Skip to main content
NCBI home page
Thoracic Cancer logoLink to Thoracic Cancer
. 2023 Oct 13;14(34):3369–3380. doi: 10.1111/1759-7714.15128

Regulation of ferroptosis‐related genes in CD8+ NKT cells and classical monocytes may affect the immunotherapy response after combined treatment in triple negative breast cancer

Zheming Liu 1, Songjiang He 1, Zhou Huang 2, Jiahui Liu 3, Yiping Gong 4,✉, Yi Yao 1,✉, Xue Zhang 4,✉
PMCID: PMC10693945  PMID: 37830388

Abstract

Background

Drug resistance has led to the failure of immunotherapy in triple negative breast cancer patients. Here we aimed to explore the mechanisms of drug resistance in patients in order to enhance their response to immunotherapy.

Methods

We downloaded publicly available single‐cell RNA‐sequencing data of peripheral blood mononuclear cells from patients after treatment to investigate the possible mechanisms of drug resistance. The publicly available TCGA transcriptomic data and somatic mutation data were used for further validation. In this study, a series of bioinformatics and machine learning methods were employed.

Results

We identified the vital roles of CD8+ NKT cells and classical monocytes in the immunotherapy response of triple‐negative breast cancer patients. The proportion of these cell types was significantly increased in group partial response. We also found that downregulation of ferroptosis‐related genes regulates the immune pathway. The analysis of scRNA data and TCGA transcriptomic data presented that DUSP1 may play a crucial role in immunotherapy resistance.

Conclusion

Overall, the composition of the tumor microenvironment affects the immunotherapy response of patients, and DUSP1 may be a potential target for overcoming drug resistance.

Keywords: drug resistance, ferroptosis, immunotherapy, triple‐negative breast cancer, tumor microenvironment

Combined therapy with atezolizumab and paclitaxel will downregulate the expression of DDIT4 in CD8+ NKT cells and DUSP1 in classical monocytes, which leading to reduced ferroptosis in both cells and consequently inhibited tumor progression.

graphic file with name TCA-14-3369-g004.jpg

INTRODUCTION

Breast cancer has the highest mortality rate among women worldwide, with triple‐negative breast cancer (TNBC) being the most aggressive subtype. 1 Chemotherapy is currently the best treatment option for TNBC, but in recent years, high tumor mutation burden (TMB) and high levels of PD‐L1 expression have been discovered in TNBC patients, making immunotherapy a potential therapeutic approach. 2 , 3 However, treatment failure often occurs due to chemotherapy‐induced cell apoptosis or intrinsic resistance in patients. Therefore, the development of effective treatment strategies has become an urgent need in TNBC treatment.

Ferroptosis is an iron‐dependent cell death pathway induced by lipid peroxidation. Since the term “ferroptosis” was proposed in 2012, the unique mechanism of ferroptosis has gained increasing graces in the field of antitumor therapy. 4 Ferrous ion (Fe2+) in the oxidation–reduction process plays a crucial role in ferroptosis, characterized by inhibiting glutathione peroxidase 4 (GPX4), lipid repair enzymes, and accumulation of lipid hydroperoxides (LPO). Therefore, the accumulation of intracellular LPO leads to cellular structural and integrity damage. 5 Several mechanisms that regulate iron and reactive oxygen species (ROS) metabolism to induce ferroptosis have been reported so far, but the mechanism of ferroptosis in breast cancer, especially in TNBC, has been rarely studied. 6 We already know that ferroptosis is a gradual process regulated by various metabolic pathways; however, a clear understanding of ferroptosis in TNBC is still lacking. 7 In recent years, single‐cell transcriptome sequencing has become a technique for revealing the overall gene expression profile within individual cells and reflecting intercellular heterogeneity. 8 Some studies 9 have used single‐cell RNA sequencing to investigate the gene expression profiles of specific cell types during TNBC occurrence and development, which may contribute to a further understanding of disease mechanisms and provide new insights for the development of targeted therapies.

Here, we reveal the characteristics and aberrant expression patterns of ferroptosis‐related genes (FRGs) in different cell types of TNBC, as well as their impact on treatment response, elucidating the potential key role of ferroptosis in TNBC patients' drug resistance.

METHODS

Data acquisition and preprocessing

We collected peripheral blood mononuclear cell (PBMC) samples from 22 advanced TNBC patients (4 partial response [PR], 7 stable disease [SD], and 11 pretreatment [Pre]) that were publicly available in the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/). These patients received treatment with paclitaxel in combination with anti‐PD‐L1 atezolizumab. Transcriptome data of 99 normal samples and 115 TNBC samples, along with their matched clinical information, were retrieved from the Cancer Genome Atlas program (TCGA, https://portal.gdc.cancer.gov/). The RNA‐Seq data were then normalized to fragments per kilobase of exon model per million mapped fragments (FPKM) values and log2 transformed. Hallmark gene sets of tumor were utilized from the Molecular Signatures Database (MSigDB, https://www.gsea-msigdb.org/gsea/msigdb/). Additionally, somatic mutation data of 115 TNBC samples were obtained from the TCGA database.

