TFT sensor array for real-time cellular characterization, stimulation, impedance measurement and optical imaging of in-vitro neural cells
Introduction
Demands for investigating the neuronal networks and the functions of the nervous system are escalating in the field of advanced in-vitro drug screening (Park et al., 2015) to cope with the central nervous system diseases such as neurodevelopmental, psychiatric, and neurodegenerative disorders (Grskovic et al., 2011). The electrogenic cells generating electrical signals facilitate the study in the electrical domain under in-vitro or in-vivo conditions (Azim et al., 2019; Cohen-Karni et al., 2010; Hess et al., 2011; Santoro et al., 2014; Shaik et al., 2020). The in-vivo study of complex neuronal networks is challenging because of the physiological complications, ethical dilemmas, and additional filtration. In-vitro neuronal study, being further classified as either extracellular or intracellular, uses measurement tools such as patch-clamp and sharp microneedle probes. Although intracellular study has an advantage of a high signal-to-noise ratio (Angle et al., 2015), it has several major drawbacks such as mechanical instability, incompatibility with multisite recording, and a risk of fluid contamination or dilution with the electrode fluid in the case of patch-clamp studies, all of these resulting in inaccurate diagnosis (Pottosin and Dobrovinskaya, 2015). Understanding the brain functions, for instance, depends mostly on the extracellular recording, which consists of direct measurement of electrical activities from recording electrodes placed close to the neuronal cell culture without causing damage to the neuron's plasma membrane (Aziz et al., 2009; Kim et al., 2015; Ruther and Paul, 2015; Shen et al., 2015). The resulting voltage signals are composed of the local field potentials (LFPs) and the spikes emitted by one or more neurons (Zanos et al., 2011).
To develop an in-vitro model through the study of neural dynamics, an ideal readout device needs to have the following features: biocompatibility, optical transparency (for fluoroscopy analyses), high fill factor, large sensor area, on-chip multiplexing, bidirectional interface, fast temporal response, and high spatial resolution. Traditional multi-electrode-array (MEA) devices are electrically passive and planar in shape with metal electrodes on top of a glass substrate (Abdelfattah et al., 2016; Foidl et al., 2018; Franke et al., 2017). However, due to the congestion problem of electrical interconnection, MEAs usually have limited number of independent electrodes, resulting in a low density and a low throughput of measurement (Cabello et al., 2018; Hood et al., 2009; Lu et al., 2016; Spira and Hai, 2013). On the contrary, complementary metal-oxide semiconductor (CMOS) arrays can accommodate a large number of electrodes at a high density. Nonetheless, silicon substrate for CMOS technology is opaque, and it would not allow visual observation by the inverted microscope (Bertotti et al., 2014; Cabello et al., 2018; Heer et al., 2006; Park et al., 2018; Snell et al., 1981).
Besides MEA and CMOS technologies, thin film transistor (TFT) technology is expected to play an increasing role for the electrophysiology. TFT is a well-known technology for flat panel display (Snell et al., 1981). Several milestones have been attained (Brody et al., 1973; Le Comber et al., 1979; Spear and Le Comber, 1993; Yamamoto, 2012) since the first TFTs were made of a compound semiconductor called cadmium sulfide (CdS), which was introduced in 1962 by Weimer (1962). Hosono's group with the Tokyo Institute of Technology, Japan, first explored the semiconducting indium gallium zinc oxide (IGZO) as a TFT channel material, and demonstrated crystalline and amorphous IGZO-TFTs in 2003 and 2004, respectively (Nomura et al., 2004, 2003).
In this paper, we present a transparent high-definition crystalline-IGZO-TFT system as a bio-measurement platform. The device chip is made by the same fabrication technology for the liquid-crystal display (LCD) fabrication. The crystalline-IGZO is an optically transparent semiconductor material with electron mobility of 20–50 times higher than that of amorphous silicon. In this study, crystalline-IGZO-TFTs are used as a transparent substrate for see-through observation of bio-tissue, and the high electron mobility is adopted for high-speed recording of the ionic behaviors. The system based on the TFT technology thus benefits from the optical transparency and high-density electronics to produce a sensor array of various cell analysis.
The purpose of this paper is to demonstrate all the potentials of TFT systems applicable to a neuron culture. Previous reports from the authors’ group demonstrated a singular modality (neuron sensing only, or impedance sensing). The different aspects of the systems are described in the following sections. The system architecture, corresponding functional units, and their integration are described in Sections 2. The electrical characterization and measurements as well as the biological measurements are described in Section 3. Sections 4 Comparison with the state-of-the-art, 5 Conclusion respectively give a detailed summary and comparison to the states-of-the-art measurement methods.
