Impedance-based Real-time Monitoring of Neural Stem Cell Differentiation

Abstract

We present here the first impedance-based characterization of the differentiation process of two human mesencephalic fetal neural stem lines. The two dopaminergic neural stem cell lines used in this study, Lund human mesencephalic (LUHMES) and human ventral mesencephalic (hVM1 Bcl-X_L), have been developed for the study of Parkinsonian pathogenesis and its treatment using cell replacement therapy. We show that if only relying on impedance magnitude analysis, which is by far the most usual approach in, e.g., cytotoxicity evaluation and drug screening applications, one may not be able to distinguish whether the neural stem cells in a population are proliferating or differentiating. However, the presented results highlight that equivalent circuit analysis can provide detailed information on cellular behavior, e.g. simultaneous changes in cell morphology, cell-cell contacts, and cell adhesion during formation of neural projections, which are the fundamental behavioral differences between proliferating and differentiating neural stem cells. Moreover, our work also demonstrates the sensitivity of impedance-based monitoring with capability to provide information on changes in cellular behavior in relation to proliferation and differentiation. For both of the studied cell lines, in already two days (one day after induction of differentiation) equivalent circuit analysis was able to show distinction between proliferation and differentiation conditions, which is significantly earlier than by microscopic imaging. This study demonstrates the potential of impedance-based monitoring as a technique of choice in the study of stem cell behavior, laying the foundation for screening assays to characterize stem cell lines and testing the efficacy epigenetic control.

Keywords: Dopaminergic neuron; ECIS; Equivalent circuit; IDE; Impedance; Neural stem cell; Stem cell differentiation.

Figure 1

Impedance measurement setup: A) Microelectrode array chip with 12 IDEs; B) zoom-in view of one IDE (the ca. 500 μm x 500 μm opening in the silicon nitride passivation layer appears as a lighter region in the center) ; C) chip holder (the lower plate accommodates a MEA chip and the upper plate provides the 600 μl cell culture chamber and an array of holes for electrical connections using spring-loaded pins; fluid tight sealing on the MEA chip is achieved by using a laser cut silicon rubber gasket); D) Printed circuit board (PCB) of the custom-made 12-channel bipotentiostat (the PCB has an opening to the cell culture vial of the chip holder to allow liquid handling and microscopic visualization; E) user interface of the data acquisition software showing recorded impedance magnitude vs. log frequency.

Figure 2

Cell Index vs. time for LUHMES cells: Initial cell density (cells/cm²) A) 30,000; B) 60,000; C) 120,000. Proliferating (blue) and differentiating (red) cells. Time in days after cell seeding. (Error bars: s.e.m., n = 6).

Figure 3

Fluorescence microscopy images of live stained (Calcein AM) LUHMES cells in growth medium (GM) and differentiation medium (DM). Initial cell density 60,000 cells/cm². Time in days after cell seeding. (Scale bars: 50 μm).

Figure 4

Cell Index vs. time for hVM1 Bcl-X_L cells: Initial cell density 120,000 cells/cm². Proliferating (blue) and differentiating (red) cells. Time in days after cell seeding. (Error bars: s.e.m., n = 6).

Figure 5

Fluorescence microscopy images of live stained (Calcein AM) hVM1 Bcl-X_L cells in growth medium (GM) and differentiation medium (DM). Initial cell density 120,000 cells/cm². Time in days after cell seeding. (Scale bars: 50 μm)

Figure 6

Equivalent circuit models for analysis of impedance spectra acquired A) in the presence and B) in the absence of cells. (Cell specific parameters: R_extra, R_cell, C_cell). For detailed description of the components, see the text.

Figure 7

Example of typical Bode plots for an electrode and the same electrode 48 h after seeding of 60,000 LUHMES cells/cm²: A) impedance magnitude and B) phase angle. Solid lines show the nonlinear least squares fit of the experimental data to the equivalent circuit models of Fig. 6.

Figure 8

Summary of the cell specific equivalent circuit components (R_cell, R_extra, C_cell) for A) proliferating and B) differentiating LUHMES cells (seeding density: 60,000 cells/cm²). Time in days after cell seeding. (Error bars: s.e.m., n = 6).

Figure 9

Summary of the cell specific equivalent circuit components (R_cell, R_extra, C_cell) for A) proliferating and B) differentiating hVM1 Bcl-X_L cells (seeding density: 120,000 cells/cm²). Time in days after cell seeding. (Error bars: s.e.m., n = 6).

Figure S1

Normalized impedance vs. frequency for A) LUHMES (60,000 cells/cm²) and B) hVM1 Bcl-X_L (120,000 cells/cm²). The spectra were acquired 48 h after cell seeding. (Error bars: standard deviation, n = 3)

Figure S2

Cell Index vs. time for proliferating hVM1 Bcl-X_L cells. Initial cell seeding density 60,000 cells/cm² and 120,000 cells/cm². (Error bars: standard deviation, n = 3)