154
Chem. Res. Toxicol. 2005, 18, 154-161
Dynamic Monitoring of Cytotoxicity on Microelectronic Sensors James Zan Xing,*,† Lijun Zhu,† Jo Ann Jackson,‡ Stephan Gabos,§ Xue-Jun Sun,| Xiao-bo Wang,‡ and Xiao Xu*,‡ Department of Laboratory Medicine and Pathology, University of Alberta, Health Surveillance, Alberta Health and Wellness, and Cross Cancer Institute, Edmonton, Alberta, Canada, and ACEA Bioscience Inc., San Diego, California Received October 7, 2004
A real-time cell electronic sensing (RT-CES) system was used for label-free, dynamic measurement of cell responses to cytotoxicants. Cells were grown onto the surfaces of microelectronic sensors, which are comprised of circle-on-line electrode arrays and are integrated into the bottom surfaces of the microtiter plate. Changes in cell status such as cell number, viability, morphology, and adherence were monitored and quantified by detecting sensor electrical impedance. For cell quantification and viability measurement, the data generated on the RT-CES system correlated well with those from the colorimetric (MTT) assay. For cytotoxicity assessment, cells growing on microelectronic sensors were treated with different cytotoxicants, such as arsenic, mercury, and sodium dichromate. The dynamic responses of the cells to the toxicants were continuously monitored by the RT-CES system. On the basis of the IC50 values, the RT-CES system displays an equal sensitivity to the neutral red uptake assay at specific time points. Furthermore, because the RT-CES system provides real-time information regarding the state of cell morphology and adhesion in addition to cell number, we were able to discern a previously unreported effect of arsenic on NIH 3T3 cells prior to cell death. Also, using the RT-CES system, we were able to monitor cytotoxicity effects that occur within a minute of compound addition. Taken together, the RT-CES system allows for realtime, continuous monitoring and quantitative recording of the whole assay process and provides new insight into the cell-toxicant interaction.
Introduction In vitro cytotoxicity assays are common alternatives to conventional animal testing in safety and hazard assessments (1-8). However, a major challenge for predicting biologic outcomes using in vitro cell models is the interactions of living cells with specific toxicants that may conditionally alter cell function in a nonlinear fashion. Upon exposure to toxic compounds, cells undergo physiological and pathological changes, including morphological dynamics (9, 10), an increase or decrease in cell adherence to the extracellular matrix (12-14), cell cycle arrest (15-17), apoptosis due to DNA damage (1518), and necrosis (19, 20). Such cellular changes are dynamic and depend largely on cell types, the nature of a chemical compound, compound concentration, and compound exposure duration (9-11). In addition, certain cellular changes, such as morphological dynamics and adhesive changes, which may not lead to ultimate cell death, are transient and occur only at early or late stages of toxicant exposure (9, 10). Measuring and modeling such diverse information sets are difficult at the analytical level, using the conventional end point cytotoxicity * To whom correspondence should be addressed. (J.Z.X.) Tel: 780492-9250. Fax: 780-492-9249. E-mail:
[email protected]. (X.X.) Tel: 858-724-0928. Fax: 858-724-0927. E-mail:
[email protected]. † University of Alberta. ‡ ACEA Bioscience Inc. § Alberta Health and Wellness. | Cross Cancer Institute.
