Tissue samples of pancreatic ductal adenocarcinomas were collected from patients undergoing a curatively aimed pancreatico-duodenectomy at the Department of Surgery at Umeå University Hospital (n = 9). Normal pancreatic tissue samples were collected from patients who underwent surgery in organs close to the pancreas, but where the disease did not involve the pancreas (n = 7). Samples were snap frozen in liquid nitrogen and stored at -80°C until analysis. This study is approved by the Research Ethics Review Board (EPN) of Northern Sweden.
Cell lines and cell culturing
Two well-characterized human pancreatic adenocarcinoma cell lines were used. HPAC (ATCC, CRL-2119), a human pancreatic adenocarcinoma cell line of ductal origin, is derived from a moderate to well differentiated pancreatic adenocarcinoma . CFPAC-1 (ATCC, CRL- 1918) was derived from a patient with a ductal adenocarcinoma . HPAC was cultured with 1:1 mixture of Dulbecco’s modified Eagle’s and Ham’s F12 medium (DMEM/F12), CFPAC-1 with 90% Iscove’s modified Dulbecco’s medium (IMDM), both supplemented with 10% fetal calf serum (FCS), penicillin (105 IU/L), and streptomycin (100 mg/L) and grown in an incubator at 37°C in an atmosphere with 5% CO2.
Cell and tissue staining
Immunofluorescence stainings were performed on 6 μm frozen sections, and on cells cultured on Falcon™ Culture Slides (BD Biosciences, Erembodegem, Belgium) using methods previously described , with the following primary antibodies and dilutions: mouse anti-human integrin α1 (MAB1973, 1:40); mouse anti-human integrin α2 (MAB1950, 1:40); mouse anti-human integrin β1 (MAB1959, 1:500), all from Millipore (Billerica, MA, USA); and rabbit anti-human type IV collagen (MP Biomedicals, LLC, Solon, OH, USA; cat. no. 10760, 1:100); goat anti-human type XVIII collagen (R&D Systems, Minneapolis, MN, USA; AF570, 1:75), goat anti-human perlecan (R&D Systems, Minneapolis, MN, USA; AF2364, 1:75), rat anti-human laminin-γ1 (Millipore, Billerica, MA, USA, Mab1914P, 1:75), rabbit anti-human nidogen (Millipore, Billerica, MA, USA, cat. no. 481978, 1:75), mouse anti-human cytokeratin 18 (DakoCytomation, Glostrup, Denmark, DC10 1:100) and sheep anti-CD31 (R&D Systems, Inc., Minneapolis, MN, USA; AF806, 1:50). Cultured cells were stained with the integrin antibodies mentioned above together with the goat anti-type IV collagen antibody (Chemicon, Billerica, MA, USA; AB769, 1:50). Double staining on tissue for type I and type IV collagen was performed with a rabbit anti-human type I collagen antibody (Cedarlane Laboratories, Burlington, NC, USA; CL50111AP, 1:200) and a mouse anti-α1(IV)NC1 (Wieslab, Malmö, Sweden; MAB1, 1:75) . Secondary antibodies used were: donkey anti-rabbit FITC, donkey anti-mouse TRITC and FITC, donkey anti-goat TRITC (all Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA; with dilutions 1:100). Sections were mounted with medium containing DAPI (Vectashield, Vector Laboratories, Inc., Burlingame, CA, USA). Negative control sections were incubated with secondary antibodies only. Hematoxylin and eosin (H&E) stainings were performed according to standard protocols.
Migration, proliferation, and apoptosis assays
Cell proliferation was measured using CellTiter-Glo® (Promega Corporation, Madison, WI, USA), a luminescence based cell viability assay, and cell apoptosis was assayed with the M30-Apoptosense® ELISA (Peviva AB, Bromma, Sweden). In the apoptosis assay the apoptosis-associated caspase-cleaved cytokeratin 18 is quantitatively determined, and high concentrations of the M30 neo-epitope indicate high apoptotic activity. Both assays were performed according to the manufacturers’ instructions. For both assays, 5–6 × 103 cells were seeded in triplicates on 96-well polystyrene plates, culture area 0.34 cm2/well (Nunc, Roskilde, Denmark), with coated or uncoated surfaces. Proliferation and apoptosis was measured after 2–3 days incubation, and the cells were ensured to be in growth phase when assayed (data not shown). For the apoptosis assay, the triplicates were pooled together and run in duplicates. Furthermore, the fraction of cells in S-phase was measured with flow cytometry, and cells were stained and analyzed according to standard techniques previously described . Transfected cells and control cells were solved in PBS and run in triplicates. DNA histograms were evaluated for ploidy (ModFit LT 3.0™, Verity Software House, Topsham, ME, USA) in order to identify the fraction of cells in the S-phase.
