Electric Cell-substrate Impedance Sensing for the Quantification of Endothelial Proliferation, Barrier Function, and Motility

Summary

This protocol reviews Electric Cell-substrate Impedance Sensing, a method to record and analyze the impedance spectrum of adherent cells for the quantification of cell attachment, proliferation, motility, and cellular responses to pharmacological and toxic stimuli. Detection of endothelial permeability and assessment of cell-cell and cell-substrate contacts are emphasized.

Abstract

Electric Cell-substrate Impedance Sensing (ECIS) is an in vitro impedance measuring system to quantify the behavior of cells within adherent cell layers. To this end, cells are grown in special culture chambers on top of opposing, circular gold electrodes. A constant small alternating current is applied between the electrodes and the potential across is measured. The insulating properties of the cell membrane create a resistance towards the electrical current flow resulting in an increased electrical potential between the electrodes. Measuring cellular impedance in this manner allows the automated study of cell attachment, growth, morphology, function, and motility. Although the ECIS measurement itself is straightforward and easy to learn, the underlying theory is complex and selection of the right settings and correct analysis and interpretation of the data is not self-evident. Yet, a clear protocol describing the individual steps from the experimental design to preparation, realization, and analysis of the experiment is not available. In this article the basic measurement principle as well as possible applications, experimental considerations, advantages and limitations of the ECIS system are discussed. A guide is provided for the study of cell attachment, spreading and proliferation; quantification of cell behavior in a confluent layer, with regard to barrier function, cell motility, quality of cell-cell and cell-substrate adhesions; and quantification of wound healing and cellular responses to vasoactive stimuli. Representative results are discussed based on human microvascular (MVEC) and human umbilical vein endothelial cells (HUVEC), but are applicable to all adherent growing cells.

Introduction

ΩΩΩHere we present Electric Cell-substrate Impedance Sensing, known as ECIS, a specific method to measure and analyze the impedance spectrum of adherent cells in culture¹. The aim of this protocol is to offer a generally applicable guide for the usage of this particular type of impedance based cellular assays and provide protocols for some of the key functions from the constantly growing number of applications. The focus will be on the study of cell proliferation, barrier function, cell junctions, and cell motility.

Since ECIS and its associated model to transform the impedance spectroscopy data in biologically relevant parameters was introduced in its current form to the scientific community by Giaever and Keese in 1991², it has often been referred to as a system for the measurement of TEER (trans-epithelial electrical resistance), which is not accurate. The differences seem marginal at first, but are important for the data interpretation. For classical TEER measurements, cells are grown on permeable filters to characterize paracellular transport mechanisms, which are dominated by epithelial tight junctions or endothelial adherens junctions³. Commonly, two electrodes positioned above and below the filter are used to apply a direct current (DC) flow over the cell layer and two other electrodes to measure the resulting voltage drop⁴. The electrical resistance is calculated using Ohm's law, which allows a numerical description of the quality of the cell barrier.

ECIS follows this basic principle and extends it. In the ECIS system, cells are grown on opposing, circular gold electrodes that are embedded in the bottom of special cell culture dishes. The number of electrodes per culture well is variable, dependent on the application and the electrodes have a standard diameter of 250 µm; in some cases a larger counter electrode is used to complete the circuit. ECIS uses a constant alternating current (AC) of 1 µA with a given frequency instead of a direct current. The impedance is calculated from the corresponding changes in voltage (in mV) between electrodes. ECIS offers the possibility to measure the impedance over a range of frequencies to study frequency dependent cellular properties, which has several advantages over TEER and will be explained in detail in this article. First, measuring complex impedance allows separating the overall impedance into cell barrier resistance and cell capacitance. In addition, by taking data at multiple frequencies and applying a mathematical model, one can differentiate between junctional impedance (tightness of cell-cell contacts) and impedance caused by cell-substrate interactions (distance of basal cell membrane to underlying matrix) as well as the contribution of the cell membrane capacitance. Second, cell proliferation and motility can be assessed, since the cells are in direct contact with the electrodes. Third, substrate and electrodes are sufficiently thin to allow for bright field and phase contrast microscopy.

