Günter Wrobel

Recording of cell action potentials with AlGaN∕GaN field-effect transistors

An AlGaN∕GaN electrolyte gate field-effect transistor array for the detection of electrical cell signals has been realized. The low-frequency noise power spectral density of these devices exhibits a 1∕f characteristic with a dimensionless Hooge parameter of 5×10−3. The equivalent gate-input noise under operation conditions has a peak-to-peak amplitude of 15μV, one order of magnitude smaller than for common silicon-based devices used for extracellular recordings. Extracellular action potentials from a confluent layer of rat heart muscle cells cultivated directly on the nonmetallized gate surface were recorded with a signal amplitude of 75μV and a signal-to-noise ratio of 5:1.

Transmission electron microscopy study of the cell–sensor interface

An emerging number of micro- and nanoelectronics-based biosensors have been developed for non-invasive recordings of physiological cellular activity. The interface between the biological system and the electronic devices strongly influences the signal transfer between these systems. Little is known about the nanoscopic structure of the cell–sensor interface that is essential for a detailed interpretation of the recordings. Therefore, we analysed the interface between the sensor surface and attached cells using transmission electron microscopy (TEM). The maximum possible resolution of our TEM study, however, was restricted by the quality of the interface preparation. Therefore, we complemented our studies with imaging ellipsometry. We cultured HEK293 cells on substrates, which had been precoated with different types of proteins. We found that contact geometry between attached cell membrane and substrate was dependent on the type of protein coating used. In the presence of polylysine, the average distance of the membrane–substrate interface was in the range of 35–40 nm. However, the cell membrane was highly protruded in the presence of other proteins like fibronectin, laminin or concanavalin-A. The presented method allows the nanoscopic characterization of the cell–sensor interface.

Time-dependent observation of individual cellular binding events to field-effect transistors

Electrolyte-gate field-effect transistors (EG-FETs) gained continuously more importance in the field of bioelectronics. The reasons for this are the intrinsic properties of these FETs. Binding of analysts or changes in the electrolyte composition are leading to variations of the drain-source current. Furthermore, due to the signal amplification upon voltage-to-current conversion even small extracellular signals can be detected. Here we report about impedance spectroscopy with an FET array to characterize passive components of a cell attached to the transistor gate. We developed a 16-channel readout system, which provides a simultaneous, lock-in based readout. A test signal of known amplitude and phase was applied via the reference electrode. We monitored the electronic transfer function of the FETs with the attached cell. The resulting frequency spectrum was used to investigate the surface adhesion of individual HEK293 cells. We applied different chemical treatments with either the serinpeptidase trypsin or the ionophor amphotericin B (AmpB). Binding studies can be realized by a time-dependent readout of the lock-in amplifier at a constant frequency. We observed cell detachment upon trypsin activity as well as membrane decomposition induced by AmpB. The results were interpreted in terms of an equivalent electrical circuit model of the complete system. The presented method could in future be applied to monitor more relevant biomedical manipulations of individual cells. Due to the utilization of the silicon technology, our method could be easily up-scaled to many output channels for high throughput pharmacological screening.

Membrane allocation profiling: a method to characterize three-dimensional cell shape and attachment based on surface reconstruction.

Three-dimensional surface reconstructions from high resolution image stacks of biological specimens, observed by confocal microscopy, have changed the perspective of morphological understanding. In the field of cell-cell or cell-substrate interfaces, combining these two techniques leads to new insights yet also creates a tremendous amount of data. In this article, we present a technique to reduce large, multidimensional data sets from confocal microscopy into one single curve: a membrane allocation profile. Reconstructed cells are represented in a three-dimensional surface from image sections of individual cells. We virtually cut segments of the reconstructed cell membrane parallel to the substrate and calculate the surface areas of each segment. The obtained membrane allocation profiles lead to morphological insights and yield an in vivo ratio of attached and free membrane areas without cell fixation. As an example, glass substrates were modified with different proteins (fibronectin, laminin, concavalin A, extracellular matrix gel, and both isomers of poly-lysine) and presented to HEK293 cells to examine differences in cell morphology and adhesion. We proved that proteins on a substrate could increase the attached portion of a cell membrane, facing the modified substrate, from an average of 32% (glass) to 45% (poly-lysine) of the total membrane surface area.

