Here, we demonstrate how to set up an inexpensive volt-amperemeter with programmable output frequency that can be used with commercially available chopstick electrodes for transepithelial/endothelial electrical resistance measurements.
Transepithelial/endothelial electrical resistance (TEER) has been used since the 1980s to determine confluency and permeability of in vitro barrier model systems. In most cases, chopstick electrodes are used to determine the electric impedance between the upper and lower compartment of a cell culture filter insert system containing cellular monolayers. The filter membrane allows the cells to adhere, polarize, and interact by building tight junctions. This technique has been described with a variety of different cell lines (e.g., cells of the blood-brain barrier, blood-cerebrospinal fluid barrier, or gastrointestinal and pulmonary tract). TEER measurement devices can be readily obtained from different laboratory equipment suppliers. However, there are more cost-effective and customizable solutions imaginable if an appropriate voltammeter is self-assembled. The overall aim of this publication is to set up a reliable device with programmable output frequency that can be used with commercially available chopstick electrodes for TEER measurement.
Epithelial and endothelial cells function as cellular boundaries, separating the apical and basolateral sides of the body. If they are connected through tight junctions, passive substance diffusion through the paracellular spaces is restricted1, resulting in the formation of a selectively permeable barrier. Several artificial barrier systems have been developed2 using microvascular endothelial cells (HBMEC, blood-brain barrier3,4,5,6,7), choroid plexus epithelial cells (HIBCPP/PCPEC, blood-cerebrospinal fluid barrier8,9,10,11,12,13,14), colorectal adenocarcinoma cells (Caco-2, gastrointestinal models15), or airway/alveolar cell lines (pulmonary models16,17). These systems typically consist of cells grown in a monolayer on permeable membranes (i.e., filter insert systems) to allow access to the apical and basolateral sides. It is important that the integrity of the model system matches the in vivo conditions. Hence, several techniques have been developed to analyze barrier function by measuring paracellular diffusion of tracer compounds across the cell layer. These substances include radiolabeled sucrose, dye-labeled albumin, FITC-labeled inulin, or dye-labeled dextrans2. However, chemical dyes can make cells unusable for further experiments. To monitor barrier systems noninvasively, measurement of transepithelial/transendothelial electrical resistance (TEER) across a cellular monolayer can be used2,18,19. Because bipolar electrode systems are influenced by the electrode polarization impedance at the electrode-electrolyte interface, tetrapolar measurements are generally used to overcome this limitation20. The underlaying technique is a four-terminal sensing (4T) that was first described in 1861 by William Thomson (Lord Kelvin)21. In brief, the current is injected by a pair of current-carrying electrodes while a second pair of voltage-sensing electrodes is used to measure the voltage drop20. Nowadays, so-called chopstick electrodes consist of a pair of double electrodes, each containing a silver/silver-chloride pellet for measuring voltage and a silver electrode for passing current2. The electrical impedance is measured between the apical and the basolateral compartment with the cell layer in between (Figure 1). A square wave signal at a frequency of typically 12.5 Hz is applied at the outer electrodes and the resulting alternating current (AC) measured. Additionally, the potential drop across the cell layer is measured by the second (inner) electrode pair. Electrical impedance is then calculated according to Ohm's law. TEER values are normalized by multiplying impedance and cell layer surface area and are typically expressed as Ω ∙ cm2.
There are systems in which cells and electrodes are arranged in a more sophisticated way, but are also based on the 4T measuring principle and can be used with the same measurement devices. EndOhm systems, for example, in which the filter is inserted, contain a chamber and cap with a pair of concentric electrodes with the same structure as the chopstick electrode. The shape of the electrodes allows for a more uniform current density flow across the membrane, thereby reducing variation between readings. Even more complex (but also more accurate) is an Ussing chamber, where a cell layer separates two chambers filled with Ringer's solution22. The chamber itself can be gassed with oxygen, CO2, or N2, and stirred or supplemented with experimental substances. As ion transport across the cell layer occurs, a potential difference can be measured by two voltage-sensing electrodes near the tissue. This voltage is cancelled out by two current-carrying electrodes placed next to the cell layer. The measured current will then give the net ion transport and the transepithelial resistance, which reflects barrier integrity, can be determined22. TEER measurement can also be applied on body-on-a-chip systems that represent barrier-tissue models23,24. These systems mimic in vivo conditions of the cells and often consist of several types of cells, stacked on top of each other in layers.
