Highly Stretchable Hydrogels as Wearable and Implantable Sensors for Recording Physiological and Brain Neural Signals

Abstract Recording electrophysiological information such as brain neural signals is of great importance in health monitoring and disease diagnosis. However, foreign body response and performance loss over time are major challenges stemming from the chemomechanical mismatch between sensors and tissues. Herein, microgels are utilized as large crosslinking centers in hydrogel networks to modulate the tradeoff between modulus and fatigue resistance/stretchability for producing hydrogels that closely match chemomechanical properties of neural tissues. The hydrogels exhibit notably different characteristics compared to nanoparticles reinforced hydrogels. The hydrogels exhibit relatively low modulus, good stretchability, and outstanding fatigue resistance. It is demonstrated that the hydrogels are well suited for fashioning into wearable and implantable sensors that can obtain physiological pressure signals, record the local field potentials in rat brains, and transmit signals through the injured peripheral nerves of rats. The hydrogels exhibit good chemomechanical match to tissues, negligible foreign body response, and minimal signal attenuation over an extended time, and as such is successfully demonstrated for use as long‐term implantable sensory devices. This work facilitates a deeper understanding of biohybrid interfaces, while also advancing the technical design concepts for implantable neural probes that efficiently obtain physiological information.

resulting suspension was cooled to room temperature and then filtered to remove aggregated particles. The microgel suspension was purified by centrifugation to remove unreacted monomers. The cleaned microgels were stored in a brown glass jar for future use.

Synthesis of vinyl groups modified microgels
50 mL microgel suspension (1.2 mmol -COOH) was bubbled by N 2 for 30 min to remove O 2 from the system. EDC (19.2 mmol, 3.68 g) and NHS (4.8 mmol, 552 mg) were added to activate carboxyl groups. After stirring for 2 h, 3-butenylamine hydrochloride (2.4 mmol, 258 mg) was added, and then the pH of the suspension was adjusted to ~8 by TEA.
The reaction proceeded at room temperature for 8 h in an N 2 atmosphere. The microgel suspension was purified by centrifugation to remove unreacted monomers and catalysts. The cleaned microgels were stored in a refrigerator for future use.

Preparation of artificial cerebrospinal fluid (ACSF):
The ASCF used in our experiments was prepared according to the methodology described by Sheng et al. [2] The ASCF was prepared in deionized water containing (in mM): 125 NaCl, 2.5 KCl, 2 CaCl 2 ·2H 2 O 1.3 MgCl 2 ·6H 2 O 1.3 NaH 2 PO 4 , 25 NaHCO 3 , and 10 glucose·H 2 O. The solution was allowed to stand for 24 h and the supernatant was taken for use. The ACSF had a pH value of 7.45.

Synthesis of hydrogels
The hydrogels were synthesized by a micellar-copolymerization method. Briefly, SDS (0.58 g, 2.0 mmol) was dissolved in 30 mL ACSF, and the mixture was stirred for 20 min.
Then, LMA (1 mmol, 290 μL) and microgel suspension (0-7.5 g) were added to the SDS/ACSF solution. The mixture was sonicated for 30 min and stirred at room temperature for 3 h to make the microgels evenly dispersed. Subsequently, AAm (7.11 g, 0.10 mol) was (15 μL, 0.10 mmol) were further added with stirring for 5 min. The obtained mixture was injected into the mold consisting of two parallel glass plates and a 2 mm silicone spacer and then polymerized at 35 °C for 8 h. The compositions of hydrogels are listed in Table S1.

Characterizations
The Fourier transform infrared (FT-IR) spectra of microgels and vinyl groups modified microgels were obtained using a Nicolet 5700 spectrometer set at 4000-600 cm -1 with a resolution of 0.4 cm -1 . Nuclear magnetic resonance (NMR) spectra were recorded at room temperature using a Bruker Avance II spectrometer operating at a frequency of 500 MHz. The dispersions of microgels were concentrated and lyophilized into powders before the test.
Transmission electron microscope (TEM) analysis was performed on a JOEL JEM1400. The morphologies of the hydrogels were performed using a ZEISS GeminiSEM. The mechanical properties of the hydrogels were evaluated using a universal testing machine (Instron 5982).
Hydrogel specimens with a width of 2 mm were prepared for the tensile test at a crosshead rate of 100 mm/min and a standard length of ~20 mm. The values presented are the arithmetic means (± standard deviation) from at least three replicates of every sample. The compression test was carried out with the same instrument. The samples were prepared into a cylinder with a diameter of ~20 mm and a height of ~10 mm. The compression speed was set to 10 mm/min.
The resistance of the hydrogel sensors was characterized by combining the universal testing machine and an LCR digital bridge (TH2830, Tonghui Electronics Co., Ltd.). The sensitivity of the hydrogel strain sensor is characterized by the gauge factor (GF). The calculation formula of the GF is: where ΔR is the resistance change, R 0 is the original resistance, and ε is the strain of the sensor.
The ionic conductivity of the hydrogels was detected by an electrochemical workstation (CH Instruments) using AC impedance mode. The average value of the impedance at a scanning frequency range of 10 4 -10 5 Hz was regarded as the resistance of the test sample. The electrocardiogram (ECG) and electromyogram (EMG) signals were obtained by an RM6240EC multi-channel physiological signal acquisition system (Chengdu Instrument Factory), with signals acquired at an acquisition frequency of 8 kHz and 20 kHz, respectively.

