Search

Supplementary Figure 1 (Fig S1) shows identification of a WNK1 inhibitor from a chemical screen. Fig S2 shows relationship among AXL, WNK1, OSR1 expression and migration in breast cancer cells. Fig S3 shows effect of WNK inhibitor on cell viability and tumor IHC. Fig S4 shows schematic representation of collaboration among WNK1, OSR1 and AXL.

Metastasis is the major cause of mortality in patients with breast cancer. Many signaling pathways have been linked to cancer invasiveness, but blockade of few protein components has succeeded in reducing metastasis. Thus, identification of proteins contributing to invasion that are manipulable by small molecules may be valuable in inhibiting spread of the disease. The protein kinase with no lysine (K) 1 (WNK1) has been suggested to induce migration of cells representing a range of cancer types. Analyses of mouse models and patient data have implicated WNK1 as one of a handful of genes uniquely linked to invasive breast cancer. Here, we present evidence that inhibition of WNK1 slows breast cancer metastasis. We show that depletion or inhibition of WNK1 reduces migration of several breast cancer cell lines in wound healing assays and decreases invasion in collagen matrices. Furthermore, WNK1 depletion suppresses expression of AXL, a tyrosine kinase implicated in metastasis. Finally, we demonstrate that WNK inhibition in mice attenuates tumor progression and metastatic burden. These data showing reduced migration, invasion, and metastasis upon WNK1 depletion in multiple breast cancer models suggest that WNK1 contributes to the metastatic phenotype, and that WNK1 inhibition may offer a therapeutic avenue for attenuating progression of invasive breast cancers.

Live imaging of explants containing MCFDCISO cells (H2B:mCherry, red, nuclei) with or without 578T cells (H2B:GFP, green, nuclei) showing cell motility.

Live imaging of the inset region of clustered spheroids containing SUM159O cells expressing control or a p63 shRNA (H2B:mCherry, red, nuclei) showing cell motility.

A) Representative images of hematoxylin and eosin (H&E) staining of MCFDCIS, MCFDCIS-Î"4 and mixed (4:1 respectively) tumors that were injected in the mammary fat pad of SCID/Beige mice after two or five weeks. MCFDCIS-Î"4 tumors were not collected at two weeks due to small tumor size. Scale bar=50μm. B) Growth of subcutaneous tumors of DCIS or DCIS-Î"4-hy in athymic nude mice. n=5. **p

A list of differentially expressed genes in tumors and cells.

A) Principal component analyses of RNA-Seq from MCFDCIS and MCF10A cell lines and the AIB1Î"4 corresponding clones. B) Relapse free survival and overall survival KM-plots in patients with ER-negative breast tumors separated by AIB1Î"4 top regulated genes in MCFDCIS cells. AIB1Î"4 signature gene list used to generate the KM-plots is in Supplementary table S3. C) Volcano plots illustrating differentially expressed genes (DEGs) regulated by AIB1Î"4 in MCF10A cell clones (|log2(FC)|>1.5 and -log(adj p-val)>1.3). Full list of DEGs in Supplementary table S1. D) Significantly enriched hallmark pathways from gene set enrichment analyses (GSEA) associated with AIB1Î"4 expression in MCF10A cells. Full list of enriched pathways in Supplementary table S2. E) Top canonical pathways altered in MCF10A-Î"4 compared to control MCF10A cells identified by Ingenuity Pathway Analysis (IPA). |z-score|>1, p-val 1.5 and -log(adj p-val)>1.3). Full list of DEGs in Supplementary table S1.

A) FDR-corrected shared and unique ChIP peaks of AIB1 and AIB1Î"4 in MCF10A parental cell line (10A) versus AIB1Î"4-expressing isogenic line (10A-Î"4). Shared peaks have at least 1bp overlap. B) AIB1 ChIP peaks in MCF7 cells treated for 3 hours with estradiol (E2) or vehicle (V) overlapped with MCF10A and MCFDCIS (parental and AIB1Î"4 derivative lines). MCF7 data set obtained from Zwart et al(32). C) Motifs within 200 bp of the peak summit for significant ChIP-Seq peaks were analyzed for motif enrichment using HOMER. In red are the motifs unique to AIB1Î"4 in MCF10A cells. D) FDR-corrected histone modification peaks in DCIS and DCIS-Î"4 overlapping with enhancer sites from human mammary epithelial cells (hMEC enh.) (42). E) The distribution of AIB1, H3K27ac or H3K4me1 ChIP peaks 500 kb up- and downstream of gene transcription start site (TSS). Genomic distance was divided into 5kb bins.