Single‐cell analysis

We analyzed single‐cell data using the R package Seurat. We selected 112 456 high‐quality single‐cell samples that met the following criteria: UMI counts more than 300 and less than 100 000, gene feature counts between 200 and 8000, and mitochondrial gene ratio less than 10%. Next, the function LogNormalize was performed to ensure that the total gene expression in each cell was equal, and the scale factor was set to 10 000. The top 2000 variable genes were turned up for downstream analysis using the FindVariableFeatures function. The R package Harmony was used to remove batch effects of the transcriptome data from different samples. 10 Subsequently, we selected the top 50 principal components for uniform manifold approximation and projection (UMAP) dimensionality reduction and visualization. Through unbiased clustering analysis, we obtained 23 cell clusters. The function FindAllMarkers was used to identify differentially expressed genes for each cell cluster. Combining reported and publicly available single‐cell surface markers, we annotated the cells to determine their cell types. Finally, we used the function AddModuleScore to score the ferroptosis‐related gene set. 11

Enrichment analysis

Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases store a wealth of gene‐related functional information based on different classification concepts. After performing differential analysis using the R package DESeq2, 12 we obtained differentially expressed genes between different groups. Subsequently, we used the R package ClusterProfiler for GO and KEGG analyses. 13

Survival analysis

We divided the 115 TCGA TNBC samples into two groups based on the median value of the target gene or other grouping criteria. The R package Survival was used to estimate the survival differences between the two groups. The Kaplan–Meier method was used to plot survival curves, and the log‐rank test was used to assess the statistical differences in prognoses between different groups.

Additional statistical analysis

All statistical analysis in this article were based on R (version 4.2.3, http://www.r-project.org). The protein–protein interaction (PPI) network was constructed using the STRING database. Z‐scores of tumor hallmark gene sets were calculated based on the TCGA transcriptome data (FPKM value) using the Z‐score algorithm provided by the R package GSVA. The R package pheatmap was used for heatmap visualization. Single nucleotide mutation (SNV) and copy number variance (CNV) data were analyzed using the R package maftools. 14 In this study, a p‐value <0.05 was considered statistically significant.

RESULTS

ScRNA‐seq analysis of peripheral blood mononuclear cells in patients with triple negative breast cancer before and after treatment determined different cell types

To explore the relationship between treatment outcomes in patients with TNBC and factors related to ferroptosis regulation, we collected PBMC samples from advanced TNBC patients treated with paclitaxel in combination with anti‐PD‐L1 atezolizumab and the RNA‐seq data of 115 TNBC patients from TCGA (Figure 1a). After rigorous data quality control, we obtained transcriptome data from 112 456 single cells. Utilizing unbiased clustering analysis, we identified 23 cell clusters. By integrating reported marker genes of PBMCs, we identified these cell subtypes as 23 distinct cell types (Figure 1b). These cell types included innate lymphoid cells, B cells, T cells, myeloid cells, and so forth, with T cells being the largest category. We further analyzed the top three specifically expressed marker genes for each cell subtype (Figure 1c). Additionally, we analyzed the proportions of each cell type in each sample group. In the PR group, T cells were the predominant cell type, with the highest variation observed in the proportion of myeloid cells (Figures 1d and S1a). Between PR and Pre groups, the proportions of certain T cell subtypes decreased while others increased. Between the SD and Pre groups, the proportions of most T cell subtypes increased, with only a few subtypes showing a decrease (Figure 1e,f). Similar characteristics were observed in the analysis of the relative proportions of each cell type and their correlation across the three groups (Figure S1a,b). Next, we conducted differential expression analysis among different groups and performed functional analysis of differentially expressed genes within each cell subtype. The results showed that among upregulated genes, there was a significant enrichment of genes involved in immune response pathways. It is noteworthy that, compared to the Pre group each subtype in the PR group was particularly enriched in neutrophil degranulation (Figure S1c). Among downregulated genes, the SD group showed enrichment in pathways related to immune response and transcriptional regulation, while the PR group was predominantly enriched in processes of translation, rRNA processing, cytoplasmic translation, viral transcription, and nuclear transcription mRNA decay (Figure S1d). Finally, we calculated the average FRGs level for each cell and displayed the statistical scores, revealing variations in ferroptosis scores for each cell type among the three sample groups, with B and T cells showing a relative decrease while myeloid cells and innate lymphoid cells showed an increase (Figure 1g). In summary, these results provide a comprehensive overview of the composition and changes in PBMC cellular composition pre‐ and post‐treatment, indicating functional differences within these cell clusters.

FIGURE 1.

FIGURE 1

Open in a new tab

Landscape of immune microenvironment in triple‐negative breast cancer (TNBC) patients after treatment. (a) Schematic representation of the analysis. Single‐cell samples were divided into three groups: partial remission (PR), stable disease (SD), and Pre‐treatment (Pre), followed by single‐cell sequencing and data analysis. (b) Uniform manifold approximation and projection (UMAP) plot exhibited 23 cell clusters. (c) Heatmap displayed the top three marker genes in each cell cluster. (d) Proportional comparison of each cluster in different groups. (e, f) Differential analysis between two groups for each cluster, calculating the fold change in the proportion of each cluster between the two groups, followed by log2 transformation. Positive bars represent clusters with increased relative proportion, while negative bars represent clusters with decreased relative proportion. (g) Scoring of ferroptosis‐related genes (FRGs) in different cell types.

