A new cuproptosis-associated lncRNA signature predicts the prognosis of clear cell renal cell carcinoma patients

Abstract

Objective

Clear cell renal cell carcinoma (ccRCC) is the most common subtype of renal cell carcinoma. Cuproptosis is a new type of programmed cell death that is mediated by protein lipid acylation and is closely linked to mitochondrial metabolism.

Methods

The Cancer Genome Atlas (TCGA) database provided us with RNA-Seq data together with the related clinical and prognostic information. Using univariate Cox, 135 prognostic lncRNAs associated with cuproptosis were identified for use in prognostic model construction. Multivariate Cox analysis was subsequently used to further integrate these lncRNAs. We used subject operating characteristic (ROC) curve analysis and Kaplan-Meier (K-M) survival curve analysis to assess the model’s prognostic ability. In order to predict immune escape and potential therapeutic drugs in the two groups, we also looked at the differences in immunological and tumor mutational burden between the high- and low-risk groups. The quantitative polymerase chain reaction (Q-PCR) was then used to validate the risk model.

Results

The lncRNA profile linked to cuproptosis is used to divide patients into two risk categories, with the lower risk group having a better prognosis. The tumor mutational burden was larger, immune escape was more likely to occur, and immunotherapy results were generally worse in the high-risk group. The traditional chemotherapeutic medications sunitinib, AKT inhibitor VIII, rapamycin, and lapatinib were more effective in low-risk individuals. Six prognostic lncRNAs were ultimately confirmed in human cell lines, including HK-2, ACHN, 769-P, and CAKI-1.

Conclusions

A risk model based on six cuproptosis-related lncRNAs has a high predictive value for ccRCC and might be a medical target for cuproptosis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12672-026-04696-9.

Introduction

Renal cell carcinoma (RCC) is one of the most prevalent cancers in the urinary tract, according to recent studies [1, 2]. It is the second most common urological cancer after bladder cancer, and its most prevalent and aggressive subtype, renal clear cell carcinoma (ccRCC), makes up about 70% of all RCC [3]. Clinically, approximately one-third of patients with renal clear cell carcinoma will develop metastases at initial diagnosis, and one-quarter of patients with limited disease will develop recurrent metastases after radical surgical resection. Metastatic ccRCC is always associated with high mortality [4–6]. Renal clear cell carcinoma is still one of the most difficult clinical issues in urology, despite the use of several treatment modalities, including surgery, chemotherapy, radiation, targeted therapy, and the recently suggested immunotherapy [7]. The primary causes are a high risk of metastases and a delayed diagnosis. Consequently, there is a pressing need to create more accurate prognostic models given the morbidity and mortality of renal clear cell carcinoma [8, 9].

The presence of adequate amounts of copper in cells is essential for cell survival because copper is a necessary trace element and a catalytic cofactor for vital enzymes involved in the regulation of energy conversion, iron collection, oxygen transport, and intracellular oxidative metabolism [10]. According to recent research, cancer patients’ blood and tumor tissues have noticeably higher copper levels than those of healthy individuals. Changes in intracellular copper levels may influence the onset and spread of cancer, even though disruption of copper homeostasis may cause cytotoxicity [11]. This mechanism has led to the use of copper chelators (tretinoin, tetrathiomolybdate, etc.) and copper ion carriers (disulfiram, dithiocarbamate, elesclomol, etc.) in anticancer therapy [12]. Cuproptosis, a novel mechanism of cell death that was initially identified by Tsvetkov et al. as resulting from direct copper binding to the lipidated components of the tricarboxylic acid cycle (TCA), has received a lot of attention lately [13, 14]. This causes iron-sulfur cluster proteins to be lost and lipid acylated proteins to aggregate, which eventually results in proteotoxic stress and cell death. Reprogramming of the tricarboxylic acid cycle (TCA cycle) is typically associated with ccRCC [15]. This reprogramming downregulates energy generation through the TCA cycle, enabling tumor cells to evade the immune system and live in hypoxic and nutrient-deficient environments. Nevertheless, there are few research on cuproptosis-associated LNCRNAs in ccRCC, and it is still unknown how these LNCRNAs affect prognosis in ccRCC patients [16–18].

Therefore, the aim of our study was to identify cuproptosis-associated LNCRNAs in ccRCC and to understand the prognostic role of cuproptosis-associated LNCRNAs, which not only reveals the signaling pathways and molecular mechanisms of cuproptosis effects in ccRCC, but also provides new targets for the treatment of ccRCC.