A) Quantification of the number of Zebrafish embryos with DCIS or DCIS-Î"4-hy cells that have extravasated out of the blood vessels and into the neighboring tissue. p=0.57. B) Quantification of the number of Zebrafish embryos with MCF10A, MCF10A-Î"4 or mixed cells that have extravasated out of the blood vessels and into the neighboring tissue. In the mixed cell population, only MCF10A parental line was fluorescently labeled and scored for extravasation. MCF10A and MCF10A-Î"4 are mixed at a 4:1 ratio. C) Invasion assay of an endothelial monolayer (HUVEC) using ECIS. MCFDCIS cells were treated with conditioned media (CM) from MCFDCIS-Î"4 cells for 4 hours before added to the endothelial monolayer, and vice versa. No cells added to the endothelial monolayer as a negative control. D) A schematic showing a three-chambered co-culture system. Top chamber has electrodes underneath a porous membrane to detect migrating cells. Middle chamber is only permeable to factors but not cells. Bottom chamber harbors cells that can crosstalk with migrating cells through secreted factors. E) Real-time migration rate of MCFDCIS and MCFDCIS-Î"4 in co-culture chambers. F) MCFDCIS and MCFDCIS-Î"4 cells were aggregated separately then embedded in 80% col I and 20% Matrigel together to monitor their crosstalk as separate spheres. Scale bar = 50μm.

Defining how interactions between tumor subpopulations contribute to invasion is essential for understanding how tumors metastasize. Here, we find that the heterogeneous expression of the transcription factor ΔNp63 confers distinct proliferative and invasive epithelial-to-mesenchymal transition (EMT) states in subpopulations that establish a leader–follower relationship to collectively invade. A ΔNp63-high EMT program coupled the ability to proliferate with an IL1α- and miR-205–dependent suppression of cellular protrusions that are required to initiate collective invasion. An alternative ΔNp63-low EMT program conferred cells with the ability to initiate and lead collective invasion. However, this ΔNp63-low EMT state triggered a collateral loss of fitness. Importantly, rare growth-suppressed ΔNp63-low EMT cells influenced tumor progression by leading the invasion of proliferative ΔNp63-high EMT cells in heterogeneous primary tumors. Thus, heterogeneous activation of distinct EMT programs promotes a mode of collective invasion that overcomes cell intrinsic phenotypic deficiencies to induce the dissemination of proliferative tumor cells.Significance:These findings reveal how an interaction between cells in different EMT states confers properties that are not induced by either EMT program alone.

A list of AIB1Î"4 signature genes that were used to generate KM-plots in Figure S5.

Live imaging of clustered spheroids containing SUM159T-dox:Î"Np63 cells {plus minus} doxycycline (H2B:GFP, green, nuclei) showing cell motility.

A) Sequence alignment of PCR amplicons of genomic DNA spanning exon 4 of AIB1 (UCSC genome browser). Black filled boxes are the sequences that align to the genome and dotted lines represent the missing/deleted sequences. AIB1 in the DCIS-Î"4 cell line is missing part of intron 3 and exon 4. Two deleted regions around exon 4 of AIB1 were identified in the 10A-Î"4 cell line; both are missing exon 4 acceptor site. B) Sanger sequencing of PCR amplicons spanning exons 2, 3, 4, 5 splice junctions. C) AIB1 mRNA between exon 2 and exon 5 in the MCFDCIS-Î"4 cell clones and the parental MCFDCIS cells. D) Western blot shows AIB1 isoforms' protein expression in MCFDCIS-Î"4 cell clones and the parental MCFDCIS cells. E) Schematic of the migration, trans-endothelial invasion and proliferation assays using real time electric cell impedance sensing (ECIS) to detect migrating cells, invasion across an endothelial monolayer or proliferation rate. F) Invasion of DCIS-Î"4-hy across an endothelial monolayer using ECIS compared to parental control DCIS cells. G) Cell proliferation of MCFDCIS (DCIS), DCIS-Î"4 and DCIS-Î"4-hy measured by electric cell impedance sensing (ECIS). A single well coated with electrodes was used to measure real time change in cell impedance as they proliferate over 80 hours (See diagram in S1E). H) Cell proliferation of MCFDCIS and MCF10A parental lines and their AIB1Î"4 expressing counterpart. ***p

