Plasmids

The plasmids for anticancer genes were obtained from the Japanese National Institute of Technology and Evaluation (NITE) human cDNA library [12]. Cloning of positive and negative controls in pME18SFL3, including RIPK1, CAS2, CAS8, and GFP has been described [13, 14]. tBID cloned in pcDNA3.1 was a kind gift from Mund’s laboratory [15]. Non-Target pLKO.1 (scrambled) shRNA and pLKO.1 harbouring shRNA against MYC were purchased from Sigma-Aldrich, Merck KGaA, Darmstadt, Germany. For low throughput methods, plasmid DNA was purified using Invitrogen’s PureLink plasmid purification kits according to manufacturer’s protocols (Invitrogen, Thermo Fisher Scientific Inc., Waltham, MA USA). For round-1 to round-3 of genetic screening (Fig. 2a), ultra-pure silica oxide large scale plasmid DNA isolation was used [16]. For clonogenic assays, FBLN5 was excised from the pME18SFL3 vector after EcoRI and XbaI (FastDigest—Thermo Scientific, Thermo Fisher Scientific Inc.) digestion and cloned into pcDNA 3.1 using standard molecular biology procedures.

Fig. 2

Sixteen novel anticancer genes cause cell death in transformed cells but not in normal cells. a Diagram depicting the process undertaken for the isolation of 16 novel anticancer genes. Screening was performed in two steps: in the first step a subset of 377 cell death inducing genes (in HEK-293 T cells) was transfected into non-transformed CV-1 cells in 4 consecutive rounds of screening where genes causing cell death above the internal threshold set for each round were gradually eliminated, resulting in a set of 78 gene candidates. DNA was isolated using silica oxide purification. In the second step, the plasmids were purified using standard DNA-miniPrep and another 35 genes were eliminated in CV-1 cells. The resulting 43 genes were transfected into HEK293T cells and 16 genes were selected for their ability to cause cell death in HEK-293T cells but not in CV-1 cells. b, c Relative cell death (CPRG ratio) in CV-1 (b) and HEK-293T (c) cells upon experimental overexpression of 16 anticancer genes. d, e Cell death in cervical cancer HeLa (d) and breast cancer MCF-7 cells (e), upon experimental overexpression of 16 anticancer genes. Cells were stained with Dioc6 (early apoptosis) and PI (late apoptosis) and subjected to flow cytometry for PI positive and/or Dioc6 negative cells. In all cases, luciferase (luc) was used for negative control and CAS2, CAS8 and RIP were used as positive controls. Histograms represent the average ± standard deviation (SD) of 3 independent transfections. Statistical significance was calculated using two-tailed Student’s t-test (*p ≤ 0.05, **p ≤ 0.005). pt: processed transcript, pr: partial overlap, nov: novel transcript

Cell culture

CV-1 [6], HeLa [6], HEK293T [6], MCF-7 [17] and COS-7 cells (a gift from T. Malik, Imperial College London) were cultured in the Dulbecco's Modified Eagle's Medium (DMEM) (with 4500 mg/L glucose, L-glutamine, sodium pyruvate, and sodium bicarbonate, Sigma-Aldrich) supplemented with 10% (v/v) heat inactivated foetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific Inc.).

Cell transfections

For high-throughput transfections, CV-1 cells were transfected using Xfect transfection reagent (Clontech, Takara Bio Inc. Shiga, Japan) whilst HEK293T cells were transfected using jetPEI transfection reagent (Polyplus-transfection, Strasbourg, France) following instructions provided by the manufacturers. For the low-throughput applications, HeLa cells were transfected using Effectene transfection reagent (Qiagen, Hilden, Germany) whilst MCF-7, CV-1 and COS-7 cells were transfected using Xfect transfection reagent by following the manufacturer’s instructions.

Stably transfected MYC-overexpressing CV1 cells were generated by transfecting cMYC in pcDNA3.0 using Xfect transfection reagent as above and cells selected initially with 2.5 mg/ml G418 (Sigma-Aldrich) and maintained thereafter with 1.5 mg/mL G418.

Stable downregulation of MYC was obtained by shRNA (in pLKO.1; Sigma-Aldrich) in MCF-7 cells using the Xfect transfection reagent as above. Transfected cells were initially selected with DMEM + 10% FBS containing 1.0 μg/mL puromycin dihydrochloride (Sigma-Aldrich) and thereafter maintained with 0.4 μg/mL puromycin.

Cell death measurementsCPRG assay

For HEK293T cells, it was performed essentially as described [13, 18]. For CV-1 cells, chlorophenol red-β-D-galactopyranoside (CPRG) was added 48 h after transfection and OD at 590 nm was measured after 8 h (CPRG1). The lysis buffer was added 49 h after transfection and OD at 590 nm was measured the following day (CPRG2). CPRG ratio was calculated by dividing 1st OD reading by 2nd OD, i.e., CPRG ratio = CPRG1 / CPRG2.

DiOC6/propidium iodide staining

Cell death in HeLa and MCF-7 cells was quantified using DiOC6/propidium iodide (PI) double staining and flow cytometry [6]. Briefly, 48 h post-transfection, floating and adherent cells were harvested, centrifuged and re-suspended in 150 μL PBS (Dulbecco’s Phosphate Buffered Saline, no calcium, no magnesium, Sigma-Aldrich) containing 40 nM 3,3-dihexaoxacarbocyanine iodide (DiOC6) (Life Technologies, Thermo Fisher Scientific Inc.) and 6 μg/mL propidium iodide (PI) (Sigma-Aldrich), incubated at 37 °C for 30 min and further incubated for 30 min at room temperature. Populations of PI-positive and/or DiOC6-negative cells were normalised for transfection efficiency using in parallel GFP transfections.