Preparation of the hydrogel electrodes
The precursor solution of HM-2 was injected into pulled glass capillaries (Sutter) with a tip diameter of ~5 µm to prepare HM-2 microelectrodes for detecting neuronal spike. In addition, fused silica capillaries (320 μm in inner diameter and 450 μm in outer diameter) were immersed in the precursor solution of HM-2 to prepare HM-2 electrodes for detecting the local field potentials (LFPs). After polymerization at 35 °C for 8 h, all the prepared electrodes were immersed in deionized water for 7 days, and the water was changed every 12 h to remove unreacted monomers, after which the electrodes were stored in ACSF before use.

Ethical statements
The wearable tests were approved by the Institutional Ethics Committee of the First Hospital of Jilin University (approval number 2021059, Changchun, China) and were carried out according to the relevant institutional guidelines and laws.

Animal Ethics Statement
The animal tests were approved by the Institutional Animal Care and Use Committee of the First Hospital of Jilin University (approval number 20210872, Changchun, China) and were carried out according to the relevant institutional guidelines and laws. For the experiment, 36 adult female Sprague-Dawley (SD) rats (~220 g) were provided by Liaoning Changsheng Biotechnology Co. Ltd (Liaoning, China). All rats could drink water and eat food freely during the experiment period. The rats were fed in the laboratory for 1 week before operation to allow them to adapt to the new environment. All surgery procedures were performed under aseptic conditions.

Implantation of the HM-2 electrode
SD rats were anesthetized with 2% isoflurane and fixed on a rat brain stereotaxic device (Shanghai Yuyan Instruments Co., Ltd.). The head skin was cut with surgical scissors to expose the skull. A skull drill was used to drill a hole where the electrode was placed, then a hole on the opposite side was drilled and screwed in a stainless-steel screw to connect the ground wire. Subsequently, the HM-2 electrode was slowly inserted into the hippocampal CA1 area of the brain (AP-3.6mm, ML -2.0 mm, DV -2.7).

Recording LFPs of free-moving rats
After implanting the HM-2 electrode in the designated position, the part where the electrode was in contact with the skull was fixed with dental cement to prevent relative displacement. The prepared plastic cap was fixed on the skull with biomedical glue (3M Vetbond, 1469SB) to prevent the electrode from being damaged. Finally, the plastic cap was bonded with the terminal of electrodes together to make the whole device stronger. After the operation, the rat was placed in an electromagnetic shielding box with artificial light. After 48 hours of rest and adaptation to the environment, the rat's LFP signals were acquired. The detective channel was connected to a preamplifier (SWF-1W, Chengdu Instrument Factory).
The signals were amplified, acquired at a sampling frequency of 2 kHz, and filtered between 0.5 Hz and 300 Hz using an RM6240EC. Matlab 2018b software was used for signal processing.

Recording neuronal spikes of rats in vivo
The implantation method of the HM-2 microelectrode is the same as that of the HM-2 electrode described above. The HM-2 microelectrode was connected to a SWF-1W by a silver wire with a diameter of 300 μm. The signals were amplified, acquired at a sampling frequency of 40 kHz, and filtered between 500 Hz and 10 kHz using an RM6240E. The signals were threshold filtered to reduce noise. Matlab 2018b software was used for spikes sorting and signal processing. [3,4] The signal-to-noise ratio (SNR) was calculated by the formula: Where V s and V n are the mean value of the signal and noise voltages, respectively.

Immunohistochemistry and imaging of HM-2, platinum, and silver interfaces
In each rat tested, a hole was drilled in the skull and the implant electrodes (HM-2, platinum, or silver) were inserted in the hippocampus CA1 area (AP: −3.6 mm, ML: 2.0 mm, DV: −2.7 mm from bregma). After waiting for 1 week, the rats were perfused for the immunochemical experiment. Firstly, the mouse was deeply anesthetized with sodium pentobarbital (1% wt/vol) and was transcardially perfused with saline buffer followed by a fixative solution. Then the brain was placed at 4 o C for 6-8 h fixation. Next, the brain was dehydrated with sucrose (25% wt/vol) overnight. Finally, coronal cryosections were cut at 50 µm on a freezing microtome (Leica RM2016) for confocal imaging and immunostaining.

Counting the number of glial cells and calculating mean fluorescence intensity.
The cell count refers to the reported method. [2] Three persons counted glial cells on each confocal image, and the median of the three counts was used to represent the image. To compare fluorescence intensity, confocal images were acquired under the same microscope parameter, and the fluorescence intensity was calculated by Image J under the same parameter. Figure S1. IR spectra of microgels.            Figure S13. Confocal micrographs of the brain slice around the electrodes after 4 weeks of implantation in rats' brains: microglia cells (green) and astrocyte cells (red).