Subcellular localization to the cell surface membrane in IBC cells

To further study the localization of matriptase and c-Met at the subcellular level, the endogenous proteins in the IBC line SUM149 were visualized by immunocytochemistry and confocal imaging (Figure ​(Figure2).2). Matriptase and c-Met were mainly expressed on the cell surface, as expected, based on the membrane topology of both proteins. Analysis of confocal images shows extensive localization of both matriptase and c-Met on the cell surface membrane with weaker cytoplasmic staining.

Open in a separate window

Figure 2

Matriptase and c-Met proteins are co-expressed in IBC cell lines

SUM149 cells were permeabilized and incubated with mouse anti-matriptase (M24) and rabbit anti-C-Met, followed by fluorophore-conjugated secondary antibodies. The cells were visualized by confocal fluorescence microscopy: Alexa Fluor 488, matriptase (upper left panel); Alexa Fluor 546, c-Met (upper right panel); DAPI/matriptase/c-Met merged (lower left panel) and matriptase/c-Met merged (lower right panel). Images were analyzed with Volocity^™ software. A representative orthogonal section from the original Z-stack was used to assess co-localization (yellow) in XY, XZ, and YZ planes. The Y and Z plane views were taken at the position indicated with an open arrow head. Scale bars, 10 μm.

Taken together with the results above, matriptase and c-Met are co-expressed in the majority of IBC patient samples analyzed and localize on the surface of IBC cells. Furthermore, the majority of IBC patient samples express E-cadherin.

Matriptase silencing abrogates pro-HGF mediated c-Met signaling in human IBC cell lines

To examine the molecular and functional relationship between matriptase and c-Met in IBC, we employed cell culture models in both 2D and 3D. SUM149 (ER-/PR-/HER2-) and SUM190 (ER-/PR-/HER2+) cells were originally isolated from primary inflammatory invasive ductal carcinoma and display characteristics commonly found in IBC tumors, including expression of E-cadherin [20–23].

Analysis of cellular proteins from SUM149 and SUM190 by western blotting showed that matriptase and c-Met are expressed in both cell lines (Figure 3A, 3A', 3B, 3B' and Supplementary Figure S2) and that stimulation with pro-HGF/HGF is required for c-Met activation since no phosphorylated c-Met was detected in the absence of pro-HGF/HGF (Figure 3A' and 3B', top panels “No HGF”). Based on these characteristics, the two IBC cell lines were deemed appropriate for further studies on matriptase/c-Met function.

Open in a separate window

Figure 3

Matriptase is essential for activation of the pro-HGF/c-Met signaling pathway and proliferation in IBC cells

RNAi silencing of matriptase in SUM149 cells (A) or SUM190 cells (B) with three independent non-overlapping synthetic RNA duplexes (siM1, siM2, and siM3). A %GC matched RNA duplex was used as negative control as well as a mock where no RNA duplex was added. (A', B') Cells were serum starved and exposed to vehicle, active HGF (2 min), or pro-HGF (15 min) as indicated. Cell lysates were separated by SDS-PAGE and analyzed for the presence of activated c-Met and its downstream targets by western blotting. As expected, no phosphorylation of c-Met was observed when no HGF was added, whereas addition of pre-cleaved, active HGF resulted in phosphorylation of c-Met in all samples. When pro-HGF was added, no phosphorylation of c-Met or its downstream targets, Gab1 and AKT, was observed in cells where matriptase was silenced with any of the three RNA duplexes. (C) SUM149 cells were grown on culture plates and matriptase was silenced using the three synthetic RNA duplexes mentioned above. The cells were serum starved and left untreated (black bars) or exposed to either active HGF (gray bars) or pro-HGF (white bars) for 24 h and counted. Bar graphs show the proliferation at 24 h after stimulation (*P < 0.05, **P <0.01). Error bars represent S.D. of triplicates within one assay. Data are representative of four independent experiments.

