br MARCoNI Assay br The MARCoNI
The MARCoNI assay was performed on PamChIP #88101 arrays (PamGene International), using whole cell lysates and AR antibody 441 (sc-7305, Santa-Cruz Biotechnology) for detection, and FITC-conjugated swine anti-rabbit (F0205, DAKO) for visualization. Cells were grown with dox and induced for 30 min with 10 nM DHT prior to cell lysis.
Pooled genome-wide CRISPR screens were performed using 2x108 LNCaP95 108708-22-1 infected with pooled lentiviral GeCKO v2 library (Fei et al., 2017). After three days of puromycin (2 mg) selection, the surviving cells were divided into three groups (0 day control, vehicle, and 10 nM DHT treatment) and cultured for four weeks prior to genomic DNA extraction. The genomic DNA was subse-quently amplified using two rounds of PCR and each library was sequenced at 30-40 million reads to achieve 300x coverage of the CRISPR library.
QUANTIFICATION AND STATISTICAL ANALYSIS
Tissue Microarray Analysis
Using Aperio ImageScope software, ARv7-stained TMA slides were annotated by outlining each TMA spot on a separate annotation layer to create regions of interest for analysis. Quantitative image analysis of the annotated regions of interest was performed using Aperio Brightfield Image Analysis Toolbox software (Leica Biosystems Pathology Imaging, Vista, CA). The data for each TMA spot was extracted into Microsoft Excel for further analysis. The quantitative analysis data for each TMA spot included total numbers and percentages of nuclei (positive and negative), average positive intensity, average positive optical density, and area of analysis (Krajewska et al., 2009; Rizzardi et al., 2012). The TMA 55 cores were then categorized into low and high ARv7 groups by selecting
those with the highest and lowest quartiles of ARv7 IHC score. The Significance Analysis of Microarrays (SAM) program (Tusher et al., 2001) was used to analyze expression differences between groups using unpaired, two-sample t-tests and controlled for multiple testing by estimation of q-values using the false discovery rate (FDR) method.
Patient Data Clustering and Analysis of Recurrence-free Survival
Patients from the Decipher-GRID cohorts (Benzon et al., 2017; Boormans et al., 2013; Den et al., 2014; Erho et al., 2013; Karnes et al., 2013; Klein et al., 2015; Ross et al., 2016; Taylor et al., 2010) were clustered according to the expression (mRNA; Z-score) of the four tumor suppressive genes (SLC30A7, B4GALT1, HIF1A and SNX14). Recurrence-free survival was assessed using the Kaplan-Meier method. Patients were censored at the time of their last clinical, tumor-free, follow-up visit. Time to PSA recurrence (as defined by each study individually) was selected as the clinical endpoint.
Western Blot Quantification
Western blots were quantified calculating the average grey level for each band after background subtraction, using a custom script in ImageJ (Schneider et al., 2012).
Differential gene expression was determined using the Visualization Pipeline for RNA-seq (VIPER)(Cornwell et al., 2018) and DEseq (Anders and Huber, 2010).
ChIP-seq data was aligned to the hg19 genome and ChIP-seq peaks determined using the ChIP-seq data quality and analysis pipe-line 2 (Qin et al., 2016). Read counts for cluster analyses were quantified using the bamliquidator tool (Loven et al., 2013; Whyte et al., 2013), with read counts being subsampled to match the lowest read count for each antibody. While evaluating the expression of the genes associated with the clusters only genes that could be associated to a unique cluster were considered. DNA motif analysis was performed using SeqPos (He et al., 2010).
FRET and FRAP Analysis Apparent abFRET efficiencies were calculated using the equation: abFRET=((CFPafter-CFPbefore) *YFPbefore)/((CFPafter*YFPbefore)-(YFPafter*CFPbefore)). Apparent abFRET efficiencies were normalized to the negative (CFP and YFP single fluorescent plasmid) and positive controls (CFP-YFP fusion protein). FRAP efficiencies were calculated according to this formula: Inorm, t = (It-Ibackground)/ (Iprebleach-Ibackground).
MARCoNI Assay Analysis
Binding was determined through array image analysis, consisting of automated spot finding, quantification and background subtrac-tion, using the BioNavigator software (PamGene International).
CRISPR Screen Analysis
CRISPR screen sequencing data was analyzed using MAGeCK (Li et al., 2014).
DATA AND SOFTWARE AVAILABILITY
The accession number for the RNA-seq and ChIP-seq data are GEO:GSE106560 and GEO: GSE106559, respectively.
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences