Microfluidic Platform for Real-Time Impedance Profiling of Transwell-Based Barrier Models

Abstract

We propose a microfluidic platform with integrated microfabricated electrodes for real-time impedance profiling of transwell-based barrier models, which can be subjected to dynamic microfluidic flow. This platform overcomes the limitations of conventional methods that are based on invasive permeability assays and single time-point impedance measurements: It enables continuous, non-invasive monitoring of tissue barrier integrity at high spatial and temporal resolution. We demonstrate the capabilities of our system by continuously monitoring the gradual loss of barrier integrity in upper-airway-tissue models exposed to non-physiological liquid-liquid interface conditions.

I. Introduction

Barrier models are indispensable for in vitro studies of physiological and pathological processes of various tissues and organs. These models replicate key attributes of epithelial and endothelial tissues, such as barrier function, selective permeability, and transport phenomena, which makes them important tools in biomedical research [1]. Transwell inserts have proven particularly effective for creating physiologically relevant in vitro barrier models, owing to permeable membranes that support polarized cell growth and provide access to both the apical and basolateral compartments for comprehensive analysis [2].

However, conventional transwell cultures have significant limitations, as they are static in nature, which obviates the recapitulation of physiologically relevant microenvironments, and they are not compatible with real-time, continuous optical monitoring for high-resolution live-cell imaging [3, 4]. Additionally, traditional methods for assessing the tissue barrier function in transwell models include permeability assays with fluorescent tracers and commercial trans-epithelial electrical resistance (TEER) measurement setups [5]. These methods are invasive and often only provide single time-point measurements and do not allow for acquiring comprehensive real-time data.

We addressed the two most important issues by developing a poly(methylmethacrylate) (PMMA)-based microfluidic device, which enables dynamic control of the culture microenvironments in both the apical and basolateral compartments of transwell inserts while continuously performing high-resolution imaging of barrier characteristics [6]. We have now further advanced the platform by integrating microelectrodes to enable real-time impedance measurements for non-invasive barrier monitoring. By utilizing glass coverslips, patterned with transparent indiumtin-oxide (ITO) electrodes, as substrates, we were able to realize precise microfluidic control along with high-resolution imaging and TEER measurements.

In our proof-of-concept study, we demonstrated the capability of our chip to monitor the disruption of barrier integrity of an upper airway model under non-physiological liquid-liquid interface conditions through real-time impedance measurements. The results were corroborated with high-resolution imaging and start/end-point standard TEER measurements. We could demonstrate that our system enables continuous barrier monitoring, precise environmental control, and detailed visualization of cellular processes, which is of great relevance in infection monitoring and drug testing.

II. Theory

A. Transwell-based upper airway model

The epithelial barrier in the upper airway acts as the first line of defense against pathogens, and its disruption can lead to severe systemic infections. To maintain physiological conditions of upper airway models, the barrier has to be cultured at the air-liquid interface (ALI) to closely reproduce the in vivo situation. A previous study [6] demonstrated that prolonged exposure to a liquid-liquid interface can significantly affect the barrier properties of airway tissues. This finding highlights the importance of real-time, continuous monitoring to detect subtle changes in barrier integrity and function that may precede visible signs of tissue damage.

B. Electrical Model

The proposed method utilizes a four-electrode system for impedance measurement. Traditional two-electrode TEER systems can be affected by the resistance of connecting wires and the high impedance of the double-layer capacitance at the electrode–electrolyte interface [7]. A four-electrode system mitigates these issues by using a separate pair of electrodes for current injection, the working (WE) and counter (CE) electrodes, and another pair for voltage measurement, the reference (RE) and sensing (SE) electrodes.

The tissue is modeled as a resistor R_TEER and capacitor C_CL (Fig. 1b) in parallel. A frequency sweep of the four-electrode setup measures a complex impedance profile of the tissue, of which the real part represents R_TEER and the imaginary part represents C_CL[8].

Fig. 1.

(a) Microfluidic platform for real-time impedance measurement. (b) Schematics of the impedance measurement setup using an upper airway barrier model in transwell inserts in a 4-electrode setup. 1: 3D-printed lid with Au-plated Cu electrode pair, 2: Transwell, 3: Silicone gasket, 4: Microfluidic chip, 5: Glass coverslip with ITO electrodes, 6: Microscope objective, 7: Upper airway epithelium.

III. Materials and Methods

A. Inverted tissue culture in the microfluidic platform

Upper airway models were generated according to the previously published work, on the lower side of the transwell insert, with the tissue layer facing down [9]. During off-chip culturing, the tissues were maintained under a physiologically relevant air-liquid interface, with ALI medium in the basolateral compartment (top), while the apical side (bottom) was exposed to air. The tissues were stained overnight with CellMask DeepRed (C10046, ThermoFisher Scientific, Waltham, MA, USA) for plasma membrane visualization.

B. Fabrication of the platform

The microfluidic chip featured a micro-milled channel with a circular opening in the center, where the cell-coated transwell was inserted as shown in Fig. 1(a). The bottom of the channel was sealed with a glass coverslip featuring a transparent ITO electrode pair. The channel and the patterned coverslips were bonded using a biocompatible epoxy layer. The basolateral side was closed with a custom-designed 3D-printed lid with an embedded gold-coated copper wire electrode pair. The platform design enabled the insertion of transwells at a pre-defined working distance of 300µm from the objective (Fig. 1(b)), thus enabling high-resolution confocal imaging of the cellular barrier at the bottom side of the transwell. The microfluidic channel featured inlet and outlet openings for connection to external pump units to establish flow.