It has been demonstrated that matriptase can cleave and activate the pro-form of HGF [24]. However, no studies describing a role for matriptase as a pro-HGF activating protease in IBC have been reported. To determine whether matriptase regulates c-Met signaling via activation of pro-HGF, SUM149 and SUM190 cells were stimulated with pro-HGF or active HGF and the phosphorylation/activation state of c-Met and its downstream signaling proteins, Gab1 and AKT, were determined by western blot analysis. HGF is secreted in its inactive pro-form by stromal cells in both human and mice, including stromal mammary fibroblasts and macrophages [25–27]. The source of pro-HGF in this experimental set-up was human mammary fibroblasts expressing pro-HGF [16]. To directly assess the role of matriptase for pro-HGF activation, experiments using RNAi mediated silencing of matriptase were performed. Silencing of matriptase with three independent non-overlapping synthetic RNA duplexes was achieved in SUM149 and SUM190 cells (Figure ​(Figure3A3A and ​and3B).3B). Addition of pro-HGF to matriptase sufficient control cells resulted in robust activation of c-Met and the downstream targets Gab1 and AKT (Figure 3A' and 3B', “pro-HGF”). In contrast, matriptase silencing led to greatly reduced or undetectable levels of c-Met pathway activation in both SUM149 and SUM190 cells. Importantly, the matriptase silenced cells displayed unaltered activation of c-Met upon stimulation with pre-cleaved active HGF, demonstrating that the impaired response to pro-HGF is caused by the lack of matriptase-mediated cleavage, while the cells remain fully HGF/c-Met signaling competent (Figure 3A' and 3B', second panel, “active HGF”).

To assess the functional consequences of abrogating the matriptase-meditated c-Met signaling in IBC cells, a proliferation assay was performed. As described above, matriptase was silenced in SUM149 with three different synthetic RNA duplexes and cells were stimulated with pro-HGF or active HGF. As expected, there was no significant proliferation difference in matriptase sufficient control cells treated with active HGF or pro-HGF. In contrast, proliferation in matriptase silenced cells was significantly impaired in response to pro-HGF (Figure ​(Figure3C,3C, white bars) in comparison to the response observed with addition of pre-cleaved active HGF (Figure ​(Figure3C,3C, grey bars).

Matriptase silencing impairs pro-HGF/c-Met mediated invasion in human IBC cells

The standard 2D cell culture models described above are optimal for biochemical analysis of matriptase-mediated c-Met signaling in IBC. However, additional models of IBC are beneficial to study the functional importance of this pathway. Breast cancer cells grown in 3D cultures are believed to better recapitulate in vivo morphology and invasion than cells grown on tissue culture plates. SUM149 cells were transiently silenced using matriptase RNA duplexes or control duplexes and transferred to 3D culture containing growth factor depleted reconstituted basement membrane (rBM) (Figure ​(Figure4).4). All cells formed 3D spheroid structures within 24 hours in this model (Figure 4A, 4B, 4D, 4E no HGF). Upon stimulation with pro-HGF or active HGF for 24 hours, matriptase sufficient control cells displayed pronounced invasive growth characterized by branching cellular structures and single cells with a flattened spindle-like appearance (Figure 4A', 4D'). In contrast, matriptase silencing (Mat KD) rendered the cells unresponsive to pro-HGF (Figure 4B', 4E'), thus, keeping the vast majority of the cells confined in compact spheroid structures (Figure ​(Figure4C).4C). Addition of pre-cleaved active HGF restored the ability of matriptase-deficient cells to display a similar phenotype to control cells, suggesting that the c-Met signaling pathway remained intact (Figure 4B″, 4E″). These data demonstrate that matriptase potentiates the invasive capacity of IBC cells by activation of pro-HGF.

