Abstract
The overall goal of this work was to assess the ability of Natural Killer cells to kill cultures of patient-derived glioblastoma cells. Herein we report impressive levels of NK-92 mediated killing of various patient-derived glioblastoma cultures observed at ET (effector: target) ratios of 5:1 and 1:1. This enabled direct comparison of the degree of glioblastoma cell loss across a broader range of glioblastoma cultures. Importantly, even at high ET ratios of 5:1, there are always subpopulations of glioblastoma cells that prove very challenging to kill that evade the NK-92 cells. Of value in this study has been the application of ECIS (Electric Cell-Substrate Impedance Sensing) biosensor technology to monitor the glioblastoma cells in real-time, enabling temporal assessment of the NK-92 cells. ECIS has been powerful in revealing that at higher ET ratios, the glioblastoma cells are acutely sensitive to the NK-92 cells, and the observed glioblastoma cell death is supported by the high-content imaging data. Moreover, long-term ECIS experiments reveal that the surviving glioblastoma cells were then able to grow and reseed the culture, which was evident 300-500 h after the addition of the NK-92 cells. This was observed for multiple glioblastoma lines. In addition, our imaging provides evidence that some NK-92 cells appear to be compromised early, which would be consistent with potent evasive mechanisms by the glioblastoma tumour cells. This research strongly highlights the potential for NK-92 cells to kill glioblastoma tumour cells and provides a basis to identify the mechanism utilised by the surviving glioblastoma cells that we now need to target to achieve maximal cytolysis of the resistant glioblastoma cells. It is survival of the highly resistant glioblastoma clones that results in tumour relapse.