C. Fabrication of the electrodes

The bottom electrodes (WE, SE) were fabricated by selective etching of #1.5 glass-ITO coverslips (30–60 Ω, Diamond Coatings, West Midlands, UK), as described in a previous study [10]. The WE was designed as a spiral with a diameter of 7 mm, and the SE was realized as a pad that partially arced around the WE (Fig. 1). The top pair of electrodes (CE, RE) is made of a thin layer of gold (NB Semiplate Au-100, NB Technologies, Bremen, Germany), electroplated on the surface of copper wires (Copper Wire, 0.2mm², 25m, Kabeltronik, Germany). The top electrodes were fixed inside a 3D-printed lid to position them above the permeable membrane inside the transwell (Fig. 1(b)).

D. Teer Measurement With Evom

Transepithelial electrical resistance (TEER) values of the upper airway models were measured with an EVOM3 Epithelial Volt Ohm Meter using the STX2-Plus Electrode (World Precision Instruments, Florida, USA) before insertion into the microfluidic platform (t_0h) and after removal from the chip (t_7h). The tissues were mucus-washed and subsequently submerged in pre-warmed HBSS (350 µL apical and 1 mL basolateral) during the measurement.

E. Real-time impedance measurement

Before conducting the impedance measurements, the transwell inserts were mounted into the chip, and a liquid-liquid interface was established by filling the microchannel with HBSS at 10 µL/min.

Measurements were conducted using an HF2-LI lock-in amplifier (Zurich Instruments AG, Zürich, Switzerland). The microfluidic devices were interfaced via a custom-made PCB to connect the lock-in amplifier to the integrated electrodes. A custom-made Matlab GUI was used for the selection of the microfluidic device and to control signal acquisition. A sinusoidal AC voltage signal with a 200 mV peak and excitation frequencies ranging from 10.5 Hz to 10 kHz was applied to the WE of the chosen transwell unit, while the CE was maintained at pseudo-ground by the HF2-LI transimpedance amplifier (HF2TA, Zürich Instruments AG) (Fig. 1(b)). The current passing through the CE, which traverses the epithelial barrier, was converted to a voltage using the HF2TA with a 1-kΩ feedback resistor and sampled by the HF2-LI. Differential voltage measurements between the basolateral and apical compartments of the barrier were recorded with the SE and the RE, while the WE and CE were used for current injection.

The voltage and current measurements were then used to determine the impedance of the barrier layer. Based on a frequency-sweep analysis, we identified 30 Hz as the optimal frequency for R_TEER measurements, minimizing electrode impedance effects while still measuring exclusively real impedance values [5]. Live imaging was carried out using a Nikon spinning-disk confocal microscope (CSU-W1, Nikon, Egg, Switzerland) with environmental control unit for live-cell imaging using a 40× water-immersion objective. The arrangement of the microfluidic TEER setup and the measurement units is shown in Fig. 2.

Fig. 2.

(a) Instrumentation setup, (b) Core measurement unit comprising PCB and microfluidic chip.

IV. Results

To demonstrate the utility of our four-electrode impedance measurement system for real-time barrier integrity profiling, we assessed the tightness of an upper airway epithelium model when immersed in a non-physiological liquid-liquid interface (LLI) condition. To validate our measurements, we used the commercial EVOM chopstick-electrode TEER assay that is compatible with transwell models. As shown in Fig. 3(a), our system detected a gradual decrease in barrier resistance over 7 hours of incubation under LLI conditions and a total resistance reduction of 52% (141 to 67 Ω/cm²), compared to a 34% decrease (294 to 193 Ω/cm²) observed with EVOM. Both chip-based and EVOM recordings included measurements of the background resistance, obtained in the absence of cells (marked “Blank”), amounting to 30 Ω/cm² and 100 Ω/cm², respectively. The difference in the decrease of the tissue’s barrier resistance between the two systems can primarily be attributed to the different measurement frequencies (f_EVOM = 12.5 Hz) and electrode configurations [5].

Fig. 3.

(a) TEER recordings of tissue-coated transwell inserts measured by the developed microfluidic setup and the commercial EVOM setup during LLI exposure, (b) representative live microscopy images evidencing the gradual barrier disruption.

Notably, our system’s capability for continuous real-time monitoring is a significant advantage. Our measurements revealed that the barrier resistance began to decrease after 2.5 hours of LLI exposure and stronger differences between tissue samples appeared from approximately 3.5 hours. The continuous measurements provided detailed, time-resolved insights into the dynamic processes of barrier disruption. Unlike the static endpoint measurements by means of the EVOM system, our approach not only confirmed the overall resistance decrease, but also captured the evolving changes in barrier integrity throughout the LLI exposure. The results were further corroborated by live microscopy images (Fig. 3 (b)) showing gradual degradation of the epithelial barrier, enhancing our understanding of the disruption mechanics.

Overall, these findings validated our developed system as a promising tool for real-time assessments of epithelial barrier integrity under conditions that can be precisely manipulated to mimic pathophysiological conditions in an in-vivo-like microenvironment.