Methods

Data retrieval and identification of cuproptosis-associated lncRNAs

We downloaded transcriptomic and clinical data for 541 ccRCC patients from the TCGA database. Copy number variation (CNV) data and somatic mutation data for ccRCC cases have also been downloaded from the TCGA data portal (https://tcgadata.nci.nih.gov/tcga/ dataaccessmatrix.htm) for download. Study participants with incomplete clinical information were excluded. Cuproptosis-related genes were obtained by literature search. Then, we assessed the correlation of cuproptosis-associated lncRNAs with cuproptosis-associated genes by Pearson correlation analysis. To identify cuproptosis-associated lncRNAs, Pearson correlation coefficients greater than 0.5 (R > 0.5) and P values less than 0.05 (P > 0.05) were required.

Construction and validation of Cuproptosis-related lncRNA features

We included 532 ccRCC patients who were randomly assigned to either the training group or the test group (Fig. 1). In the training group, a cuproptosis-related lncRNAs signature was constructed by combining the analysis of univariate Cox regression, LASSO Cox regression, and multivariate Cox regression. Finally, a risk score was calculated for each individual using this prognostic signature. The risk score was calculated using the following formula: risk score= (normalized expression level of each cuproptosis-associated lncRNA ∗ corresponding correlation coefficient). Patients in the training cohort were divided into low and high risk groups based on the median value of the risk score. Kaplan-Meier curves were generated by using the “SurvMiner” R package and log-rank test to compare overall survival between the high/low risk groups (OS). Subject operating characteristic curves (ROC) were generated to assess the accuracy of the predicted signal by the “TIMEROC” R package. To assess the feasibility of the model, risk scores were calculated in the validation cohort based on the same formula in the training cohort, and then the same validation method as described above was performed.

Fig. 1.

Identification of cuproptosis-related lncRNA. A Cuproptosis-related lncRNA by Pearson correlation analysis in TCGA database. B Forest plots show the results of Cox univariate regression: The first 27 prognostically differentially expressed Cuproptosis-associated lncRNAs C, D lncRNAs screened by the LASSO-Cox regression model. E Heat map of the correlation between six genes and Cuproptosis-related Gene in the multivariate Cox model

Creating and validating a predictive nomogram

Using the R package of “rms”, the clinical features (age, gender, grade, T stage, M stage, and N stage) and risk score were utilized to construct a prognostic nomogram to predict the 1-,3-, and 5-year OS of patients with ccRCC. Each variable in the nomogram scoring system was matched with a score, and the overall score was calculated by summing the scores from all variables in each sample. The nomogram calibration plots were utilized to show the predictive value between the forecasted 1-, 3-, and 5-year OS and the practically observed results.

Analysis of tumor immunity and tumor mutational load between patients in the high-risk and low-risk groups. We analyzed the tumor mutational load, mutational differences, and survival differences between the high and low risk groups. tumor immune dysfunction and rejection (TIDE) in predicting patient response to immunotherapy .

Functional enrichment analysis

The genes differentially expressed between the high-risk and low-risk groups were identified (|log2(fold change) |>1 and FDR < 0.05) with the ‘edgeR’ R package and functionally annotated based on the Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) with the clusterProfiler’ R package (adjusted p value < 0.05).

Analysis of tumor immunity and tumor mutational load

We analyzed differences in immune-related functions, tumor mutational load, mutational differences, and survival differences between the high- and low-risk groups. Tumor immune dysfunction and rejection (TIDE) were analyzed in predicting patient response to immunotherapy.

Quantitative polymerase reaction (q-PCR) with cell culture

The HK-2, ACHN, 769-P, and CAKI-1 cell lines were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China) and then were cultivated in DMEM (HyClone, Logan, UT, USA) media supplemented with 1% sodium penicillin G/streptomycin sulfate and 10% fetal bovine serum (FBS) (Gibco, Waltham, MA, USA) at 37 °C in a humidified environment with 5% CO2. The Rizol reagent (Invitrogen) was used to extract total RNA from the cells. After that, cDNA was extracted by reverse transcription using a cDNA synthesis combination, and quantitative PCR was used for analysis. The internal standard was GAPDH. The 2-ΔΔCt statistic was utilized to determine the levels of gene expression. Primers for SNHG3 were: forward, 5 ′-AGTGGTCGCTTCTTCTCCTTG-3 ′ and reverse, 5 ′-GATTGTCAAACCCTCCCTGTTA-3 ′.GAPDH were ：forward,5′-GGAGCGAGATCCCTCCAAAAT-3′ and reverse， 5′-GGCTGTTGTCATACTTCTCATGG-3′.

ShRNA transfection for SNHG3 knockdown in ACHN and 769-P Cells

ShRNA targeting SNHG3 or scramble controls were transfected into ACHN and 769-P cells using Lipofectamine 3000 according to the manufacturer’s instructions. After 48–72 h, cells were harvested, and knockdown efficiency was confirmed by qRT-PCR.

CCK-8 Assay for cell proliferation

Cell viability was assessed using the CCK-8 assay (Control and SNHG3 knockdown ACHN or 769-P cells were seeded in 96-well plates, incubated with CCK-8 reagent for 2 h, and absorbance at 450 nm was measured to evaluate proliferation.

