IL-17A Increases Doxorubicin Efficacy in Triple Negative Breast Cancer

Abstract

Due to lack of targetable receptors and intertumoral heterogeneity, triple negative breast cancer (TNBC) remains particularly difficult to treat. Doxorubicin (DOX) is typically used as nonselective neoadjuvant chemotherapy, but the diversity of treatment efficacy remains unclear. Comparable to variability in clinical response, an experimental model of TNBC using a 4T1 syngeneic mouse model was found to elicit a differential response to a seven-day treatment regimen of DOX. Single-cell RNA sequencing identified an increase in T cells in tumors that responded to DOX treatment compared to tumors that continued to grow uninhibited. Additionally, compared to resistant tumors, DOX sensitive tumors contained significantly more CD4 T helper cells (339%), γδ T cells (727%), Naïve T cells (278%), and activated CD8 T cells (130%). Furthermore, transcriptional profiles of tumor infiltrated T cells in DOX responsive tumors revealed decreased exhaustion, increased chemokine/cytokine expression, and increased activation and cytotoxic activity. γδ T cell derived IL-17A was identified to be highly abundant in the sensitive tumor microenvironment. IL-17A was also found to directly increase sensitivity of TNBC cells in combination with DOX treatment. In TNBC tumors sensitive to DOX, increased IL-17A levels lead to a direct effect on cancer cell responsiveness and chronic stimulation of tumor infiltrated T cells leading to improved chemotherapeutic efficacy. IL-17A's role as a chemosensitive cytokine in TNBC may offer new opportunities for treating chemoresistant breast tumors and other cancer types.

Keywords: 4T1; IL-17A; chemoresistance; doxorubicin; single cell RNA seq; triple negative breast cancer; γδ T cells.

Figure 1

Syngeneic 4T1 tumors differentially respond to doxorubicin (DOX) treatment. (A) DOX treated BALB/c 4T1 mammary fat pad tumor experimental design. (B) Syngeneic tumor growth rates of tumors in response to DOX. Relative tumor volume was normalized to initial volume and calculated from caliper measurements throughout the experiment. (C) Relative terminal tumor volume at Day 8 post DOX initiation from saline-treated controls or DOX treated tumors in immunocompetent BALB/c mice binned into resistant or sensitive populations (n=10-16). (D) Relative terminal tumor volume at Day 8 post DOX initiation from saline-treated controls or DOX treated tumors in immunodeficient NSG mice (n=3-9). ns: non-significant p > 0.05, ****p < 0.0001.

Figure 2

Single cell RNA sequencing (scRNA-Seq) of tumors following DOX administration. (A) UMAP plot representing all tumor cells following DOX treatment. (B) Colors depict cells derived from tumors of different treatment/response. (C) T cell [Ptprc (CD45)+ CD3d+] relative abundance derived from single cell RNAseq data. (D–G) Feature plots of gene markers used in identifying cell clusters. (H) Flow cytometric quantification of T cell (CD3ϵ+) abundance in syngeneic 4T1 tumors. (I) Flow cytometric quantification of cytotoxic T cell (CD8+CD3ϵ+) abundance in syngeneic 4T1 tumors. (J) Flow cytometric quantification of immune [Ptprc (CD45)+] abundance in syngeneic 4T1 tumors. (K) Flow cytometric quantification of immune [Ptprc (CD45)+] abundance in bone marrow of tumor bearing mice. (L) Flow cytometric quantification of immune [Ptprc (CD45)+] abundance in spleens of tumor bearing mice. (M) Flow cytometric quantification of immune [Ptprc (CD45)+] abundance in spleens of tumor bearing mice. ns: non-significant p > 0.05; *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001.

