Virtakoivu Et Al 2015 Supplemental Information

Virtakoivu et al 2015 – Supplemental Information

Supplementary figures 1-5 and associated legends

Supplementary materials and methods

Supplementary references

Figure S1. ERK silencing decreases breast cancer cell migration and invasion but does not influence cell proliferation. (A) Migration of control-, ERK2-, vimentin- and Slug-silenced MDA-MB-231 cells (top panel) was followed with time-lapse imaging for 13 h at 10 min intervals (n = 34-43 cells/condition). For invasion assays (bottom panel) control- and ERK2-silenced MDA-MB-231 cells were allowed to invade in Matrigel for 4 days, stained with Alexa Fluor 488 phalloidin and imaged with confocal microscopy. Invasion area was calculated from the side view (z-axis) of invading cells and the arrow indicates the direction of cell invasion (n = 4). Silencing efficiencies were analysed by western blot 72 hours post transfection. (B) Proliferation of control-, ERK-, Slug- and vimentin-silenced MDA-MB-231 cells was analysed using WST-1 reagent (n = 2; 5-8 wells per experiment). Silencing efficiencies were analysed by western blot 72 hours post transfection (mean ± SEM; * p < 0.05, *** p < 0.001). (C) Immunohistochemical staining of DAPI in orthotopically implanted MDA-MB-231 primary tumors and representative contralateral lymph node metastasis from three animals (n = 7 lymph node lesions and n = 5 primary tumors; these are from the same samples as the vimentin staining shown in Fig 1). Scale bar: 300 µm and 100 µm for region of interest (ROI) images.

Figure S2. ERK2 does not interact with actin or keratin 8 and ERK2 phosphorylation is supported by vimentin but not actin. (A) Representative western blots of total ERK and pERK levels in control or vimentin silenced MDA-MB-231 cells (n = 3). (B) ERK2 does not co-immunoprecipitate (IP) with keratin 8 or actin. Inputs are shown for keratin 8 and actin protein levels. (C) Quantification of ERK2 phosphorylation in alkaline phosphatase (AP) protection assays. Recombinant active 32P-ATP phosphorylated ERK2 alone or in combination with vimentin, was exposed to AP for the indicated times and protein phosphorylation detected by autoradiography (n = 4). (D) AP protection assay for active phosphorylated ERK2 in the presence of recombinant actin. ERK2 alone or with actin was exposed to AP for the indicated times. Quantification of pERK levels relative to total ERK are shown (n = 4). (E) Induction of stress fibers in MDA-MB-231 cells transiently transfected with active RhoA mutants (GFP-RhoA QL or GFP-RhoA V14) or GFP alone. Representative confocal images of actin stress fibres (phalloidin staining) and western blots for active ERK (pERK), total ERK, actin and GFP are shown. (n = 3 experiments). Scale bar: 20 µm. (mean ± SEM; * p < 0.05).

Figure S3. Fluorescence recovery of wild-type Slug-GFP after photobleaching (FRAP). (A-B) Fluorescence intensities were measured from the indicated regions of interest (ROI) in the nucleus, cytoplasm, and background and from another cell nucleus as a reference before and after photobleaching. (A) A representative GFP-Slug signal in a cell at the indicated time points. (B) Fluorescence intensity changes over time following photobleaching. Mono exponential fits were used for recovery curves. (C) Quantification of the recovery half-time, mobile and immobile fractions of GFP-Slug from 12 cells. The intensity values of analysed ROIs were corrected for intensity changes of background and reference cells.

Figure S4. Mass spectrometric identification of ERK1/2-dependent phosphorylation sites on Slug and validation using a newly generated Slug phospho-serine-87 antibody. (A) Slug was preincubated with ERK1 or ERK2 in in vitro kinase assays. Following in-gel trypsin digestion and TiO2 phosphopeptide enrichment, MS/MS analysis identified three phosphopeptides from ERK2-phosphorylated Slug. (B) Label-free quantification of two of the identified phosphopeptides before and after TiO2 phosphopeptide enrichment. Phosphopeptide abundance was normalized to Slug protein abundance. (C) Specificity of Slug phospho-serine-87 antibody (anti-pSlug S87) was analysed by measuring Slug phosphorylation levels in H-Ras transformed MCF10A-DCIS cells ± MEK inhibitor (U0126, 24 or 48 h). Quantification of pSlug S87 levels from 3 independent experiments is shown. (D) Antibody specificity was further validated in MCF10A cells expressing Flag-Slug wt or Flag-Slug phospho-site mutants (S87A and S87,104AA constructs).

