Microfluidic Isolation and Identification of True Glioblastoma Stem Cells

Introduction:: Glioblastoma (GBM) is the most lethal and most malignant primary brain tumor, representing 57.3% of all gliomas¹. There are expected to be more than 13,000 GBM diagnoses and more than 10,000 deaths due to GBM in the United States this year. The median survival of GBM patients post-diagnosis is ~15 months and there is a dismal 7.2% 5-year survival rate in the United States. There is also a >90% recurrence rate for GBM patients². GBM recurrence is primarily attributed to the highly drug resistant and tumorigenic glioblastoma stem cells (GSCs)³. These cells, like all classical stem cells, have two basic properties: self-renewal and multi-lineage differentiation⁴. Conventional methods for GSC isolation including fluorescence-activated cell sorting (FACS) and magnetic-activated cell sorting (MACS), are not reliable in isolation of GSCs considering that no known marker is definitive of all GSCs⁵. It was reported previously by our lab that breast cancer stem cells can be isolated in vitro using a biomimetic hydrogel core-hydrogel shell microcapsule with one cancer cell encapsulated per microcapsule (“1-cell” culture)⁶. This work here demonstrates the first isolation of “true” GSCs using 1-cell culture and identification of microRNA (miR) targets for these isolated, true GSCs with miR sequencing. In combination with chemotherapy, this will create a therapeutic strategy for not only destroying GBM tumor but also significantly reducing GBM recurrence, the major cause of GBM-related mortality.


Materials and Methods:: A non-planar microfluidic device was used to produce hydrogel core-hydrogel shell microcapsules with one cell in each microcapsule for 1-cell culture. The device included three inlets for flowing 1% alginate, 0.5% hyaluronic acid, and GBM cells (“core”); 2% alginate (“shell”); and a mineral oil-CaCl₂ emulsion, respectively. The core and shell solutions were made in 0.25 M mannitol and 10 mM HEPES buffer to maintain isotonicity and neutral pH. Microcapsules were monitored for 2-3 weeks to study clone formation percentage. Serial passaging and multi-lineage differentiation (e.g., endothelial, cardiac and undirected) were performed to confirm GSC identity. Undifferentiated and differentiated clones were immunostained for stemness and lineage-specific markers, respectively. Clones were also gathered for miR sequencing and miR targets which are differentially expressed in conventional suspension-cultured GBM cells were identified. Statistical analysis was performed using an unpaired, two-tailed t-test, assuming equal variance.


Results, Conclusions, and Discussions:: One single cell encapsulation for patient-derived GBM cells results in core-shell structured microcapsules with core and overall diameters of ~250 μm and ~350 μm, respectively. A total of ~31% of microcapsules with one cell produce clones of ~223 ± 46 μm in diameter in ~21 days (Fig. 1A). When the clones from 1-cell culture are immunostained for putative GSC markers including CD133 and NESTIN, the expressions of all these proteins are visibly greater compared to the cells from conventional 3D suspension (3Dsus) culture (Fig. 1B). The serial passage assay further confirms the clone cells from 1-cell culture are capable of self-renewal to generate new clones under the 1-cell culture for multiple passages. GSC clones collected from the microcapsules demonstrate differentiation capability into endothelial cells much more efficiently than conventional 3Dsus GBM cells. Tube formation and CD31 expression (Fig. 1C) is more extensive for cells derived by endothelial differentiation from the clone cells than 3Dsus GBM cells. From miR sequencing, several miR candidates, including miRs-196a, 9, 23b, 455, and 590 show differences in expression between high and low-tumorigenicity GBM cell lines and can be potential targets of GSCs for their destruction.



This novel approach for designing and testing a potential GSC-targeted GBM therapy will be the first to target true GSCs. This will also advance knowledge of GSC genotypic and phenotypic properties, including druggable targets that can be scoped in future GBM therapeutic development focusing on eliminating GBM recurrence, the major cause of death due to GBM.










Acknowledgements (Optional): :

References (Optional): : ¹Ostrom et al. (2019). Neuro Oncol 21, 5. ²Weller et al. (2013). Neuro Oncol 15, 1. ³Safa et al. (2015). Genes Dis 2, 2. ⁴Lathia and Liu (2017). Target Oncol 12, 4. ⁵Hassn et al. (2020). Cells. 9, 5. ⁶Wang et al. (2020). Adv Sci 7, 2000259.