Crosslinked hyaluronic acid-doped polypyrrole: Stable, nonbiofouling implantable bioelectrodes for in vivo signal recording

Abstract

Implantable bioelectrodes with biomimetic and antibiofouling properties are essential for reliable long-term signal acquisition in biomedical applications. Hyaluronic acid (HA), a natural biopolymer, has been incorporated into conductive polymers such as polypyrrole (PPy) to produce biomimetic bioelectrodes with enhanced biocompatibility. However, the poor enzymatic stability of HA limits the longevity of HA-doped PPy (PPy/HA) electrodes, resulting in fibrosis and the loss of signal sensitivity. In this study, we developed PPy electrodes doped with crosslinked HA (PPy/cHA) to improve their enzymatic resistance and electrochemical stability in vivo. PPy/HA electrodes were prepared via electrochemical polymerization and subsequently treated with 1,4-butanediol diglycidyl ether to covalently crosslink the surface-exposed HA moieties. This post-crosslinking did not significantly alter the surface morphology, hydrophilicity, impedance, or nonfouling properties of the electrodes. Importantly, upon hyaluronidase treatment, the PPy/HA electrodes exhibited increased impedance and fibroblast adhesion, whereas PPy/cHA retained its original physicochemical and antibiofouling characteristics. Subcutaneous implantation in mice for three weeks revealed significantly reduced fibrotic tissue around the PPy/cHA electrodes. Furthermore, electrocardiogram monitoring revealed that compared to controls, the PPy/cHA electrodes sustained stable signal transmission with a higher signal-to-noise ratio over three weeks. These results highlight the potential of the PPy/cHA electrodes as reliable, nonbiofouling, and implantable bioelectrodes for diverse biomedical applications.

Graphical abstract

1. Introduction

Implantable bioelectrodes facilitate precise transmission of electrical signals between biological tissues and electronic devices [1]. Various types of implantable bioelectrodes have been developed for various purposes, such as neural recording [2], cardiac monitoring [3], and biosensing [4]. Conducting polymers, particularly polypyrrole (PPy), have emerged as promising bioelectrode materials owing to their high electrical conductivity, biocompatibility, and diverse modifications [5]. Recently, substantial efforts have been made to fabricate biomimetic PPy-based bioelectrodes by incorporating extracellular matrix (ECM) components, such as hyaluronic acid (HA), to improve tissue integration and minimize adverse foreign body reactions [[6], [7], [8], [9]]. However, HA-based biomaterials, including HA-doped PPy bioelectrodes, suffer from rapid enzymatic degradation by hyaluronidase (HAase) and hydrolysis, leading to the loss of their original biocompatibility, thereby limiting their reliability and functionality in vivo [[9], [10], [11], [12]]. Consequently, novel strategies for enhancing the stability of HA-doped PPy electrodes, while preserving their original beneficial properties, are highly desirable.

One of the major challenges in developing implantable bioelectrodes is biofouling, which involves the nonspecific adsorption of proteins and cells onto electrode surfaces [13,14]. Biofouling generally increases interfacial resistance, triggers inflammatory responses, and ultimately compromises signal sensitivity and longevity [15,16]. Previously, we demonstrated that HA-doped PPy (PPy/HA) electrodes exhibited excellent non-cell-adhesive characteristics and reduced fibrotic encapsulation after implantation [6]. The incorporation of HA, a naturally occurring polyanionic polysaccharide, into PPy can enhance its hydrophilicity and antibiofouling abilities. However, the HA in PPy/HA is highly susceptible to enzymatic degradation. HA degradation by HAase treatment impairs the electrical/electrochemical properties of PPy/HA due to a decrease in the interfacial charge density. Importantly, the PPy/HA electrodes eventually lose their nonbiofouling properties and exhibit eventual decreases in signal sensitivity in vivo after implantation in the body, likely due to degradation of the HA moieties [9].

Generally, HA-based implants undergo enzymatic and oxidative degradation after implantation, resulting in substantial changes in their original functional and structural characteristics [17,18]. Consequently, extensive efforts have been made to enhance the stability of HA-based biomaterials, including PPy/HA [19]. For example, the chemical crosslinking of HA hydrogels using 1,4-butanediol diglycidyl ether (BDDE) or divinyl sulfone has been widely employed to improve the resistance against enzymatic degradation and hydrolysis [[20], [21], [22]]. Moreover, chemical modifications of HA, including sulfation, methacrylation, and hydrophobic conjugation, have been investigated to enhance its structural integrity and prolong its in vivo functionality [23,24]. However, most previous approaches focused on tissue engineering or scaffold applications, and their effects on the electrochemical stability of HA-based bioelectrodes remain unexplored.

