Impact of genome abnormalities in glioblastoma stem cells on proteome dynamics

Background

Next-generation sequencing has revolutionised our ability to predict disease endotypes and stratify patient groups based on genetic variation. However, our understanding of how mutations affect the proteome, and consequently disease or treatment outcomes, remains limited. For example, gene copy number alterations (CNAs) are extremely common in many types of cancers, including glioblastoma multiforme. Such gene amplifications or losses result in proportionally changed mRNA levels, but the corresponding protein abundances are often buffered or attenuated towards wildtype levels. Identifying which CNAs result in altered protein levels, and how the buffering mechanism works, could significantly improve our understanding of how genetic variation contributes to disease outcomes.

To address such questions, the Kustatscher lab has recently established DIA-pulse-SILAC, a proteomics technique that enables us to determine protein synthesis and degradation rates with high speed and sensitivity. This works by pulse-labelling newly synthesised proteins with non-radioactive amino acid isotopes, distinguishing them from unlabelled, pre-existing proteins in the cell. Moreover, nascent proteins can be tagged with enrichable methionine-analogues, allowing us to selectively analyse newly translated proteins. These techniques could be applied to study if, for example, cells compensate for gene duplications through decreased translation or enhanced degradation.

Until recently, mass spectrometry-based proteomics was not routinely used to screen clinical or research samples, as it did not achieve the required throughput and robustness. However, the recent advent of high-throughput proteomics (HTP) is rapidly changing that. In collaboration with Markus Ralser (Charité hospital, Berlin), we recently reported the largest cellular proteomics study to date, mapping the impact of nearly 5,000 gene deletions on the yeast proteome (Messner et al, Cell, 2023). Now, as part of the upcoming Wellcome Discovery Research Platform for Hidden Cell Biology, we are setting up a similar HTP platform in Edinburgh.

The Pollard group is studying Glioblastoma multiforme, the most common and aggressive form of adult brain cancer, with a median survival time of 15 months. Glioblastomas are initiated and fueled by glioblastoma neural stem cells (Gangoso et al, Cell, 2021). The group has established the unique Glioma Cellular Genetics Resource (GCGR), which contains early-passage neural stem cell lines established directly from healthy and cancer patient tissues. Normal (NS) and glioblastoma-derived neural stem cells (GNS) are phenotypically very similar. However, NS cells are cytogenetically normal, whereas GNS cells have many CNAs. In these cell lines, CNAs were already mapped at high resolution.

Aims

The aim of this project is to understand how glioblastoma-derived stem cell lines of different patients differ in terms of proteome dynamics, and how this relates to normal neural stem cells. We will integrate these data with orthogonal data already acquired for these cell lines, including genome and transcriptome sequences. The student can investigate differences in the occurrence and buffering of CNAs, whether buffering is linked to adaptations of translation or degradation, and if computational models can predict which CNAs will result in altered protein levels. We will compare the nascent proteome of NS and GNS cell lines, with a focus on any unstable, stress-induced proteins originating from non-coding genomic regions, which could be exploited for immunotherapy and vaccine development.

We will take further advantage of the HTP platform to determine the response of these cell lines to pharmacological treatments that are under development in the Pollard lab, which trigger the degradation of proteins required for GNS cell survival.

Training outcomes

The student will learn how to cultivate and isotope-label patient-derived stem cells, enrich nascent proteomes, and analyse samples by proteomics. They will have the opportunity to develop quantitative skills, including proteomics data processing, R programming and statistical analysis of large datasets, including integration with orthogonal omics data.

References

Messner CB, Demichev V, Muenzner J, Aulakh SK, Barthel N, Röhl A, Herrera-Domínguez L, Egger AS, Kamrad S, Hou J, Tan G, Lemke O, Calvani E, Szyrwiel L, Mülleder M, Lilley KS, Boone C, Kustatscher G, Ralser M. The proteomic landscape of genome-wide genetic perturbations. Cell. 2023 Apr 27;186(9):2018-2034.e21. doi: 10.1016/j.cell.2023.03.026.

Gangoso E, Southgate B, Bradley L, Rus S, Galvez-Cancino F, McGivern N, Güç E, Kapourani CA, Byron A, Ferguson KM, Alfazema N, Morrison G, Grant V, Blin C, Sou I, Marques-Torrejon MA, Conde L, Parrinello S, Herrero J, Beck S, Brandner S, Brennan PM, Bertone P, Pollard JW, Quezada SA, Sproul D, Frame MC, Serrels A, Pollard SM. Glioblastomas acquire myeloid-affiliated transcriptional programs via epigenetic immunoediting to elicit immune evasion. Cell. 2021 Apr 29;184(9):2454-2470.e26. doi: 10.1016/j.cell.2021.03.023.