Figure S5. The phosphorylation status of Slug does not influence repression of E-Cadherin, Slug nuclear localisation or stability. (A and B) E-cadherin promoter activity, measured with Envision, in HEK293 and MCF7 cells co-transfected with luciferase construct for E-cadherin, pIRES_slug wt/S87A/S87,104AA/dominant negative (DN) and pRL-TK plasmids (n = 3-5 for HEK293 and n = 2 for MCF7 cells). (C) Western blot analysis of E-cadherin protein levels in retrovirally-transduced MCF10A_GFP and MCF10A_GFP slug wt/mutant cells. (D) DNA binding assays analysing GST-Slug wt or S87A/S104A mutant (Slug_AA) interaction with vimentin or E-cadherin promoter DNA. Slug constructs were preincubated with active ERK2 and ATP in in vitro kinase reaction buffer prior to performing the DNA binding assays (n = 3). (E) Slug localisation was analysed by confocal microscopy in MCF10A cells transfected with GFP, GFP_Slug wt or Slug phospho-site mutant. Cells were fixed and the localization of GFP_slug was quantified 24 h post transfection (n < 90 cells/transfection). (F) Quantification of Slug localisation in fractionated control or vimentin-silenced MDA-MB-231 cells (nuclear and cytosolic fractions) (n = 3). (G) Slug protein stability was monitored in HEK293 cells transfected with pIRES_slug wt or pIRES_slug S87,104AA plasmids. Cells were treated with cycloheximide (CHX) 10 µg/ml 24 h post transfection, lysed at the indicated time points and subjected to western blot analysis. Quantification of band intensities from six independent experiments analysed by ImageJ is shown (mean ± SEM; * p < 0.05, *** p < 0.001).

Supplementary Materials and Methods

Antibodies, Reagents, siRNAs and Plasmids. The following antibodies were used in this study: rabbit mAb antibodies against Slug (C19G7), p44/42 MAPK (Erk1/2) (137F5) and Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (Cell Signaling Technology) and 12G10 anti-alpha-tubulin (Developmental Studies Hybridoma Bank), mouse monoclonal anti-vimentin, anti-Axl H-124, mouse monoclonal IgG1 AXL (Z49M), mouse monoclonal lamin A/C IgG 2b (Santa Cruz Biotechnology), goat anti-hAXL (R&D Systems), mouse monoclonal anti-beta-actin antibody (Sigma), rat monoclonal anti-keratin-8 antibody (Troma I, XYZ, Developmental Studies, Hybridoma Bank, NIH, USA), anti-integrin β1 MA2252 clone 29 (Millipore) and mAb GST antibody (GenScript). For immunohistochemistry, primary polyclonal rabbit anti-human antibody for Slug (RB 1398) was purchased from Abgent (San Diego, CA, USA) and for vimentin, clones V9 and HPA001762, from Biogenex (Fremont, CA, USA) and Sigma, respectively. Antibodies directed against ERK and pERK were from Cell Signaling.

Growth factors and/inhibitors used in this study included TGF-β1 (R&D systems) at 5 ng/mL for 20 h (or for 5 days in TGFβ induction experiments), EGF (Sigma) at 50 ng/mL for 20 min and MEK inhibitor U0126 (Sigma) at 10 µM for 20 h.

The following siRNAs were used in this study: ERK1 (MAPK3) siRNA (ON-TARGETplus Smart pool Human MAPK3 siRNA LU-003592-00-0002: Dharmacon, Chicago, IL, USA), ERK2 (MAPK1) siRNAs (ON-TARGETplus Smart Pool Human MAPK1 siRNA LU-003555-00-0002, Individual ON-TARGETplus MAPK1 siRNA J-003555-11-0002: Dharmacon), Slug siRNAs (ON-TARGETplus Smart Pool siRNA L-017386-00-0005: Dharmacon, Hs_SNAI2_5 Flexitube siRNA SI03034416: Qiagen), Vimentin siRNAs (Smart Pool M-003551-01-005: Dharmacon, #SI00302190: Qiagen) and scrambled siRNA (AllStars negative control: Qiagen).

The following plasmids were used in this study: pcDNA3 wild type vimentin (full length), pCMV_MEK_ERK2 plasmid, empty pCMV plasmid (used as a negative control), CRU5-IRES-GFP-Slug, p3XFLAG-CMV-Slug and pEGFP-C1-Slug. The CRU5-IRES-GFP retroviral vectors for expressing hSlug, shLuc and shAxl were conducted as described in (1). Point mutations in CRU5-IRES-GFP-Slug, p3XFLAG-CMV-Slug and pEGFP-C1-Slug vectors were performed using Quickchange II XL Site-directed mutagenesis kit according to the manufacturer’s instructions (Agilent Technologies). SNAI1/Snail and SNAI2/Slug coding sequences were cloned into the pGEX-4T-1 vector (Promega) and sequence verified. GST fusion proteins were expressed in Escherichia coli (Rosetta BL21DE3) and purified according to the manufacturer´s instructions (BD Biosciences); the GST moiety was cleaved with thrombin.

