Female C57BL/6 mice were purchased from the Mossakowski Medical Research Centre, Polish Academy of Sciences, Warsaw, and housed with ample access to food and water. They were used at 7–8 weeks of age. All experimental procedures were approved by the Animal Experiments I Local Ethics Committee, Kraków (Approval No 11/2007) and all efforts were made to minimize animal suffering.
Murine colon adenocarcinoma MC38 
, stably expressing human carcinoembrional antigen (MC38CEA) was a gift from Dr. Michał Bereta (Jagiellonian University, Kraków) 
. The cells were cultured in DMEM (Lonza) supplemented with 10% heat-inactivated fetal calf serum (FCS) (Lonza) at standard conditions. Every few passages the culture medium was enriched in geneticin (1.5 mg/ml) and for mock-transfected and ADAM17-silenced MC38CEA (see below) additionally in hygromycin (400 µg/ml). The cell cultures were routinely tested by PCR for mycoplasma contamination using mycoplasma-specific primers.
Generation of MC38CEA cell lines with stably silenced ADAM17
MC38CEA cells were transfected with pGeneClip™Vector (Promega) coding for ADAM17 shRNA or for non-interfering control shRNA using lipofectamine-2000 (Invitrogen Life Technologies) according to the manufacturer's protocol. Stably transfected cells were obtained by hygromycin selection. On the basis of our preliminary results three cell lines with the highest level of ADAM17-silencing (S2, S3, S35) and three control cell lines (mock-transfected; M1, M2, M3) were chosen for the experiments. The coding sequences of ADAM17 shRNA and control shRNA were AACGAATGCTGGTGTATAAGT and GCTCAGATATCCAGTCATGTT, respectively.
TNF shedding assay
MC38CEA cells (5×105) were incubated for 30 min in 100 µl of DMEM with PMA (100 ng/ml) or with corresponding volume of DMSO (solvent). The level of TNF released to the medium was evaluated using ELISA (R&D Systems).
Neuregulin-1 shedding assay
MC38CEA cells were cultured for 24 h in 12-well plates (2.5×105 cells/well) in DMEM containing 5% FCS and for the next 24 h in 0.5 ml of DMEM/F12 enriched in BSA (0.5 mg/ml), human holotransferrin (5 µg/ml) and sodium selenite (2 ng/ml) (complete DMEM/F12). Concentration of NRG-1 released to the medium was evaluated by ELISA (Uscn Life Science Inc.) according to the manufacturer's instruction.
Analysis of VEGF-A secretion
MC38CEA cells were plated in 96-well plates (5×103 cells/well). Next day medium was changed for DMEM containing 2% FCS and the cells were cultured for 48 h. Concentration of VEGF-A in the culture media was measured using mouse VEGF ELISA Duoset (R&D Systems) according to the manufacturer's protocol.
Proliferation/viability assay (MTT test)
MC38CEA cells were plated in 96-well plates (5×103
cells/well) and cultured for 5 days in DMEM containing 0.5% FCS or in DMEM without serum. Every day cell metabolic activity being an indicator of cell number and viability was measured using MTT assay 
. The absorbance of solubilized formazan was measured at 545 nm.
MC38CEA cells were cultured for 24 h in 12-well plates (2.5×105 cells/well) in DMEM containing 5% FCS and for the next 24 h in fresh, serum-free DMEM or in DMEM containing 5% FCS. For the last 6 h of incubation, 1 µCi of 3H-thymidine was added to each well. After intensive washing with PBS, the cells were permeabilized with ice-cold methanol, washed again with PBS, and the radioactivity incorporated into DNA was determined by liquid scintillation counting in a Wallac β-counter (Perkin Elmer).
