Sensitive and prolonged intracellular electrophysiological recording by three‐dimensional nanodensity regulation

With the advancement of micro/nanotechnologies, multielectrodes arrays (MEAs) with different three‐dimensional (3D) micro/nanostructures have been developed to achieve intracellular action potential recording of cardiomyocytes. However, the effect of the 3D micro/nanostructures density on the intracellular recording has not been fully investigated. In this work, 3D tunable nanodensity electrode arrays (TNDEA) are fabricated by hydrothermal synthesis and standard microfabrication to explore the effect of nanodensity regulation on intracellular biosensing. By low‐voltage electroporation, the signal quality of intracellular potentials recorded by the low‐nanodensity TNDEA showed was significantly improved compared to those recorded by the high‐nanodensity TNDEA. The low‐nanodensity TNDEA improved the amplitude (up to 7.7 mV), signal‐to‐noise ratio (SNR) (up to 69.46 dB), recording duration (up to 83 min), and recording yield (∼100%). The 3D nanodensity regulating strategy has achieved sensitive and prolonged intracellular biosensing of action potentials and is expected to be a powerful electrophysiological research tool in the biomedical field.


INTRODUCTION
Cardiac arrhythmias as one of the cardiovascular diseases (CVDs) have been reported to result in human death and disability. 1,2Accurate monitoring of cardiac arrhythmias is therefore necessary for cardiac electrophysiological study and novel drug development.Recent studies have reported various strategies to pave the way for accurate monitoring of cardiac arrhythmias from cells to animals.Among these strategies, electrophysiological recording of cardiomyocyte offers a precise model at the cellular level that balances the cost and accuracy of the cardiac arrhythmias study.][5][6] Compared to extracellular potential recording, more information can be obtained from cellular processes based on high-fidelity intracellular potential recording, which contributes greatly to both biological studies and biomedical applications. 7,8][11] The patch clamp technique, a gold-standard technique with high signal-to-noise ratio (SNR), temporal resolution, and signal fidelity, is a traditional intracellular recording method to measure transmembrane potential by directly accessing the cell interior with a glass pipette. 12,13However, the application of patch clamp technique in longterm period and large-scale parallel electrophysiological investigations is hampered due to its high invasiveness to cells, limited recording time, and low experimental throughput.5][16] However, MEAs conventionally applied in extracellular recordings produce low-quality and distorted extracellular signals, limiting their applications in measurement of the transmembrane action potentials.
][19][20][21] Meanwhile, various cellular perforation techniques including chemical modification, 22,23 electroporation, 24,25 and plasmonic optoporation 19,26 have been integrated into the 3D structure-based MEAs to improve the recordings of intracellular potentials.Compared to the MEAs with planar structure, MEAs with different 3D micro/nanostructures can form a tighter cell-electrode coupling to improve the signal quality.For example, nanopillar electrodes 27 and nanotube electrodes 24 cultured with cardiomyocytes could achieve high signal strength of recorded action potentials by nanoscale electroporation and resealing processes.However, these 3D MEAs usually requires complicated and time-consuming fabrication procedures such as focused ion beam or e-beam lithography, hampering their high efficiency and large-scale manufacturing. 28Although various geometry of 3D micro/nanostructure and the corresponding cellular penetration strategies have been rapidly developed, the fundamental effect of 3D micro/nanostructures on the electrophysiological recording has not been fully understood to further improve these 3D devices.
Here we propose a universal strategy to achieve sensitive and prolonged intracellular recording of action potentials by regulating 3D nanodensity (Figure 1).The 3D tunable nanodensity electrode array (TNDEA) is prepared by hydrothermal synthesis and standard microfabrication in a large-scale manner.Low-nanodensity and high-nanodensity TNDEA could form different cardiomyocyte-nanostructure coupling.When low-voltage electroporation is applied through the conductive nanointerface, the intracellular electrophysiological signals of cardiomyocytes could be acquired through perforating cell membrane.Results show that the low-nanodensity TNDEA effectively improved the quality of intracellular recordings with significant enhancement on amplitude, SNR, recording duration, and recording yield, when compared to the high-nanodensity TNDEA and conventional planar MEA.This universal nanodensity regulation on multielectrode device offers a novel and effective strategy to achieve high-performance intracellular recordings and is a highly promising and powerful platform in the electrophysiological study.
F I G U R E 1 Schematic of cardiomyocyte-electrode biointerface for intracellular electrophysiological recording.Cardiomyocyte will form different cell-electrode coupling on electrodes with different structures.By nanodensity regulation, electroporation can be performed by different nanodensity devices to obtain intracellular recordings of cardiomyocytes with different signal qualities.Low-nanodensity tunable nanodensity electrode array (TNDEA) presented higher signal quality than those of high-nanodensity TNDEA and planar multielectrodes array (MEA) in terms of amplitude, signal-to-noise ratio, and recording duration.