Cell Culture, Stable Cell Lines and Transfections. MDA-MB-231, MCF7 and HEK293 cells were cultured in Dulbecco´s modified Eagle´s medium supplemented with 10 % FBS, 1 % L-glutamate and non-essential amino acids (not included for MCF7 cells). MCF10A cells were cultured in DMEM/F12 1:1 supplemented with 5 % horse serum, 20 ng/mL EGF, 0.5 μg/mL hydrocortisone, 100 ng/mL cholera toxin, 10 μg/mL insulin and 100 μg/mL streptomycin. Vimentin WT and -/- mouse embryonic fibroblasts (MEFs) were immortalized by retroviral infection of pBABE-Large T antigen (Biomedicum Genomics, Helsinki, Finland). Cells were split to 50% confluence one day prior to the infections. Medium was changed the next day to include growth medium (Dulbecco´s modified Eagle´s medium supplemented with 10 % FBS and 1 % L-glutamate) and virus supernatant in 1:2 ratio. Cells were incubated with virus-containing medium for 7h prior to media change, split 2 days after the virus infections and selected with 250 μg/ml geneticin. Antibiotic selection was continued for 4 weeks.

siRNA transfections were performed using HiPerFect transfection reagent (Qiagen, Valencia, CA, USA) according to the manufacturer´s protocol. Plasmid transfections were done using Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA) according to the manufacturer´s protocol.

Tissue Preparation and Immunohistochemistry. Paraffin-embedded samples were cut into 4 µm sections, mounted on SuperFrost Plus slides (Menzel-Gläser, Germany), deparaffinized in xylene, rehydrated in graded alcohols, and rinsed in 0.01M phosphate-buffered saline (PBS). Microwave-stimulated antigen retrieval was performed in 10 mM citrate buffer, at pH 6.0. An avidin-biotin-peroxidase method was applied using Dako Envision Kit (Dakopatts, Copenhagen) or HistostainTM Kit (Zymed Laboratories Inc). Endogenous peroxidase was blocked with aqueous 0.3% H2O2 for 15 min. Primary antibodies were diluted in PBS containing 1% bovine serum albumin (BSA) and 0.02 M glycine and PBS was used in all subsequent washing steps. Sections were incubated overnight with primary antibodies directed against slug (dilution 1:100), vimentin (V9 clone, 1:3000), ERK (1:100) and pERK (1:600). Diaminobenzidine (Sigma) was used as the chromogen and the sections were lightly counterstained with Mayer’s hematoxylin. Samples were evaluated for the absence (-) or presence (+) of nuclear or cytoplasmic immunoreactivity. The threshold for determining whether a sample was positive for the expression of the indicated proteins was set at 20 % (i.e. percentage of cells exhibiting positive IHC staining). Samples under this threshold were considered negative.

Immunohistochemical staining of frozen sections was carried out on Tissue-Tek® O.C.T.™ Compound (Sakura)-embedded sections. Following acetone fixation (10 min at -20oC) samples were permeabilized in 0.3% Triton-X/2% BSA, blocked in 30 % horse serum-PBS (30 min) and stained with anti-vimentin (1:400; HPA001762 clone) for 1 h at RT followed by PBS washes and incubation with Alexa Fluor anti-rabbit-555 (1:400; Life technologies) and Dapi (0.1 ug/ml).

Tumor xenografts on chick embryo chorioallantoic membranes. Fertilized chicken eggs were incubated like previously described in Hagedorn et al.2002. Shortly, to start the development the eggs were washed and placed at 37 C incubator. After that on day 3, a small hole was made in the eggshell in order to drop the chorioallantoic membrane (CAM). On the developmental day 10, plastic ring was placed on CAM and 1 million control or vimentin silenced cells were implanted inside the ring in 40 ul of 50 % matrigel. After 3 days tumors were imaged and dissected. Tumors were lysed in lysis buffer (50 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5% TX-100, 5% glycerol, 1% SDS, 1 mM Na3VO4, 10 mM NaF, and 1 mM phenylmethanesulfonyl fluoride) and homogenasized with MagNA lyser green beads in MagNA Lyser (Roche) (2).

