation of
   CyclinD1 and CDK4. In conclusion, these results emphasize the
   overexpression and oncogenic function of KIF14 in ccRCC, suggesting
   that KIF14 may serve as a potential therapeutic target for the
   treatment of this disease.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [173]12894_2025_1732_MOESM1_ESM.tif^ (2.5MB, tif)

   Supplementary Material 1: Supplementary Figure 1. Enrichment analysis
   via Metascape.
   [174]12894_2025_1732_MOESM2_ESM.tif^ (508.9KB, tif)

   Supplementary Material 2: Supplementary Figure 2. Correlation analysis
   between CDK4/cyclinD1 and KIF14.
   [175]Supplementary Material 3^ (2MB, tif)
   [176]Supplementary Material 4^ (1.4MB, tif)
   [177]Supplementary Material 5^ (1.6MB, tif)

Acknowledgements