Propidium iodide staining

Cell death in CV-1 and CV-1 MYC cells was measured by PI staining. Floating and adherent cells were stained with 20 μg/mL propidium iodide (Sigma-Aldrich) and cell death was calculated as above.

Clonogenic cell death assay

For CV-1 cells, 70,000 cells/well were seeded in 6-well plates and transfected the following day. Forty-eight hours post-transfection, media was replaced with fresh one containing 2.5 mg/mL G418 and changed every third day till there were no cells left in the untransfected population (~ 2.5 weeks). Media was removed and the adherent cells were washed with PBS, fixed with 4% (w/v) paraformaldehyde (Sigma-Aldrich), stained with 0.2% (w/v) crystal violet (Sigma-Aldrich) and imaged using a conventional table-top scanner. For COS-7 cells, 20,000 cells/well were seeded in 24-well plates and transfected a day after. Twenty-four hours post-transfection, cells were trypsinized and resuspended in 1 mL DMEM + 10% FBS (final volume). Out of 1 mL cell suspension, 10 µL were re-seeded in the corresponding wells of 6-well plates. Twenty-four hours after re-seeding, media was replaced with fresh one containing 1.0 mg/mL G418 and changed every third day till there were no cells left in the untransfected (‘Untreated’) population (~ 10 days). Cells were fixed and stained with crystal violet as described above.

Phenotype inspection for cell death

CV-1 and COS-7 cell were co-transfected with GFP and either FBLN5, luciferase or tBID in 1:4 ratio (1 GFP: 4 test plasmid) in 96-well plates. At 48 h post-transfection, fluorescent green cells were imaged using a IN Cell Analyzer 2000 (GE Healthcare, Chicago,IL, USA) with FITC wavelength filter and counted manually (~ 3000 COS-7 cells and ~ 2500 CV-1 cells).

Transcriptomic analyses

Cells were seeded in 24-well plates and transfected (for FBLN5 overexpression) as described above. Lysates from 12 wells (100 μL Buffer RLT Plus, Qiagen, per well) were pooled and RNA purified using RNeasy Plus Mini Kit (Qiagen) following manufacturer’s instruction. In addition, optional DNA digestion was performed using On-Column RNase-Free DNase set (Qiagen). Integrity of RNA was confirmed using the RNA Pico Chip (Agilent Technologies, Santa Clara, CA, USA). For each sample, 10 ng of RNA was converted into labelled cDNA using the NuGEN Ovation Pico WTA System V2 (NuGEN, Tecan Group Ltd., Männedorf Switzerland), followed by biotinylating step using the NuGEN Encore® biotin Module (NuGEN). Labelled cDNA was hybridised to Affymetrix GeneChip Clariom D human microarrays (Applied biosystems, Thermo Fisher Scientific Inc.) for 20 h at 45° C, washed, stained (GeneChip Fluidics Station 450) and scanned (GeneChip Scanner 3000 7G) according to the manufacturer’s instructions (NuGEN and Affymetrix). Data processing was conducted using Transcriptome Analysis Console v 4.0.1 (TAC) (Applied biosystems, Thermo Fisher Scientific Inc.). Datasets are available through ArrayExpress [19] accession E-MTAB-11449. Unless otherwise stated, gene expression signatures were filtered through a criteria of fold change (± 2.0) and p value and false discovery rate of 0.05 (Fig. 4a). Pathway and comparison analyses were conducted using Ingenuity Pathway Analysis (IPA) [20, 21].

Cell cycle analysis

Cell cycle was assessed by PI staining as described [6]. Briefly, 20,000 cells/well were seeded in a 24-well plate and cultured for 72 h. Cells were harvested, centrifuged, and re-suspended in 300 μL lysis buffer (20 μg/mL propidium iodide, 0.1% sodium citrate and 0.1% triton X-100 in PBS), incubated for 10 min at room temperature and subjected to flow cytometry. Data analysis was performed using Flowjo v7.6.2.

MTT assay

Cells were seeded in a 96-well plate and cell proliferation was measured by adding 20 μL of 5 mg/mL MTT solution (Sigma-Aldrich). Following a 3.5 h incubation, MTT solution was replaced with 150 μl of MTT solvent (4 mM HCl, 0.1% NP-40) in isopropanol. The plate was incubated while shaking for 15 min at room temperature, followed by the measurement of OD at 590 nm.

Live cell imaging

Poly-D-Lysine coated, 35 mm glass bottom dishes (ibidi GmbH, Gräfelfing, Germany) were used for live imaging. Cells were co-transfected with GFP and either FBLN5, luciferase or tBID in 1:4 ratio (1 GFP: 4 test plasmid) as described above and 8 h post-transfection images were captured at 15–20 min intervals. Excitation at 488 nm with 10 mW laser power and 50 ms exposure time was used, emission was detected at 495–550 nm for GFP and halogen lamp for illumination and a full visible spectrum filter cube for phase contrast. During imaging, cells were maintained at 37° C and 5% CO2 under high relative humidity in a stage top incubator (Tokai Hit, Shizuoka, Japan) on the Zeiss Elyra wide-field microscope. Image analysis of the Z-stack time series was performed using Fiji [22].