Supplementary Data from A Novel ACKR2-Dependent Role of Fibroblast-Derived CXCL14 in Epithelial-to-Mesenchymal Transition and Metastasis of Breast Cancer

Abstract

<p>FIGURES. Fig S1, Bioinformatics reveal potential CXCL14-receptor candidates; Fig. S2, Expression of CXCL14 correlates with EMT in breast cancer patients; Fig. S3, Expression of CXCL14 correlates with EMT in breast, prostate and ovarian cancer patients; Fig. S4, Enhanced expression of CXCL14 and ACKR2 correlates with EMT in ovarian and prostate cancer patients; Fig. S5, CXCL14-fibroblasts stimulate the EMT program in xenograft tumors; Fig. S6, Soluble factors secreted by CXCL14-fibroblasts induce EMT of breast cancer cells in vitro; Fig. S7, CXCL14-fibroblasts enhance invasion of breast cancer cells; Fig. S8, CXCL14-responsive and non-responsive breast cancer cell lines; Fig. S9, CXCL14 signals through a Gαi-coupled receptor; Fig. S10, CXCL14 responses depend on ACKR2; Fig. S11, Downregulation of ACKR2 abolishes CXCL14-induced ERK signaling; Fig. S12, CXCL14 is not a direct ligand of ACKR2; Fig. S13, Correlation of CXCL14 expression levels and EMT does not reflect stroma abundance; Fig. S14, Enhanced expression of CXCL14 and ACKR2 predicts worse outcome in breast cancer patients; Fig. S15, Identification of potential mediators of CXCL14-fibroblast- stimulated breast cancer EMT; Fig. S16, A CXCL14-ACKR2-NOS1 axis in fibroblasts stimulates EMT and invasion of breast cancer cells // TABLES. Table S1, Primers for the amplification of human and murine genes by qRT-PCR; Table S2, Scored levels of E-Cadherin and CytK8/18 in MCF7 breast xenograft tumors, and CytK8/18 in LnCaP prostate xenograft tumors; Table S3, CXCL14 candidate receptor expression in CXCL14-responsive and non-responsive cell lines; Table S4, Multivariable analysis for CXCL14high/ACKR2high subgroup of breast cancer patients from the TCGA gene expression dataset; Table S5, Univariable cox-regression analyses for the CXCL14 (high)/ACKR2(high) subgroup (90-percentile) in publicly available gene expression datasets (TCGA) from 13 different tumor types.</p>