Heterogeneity and regulatory module of cell specific ferroptosis in TNBC patients before and after treatment

Based on the 260 reported FRGs, we performed unsupervised clustering analysis using the R package Seurat. The result showed that the expression mode of FRGs exhibits significant cell type specificity. Furthermore, there were significant compositional changes in the ferroptosis‐expressing clusters before and after treatment (Figures 2a and S2a). By analyzing the proportion features of ferroptosis‐expressing clusters in each functional cell type, we found that the overall changes in F1, F3, and F4 were the most significant, and F2 was the main component in myeloid cells, with the largest increase in the PR group (Figures 2b and S2b). These results suggest that different subtypes of ferroptosis expression pattern may exist within the same cell type, and the relative proportions of these subtypes are highly correlated with the disease treatment status. We also found that different ferroptosis‐expressing clusters exhibited highly specific expressions of FRGs (Figure 2c). These ferroptosis‐expressing clusters showed common and specific features in functional enrichment analysis. For example, F2, F11, and F9 were enriched in autophagy and cellular oxidative stress response, while F3 showed specific enrichment in transcription regulation. Additionally, F1 and F2 were particularly enriched in the apoptosis process. F1 mainly consisted of T cells and innate lymphoid cells, while F2 mainly consisted of myeloid cells (Figures 2d and S2c). Finally, we showed the T cell specifically expressed gene PEBP1 and one of the myeloid cell FTL was among the top three highly expressed marker genes (Figure 2e,f).

FIGURE 2.

FIGURE 2

Open in a new tab

Heterogeneity of ferroptosis‐related gene (FRG) expression patterns among different cell types. (a) Unsupervised clustering based on FRGs, with uniform manifold approximation and projection (UMAP) plot visualization. (b) Composition of ferroptosis‐expression mode clusters in different clusters from different groups. (c) Heatmap displayed the top three marker genes in each ferroptosis‐expressing cluster. (d) Gene ontology (GO) enrichment analysis of the top 50 genes in each ferroptosis‐expressing cluster. (e, f) Expression distribution characteristics of PEBP1 and FTL in the ferroptosis‐expressing cluster.

Functional FRGs in CD8+ NKT cells of TNBC patients before and after treatment are regulated

Next, we extracted T cells and performed UMAP clustering visualization (Figure 3a). Different subsets of T cells exhibited different changes in the proportions of cell clusters among the three groups, with a significant increase in C8: CD8+ NKT cells in the PR group (Figure 3b). PRF1, GNLY, and FCGR3A were the top three highly expressed marker genes for C8 (Figure 3c). In addition to being enriched in immune response pathways, the top 50 highly expressed marker genes of C8 were particularly enriched in cell defense response and neutrophil degranulation (Figure S3a,b). Therefore, we further analyzed the different FRGs and regulatory T cell genes in the C8 (Figure 3d,e) and found that the downregulated gene DDIT4 specifically regulated immune response, while the upregulated genes regulated processes such as translation, metabolism, and transcription (Figure 3f). Furthermore, we observed a significant decrease in DDIT4 expression and a significant increase in FTH1 expression in the C8 (Figure 3g,h).

FIGURE 3.

FIGURE 3

Open in a new tab

CD8+ NKT cells plays an important role in immunotherapy response. (a) Uniform manifold approximation and projection (UMAP) plots showed the distribution of T cell subtypes in different groups. (b) Proportional of T cell subtypes among three groups. (c) Heatmap displayed the top three marker genes in T cell subtypes. (d) Heatmap shows differential ferroptosis‐related genes (FRGs) in C8: D8 + NKT cells. (e) Network plot displayed the coexpression of differential genes and FRGs in C8: CD8+ NKT cells. (f) Gene ontology (GO) functional pathways of differential FRGs in C8: CD8+ NKT cells. (g, h) Expression distribution of FRGs (DDIT4 and FTH1) in T cell populations.

Functional FRGs in classical monocytes of TNBC patients before and after treatment are largely regulated

According to the results shown in Figure 2, we determined that there were significant differences in the expression patterns of FRGs in different subpopulations of myeloid cells. Therefore, we performed UMAP visualization of myeloid cells respectively (Figure 4a). We observed that the proportions of different myeloid cell subgroups changed in the three groups, with a significant increase in the C2: classical monocytes (Figure 4b). NEAT1, VCAN, and CSF3R were the top three highly expressed marker genes in the C2 (Figure 4c). The top 50 highly expressed marker genes in the C2 were mainly enriched in processes such as neutrophil degranulation, immune response, and inflammation response (Figure S4a). Subsequently, we analyzed the differentially expressed FRGs in the C2 among the three groups and performed coexpression and functional enrichment analysis of the up‐ and downregulated FRGs (Figure 4d–f). The results showed that these differentially expressed FRGs mainly regulated biological processes such as immune response, apoptosis, and inflammation response in the C2, indicating their crucial role in this subgroup (Figures 4g,h and S4b). We specifically focused on two genes, DUSP1, which showed a significant decrease in the C2, and CD44 which showed a significant increase (Figure 4i,j), and both genes exhibited a significant correlation with neutrophil degranulation and immune response pathways (Figure 4h).