Basis of impedance measurements: The complex impedance

The basis for the measurement of the electrical impedance of biological objects (e.g. cells) is Ohm's law, a basic electro-technical principle, which describes the relation between resistance (R), current (I) and voltage (U) in an electrical circuit at a given time (t).
Applicable in DC circuit: R(t) = U(t)/I(t)

When working in the AC system, current and voltage not only differ in their amplitude, but also in their phase (φ). Now, resistance is no longer sufficient to describe these relations. Instead, the complex impedance (Z) or in most cases the magnitude of the impedance (|Z|) are used, containing the previously described ohmic resistance plus reactance (X), which results from AC flow through capacitors and inductors driving the phase shift between voltage and current⁵.
Applicable in AC circuit: |Z(f)| = √(R²+X(f)²)
φ = arctan(X/R)

When performing impedance measurements on intact cells, due to the characteristics of their membrane, cells act as a parallel connection of resistor and capacitor. Here, resistance represents the opposition to current flow, whereas capacitance (C) describes the separation of electric carriers at the insulating bi-layer of the cell membrane that causes polarization of the cell. Thereby X is dominated by the capacitive properties of the cell membrane.
X(f) ≈ (2*Pi*f*C_Cell)^-1

Since X is frequency dependent, variation of the measurement frequency enables study of different functional and structural properties of the cell. The ECIS device measures both R and X, allowing calculation of |Z|, C and φ.

Quantifying entire cell layers with impedance spectroscopy: The electrical equivalent circuit.

As previously explained, when a cell is brought into an electrical field, it shows properties of passive electronic components. If now, instead of a single cell, an entire cell layer grown on top of electrodes and supplemented with cell culture medium is investigated, the simple model of resistor and capacitor needs to be extended to an entire electrical network. In this so-called equivalent circuit, resistance of the culture medium (R_Med) as well as capacitance (C_Electr) and resistance (R_Electr) characterizing the electrode/electrolyte interaction need to be considered^3,6.

A simplified, general example of such an equivalent circuit for an adherent growing cell layer can be found in Figure 1. The advantage of such a mathematical approach to describe a biological system is that those circuits can be refined ad libitum and adjusted to the specific experimental questions, e.g. by considering impedance caused by intra-cellular organelles or to distinguish influences of cell-cell (R_Junc) and cell-substrate adhesions (R_Sub) on overall impedance^7,8. Nevertheless the aim for the modeling should always be to use the smallest number of elements describing all features of the measured impedance spectrum to allow meaningful correlations.

Figure 1. Schematic of the ECIS system and representative equivalent circuit for an adherent growing cells layer. A) Cross section of an ECIS culture well. The cells are growing on top of sensing and counter electrode and are covered with culture medium. The electrodes are connected to a lock-in amplifier and an AC signal is applied via a 1 MΩ resistor to create a constant current source. Stimuli can be added directly to the culture medium at any point in time. B) ECIS measures the sum of all individual contributions to the impedance. Resistance of the culture medium (R_Med) as well as impedance caused by the electrode/electrolyte interface, which is for simplicity presented as a parallel combination of a resistor (R_Electr) and a capacitor (C_Electr), and also the electrical properties of the cell membrane, described by a parallel connection of resistance (R_Cell) and capacitance (C_Mem), all need to be considered. R_Cell is variable, since it is dependent on the cell permeability towards the current. The equivalent circuit can be extended and refined ad libitum. As an example junctional (R_Junc) as well as subendothelial (R_Sub) resistance were added to the circuit. Please click here to view a larger version of this figure.

Impedance parameters and their biological meaning

The two most direct parameters derived from impedance measurements are resistance and capacitance of cells. Resistance represents quality and function of the cell barrier and therefore takes into consideration the resistance towards para- and trans-cellular current flow. Capacitance provides an overall measure of electrode coverage. The distinctive feature of the ECIS is that with the help of equivalent circuits and modeling those global parameters provide insights on many more cellular properties, including cell-cell and cell-substrate adhesions, which will be discussed later in this article.

Before starting: Experimental considerations

Measurement setup - The setup consists of several separate components: ECIS device with measurement electronics; PC for data acquisition; array holder for the 8- or 96-well system; ECIS arrays and the cell culture of choice. The array holder must be placed in an incubator and connected to the ECIS device outside the incubator. The PC needs to be equipped with the ECIS software (1.2.123.0 14th February 2013) and connected to the ECIS device.