Fabrication and characterization of TNDEA
By integrating the hydrothermal synthesis with standard microfabrication, the 3D TNDEA devices could be obtained with large-scale manufacturing.As shown in Figure 2A, ZnO seeds were first prepared by magnetron sputtering.The thicknesses of ZnO seeds were controlled by sputtering time, and the thickness is key factor to control the ZnO nanopillar growth.Then, the low-and high-nanodensity ZnO nanopillars were synthesized using hydrothermal synthesis.SEM imaging were used to characterize the nanodensity of TNDEA device.High nanodensity and low nanodensity were 10 nanopillars/µm 2 and 0.56 nanopillars/µm 2 , respectively (Figure 2B and Figure S1).Surface coverage of low-nanodensity structure was lower than that of the high-nanodensity structure, while the diameter of the low-nanodensity nanopillars (0.49 µm) was larger than that of the high-nanodensity nanopillars (0.13 µm).Following the synthesis of nanopillars, multichannel electrodes were defined on the nanopatterned substrate by the standard photolithography with a Ti/Au (10/100 nm) conductive layer (Figure S2).Moreover, SU-8 insulating layer was fabricated on the substrate with 32 circular openings in diameter of 10 µm to insulate the nonelectrode region, reducing capacitive current leaks from the electrical leads (Figure 2C and Figure S3).The exposed conductive electrode was coupled with the cell for electrophysiological signal recording.
Figure 2D describes the assembly of TNDEA device.The TNDEA was fixed on a customized printed circuit board (PCB) using polydimethylsiloxane (PDMS) glue.The TNDEA device and PCB were electrically connected by conductive silver glue through corresponding pads.A glass ring was attached to center of TNDEA device by PDMS glue to form a cell culture well, and the pin headers was soldered on the PCB to match the self-developed multichannel integrated electrical signal recording and stimulating system (Figure 2E).A platinum/titanium wire was introduced from the top of the cell culture well and immersed in the culture medium to serve as a reference electrode.Before cell culture, the devices were sterilized with 75% ethanol soak and ultraviolet exposure in the biosafety cabinet.Primary cardiomyocytes were then cultured on the ZnO/Au nanopillars.Figure 2F and Figure S4 showed the cross-sectional images of cardiomyocyte-nanostructure biointerface.In contrast to high-density ZnO/Au nanopillars, the low-density ZnO/Au nanopillars formed a tight interface with cell membrane, which was beneficial for high-quality intracellular signal recording after electroporation.Additionally, electrochemical impedance spectroscope revealed that low-density TNDEA possesses lower electrical impedance (1.6 ± 0.061 MΩ at 1 kHz) than those of high-density TNDEA (7.7 ± 7.49 MΩ at 1 kHz) and planar MEA (11.2 ± 3.49 MΩ at 1 kHz) (Figure 2G).