Colony formation assay

Control and SNHG3 knockdown cells were seeded in 6-well plates and cultured for 10–14 days, then fixed, stained with crystal violet, and colonies were imaged and counted manually.

Flow cytometric analysis of apoptosis

Control and SNHG3 knockdown cells were stained with Annexin V-FITC/PI and analyzed by flow cytometry to quantify apoptotic cells, with results expressed as percentages of total cells.

Transwell invasion assay

Control and SNHG3 knockdown cells were seeded in Matrigel-coated Transwell inserts; after 24–48 h, invaded cells were fixed, stained with crystal violet, and counted under a microscope.

Wound healing assay

Control and SNHG3 knockdown cells were scratched at confluence, washed, and cultured in serum-free medium; wound closure was imaged at set times, and migration was quantified.

Immunofluorescence

Control and SNHG3 knockdown cells on coverslips were fixed, permeabilized, blocked, and incubated with N-cadherin and E-cadherin antibodies, followed by fluorescent secondary antibodies and DAPI; images were captured and quantified by microscopy.

Seahorse glycolysis assay

Control and SNHG3 knockdown cells were seeded in Seahorse XF plates, incubated in base medium, and subjected to the Glycolysis Stress Test with sequential glucose, oligomycin, and 2-DG injections; ECAR was measured to assess glycolytic function.

Subcutaneous tumor formation

The nude mice aged eight weeks were procured from Charles River Laboratories (Beijing, China) and kept in a specific pathogen-free environment. Four nude mice per group were injected subcutaneously with 5 × 10⁶ control or SNHG3 knockdown cells. Tumor growth was monitored every 3–4 days, and mice were euthanized after 4–6 weeks for tumor excision and final weight measurement.

Mice were anesthetized with 2–3% isoflurane in oxygen during all surgical and imaging procedures. For euthanasia, mice were first deeply anesthetized with isoflurane and then sacrificed by cervical dislocation. All procedures were performed to minimize animal suffering in accordance with the institutional ethical guidelines.

According to the ethical guidelines, the maximal tumor size permitted by the ethics committee was 2 cm in diameter or 2000 mm³ in volume. During the experiments, the tumor burden in all animals did not exceed this limit.

Western blotting

Cells were lysed on ice using RIPA buffer (Beyotime, Shanghai, China) supplemented with 1% phosphatase inhibitor (Servicebio, Wuhan, China). After 30 min on ice, the lysates were centrifuged at 13,000 rpm for 15 min, and the supernatants were collected for protein quantification with a BCA assay. Equalized samples were mixed with loading buffer, boiled at 100 °C for 15 min, and stored at − 80 °C. Proteins were separated on 12% SDS–PAGE gels and transferred onto PVDF membranes (Millipore, NJ, USA). Membranes were blocked for 15 min, incubated with primary antibodies overnight at 4 °C, and then with secondary antibodies for 70 min. Protein signals were detected using an ECL system (Beyotime, Shanghai, China) and quantified with ImageJ software. Antibodies are detailed in Table S1.

Statistical analysis

R was used to conduct all statistical analyses (v4.0.5). If the statistical significance level was not specifically indicated, it was assumed to be P < 0.05.

Results

Construction of a cuproptosis-associated lncRNA signature

Creation of a lncRNA Signature Associated with Cuproptosis. We recruited 532 ccRCC patients, who were randomized in a ratio of almost 1:1 to either the training group (n = 266) or the testing group (n = 266). Table S1 lists all of the patients’ clinical information. The clinical features of the training group and the testing group did not differ statistically significantly. After 19 cuproptosis-associated genes were identified from the literature study, 286 cuproptosis-associated lncRNAs were found by calculating Pearson’s correlation with these genes (Fig. 1A). First, we identified 135 cuproptosis-associated lncRNAs with a predictive significance in our training group using the univariate Cox regression analysis (Fig. 1B). Second, 16 lncRNAs were chosen after multicollinearity was reduced using LASSO Cox regression (Fig. 1C, D). Third, six cuproptosis-associated lncRNAs (SNHG3, SMARCA5-AS1, LINC01711, SBF2-AS1, CDK6-AS1, PINK1-AS) are identified for prognosis based on the lowest AIC in a later multivariate analysis. The expression levels of each lncRNA are used to calculate the risk score, which is as follows: SNHG3* SMARCA5-AS1, LINC01711, SBF2-AS1, CDK6-AS1, PINK1-AS (Fig. 1E). Displays the correlation between six genes and copper death genes. The heatmap in Fig. 2 provides a visual overview of the clustering patterns, showing a distinct separation in expression levels between normal and tumor samples. Furthermore, the plots in Fig. 2B–G reveal that all six lncRNAs are significantly upregulated in KIRC tissues, suggesting that these cuproptosis-related genes may play a critical role in the oncogenesis or progression of the disease.Using the median risk score, patients in the training group were divided into two groups: high-risk (n = 130) and low-risk (n = 136). . By integrating expression profiling (Fig. 3A-C), risk scoring (Fig. 3D-F), and longitudinal survival data (Fig. 3G-L), the results demonstrate that high expression of these six lncRNAs correlates with elevated risk scores and a significantly higher probability of mortality. The high degree of consistency between the training and testing cohorts underscores the model's reliability and its potential utility as a molecular diagnostic tool for risk stratification and personalized management in renal clear cell carcinoma. Figure 4B, E, and J show the lncRNA expression levels, risk status, and survival result for each patient. The OS in the high-risk group was clearly poorer than that in the low-risk group, as shown by the Kaplan–Meier analysis (Fig. 4L).