Array selection - There is a continuously growing variety of ECIS arrays, designed for multiple applications. The standard arrays are the 8W1E and the 8W10E arrays, which are composed of 8 culture wells (indicated by W) comprising 1 or 10 measurement electrodes (indicated by E), respectively. A large counter electrode completes the circuit, but its impedance is essentially negligible in the actual measurement⁶. The standard 8-well array holder can host two arrays, resulting in a total number of 16 culture wells. The gold electrodes are 50 nm thick, delineated with an insulating film and mounted on either an optically clear Lexan polycarbonate substrate or a printed circuit board (PCB). The PCB arrays are more robust and cost efficient. The transparent slides allow for light and immunofluorescence microscopy. What must be considered is that the 1E array enhances fluctuations in the resistance signal caused by cell motions and is needed for wound healing studies. In addition, single electrodes allow correlation of electrical and optical signals. In the multi electrode arrays, the signal is averaged over several electrodes, which due to the increased measurement area includes more cells in the measurement, limits bias of the data by uneven inoculation and growth of the cells and reduces blurring of the signal due to cell motions. Therefore, the multi electrode arrays are useful to study cell proliferation and barrier formation. Next to standard arrays there are special arrays available for the application of shear stress⁹, to study chemotaxis¹⁰, cell migration, and proliferation as well as 96-well plates for high throughput screenings. To conclude, the array to be used is strongly dependent on scientific question and cell type and should be selected and tested carefully.

Measurement frequency - The modeling of Rb and alpha (see data analysis) requires multi frequency measurements (MFT). Otherwise impedance can be measured over time at one cell type specific frequency (SFT), with the advantage that data can be collected with a higher temporal resolution. The most sensitive measurement frequency for a specific cell type can be found by frequency scans. When plotting impedance respectively resistance vs. frequency in a log-log graph the frequency where the difference between cell-free and cell-covered electrode is biggest is the frequency where the cells block the current most effectively. In case of endothelial cells (EC) this frequency is at about 4 kHz.

Seeding density - As in every regular cell-based experiment seeding density depends on the scientific problem. When studying adhesion and spreading or barrier formation, endothelial cells should be seeded with a high density of 40,000-60,000 cells/cm² to guarantee a confluent, stable barrier after 48 hr. If the focus of the experiment is on proliferation, endothelial cells should be seeded with a low density of about 2,000-10,000 cells/cm².

Protocol

1. Preparation of Measurement System

1. Prewarm the array holder in an incubator to 37 °C before starting the experiment to prevent condensation.
2. Fill water casket of the incubator with distilled water to avoid drying of the wells.
3. Perform all manipulations on cells and arrays in a laminar flow cabinet under sterile conditions.

2. Preparation of Arrays and Inoculation of Cells

NOTE: The ECIS arrays need to be cleaned and stabilized before an experiment to prevent electrode drift, to improve well-to-well reproducibility and signal-to-noise ratio. Therefore two options are available: A) pretreatment of the electrodes with L-cysteine and B) the stabilization function of the ECIS.

Option A: L-cysteine is recommended as the most consistent pretreatment, but might in some special cases interact with protein coatings on the electrodes. The volumes presented are used for standard 8-well arrays.

1. Add 200 µl/well 10 mM L-cysteine. Cysteine cleans and modifies the electrode surface providing a high and reproducible electrode capacitance.
2. Incubate 15 min at room temperature (RT).
3. Wash twice 500 µl/well with ultra-pure water (type 1). Do not use phosphate buffers, which can interfere with the adsorption of some proteins. Also avoid serum-containing solutions before coating, since the gold surface will adsorb the first macromolecules brought in contact.
4. Coat with 200 µl/well warm 1% gelatin.
5. Incubate 30 min at 37 °C.
6. Remove gelatin, but don't let surface dry. This can damage the electrodes.
7. Wash once with ultra-pure water (type 1).
8. Add 400 µl/well complete cell culture medium (containing serum, antibiotics and all growth factors).
9. Slide array into array holder and make sure that the array is positioned such that the nine gold square pads are properly contacted by the spring-loaded pogo pins and fix adjustment screw hand tight. Be careful with transparent arrays, the gold layer can easily be damaged.
10. Open the measurement software and press Setup followed by Check in the "Collect Data" section to perform a quick impedance measurement of each individual well. The resulting values are the measurement baseline and will automatically be stored in the "Comments".
11. The well check will automatically determine the type of ECIS arrays used. Verify the selected arrays in the "Well Configuration" section.
12. The array diagram lights up green for properly connected wells and red for empty or incorrect connected wells. Still, always make sure that the arrays were placed correctly in the holder.
    NOTE: If cleaning and coating were successful the 8W1E ECIS arrays have a capacitance of about 5 nF and the 8W10E arrays of 50 nF, which results in a baseline resistance of about 2,000 and 200 Ω, respectively.
13. Remove array from holder.
14. Seed cells as a single cell suspension in 400 µl/well complete cell culture medium.