Supplemental Figures S1-S5 and Supplemental Legends

A list of enriched pathways using Gene Set Enrichment Analysis (GSEA).

Live imaging of clustered spheroids containing SUM159O cells expressing control or a p63 shRNA (H2B:mCherry, red, nuclei) showing cell motility.

AIB1Δ4 is an N-terminally truncated isoform of the oncogene amplified in breast cancer 1 (AIB1) with increased expression in high-grade human ductal carcinoma in situ (DCIS). However, the role of AIB1Δ4 in DCIS malignant progression has not been defined. Here we CRISPR-engineered RNA splice junctions to produce normal and early-stage DCIS breast epithelial cells that expressed only AIB1Δ4. These cells showed enhanced motility and invasion in 3D cell culture. In zebrafish, AIB1Δ4-expressing cells enabled invasion of parental cells when present in a mixed population. In mouse xenografts, a subpopulation of AIB1Δ4 cells mixed with parental cells enhanced tumor growth, recurrence, and lung metastasis. AIB1Δ4 chromatin immunoprecipitation sequencing revealed enhanced binding to regions including peroxisome proliferator-activated receptor (PPAR) and glucocorticoid receptor (GR) genomic recognition sites. H3K27ac and H3K4me1 genomic engagement patterns revealed selective activation of breast cancer-specific enhancer sites by AIB1Δ4. AIB1Δ4 cells displayed upregulated inflammatory response genes and downregulated PPAR signaling gene expression patterns. In the presence of AIB1Δ4 enabler cells, parental cells increased NF-κB and WNT signaling. Cellular cross-talk was inhibited by the PPARγ agonist efatutazone but was enhanced by treatment with the GR agonist dexamethasone. In conclusion, expression of the AIB1Δ4-selective cistrome in a small subpopulation of cells triggers an “enabler” phenotype hallmarked by an invasive transcriptional program and collective malignant progression in a heterogeneous tumor population.Significance:A minor subset of early-stage breast cancer cells expressing AIB1Δ4 enables bulk tumor cells to become invasive, suggesting that selective eradication of this population could impair breast cancer metastasis.

Live imaging of the inset region of clustered spheroids containing SUM159T-dox:Î"Np63 cells {plus minus} doxycycline (H2B:GFP, green, nuclei) showing cell motility.

Supplementary Methods

Live imaging of SUM159T cells embedded in ECM containing fluorescent microbeads (blue) showing ECM reorganization through bead movement.

MOV file - 314K, Live-imaging of a HCC1806-H2B:GFP spheroid.

MOV file - 125K, Live-imaging of a MCFDCIS-mCherryCAAX spheroid.

MOV file - 386K, Live-imaging of a HCC1954-H2B:GFP spheroid co-cultured with fibroblasts.

MOV file - 73K, Live-imaging of a T47D-H2B:GFP spheroid co-cultured with fibroblasts.

