Journal of Cancer Stem Cell Research
Background: Despite recent advancements, Glioblastoma multiform remains the most fatal type of brain cancer with median
survival rate of 14.6 months. Glioblastoma stem cells (GSCs) are a malignant subpopulation with self-renewal properties that
resist therapies and cause recurrence. Isolating GSCs from tumor bulk has been challenging, due to lack of appropriate markers.
In this study we evaluate the feasibility of using dielectrophoresis (DEP) to isolate glioblastoma cells with increased stemness.
DEP is a technique that uses a non-uniform electric field to polarize non-charged particles. Polarized particles experience the DEP
force, which is a function of applied electric field as well as physical properties of the particles (i.e. their permittivity and
Cell Culture: Reynold and Weiss  neurosphere assay has become a common method for developing an in vitro model for cells
with increased stemness. Following a similar protocol , we cultured a glioblastoma cell line (U251) as neurospheres. Spheroid
forming cells (SF-U251) and U251 cells cultured under normal conditions, were characterized with immunostaining and western
blot assays. Both subpopulations showed expression of three stem cell markers, CD133, CD44 and Nestin, as well as a glial
marker, GFAP. However, our western blot assays showed increased Nestin expression as well as decreasedGFAP expression for
SF-U251, suggesting that SF-U251 cells have increased stem-like properties.
Microfluidic device: We evaluated the dielectric properties of U251 and SF-U251 in a microfluidic device, described previously
. In our microfluidic platform, cells flow through a channel with an array of 20-micron posts. Maximum DEP force is applied on
cells near the posts. The DEP force can be positive (attracting) or negative (repelling) based on the applied frequency. In the
positive domain, cells get trapped to the posts when the DEP force is equal to or larger than the drag force, exerted by the flow.
Altering the applied electric field, we characterized the trapping range for both U251 and SF-U251 subpopulations. Next
we compared the applied voltages needed for the cells to reach their maximum trapping capacity. Interestingly, we showed that
SF-U251 cells needed higher voltages to reach the maximum trapping. Our results suggest that the two subpopulations are
separable using our microfluidic device.
Conclusion: We developed a malignant subpopulation with increased stemness, following a spheroid culture assay. Both U251
and SF-U251 cells showed expression of stem cell markers. However, western blotting showed increased stemness in SF-U251
cells. Since all four markers were expressed in both phenotypes, separating these two subpopulations using marker-based
sorting methods such as FACS would be challenging.