assays, since current conventional cytotoxicity assays can only detect very specific cellular changes such as viability, occurring after certain durations of toxicant exposure (911), which provides no dynamic information with respect to living cells in response to toxicants. Thus, to accurately assess toxicant-induced cellular damage and to understand the mechanism of action for toxic compounds, it is pertinent to examine multiple parameters in the same assay under dynamic conditions. To develop a cell-based assay system allowing for dynamic detection of a broad range of physiological and pathological responses to toxic agents in living cells, we designed a novel electrical impedance sensor array, termed circle-on-line electrode sensors. In contrast to previous cell impedance sensors, the new sensors, which are integrated onto the bottom of a standard microtiter plate, cover the 80% surface area of the well, allowing for more sensitive and quantitative detection of living cells. On the basis of the design of the new sensor, a realtime cell electronic sensing (RT-CES) system was developed. The basic principle of the RT-CES system is similar to the electric cell impedance detection systems described previously (21, 22). Electronic impedance of an electrode is primarily determined by the ion environment both at the electrode/solution interface and in the bulk solution (21, 22). Upon the application of an electrical field, ions undergo field-directed movement and concentration gradient-driven diffusion, leading to frequency-dependent
10.1021/tx049721s CCC: $30.25 © 2005 American Chemical Society Published on Web 02/04/2005
Dynamic Monitor Cytotoxicity by Electronic Sensor
Chem. Res. Toxicol., Vol. 18, No. 2, 2005 155
improved sensitivity and high information contents for cytotoxicity measurements performed on the RT-CES system are demonstrated.
Experimental Procedures RT-CES System. The RT-CES system (ACEA Biosciences, San Diego, CA) has three components: an electronic sensor analyzer, a device station, and 16× sensor device with 16 plastic wells with 9 mm well-to-well spacing. The circle-on-line electrode sensors (Figure 1a), specially designed and optimized to cover about 80% of the bottom surface area of a well, were fabricated on glass slides with lithographical microfabrication methods, and the electrode-containing slides were assembled to sensor devices so that the electrode sensor array formed the well bottom (Figure 1a). The device station received the 16× devices and was capable of electronically switching any one of the wells to the sensor analyzer for impedance measurement. In operation, the devices with cultured cells were mounted to a device station that was placed inside a CO2 incubator. Electrical cables connected the device station to the sensor analyzer. Under the RT-CES software control, the sensor analyzer can automatically select wells to be measured and continuously conduct measurements on wells. The electronic impedance can then be transferred to a computer and plotted. In this report, all of the experimental data were generated on the 16× sensor device, which has the same sensor well format as the 96× sensor device. CI. A parameter termed CI is derived to represent cell status based on the measured electrical impedance. The frequencydependent electrode impedance (resistance) without or with cells present in the wells is represented as Rb(f) and Rcell(f), respectively. The CI is calculated by
CI ) max
i)1,...,N
Figure 1. Microelectrode array device and cell electronic sensing. (A) Sixteen× sensor devices with 16 circle-on-line electrode array units are fabricated on glass slides with lithographic microfabrication methods, which represent 16 individual detection units with 9 mm as the center to center spacing. An enlarged unit of the circle-on-line electrode array is shown in the top panel. (B) A schematic illustration of the principle of cell detection on the sensor. The presence of cells affects the local ionic environment at the electrode/solution interface, leading to an increase in the electrode impedance. If more cells attach to the electrodes, there will be a large increase in electrode impedance, leading to a large CI. The CI decrease, when cells die off, resulted from the exposure to a toxicant.