Fig. 2.

Expression levels and differences of six Cuproptosis-associated lncRNAs in KIRC and normal tissues. A The expression levels of six Cuproptosis-related lncRNAs. B–G Differential expression of SNHG3, SMARCA5-AS1, LINC01711, SBF2-AS1, CDK6-AS1 and PINK1-AS in KIRC and normal tissues

Fig. 3.

Prognostic signature of the riskScore analyses of the 6 Cuproptosis -related lncRNAs in the TCGA overall, training and testing groups. A–C Clustering analysis heatmap in the overall, training and testing groups (D–F) The distribution of the risk scores in the overall, training and testing groups. G–I The distributions of overall survival status and risk score in the overall, training and testing groups (J–L) The Kaplan–Meier curves for survival status and survival time in the overall, training and testing groups

Fig. 4.

The correlation between the predictive signature and the prognosis of KIRC patients. A Kaplan-Meier analysis of the progression free survival rate of KIRC patients in the high and low-risk groups. B Univariate Cox regression analysis of the clinical characteristics and riskScore with the OS. C Multivariate analysis of the clinical characteristics and riskScore with the OS. D AUC of ROC curves comparing the prognostic accuracy of the risk score and other prognostic factors in the overall groups. E The receiver operating characteristic (ROC) curve shows the potential of the prognostic Cuproptosis-related lncRNAs signature in predicting 1-, 2-, and 3-year overall survival (OS) in the overall groups. F Index curves for risk scores and other prognostic factors

Verification of the cuproptosis-associated lncRNA signature

Using the same technique as in the training group, we determined a risk score for every member of the testing group in order to confirm the correctness of the cuproptosis-associated lncRNAs signature. Next, using the same cut-off values as the training group, the patients were divided into two groups: a high-risk group (n = 133) and a low-risk group (n = 133). Figure 4C, F, and I show the lncRNA expression levels, risk status, and survival result for each patient. For ccRCC patients in the high-risk category, the Kaplan–Meier analysis findings showed a comparatively dismal prognosis (Fig. 4L). Lastly, we compared all patients using the same cut-off values and got consistent findings, much as the training and testing groups. Figure 4A, D, and G show the lncRNA expression levels, risk status, and survival result for each patient. For ccRCC patients in the high-risk category, the Kaplan–Meier analysis findings showed a comparatively dismal prognosis (Fig. 4J).

Correlation between the predictive signature and the prognosis of ccRCC patients

Cox regression analysis was used to ascertain if the predictive signature is a stand-alone prognostic factor for patients with ccRCC. Age, stage, grade, and risk score were all substantially correlated with the OS of BC patients, according to univariate Cox regression analysis (Fig. 5B). Age, stage, grade, and risk score were all independent predictors of OS in patients with ccRCC, according to multivariate Cox regression analysis (Fig. 5C). Age, stage, grade, and risk score had the following AUCs: 0.653, 0.799, 0.720, and 0.744. Good predictive performance was demonstrated by the AUCs of 1, 3, and 5-year survival, which were 0.747, 0.713, and 0.761, respectively (Fig. 5E).

Fig. 5.

The development and assessment of a predictive nomogram. A The nomogram forecasts the possibility of 1-, 3-, and 5-year overall survival. B The calibration plot of the nomogram forecasts the likelihood of the 1-, 3-, and 5-year overall survival

Construction of a nomogram to predict patients’ survival

We subsequently integrated the cuproptosis-associated lncRNA risk scores with clinicopathological features to develop a hybrid nomogram model for predicting 1-, 3-, and 5-year OS in light of the inconvenient clinical utility of the cuproptosis-associated lncRNA risk score in predicting OS of patients with ccRCC (Fig. 6A). The risk score and gender were among the predictors. According to the following calibration plots, the suggested model’s performance was comparable to that of an ideal model (Fig. 6B).

Fig. 6.

