Introduction

One of the most exciting challenges for basic and applied sciences is measuring cellular properties under noninvasive conditions in real-time. In this respect, impedance spectroscopy represents a promising tool, since this technique fulfills the requirements for a nondestructive, real-time measurement of almost all living cells and tissues (1). For impedimetric recordings, an alternating current can be applied to single cells or complex tissues. The current flows from a working electrode through the medium—cell and/or tissue—to a counter electrode with which the cell itself behaves as a frequency-dependent resistor and capacitor. Depending on the applied frequency, alterations of intra- and extracellular properties or processes (e.g., apoptosis, necrosis, proliferation, and migration of cells) can be measured via the changed cellular impedance (2,3,4,5). However, the bottleneck for applying this technique to a broad scientific and pharmaceutical community is the limited commercial availability of cost-effective, feasible, and reliable chip-based sensors with embedded microelectrodes for impedimetric measurement of living cellular systems. To overcome this problem, we used commercially available multi-electrode arrays (MEAs) and combined these with an already existing impedance analyzer. In principle, MEAs are composed of 60 substrate-embedded microelectrodes, predominantly consisting of noble metals such as gold, titanium, or platinum (i.e., materials that have been commonly used for traditional impedance spectroscopy). Since MEAs have been originally developed for extracellular recording of electrically active cells, all components of the chip are tested for biocompatibility; therefore cytotoxic effects can be excluded (6,7). To test whether MEAs can be used for impedance measurement of cellular alteration, MCF-7 cells derived from human breast cancer cells were cultured as a confluent monolayer on the microelectrodes. MCF-7 cells were chosen since they (i) have already been used for impedance measurements (8) and (ii) represent a well-suited model system for inducing various morphological alterations caused by activation of intracellular signaling pathways, especially in terms of phorbol 12-myristate 13-acetate (PMA)-induced protein kinase C (PKC) signaling pathways (9,10,11,12,13). Here, we demonstrated the successful use of commercially available multimicroelectrode arrays for feasible, reliable, and cost-effective impedimetric measurements of PMA-induced intracellular changes.

Materials and Methods

Design of Multielectrode Arrays

Each MEA (purchased from Multi Channel Systems, Reutlingen, Germany) consists of 60 planar round nanocolumnar TiN microelectrodes with a diameter of 30 µM and an electrode distance of 200 µM (Figure 1, A–C). The microelectrodes are arranged in an 8 × 8 array on a glass substrate and have been passivated with silicon nitrite. A large electrode on the edge of the array serves as counter electrode. To maintain stable culture conditions during measurement, the chip containing the MEA adapter was placed into a CO₂ incubator (Figure 1D).

Figure 1. Multielectrode arrays (MEAs) and measuring setup. (Click to enlarge)

Cell Culture on the Chip

For impedance measurement, 5 × 10⁵ MCF-7 cells were resuspended in 1000 µL culture medium [RPMI, 10% fetal calf serum (FCS), 1% glutamax, 1% penicillin/streptomycin, 1 mM sodium pyruvate, and 10 µg/mL insulin] and transferred into the chip reservoir. Cells were maintained under standard culture conditions (37°C, 5% CO₂, and 95% air). Attachment and growth of cells were monitored by light microscopy and impedance spectroscopy.

Experiments were not started until cultures reached a confluent stage to ensure that almost all 59 electrodes— with the exception of the counter electrode—were covered by cells. Medium was changed every 2 days. Twenty-four hours prior to measurements, medium was replaced by serum-free medium (RPMI, N2 supplement, 1% glutamax, 1% penicillin/streptomycin, 1 mM sodium pyruvate, and 10 µg/mL insulin) to avoid nonspecific effects caused by unknown serum components. For repeated use, cells were removed from the chip by tryptic digestion (0.25% trypsin/0.04% EDTA) overnight and the MEA was sonicated for 10 min in a 2% ultrasonol solution, pH 7.0. Thereafter, MEAs were extensively rinsed with distilled water and sterilized by autoclaving at 121°C for 20 min. Before MCF-7 cells were cultured on MEAs, the chip surface was coated with laminin [5 µg/mL phosphate-buffered saline (PBS)] for at least 2 h at 37°C or overnight, respectively.