Single cell recordings with pairs of complementary transistors

Floating gate field-effect transistors (FETs) for the detection of extracellular signals from electrogenic cells were fabricated in a complementary metal oxide semiconductor process. Additional passivation layers protected the transistor gates from the electrolyte solution. To compare the signals from n- and p-FETs, two electronically separated, but locally adjacent transistors were combined to one measuring unit. The paired sensing area of this unit had the dimension of a single cell. Simultaneous recordings with n- and p-channel floating gate FETs from a single cell exhibited comparable amplitudes and identical time courses. The experiments indicate that both types of FETs express similar sensitivities.

To establish a pharmacological experimental platform for the study of cardiac hypoxia using the microelectrode array

INTRODUCTION Simultaneous recording of electrical potentials from multiple cells may be useful for physiological and pharmacological research. The present study aimed to establish an in vitro cardiac hypoxia experimental platform on the microelectrode array (MEA). METHODS Embryonic rat cardiac myocytes were cultured on the MEAs. Following >or=90 min of hypoxia, changes in lactate production (mM), pH, beat frequency (beats per min, bpm), extracellular action potential (exAP) amplitude, and propagation velocity between the normoxic and hypoxic cells were compared. RESULTS Under hypoxia, the beat frequency of cells increased and peaked at around 42.5 min (08.1+/-3.2 bpm). The exAP amplitude reduced as soon as the cells were exposed to the hypoxic medium, and this reduction increased significantly after approximately 20 min of hypoxia. The propagation velocity of the hypoxic cells was significantly lower than that of the control throughout the entire 90+ min of hypoxia. The rate of depolarisation and Na(+) signal gradually reduced over time, and these changes had a direct effect on the exAP duration. DISCUSSION The extracellular electrophysiological measurements allow a partial reconstruction of the signal shape and time course of the underlying hypoxia-associated physiological changes. The present study showed that the cardiac myocyte-integrated MEA may be used as an experimental platform for the pharmacological studies of cardiovascular diseases in the future.

Probing the Adhesion and Viability of Individual Cells with Field-Effect Transistors

We describe a novel method for the non-invasive, electronic monitoring of the cell-substrate adhesion. We utilize open-gate field-effect transistor (FET) chips for our adhesion measurements. These chips were developed, fabricated, and previously used for the extracellular recording from electrogenic cells. We developed a miniaturized and portable 16-channel amplifier system, which monitors the electronic transfer function of the FETs. We present cellular adhesion measurements of HEK 293 cells on an individual cell level. The system can additionally be used to probe the viability of individual cells.

Interfacing Neurons and Silicon-Based Devices

The combination of biological signal processing elements such as membrane proteins, whole cells, or even tissue slices with electronic transducers for the detection of physical signals creates functional hybrid systems that bring together the living and the technical worlds. Functional coupling of physiological processes with microelectronic and nanoelectronic devices will have great impact for a wide range of applications. The high sensitivity and selectivity of biological recognition systems with a manufactured signal-detection and processing system will open up exciting possibilities for the development of new biosensors as well as for new approaches in neuroscience and computer science. This includes: (a) pharmacological as well as toxicologically lab on a chip concepts, which allows fast, high-throughput screening of potential drugs; (b) the use of the high sensitivity and selectivity of biological recognition systems with signal-amplification cascades for the development of biosensors with unprecedented detection threshold; and (c) the multisite interfacing of neuronal networks with arrays of electronic devices on the microscopic level of individual nerve cells or cell processes would facilitate spatiotemporal mapping of brain dynamics.