FIGURE 4.

FIGURE 4

Open in a new tab

Classical monocytes take part in the response to immunotherapy through regulating ferroptosis‐related genes (FRGs). (a) Uniform manifold approximation and projection (UMAP) plots showed the distribution of myeloid subtypes in different groups. (b) Proportional of myeloid subtypes among three groups. (c) Heatmap displayed the expression of the top three marker genes in myeloid subtypes. (d) Heatmap showed differential FRGs in C2: classical monocytes. (e, f) Venn diagrams showed the overlap genes of upregulated and downregulated FRGs in C2: classical monocytes among the three groups. (g) Network plot displayed the coexpression of differential genes and FRGs in C2: classical monocytes. (h) Gene ontology (GO) pathways of differential FRGs in C2: classical monocytes. (i, j) Expression of DUSP1 and CD44 in myeloid cell populations.

Prognostic verification of TNBC patients in TCGA

There are 70 differentially expressed FRGs in TNBC when compared with normal samples (Figure 5a). Principal component analysis (PCA) revealed significant differences in gene expression between cancer tissues and adjacent normal tissues, showing distinct separation (Figure 5b). FRGs displayed differential expression patterns between tumor and normal samples (Figure 5c). Functional pathway enrichment analysis of up‐ and downregulated genes in cancer tissues and adjacent normal tissues revealed that upregulated genes were mainly enriched in immune response, cell division, and cell cycle pathways, while downregulated genes were mainly enriched in signal transduction and cell adhesion pathways (Figure S5a,b). Differentially expressed FRGs were primarily enriched in oxidative stress and redox pathways (Figure S5c). Cox regression analysis demonstrated that four differential FRGs (SLC2A3, JDP2, DUSP1, DDIT4) of C8: NKT cells and C2: classical monocytes significantly contributed to survival (Figure 5d). DUSP1 and DDIT4 were the genes overlapping with the significantly differentially expressed results (Figure 5e). We compared the expression of DUSP1 and DDIT4 in different stages of TNBC patients and found that DUSP1 was highly expressed in samples from stage II while DDIT4 showed significant upregulation in stage IV (Figure 5f). Furthermore, we investigated the impact of DUSP1 and DDIT4 on survival and discovered that TNBC patients with high expression of DUSP1 had significantly shorter survival time, whereas patients with high expression of DDIT4 exhibited a noticeable survival difference only after 3 years (Figure 5g,h). During the development and progression of tumors, tumor cells acquire a series of malignant hallmarks. 15 As a regulator of ferroptosis and autophagy, the downregulation of DUSP1 in tumor tissues allows tumor cells to have a longer lifespan. 16 However, we found that at higher levels of DUSP1 expression, cancer malignancy features and relevant pathways in TNBC patients were significantly activated, which directly led to poorer prognosis (Figures 5i and S5d).

FIGURE 5.

FIGURE 5

Open in a new tab

DUSP1 can be a potential marker to reflect prognosis of triple‐negative breast cancer (TNBC) patients. (a) Bar plot showed the number of differential genes and differential ferroptosis‐related genes (FRGs) in TNBC. (b) Principal component analysis (PCA) analysis exhibited the heterogeneity of cancer and adjacent tissues. (c) Heatmap displayed the expression patterns and characteristics of differential FRGs. (d) Cox regression analysis showed the impact of differential FRGs in CD8+ NKT cells and classical monocytes on the survival of TNBC patients. (e) DDIT4 and DUSP1 are genes with significant significance in both differential analysis and Cox regression analysis. (f) Box plot showed the expression profiles of DDIT4 and DUSP1 at different stages. (g, h) Survival curves plotted the prognosis of DDIT4 and DUSP1. (i) Z‐score of tumor malignant hallmarks in different DUSP1 expression groups.

Alteration of DUSP1 lead to worse prognosis of TNBC patients

We observed the mutation status of differential expressed FRGs in C8: NKT cells and C2: classical monocytes among TNBC patients (Figure 6a). We found a significant comutation of NCOA4 and TXNIP (p < 0.05) among these genes (Figure 6b). Next, we demonstrated the CNV of these differential genes in the chromosome, which indicated a significant increase in copy numbers of DUSP1 and DDIT4 (Figure 6c). The lollipop plot displayed the results of SNV of these two genes, showing a missense mutation in DUSP1 and a nonsense mutation in DDIT4 (Figure 6d,e). Survival analysis revealed a poorer prognosis in the DUSP1 altered group (Figure 6f,g).

FIGURE 6.

FIGURE 6

Open in a new tab

Alteration of ferroptosis‐related genes (FRGs) attribute to progression of triple‐negative breast cancer (TNBC). (a) Oncoplot exhibited the mutations of differential FRGs in CD8+ NKT cells and classical monocytes. (b) Comutation analysis between differential FRGs in CD8+ NKT cells and classical monocytes. (c) Copy number variation between differential FRGs in CD8+ NKT cells and classical monocytes. (d, e) Lollipop plots showed the single nucleotide mutations of DUSP1 and DDIT4. (f, g) Survival analysis demonstrated the differences of survival between altered and unaltered groups of DUSP1 and DDIT4.