Kaplan-Meier survival curves of high- and low-risk groups among patients sorted according to different clinicopathological variables. A, B Age. C, D Gender. E, F T stage. G, H M stage. I, J Grade K, L Stage

Relationship between the predictive signature and the prognosis of ccRCC patients in different clinicopathological variables

CcRCC patients were divided into groups according on age, grade, stage, T stage, and M stage in order to examine the association between the predictive signature and the prognosis of patients sorted based on various clinicopathological factors. The OS of patients in the high-risk group was significantly shorter than that of individuals in the low-risk group for each of the other classes (Fig. 7). These findings imply that the predictive signature, independent of clinicopathological factors, may forecast the prognosis of patients with ccRCC.

Fig. 7.

Principal component analysis between the low-risk and high-risk subgroups based on all genes, cuproptosisgene, cuproptosisLncRNA, and risk LncRNA (A–D)

Principal component analysis (PCA) using the “ggplot2” R package is used to determine the difference between high-risk and low-risk individuals based on the entire examined genes (Fig. 8A), 19 cuproptosis-associated genes (Fig. 8B), 286 cuproptosis-associated lncRNAs (Fig. 8C), and 6 cuproptosis-associated lncRNAs of the signature (Fig. 8D). The findings demonstrated that patients at high and low risk have a comparatively broad spectrum of gene expression.

Fig. 8.

Results of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. GO (A, C) and KEGG (B, D) analysis of high-risk group and low-risk group based on the cuproptosis-related lncRNA prognostic signature

Functional enrichment analysis

Differentially expressed genes (DEGs) in the high- and low-risk groups were analyzed using KEGG pathway analyses and GO enrichment. The results of KEGG pathway analyses revealed that DEGs in the high-risk and low-risk groups were primarily enriched in biological processes related to immunity, such as humoral immunity, lymphocyte-mediated immunity, and adaptive immunity (Fig. 9A, C). The results of KEGG pathway analyses showed that DEGs in the high-risk and low-risk groups tended to be enriched in numerous pathways, including PPAR signaling pathway, complement and coagulation cascades, PI3K-Akt signaling pathway, and cholesterol metabolism (Fig. 9B, D).

Fig. 9.

Genetic alterations and tumor mutation load analysis. A, B Top 10 most substantially changed genes in the high- and low-risk score subgroups. (C) Differences in tumor mutational burden between high- and low-risk groups. (D) Kaplan-Meier survival curves of high and low tur mutation burden groups (E) Kaplan-Meier survival curves of tumor mutational burden combined with risk scores

Cancer-related gene mutation and tumor immunology research

Next, we looked at how the somatic mutation distribution changed between the high-risk and low-risk groups. We found that the high-risk groups had a mutation rate of 111 (82.34%) of 134 samples, with the top three mutated genes being VHL, PBRM1, and TTN, while the low-risk groups had a mutation rate of 150 (77.32%) of 194 samples, with the top three mutated genes being VHL, PBRM1, and TTN. Notably, compared to those at high risk, a considerably greater number of VHL and TTN mutations were found in low-risk patients. On the other hand, PBRM1 mutation levels showed the exact opposite pattern (Fig. 10A, B). TMB is a marker linked to improved ICB therapy response. According to the examination of patient mutation data, the high-risk group had a greater TMB than the low-risk group (Fig. 10C), which may indicate that immunotherapy would be beneficial for the high-risk group. Patients with a greater TMB had a superior overall survival rate (Fig. 10D). The TMBs in the high-risk and low-risk groups were further contrasted. According to Fig. 10E, patients in the high-risk group with lower TMB had the poorest chance of surviving, while those in the low-risk group with higher TMB had the greatest chance. The two primary pathways of tumor immune escape may be modeled using the TIDE computational framework, which can yield immunotherapy prediction findings. In order to better demonstrate the predictive power of risk scores for immunotherapy, we applied TIDE to our patients. TIDE was positive correlated with the risk scores (Fig. 11B). In addition, we analyzed the immune cell-related functions in the high and low analysis groups. According to the correlation analysis based on GSVA package ssGSEA, immune-related functions such as type II IFN_response, APC_co_stimulation, CCR, parainflammation, APC\u cou inhibition, HLA, cytolytic activity, checkpoint, T\u cell co-stimulation, inflammation promotion, T\U cell co-inhibition, MHC\u class I and type I IFN\u response in the low-risk group were significantly more abundant (Fig. 11A). To further explore the difference in the two risk groups about the drug resistance potential. We compared the estimated IC50 levels of 50 chemotherapy drugs or inhibitors in the two groups. Among those, 11 representative drugs are shown in Fig. 11C–K. We found that sunitinib, AKT inhibitor VIII, rapamycin, lapatinib, mitomycin, parthenolide, PAC–1, midostaurin, obatoclax mesylate may not be ideal for patients in the high-risk group.

Fig. 10.

