ccRCC is generally involved with the hypoxia circumstance along with the reprogrammed TCA cycle which gradually confers the tumor cells in an immune escaping milieu[11, 12]. Recently, many scientists have devoted their effort to investigating the relationship between cuproptosis and tumors. However, the influence of cuproptosis in ccRCC TME remains elusive. In the present study, we conducted an innovative cuproptosis- and immune-infiltration-related prognostic risk score via hub genes screened out from 10 cuproptosis-related and 40 TME-pathway-related genes in ccRCC tissue. By reinforcing the prognosis score in the validation cohort and entire cohort with OS, the hub genes show a convincing association with the outcome of ccRCC patients. Subsequently, we identified that these hub genes were associated with mTORC1 signaling, glycolysis pathway and the mitotic spindle pathway enrichment, along with the alterations of immune infiltration in ccRCC TME.
Compared with kidney tissues, it is consistent with the findings of Bian et al.[9] that 6 genes exhibited a significant down-regulation in tumor tissues, while CDKN2A is the only up-regulated gene, which reveals a potential relationship between ccRCC and cuproptosis. However, after combining them with the TME-pathway signatures, only FDX1, PDHB and CDKN2A were selected as the hub genes. Expressed in the matrix of human mitochondria, FDX1 residues closely resemble iron-sulfur cluster and plays a vital role as an electron donor in this complex [13], which participates in the reduction progress of Cu2+ to its toxic form Cu+ [14]. Recently, Peter Tsvetkov et al. proved that FDX1 and 6 other protein lipoylation, including PDHA, are the critical regulators of cuproptosis[6], among which our results exhibited that only FDX1 and PDHB were possibly associated with the immune infiltration of ccRCC TME and had a significant impact on the OS of ccRCC patients. PDHB encodes a mitochondrial multienzyme complex that catalyzes the overall conversion of pyruvate to acetyl-CoA and carbon dioxide[15], and provides the primary link between glycolysis and the TCA cycle[16, 17]. As a protein-coding gene, CDKN2A, a famous tumor-suppressing gene, is reported to be involved in at least 3 alternatively spliced variants encoding distinct proteins, and the mutation and deletion of CDKN2A are linked with various types of cancers[18–20]. According to the risk score, High-Score patients displayed a lower score for the essential genes of cuproptosis, FDX1 and PDHB, which is consistent with the work of Peter Tsvetkov et al. who detected the expression of FDX1 in several cancer cell lines and found the abundance of FDX1 and lipoylated proteins, such as PDHB, which is associated with the sensitivity of cuproptosis.
In addition to the cuproptosis-related genes, our risk score also contains 7 genes of the TME pathway. IL2RA and IL2RG are two IL-2 receptor protein coding genes, whose dysfunction will lead to immunodeficiency and are documented to participate in the immunosuppressive microenvironment[21–25]. As a member of the TLR family, TLR7 recognizes single-stranded RNA oligonucleotides from RNA viruses in the endosomes of plasmacytoid dendritic cells and B cells[26, 27], which is also involved in tumor development[28]. Our results show a negative relationship with the risk score, which suggests that TLR7 would benefit from the immune response in ccRCC TME. NFKB2, the gene encoding NF-κB2, exists in numerous cell types and roles, including inflammation and immune functions, and the dysregulation of NFKB2 is related to immunodeficiency and several TME pathways[29–32]. The IL4, TGFB1 and PDCD1, which encodes the famous immunosuppressive factors IL-4, TGF-β1 and PD-1, respectively, are always suggesting a worse prognosis[33–35]. All of these immunosuppressive genes are positively related to the constructed risk score, which indicates that the High-Score reveals the worse immune escaping TME.