Western blotting analysis of ErbB2 phosphorylation
MC38CEA cells (2.5×105
cells/well) were incubated in complete DMEM/F12. The cells were lysed on ice in buffer containing 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 0.1% SDS, 1% NP-40 and 1% CHAPS, Complete Protease Inhibitor Cocktail (Roche Applied Science) and additionally 1 mM PMSF, 5 mM NaF and 2 mM Na3
. Western blotting analysis was performed according to the standard protocol 
. Membranes were probed with rabbit polyclonal anti-ErbB2 (phospho Y1248) antibody (Abcam) at 1
1000 dilution and mouse monoclonal anti-GAPDH antibody clone 6C5 (Biodesign Int.) (0.5 µg/ml) and HRP-conjugated appropriate secondary antibodies. Bands were visualized using Immobilon Western Chemiluminescent HRP Substrate (Millipore). Quantification: Serial dilutions of a lysate obtained from the M1 cells incubated for 10 min with rmNRG-1 were subjected to Western blotting with anti-phosphorylated ErbB2 in order to find a slope and a range of linearity of the standard curve. The curve slope allowed to convert relative autoradiographic signals to relative levels of phosphorylated ErbB2 in the parallel experiments. For each sample, the autoradiographic signal of phosphorylated ErbB2 was normalized to the GAPDH signal. The values obtained for mock-transfected M1 cell line were taken as 100% in all performed experiments.
Caspase activity assay
MC38CEA cells were cultured for 24 h in white-walled 96-well plates (5×103 cells/well) in DMEM containing 5% FCS. Then the medium was changed for DMEM containing 2% FCS and the cells were incubated for 6 h with 120 µl of medium alone, or with recombinant human TNF (10 ng/ml), or with recombinant mouse IFNγ (10 ng/ml) or with both cytokines. The activity of caspases 3 and 7 in 30 µl samples of the media was evaluated using Caspase-Glo 3/7 Assay (Promega) according to the manufacturer's instruction. Chemiluminescence was measured using Infinite 200 PRO chemiluminometer (Tecan).
Analysis of various mRNA levels using quantitative reverse-transcription PCR (qRT-PCR)
RNA was isolated from cells by guanidinium thiocyanate-phenol-chloroform extraction and 2 µg of each sample was used to synthesize cDNA using the reverse transcription kit (Promega). For obtaining ErbB4 cDNA the specific primer (5′-CTTTTTGATGCTCTTTCTTCTGAC-3′) apart from oligo(dT) was used. Samples of cDNA (~5 ng) were amplified using one step real time PCR kit (Kapa Biosystems) and the RotorGene-3000 thermocycler (Corbett Research, Australia). Following primers (final concentration 200 nM) were used:
ADAM17 – F: CCAGGAGCGCAGCAACAAGGT, R: TCCTATCACTGCACTGCACACCCG; TNF – F: CATCTTCTCAAAATTCGAGTG, R: TGGGAGTAGACAAGGTACAA;
IFNγ – F: GCTTTGCAGCTCTTCCTCAT, R:GTCACCATCCTTTTGCCAGT;
MCP-1 – F: AGCACCAGCCAACTCTCACT, R: GCTGCTGGTGATCCTCTTGT;
VEGF-A – F: ATGCGGATCAAACCTCACCAAGGC, R: TTAACTCAAGCTGCCTCGCCTTGC;
TGFα – F: TGATCCACTGCTGTCAGCTC, R:CTTGGTTGGGCTGTCATCGG;
EGFR – F: ACACTGCTGGTGTTGCTGAC, R:CCCAAGGACCACTTCACAGT;
ErbB2 – F: AGGGGCTGGCTCCGATGTGTTT, R: GGCTGGTTCACATACTCGGGCT;
ErbB3 – F: CCACGTGGGAGCCAGAGTCTTT, R: TTCAGCCTCAGAGCCCGTCACA;
ErbB4 – F: GAAATGTCCAGATGGCCTACAGGG, R: CTTTTTGATGCTCTTTCTTCTGAC;
EF2 – F: GACATCACCAAGGGTGTGCAG, R: GCGGTCAGCACAATGGCATA.
The levels of various mRNA in individual cell lines (mock-transfected and ADAM17-silenced were compared to that in wild type MC38CEA using the “delta delta Ct” relative quantitation method (Applied Biosystems). EF2 cDNA was used as reference.