Option B: The stabilization function of the ECIS can be used as an alternative or in combination with the L-cysteine treatment and if problems with electrode drift or large differences in well-to-well resistance and capacitance occur.

1. Perform cysteine treatment (optional) and precoat electrodes with protein (optional).
2. Add 200 µl/well of complete medium to each well and place the array in the array holder.
3. Open measurement software and press Setup followed by Check to ascertain the array is properly connected and to measure initial Z, R, and C.
4. Stabilize the electrodes by pressing Stabilize in the "Collect Data" section. The ECIS will apply a high current (ca. 3 mA), high frequency (64 kHz) pulse to clean the electrodes.
5. Check electrodes again. Capacitance should be increased and resistance values should be in a comparable range.
6. If capacitance and resistance values are still not satisfactory, repeat the stabilization.
7. Remove array from the holder and inoculate cells.

3. Setting up Measurement Software and Data Acquisition

1. Place the array in the array holder as described before.
2. Open measurement software and press the Setup button in the "Collect Data" section to connect the ECIS software to the ECIS instrument and run another well Check.
3. Verify the selected arrays in the "Well Configuration" section.
4. Select the wells to be measured in the "Well Configuration" section.
5. Select the measurement mode from the "Collect Data" options:
   1. For a time series at a fixed frequency, select Single Freq. / Time (SFT) and choose the measurement frequency from the drop down menu.
   2. For measurement of electrode coverage and modeling of Rb and Alpha, select Multiple Freq. / Time (MFT). The ECIS device will measure automatically at all available frequencies.
   3. For micromotion (see data analysis) select Rapid Time Collect (RTC) and adjust measurement frequency and sampling frequency. Here the standard temporal resolution is one sample per sec (1 Hz), but sampling frequency can also be increased to track fast changes in impedance up to 25 Hz (for the Z-Theta). Important note: in RTC mode only a single well is measured.
      NOTE: If micromotion need to be measured in several wells subsequently, go to Help > Show expert toolbar/menu items > Acquire > Multi-Well RTC. Specify the Time limit (hr) in the Collect Data section; the ECIS will now measure the selected wells in numerical order. After starting the measurement the software will ask to specify the numbers of cycles. Enter the value for how often the measurement needs to be repeated.
6. For SFT and MFT, select the time Interval (sec) between measurements or if data need to be acquired as fast as possible leave this option unchecked.
   NOTE: To acquire data the ECIS device uses a multiplexer to switch from one well to another with a minimum SFT acquisition rate of 0.25 sec and 7.5 sec for MFT. Meaning the interval is dependent on number of selected wells and frequencies. For measurements on slow proliferating cells or a simple recording of resistance vs. time, use intervals of 600 sec or more.
7. Press Start, name the file and select location to store the data.
   NOTE: File size should not be a problem, since ECIS data are saved in a kind of plain text format called *.abp, which never exceeds several hundred MB even with all measurement frequencies and number of wells selected.
8. Stop run by pressing Finish.

4. Medium Change and Cell Manipulation

1. Press Pause, this will stop the data acquisition, but continue to run the experimental clock.
2. Remove the array from the holder and manipulate it under a laminar flow bench (i.e. change medium).
3. Place array back in the holder and either click Check Connection or Resume Experiment to continue data acquisition. The measurement software will place a time mark in the data set.

5. Acute Stimulation During Data Acquisition

NOTE: A second option is to manipulate the cells during data acquisition to follow immediate changes. This is only possible when samples don't needs to be sterile in the further course of the experiment.