Performance enhancement of TNDEA with nanodensity regulation
To determine the performance of the TNDEA with different nanodensities, we carried out electrophysiological signals recordings of cardiomyocytes using low-density TNDEAs, planar MEAs, and high-nanodensity TNDEAs.The primary neonatal rat cardiomyocytes can be cultured for 7-9 days.Conventionally, after three-day culture, the cardiomyocytes fused to form the monolayer on the entire surface of different TNDEA devices and exhibited rhythmic spontaneous beating, and the extracellular potentials of cardiomyocytes can be measured.Except for refreshing the culture medium, the device and recording system are kept in the 37 • C and 5% CO 2 cell incubator during the entire experiment.Figure 3A illustrates coupling of the cardiomyocyte membrane with different nanodensity electrodes.Compared to high-density TNDEA and planar MEA, low-density TNDEA produced higher cell membrane curvature at the cell-electrode interface, leading to a tighter cell-electrode coupling and higher seal resistance (Figure S5).An electroporation was applied transiently on cell membrane to form nanopores and record the intracellular potentials, and the penetration of electrodes in inter-cellular experiments can be ensured by the signal quality and profile.Figure 3B(i) shows the sequential recording of typical spontaneous extracellular action potentials by low-density TNDEA (amplitude of ∼50-1500 µV), planar MEA (amplitude of ∼100-500 µV), and high-density TNDEA (amplitude of ∼50-500 µV) from cardiomyocytes.The partially enlarged signals exhibit significant biphasic extracellular potentials recorded by these devices.For the intracellular recording, a highly localized electric field was induced through the TNDEA to cause transient electroporation on cell membrane.This electroporation with 20 pulses and 3 V in amplitude for 1s induced the nanopores on cell membrane, lowered the cell-electrode junction impedance, and improved electrophysiological recording (Figure 3B(ii)).The amplitudes of intracellular recording using low-density TNDEA, planar MEA, and high-nanodensity TNDEA electroporation are up to 7.7, 2, and 2.2 mV, respectively.As shown in Figure 3B(ii), the enlarged profiles of electrophysiological signals exhibit intracellular-like potentials, which shows significant depolarization, repolarization, and rest phases.The profile of TNDEA recorded intracellular potentials agrees with that of patch-clamp recording (Figure S6).The partially enlarged signals demonstrate that these devices could record both extracellular and intracellular electrophysiological signals.Furthermore, TNDEA can be demonstrated to have good mechanical strength and stability from two aspects.On the one hand, there is seldom difference in the surface of the nanopillars before and after cell culture, so the mechanical strength is strong enough for cell culture and electrophysiological experiments (Figure S7).On the other hand, these TNDEA devices can be reused 3-8 times in this work to record the extracellular potentials of cardiomyocytes and intracellular potentials after cardiomyocyte electroporation, which also indicates good mechanical strength and stability.
To further compare the performance of low-nanodensity TNDEA, planar MEA, and high-nanodensity TNDEA, we carried out quantitative statistical analysis for amplitudes, SNR, and intracellular yield.As shown in Figure 3C, despite different electrode impedance and cell-electrode coupling of different devices, there was no significant difference between the mean values of extracellular action potentials amplitudes recorded by low-nanodensity TNDEA (0.213 ± 0.301 mV), planar MEA (0.247 ± 0.128 mV), and high-nanodensity TNDEA (0.099 ± 0.1 mV).It can be explained by the cell-electrode interface model (Figure S8) that electrode impedance, which is usually much smaller compared to input impedance, 29 has little contribution to the improvement of the extracellular signal recording. 30Moreover, the high seal resistance, as a result of tight cell-electrode coupling, has significant contribution only to intracellular signal recording. 29,31ne the other hand, the amplitude of intracellular electrophysiological signals recorded by low-density TNDEA (amplitude of 1.688 ± 2.347 mV) possesses higher quality than those of planar MEA (amplitude of 0.827 ± 0.398 mV) and high-density TNDEA (0.81 ± 0.65 mV).Moreover, lownanodensity TNDEA showed much larger SNR in some individual recordings than those of high-nanodensity TNDEA and planar MEA, although the mean of SNR of low-nanodensity TNDEA was slightly larger than those of high-nanodensity NRMEA and planar MEA (Figure 3D).The maximum SNR increase of low-nanodensity TNDEA, planar MEA and high-nanodensity are 59.71 dB, 35.80 dB, and 30.23 dB, respectively (Figure 3E).Furthermore, compared to the planar MEAs, the low-nanodensity TNDEAs and high-nanodensity TNDEAs have higher performance in achieving higher rates of intracellular recording under the same condition (low-nanodensity nanopillars vs. planar vs. high-nanodensity: 98.38% ± 2.55% vs. 82.16%± 15.33% vs. 97.22%± 5.56%) as shown in Figure 3F.For more intuitive comparison, the distributions of amplitudes are demonstrated in Figure 3G.It can be observed that the intracellular action potentials recorded by low-nanodensity TNDEA possess a larger proportion in high amplitude than those recorded by planar MEA, and high-nanodensity TNDEA.Overall, low-nanodensity TNDEA showed excellent performance in electrophysiological signal recording of cardiomyocytes.It agrees with the cell-electrode interface model (Figure S8) since low-nanodensity TNDEA is shown to have tighter cellelectrode coupling that leads to higher seal resistance and better intracellular signal quality. 31esides nanodensity-regulating, other geometryregulating strategies have been explored (Figure S9).Regulating the diameters of the nanoporous polyethylene terephthalate membranes templates and adjusting the height of the Au-Al 2 O 3 nanopillar electrodes are two strategies that improve the intracellular action potentials signal recording quality.Although these strategies can also achieve better signal recording quality, the underlying parameters they optimize are different.They either only focus on adjusting the shape of electrodes to achieve better cell-electrode coupling or vary the shape and density of electrodes at the same time to obtain an overall better result.The effect of the density of electrodes as an essential parameter of signal recording quality is not clearly revealed.The nanodensity-regulating strategy, on the other hand, solely shows the optimization of electrode density to achieve high signal recording quality.It reveals the direct correlation between the electrode density and the improvement of signal recording quality.Additionally, among these three strategies, regulating nanodensity of TNDEA provides the best signal recording quality.
In short, effective 3D nanodensity regulation, as a strategy to solely optimize the electrode density, helps form a tight cell-nanostructure junction and can improve the rate of success to achieve high-quality intracellular recording, facilitating the studies on the characteristics of electrophysiology.

