Long-term severe hypoxia adaptation induces non-canonical EMT and a novel Wilms Tumor 1 (WT1) isoform

The majority of cancer deaths are caused by solid tumors, where the four most prevalent cancers (breast, lung, colorectal and prostate) account for more than 60% of all cases (1). Tumor cell heterogeneity driven by variable cancer microenvironments, such as hypoxia, is a key determinant of therapeutic outcome. We developed a novel culture protocol, termed the Long-Term Hypoxia (LTHY) time course, to recapitulate the gradual development of severe hypoxia seen in vivo to mimic conditions observed in primary tumors. Cells subjected to LTHY underwent a non-canonical epithelial to mesenchymal transition (EMT) based on miRNA and mRNA signatures as well as displayed EMT-like morphological changes. Concomitant to this, we report production of a novel truncated isoform of WT1 transcription factor (tWt1), a non-canonical EMT driver, with expression driven by a yet undescribed intronic promoter through hypoxia-responsive elements (HREs). We further demonstrated that tWt1 initiates translation from an intron-derived start codon, retains proper subcellular localization and DNA binding. A similar tWt1 is also expressed in LTHY-cultured human cancer cell lines as well as primary cancers and predicts long-term patient survival. Our study not only demonstrates the importance of culture conditions that better mimic those observed in primary cancers, especially with regards to hypoxia, but also identifies a novel isoform of WT1 which correlates with poor long-term survival in ovarian cancer.

Cells are transduced with lentivirus in a 24 well format as previously described.One day post transduction, they are passaged into one well of a 6 well plate.A well of untransduced cells of the same cell line is also seeded in one well of a 6 well format.Three days post transduction, media is replaced with media supplemented with 1µg/mL of Puromycin.Media was changed every other day with Puromycin supplemented media for six days, or until 100% of the untransduced control cells have died.
Single cell sorting was performed at the IRIC Flow Cytometry platform using the BD FACSAria III sorter.Each positive cell was sorted directly into a well of a 96 well flat bottom adherent plate containing 150uL of conditioned media (45% fresh cell media, 45% media used to grow the same cell line for 24 hours, and an additional 10% FBS).

Confocal microscopy & cell morphology picture:
Cells were stained with Hoechst (2mg/mL) for 20 minutes, washed with PBS and then fresh DMEM free of phenol red was added prior to live cell imaging.All fluorescent microscopy images were taken using an LSM-880 (Zeiss).GFP was acquired using an Argon-488nm laser, and mCherry was acquired using an Argon-561nm laser.
Images were processed using ImageJ.The cell morphology photos (Fig. 1E) were taken using a white light Nikon Eclipse TS100 microscope, the 10x magnification objective lens, and a Nexus 5 smartphone.
Cell proliferation assay: B16-WT and B16-HG cells were seeded at equal density in triplicate.
Every 24 hours, cells were harvested for counting using a Beckman Coulter cell counter, using the manufacturer's protocol.

Molecular Biology and Biochemical protocols:
Western blot: Proteins were prepared from cell RIPA cell lysate using the Wessel-Fluegge method.
All Proteins were electrophoresed on pre-cast NuPage 4-12% Bis-Tris gel (Life Technologies: NP0321BOX), and migrated at 120V in MES buffer (Life Technologies: J62138.AP).Proteins were transferred to a methanol-soaked Polyvinylidene Fluoride (PVDF) membrane (Cytiva Life Sciences: 10600021) using a wet transfer box set to 200mA for 30 minutes in Towbin buffer.
Primary antibody solutions are prepared in TBST (Tris-buffered saline with 0,5% tween) +3% BSA supplemented with 0.002% (g/ml) of sodium azide (Bioshop: SAZ001), and primary antibody.PVDF membrane sections were incubated in primary antibody solution at Images were processed using ImageJ (US National Institutes of Health).Biosystems).G/P specific PCR products were generated using isoform specific forward primers (G: CCAACTATTTTTGTTCTCCACAGCAC; P: CCAAATGGCGACTGTGAGC) and a common reverse primer (GGACCGGGAGAACTTTCGC).PCR was performed using Pfusion as per the manufacturer's protocol (New England Biolabs: M0530S).

