Quantitative Analysis of Signaling Networks across Differentially Embedded Tumors Highlights Interpatient Heterogeneity in Human Glioblastoma

Abstract

Glioblastoma multiforme (GBM) is the most aggressive malignant primary brain tumor, with a dismal mean survival even with the current standard of care. Although in vitro cell systems can provide mechanistic insight into the regulatory networks governing GBM cell proliferation and migration, clinical samples provide a more physiologically relevant view of oncogenic signaling networks. However, clinical samples are not widely available and may be embedded for histopathologic analysis. With the goal of accurately identifying activated signaling networks in GBM tumor samples, we investigated the impact of embedding in optimal cutting temperature (OCT) compound followed by flash freezing in LN2 vs immediate flash freezing (iFF) in LN2 on protein expression and phosphorylation-mediated signaling networks. Quantitative proteomic and phosphoproteomic analysis of 8 pairs of tumor specimens revealed minimal impact of the different sample processing strategies and highlighted the large interpatient heterogeneity present in these tumors. Correlation analyses of the differentially processed tumor sections identified activated signaling networks present in selected tumors and revealed the differential expression of transcription, translation, and degradation associated proteins. This study demonstrates the capability of quantitative mass spectrometry for identification of in vivo oncogenic signaling networks from human tumor specimens that were either OCT-embedded or immediately flash-frozen.

Quantitative Phosphoproteomics Reveals Wee1 Kinase as a Therapeutic Target in a Model of Proneural Glioblastoma

“…Once excised, flash frozen tumors and brain tissues to be analyzed by MS and western blotting were homogenized (Polytron) in ice-cold lysis buffer consisting of 8M urea supplemented with 1 mM sodium orthovanadate, 0.1% Nonident P-40 (NP-40), and protease inhibitor and phosSTOP tablets (Roche), as described by Johnson and White (14). Samples were homogenized on ice using 5×10 sec pulses, with 10 sec intervening periods to prevent tissue heating.…”

High risk glioblastoma cells revealed by machine learning and single cell signaling profiles

“…Resected tissues were immediately dissociated into single-cell suspensions as previously reported 38 . Cells were stained with a customized antibody panel for mass cytometry designed to capture the expression of known cell surface proteins, intracellular proteins, and phospho-signaling events (Figure 1, 2, Supplementary Table 2, Supplemental Information) that are critical for gliomagenesis and pathogenesis 32,39,40,30,34,35 . Collectively, the antigens included in this panel positively identified >99% of viable single cells within any given tumor sample.…”

“…These highly aggressive tumors are composed of both tumor and stromal cells, which harbor diverse genomic, transcriptomic, and proteomic expression profiles reflecting abnormal neural lineages 24,29,30 . Previous studies in glioblastomas have either measured signaling states in bulk primary tumors [31][32][33] or characterized genomic and transcriptomic profiles in a modicum of single cells (<1000) 29,30,34,35 . These approaches, however, have not yet improved clinical practice or outcome for patients with glioblastoma.…”

Comprehensive proteome analysis of fresh frozen and optimal cutting temperature (OCT) embedded primary non-small cell lung carcinoma by LC–MS/MS

“…However, the polymers in OCT medium (i.e., PVA and PEG) interfere with MS analysis by suppressing ion formation [7], and thus need to be removed from samples. A few proteomic studies have been conducted in OCTembedded tissues [8][9][10][11][12]. In some studies, OCT was cleared in subsequent experimental steps (e.g., any separation approach that involved running polyacrylamide gels) [8,9,12].…”

“…A few proteomic studies have been conducted in OCTembedded tissues [8][9][10][11][12]. In some studies, OCT was cleared in subsequent experimental steps (e.g., any separation approach that involved running polyacrylamide gels) [8,9,12]. While in other studies, the number of proteins identified was compromised due to incomplete elimination of OCT medium [10,11].…”