Open in a separate window

Figure 4

Matriptase mediates pro-HGF induced invasion of IBC in 3D culture

Matriptase was silenced in SUM149 cells (Mat KD) using synthetic RNA duplexes (B, B', B″). A %GC matched duplex was used as negative control (A, A', A″). Cells formed spheroids in 3D rBM overlay culture (A, B) before addition of either pro-HGF (A', B') or active HGF (A″, B″) and imaged by confocal microscopy. Cells were labeled with Cell Tracker Orange and nuclei with Hoechst (blue). When no growth factors were added, control and Mat KD spheroids remained intact. In contrast, addition of pro-HGF induced extensive invasive outgrowths in matriptase sufficient control cells (A') whereas the majority of Mat KD spheroids remained intact (B'). Both control and Mat KD spheroids responded to active two-chain HGF (A″, B″). Scale bars, 50 μm (C) 3D rBM cell cultures were scored for the number of intact control spheroids (black bars) and intact Mat KD spheroids (white bars) when left untreated or exposed to either active HGF or pro-HGF. Data represent mean of the number of intact spheroids before treatment relative to the number after treatment (*P < 0.05). Error bars represent S.D. of triplicates within one assay. Data are representative of four independent experiments. (D–E) Low magnification photos of independent experiment performed as above. Pro-HGF (D', E'), HGF (D″, E″), or no growth factor (D, E). When no HGF was added, control and Mat KD spheroids remained intact (D, E) (black arrowheads). In contrast, addition of pro-HGF induced extensive scattering and invasion in matriptase sufficient control cells (D') (indicated with open arrowheads) whereas the majority of spheroids in Mat KD cells remained intact (E'). Both control and Mat KD responded to active two-chain HGF (D″, E″) Scale bars, 200 μm.

Immunohistochemistry

Tissue slides were deparaffinized with xylene and hydrated with graded ethanol solutions. Antigen retrieval was performed using citrate buffer, pH 6.0 (Bethyl Laboratories, Montgomery, TX) and incubation for 1 h in a 73^°C water bath. The slides were blocked with 2% bovine serum albumin in PBS, and immunostained overnight at 4^°C. Primary antibodies were rabbit anti-matriptase (Calbiochem/EMD Millipore, San Diego, CA), rabbit anti-c-Met (Leica Microsystems Inc., Buffalo Grove, IL), and mouse anti-E-cadherin (Pharmingen/BD Biosciences, San Jose, CA). As negative controls, non-immune mouse IgG (Sigma, St. Louis, MO) or non-immune rabbit IgG (NeoMarkers, Fremont, CA) were used. Bound antibodies were visualized using biotin-conjugated anti-rabbit or anti-mouse (Vector Laboratories, Burlingame, CA) secondary antibodies and a Vectastain ABC kit (Vector Laboratories). 3,3′-diaminobenzidine (DAB) was used as substrate (Sigma, St. Louis, MO) and arrays were counterstained with hematoxylin. All microscopic images were acquired on a Zeiss Scope A.1 using digital imaging.

Human breast cancer cell lines

The SUM149 and SUM190 cell lines were a gift from Dr. Stephen Ethier (Medical University of South Carolina, Charleston, SC). SUM149 cells were grown in 5% IH media (Ham's F-12 media, supplemented with 5% FBS, 1μg/ml hydrocortisone, and 5 μg/ml insulin). SUM190 cells were grown in SFIH media (Ham's F-12 media, supplemented with 1 μg/ml hydrocortisone, 5 μg/ml insulin, 5 mM ethanolamine, 10 mM HEPES, 5 μg/ml transferrin, 10 nM triodo-thyronine, 50 μM sodium selenite, and 5% BSA).

Immunocytochemistry

SUM149 cells were fixed with 10% neutral-buffered zinc formalin (Z-fix) (Anatech, Battle Creek, MI) for 20 min, blocked in 5% BSA with 0.1% Triton X-100 for 1 hr, and incubated with rabbit anti-c-Met (Cell Signaling Technology, Beverly, MA) and mouse monoclonal anti-matriptase M24 antibody [13] overnight at 4^°C. The following day, cells were washed with PBS, and incubated with anti-rabbit 546 or anti-mouse 488 antibodies (Invitrogen, Carlsbad, CA) in blocking solution for 2 hrs at RT. Cells were washed with PBS and mounted with Prolong gold with DAPI (Invitrogen). Confocal images were acquired on the Zeiss LSM 780 scope at the Microscopy Imaging and Cytometry Resources Core at Wayne State University School of Medicine. Images were analyzed using Volocity^™ software.

RNAi silencing

For matriptase knockdown in SUM149 and SUM190 IBC cells, three independent Stealth RNAi^™ siRNA duplexes targeting matriptase (siM1-ST14HSS186125, siM2-ST14HSS186126, siM3- ST14HSS110268), as well as %GC matched negative controls were used as previously described (Invitrogen) [14]. Two days after transfection, the cells were serum starved and used in the pro-HGF/HGF activation assay, in the 3D invasion assay, and in the proliferation assay described below.