DISCUSSION

The composition of the immune microenvironment is crucial for the response to immunotherapy. 17 In this study, we focused on examining the changes in the immune landscape in PBMCs from patients in different treatment groups (Pre, PR, SD) and the impact of these changes on treatment outcomes. We observed a significant increase in the proportion of myeloid cells in the PR and SD groups, along with distinct alterations in the composition of certain T cell subsets. Recent studies 18 have indicated a correlation between the abundance of TH1 cells, cytotoxic T cells, and extended survival in cancer patients receiving immunotherapy. Additionally, researchers have been gradually exploring the mechanisms through which myeloid cells develop resistance to immunotherapy. 19 , 20 , 21 , 22 , 23 Subsequently, we conducted functional enrichment analysis on the upregulated genes in the PR and SD groups, revealing significant upregulation of immune response, neutrophil degranulation, and other immune‐related pathways in these two groups. Recent studies 24 , 25 , 26 have elucidated the close relationship between ferroptosis expression pattern in CD8+ T cells and their activity, directly impacting the antitumor process. Therefore, further investigation found that the differential cellular composition in the tumor microenvironment (TME) might be attributed to distinct FRGs expression patterns exhibited by these cells. By reclustering the cells based on FRGs we found that populations enriched with T cells and myeloid cells exhibited specific upregulation of the autophagy pathway. Recent studies 27 , 28 have demonstrated the critical role of autophagy in metabolic regulation during various stages of T cell development and function, particularly in the maturation of precursor T cells. Subsequently, during thymic development, autophagy regulates the presentation of peptides by stromal cells and professional antigen‐presenting cells, mediating the selection of thymocytes. Furthermore, as mature T cells enter the periphery and become activated, their metabolic changes also depend on autophagy. In summary, autophagy can prevent premature aging and ensure the maintenance of memory T cells. Therefore, we speculate that after TNBC treatment, T cells, and myeloid cells are influenced by regulating ferroptosis, and inhibiting ferroptosis may enhance the regulatory T cell activity and strengthen the antitumor response.

Increasing the number of antitumor cells is an attractive therapy as it is likely to bring significant benefits to patients receiving immunotherapy. Therefore, we further subdivided the T cells and myeloid cells in the dataset. We observed a similar increasing trend in C8: CD8+ NKT cells and C2: classical monocytes in the immunotherapy response group, while dendritic cells and nonclassical monocytes derived from classical monocytes 29 decreased, this suggests that the differentiation barrier of classical monocytes may contribute to the antitumor response. CD8+ NKT cells, as well‐known as a main antitumor component, 30 , 31 have always been of interest, and an increase in their number undoubtedly improves the treatment effectiveness for patients. In recent years, the role of myeloid cells in tumors has gradually attracted the interest of numerous scientists. 20 , 32 , 33 , 34 Some new studies 35 , 36 have shown that tumor‐associated macrophages derived from monocytes are associated with immune suppression and antigen presentation, indicating their significant role in tumor progression. In our study, we explored the significant differences in FRGs expression between CD8+ NKT cells and classical monocytes from different treatment groups and conducted coexpression analysis and enrichment analysis of differential genes. The results showed that downregulated ferroptosis‐related gene (DDIT4) in CD8+ NKT cells was significantly associated with the immune pathway, suggesting that DDIT4 may inhibit the activity or proliferation of CD8+ NKT cells. Similarly, in classical monocytes, we found that FTL, BID, CD44, SLC2A3, DUSP1, DDIT4, and JUN were significantly associated with the immune response.

RNA‐seq data have been used to explore genes of greater significance and validate their roles in prognosis. We employed Cox regression analysis and differential analysis and found that DDIT4 and DUSP1 were associated with patient survival and exhibited significant expression differences in tumor tissues. The expression level of DDIT4 showed no significant difference in patient prognosis, but patients with higher expression of DDIT4 experienced significantly worsened prognoses after the mid‐term (3 years). DUSP1 exhibited notable differences in early prognosis, and patients owned poorer prognoses when showing higher expression of DUSP1. In a recent study, Jiang et al. 37 demonstrated that lncRNA DDIT4‐AS1 recruits RNA‐binding protein AUF1 to facilitate the interaction between AUF1 and DDIT4 mRNA, thereby stabilizing DDIT4 mRNA and inhibiting the mTORC1 signaling pathway. It induces the regulation of autophagy in TNBC cells and promotes the proliferation, migration, and invasion of TNBC cells. Additionally, in vitro and in vivo experiments demonstrated that downregulation of DDIT4‐AS1 enhanced the sensitivity of TNBC cells to chemotherapy drugs such as paclitaxel. Recent reports by Azam et al. 38 also suggested that combined therapy targeting the signaling proteins c‐Fos and DUSP1 could potentially cure various kinase‐driven refractory leukemias and solid tumors, including acute myeloid leukemia (AML), lung cancer, breast cancer, and chronic myeloid leukemia (CML). These findings are consistent with our discoveries. However, there is currently limited literature available on the relationship between DUSP1 and drug resistance in triple‐negative breast cancer, which will be further elucidated in our subsequent research.

Gene mutations often drive tumor progression and drug resistance. 39 We investigated the mutation profiles of differentially expressed FRGs. The results showed that the average copy numbers of the DUSP1 and DDIT4 genes were increased in patients with TNBC. However, the DUSP1 gene had missense mutations, while the DDIT4 gene had nonsense mutations. These differences resulted in significantly worse prognoses for TNBC patients with DUSP1 gene mutations, while the prognoses of patients with DDIT4 gene variations did not show any significant difference compared to those without mutations. These findings suggest that DUSP1 gene mutations may be an important factor contributing to the development of tumor development and drug resistance.

Taken together, our study demonstrates that differential expression patterns of FRGs contribute to varieties of the TME of TNBC patients, leading to their resistance to immunotherapy. DDIT4 and DUSP1 play crucial regulatory roles in this process, and DUSP1 gene mutations may serve as potential therapeutic targets for immunotherapy resistance in TNBC patients.

AUTHOR CONTRIBUTIONS

Zheming Liu, Songjiang He, and Zhou Huang are responsible for data analyses and composing the article, Jiahui Liu in charge of the data collection, Yiping Gong, Yi Yao, and Xue Zhang revised the current study.

CONFLICT OF INTEREST STATEMENT

The authors confirm there are no conflicts of interest.

Supporting information

Figure S1. Different landscapes of functional pathways among three treatment groups. (a) Comparison of the relative proportions of each cluster in different groups. (b) Scatter plot showed the changes in the relative proportions of cell populations in two groups. The x‐ and y‐axes represent log2 (cell proportion). (c) Heatmap exhibited the gene ontology (GO) pathways of upregulated genes in each cell population in the partial response (PR) versus stable disease (SD). (d) Heatmap showed the Kyoto encyclopedia of genes and genomes (KEGG) pathways of upregulated genes in each cell population in the PR versus SD.

Click here for additional data file. (420.3KB, pdf)

Figure S2. Differences of biological function in varies ferroptosis‐expressing clusters. (a) Clustering results of 12 cell populations obtained by unsupervised clustering based on ferroptosis‐related genes (FRGs), visualized by UMAP plot. (b) Proportion of ferroptosis‐expressing clusters in each cell type. (c) KEGG enrichment analysis of the top 50 marker FRGs in each ferroptosis‐expressing clusters.

Click here for additional data file. (516.4KB, pdf)

Figure S3. Functional pathway of up‐ and down‐regulated ferroptosis‐related genes (FRGs) in CD8+ NKT cells. (a) gene ontology (GO) enrichment analysis of the top 50 marker FRGs in T cell populations. (B) Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis of differentially genes co‐expressed with FRGs in C8: CD8+ NKT cells.

Click here for additional data file. (397.1KB, pdf)

Figure S4. Functional pathway of differential ferroptosis‐related genes (FRGs) in classical monocytes. (a) Gene ontology (GO) enrichment analysis of the top 50 marker FRGs in myeloid cell populations. (b) Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis of differential genes co‐expressed with FRGs in C2: classical monocytes.

Click here for additional data file. (407.6KB, pdf)

Figure S5. Validation of ferroptosis‐related genes (FRGs) in TCGA set. (a, b) Upregulated and downregulated gene ontology (GO) pathways in cancer versus adjacent tissues. (c) Kyoto encyclopedia of genes and genomes (KEGG) enrichment of differential FRGs. (d) Correlation analysis between expression level of DUSP1 and Z‐score of malignant hallmarks of tumor.

Click here for additional data file. (535.9KB, pdf)

Liu Z, He S, Huang Z, Liu J, Gong Y, Yao Y, et al. Regulation of ferroptosis‐related genes in CD8+ NKT cells and classical monocytes may affect the immunotherapy response after combined treatment in triple negative breast cancer. Thorac Cancer. 2023;14(34):3369–3380. 10.1111/1759-7714.15128

Zheming Liu, Songjiang He, Zhou Huang contributed equally to this study.

Contributor Information

Yiping Gong, Email: xzhangrm@whu.edu.cn, Email: gongyp@whu.edu.cn.

Yi Yao, Email: yaoyi2018@whu.edu.cn.

Xue Zhang, Email: xzhangrm@whu.edu.cn.