The following protocol explains how to set up a cost-effective and reliable voltammeter with programmable output frequency that produces no statistically significant differences in TEER compared to commercially available measurement systems.
1. Assembly of a basic volt-amperemeter for TEER measurement
2. Programming the microcontroller
3. Recording of voltage oscillograms (optional)
4. Cell cultivation and TEER measurement
To compare the operation of a self-assembled voltammeter with its commercially available counterpart, a voltage oscillogram of both devices was recorded.
As shown in Figure 2A, the reference instrument generated a square wave signal with an amplitude of 80 mV and an oscillation time of 80 ms, which corresponds to a frequency of 12.5 Hz, when operating on-load with a 1 kΩ test resistor.
In contrast, the microcontroller of the self-assembled device switched the supply voltage to a square wave signal with an amplitude of 5 V (Figure 2B) if no pre-resistor was set in. It became apparent that the resulting current destroys any barrier function and is not applicable for cell culture experiments (data not shown). A further issue is that, in this setup a 1 kΩ test resistor caused an overload with a resulting decline of voltage (Figure 2B). Additionally, the effective oscillation time of the microcontroller was 60 ms (frequency = 16.7 Hz) and thereby differed from the programmed delay time due to the inaccuracy of the time emitter. If a 120 kΩ preresistor was installed, the amplitude decreased to a value of 40 mV, which was suitable for cell culture (Figure 2C). As seen in the oscillogram, signal-to-noise ratio was considerably impaired (Figure 2C) but did not affect measurements noticeably.
Both devices were used to determine the impedance of an artificial blood-cerebrospinal fluid barrier (simplified circuit diagram shown in Figure 2D). HIBCPP cells were cultivated on cell culture filter inserts and TEER was measured over 6 days: starting one day before cells were moved to serum-free conditions (Day -1) and up to 4 days after changing the medium (Day 4). All measurements were done in quadruplicates using four HIBCPP filters prepared in the same manner. Similar values were obtained for the reference instrument and the self-assembled voltammeter (Figure 3). Measurements were reproducible, and standard deviations were within the same range. TEER values ranged from 20−550 Ω ∙ cm2. Using 0.33 cm2 filters, this equates to an absolute impedance of 83−1,660 Ω.
Figure 1: Layout diagram of a basic volt-amperemeter for TEER measurement. Please click here to view a larger version of this figure.
Figure 2: Oscillograms and measurement setup. (A) Commercially available EVOM. (B) Self-assembled voltammeter without pre-resistor. (C) Self-assembled voltammeter with 120 kΩ pre-resistor. (D) Circuit diagram of measurement setup. Note that Celectrode only appears in the electrical circuits when bipolar systems are used. Please click here to view a larger version of this figure.
Figure 3: TEER measurements of HIBCPP cell layers on cell culture filter inserts before switching to serum-free culture medium (Day -1), on the day of switching (Day 0), and up to 4 days after (Days 1−4). Error bars indicate the standard deviation of the four HIBCPP filters that were prepared in the same manner. Please click here to view a larger version of this figure.
Supplemental Coding File 1. Please click here to download this file.
Supplemental Coding File 2. Please click here to download this file.
Supplemental Coding File 3. Please click here to download this file.
Before a self-made voltammeter can be used in a daily routine, it is essential to check the device for proper function. In our case, a half-time of oscillation of 40 ms (12.5 Hz) was programmed, but the effective oscillation time turned out to be 60 ms (16.7 Hz). This inaccuracy of the microcontroller's time emitter had no detectable impact on TEER measurements. It might be best to determine the actual frequency using the frequency setting of one of the multimeters. If any deviation is found, the source code can be adjusted accordingly. Further, it is strongly recommended to check whether a test resistor or other defined setups give correct and reproducible results. If working with artificial cellular barrier systems, it might be best to always correlate molecule flux with impedance measurement.