For the matrix studies, type IV and type I collagen (BD Biosciences, Two Oak Park, Bedford, MA, USA) and Bovine Serum Albumin (BSA, used as unspecific control protein) were coated to the wells. For the blocking studies, wells were coated with antibodies directed against the NC1, CB3, and 7S domains of the type IV collagen. These antibodies were: mouse-anti-α1(IV)NC1 (Wieslab, Malmö, Sweden, MAB1); mouse-anti-α2(IV)7S (Chemicon International, Inc., Billerica, MA, USA; MAB1910); mouse anti-human collagen type IV antibody (Acris Antibodies GmbH, Herford, Germany; clone CIV22) ; and whole mouse IgG (used as control, Jackson ImmunoResearch Laboratories, Inc.). The coated proteins were diluted in 2 mM hydrochloric acid or 10 mM acetic acid, followed by incubation for 2 hours in the wells. The wells were then washed, cells added, and grown under FCS free conditions.
For the integrin blocking studies, cells were grown in FCS supplemented medium for one day, washed and thereafter, the blocking antibody diluted in serum free medium was added. Antibodies known to functionally block the integrin receptors and a control IgG antibody were used (all mentioned above).
Migration was studied in a wound-healing assay, in which cells were grown on a 24-well wound healing assay plate (Cell Biolabs, Inc., San Diego, CA, USA), and standardized wounds (0.9 mm) were generated according to the manufacturer’s instructions. Two or more wells were used for each experiment, and pictures were taken every 10 minutes (Olympus IX81 with Cell-R software) with time-lapse Differential Interference Contrast microscopy (DIC), and the time of wound closure was determined at two different locations for each wound. Cells were grown under FCS supplemented conditions.
CFPAC-1 cells were transfected with a pool of 3 different target-specific siRNAs designed to knock down the gene expression of the COL4A1 gene, and control the siRNA design to not target any known human gene (Santa Cruz Biotechnology, Inc., CA, USA, sc-43064 and sc-37007). Subsequently, 5 x 104 cells/ml were seeded in 6 well plates (Nunc, Roskilde, Denmark), and transfected according to the manufacturer’s protocol. The down-regulation of type IV collagen synthesis was verified by quantitative real time-PCR. Approximately 2x105 cells from each cell culture were harvested by centrifugation, and the total RNA was extracted using the RNeasy Mini Kit (Qiagen GmbH, Hilden, Germany). Equal amounts (1 μg) of RNA were reverse transcribed into cDNA using the QuantiTect Reverse Transcription Kit (Qiagen). Gene expression was quantified using the Bio-Rad iQ SYBR Green Supermix and the iQ5 iCycler according to the manufacturer’s instructions. The primer set 5′-TGCACCACCAACTGCTTAGC-3′ and 5′-GGCATGGACTGTGGTCATGAG-3′ for the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) housekeeping gene was used for normalization, and the primer set 5′-CTCTACGTGCAAGGCAATGA-3′ and 5′-AGAACAGGAAGGGCATTGTG-3′, designed using the web-based PCR primer design tool Primer3 (Rozen and Skaletsky, 2000) was used for quantification of the COL4A1 gene. All reactions were run at 95°C for 3 minutes, followed by 40 cycles at 95°C for 10 seconds and at 60°C for 30 seconds.
Additionally, the type IV collagen protein expression was detected in the transfected cells using immunofluorescence staining with the MAB1 antibody. The transfected cells were assayed for proliferation, apoptosis, and migration as described above. For the cell proliferation assay, the experiments were repeated five times with similar results both under FCS free and serum supplemented conditions.
All assays performed on 96-well plates were run in triplets. For the M30-Apoptosense® ELISA, the triplicates were subsequently pooled and the final analysis was performed on duplicates of the pooled samples. Normal distribution was assumed. Unpaired t-test was used when comparing two groups and ANOVA with Bonferroni post-hoc test if more than two groups were compared. P < 0.05 was considered significant. Bars in the figures illustrate the standard deviation.