Global phosphotyrosine survey in triple-negative breast cancer reveals activation of multiple tyrosine kinase signaling pathways

Abstract

Breast cancer is the most prevalent cancer in women worldwide. About 15-20% of all breast cancers are triple negative breast cancer (TNBC) and are often highly aggressive when compared to other subtypes of breast cancers. To better characterize the biology that underlies the TNBC phenotype, we profiled the phosphotyrosine proteome of a panel of twenty-six TNBC cell lines using quantitative high resolution Fourier transform mass spectrometry. A heterogeneous pattern of tyrosine kinase activation was observed based on 1,789 tyrosine-phosphorylated peptides identified from 969 proteins. One of the tyrosine kinases, AXL, was found to be activated in a majority of aggressive TNBC cell lines and was accompanied by a higher level of AXL expression. High levels of AXL expression are correlated with a significant decrease in patient survival. Treatment of cells bearing activated AXL with a humanized AXL antibody inhibited cell proliferation and migration in vitro, and tumor growth in mice. Overall, our global phosphoproteomic analysis provided new insights into the heterogeneity in the activation status of tyrosine kinase pathways in TNBCs. Our approach presents an effective means of identifying important novel biomarkers and targets for therapy such as AXL in TNBC.

Keywords: AXL; kinase; protein phosphorylation; proteomics; triple negative breast cancer.

Figure 1. Systematic phenotyping and phosphotyrosine profiling of triple negative breast cancer cell lines

A. Characterization of TNBC cell lines according to the extent of invasion in Boyden matrigel chamber (x axis) and colony formation in soft agar (y axis). B. Global protein tyrosine phosphorylation pattern across the panel of cell lines. Anti-phosphotyrosine antibody was used for immunoprecipitation and western blotting to detect tyrosine-phosphorylated proteins.

Figure 2. Mass spectrometry-based quantitative phosphotyrosine profiling

A. A schematic illustration of SILAC spike-in based quantitative phosphotyrosine profiling approach. Lysates extracted from each TNBC cell line cultured in “light” medium were spiked-in with “heavy” SILAC labeled MDA-MB-231 cell lysates. The lysates were digested with trypsin followed by enrichment of phosphorylated tyrosine-containing peptides using anti-phosphotyrosine antibody (pY100). Enriched phosphotyrosine peptides were then analyzed by LC-MS/MS. B. Hierarchical clustering of phosphotyrosine peptides showing association with aggressive phenotypes (invasion and soft agar colony formation) of TNBC cell lines. Gene symbols along with the corresponding phosphopeptide sequences are shown. The phosphorylated tyrosine residue is indicated by a small “y”.

Figure 3. Functional validation of tyrosine kinases in aggressive phenotype of TNBC cells

A. Western blot analysis to assess the efficiency of siRNA-based knockdown of AXL, TNK2, DYRK2, EPHA2 and TYK2 in MDA-MB-231 cells. B–D. The effect of siRNA-based knockdown of AXL, TNK2, DYRK2, EPHA2 and TYK2 on colony forming ability (B), proliferation (C) and invasive ability (D) of three TNBC cell lines, SUM159, MDA-MB-231 and HCC1395. Mann-Whitney tests were performed for statistical analyses.

Figure 4. Humanized AXL antibody hMAb173 inhibits TNBC cell proliferation and migration in vitro

A. AXL phosphorylation level correlates with aggressive phenotypes of TNBC cells. Top panel: Western blot analysis of the expression of AXL in the panel of TNBC cell lines. Color-coded plots showing the expression level of AXL (top row), pY702 AXL phosphorylation level (second row), invasiveness (third row) and colony formation ability (bottom row) across the panel of TNBC cells. Spearman's rank correlation was performed for statistical analysis. *p < 0.05. B. Activation of AXL signaling pathway in TNBC cells. Protein names in red represent the proteins commonly phosphorylated in AXL activated TNBC cells and proteins names in black represent the proteins not identified or phosphorylation not detected in all 6 AXL activated cell lines. Solid lines indicate direct phosphorylation or interaction events based on literatures and databases (KEGG, PhosphoSite). C. Western blot analysis to assess the AXL expression and phosphorylation levels of AXL-dowstream signaling proteins including MET (Y1003), AKT (T308), FAK (Y397), SRC (Y17) and P130Cas (Y249) in MDA-MB-231 cells treated with AXL antibody hMAb173. β-actin serves as the loading control. D. Cell proliferation assays of indicated cell lines treated with different doses of hMAb173 (20 μg/ml or 100 μg/ml). Mann-Whitney tests were performed for statistical analyses. E. Wound-healing assays to measure the cell migration ability of HCC1395 cells treated with 20 μg/ml or 100 μg/ml hMAb173. Human IgG (100 μg/ml) served as the control treatment. Microscopic observations were recorded 0, 6, 12, 24 and 36 hours after scratching the cell surface.

Figure 5. Humanized AXL antibody hMAb173 inhibits TNBC cell tumor formation in vivo

A. NOD-SCID mice were implanted with MDA-MB-231 cells. When tumor sizes reached approximately 50 mm³, mice were treated by intraperitoneal injection of hMAb173 (20 mg/kg) or Ctrl IgG, 2 times a week. Tumor volume was measured 3 times a week and plotted. The P value was calculated by Student's t test. Mean ± SEM is shown. B. Top panel: Immunofluorescent staining of AXL, Ki67 and pS6 for MDA-MB-231 xenograft tumors treated with control or hMAb173 antibodies. Apoptosis was examined with TUNEL assay. Nuclei were counterstained with DAPI. Hematoxylin and eosin staining was also performed. Bottom panel: The intensity of staining and the positive signal coverage area were quantified with ImageJ (NIH) and plotted. A Student's t-test (two-tailed, unpaired) was used to calculate P values between groups where indicated. C. Kaplan-Meier plot of 172 high-grade ER-negative, basal-like breast cancer patients stratified with high or low AXL gene expression. The red line represents the survival curve of patients with high expression of AXL and the black line represents the survival curve of patients with low expression of AXL.

Figure 6. Expression of AXL protein correlates with poorer survival of TNBC patients

A. Representative IHC staining of AXL in TNBC on breast cancer TMA indicating negative and positive staining. B. Kaplan-Meier plot of 57 TNBC cases stratified according to positive and negative expression of AXL, showing that AXL expression is a significant predictor of patient survival (p = 0.038, Mantel-Cox test).