Time-lapse monitoring of individual cell movement
MC38CEA cells (1.5×105
cells) were plated in a T25 culture flask in DMEM containing 10% FCS. After 24 h cell movement was recorded for 8 h at 5 min intervals using Leica DM IRE 2 microscope equipped with FW4000 software. The trajectories of 50 individual, randomly chosen cells were analyzed as previously described 
in order to obtain: the total length of cell trajectory (TLT), the velocity of cell movement defined as a total length of cell trajectory/time of recording, the total length of displacement (TLD) and the coefficient of movement efficiency (CME) defined as the ratio of total length of cell displacement to total length of cell trajectory (TLD/TLT).
Wound healing assay
MC38CEA cells (2×105) were grown in a 6-well plate in DMEM containing 10% FCS until they reached monolayer. The medium was aspirated, a scratch wound was made across each well using a tip, monolayers were washed to remove detached cells and the cells were incubated in fresh, serum-free DMEM for 24 h. The pictures were taken at time 0 and 24 h after wounding, and the widths of the wounds were measured. In some experiments the cells were starved for 3 h before wounding and then wounded monolayers were incubated in serum-free DMEM alone or in the medium enriched in rmNRG-1. The generation and testing of rmNRG-1 is described in Supporting Information. The lengths of cell displacement are expressed as the difference between average pre- and post-healing wound widths. The average width was calculated on the basis of 5 measurements at random positions along each wound.
In vivo tumor growth
MC38CEA cells (ADAM17-silenced, mock-transfected and wild type) were trypsinized and washed twice with PBS. 5×105 cells in 100 µl of PBS were injected subcutaneously into one flank of each mouse. Tumor sizes were evaluated by caliper every 2–3 days and the tumor area was determined by multiplying measurements of two perpendicular diameters.
Cytometric bead array
The levels of cytokines [interleukin-6 (IL-6), IL-10, monocyte chemoattractant protein 1 (MCP-1), IFNγ, TNF, and IL-12p70] were measured in serum and tumor lysates using the BD Cytometric Bead Array (CBA) Mouse Inflammation Kit (Becton Dickinson), according to the manufacturer's instruction using 50 µl of serum or 100 µg of total lysate protein obtained using RIPA buffer (25 mM Tris-HCl, pH 7.6, 150 mM NaCl, 1% NP-40, Complete Protease Inhibitor Cocktail). The fluorescence of samples was acquired on the LSRII flow cytometer (Becton Dickinson) and analyzed using Becton Dickinson FCAP Array™ Software.
Immunohistochemical staining of blood- and lymphatic vessels
Immunochemical staining of frozen tissue sections was performed according to the standard protocol 
. Briefly, 10 µm-thick cryosections were air-dried and fixed with a zinc-based fixative for 24 h, blocked with 10% donkey- and 10% goat serum in PBS and stained with (i
) hamster anti-CD31/Alexa fluor 546-conjugated anti-hamster IgG; (ii
) rabbit anti-lymphatic endothelium specific antigen-1 (LYVE-1)/Alexa Fluor 488-conjugated anti-rabbit IgG; (iii
) rat anti-CD11b/Alexa fluor 594-conjugated anti-rat IgG; (iv)
APC-conjugated rat anti-CD45. In negative controls primary antibodies were omitted. Cell nuclei were counterstained with DAPI. Sections were imaged on a Zeiss LSM 510 Meta confocal microscope at magnification 400×. Five randomly chosen fields of peritumoral area and five randomly chosen fields of tumor peripheral area were analyzed in each section. The area covered by CD31+
blood vessels in tumor periphery were calculated using MetaMorph imaging software and expressed as the percentage of the total area.
Zymography was performed according to a standard protocol. Cells were cultured for 24 h in 6-well plates (5×105 cells/well) in DMEM containing 5% FCS and for the next 24 h in fresh, serum-free DMEM. The samples of cell media (20 µl) were subjected to SDS-PAGE in 10% gels containing 0.1% gelatin. The gels were stained with Coomassie Brilliant Blue, resulting in a blue background of nondegraded gelatin with cleared bands of proteolytic activity.
Statistical analysis was performed using the Student's t-test, with P<0.05 being considered significant.