Protein Tyrosine Phosphatase SHP2 Controls Interleukin-8 Expression in Breast Cancer Cells

Abstract

Treatment of metastasis remains a clinical challenge and the majority of breast cancer-related deaths are the result of drug-resistant metastases. The protein tyrosine phosphatase SHP2 encoded by the proto-oncogene PTPN11 promotes breast cancer progression. Inhibition of SHP2 has been shown to decrease metastases formation in various breast cancer models, but specific downstream effectors of SHP2 remain poorly characterized. Certain cytokines in the metastatic cascade facilitate local invasion and promote metastatic colonization. In this study, we investigated cytokines affected by SHP2 that could be relevant for its pro-tumorigenic properties. We used a cytokine array to investigate differentially released cytokines in the supernatant of SHP2 inhibitor-treated breast cancer cells. Expression of CXCL8 transcripts and protein abundance were assessed in human breast cancer cell lines in which we blocked SHP2 using shRNA constructs or an allosteric inhibitor. The impact of SHP2 inhibition on the phospho-tyrosine-proteome and signaling was determined using mass spectrometry. From previously published RNAseq data (Aceto et al. in Nat. Med. 18:529-37, 2012), we computed transcription factor activities using an integrated system for motif activity response analysis (ISMARA) (Balwierz et al. in Genome Res. 24:869-84, 2014). Finally, using siRNA against ETS1, we investigated whether ETS1 directly influences CXCL8 expression levels. We found that IL-8 is one of the most downregulated cytokines in cell supernatants upon SHP2 blockade, with a twofold decrease in CXCL8 transcripts and a fourfold decrease in IL-8 protein. These effects were also observed in preclinical tumor models. Analysis of the phospho-tyrosine-proteome revealed that several effectors of the mitogen-activated protein kinase (MAPK) pathway are downregulated upon SHP2 inhibition in vitro. MEK1/2 inhibition consistently reduced IL-8 levels in breast cancer cell supernatants. Computational analysis of RNAseq data from SHP2-depleted tumors revealed reduced activity of the transcription factor ETS1, a direct target of ERK and a transcription factor reported to regulate IL-8 expression. Our work reveals that SHP2 mediates breast cancer progression by enhancing the production and secretion of the pro-metastatic cytokine IL-8. We also provide mechanistic insights into the effects of SHP2 inhibition and its downstream repercussions. Overall, these results support a rationale for targeting SHP2 in breast cancer.

Keywords: Breast cancer; CXCL8; Cytokines; ETS1; IL-8; MAPK; Microenvironment; PTPN11; Phosphatase; SHP2.

Fig. 1

SHP2 blockade downregulates IL-8. (A) Bar graph representing the quantification of the nine most downregulated cytokines in SUM159 cell supernatants treated with SHP099 for 48 h relative to vehicle-treated cells. Data shown are dot quantification by pixel density from cytokine-array scans. (B) Bar graph representing IL-8 protein abundance in the supernatant of the indicated cell lines treated for 48 h at the SHP099 concentrations shown. Data are IL-8 protein concentrations normalized by total protein ± standard deviation (S.D.) (n = 3, ****P ≤ 0.0001, Two-way ANOVA test). (C) Bar graph representing CXCL8 mRNA expression in the indicated cell lines after 120 h of SHP099 inhibition (n = 3, **P ≤ 0.001, Two-way ANOVA test). (D) Bar graph representing IL-8 protein abundance in the supernatant of the indicated cell lines treated for 24 h with DMSO or 10 µM SHP099 normalized to the protein content of each well determined by SRB staining. Washout conditions were collected 24 h after media exchange. (n = 3, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, Two-way ANOVA test). (E) Bar graph representing IL-8 protein abundance in the supernatants of the indicated cell lines upon SHP2 knockdown. Data are shown ± S.D. (n = 3, ****P ≤ 0.0001, Two-way ANOVA test). (F) Bar graph representing CXCL8 mRNA expression in the indicated cell lines upon SHP2 knockdown. Data are shown ± S.D. (n = 3, *P ≤ 0.05, **P ≤ 0.01; Two-way ANOVA test)

Fig. 2

SHP2/MAPK enhances IL-8 levels via the transcription factor ETS1. (A) Scatter plot representing ranked phospho-tyrosine peptides depleted (left) or enriched (right) in SUM159 cells after 30 min of SHP099 treatment. Dark blue dots indicate phospho-peptides known to be involved in MAPK pathway regulation. The dotted line represents an arbitrary cutoff value. (B–C) Curves representing IL-8 protein abundance in the supernatants of the indicated cell lines upon MEK162 treatment at the indicated time points. Data shown are IL-8 protein concentrations normalized by total protein. Fold changed based on the 2h time point, ± S.D. (n = 3, ***P ≤ 0.001, ****P ≤ 0.0001, Two-way ANOVA test). (D) Bar graphs showing the ETS1 transcription factor activity profile (left) in SHP2 knockdown SUM159 tumors and MCF10A-HER2/HER3 3D cultures and its DNA-binding motif (right). (E) Bar graphs representing ETS1 and CXCL8 mRNA expression in the indicated cell lines 96 h after addition of siRNA (n = 3, *P ≤ 0.05, ***P ≤ 0.001, Two-way ANOVA test)