Performance of TNDEA in long-term intracellular recording
Duration time is another important parameter to assess the quality of intracellular recordings.We therefore performed long-term recording by low-nanodensity TNDEA, planar MEA, and high-nanodensity MEA to assess their capacity of long-term intracellular recordings.As shown in Figure 4A, low-density TNDEA allows continuous intracellular recordings for up to 83 min.After multisite electroporation by nanopillar electrodes, the amplitude of intracellular action potential surges to 2.62 ± 0.059 mV within 1 min followed by 10 min duration without significant change.The amplitude then gradually decays to a low value (1.52 ± 0.12 mV at 10 min, 277.1 ± 4.9 µV at 60 min, and 139.2 ± 33.85 µV at 83 min) (Figure 4B).The working time of intracellular access is defined as the time when the amplitude decays to 10% of the initial amplitude.Due to the electroporation-induced transient nanopores, the dramatical decrease of junction resistance of cell-nanopillars interface improves the intracellular recording, and the intracellular recording becomes weak due to the increased junctional resistance when the transient nanopores on cell membrane reseal.Notably, the action potential recorded by the planar MEAs and high-nanodensity TNDEAs decays fast, usually only in a few minutes (planar MEA:3.45 ± 1 min; high-nanodensity MEA: 0.7 ± 0.3 min) (Figure 4C).Meanwhile, the low-nanodensity TNDEA could significantly prolong the intracellular access time and achieve the long-term intracellular recording (12.7 ± 14.25 min).
This could be attributed to the enhanced multisite electroporation and delayed cell resealing due to the tighter cell-electrode coupling.In short, low-nanodensity TNDEA shows significant improvement of the intracellular recording quality and can be a powerful tool for cardiology.

CONCLUSIONS AND PERSPECTIVES
In this work, a universal 3D nanodensity regulating strategy for biosensing device is proposed to enhance performance of intracellular recording.Tunable nanodensity device is developed to achieve electrophysiological signal recordings of cardiomyocytes.Compared to the highnanodensity and planar TNDEA, the low-nanodensity TNDEA is demonstrated to possess superiority in intracellular biosensing based on the tight cell-electrode coupling for electroporation.The cell membrane can form tighter sealing with the 3D nanopillars on lownanodensity TNDEA, resulting in a larger contact specific surface area for the high-efficiency electroporation and higher seal resistance for the better signal recording.
The low-nanodensity TNDEAs can obtain high-quality and long-term intracellular electrophysiological signals by applying the low-voltage electroporation.Furthermore, this TNDEA device is developed by combining hydrothermal synthesis and standard microfabrication without complex fabrication operations, facilitating the large-scale and low-cost applications.To further improve the detection platform based on nanodevice, the following aspects can be investigated in future.(i) To further simplify the manufacturing process, the conductive nanopillar electrodes can be directly prepared by the conductive materials, such as Au and Pt.(ii) To obtain the gentle, stable, and long-term recordings, the morphology of the nanopillar and its substrate should be tunable to form a high seal and tight cell-nanopillar coupling biointerface.
(iii) Parameters including the thickness of gold conductive coating layer, SU-8 insulating layer, the size of recording site, and the arrangement of nanostructure have potential effects on cell-coupling and seal resistance, which can significantly influence signal quality of electrophysiological recording.Hence, these parameters in nanodevice fabrication should be further optimized to obtain a better performance on intracellular biosensing.(iv) To improve the biocompatibility of the nanopillar multielectrode device, some biocompatible photoresist with electrical insulation (e.g.silk fibroin 32 ) can be used instead of the SU-8 to produce the insulating layer and maintain a long-term