1. Add the stimulus directly to the well and make sure to mix careful with a 200 µl pipette.
2. Mark time point manually by pressing Mark and add a comment.

6. Electrical Wounding

NOTE: Finding the right settings for the used cell type is key for electrical wounding. If wounding is too short this can result in insufficient removal of cells, whereas if wounding is too long and/or rough this can damage the electrodes (especially on the transparent arrays). Therefore, several short wounding pulses are recommended. For HUVEC cultures two 10-20 sec long wounding pulses of 5 V at 60 kHz each have been found optimal using the ECIS 1600R. In general it is recommended to use high frequencies >20 kHz to maintain uniform high fields across the cells and to avoid any damage to the electrodes.

1. Perform wounding during an active ECIS measurement.
2. Enable Wound / Electroporate Setup, then click Wound and fill in the time (sec), wound voltage (V for 1,600 R) or current (µA for Z and Z-Theta) and frequency (Hz). The displayed default values are based on the type of array and ECIS device.
3. Select the wells to wound in the "Well Configuration" section. Only the checked wells will be wounded.
4. Enable Delay Until Hour and fill in the delay time to activate the delay wound function. This function can be stopped at any time by pressing Stop or applying a wound manually. Only one delay command can be active at a time.
5. Press Activate and click OK in the pop-up window to start wounding.
6. Make sure that the resistance signal drops to the offset impedance after wounding, otherwise repeat the wounding.

7. Data Analysis

1. Data representation and exporting.
   1. Finish measurement or load data set via File -> Open.
   2. Select wells to be displayed from the "Well Configuration" window. Group individual wells to present the average.
   3. Chose parameter to be displayed from the upper toolbar (Z, R or C).
   4. Select graph type from the upper toolbar (T = time, F = frequency, 3D = waterfall blot).
   5. Define graph offset and range with the sliders below the graph window or fill in the exact values in the adjacent controls.
   6. Chose from the basic "Time-Series Options" in the "Analyze" section to perform standard deviation, running average or normalization of the data.
   7. Activate Expert Toolbar in the Help menu for more advanced operations like Nyuist plot and Fourier transformation.
   8. Use "Display Options" in the "Analyze" section to define range of x- and y-axes and export graph.
   9. Export selected data to an Excel file via File -> Export Data -> To Excel (Selected) and use a third party software of choice for in depth analysis, statistics and representation as needed.
2. Modeling of Rb and Alpha.
   NOTE: For modeling it is recommended to use a cell-free, medium filled well as reference.
   1. Finish MFT measurement or load MFT data set.
   2. If a cell-free reference well is available press either Find to automatically select the reference value or Set to select the cell free time point manually from "Freq. Scan Modeling" in the "Analyzing" section.
   3. If there is no cell-free reference, import a reference value from another data set. Open the reference data set (make sure that the same array and medium was used) and use Find or Set to select the reference value. Then navigate to the measured data and press Get.
   4. If there is no cell-free reference data set available use the factory default.
   5. Press Model to start the calculations. Modeling might take several minutes dependent on the size of the data set.
3. Calculation of micromotion.
   NOTE: This calculation is not part of the measurement software, but can be performed with any signal processing software.
   1. Finish RTC measurement or load RTC data set.
   2. Normalize data set by the averaged resistance of the entire period.
   3. Break data set into segments with a length of 256-data points and clean each segment by a Hanning window.
   4. Run a 256-point Fourier transformation and average resulting frequency spectra to a single-sided 129-point power spectrum to minimize random noise and enhanced the biological signal.
   5. Plot frequency against amplitude in a log-log graph and apply least square linear fit.
   6. The slope of the linear fit is a measure of overall noise in the signal and thereby micromotion of the cells.

Discussion

ECIS is an excellent tool for the screening of cell properties and behavior as well as for the quantification of the effects of known and unknown substances. Thereby the cells are held under standard culture conditions, impedance can continuously be monitored with a high temporal resolution and correlated to optical signals. That way the optimal time point for cell manipulations can be chosen on the basis of the morphological and functional cell status. Unfortunately, this high measurement resolution comes with the price that small changes in temperature, pH or mechanical stimulation of cells (medium change) will influence the impedance signal immediately.