impedance dispersion. The presence of the cells will affect the local ionic environment at the electrode/solution interface, leading an increase in the electrode impedance. The more cells there are on the electrodes, the larger the electrode impedance (Figure 1b). Furthermore, the impedance also depends on the extent to which cells attach to the electrodes. For example, if cells spread, there will be a greater cell/electrode contact area, resulting in larger impedance. Thus, the cell biological status including cell viability, cell number, cell morphology, and cell adhesion will all affect the measurement of electrode impedance that is reflected by cell index (CI) on the RT-CES system. Therefore, a dynamic pattern of a given CI curve may indicate sophisticated physiological and pathological responses of the living cells to a given toxic compound. In this paper, the RT-CES system is applied for realtime, quantitative detection of environmental toxicants, such as arsenic, mercury, and sodium dichromate. The
[
Rcell(fi) Rb(fi)
-1
]
where N is the number of the frequency points at which the impedance is measured. Several features of the CI can be derived (Figure 1b): (i) Under the same physiological conditions, if more cells attach onto the electrodes, the larger impedance value leading to a larger CI value will be detected. If no cells are present on the electrodes or if the cells are not well-attached onto the electrodes, Rcell(f) is the same as Rb(f), leading to CI ) 0. (ii) A largeRcell(f) value leads to a larger CI. Thus, CI is a quantitative measure of the number of cells attached to the sensors. (iii) For the same number of cells attached to the sensors, changes in cell status will lead to a change in CI. For example, an increase in cell adhesion or cell spread leading to a larger cell/electrode contact area will lead to an increase in Rcell(f) and a larger CI. On the other hand, cell death or toxicityinduced cell detachment or cells rounding up will lead to a smaller Rcell(f) and thus smaller CI. To initiate an experiment, the target cells are loaded into each well of the sensor devices to allow attachment and growth on the sensor. To ensure a consistent initial cell condition for cytotoxicity assays, the toxic compounds were added after CI reached a range of 1-1.2, dependent on cell types, indicating about 40-50% cell confluence. The sensor devices with cells and toxicant were then mounted back to device stations placed inside a CO2 incubator. The CI was automatically and continuously monitored by the RT-CES system. Chemical Compounds and Reagent Kits. Sodium arsenite [As(III)], mercury(II) chloride, sodium dichromate [chromium(VI)], and neutral red solution (0.33%) were purchased from Sigma-Aldrich (Milwaukee, WI). A stock solution and dilutions of As(III) were prepared as previously described (23). Mercury(II) chloride and chromium(VI) were dissolved in distilled water and filtered (0.22 µM filter). The freshly prepared solutions were added to cultures at the indicated concentration. The MTT CellTiter 96 nonradioactive cell proliferation assay kit and LDH CytoTox-ONE Homogeneous Membrane Integrity
156
Chem. Res. Toxicol., Vol. 18, No. 2, 2005
Assay kit were purchased from Promega (Madison, WI). The florescent dyes rhodamine-phalloidin and Hoechst 33258 were purchased from Molecular Probes (Eugene, OR). Cell Lines and Culture Conditions. All of the cell lines (NIH 3T3, BALB/c 3T3, and CHO-K1) were purchased from the American Type Culture Collection (ATCC). Cell culture media (Dulbecco’s modified Eagle media, DMEM, and Ham’s F12 media, F12), penicillin and streptomycin, fetal bovine serum (FBS), and new born calf serum (NBCS) were purchased from Gibco (Invitrogen, Builington, ON, Canada). NIH 3T3 cells and CHO-K1 cells were cultured with DMEM media or Ham’s F12 medium, respectively, containing 10% FBS, 100 µg/mL penicillin, and 100 units/mL streptomycin at 37 °C and 5% CO2. BALB/c 3T3 cells were cultured with DMEM media with 10% NBCS, 100 µg/mL penicillin, and 100 units/mL streptomycin at 37 °C and 7.5% CO2. Cell Culture and Cell Quantification. The cell culture condition on the 16× sensor device is the same as that in a tissue culture flask as described above. Because the sensor device format is compatible to a standard microtiter plate, the starting cell number and cell culture medium volume used for the 16× sensor device were similar to the cell culture in the standard 96 well microtiter plate (3). For real-time calculation of CI, based on impedance values as described above, the background of impedance for each sensor well was measured in the absence of the cells. To do so, 50 µL of cell culture media was added into each well for the baseline measurement, followed by addition of the cells to the sensor wells. Once