In this study, we developed crosslinked HA-doped PPy electrodes (PPy/cHA) to enhance the enzymatic resistance while maintaining the electrochemical stability and nonbiofouling properties. Unlike conventional bulk HA hydrogels used in tissue engineering or dermal fillers, our strategy specifically targets the free HA moieties exposed at the bioelectrode interface. This localized surface stabilization is uniquely tailored for implantable bioelectronics; it preserves the bulk electrical conductivity of the underlying conductive polymer while preventing degradation-induced impedance increase, thereby promoting long-term performance. Specifically, PPy/HA was crosslinked using BDDE, a widely used crosslinker, to reduce its susceptibility to enzymatic hydrolysis. By crosslinking the HA moieties present on the PPy electrodes, we hypothesized that PPy/cHA would have superior resistance to HAase-mediated degradation and reduce the degradation-associated impedance increase, while preserving its hydrophilic, nonbiofouling nature.

We systematically evaluated the surface characteristics, electrochemical properties, and biological performance of PPy/cHA electrodes and compared them with those of noncrosslinked PPy/HA electrodes. Finally, we assessed the extended signal recording capability of PPy/cHA by monitoring electrocardiogram (ECG) signals in vivo over two weeks to assess its reliability and stability (Fig. 1).

Fig. 1.

Schematic illustration of the fabrication, nonbiofouling property, and performance of BDDE-crosslinked hyaluronic acid-doped polypyrrole (PPy/cHA) bioelectrodes for long-term high-quality electrical signal recording in vivo.

2. Materials and methods

2.1. Materials

Pyrrole, BDDE, aluminum oxide, phosphate-buffered saline (PBS), toluidine blue O (TBO), HAase from bovine testes (400–1000 U/mg), sodium chloride, sodium hydroxide, hydrochloric acid, acetic acid, and fetal bovine serum (FBS) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ultra-high molecular weight HA (3 × 10⁶ Da) was obtained from LG Chem (Seoul, Republic of Korea). Dulbecco's modified Eagle's medium (DMEM) and Dulbecco's phosphate-buffered saline (DPBS) were purchased from Welgene (Gyeongsan, Republic of Korea). The EZ-Cytox assay kit (WST-1; 4-[3-(4-Iodophenyl)-2-(4-nitro-phenyl)-2H-5-tetrazolio]-1,3-benzene sulfonate) was purchased from DoGenBio (Seoul, Republic of Korea). Trypsin/EDTA (0.05%) and antibiotic–antimycotic (100 × ) was obtained from Gibco (Grand Island, NY, USA). Acetone and methanol were purchased from Duksan (Seoul, Republic of Korea). A LIVE/DEAD™ viability/cytotoxicity kit for mammalian cells was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Hematoxylin & eosin Y (H&E) and Masson's trichrome staining kits were purchased from BBC Biochemical (Mount Vernon, WA, USA).

2.2. Electrochemical synthesis of PPy/HA electrodes

Glass slides were cleaned by immersion in acetone and methanol for 10 min each in a bath sonicator (Branson 2510, St. Louis, MO, USA). The cleaned glass slides were coated with chromium (8 nm) and gold (70 nm) by using a DC sputtering system (KVS-2003L; Korea Vacuum Tech, Gimpo, Republic of Korea). The gold-coated slides were further cleaned by immersion in acetone and methanol for 10 min each in a bath sonicator. Pyrrole was purified by passing it through an alumina oxide column prior to use. A polymerization solution containing 0.1 M pyrrole and 2 mg/mL of HA in double-deionized water (DDIW) was prepared and sonicated for 360 s. Electrochemical polymerization was performed using a potentiostat (VersaSTAT 3, Princeton Applied Research, Oak Ridge, TN, USA) and a three-electrode setup with a platinum mesh as the counter electrode and a saturated calomel electrode (SCE) as the reference electrode. PPy/HA (1 × 1 cm) was electrochemically deposited onto a gold-coated glass slide in the chronopotentiometry mode at a constant current density of 1 mA/cm² for 500 s to a final charge density of 50 mC/cm². The electrodes were washed with DDIW and incubated in DDIW at 25 °C for 2 days to remove residual monomers. For electrochemical impedance spectroscopy (EIS) and cyclic voltammetry (CV), the gold electrode was covered with a tape film with a hole in the center to achieve an identical electrode area (0.283 cm²) for each sample and was used for the electrochemical deposition of PPy/HA under the same conditions as described above.