A Heatmap of ssGSEA for the association between immune cell subpopulations and related functions. B Prediction of immune escape between high and low risk groups (C–K) Boxplot showing the mean differences in estimated IC50 values of 9 representative drugs (sunitinib, AKT inhibitor VIII, rapamycin, lapatinib, mitomycin, parthenolide, PAC–1, midostaurin, obatoclax mesylate) between the two risk groups

Fig. 11.

Overall survival and ROC curve. A–F Overall survival analysis of cuproptosis-associated lncRNAs. G–L ROC curve of cuproptosis-associated lncRNAs

SNHG3 is required for ccRCC cell proliferation and survival

Among the candidate lncRNAs, SNHG3 deserves special attention. In hepatocellular carcinoma, SNHG3 can sponge miR-1306-5p via the competing endogenous RNA (ceRNA) mechanism, thereby upregulating the expression of pyruvate dehydrogenase E1 subunit alpha 1 (PDHA1). Notably, PDHA1, together with FDX1, LIAS, and DLAT, are all key regulators of the tricarboxylic acid cycle and the cuproptosis pathway, and their dysfunction directly affects the copper ion-mediated cell death process [19]. Furthermore, in various cancers including clear cell renal cell carcinoma (ccRCC), SNHG3 generally regulates target gene expression via the ceRNA mechanism to promote tumor progression. The conservation of this regulatory pattern further supports that SNHG3 may regulate cuproptosis-related genes in ccRCC through a similar pathway [20]. Most importantly, SNHG3 is closely associated with glycolysis, yet its function in clear cell renal cell carcinoma has not been fully investigated [21–23]. To investigate the biological role of SNHG3 in renal cell carcinoma, we first examined its expression in several RCC cell lines. Quantitative RT–PCR analysis revealed that SNHG3 expression was markedly elevated in ACHN, 769-P, and Caki-1 cells compared with normal renal tubular epithelial HK-2 cells (Fig. 12A). To elucidate its function, SNHG3 was silenced in ACHN and 769-P cells using two independent short hairpin RNAs (shRNAs), both of which significantly reduced SNHG3 transcript levels (Fig. 12B, C).

Fig. 12.

Expression and functional impact of SNHG3 on proliferation and apoptosis in ccRCC. A Expression levels of SNHG3 in HK-2, ACHN, 769-P, and CAKI-1 cell lines. B, C The efficiency of SNHG3 knockdown was verified by qRT-PCR. D, E The effect of SNHG3 knockdown on the proliferation of ACHN and 769-P cells was assessed by the CCK-8 assay. F–I The effect of SNHG3 knockdown on the proliferation of ACHN and 769-P cells was evaluated by colony formation assay. J–M Representative immunofluorescence images of Ki67 in control and SNHG3 knockdown groups, scale bar: 50 μm. N–Q Representative flow cytometry plots of apoptosis in control and SNHG3 knockdown groups. R, S Representative Western blot of Bax, Bcl2, and cleaved-caspase-3 in control and SNHG3 knockdown groups. *P < 0.05, **P < 0.01, ***P < 0.001

CCK-8 assays demonstrated that SNHG3 knockdown markedly inhibited cell proliferation in both ACHN and 769-P cells in a time-dependent manner (Fig. 12D, E). Consistently, colony formation assays revealed a significant reduction in the number and size of colonies following SNHG3 silencing (Fig. 12F–I). Immunofluorescence staining for Ki67 further confirmed a decrease in proliferating cells after SNHG3 depletion (Fig. 12J–M).

Flow cytometric analysis indicated a substantial increase in apoptotic cell populations upon SNHG3 knockdown in both cell lines (Fig. 12N–Q). Western blotting showed downregulation of the anti-apoptotic protein Bcl-2 and upregulation of pro-apoptotic Bax and cleaved caspase-3 (Figure 12R, S), further supporting that SNHG3 depletion induces apoptosis in ccRCC cells. Collectively, these findings suggest that SNHG3 acts as an oncogenic lncRNA that promotes cell proliferation and inhibits apoptosis in ccRCC.

SNHG3 drives ccRCC progression by reprogramming glycolytic metabolism to facilitate migration, invasion, EMT, and tumor growth

Transwell invasion assays (Fig. 13A–D) showed that compared with the control group, the number of invasive cells in ACHN and 769-P cells with shRNA1 or shRNA2 knockdown was significantly reduced, indicating impaired invasive capacity. For cell migration, wound healing assays (Fig. 13E–H) demonstrated that shRNA treated cells exhibited slower wound closure from 0 to 24 h, with lower migration rates than the control group in both cell lines.

Fig. 13.