Genomic PCR validation of canonical
ChIP-qPCR analyses: B16-E7-WT1-GFP or B16wt cells were cultured following the LTHY protocol and treated with Dox once the cells were put into the 0.5% O2 condition for 48 hours.
Following this, cells are processed for ChIP as previously described (2).Briefly, cells are fixed with 1% formaldehyde for 10 minutes with shaking.Cross-linking is quenched with 150mM Glycine.Cells are then washed with PBS, collected by scrapping, and gently lysed with a 0.5% NP40 based buffer, which maintains nuclear membrane integrity.Nuclei are retained, then lysed with a 1% SDS buffer, and sonicated.Sonication was performed using a Bioruptor (Diagenode), All probe sets are presented in Supplemental table 5.The negative control, Kmda3, and Vegfa probe sets were used as previously described; all other probe sets were designed by Primer-BLAST (3)(4)(5).The qPCR mixes, acquisition machine, and run settings were the same as for a regular cDNA qPCR run.Experiment was done with biological triplicates.Quantification was done using the fold-enrichment method (6).
Mapping of the WT1 motif to called peaks was done manually in Python using the Biopython module.Generation of gene coverage figures was done using the Spark python tool, using the following flags: -gs, -sm 10) (11.Annotation cluster bubbleplots were generated using DAVID functional annotation enrichment and rendered using Python (13,14).

Characterization of E7-WT1-GFP by Mass Spectrometry: HEK293T cells stably expressing
pCW-E7K-tWt1-GFP were incubated with 2µg/mL Dox for 48 hours prior to lysis.Cells were lysed using the RIPA method as previously described, and (2,5µg) anti-GFP was used to IP the -GFP fusion protein using Protein A Sepharose beads.The IP sample was resuspended in 1X LDS (Invitrogen) supplemented with 10mM DTT and run on a NuPage 4-12% Bis-Tris gel in MES buffer (Invitrogen).Following electrophoresis, bands corresponding to the expected size were excised from the gel and sent to the IRIC Proteomics platform digestion and analyses.The theoretical E7-WT1 protein sequence was used to search for peptide coverage, including posttranslational modifications, using the search engine PEAKS Studio v10.5.Coverage of the tWT1-GFP CDS was calculated using Scaffold v4.8.3.
Small RNA library preparation and next generation sequencing was performed by the McGill University and Genome Quebec Innovation Center.Library preparation was done using the NEB miRNA library preparation protocol (NEB: E7330S).miRNAseq was performed using single end 50bp reads on an Illumina HiSeq.Raw RNAseq reads were trimmed using cutadapt (v1.15) with options that favor specifity (--quality-cutoff 22,20 --error-rate 0.33 --overlap 2 -minimum-length 17 --maximum-length 30 --match-read-wildcards --trim-n) to maximize genomic mapping rate.Genomic mapping was done using miRDeep2 (v2.0.0.8) and bowtie1 (v1.2).Aside from the following options, all settings were set to default: reads shorter than 17nt are discarded; reads can map to up to ten places in the genome; at most; 1 mismatch is permitted per read.After genomic mapping, miRs are counted from the surviving reads.Differential expression was calculated using DESeq2 (v1.14.1) and the Benjamini-Hochberg p-value adjustment (16).
30 seconds ON, 30 seconds OFF at medium intensity for 10 total minutes at 4 o C. DNA fragmentation quality is confirmed via agarose gel migration.50µg of sonicated DNA is used per IP, diluted 1:20 with a dilution buffer and incubated overnight with 2.5µg of rabbit anti-GFP antibody (Invitrogen: #A6455) or Anti-HIF1α (Novus Bio nb100-134) with rotation at 4 o C The following day, samples are incubated with Dynabeads (Thermofisher) for 2 hours at 4 o C.After incubation, proteins are removed from the beads using a 1% SDS based buffer and incubation at 65 o C. Proteins are then digested using Proteinase K (ThermoFisher EO0491) at 55 o C for hour with shaking.DNA is finally purified using a DNA gel recovery kit (Zymo Research: D4001).
ChIPseq sample preparation was performed using the ChIP-qPCR sample preparation protocol with the following adjustment.Nuclear lysate Bioruptor sonication settings are increased to 1x12 minutes 30 seconds ON, 30 seconds OFF, Medium intensity to increase DNA fragmentation to the 200-600bp range.ChIP samples were submitted to the IRIC Genomics platform for library preparation using the KAPA library preparation kit and NGS on an Illumina OVCAR3 RNA and cDNA were prepared using TriZol and High Capacity cDNA Reverse Transcription Kit as per the manufacturer's protocols (ThermoFisher & Applied (ThermoFisher: 15596026) as per the manufacturers protocol.Reverse transcription was performed using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) and 2ug of RNA.qPCR was performed using Advanced qPCR MasterMix HI-ROX (Wisent Bioproducts) as per the manufacturer's protocol.qPCR reactions were run in 10uL technical triplicates, in either 96 well or 384 well formats.Human melanoma qPCR data was acquired using biological duplicates.Ovarian cancer qPCR was acquired using technical triplicates.Primer sequences are detailed in Supplemental table 4.