Epithelioid peritoneal mesothelioma: a hybrid phenotype within a mesenchymal-epithelial/epithelial-mesenchymal transition framework

The aim of this study was to reconsider the biological characteristics of epithelioid malignant peritoneal mesothelioma (E-MpM) in the light of new concepts about epithelial mesenchymal transition and mesenchymal epithelial reverse transition (EMT/MErT) and the role of epigenetic reprogramming in this context. To this end we profiled surgical specimens and derived cells cultures by a number of complementary approaches i.e. immunohistochemistry, immunofluorescence, in situ hybridization, biochemistry, pluripotent stem cell arrays, treatments with cytokines, growth factors and specific inhibitors. The analyses of the surgical specimens showed that i) EZH2 is expressed throughout the spectrum of MpM, ii) that E-MpM (including the high-grade undifferentiated form) are characterised by c-MYC and miRNA 17-5p expression, and iii) that progression to sarcomatoid MpM is dictated by EMT regulators. They also showed that E-MpM expressed c-MET and are enriched in E- and P-cadherins- and VEGFR2-expressing CSCs, thus strongly supporting a role for MErT reprogramming in endowing E-MpM tumour cells with stemness and plasticity, and hence with a drug resistant phenotype. The cell culture-based experiments confirmed the stemness traits and plasticity of E-MpM, and support the view that EZH2 is a druggable target in this tumor.


VEGFR2 mRNA ISH
VEGFR2 mRNA ISH was carried out manually using the RNAscope kit (Advanced Cell Diagnostics, Inc., Hayward, CA) in accordance with the manufacturer's instructions. In brief, 5 μm sections of formalin-fixed, paraffin-embedded tissue were pre-treated with heat and protease before being hybridised with the Hs-VEGFR2 probe (Cat. No. 312121, Advanced Cell Diagnostics, Inc.). A horseradish peroxidase-based signal amplification system has then bound to the target probes followed by colour development with diaminobenzidine. Positive staining appeared as brown dots in the nucleus and/or cytoplasm. The bacterial gene 4-hydroxytetrahydrodipicolinate reductase (DapB) (negative control) and the housekeeping gene ubiquitin C (UbC) (positive control) were included in each run.

miRNA ISH and quantification
MiRNA ISH was carried out as previously described [Gualeni et al, J Clin Pathol. 2015vol 68, pp661-664 2015. Briefly, after tissue preparation and permeabilisation, tumoral sections were hybridised for two hours with a double-DIG-LNA probe for hsa-miR-17 Exiqon,Vedbaek,Denmark) at 61°C, an hsa-miR-21 probe (positive control; Cat. No. 38102-15, Exiqon) at 53°C, and a scramble-miR probe (negative control; Cat. No. 99004-15, Exiqon) at 57°C. Signal detection was automated using the OptiviewDAB Detection Kit (Ventana Medical Systems, Tucson, Arizona, USA) on a Ventana BenchMark ULTRA (Ventana Medical Systems). Positive miRNA 17-5p staining appeared as brown dots usually localised in the cytoplasm. The size and intensity of the dots varied from case to case.

Immunofluorescence (IF)
VEGFR2/E-cadherin, E-cadherin/P-cadherin, and E-cadherin/Gata-4 co-expression was investigated by means of IF. The antibodies were diluted as shown in Supplementary Table S2, and the antigen retrieval was carried out using the fully automated BenchMark ULTRA (Ventana) instrument in accordance with the manufacturer's instructions. The slides were incubated with the specific secondary Alexa Fluor antibodies (Alexa Fluor 488 and Alexa Fluor 546, Thermo Fischer Scientific, MA, USA) at room temperature for one hour, and then mounted using Vectashield mounting medium with DAPI (Vector Labs, Burlingame, CA, USA). The samples were observed by means of a Leica DM6000B microscope equipped with a 100 W mercury lamp, with excitation being obtained using Spectrum Orange (546 nm), Spectrum Green (488 nm) and DAPI excitation filters. The images were acquired through 20x and 40x oil immersion objectives, and analysed using Cytovision software. The images from each channel were collected sequentially in order to limit fluorescence cross-talk. The published images represent extended depth-of-field frames in stack, with focus regions selected on the basis of their maximum intensity.

Biochemistry
The proteins from tissue samples stored at -80°C were homogenised at 4°C in lysis buffer (50 mmol/L HEPES, 150 mmol/L NaCl, 10%glycerol, 1% Triton X-100, 1.5 mmol/L MgCl 2 , 1 mmol/L EGTA, 10 mmol/L Na 4 P 2 O 7 , and 100 mmol/L NaF) supplemented with protease and phosphatase inhibitors (Cocktail Inhibitors I and II, Sigma, St. Louis, MO). The lysing involved frequent vortexing, and the lysates were then cleared by means of centrifugation at 13,000 rpm for 30 minutes at 4°C and the proteins were measured using a Bio-Rad protein assay. Western blotting (WB) was carried using standard procedures and 20 μg of proteins. The antibodies used are listed in Supplementary Table S3.

Cell cultures
Fresh aseptic surgical MpM samples were minced and incubated with collagenase (Cat. No. C6885, Sigma) for three hours, and the obtained cell suspension was filtered through a 45 μL nylon mesh, washed with RPMI/ FCS 0.5%, diluted in PBS/BSA 1%, and seeded in RPMI (Cat. No. 21875-034, Invitrogen) supplemented with 10% of fetal bovine serum.

Biochemistry
When sub-confluence was reached, the cells were detached as described above for the surgical samples, and analysed by means of WB using the antibodies and conditions shown in Supplementary Table S3.

Flow cytometry (FC)
FC staining was used in order to distinguish MpM and non-MpM cells in the primary cultures. Cells obtained after detachment with trypsin-EDTA (Cat. No. 15400, Invitrogen) were washed twice in PBS, counted, and diluted to an appropriate concentration for staining with CAM 5.2 FITC (Cat. No. 35303,Becton Dikinson). Briefly, 2x10 5 cells were placed in a 12 x 75 mm roundbottomed polystyrene tube and, after washing with PBS/ BSA, the cells were labelled with CAM 5.2 FITC using the Fix and Perm Kit (Cat. No. GAS001,Invitrogen) in accordance with the manufacturer's instructions. At least 5x10 4 cells were acquired using a FACS-Canto 2 cytometer, and analysed using FACS DIVA or Win MDI software.

Immunoistochemistry (IHC)
The primary and stabilised cell cultures were detached using trypsin, and fixed for one hour at room temperature in a 10% solution of buffered formalin. Formalin-fixed, paraffin-embedded (FFPE) cell blocks suitable for IHC were then obtained using standard procedures. The IHC experiments were performed under the conditions shown in Supplementary Table 2.

RNA analyses
RNA was extracted from the cells using an RNeasy minikit (Cat. No. 74104,Qiagen), and quantified using nano-drop device. Five hundred nanograms of of RNA was used for RT-PCR and cDNA synthesis. cDNA retrotranscribed from 500 ng of RNA obtained from U-87 glioma cell line was used as a reference sample.