This article describes our systematic investigation into the utility of employing electronic impedance-based cell sensing measurement systems to evaluate changes in cell behavior. We investigated changes resulting from routine passaging (data not shown) and intentional perturbation. Insights gathered from this and other studies present us with new opportunities for introducing advanced levels of cellular characterization and contribute towards development of superior cellular assays at ATCC.
Changes in cellular morphology or growth rate can be early indicators of adverse cellular events. Detection and quantitation of these events with impedance-based cell sensing measurement systems appear to offer improved, real-time, label-free and non-invasive analysis of key cellular events. When adherent cells attach and spread on the sensor surface of an electrode, increases in impedance are recorded. Conversely, cells that round up or detach, even for a short time, cause impedance values to drop. One system well equipped to measure cell-impedance is the xCELLigence System from Roche Applied Science.
Numerous compounds, including toxins and viruses, are known to induce changes in cell morphology and growth. We explored the ability of the Roche xCELLigence System to optimize our current assays and to develop new assays for our studies of toxins and viruses. Our first priority was to establish baseline measurements for Madin Darby Canine Kidney epithelial cells [MDCK (ATCC CCL-34)] and Green monkey kidney cells [Vero (ATCC CCL-81)]. These two cell lines are utilized extensively both within and outside ATCC, including use in vaccine production. Figure 1 illustrates growth curve data generated from individual wells containing Vero cells (Figure 1A) or data plotted as average and standard deviation from replicate wells of MDCK cells (Figure 1B). The changes in impedance are expressed as Cell Index (CI) measurement and the Real-Time Cell Analyzer (RTCA) software (included with the xCELLigence System) permits graphical representation of data plotted as well as an efficient approach to measure growth parameters, including calculating celldoubling times. Our experimentally derived doubling times of 13.7 h and 13.2 h generated using the xCELLigence System and a more traditional approach, respectively, show good correlation. However, the xCELLigence System approach requires far less labor and reagents.
We next confirmed the xCELLigence System's ability to measure early attachment events such as adhesion and spreading (data not shown). Replicate wells were coated with collagen, laminin, matrigel and fibronectin respectively. Measurements were performed every 15 s over a 4-h period in an effort to separate early cell adhesion and spreading from later cell replication events. Fibronectin induced the greatest overall increase in cell spreading and impedance with collagen coating showing the next highest levels. Dose-dependent changes in impedance were observed in wells with varying concentrations of fibronectin with parallel studies supporting our observations that the observed CI changes in these wells were due to morphology changes and not from dramatic differences in the number of adherent cells.
The impact of viral infection on Vero cell CI profiles was investigated in numerous studies including the comparison of profiles generated by known cytopathic effects (CPE)-inducing viruses as compared to those that induce syncytia formation. Figure 2A is an example of growth profiles from cells infected with a syncytia-forming virus (Respiratory Syncytial Virus, RSV) compared to those infected with a CPE-inducing virus (Dengue Virus type 2-New Guinea C, DV). In general, large reductions in CI were seen in most lytic virus infections, and smaller reductions were seen in infections triggering cellrounding and shrinking (data not shown). The increase in the CI in the RSV-infected cells in Figure 2A is consistent with cell fusion and enlargement of cells typically seen in syncytia forming infections. Figure 2B shows a clear dose-dependent impact of Vero cells infected with Dengue Virus type 2 at a multiplicity of infection (MOI) of 5, 2.5, 1.25 and 0.63.
Cytotoxicity is a hallmark effect of a range of reagents including toxins. In an effort to optimize our cytotoxicity testing parameters and to extend our toxin characterization efforts, we employed the xCELLigence System in conjunction with a battery of related assays. These experiments were performed using replicate wells seeded with human adenocarcinoma [Caco-2 (ATCC HTB-37)] cells. Following a 24-h incubation the cells were treated with varying concentrations of Clostridium perfringens enterotoxin (ATCC BTX-120) and measurements were taken every 15 min over 24 h. Figure 3 shows CI tracings normalized to a point prior to addition of toxin and illustrates the impact of several concentrations of Clostridium perfringens enterotoxin (in the nanomolar to picomolar range) on cell behavior. The RTCA software was utilized to convert the CI measurements from numerous wells and a range of time points into EC50 values. Select EC50 results were verified using other standard cytotoxicity assays; some of these were significantly more time consuming and labor intensive.
A natural extension of the work described above were studies designed to assess the unit's ability to rapidly screen for cytotoxicity induced by toxins secreted by select bacterial strains or to explore the susceptibility of a range of cell lines to a particular toxin. We also used it to investigate the ability of an inhibitor to block the toxin's influence on the cells. Together this information will be utilized for assay development, assay optimization and for gaining a better understanding of the functional behavior of our reagents.
The primary goal of this technical assessment was to explore ways in which impedance-based sensing tools could be used to evaluate changes in cellular behavior. Our studies allowed us to evaluate multiple aspects of the system including sensitivity and flexibility. The xCELLigence platform offers dynamic, real-time, label-free and non-invasive analysis of a variety of cellular events. In some scenarios, it offers considerable labor and reagent savings when compared to classical approaches. Our results demonstrate that the xCELLigence System possesses the speed and sensitivity to measure early changes in cell behavior such as cell adhesion and spreading and is able to discriminate dose-dependent changes resulting from virus or toxin addition. The xCELLigence System offers an efficient way to optimize our cell-based assays and provides us with data that is almost impossible to capture using typical endpoint assays. Studies aimed at supporting our preliminary observations and further evaluating this technology for additional screening and characterization applications are ongoing.
Application note sponsored by Roche Applied Science. To see the full length article, visit www.xcelligence.roche.com. ATTC: www.atcc.org. xCELLigence is a trademark of Roche. ACEA Biosciences is a registered trademark of ACEA Biosciences, Inc. in the US. For life science research only. Not for use in diagnostic procedures.
The author would like to thank Matthew Boley, Dr. Fanching Lin, Jessica Shifflett, Lilit Vardanian, and Dr. Melissa Willis for their critical contributions to this project.
