Modelling invasion, migration and cellular networks in Glioblastoma stem cells

Glioblastoma (GBM) is the most common and aggressive brain tumour in adults. The prognosis for the average patient is very poor with median survival times of only 12-15 months. Despite treatment (surgical debulking and adjuvant chemoradiotherapy), glioblastoma tumours inevitably recur, which is attributed to the existence of cancer stem cell subpopulations, known as glioma stem cells (GSC). GSCs are highly infiltrative and have been shown to significantly contribute to tumour growth and recurrence. However, GSCs remain moving targets in terms of their adaptive molecular phenotype which hampers targeted treatment strategies against GSCs. Here we have developed model systems to investigate aspects of adaptive GSC phenotypic behaviour including invasion, epithelial-mesenchymal transition (EMT) and cellular networks, which all contribute to the overall aggressiveness of GBM tumours.

Specifically, we have developed a human-human assembloid assay, whereby patient derived GSC spheroids spontaneously fuse and infiltrate human cerebral organoids (hCOs). This enabled us to dynamically model and quantify the GSC tissue invasion process. Importantly, the assay differentiates between the invasive behaviour of malignant and non- malignant control cell types, also indicating different invasion timelines across different patient-derived GSC models.

Additionally, we have explored conditions that promote the transcriptional upregulation of an EMT profile in GSC subpopulations derived from GBM tumour specimens. These GSCs have characteristics of a mesenchymal GSC subpopulation that warrants further functional investigation. We propose that this previously unidentified cell population may contribute to a more infiltrative tumour phenotype through the EMT pathway.

We have also further validated an inducible cellular network in GSCs using chemical inhibition of Rho associated protein kinase (ROCK). Such an in vitro system recapitulates features of cellular networks discovered in vivo in an accessible manner for the study of functional cell co-operation mechanisms. Specifically, we have determined the ability of networked cells to transfer mitochondria intercellularly and propose this as a mechanism that critically contributes to a radioresistant GSC phenotype.

By modelling the phenotypic plasticity of GSCs, with the ability of cells to invade and infiltrate, demonstrate EMT transcriptional upregulation as well as the formation of cellular networks, we have paved a way for the interrogation of GSC functional biology ex vivo. This is expected to complement time and resource-intense in vivo experimentation in rodents, hence aiding the identification of GSC-directed therapeutic strategies that target key adaptive GSC phenotypes in the treatment of GBM.