The Proteome of Primary Prostate Cancer Diego Iglesias-Gato, Pernilla Wikström, Stefka Tyanova, Charlotte Lavallee, Elin Thysell, Jessica Carlsson, Christina Hägglöf, Jürgen Cox, Ove Andrén, Pär Stattin, Lars Egevad, Anders Widmark, Anders Bjartell, Colin C. Collins, Anders Bergh, Tamar Geiger, Matthias Mann, Amilcar Flores-Morales  European Urology  Volume 69, Issue 5, Pages 942-952 (May 2016) DOI: 10.1016/j.eururo.2015.10.053 Copyright © 2015 European Association of Urology Terms and Conditions

Fig. 1 Workflow for the quantitative mass spectrometry analysis of formalin-fixed paraffin embedded prostatectomy samples. (A) Epithelium-derived LNCaP (androgen sensitive, AR+), 22rv1 (androgen insensitive, AR+), and PC3 (androgen insensitive, AR-) cells, and stroma-derived WPMY-1 cells were metabolically labeled with heavy arginine and lysine for 10 generations. Whole cell protein extracts of each cell line were mixed in the indicated proportions (%) to create the spike-in standard. Whole protein extracts from formalin-fixed paraffin-embedded prostatectomy samples were mixed with the standard in a 1:1 ratio. Combined extracts were digested with trypsin and the resulting peptides were further fractionated by strong anionic exchange to simplify the complexity of the peptide mixture. Each fraction was analyzed by online liquid chromatography–Mass spectrometry/mass spectrometry using an Exactive-Q mass spectrometer. Peptides were identified and quantified using MaxQuant software suit. (B) Scatter plot of the relative abundance of all proteins shown as the log10 of the median intensity of the tumors (n=28) and the benign neighboring prostate tissue (controls; n=8). Relative density is color-coded and the names of several prostate cancer relevant proteins are indicated. (C) Number of proteins with valid ratios: light (from prostatectomy origin)/heavy (from cell line origin) for each tumor and control sample. FASP=filter-aided sample preparation; FFEP=formalin-fixed paraffin-embedded; LC-MS/MS=liquid chromatography–mass spectrometry/mass spectrometry; SAX=strong anion exchange; SILAC=stable isotope labeling by amino acids in cell culture. European Urology 2016 69, 942-952DOI: (10.1016/j.eururo.2015.10.053) Copyright © 2015 European Association of Urology Terms and Conditions

Fig. 2 Network representation of gene ontology terms enriched among proteins differentially regulated between tumors and control tissue. Functional categories overrepresented among the proteins with elevated (orange) or reduced (blue) expression in the tumors compared to the controls. Categorical enrichment was calculated using DAVID and enrichment results were plotted using Cytoscape. Line thickness represents number of shared proteins between categories. On each node, the size of the red inner circle and the thickness of the red ring relates to the number of proteins up-regulated and down-regulated, respectively. loc=localization. European Urology 2016 69, 942-952DOI: (10.1016/j.eururo.2015.10.053) Copyright © 2015 European Association of Urology Terms and Conditions

Fig. 3 Elevated mitochondrial content and activity in prostate tumors. (A) Immunohistochemical staining of mitochondrial proteins ACAD9 and NDUFAF1 in radical prostatectomy specimens containing both tumor (T; arrows) and benign prostate tissue (C; arrowheads). Immunoreactivity was estimated in a 0–3 scale (right panel). Student t test p values are indicated. (B) Mitochondrial complex IV activity was measured with histochemistry in frozen radical prostatectomy specimens containing both tumor (arrows) and benign prostate tissue (arrowheads). Staining intensity was defined as low (1) or high (2). Student t test p values are indicated. (C) Analysis of cell proliferation of 22rv1 prostate cancer cells treated with increasing amounts of the carnitine palmitoyltransferase II (CPT2) inhibitor L-aminocarnitine, or the fatty acid oxidation inhibitor trimetazidine. * Student t test p value < 0.05 at the end of the experiment. European Urology 2016 69, 942-952DOI: (10.1016/j.eururo.2015.10.053) Copyright © 2015 European Association of Urology Terms and Conditions

Fig. 4 Increased proneuropeptide-Y (pro-NPY) expression correlates with poor prognosis in prostate cancer. (A) Scatter plot of the different relative expression of proteins quantified in prostate tumors with primary Gleason Grade 4 (high risk) versus Gleason Grade 3 (low risk) (expressed as Log2 values) distributed according to the probability of being true expressed as –log10 of Student t-test p value. Red dots represent proteins up or down-regulated more than 2-fold with a p value < 0.05. (B) Mass spectrometry/mass spectrometry spectrum of the most commonly identified NPY peptide across all samples. Peptide matches to the c-terminal domain of NPY (pro-NPY), depicted in orange. (C) Immunohistochemical staining of pro-NPY. Pro-NPY immunoreactivity (IR) was scored as negative (IR=0), weak (IR=1), moderate (IR=2), and strong (IR=3). (D) Pro-NPY expression was examined on tissue microarrays from 10 different common cancer types (n=40 patients each, two cores per patient). IR was scored as negative (IR=0), weak (IR=1), moderate (IR=2), and strong (IR=3). (E) Pro-NPY IR in prostate tumors and benign neighboring tissue was measured on a “watchful waiting” cohort. Chi square p value is shown. (F) Kaplan-Meier analysis of pro-NPY expression relative to prostate cancer-free survival. (G) Kaplan-Meier analysis of pro-NPY expression relative to prostate cancer-free survival of patients with Gleason score > 7 (left) or Gleason score ≤ 7 (right). In Kaplan-Meier analysis pro-NPY IR score of 2 (moderate) and 3 (strong) were considered “high NPY”, while IR score of 0 (negative) and 1 (low) were considered “low NPY”. GS=Gleason score; HR=hazard ratio; IR=immunoreactivity; PCA=prostate cancer; pro-NYP=proneuropeptide-Y. European Urology 2016 69, 942-952DOI: (10.1016/j.eururo.2015.10.053) Copyright © 2015 European Association of Urology Terms and Conditions

Fig. 5 Increased proneuropeptide-Y expression in combination with positive ERG status correlates with poor prognosis in prostate cancer. (A) Kaplan-Meier analysis of proneuropeptide-Y expression dichotomized as high (moderate or strong immunoreactivity) or low (negative or weak immunoreactivity) in combination with the ERG status (positive “ERG +” or negative “ERG -”) relative to prostate cancer-free survival. (B) Same analysis as in (A) after dichotomizing the patients by the Gleason score of the tumors at diagnosis. GS=Gleason score; HR=hazard ratio; IR=immunoreactivity; PCa=prostate cancer; pro-NYP=proneuropeptide-Y. European Urology 2016 69, 942-952DOI: (10.1016/j.eururo.2015.10.053) Copyright © 2015 European Association of Urology Terms and Conditions