REFERENCES

  • 1. Allemani C, Matsuda T, Di Carlo V, Harewood R, Matz M, Niksic M, et al. Global surveillance of trends in cancer survival 2000‐14 (CONCORD‐3): analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population‐based registries in 71 countries. Lancet. 2018;391(10125):1023–1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Keenan TE, Tolaney SM. Role of immunotherapy in triple‐negative breast cancer. J Natl Compr Canc Netw. 2020;18(4):479–489. [DOI] [PubMed] [Google Scholar]
  • 3. Cortes J, Cescon DW, Rugo HS, Nowecki Z, Im SA, Yusof MM, et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for previously untreated locally recurrent inoperable or metastatic triple‐negative breast cancer (KEYNOTE‐355): a randomised, placebo‐controlled, double‐blind, phase 3 clinical trial. Lancet. 2020;396(10265):1817–1828. [DOI] [PubMed] [Google Scholar]
  • 4. Lei G, Zhuang L, Gan B. Targeting ferroptosis as a vulnerability in cancer. Nat Rev Cancer. 2022;22(7):381–396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Wu SZ, Roden DL, Wang C, Holliday H, Harvey K, Cazet AS, et al. Stromal cell diversity associated with immune evasion in human triple‐negative breast cancer. EMBO J. 2020;39(19):e104063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Yao X, Xie R, Cao Y, Tang J, Men Y, Peng H, et al. Simvastatin induced ferroptosis for triple‐negative breast cancer therapy. J Nanobiotechnology. 2021;19(1):311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Yang F, Xiao Y, Ding JH, Jin X, Ma D, Li DQ, et al. Ferroptosis heterogeneity in triple‐negative breast cancer reveals an innovative immunotherapy combination strategy. Cell Metab. 2023;35(1):84–100 e8. [DOI] [PubMed] [Google Scholar]
  • 8. Shao X, Lu X, Liao J, Chen H, Fan X. New avenues for systematically inferring cell‐cell communication: through single‐cell transcriptomics data. Protein Cell. 2020;11(12):866–880. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Zhang Y, Chen H, Mo H, Hu X, Gao R, Zhao Y, et al. Single‐cell analyses reveal key immune cell subsets associated with response to PD‐L1 blockade in triple‐negative breast cancer. Cancer Cell. 2021;39(12):1578. [DOI] [PubMed] [Google Scholar]
  • 10. Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, et al. Fast, sensitive and accurate integration of single‐cell data with harmony. Nat Methods. 2019;16(12):1289–1296. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM 3rd, et al. Comprehensive integration of single‐cell data. Cell. 2019;177(7):1888. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA‐seq data with DESeq2. Genome Biol. 2014;15(12):550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. Omics. 2012;16(5):284–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Mayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. 2018;28(11):1747–1756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Hanahan D. Hallmarks of cancer: new dimensions. Cancer Discov. 2022;12(1):31–46. [DOI] [PubMed] [Google Scholar]
  • 16. Chen X, Yu C, Kang R, Kroemer G, Tang D. Cellular degradation systems in ferroptosis. Cell Death Differ. 2021;28(4):1135–1148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Pitt JM, Marabelle A, Eggermont A, Soria JC, Kroemer G, Zitvogel L. Targeting the tumor microenvironment: removing obstruction to anticancer immune responses and immunotherapy. Ann Oncol. 2016;27(8):1482–1492. [DOI] [PubMed] [Google Scholar]
  • 18. Galon J, Bruni D. Approaches to treat immune hot, altered and cold tumours with combination immunotherapies. Nat Rev Drug Discov. 2019;18(3):197–218. [DOI] [PubMed] [Google Scholar]
  • 19. DeNardo DG, Ruffell B. Macrophages as regulators of tumour immunity and immunotherapy. Nat Rev Immunol. 2019;19(6):369–382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Engblom C, Pfirschke C, Pittet MJ. The role of myeloid cells in cancer therapies. Nat Rev Cancer. 2016;16(7):447–462. [DOI] [PubMed] [Google Scholar]
  • 21. DeNardo DG, Brennan DJ, Rexhepaj E, Ruffell B, Shiao SL, Madden SF, et al. Leukocyte complexity predicts breast cancer survival and functionally regulates response to chemotherapy. Cancer Discov. 2011;1(1):54–67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Wang T, Xiao M, Ge Y, Krepler C, Belser E, Lopez‐Coral A, et al. BRAF inhibition stimulates melanoma‐associated macrophages to drive tumor growth. Clin Cancer Res. 2015;21(7):1652–1664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Kumagai S, Koyama S, Nishikawa H. Antitumour immunity regulated by aberrant ERBB family signalling. Nat Rev Cancer. 2021;21(3):181–197. [DOI] [PubMed] [Google Scholar]
  • 24. Xu H, Ye D, Ren M, Zhang H, Bi F. Ferroptosis in the tumor microenvironment: perspectives for immunotherapy. Trends Mol Med. 2021;27(9):856–867. [DOI] [PubMed] [Google Scholar]
  • 25. Wang W, Green M, Choi JE, Gijon M, Kennedy PD, Johnson JK, et al. CD8(+) T cells regulate tumour ferroptosis during cancer immunotherapy. Nature. 2019;569(7755):270–274. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Liu T, Zhu C, Chen X, Guan G, Zou C, Shen S, et al. Ferroptosis, as the most enriched programmed cell death process in glioma, induces immunosuppression and immunotherapy resistance. Neuro Oncol. 2022;24(7):1113–1125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Bronietzki AW, Schuster M, Schmitz I. Autophagy in T‐cell development, activation and differentiation. Immunol Cell Biol. 2015;93(1):25–34. [DOI] [PubMed] [Google Scholar]
  • 28. Xia H, Green DR, Zou W. Autophagy in tumour immunity and therapy. Nat Rev Cancer. 2021;21(5):281–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Guilliams M, Mildner A, Yona S. Developmental and functional heterogeneity of monocytes. Immunity. 2018;49(4):595–613. [DOI] [PubMed] [Google Scholar]
  • 30. Pellicci DG, Koay HF, Berzins SP. Thymic development of unconventional T cells: how NKT cells, MAIT cells and gammadelta T cells emerge. Nat Rev Immunol. 2020;20(12):756–770. [DOI] [PubMed] [Google Scholar]
  • 31. Huntington ND, Cursons J, Rautela J. The cancer‐natural killer cell immunity cycle. Nat Rev Cancer. 2020;20(8):437–454. [DOI] [PubMed] [Google Scholar]
  • 32. Mantovani A, Allavena P, Marchesi F, Garlanda C. Macrophages as tools and targets in cancer therapy. Nat Rev Drug Discov. 2022;21(11):799–820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Li K, Shi H, Zhang B, Ou X, Ma Q, Chen Y, et al. Myeloid‐derived suppressor cells as immunosuppressive regulators and therapeutic targets in cancer. Signal Transduct Target Ther. 2021;6(1):362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Wang Y, Johnson KCC, Gatti‐Mays ME, Li Z. Emerging strategies in targeting tumor‐resident myeloid cells for cancer immunotherapy. J Hematol Oncol. 2022;15(1):118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Zhu Y, Herndon JM, Sojka DK, Kim KW, Knolhoff BL, Zuo C, et al. Tissue‐resident macrophages in pancreatic ductal adenocarcinoma originate from embryonic hematopoiesis and promote tumor progression. Immunity. 2017;47(2):323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Xiang X, Wang J, Lu D, Xu X. Targeting tumor‐associated macrophages to synergize tumor immunotherapy. Signal Transduct Target Ther. 2021;6(1):75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Jiang T, Zhu J, Jiang S, Chen Z, Xu P, Gong R, et al. Targeting lncRNA DDIT4‐AS1 sensitizes triple negative breast cancer to chemotherapy via suppressing of autophagy. Adv Sci (Weinh). 2023;10(17):e2207257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Kesarwani M, Kincaid Z, Gomaa A, Huber E, Rohrabaugh S, Siddiqui Z, et al. Targeting c‐FOS and DUSP1 abrogates intrinsic resistance to tyrosine‐kinase inhibitor therapy in BCR‐ABL‐induced leukemia. Nat Med. 2017;23(4):472–482. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Jiang Z, Lu Y, Liu Z, Wu W, Xu X, Dinnyes A, et al. Drug resistance prediction and resistance genes identification in mycobacterium tuberculosis based on a hierarchical attentive neural network utilizing genome‐wide variants. Brief Bioinform. 2022;23(3):bbac041. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1. Different landscapes of functional pathways among three treatment groups. (a) Comparison of the relative proportions of each cluster in different groups. (b) Scatter plot showed the changes in the relative proportions of cell populations in two groups. The x‐ and y‐axes represent log2 (cell proportion). (c) Heatmap exhibited the gene ontology (GO) pathways of upregulated genes in each cell population in the partial response (PR) versus stable disease (SD). (d) Heatmap showed the Kyoto encyclopedia of genes and genomes (KEGG) pathways of upregulated genes in each cell population in the PR versus SD.