Most ductal breast carcinoma cells are weakly invasive in vitro and in vivo, suggesting that components of their microenvironment may facilitate a transition from in situ to invasive stages during progression. Here, we report that coculture of mammary fibroblasts specifically triggers invasive behavior in basal-type breast cancer cells through a ligand independent mechanism. When cultured alone in organotypic culture, both basal- and luminal-type breast cancer cells formed noninvasive spheroids with characteristics of ductal carcinoma in situ (DCIS). In contrast, when cocultured with mammary fibroblasts, basal-type spheroids exhibited invasive character whereas the luminal-type spheroids retained a benign and noninvasive duct-like architecture. Real-time imaging and functional studies revealed that the specificity of invasion was linked to a unique capacity of basal-type breast cancer cells to move within spheroids. Mammary fibroblasts induced invasion by triggering basal-type breast cancer cells to convert from a noninvasive program of mammary epithelial morphogenesis to an invasive program of sprouting endothelial angiogenesis. Contrary to the existing invasion models, soluble ligands produced by the fibroblasts were not sufficient to trigger invasion. Instead, basal-type invasion relied upon a Cdc42-dependent reorganization of collagen fibers in the extracellular matrix by fibroblasts. Inhibiting basal-type cell movement with clinically relevant drugs blocked invasion both in organotypic culture and in animals, suggesting a new treatment strategy for early-stage patients. Together our findings establish that fibroblast recruitment by basal-type breast cancer cells into early-stage tumors is sufficient to trigger their conversion from a benign, noninvasive DCIS-like stage to a malignant invasive stage. Furthermore, our findings suggest that different subtypes of breast cancer may require distinct types of contributions from the microenvironment to undergo malignant progression. Cancer Res; 71(21); 6857–66. ©2011 AACR.

MOV file -434K, Live-imaging of a MCFDCIS-H2B:mCherry spheroid.

MOV file - 176K, Live-imaging of a HCC1428-H2B:GFP spheroid co-cultured with fibroblasts.

Contains Supplementary Figures S1-S7, Supplementary Figure Legends and Supplementary Methods.

MOV file - 36K, Live-imaging of a MCFDCIS-H2B:mCherry spheroid co-cultured with mammary fibroblasts treated with diluent.

MOV file - 93K

MOV file - 50K, Live-imaging of a HCC1954-H2B:GFP spheroid used for tracking cell movement.

MOV file - 2MB, Phase contrast and H2B:mCherry images from live-imaging of a MCFDCIS-H2B:mCherry spheroid.

MOV file - 35K, Live-imaging of a T47D-H2B:GFP spheroid used for tracking cell movement.

PDF file - 2MB, Expression of Cdc42 in mammary fibroblasts transfected with siRNAs.

PDF file - 3.9MB, Residual tumor lesions after MEK1/2 inhibitor treatment are composed of MCFDCIS cells.

Supplementary tables related to miRNA and mRNA expression analysis.

MOV file - 773K, Live-imaging of MCFDCIS-mCherryCAAX cells invading towards mammary fibroblasts expressing GFP.

MOV file - Live-imaging of a small MCFDCIS-H2B:mCherry spheroid co-cultured with mammary fibroblasts.

MOV file - 232K, Live-imaging of a HCC1954-H2B:GFP spheroid.

MOV file - 76K, Live-imaging of a T47D-H2B:GFP spheroid.

PDF file - 7MB, Fibroblast induced collective invasion does not require a loss of epithelial character.

PDF file - 1.6MB, EGFR and ERK1/2 are necessary for HCC1806 cell morphogenic movement and fibroblast induced collective invasion.

MOV file - 162K, Live-imaging of a HCC1428-H2B:GFP spheroid.

PDF file - 5MB, Cancer cells in the spheroids follow the path of the initial invading cells.

MOV file - 200K, Live-imaging of a HCC1806-H2B:GFP spheroid used for tracking cell movement.

MOV file - 45K, Live-imaging of a MCFDCIS-H2B:mCherry spheroid co-cultured with mammary fibroblasts treated with the MEK1/2 inhibitor PD0325901.

MOV file - 3.1MB, Live-imaging of a HCC1428-H2B:GFP spheroid used for tracking cell movement.

PDF file - 2MB, Tracking the movement of breast cancer cells in spheroids.

MOV file - 1.5MB, Live-imaging of a MCFDCIS-mCherryCAAX spheroid and FITC-Collagen-1 when co-cultured with mammary fibroblasts.

Supplementary tables related to miRNA wounding screen.