In this case, the applied current was limited using a 120 kΩ pre-resistor. Assuming that typical TEER values range from 100 Ω−2,000 Ω, the voltage drop across the cell layer can be calculated to be 4−83 mV. A TEER of 1 kΩ was simulated by a test resistor and the resulting potential drop was confirmed to be 40 mV (Figure 2C).
Commercially available devices often provide a measurement range switch to toggle the pre-resistor and thus limit the output current to different values. In this case, it is feasible to install different pre-resistors or to even replace the resistor with a potentiometer.
The shown setup represents a cost-effective alternative to commercially available instruments for TEER measurement. Values that have been measured with the self-assembled voltammeter were comparable to the reference device over a broad range. The same is true for the standard deviations. The noise in the square wave signal did not affect measurements notably. The protocol can support scientists who are restricted by limited financial resources or who want to perform preliminary experiments at low costs.
Further, the microcontroller can be easily programmed to different output frequencies. This may be beneficial, as the apparent impedance consists of Rmedium, RTEER, as well as the capacity Ccell layer26 (Figure 2D). Additionally, Celectrode appears if bipolar systems are used, whereas the influence from the electrode polarization impedance is reduced in tetrapolar systems. This means that the measured impedance will be dominated by RTEER at low frequencies and, in bipolar systems, by the capacity of the electrodes, whereas at high frequencies the total impedance converges to the resistance of the medium26,27. In between, the impedance is influenced by Ccell layer, which is therefore accessible using electrical impedance spectroscopy28.
We provide two (untested) example codes to give an idea how the device could be optimized or reprogrammed for different applications. First, a very basic impedance spectroscopy could be realized by alternating the output frequency in 20 second intervals between 12.5, 500 and 5000 Hz (supplemental coding file 2). In this case, a tetrapolar20,28 or bipolar27 electrode could be used. Applied frequency could be shown by the build-in multimeter (or any display or LED connected to the microcontroller). Second, the device could be used to measure conductivity of buffers and media. This is typically done using tetrapolar electrodes with high frequencies in a range of 1-110kHz. The code in supplemental coding file 3 contains no delay time and (with our device) generated a frequency of approximately 70kHz.
The authors have nothing to disclose.
The authors would like to thank Herman Liggesmeyer and Marvin Bende for their expert advice in electrotechnics and informatics.
120 kOhm resistor | General (generic) equipment | ||
Banana plug cables | General (generic) equipment | ||
Cables | General (generic) equipment | ||
Chopstick electrode | Merck Millicell | MERSSTX01 | |
Chopstick electrode (alternative) | WPI World Precision Instruments | STX2 | |
Crimping tool | General tool | ||
Digispark / ATtiny85 | AZ-Delivery Vertriebs GmbH | Digispark Rev.3 Kickstarter | |
DMEM:F12 | Gibco (Thermo Fisher) | 31330038 | |
Fetal calf serum (FCS)/Fetal Bovine Serum (FBS) | Life Technologies | 10270106 | |
Filter inserts 3µm translucent | Greiner Bioone | 662631 | |
HIBCPP | Hiroshi Ishikawa / Horst Schroten | ||
Insulation stripper | General tool | ||
Luster terminal | General (generic) equipment | ||
Oscilloscope | HAMEG | Digital Storage Scope HM 208 | |
Plotter | PHILIPS | PM 8143 X-Y recorder | |
Software Arduino | https://www.arduino.cc | Arduino 1.8.9 | |
Soldering iron | General tool | ||
Soldering lugs | General (generic) equipment | ||
Telephone cable with RJ14 (6P4C) connector | General (generic) equipment | ||
Test resistor | Merck Millicell | MERSSTX04 | |
True-RMS multimeters | VOLTCRAFT | VC185 | |
USB charger | General (generic) equipment | ||
USB extension cord | General (generic) equipment | ||
Voltohmmeter for TEER measurement | WPI World Precision Instruments | EVOM | |
Voltohmmeter for TEER measurement (alternative) | Merck Millicell | ERS | |
Wire end ferrules | General (generic) equipment |