Click here for additional data file. (420.3KB, pdf)

Figure S2. Differences of biological function in varies ferroptosis‐expressing clusters. (a) Clustering results of 12 cell populations obtained by unsupervised clustering based on ferroptosis‐related genes (FRGs), visualized by UMAP plot. (b) Proportion of ferroptosis‐expressing clusters in each cell type. (c) KEGG enrichment analysis of the top 50 marker FRGs in each ferroptosis‐expressing clusters.

Click here for additional data file. (516.4KB, pdf)

Figure S3. Functional pathway of up‐ and down‐regulated ferroptosis‐related genes (FRGs) in CD8+ NKT cells. (a) gene ontology (GO) enrichment analysis of the top 50 marker FRGs in T cell populations. (B) Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis of differentially genes co‐expressed with FRGs in C8: CD8+ NKT cells.

Click here for additional data file. (397.1KB, pdf)

Figure S4. Functional pathway of differential ferroptosis‐related genes (FRGs) in classical monocytes. (a) Gene ontology (GO) enrichment analysis of the top 50 marker FRGs in myeloid cell populations. (b) Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis of differential genes co‐expressed with FRGs in C2: classical monocytes.

Click here for additional data file. (407.6KB, pdf)

Figure S5. Validation of ferroptosis‐related genes (FRGs) in TCGA set. (a, b) Upregulated and downregulated gene ontology (GO) pathways in cancer versus adjacent tissues. (c) Kyoto encyclopedia of genes and genomes (KEGG) enrichment of differential FRGs. (d) Correlation analysis between expression level of DUSP1 and Z‐score of malignant hallmarks of tumor.

Click here for additional data file. (535.9KB, pdf)

Articles from Thoracic Cancer are provided here courtesy of Wiley

RESOURCES