the cells were added to the sensor wells, the sensor devices were placed into the incubator and the real-time CI data acquisition was initiated by the RTCES analyzer under the control of the RT-CES software. To test cell quantification on the RT-CES system, a cell titration experiment was performed using NIH 3T3 cells. After the background of impedance was measured, 100 µL of the cell suspensions, containing 125, 250, 500, 1000, 2000, 4000, 8000, and 16000 cells, respectively, were added into each well of 16× sensor devices. The sensor devices were placed into the CO2 incubator, and CI values were determined every 30 min automatically by the RT-CES system for up to 6 h. In a parallel experiment, the same cell titration was assayed using a tetrazolium compound-based colorimetric method (MTT assay, Promega Celltiter 96) following the manufacturer’s protocol. Cytotoxicity Assays. Three environmental toxicants, As(III), mercury(II) chloride, and chromium(VI), were used for cytotoxicity assessment on the 16× sensor device. Three cell lines, namely, NIH 3T3, BALB/c 3T3, and CHO-K1, were tested. The starting cell numbers were 10000, 10000, and 15000 cells per sensor wells, respectively, for NIH 3T3, BALB/c 3T3, and CHO-K1 cells. The cell growth on the sensor device was monitored in real time by the RT-CES system. When the CI values reached a range between 1.0 and 1.2, the cells were then exposed to either As(III), mercury(II) chloride, or chromium(VI) at different concentrations. After compound addition, the cell responses to the compounds were continuously and automatically monitored every hour by the RT-CES system for up to 48 h. The IC50 value at a given time point was calculated based on the concentration producing 50% reduction of CI value relative to solvent control CI value (100%) and expressed as mmol/L. The mean and standard deviation (SD) of IC50 were calculated based on the data obtained from three separate experiments, and six replicas were performed for each concentration dose. IC50 values determined using RT-CES systems were compared with those determined by MTT, LDH, and neutral red uptake (NRU) assays (3), which were performed in parallel with the RT-CES system. To monitor the fast cell killing on the RTCES system, NIH 3T3 cells cultured in the 16× sensor device were treated with the IC50 and high concentrations (3-fold higher than IC50 values) of mercury(II) chloride and 0.1% Triton X-100. The cell response to the compound treatment was then immediately and continuously recorded by the RT-CES system at a time interval of 1 min for 30 min.
Xing et al. Cell Morphology Analysis by Confocal Microscopy. Cell morphological dynamics after As(III) treatment was examined by confocal microscopy. Cells were seeded in the 16× sensor device and treated with As(III). Hoecsht was added and incubated with the test cells for 1 h ahead of each collecting time point. The treated cells were then collected at different time points and fixed with 4% parafarmaldehyde for 30 min. The fixed cells were washed 3× with PBS, permeablized in PBS containing 0.2% TX-100, and blocked in PBS containing 0.5% BSA. The cells were stained with rhodamine-phalloidin for 30 min, washed 3× with PBS, and visualized and imaged with a Zeiss LSM510 NLO multiphoton Laser Scanning Microscope that was mounted on a Zeiss Axiovert 200 M microscope.
Results and Discussion Principle of the RT-CES System and Improved Cell Quantification. To determine if the CI value obtained on the RT-CES system is quantitatively correlated with the cell growth and cell numbers, NIH 3T3 cells were titrated and seeded on the electrode sensor surface (Figure 2, top panel). The CI values as determined on the RT-CES system were found to be linearly correlated with cell numbers over the range between 125 and 16000 cells, having a correlation coefficient of 0.995 (Figure 2, bottom left panel). The variation in CI as measured over three replicates of appropriate cell numbers and CV values ranged from 1.4% for high cell numbers to 24.8% for 125 cells (Figure 2, middle panel). The cell titration assay with the same range of cell numbers was also performed using the MTT assay. The MTT assay had a correlation coefficient of 0.996 (Figure 2, bottom right panel) and was similar to those obtained from the RT-CES system. This indicates that for viable cell counting and quantification, the data generated on the RT-CES system are consistent with that obtained for the MTT assay. The major improvement in the RT-CES system over previous impedance technique such as electric cell impedance sensing (ECIS) is the development and the use of the circle-on-line microelectrode array that covers about 80% of the area on the bottom of a well (Figure 1a). In the ECIS system, small detection electrodes [
