Cuproptosis-Related MiR-21-5p/FDX1 Axis in Clear Cell Renal Cell Carcinoma and Its Potential Impact on Tumor Microenvironment

Abstract

As a newly identified type of programmed cell death, cuproptosis may have an impact on cancer development, including clear cell renal cell carcinoma (ccRCC). Herein, we first noticed that the expression levels of cuproptosis regulators exhibited a tight correlation with the clinicopathological characteristics of ccRCC. The cuproptosis-sensitive sub-type (CSS), classified via consensus clustering analysis, harbored a higher overall survival rate compared to the cuproptosis-resistant sub-type (CRS), which may have resulted from the differential infiltration of immune cells. FDX1, the cuproptosis master regulator, was experimentally determined as a tumor suppressor in ccRCC cells by suppressing the cell growth and cell invasion of ACHN and OSRC-2 cells in a cuproptosis-dependent and -independent manner. The results from IHC staining also demonstrated that FDX1 expression was negatively correlated with ccRCC tumor initiation and progression. Furthermore, we identified the miR-21-5p/FDX1 axis in ccRCC and experimentally verified that miR-21-5p directly binds the 3'-UTR of FDX1 to mediate its degradation. Consequently, a miR-21-5p inhibitor suppressed the cell growth and cell invasion of ACHN and OSRC-2 cells, which could be compensated by FDX1 knockdown, reinforcing the functional linkage between miR-21-5p and FDX1 in ccRCC. Finally, we evaluated the ccRCC tumor microenvironment under the miR-21-5p/FDX1 axis and noted that this axis was strongly associated with the infiltration of immune cells such as CD4⁺ T cells, Treg cells, and macrophages, suggesting that this signaling axis may alter microenvironmental components to drive ccRCC progression. Overall, this study constructed the miR-21-5p/FDX1 axis in ccRCC and analyzed its potential impact on the tumor microenvironment, providing valuable insights to improve current ccRCC management.

Keywords: FDX1; cuproptosis; miR-21-5p; microenvironment; renal cell carcinoma.

Figure 1

Clinical significance of cuproptosis regulators in ccRCC. (A) Expression levels of cuproptosis regulators between ccRCC and normal kidney tissues. (B) Methylation on promoter regions of cuproptosis regulators between ccRCC and normal kidney tissues. (C–I) High expression of FDX1 (C), DLAT (D), DLD (E), PDHB (F), LIAS (G), LIPT1 (H), and PDHA1 (I) predicted longer survival in ccRCC according to Kaplan–Meier survival analyses. * p < 0.05, *** p < 0.001, **** p < 0.0001; n.s. = no significance.

Figure 2

Consensus clustering analysis to evaluate the potential role of cuproptosis in ccRCC. (A) CDF and delta area analysis of category number k. (B) Consensus matrix when k = 2. (C) PCA analysis of CSS and CRS. (D) Expression heatmap of cuproptosis regulators in CSS and CRS. (E) A comparison of clinicopathological features between CSS and CRS. (F) CSS patients experienced longer survival as compared to CRS counterparts. (G) Estimated proportion of immune cells in CSS patients. (H) Estimated proportion of immune cells in CRS patients. (I) Comparison of immune cell infiltration between CSS and CRS patients. * p < 0.05, *** p < 0.001, **** p < 0.0001; n.s. = no significance.

Figure 3

Cuproptosis risk score has a prognostic value in ccRCC. (A) Multivariate Cox regression analysis of seven cuproptosis regulators. (B) Cuproptosis_score high patients harbored longer survival as compared to cuproptosis_score low controls. (C) A fitness test of cuproptosis risk score in another independent dataset. (D) AUC analysis to evaluate the analysis of C. (E) Correlation between the CD⁴⁺ T memory resting cell population and the cuproptosis risk score. (F) Correlation between the T regulatory Treg cell population and the cuproptosis risk score.

