Multi-Omics Analysis of Glioblastoma Cells’ Sensitivity to Oncolytic Viruses

Abstract

Oncolytic viruses have gained momentum in the last decades as a promising tool for cancer treatment. Despite the progress, only a fraction of patients show a positive response to viral therapy. One of the key variable factors contributing to therapy outcomes is interferon-dependent antiviral mechanisms in tumor cells. Here, we evaluated this factor using patient-derived glioblastoma multiforme (GBM) cultures. Cell response to the type I interferons’ (IFNs) stimulation was characterized at mRNA and protein levels. Omics analysis revealed that GBM cells overexpress interferon-stimulated genes (ISGs) and upregulate their proteins, similar to the normal cells. A conserved molecular pattern unambiguously differentiates between the preserved and defective responses. Comparing ISGs’ portraits with titration-based measurements of cell sensitivity to a panel of viruses, the “strength” of IFN-induced resistance acquired by GBM cells was ranked. The study demonstrates that suppressing a single ISG and encoding an essential antiviral protein, does not necessarily increase sensitivity to viruses. Conversely, silencing IFIT3 and PLSCR1 genes in tumor cells can negatively affect the internalization of vesicular stomatitis and Newcastle disease viruses. We present evidence of a complex relationship between the interferon response genes and other factors affecting the sensitivity of tumor cells to viruses.

DirectMS1Quant: ultrafast quantitative proteomics with MS/MS-free mass spectrometry

“…To explore the DirectMS1Quant sensitivity for quantitative description of cellular response to stimuli, two glioblastoma, GBM3821 and GBM4114, and the astrocyte (AN) cells treated with interferon alpha (IFN α ) were used. Earlier it has been demonstrated that AN and GBM3821 cells respond to IFN α by upregulation of dozens of IFN-induced proteins, while GBM4114 has impaired interferon-dependent antiviral mechanisms and does not respond to IFN stimulation 23 . In this work, IFN response of these cellular models was quantified using different workflows shown in Table 1 .…”

“…MS/MS-based analysis of glioblastoma samples was performed using Q-Exactive HF mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA). Experimental label-free data for DDA mode were published elsewhere 23 . Experimental data for TMT were acquired in two ways: single-shot LC-MS/MS runs and deep proteome analysis with sample pre-fractionation.…”

“…Label-free LC-MS/MS experiments with two low-passage cultures of GBM (3821, 4114) and normal astrocytes (AN) cell lines were used from previous study 23 (ProteomeXchange id PXD022906). Briefly, cell cultures were grown in four replicates and treated with recombinant interferon (IFN) α-2b (4 control, 4 treated samples).…”

“…Differentially expressed proteins were selected according to the following criteria: |log 2 FC| >= 1 standard deviations of background fold changes, fdr_bh <= 0.05. IFN-stimulated gene products were recognized using public databases 23,36,37 To further demonstrate the quantitative capabilities of ultrafast DirectMS1 method, ranking of interferon pathway functionality was performed at the level of gene ontology enrichments (Figure 4). Processes were ranked by GO_score 38 calculated using the following equation: where fdr_BH is the Benjamini-Hochberg false discovery rate and E is the enrichment, both calculated by STRING db.…”

“…Peptides were desalted using cartridges for solid-phase extraction (Oasis HLB, 1 cc, 30 mg, 30 µm particle size, Waters, Milfold, CT, USA) and dried. Note, that glioblastoma cell lines used in this work were biological samples from the previous study 23 .…”