Human prostate adenocarcinoma PC-3 cells, human prostate carcinoma LnCaP and DU145 cells, and human normal prostate stroma/fibroblast WPMY-1 cells were obtained from the American Type Culture Collection (ATCC) and cultured as previously described22. The human melanoma cell line SKMEL5 (HTB-70) was obtained from the ATCC via LGC Standards Ltd and cultured according to ATCC protocols. The promyelocytic leukemia derived cell line HL60 was also obtained from the ATCC (via LGC Standards Ltd) and was cultured as previously described23. Growth conditions for all the cell lines used are presented in Supplementary Table?1.
Plasmid transfections for forced expression of EN2 tagged with GFP were carried out following the manufacturer’s instructions using Lipofectamine? 2000 (Life Technologies, UK) for DU145 
cells (6?μl of reagent per 1?μg of DNA) or ViaFect? (Promega, UK) for the other cell lines, using the following reagent(μl):DNA(μg) ratios: PC3 (3:1), LnCaP (5:1), WPMY-1 (6:1), SKMEL-5 (5:1). During transfection, cells were starved by reducing fetal bovine serum levels in fresh media from 10% to 0.5%. Cell lines were transfected with DNA plasmid created using Myc-DDK tagged ORF clone of Homo sapiens EN2 as transfection ready DNA (accession number: NM_001427) and PrecisionShuttle mammalian vector with N-terminal mGFP (OriGene Technologies, USA), which enabled the expression of N-terminal GFP-tagged EN2. As a control, cell lines expressing GFP only were created in the same way using the PrecisionShuttle mammalian vector with N-terminal mGFP (OriGene Technologies, USA) alone.
siRNA transfections were carried out using siPORTTM NeoFXTM transfection reagent (Invitrogen, en2 UK) following the manufacturer’s protocol. Briefly, EN2 siRNA (4676; mRNA sequence accession number NM_001427.3, targeted exon 2, siRNA location 1110 bp) or negative control siRNA diluted in Optimem and siPORTTM NeoFXTM transfection reagent diluted in Optimem were mixed at 1:1 ratio, added to PC3 cells and EN2 protein detected 48 h after transfection using immunocytochemistry.
Total RNA was isolated using the RNeasy Plus Micro Kit (Qiagen, UK) followed by a two-step cDNA synthesis reaction using nanoScript Reverse Transcription kit (Primer design Ltd, UK) with annealing step preceding the extension step, all according to the manufacturer’s protocol. The mRNA expression of EN2, GAPDH and MX2 was quantified using the Stratagene Mx3005P qPCR system (Agilent Technologies, USA) with SYBR Green fluorescence technology with the following primer sequences: EN2 NM_001427 F: GAACCCGAACAAAGAGGACA and R: CGCTTGTTCTGGAACCAAAT, GAPDH NM_002046.5, F: GAACCCGAACAAAGAGGACA and R: CGCTTGTTCTGGAACCAAAT, and MX2 NM_002463: F: AAGCAGTATCGAGGCAAGGA and R: TCGTGCTCTGAACAGTTTGG. EN2 and MX2 expressions were normalised to GAPDH using the ΔCt relative quantification method.
Immunocytochemistry was used to demonstrate the? presence of EN2 in cells. Cell lines were grown overnight in an 8-chambered polystyrene culture treated glass slides (BD Biosciences, UK), fixed with 4% paraformaldehyde (Sigma, UK) and washed with PBS three times. Next, the cells were incubated with 10% horse serum (Jackson ImmunoResearch, USA) for 20?minutes and further with anti-EN2 goat polyclonal IgG primary antibody (Abcam, UK) diluted 1:100 in PBS/1% BSA at room temperature for 2?hours or overnight at 4?°C. A ‘no primary antibody’ sample was included containing PBS/1% BSA alone for use as a negative control. After primary antibody incubation, the cells were washed three times with PBS and incubated with donkey anti-goat secondary antibody (Abcam, UK) at 1:200 in the dark for 45?minutes. After incubation, cells were washed three times with PBS, chambers removed and slides mounted using propidium iodide (PI) Vectashield? mounting medium (Vector Laboratories, USA). The slides were imaged at X40 magnification using the Nikon A1M confocal microscope and NIS elements acquisition software (Nikon, en2 UK). Images were processed with ImageJ.
A panel of 19 polyclonal sheep antibodies, Ab2 - Ab33 (kindly provided by Bioventix, UK, as part of a collaborative project) (Supplementary Table?2), were made against peptides covering the whole length of the EN2 protein. Each peptide was 20 amino acids long and overlapped its neighboring peptides by 10 amino acids. These antibodies were generated in sheep using a nested peptide series conjugated to the metalloprotein, KLH. Immunocytochemical staining was performed as described above, except 5% horse serum was used for blocking and donkey anti-sheep secondary antibody (Abcam, UK) at 1:10,000 dilution was used for visualisation.
PC3 cells were seeded in a 4-chamber glass-bottom dish (MatTek, USA) and 24?hours later transfected with 0.5?μg of either GFP-EN2 or GFP (control) plasmids as described above. After further 24?h incubation, the media was changed to Gibco? FluoroBrite? DMEM (Life Technologies, UK) in order to enhance the fluorescent signal. GFP fluorescence was detected using confocal microscopy. Time-lapse mode was selected and the areas to be imaged were chosen by focusing on the cells as well as labelling and setting the X/Y parameters. Images were set to be taken every 5?minutes over 24?hours. Fold difference in fluorescence from T0 was calculated and plotted over time.
To investigate the intercellular transfer of EN2, stable PC3 LifeAct? (Ibidi, Germany) expressing clones were created as described above. F-actin was visualized in the cells to reveal the cytoskeleton without compromising cellular processes. These cells were then directly co-cultured with PC3 EN2GFP (green fluorescence) or PC3 GFP cells in 6-well plates at half the normal density in DMEM culture media. Images were taken after 72?h using a fluorescence microscope at X20 or X40 magnification. The assay was repeated and images taken after 96?h with addition of NucBlue? Live ReadyProbes? Reagent to demarcate the nucleus.
MX2 expression was investigated in prostate cancer and normal prostate using a 3 μm-thick, formalin fixed, paraffin embedded tissue array (PR2085a, US Biomax, Rockville, MD, USA), with a rabbit polyclonal MX2 antibody (Abcam 224479) diluted 1:100 and the ABC detection method with peroxidase block (DakoCytomation). Antigen retrieval was performed using pH 6.0 citrate/EDTA buffer (DakoCytomation) and heating in a microwave for 23?minutes. DAB Peroxidase (HRP) Substrate Kit (Vector Laboratories) was used in the final detection step and images taken under a light microscope at?×?40 magnification.
Statistical analyses were performed using GraphPad-PRISM software (CA, USA) based on a minimum of two independent experiments using Student’s t-test for comparison between two groups. Significance was scored as: ****P?<?0.0001; ***P?<?0.001; **P?<?0.01; en2 *P?<?0.05.