Matriptase synthetic inhibitor

A selective, slow, tight-binding inhibitor of matriptase containing a ketobenzothiazole serine trap (IN-1) was designed based on the auto-catalytic domain (RQAR) of matriptase. Details on the synthesis and characteristics of the inhibitor have been described previously [15]. IN-1 was solubilized in PBS and used at concentrations ranging from 10 nM to 10 μM.

Pro-HGF/SF activation assays

Pro-HGF/SF was produced in the immortalized human fibroblasts cell line, RMF-HGF, as described [16, 17]. In order to estimate the concentration of each preparation and ensure quality, western blot analysis was performed using a goat anti-HGF primary antibody (R&D Systems, Minneapolis, MN) with recombinant human recombinant active HGF (R&D Systems) as a standard. Human IBC cell lines were plated in 6-well plates, grown to confluence, and serum starved for a minimum of 3 hrs. The media was then changed to either fresh serum free media, serum free media with 100 ng/ml recombinant active two-chain HGF (R&D Systems), or pro-HGF with or without IN-1. To determine the effect on activation of c-Met and the downstream signaling molecules, Gab1 and AKT, the cells were lysed in ice cold RIPA lysis buffer (150 mM NaCl, 50 mM Tris, 1% NP-40, 0.1% SDS, pH 7.4) with protease inhibitor cocktail and Sodium Orthovanadate (Sigma). Activation levels of c-Met, Gab1, and AKT were determined by assessing the levels of the phosphorylated forms relative to the level of the total amount of the specific signaling molecules by western blot analysis.

Western blot analysis

The protein concentration in cell lysates was determined by BCA assay (Pierce, Rockford, IL) and lysates were separated by 4–12% reducing SDS-PAGE and blotted onto polyvinylidene difluoride (PVDF) membranes (Invitrogen). The following primary antibodies were used for detection: rabbit anti- matriptase (CalBioChem, Philadelphia, PA), rabbit anti-phospho cMet (Y1234/1235), anti-phospho Gab1 (y627), anti-Gab1, anti-phospho AKT (S473), anti-AKT (C73H10), mouse anti-human c-Met (25H2) (all from Cell Signaling Technology), and mouse anti-beta-actin (Sigma). For detection, secondary antibodies conjugated with either alkaline phosphatase (Sigma) or horseradish peroxidase (Chemicon, Temecula, CA) in combination with either the chromogenic substrate nitro-blue tetrazolium and 5-bromo-4-chloro-3′-indolyphosphate (Roche, Indianapolis, IN) or Super-SignalWest Femto Chemiluminescent Substrate (Pierce, Rockford, IL) were used.

3D cell culture

The 3D protocol described in [18] was used. Briefly, 1 × 10⁴ cells were plated on 12 mm circular cover glasses coated with Matrigel Matrix Growth Factor Reduced (MMGFR) (BD Biosciences) with serum free Mammary Epithelial Cell Growth Medium (MEGM™) media (Lonza, Walkersville, MD) containing the BulletKit™ growth supplement (BPE, hydrocortisone, GA-1000, Insulin). An overlay of MEGM growth media containing 2% MMGFR was added 1 hr after seeding. Spheroids were allowed to form overnight before addition of 100 ng recombinant HGF (R&D Biosystems) or pro-HGF conditioned media for 24 hrs after which cells were stained with 5 uM Cell Tracker Orange (Invitrogen) for 45 min and Hoechst 33342 (Life Technologies Carlsbad, CA) for 5 min prior to imaging. Confocal images were acquired on the Zeiss LSM 780 scope at the Microscopy Imaging and Cytometry Resources Core at Wayne State University School of Medicine.

This work was supported by a NIH Ruth L. Kirschstein National Research Service Award T32-CA009531 (G.L.Z.), The American Cancer Society PF-14-168-01-CSM (L.M.T), NIH/NCI NCI RCA60565A grant (K.L.), and KG100892 Susan G. Komen Research Programs (K.L.). J.E.L was supported in part by award number P30CA014089 from the National Cancer Institute. The RMF-HGF cell line was kindly provided by Dr. Charlotte Kuperwasser, Tufts University. We are grateful to Dr. Alfredo Molinolo, University of California San Diego, for providing input and guidance.