Characterized as the copper ionophores related cell death, cuproptosis is distinct from other known cell death such as apoptosis, ferroptosis, necroptosis and oxidative stress[6], and agents refer to both copper ionophores and copper chelators that have already been promoted to different phases of clinical trials[7, 36–39]. However, independent of traditional cell death progression, such as apoptosis, ferroptosis and necroptosis[6], it piqued our interest in whether the novel cuproptosis would influence the OS and TME of ccRCC patients. By incorporating the cuproptosis and TME-related genes, we constructed the risk score according to the top 10 screened out hub genes. Our findings indicate that the High-Score patients, though exhibiting the worse OS, may respond better than the Low-Score patients, along with the hypoxia environment, which would also lead to the immunosuppressive TME. Besides that, cuproptosis-related genes, especially FDX1 and PDHB, showed a negative relation with not only the risk score but also the immune response. With the functional enrichment, we found that the High-Score group is strongly associated with the mTOR 1 and glycolysis pathway. Characterized by a hypoxia environment, ccRCC is recognized as an mTOR inhibitor-responsive cancer, and the mTOR1 pathway links to the expression of several immunosuppressive cytokines[40–42]. Additionally, Kevin D Courtney et al. demonstrated the highest enrichment in the glycolysis pathway and lowest enrichment in TCA cycle pathway[43], which is consistent with our results and also corroborates the reduced cuproptosis of tumor cells. Our findings also illustrated the feasibility of cuproptosis-related treatment for ccRCC patients.
Interestingly, althoug the ccRCC patients from the High-Score group exhibited the worse prognosis, the functional enrichment, ESTIMATE Score and TMESCORE revealed a better inflammatory response in these patients, respectively. Similarly, the CIBERSORT algorithm also discloses the better immune infiltration of those from the High-Score group because the proportion of the anti-tumor lymphocytes, such as CD8 + T cells, activated NK cells, and activated memory T cells was significantly increased along with the decreased proportion of immunosuppressive lymphocytes M2 such as macrophages. However, our findings also revealed a significant increase in the proportion of T follicular helper cells and regulatory T cells, who are the critical immune regulators leading to T cell exhausting and immunosuppressive TME[44–46]. RCC is widely regarded as a highly immunogenetic cancer, leading to abundant infiltration of immune lymphocytes, including the cytotoxic CD8 + T cells[47–49]. Nevertheless, distinct in many solid tumors, the increase of CD8 + T cells does not appear to have a close association with the prognosis of RCC patients. Even the PD-1 and PD-L1 fail to convincingly predict the clinical response to immune checkpoint inhibitors[49–51], owing to the defective antigen presentation[50, 51]. Our risk score model combined the traditional TME-pathway genes with cuproptosis-related genes and constructed a novel algorithm with hub genes which not only effectively evaluated the prognosis of ccRCC patients, but also connected it to the inflammatory response and the immune cell recruitment talent. According to our risk score model, the High-Score group, though exhibiting a shortened OS, showed better T cells and cDCs chemotactic ability and inflammatory response, which should be considered as a potential ICI treatment indicator.
Our model possesses several advantages. First and foremost, we set up a cuproptosis and TME-related risk score, which effectively predicted the prognosis of ccRCC patients. As a novel cell death pathway, the influence of tumor cell fate and the following impact on TME remains to be expounded. We revealed the positive relationship between cuproptosis-related genes, especially the key regulators FDX1 and PDHB, and the outcome of ccRCC patients. Meanwhile, the abundance of FDX1 and PDHB may also alter the TME to the active immune condition. However, another cuproptosis-related gene, CDKN2A, showed a negative relationship with the outcome of ccRCC patients and may conversely lead to the immunosuppressive TME. Furthermore, although the patients from the High-Score group have a worse OS, they also displayed a better immune response from our analysis, which may be attributed to the T cells priming but exhausted by the immunosuppressive TME.
Nevertheless, the limitations are keenly awaiting further study. As all the results came from the TCGA analysis, our findings revealed the positive relationship between the vital cuproptosis regulators FDX1 and PDHB with the OS of ccRCC patients. However, whether the copper ionophore agents would work with specific up-regulation with FDX1 and PDHB remains to be discovered. Besides that, our model exhibited the better activated TME in the High-Score group with the lower cuproptosis. It awaits to be studied whether the cuproptosis of ccRCC tumor cells would further activate the anti-tumor immune systems via antigen presentation. In addition, it is still necessary to conduct further research on the hub genes with the functional detection in ccRCC tumors and TME in vivo and in vitro.