The Role of SNHG3 in Invasion, Migration, Glycolysis, and Subcutaneous Tumorigenesis in ccRCC. A–D Representative images of the cell invasion assay using Matrigel-coated Transwell chambers in control and SNHG3 knockdown groups, scale bar: 100 μm. E–H Representative images of the wound healing assay at 0 h and 24 h in control and SNHG3 knockdown groups, scale bar: 200 μm. I, J Representative Western blot images of E-cadherin, N-cadherin, and Vimentin in control and SNHG3 knockdown groups. K–R Representative immunofluorescence images of E-cadherin and N-cadherin in control and SNHG3 knockdown groups, scale bar: 50 μm. S, T Measurement of glycolytic function by the extracellular acidification rate (ECAR) in control and SNHG3 knockdown groups using the Seahorse XF Analyzer. W Representative images of subcutaneous xenograft tumors in nude mice and quantitative analysis of final tumor weights. X Representative IHC staining of the proliferation marker Ki67 in tumor sections, scale bar: 50 μm. *P < 0.05, **P < 0.01, ***P < 0.001

Western blot analysis (Fig. 13I, J) revealed that in shRNA-treated ACHN and 769-P cells, the expression of the epithelial marker E-cadherin was upregulated, while the mesenchymal markers N-cadherin and Vimentin were downregulated, suggesting a reversal of EMT. Immunofluorescence staining (Fig. 13K–R) further confirmed these changes: N-cadherin fluorescence intensity decreased, and E-cadherin intensity increased in shRNA-treated ACHN (Fig. 13K–N) and 769-P (Fig. 13O–R) cells, consistent with the Western blot results. Extracellular acidification rate (ECAR) measurements (Fig. 13S–T) showed that shRNA knockdown led to a decrease in glycolytic activity, as reflected by reduced ECAR over time and lower glycolysis rates in treated cells compared with the control.

In vivo tumorigenicity assays (Fig. 13W) indicated that shRNA-treated groups had smaller tumor sizes and lower weights. Immunohistochemistry for Ki67, a proliferation marker, showed fewer Ki67-positive cells in shRNA-treated tumor tissues (Figure 13X), confirming reduced tumor cell proliferation.

Collectively, these results suggest that the SNHG3 promotes renal cancer cell invasion, migration, EMT, glycolysis, and in vivo tumor growth.

Discussion

The prognosis for individuals with ccRCC is still dismal even though vigorous multidisciplinary treatment—which includes surgery, radiation, chemotherapy, and immunotherapy—significantly increases survival [1, 24]. Patients’ prognoses and treatment results might differ significantly, even if they share the same clinical risk factor. Consequently, the diagnosis and therapy of ccRCC depend heavily on the discovery of efficacious therapeutic targets [25, 26]. Deciphering the molecular mechanism of copper deposition within tumor cells may reveal new therapeutic targets, since Tsvetkov et al. recently discovered that cuproptosis represents a unique form of programmed cell death with multiple roles in tumor growth and treatment. Numerous lncRNAs are produced by human cells and are crucial for a number of biological activities, such as cell differentiation and genome expression [27, 28]. According to recent research, abnormal LncRNA expression may contribute to the development and progression of cancer [29]. Nevertheless, not much research has looked at lncRNAs linked to cuproptosis. Notably, this is the first thorough investigation of how lncRNAs linked to cuproptosis contribute to the development of ccRCC.

Before identifying 135 noteworthy ferroptosis-associated lncRNAs, we initially found 286 ferroptosis-associated lncRNAs in this work. A model of six ferroptosis-associated lncRNA signatures was eventually developed. Patients with ccRCC were then divided into low-risk and high-risk groups with significantly different OS based on the medium risk score. Furthermore, after adjusting for conventional clinical risk factors, it was demonstrated that the six-ferroptosis associated lncRNA signature model was an independent predictor of ccRCC. This suggested that the prognosis of patients with ccRCC might be accurately predicted by the six-ferroptosis-related lncRNA signature. The variations between the high- and low-risk groups’ tumor gene mutations, immune cell-related functions, tumor mutational burden, and TIDED were next examined.

Our research discovered a potential biomarker and treatment target in the Copper Death signaling system. Lastly, we validated the risk model using q-PCR.

Prognosis prediction is made more challenging by the fact that CcRCC is a very diverse cancer. The prognosis of patients with ccRCC and other cancers is increasingly being considered in relation to cuprotosis. Numerous research have examined the effects of cuprotosis on tumor growth and treatment. It is anticipated that lncRNAs will be a viable and legitimate molecular target for the therapy of ccRCC, as research has shown that they significantly influence the prognosis of the disease. Numerous prognostic signature models for cancers based on ferroptosis and lncRNA have been described to far. Notably, recent studies have demonstrated that a number of lncRNAs can promote ferroptosis, which in turn can regulate the incidence and progression of illnesses. Currently, several cancers have been shown to use the cuproptosis-associated lncRNA prognostic signature models. The Cuprotosis-associated lncRNA prognostic signature concept in ccRCC, however, has hardly been studied. In order to accurately predict the prognosis of patients with ccRCC, we first developed a six-ferroptosis-associated lncRNA prognostic signature model. Three high-risk genes (SNHG3, SBF2-AS1, and CDK6-AS1) and three low-risk genes (SMARCA5-AS1, LINC01711, and PINK1-AS) were included in the six-cuprotosis-associated lncRNA signature model that was developed in this investigation. According to research, patients with ccRCC who had high expressions of the high-risk genes SNHG3 and SBF2-AS1 had a shorter overall survival time and a poorer prognosis, which is in line with our results [30–32]. Our results are in line with research showing that ccRCC patients with high expressions of SMARCA5-AS1, LINC01711, and PINK1-AS among the low-risk genes had a longer OS and a better prognosis.