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflict of interest.

F I G U R E 3
Electrophysiological recordings of the primary neonatal rat cardiomyocytes by low-nanodensity tunable nanodensity electrode array (TNDEA), planar multielectrodes array (MEA), and high-nanodensity TNDEA electroporation.(A) Schematics of cells interfacing with low-nanodensity TNDEA, planar MEA, and high-nanodensity TNDEA.(B) Extra-and intra-cellular recordings of primary neonatal rat cardiomyocytes before and after electroporation with different nanodensities.(i) Spontaneous extracellular action potentials recording before electroporation and their partially enlarged views in dashed boxes.(ii) Intracellular recording of action potentials after electroporation and their partially enlarged views in dashed boxes.(C) Statistical plotting of mean amplitude of ten maximal intracellular and extracellular potentials from channels of low-nanodensity TNDEA, planar MEA, and high-nanodensity TNDEA.(D) Statistical plotting of signal-to-noise ratio (SNR) calculated from mean amplitude of the ten maximal intracellular and extracellular potentials from each channel of

F I G U R E 4
low-nanodensity TNDEA, planar MEA, and high-nanodensity TNDEA.(E) Statistical plotting of SNR changes before and after electroporation.For (C), (D) and (E), N ≥ 45. ****p < .0001 in one-way ANOVA and two-way ANOVA.(F) The intracellular yield of low-nanodensity and high-nanodensity TNDEA devices (N ≥ 4).(G) Statistical comparison showing the proportion of different amplitudes.Intracellular recording of cultured cardiomyocyte by the low-nanodensity tunable nanodensity electrode array (TNDEA).(A) Recording of intracellular action potentials at different time points (t = 1 min, 10 min, 83 min) and their corresponding enlarged views.(B) Action potential amplitude (APA) (n ≥ 23).(C) Plot of the duration time of low-nanodensity TNDEA, high-nanodensity TNDEA, and planar multielectrodes array (MEA).n ≥ 50.**** indicates p < .0001 in an unpaired t-test with Welch's correction.cellactivity and realize high-quality signal recording.The ongoing exploration of cardiomyocytes intracellular signal recording will eventually achieve a high-throughput, long-term, minimally invasive, and reliable detection platform for cardiology and pharmacology research.A U T H O R C O N T R I B U T I O N SX.L. performed the investigation, established the methodology, prepared the manuscript, and acquired funding.D.X. prepared the manuscript, performed the investigation, and established the methodology.M.Z.provided software, and established the methodology.J.F. performed the investigation and established the methodology.H.L. provided software, and established the methodology.W.Y., Y.W., and Z.X.prepared the manuscript.Y.W. prepared the manuscript.N.H. and D.Z. conceptualized the study, wrote a review of the literature, edited the paper, supervised the research, managed the project, and acquired funding.All the authors discussed the results and reviewed the manuscript.A C K N O W L E D G M E N T SThe work is supported in part by the National Natural Science Foundation of China (grant numbers: 62171483, 82061148011, and 62104264), Zhejiang Provincial Natural Science Foundation of China (grant number: LZ23F010004, LQ23C050004), Guangdong Basic and Applied Basic Research Foundation (grant numbers: 2020A1515110940, 2020A1515111210, and 2021A1515011609), and Key Research Project of Zhejiang Lab (grant number: 2022MH0AC01).