The application of the small measurement current to the cells makes the ECIS measurement noninvasive, nondestructive, and label-free, but as a result only passive bioelectrical properties can be measured (no action potentials). A key feature is that a number of parameters can be derived from a single measurement, combining the information from several classical assays, like permeability or wound healing assays. Here the particular interesting aspect is that mathematically modeled data can be used to explore changes in resistance and capacitance and refer them to distinct cellular structures (e.g. cell-contacts or cell membrane). Important to note is that impedance spectroscopy always provides an averaged signal from all cells on the sensing electrode, which does not allow for studies on single cells and also the mathematical model is only valid in confluent cell layers. Therefore endothelial cells should be held in the confluent state for at least one day before used for modeling to ensure mature cell-adhesions and quiescent cells. Equally, electrical wounds should only be applied to confluent cell layers using multiple short wounding pulses with high frequencies, to achieve optimal wounding efficiency and prevent damage of the electrodes.

To get the maximal amount of information from an ECIS measurement, as in every assay, several parameters such as the combination of array substrate, coating and seeding density for the individual cell type need to be tested and optimized before an experiment.

A major limitation of ECIS is that the measurement does not provide direct information on the molecular level. Thus ECIS measurements are usually most informative at the beginning of an experimental series to help associate a scientific problem with cellular structures or properties and provide significant input for the generation of a testable hypothesis. Therefore the modular design of ECIS provides a wide spectrum of applications with the possibility for tailor made arrays. The latest array developments indicate a future focus on high throughput impedance screenings for cell proliferation and electrical wounding and the advance of special flow arrays for the simulation of in vivo shear stress with different flow profiles.

Further literature
Please also refer to the webpage of Applied BioPhysics (www.biophysics.com) for application notes, webinars and a detailed list of publications covering the entire ECIS spectrum.

Materials

Name	Company	Catalog Number	Comments
Name of Material/Equipment	Company	Distributor Europe	Catalog Number	Comments/Description
ECIS Zθ	Applied Biophysics Inc., Troy, NY, USA	ibidi GmbH, Munich, Germany	71617	http://www.biophysics.com/products-ecisz0.php
8W1E PCB Electrode Array	Applied Biophysics Inc., Troy, NY, USA	ibidi GmbH, Munich, Germany	70004	http://www.biophysics.com/cultureware.php
8W1E Lexan Electrode Array	Applied Biophysics Inc., Troy, NY, USA	ibidi GmbH, Munich, Germany	70001	http://www.biophysics.com/cultureware.php
8W10E PCB Electrode Array	Applied Biophysics Inc., Troy, NY, USA	ibidi GmbH, Munich, Germany	70010	http://www.biophysics.com/cultureware.php
L-Cystein	Merck Millipore	-	102838	http://www.merckmillipore.com/germany/chemicals/l-cystein/MDA_CHEM-102838/p_GLOb.s1L7V8AAAEWL.EfVhTl?WFSimpleSearch_NameOrID=L-cystein&BackButtonText=search+results
Gelatin	Merck Millipore	-	104070	http://www.merckmillipore.com/germany/chemicals/gelatine/MDA_CHEM-104070/p_S7Wb.s1LadYAAAEWaOEfVhTl?WFSimpleSearch_NameOrID=gelatine&BackButtonText=search+results
Bovine Thrombin	Merck Millipore	-	604980	http://www.merckmillipore.com/germany/life-science-research/thrombin-bovine-high-activity/EMD_BIO-604980/p__pmb.s1LNsYAAAEWS2IfVhTm?PortalCatalogID=merck4biosciences
Albuman (Human Serum Albumin)	Sanquin	-		http://www.sanquin.nl/en/products-services/plasma-products/
Y-27632	Tocris Biosciences	-	1254	http://www.tocris.com/dispprod.php?ItemId=1382#.UbiFNNish8E
EGM-2 BulletKit	Lonza	-	CC-3162	http://www.lonza.com/products-services/bio-research/primary-and-stem-cells/human-cells-and-media/endothelial-cells-and-media/endothelial-cell-growth-media-kits.aspx
Pen/Strep	Lonza	-	17-602E	http://www.lonza.com/products-services/bio-research/cell-culture-products/reagents/antibiotics-antimycotics/penicillin-streptomycin.aspx