br DepArray analysis and next generation sequencing br
2.3. DepArray analysis and next-generation sequencing
In order to determine the existence, frequency and origin of CK+/VIM+ cells in the tumor Talaporfin population, we chose a
DepArray method (Silicon Biosystems, Bologna) . Briefly, samples in tubes were sent to Silicon Biosystems, where they were deparaffinized and disaggregated into a single cell sus-pension. Next, heat-induced epitope retrieval and immunoflu-orescence labeling for CK-Alexa488, VIM-Alexa647 and DAPI were performed. The stained single cell suspension was pipetted into a microfluidic silicon-biochip single use car-tridge, where the single cells were distributed into individual di-electrophoresis chambers, and inserted into the DEPArray instrument. The trapped cells of interest (ie, cells positive for DAPI, VIM, and CK) could be individually selected, separated by di-electrophoresis, counted and recovered. In a last step, next-generation sequencing was performed on the recovered cells using the Ion AmpliSeq Cancer Hotspot Panel v2 and analyzed with TorrentSuite v4.4 (Thermo Fischer).
2.4. Fluorescence-activated cell sorting
Tissues from 6 colorectal cancers were carefully selected for fluorescence-activated cell sorting (FACS) analysis. Blocks were chosen such that no normal tissue, fat, mucinous components or CK-positive stromal areas were included in the blocks. Nine 50 μm tissue rolls were cut and packed into nylon biopsy bags to prevent cell loss. Deparaffinization was per-formed in a large volume of xylol, followed by rehydration in decreasing percentages of ethanol. Antigen retrieval was performed in citrate buffer for 2 hours at 80°C. The tissue was enzymatically dissociated for 60 min in RPMI containing 0.2% collagenase la and dispase. Antibody labelling was performed using CK (MNF116 – Dako and AE1/AE3 – Millipore) and VIM (3B4 – Dako) as primary, and AF488 IgG1 and AF647 IgG2a (Life Technologies) as secondary antibodies. Additionally, the nucleus was labeled with DAPI.
2.5. Matched CK-positive and CK-negative ngTMA
Eighty-nine mixed stage colon cancers cases with CK-positive stroma, treated at the Department of Surgery at the Tech-nical University Munich hospital, Munich, Germany, between 1993 and 2005 were included in this part of the study. A next-generation tissue microarray (ngTMA) of matched CK-positive and negative stromal areas was constructed  (Supplementary Fig. 1). Briefly, the CK slide was scanned and annotated using a TMA tool of 0.6 mm diameter in one area of CK positivity and one area of negativity. The annotated scans were then aligned to the image of the donor block and the precise area was cored out for ngTMA construction. The end result was a matched CK-positive and negative ngTMA used for subse-quent analysis of stromal markers by immunohistochemistry.
to whole tissue slides from the five cases sent for DepArray analysis as well as the tumor invading vessel ngTMA. Slides were cut at 2.5 um. Fluorescent labelling was performed on the Bond RX (Leica Microsystems). Antigen retrieval with citrate buffer at 100°C for 30 min was applied. Subsequently, we applied 0.2% (v/v) Triton X-100 in 1× phosphate-buffered saline (PBS) for 20 min. The blocking step was implemented using 1% bovine serum albumin (w/v) and 5% horse serum (w/v) for 20 min. Primary antibodies (listed in Supplementary Table 1), mouse pan-CK (AE1/AE3) antibody with rabbit VIM antibody and mouse E-cadherin antibody with rabbit VIM antibody were incubated for 60 min. Goat anti-rabbit Alexa Flour 488 and goat anti-mouse Alexa Fluor 546 (Invitrogen), were used as secondary antibodies, in a dilution of 1:1000 for both antibodies, for 60 min. Mounting medium with DAPI (Vectashield). Images were taken on an Olympus FluoView-1000 (Olympus) confocal microscope at 60× magnification.
Single and double immunohistochemistry was performed on the Bond RX (Leica Microsystems) on the matched stroma ngTMA for the following makers: (1) E-cadherin and VIM in order to determine the epithelial and mesenchymal origin respectively; (2) Snail1, Twist1, Zeb1 and Zeb2 to determi-nate the activated EMT/mesothelial-mesenchymal transition (MMT) process; (3) COL11A1, SPARC, CD90, α-SMA, and FAP to determine the “CAF/myofibroblast” phenotype; (4) WT1 determine mesothelial phenotype. In addition, whole tissue slides from one tumor with CK-positive stroma and known BRAF V600E mutation underwent immunohistochemistry for the VE1 antibody and CK20. This was performed in order to determine whether the CK-positive stroma derived from the tumor itself. All primary antibodies and protocols are listed in Supplementary Table 1. All single stainings as well as the first step in double stainings were visualized with Bond polymer refine detection, using 3,3′-diaminobenzidine (DAB) as brown chromogen (Leica Bio-systems). The second step of double stainings were visualized with Red polymer refine Detection, using fast red as red chro-mogen (Leica Biosystems, Ref DS9390). Finally, the sam-ples were counterstained with hematoxylin and mounted with Aquatex (Merck). Slides were scanned using Pannoramic P250 (3DHistech, Budapest).