Introduction and findings
Glioblastoma (GBM), an aggressive type of brain tumour, is one of the deadliest forms of cancer. Development of effective therapeutic approaches for GBM faces multiple hurdles, including inter- and intra-tumour cellular and transcriptomic heterogeneity. As a result, remnants of tumour can survive by not reacting to anti-cancer drugs. For example, according to previous single-cell based studies, GBM initiating cells (GICs) possess transcriptomes that mimic neural stem cells (NSCs), which in part leads to poor response to Temozolomide (TMZ) treatment, currently the only U.S. FDA-approved GBM medication (Couturier et al., 2020). However, further comparative studies on the transcriptomes and epigenomes of GICs and NSCs is hindered by the lack of appropriate controls and the fact that most studies occur in cell lines that lack physiological contexts. In a recent report, Vinel et al. tackled such problems by utilizing primary GICs and patient-matched expanded potential stem cell (EPSC)-derived NSC (iNSC) to identify both shared and patient-specific dysregulated and druggable pathways.
Patient-matched GICs and iNSCs were cultured and derived from surgical tumours and fibroblasts from the dura mater, respectively. Since human endogenous NSCs (eNSCs) were unobtainable, the authors aimed to show that iNSC can serve as a suitable proxy. Mouse eNSCs and iNSCs from the same mouse were derived and shown to have similar transcriptomes and epigenomes.. By using this patient-matched, or syngeneic models for comparative analysis, in specific group of GBM patients with higher number of regulatory T cells in their tumours, Vinel and colleagues were able to identify dysregulated molecular pathways in GICs, such as the glycosaminoglycan genes. Losing glycosaminoglycan can contribute to decreased cellular immunity, which is commonly found in GBM (El Andaloussi et al., 2006).
In order to develop patient-specific therapeutic approaches, this study also focused on identifying the dysregulation of genes in specific individuals by comparing the methylomes and transcriptomes of GICs and matched iNSCs. Integrating information from the drug-gene interaction database (DGIdb), the selected candidate drugs were able to inhibit growth of GBM cells or tumours, validated in mouse xenografts and in 3D cerebral organoid glioma (GLICO) models.
Opinions
This study is significant in several aspects. Firstly, the authors were able to develop an appropriate control model for analysing GICs. The iNSCs induced from the same patients were highly similar biologically to eNSC. This is a remarkable achievement, as GMB had been reported to originate from NSCs. Furthermore, this syngeneic model can also avoid variations from non-matched samples and controls. In addition, by integrating results from data such as cell type composition in tumours, and comparative analysis of transcriptomes and methylomes between GIC and iNSC, the syngeneic model was able to identify patient-specific and GBM-related molecular phenotypes, such as the perturbation of glycosaminoglycans sulfation in GIC. This information may help in personalizing therapeutic regimes for individual patients.
The findings in this study provide the groundwork for many possible future directions. Recent single cell transcriptomic studies have re-categorized GBM into several subtypes (Neftel et al., 2019) in addition to the widely adopted classifications from The Cancer Genome Atlas (TCGA). As demonstrated by the association between enrichment of regulatory T cells and the glycosaminoglycans sulfation pathway found from this paper, similar approaches could be taken to potentially identify more GBM subtype-specific dysregulated transcriptional pathways. Unfortunately, the comparative analyses in this study was largely restricted to protein-coding genes. Epigenetic dysregulation in cancers can also reactivate non-coding sequences, which can play vital regulatory roles in gene expression (Jang H.S. et al.,2019). A parallel inspection of specific transcriptomic and epigenomic patterns of the non-coding genome in GBM patients may provide more insights for customized drug treatment.
References
Couturier, C. P., Ayyadhury, S., Le, P. U., Nadaf, J., Monlong, J., Riva, G., Allache, R., Baig, S., Yan, X., Bourgey, M., Lee, C., Wang, Y., Wee Yong, V., Guiot, M. C., Najafabadi, H., Misic, B., Antel, J., Bourque, G., Ragoussis, J., & Petrecca, K. (2020). Single-cell RNA-seq reveals that glioblastoma recapitulates a normal neurodevelopmental hierarchy. Nature communications, 11(1), 3406. https://doi.org/10.1038/s41467-020-17186-5
El Andaloussi, A., & Lesniak, M. S. (2006). An increase in CD4+CD25+FOXP3+ regulatory T cells in tumor-infiltrating lymphocytes of human glioblastoma multiforme. Neuro-oncology, 8(3), 234–243. https://doi.org/10.1215/15228517-2006-006
Jang, H.S., Shah, N.M., Du, A.Y. et al. Transposable elements drive widespread expression of oncogenes in human cancers. Nat Genet 51, 611–617 (2019). https://doi.org/10.1038/s41588-019-0373-3
Neftel, C., Laffy, J., Filbin, M. G., Hara, T., Shore, M. E., Rahme, G. J., Richman, A. R., Silverbush, D., Shaw, M. L., Hebert, C. M., Dewitt, J., Gritsch, S., Perez, E. M., Gonzalez Castro, L. N., Lan, X., Druck, N., Rodman, C., Dionne, D., Kaplan, A., Bertalan, M. S., … Suvà, M. L. (2019). An Integrative Model of Cellular States, Plasticity, and Genetics for Glioblastoma. Cell, 178(4), 835–849.e21. https://doi.org/10.1016/j.cell.2019.06.024