Figure 4

Experimental validation supports the tumor-suppressive role of FDX1 in ccRCC. (A) Western blotting to examine FDX1 expression in 10 paired ccRCC samples. (B) IHC staining of FDX1 expression in ccRCC tissue microarray (27/62 ccRCC with the paired adjacent tissues). Left, representative image of IHC staining. Right, statistical analysis of IHC staining. Scale bar: 200 μm for the left images and 100 μm for the right images. (C,D) Knockdown efficiency of FDX1 in OSRC-2 cells via qPCR (C) and WB (D). (E) Knockdown of FDX1 significantly increased cell growth of OSRC-2 and ACHN cells. (F) Knockdown of FDX1 remarkably increased cell invasion of OSRC-2 and ACHN cells. Left, representative image of invaded cells. Right, statistical analysis of invaded cells. Scale bar = 100 μm. (G) Top, GSEA analysis showed that JAK_STAT signaling was enriched in FDX1_low ccRCC patients. Bottom, FDX1 siRNA activated STAT3 signaling in OSRC-2 cells. GAPDH served as a loading control. * p < 0.05, ** p < 0.01, **** p < 0.0001.

Figure 5

FDX1 correlates with the infiltration of CD4⁺ T cells in ccRCC. (A) Representative IHC images of FDX1 in different histological T stages of ccRCC. Scale bar: 200 μm for the top images and 100 μm for the bottom images. (B) Representative IHC images of FDX1 in different grades of ccRCC. Scale bar: 200 μm for the top images and 100 μm for the bottom images. (C) FDX1 was expressed at a higher level in ccRCC patients with histological T stage < T2 (left) or grade G1 + G2 (right) compared to their corresponding counterparts. n = 25 in T stage < T2; n = 26 in stage ≥ T2; n = 25 in G1 + G2; n = 26 in G3 + G4. (D) The FDX1 IHC score was positively correlated with the infiltration of CD4⁺ T cells in ccRCC (r = 0.2662, p = 0.03, n = 62). (E) Representative images of the correlation between FDX1 intensity and CD4⁺ T cell infiltration. Scale bar: 200 μm for the top images and 100 μm for the bottom images. (F) Immune cell infiltration analysis of the TCGA-KIRC dataset using TIMER 2.0 showed that FDX1 expression was positively correlated with CD4⁺ and CD8⁺ T cell populations. * p < 0.05.

Figure 6

Identification of miR-21-5p as the upstream regulator of FDX1. (A) A heap map of miRNAs differentially expressed between normal kidney tissues and ccRCC. (B) Prognosis-based LASSO analysis of the differential miRNAs. (C) A list of 14 miRNAs highly related to the prognosis. (D) A Venn diagram of 14 miRNAs and FDX1-targeted miRNAs. (E) High expression of miR-21-5p was associated with a short overall survival of ccRCC. (F) Top, the predicted binding site between the 3′-UTR of FDX1 and miR-21-5p. Bottom, miR-21-5p inhibitor treatment elevated FDX1 expression in both ACHN and OSRC-2 cells. GAPDH was the loading control. (G) Luciferase report assay showed that miR-21-5p mimics remarkably suppressed the activity of 3′-UTR of FDX1. (H) miR-21-5p inhibitor-suppressed cell growth of ACHN and OSRC-2 cells could be rescued by FDX1 knockdown. (I) miR-21-5p inhibitor-suppressed cell invasion of OSRC-2 cells could be blocked by FDX1 knockdown. Left, representative invading cells. Right, statistical analysis. Scale bar = 100 μm. (J) Cu²⁺ level in OSRC-2 cells before and after miR-21-5p inhibitor or siFDX1 treatment. (K) miR-21-5p expression was inversely correlated with FDX1 in TCGA-KIRC. * p < 0.05, ** p < 0.01, *** p < 0.001; n.s. = no significance.

Figure 7

Immune cell infiltration and prognostic prediction model under the miR-21-5p/FDX1 axis. (A) Estimated proportion of tumor microenvironmental components in miR-21-5p_low ccRCC patients. (B) Estimated proportion of tumor microenvironmental components in miR-21-5p_high ccRCC. (C) Comparison of tumor microenvironmental components between miR-21-5p_high and miR-21-5p_low ccRCC. (D) Correlation of miR-21-5p/FDX1 with immune checkpoints. **p < 0.01; ****p < 0.0001.