Figure 3

ScRNA-Seq of tumor infiltrating T cells. (A) UMAP plot of tumor infiltrated T cells. 12 populations identified via transcriptional profiles are denoted by color. (B) UMAP plot with colors denoting tumor response to DOX. (C) Heat map showing relative gene expression for genes of interest per cell in each cluster. (D) Distribution of T cell subtypes in sensitive and resistant tumors. (E) Normalized T cell subtype abundance in sensitive or resistant tumors relative to all tumor cells. (F) Shifts in abundance of tumor infiltrating T cells in sensitive and resistant tumors. Blue bars denote populations with increased abundance in sensitive tumors, red bars denote populations increased in resistant tumors. Bars outside the grey shaded region represent significant (>2-fold) shifts in cell type abundance. Exh/Eff CD8, exhausted effector CD8 T cells; Eff CD8, effector CD8 T cells; Act CD8, activated CD8 T cells; CD4, CD4 T cells; Prolif-1, proliferating T cells-1; Naïve, naïve T cells; IFN CD8, interferon stimulated CD8 T cells; Prolif-2, proliferating T cells-2; NK Cell, natural killer cells; Treg, T regulatory cells; γδ, γδ T cells; AP, antigen presenting T cells.

Figure 4

T cell behavior in response to DOX sensitivity. (A) Violin plots denoting expression of Pdcd1 per T cell in sensitive or resistant tumors. (B) Percentage of tumor residing T cells expressing Pdcd1 as quantified in scRNA-Seq data. (C) PD-1+ T cells quantified from tumor infiltrated T cells following DOX treatment using flow cytometric analysis. (D) ScRNA-Seq expression levels of Pdcd1 by cluster ID depicted as violin plots. (E) Fold upregulation of chemokine and (F) cytokine genes in sensitive vs resistant T cell populations inferred from scRNA-Seq data. (G–I) Activation gene markers (CD69) and cytotoxic genes (Prf1, GzmB) associated with T cell activity expression in sensitive or resistant tumors depicted as violin plots. (J, K) Dimeric transcription factors comprising the AP-1 transcription factor indicative of T cell activation expression segregated by cell types identified in single cell transcriptomic data. *p ≤ 0.05.

Figure 5

Increased γδ IL-17+ T cells in the DOX sensitive tumor microenvironment. (A) UMAP projection of 4T1 syngeneic tumor cells identifying that IL17a expression is restricted to γδ T cells. (B) Ratio of IL17a+ cells in sensitive and resistant tumors extrapolated from scRNA-Seq data. (C, D) Representative immunohistochemistry images of IL17A expression in 4T1 tumor sections from sensitive and resistant tumors following DOX therapy. (E) Protein abundance quantification in tumors normalized to GAPDH expression (n=3). (F, G) Representative flow cytometry plots identifying γδ IL-17+ T cells in T cell populations. (H) Quantitation of IL17A+ T cells from DOX sensitive and resistant tumors. (I) Quantitation of γδ IL17A+ T cells from DOX sensitive and resistant tumors (n=4). (J) Distribution of γδ IL17A+ T cells identifying expanded CD8 and double positive (CD4+CD8+) cells in sensitive tumors (n=4). *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 6

IL-17A co-administration with DOX directly affects chemotherapeutic efficacy in cancer cells. (A) Ex vivo DOX sensitivity from 4T1-Thy1.1 cells isolated from primary tumors or from in vitro culture (n=12 from 3 independent tumors). (B) IC50 values extrapolated from dose response curves with error bars represent 95% CI. (C) Relative viability of 4T1-Thy1.1 cells cultured in the presence of DOX and/or IL-17A for 48 hours (n=9-11). ns: non-significant p > 0.05; *p ≤ 0.05; ****p ≤ 0.0001.

Figure 7

Transcriptomic analysis of 4T1 cells in response to DOX and IL-17A. (A) Venn diagrams depicting overlapping differentially expressed genes from 4T1 cells exposed to DOX +/- IL-17A for 48 hours. (B) Normalized gene expression of CD274 (PD-L1) across in vitro conditions. (C) GSEA analysis of enriched Reactome pathways in common to DOX treated cells regardless of IL-17A. (D) Upregulated pathways with differential enrichment. (E) Downregulated pathways with differential enrichment (n=3-4). ns non-significant p > 0.05; ***p ≤ 0.001; ****p ≤ 0.0001.

Figure 8

Summary schematic.