Tumor staging and tumor grading are important factors to take into account when estimating the prognosis of patients with ccRCC, as is well known. The American Joint Committee on Cancer (AJCC)-TNM, which is often used by surgeons and has a limited prognostic prediction value of ccRCC patients, is one of the many staging systems currently developed for the prognosis prediction of ccRCC patients [33, 34]. It’s interesting to note that we discovered that clinical characteristics can also predict ccRCC patients’ OS. However, multivariate Cox regression analysis showed that the risk model’s anticipated impact was near to the clinicopathological characteristics. The risk model outperforms clinical characteristics in predicting the prognosis of ccRCC, as evidenced by the fact that the AUC of the risk score was close to clinical features. Therefore, our results showed that in patients with ccRCC, the unique six-ferroptosis-related lncRNA signature was a strong predictor of OS.

Recent studies have discovered a number of aberrant signaling pathways in ccRCC. Certain aberrant molecules of these signals, such the PI3K/AKT-route, P53 signaling pathway, and Wnt targets signaling system, may be utilized to find innovative treatments. In this study, we discovered that the low-risk and high-risk subgroups have distinct immunoregulatory mechanisms, which might help direct future ccRCC therapy. Future therapy of individuals with ccRCC may also be guided by the variable anti-tumor immunity of patients in the high-risk and low-risk categories.

Immunotherapy is currently a promising new treatment option for ccRCC. Immunotherapy based on immune checkpoint inhibition, however, did not cause any reaction in most patients. Cuprotosis induction is intimately associated with anti-tumor immunity, as it not only destroys tumor cells via ICI-activated T cells but also directly modifies the activity of other cancer immune synergistic cells, suggesting treatment. Even in ICI-resistant types, cuprotosis and ICIs can work in concert to increase antitumor activity. Research on the connection between ICI and cuprotosis is still scarce. To investigate the relationship between cuprotosis and ICIs, a six-cuprotosis-related lncRNA signature was created [35]. The majority of ICIs were expressed at higher levels in the high-risk segment of our research than in the low-risk category. This suggested that immune checkpoint blockade immunotherapy may be guided by the six-cuprotosis-related lncRNA signature, which could also be used to predict the degree of ICIs expression. Combining ICIs with cuprotosis inducers may encourage malignant cell ferroptosis in high-risk ccRCC patients, improving their prognosis overall. For individuals with ccRCC, this combination of ICIs and cuprotosis inducers may thus result in new treatment choices down the road. To this end, we selected SNHG3 from the candidate lncRNAs for further investigation. A series of in vitro and in vivo experiments collectively demonstrated its role in promoting ccRCC progression.

There are many limitations to our research. The TCGA public database is the source of the research data, and further prospective, multicenter, and practical data are required to validate our model. Second, the trustworthiness of our findings is questionable since clinical samples were not used to validate the research findings. Furthermore, considering the limits of clinical data, the results should be used with caution.

Conclusions

In conclusion, our research demonstrates that lncRNA linked to six-cuprotosis may accurately predict the prognosis of individuals with ccRCC. Furthermore, this study may offer hints for enhancing anti-tumor immunity and developing innovative treatment approaches for ccRCC.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Mu He: Investigation, Methodology, Software, Writing—original draft. Wei Hu: Data curation, Software, Writing—original draft. Yi Guan and Yan Leng: Data curation, Software, Writing—original draft. Wenjing Shu: Supervision, Validation, Writing—original draft. Hengcheng Zhu and Kang Yang: Investigation, Supervision, Writing—review and editing.

Funding

This work was supported by the National Natural Science Foundation of China (No. 82100703) and the Natural Foundation of Hubei Province (No. 2022CFC015).

Data availability

The data used in our study were obtained from the The Cancer Genome Atlas Program (TCGA). (https://portal.gdc.cancer.gov/, accession ID: TCGA-KIRC-phs000178).

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Laboratory Animal Welfare Ethics Committee of Renmin Hospital of Wuhan University (Approval No. WDRM-2024032) and were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and the institutional regulations of Renmin Hospital of Wuhan University.

Consent for publication

All participants agree to their publication.

Generative AI statement

The authors declare that no Generative AI was used in the creation of this manuscript.

Competing interests

The authors declare no competing interests.