Section snippets
System architecture
Crystalline-IGZO-TFT in this work is designed to perform multiple stimulation/measurement functions as schematically shown in Fig. 1(a) by appropriately choosing the electrode sites (Shaik et al., 2018, 2017). The block diagram in Fig. 1(b) illustrates the system architecture, where all the functional units are depicted. Typical chip has 175 × 175 TFT microelectrodes over an area of 243 mm2. TFTs plates of several cm2 or more are achievable, which is useful for organ-on-chip study (Wikswo et
Device characterization
Characterization of the TFT electrodes has been performed to ensure precise and selective stimulation of individual neurons. Typical performances are shown in Fig. 3, including (a) drain voltage response, (b) gate voltage response, (c) frequency response, (d) total harmonic distortion (THD), (e) drain current as a function of transistor gate width, (f) electrolysis test, (g) stress characterization, and (f) optical characterization. The output curves give qualitative information such as the
Comparison with the state-of-the-art
The performance of the developed TFT chip has been summarized and compared with other states-of-the-art devices used for electrogenic cells study as shown in Table 1. In the past, a large number of studies have been carried out on opaque substrates such as silicon-based MOSFET and ion-sensitive field-effect transistor. A trade-off exists in the conventional MEAs between the substrate transparency and the electrode density; achieving the high transparency required a sparse electrode density and
Conclusion
An arrayed biosensor chip with six modalities for neuronal ensemble investigation has been developed using the thin-film-transistor (TFT) technology, and characterized from electrophysiological, electrochemical and optical points of view. The electrodes are designed in such a way that they can accommodate all the necessary features as an ideal readout system to understand the complex neuronal behaviors. Besides CMOS, the high definition TFT system includes the largest number of measurement or
CRediT authorship contribution statement
Faruk Azam Shaik: Writing- original draft. Hiroshi Toshiyoshi: Writing - review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
This work was supported by JSPS Grant-in-Aid for Scientific Research (B) Grant Number 15H03984.
References (58)
- et al.
Curr. Opin. Neurobiol.
(2015) - et al.
Sensor. Actuator. B Chem.
(2018) - et al.
Brain Res. Protoc.
(1997) - et al.
J. Neurosci. Methods
(2018) - et al.
Neuron
(2012) - et al.
Curr. Opin. Neurobiol.
(2015) - et al.
Biosens. Bioelectron.
(2012) - et al.
Solid State Commun.
(1993) - et al.
J. Neurosci.
(2016) - et al.
J. Microelectromechanical Syst.
(2019)
IEEE J. Solid State Circ.
Nat. Mater.
J. Neurosci. Res.
IEEE Trans. Electron. Dev.
IEEJ Trans. Sensors Micromachines
J. Biomed. Nanotechnol.
IEEE Trans. Biomed. Circuits Syst.
Sci. Rep.
Nano Lett.
Microsystems for Pharmatechnology
IEEE J. Solid State Circ.
Nature
Nat. Rev. Drug Discov.
Front. Neurosci.
Some typical dimensions of dendrites for a few types of neurons
Cited by (20)
Organ-on-a-chip platforms integrated with biosensors for precise monitoring of the cells and cellular microenvironment
2024, TrAC - Trends in Analytical ChemistryElectrical instabilities of a-IGZO TFTs under different conditions of bias and illumination stress
2023, Microelectronics Reliabilitya-IGZO thin-film transistors with transparent ultrathin Al/Ag bilayer source and drain for active neural interfaces
2023, Materials Science in Semiconductor ProcessingCitation Excerpt :The shortcomings of the transparency of neural interfaces are surmountable by thin-film transistors (TFTs). Fabrication of high-density electrodes and usage of transparent glass substrates are possible with TFT-based active-matrix systems [9–11]. Moreover, the TFT itself can be transparent and flexible when wide bandgap semiconducting materials such as a-IGZO are used, which is well known for high optical transparency and mobility in display applications.
Recent trends of biomaterials and biosensors for organ-on-chip platforms
2022, BioprintingCitation Excerpt :The focused ultrasound was integrated vertically on the microfluidic device so the pressure waves were delivered perpendicularly. Shaik et al. [100] presented a thin-film-transistor (TFT) array biosensor chip for neuronal ensemble investigation. The biosensor allowed the recording of local field and action potentials, as well as optical imaging, chemical stimulation, current or voltage stimulation and electrical impedance measurement.
A real-time mirror-LAPS mini system for dynamic chemical imaging and cell acidification monitoring
2021, Sensors and Actuators, B: ChemicalCitation Excerpt :Nonoptical sensors with electrical signal detection for noninvasive and long-term monitoring were applied to measure pH changes since protons, lactic acid or CO2 products of cellular metabolism are generated in glycolysis and respiration processes [1,6,7]. With the scaling of microfabrication technology, nonoptical sensors with dimensions down to the micrometer level, including microelectrode arrays (MEAs) [8,9], ion-sensitive field effect transistors (ISFETs) [6,7,10–12] and light-addressable potentiometric sensors (LAPSs) [13–15], have been developed, which could be applied in the biochemical measurement of living cells and tissues. All of them can achieve two-dimensional (2D) electrical measurements, but the resolution resulting from the dimensions and spacing of the sensing points of MEAs and ISFETs is still limited by the complexity of fabrication and high cost of arrays.
Advancing Intelligent Organ-on-a-Chip Systems with Comprehensive In Situ Bioanalysis
2024, Advanced Materials