Luciferase reporter assays and DNA binding ELISA. Dual-luciferase reporter assays were performed according to manufacturer´s instructions (Promega, Madison, USA). Briefly, cells were triple-transfected with pRL-TK, pE-cad-571-Luc (3) or pVIM-571-Luc (4, 5) and CRU5-IRES-GFP/-Slug wt/-Slug mutant plasmids. Cells were lysed with Passive Lysis Buffer 48 hours post transfection and plated on Costar 96-well plates (Corning, NY, USA). Luciferase Assay Substrate was added to each well and luminescence measured with Envision multilabel plate reader (Perkin Elmer-Cetus, Waltham, MA, USA). Stop & Glow reagent was added and the luminescence signal was re-measured. The luciferase signals from E-cadherin and vimentin promoters were normalized for the renilla luciferase signal (pRL-TK).

In DNA binding ELISA assays, 96-well plates (Corning, NY, USA) were incubated with GST- Slug/SNAI2 WT or Slug S87A/S104A mutant for 1 hour at RT. Plates were blocked with blocking buffer (5% BSA in TBS-T, 10 µM ZnCl2 and 1mM DTT) for 1 hour at RT and incubated with DNA (300 ng/ml) in blocking buffer for 3 hours at RT. Plates were washed twice with blocking buffer and incubated with SYBR® Gold (Molecular Probes) in 50 mM Tris pH 7.5. DNA was detected by measuring sample emission at 520 nm using Envision 2100 multilabel reader (Perkin Elmer).

qRT-PCR. Cellular RNA from transfected cells was isolated with Qiagen RNeasy kit and 20 ng of the extracted RNA was used as a template for the reverse transcriptase reaction. 1/10 of synthesized cDNA was then used for qRT-PCR. Primers for indicated proteins (designed using Roche´s Universal ProbeLibrary) were used at 300 nM final concentration and probes at 100 nM. The expression of indicated proteins was detected by the quantitation method using GAPDH as an internal control.

Immunofluorescence, Fluorescence recovery after photobleaching and Flow Cytometry. For immunofluorescence, cells were washed with cold PBS, fixed in cold 4% paraformaldehyde (PFAH), permeabilized with 0.1% Triton X-100 in 2% BSA/PBS for 15 min at RT and blocked with 30 % horse serum in PBS for 1 hour at RT. The indicated primary antibodies diluted in blocking buffer were added and incubated for 1 h at RT. After washing three times with PBS, Alexa-conjugated secondary antibodies and DAPI (4´6-diamidino-2-phenylindole) were added for 1 hour at RT. Coverslips were washed with PBS and MQH2O and mounted with Vectashield mounting medium (Vector Labs). Confocal 3D images were taken with Zeiss Axiovert 200 M with spinning disc confocal unit Yokogawa CSU22 and Zeiss Plan-Neofluar 63×Oil/1.4 NA objective and analyzed with NIH ImageJ.

For 3D immunofluorescence, MDA-MB-231 cells were embedded into 50% Matrigel and allowed to form cell spheroids for 5 days. Cell invasion was then induced with 50 ng/ml EGF. 24 h later, the cells were fixed and permeabilized simultaneously with 2% PFA in PBS supplemented with 0.5% Triton X-100 for 1.5 h at RT. After 3 consecutive washing steps (10-15 min per wash at RT) with glycine buffer (130 mM NaCl, 7 mM Na2HPO4, 3.5 mM NaH2PO4, and 100 mM glycine in PBS), blocking was done for 2 h at RT with buffer containing 130 mM NaCl, 7 mM Na2HPO4, 3.5 mM NaH2PO4, 7.7 mM NaN3, 0.1% BSA, 0.2% Triton-X100, 0.05% Tween20, and 10% horse serum in PBS. Primary antibodies were used at 5–10 mg/ml concentrations in blocking buffer and incubated overnight at 4 °C. The cells were then washed three times (for 20 min each step) with blocking buffer at RT with gentle rocking. Alexa-conjugated secondary antibodies, at a concentration of 5 μg/ml, were incubated at RT for 1 h, followed by another round of washing. The spheroids were mounted with Mowiol containing anti-fading reagent (Vectashield, Vector Labs) for 1 h at 37°C. Immunofluorescent samples were analysed with an inverted wide-field microscope (Carl Zeiss) with a confocal unit, Orca-ER camera (Hamamatsu Photonics), Plan-Neofluar 63× oil/1.4 NA objective (Carl Zeiss), and SlideBook 5.0 imaging software (Intelligent Imaging Innovations, Inc.).