2.3. BDDE-mediated crosslinking of PPy/HA electrodes

BDDE-mediated crosslinking of PPy/HA was performed by incubation in an aqueous solution containing 10 v/v% BDDE (in 0.25 M NaOH) at 37 °C for 24 h. Then, residual BDDE was removed by incubation of PPy/cHA in DDIW for 2 days before use.

2.4. Hyaluronidase treatment of the electrodes

For the enzymatic degradation study, PPy/HA and PPy/cHA electrodes were incubated in a HAase solution (2.5 U/mL in PBS) at 37 °C for 24 h, followed by three washes with DDIW.

2.5. Surface characterization of the electrodes

The surface roughness of each sample was analyzed using atomic force microscopy (AFM, XE-100, Park Systems, Seoul, Republic of Korea) in non-contact mode at a scan rate of 1 Hz. The surface roughness (R_a) was calculated using XEI software 4.3 (Park Systems, Republic of Korea).

Time-of-flight secondary ion mass spectrometry (ToF-SIMS) (ION-TOF, Münster, Germany) was used to analyze and compare the surface chemical compositions of PPy/HA and PPy/cHA. Prior to analysis, the PPy films were thoroughly dried in a dry oven at 25 °C. Positive secondary ion spectra were acquired from the film surface using a 25 keV Bi₃⁺ liquid metal ion gun over a 100 × 100 μm area with a target current of 0.4 pA.

The water contact angle (WCA) was measured using a goniometer (Phoenix 300, SEO Co., Suwon, Republic of Korea) equipped with a digital camera. A droplet (5 μL) of DDIW was dropped at 25 °C on the surface of each sample, and images were acquired. The obtained images were analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) to determine the WCA.

Free carboxyl groups on each electrode surface were quantified by TBO assay. Each sample was incubated in an aqueous TBO solution (0.005% TBO, 0.2% NaCl, and 0.01 M HCl) at 25 °C for 10 min. The samples were then washed with 10 mM NaOH solution to remove unbound TBO. Bound TBO was extracted by incubating each sample in a 50% acetic acid solution for 20 min. The amount of TBO in the extracted solution was quantified by measuring the absorbance at 633 nm using ultraviolet–visible (UV-Vis) spectroscopy (Biodrop Duo, Biodrop, UK).

2.6. Electrochemical characterization of the electrodes

EIS was performed to measure the electrochemical impedance of each electrode using the aforementioned three-electrode system. The EIS spectra were recorded in DPBS in a frequency range of 10⁻¹ to 10⁵ Hz with an applied sinusoidal voltage of 10 mV and a DC potential of 0 V versus the open circuit potential. The impedance spectra were analyzed and fitted using ZView software (AMETEK, Inc., Berwyn, PA, USA) to extract the circuit parameters.

2.7. In vitro cell adhesion

Mouse NIH-3T3 fibroblasts (Korean Cell Line Bank, Seoul, Republic of Korea) were seeded on each sample to evaluate cell adhesion. The cells were maintained in a high-glucose DMEM supplemented with 10% FBS and 1% antibiotic–antimycotic in a humidified 5% CO₂ incubator at 37 °C. Before cell seeding, the samples were sterilized by incubation in ethanol for 10 min, followed by two washes with sterile DPBS. Cells were seeded at a density of 2.0 × 10⁴ cells/cm² and further incubated in a humidified 5% CO₂ incubator at 37 °C for 24 h. Then, each sample was carefully washed twice with DPBS to remove non-adherent cells. Optical microscopic images were obtained, and the number of adherent cells was counted from randomly selected images using ImageJ software (n = 4).

2.8. In vitro cytocompatibility

The cytotoxicities of the PPy/HA and PPy/cHA samples were evaluated in accordance with ISO 10993-12. Each sample was incubated in the culture medium at 37 °C in a humidified incubator for 24 h to obtain the extracts. NIH-3T3 cells were seeded in 24-well plates at a density of 4.0 × 10⁴ cells/well and cultured for 24 h. The cells were subsequently incubated with the extract solutions for an additional 24 h. Cytocompatibility was assessed by the LIVE/DEAD assay according to the manufacturer's instructions (Thermo Fisher Scientific). In brief, each sample was incubated with the staining solution (4 nM EthD-1 and 2.5 μM calcein AM in DPBS) at 37 °C for 10 min and washed twice in DPBS for 5 min. Fluorescence images were acquired using a fluorescence microscope (Leica DMI 3000B). The numbers of live (green) and dead (red) cells were counted using ImageJ software, and cell viability was determined as the percentage of live cells relative to the total cell number. Cell metabolic activity was evaluated using a WST-1 assay kit. Briefly, WST-1 solution was diluted with culture medium at a 1:10 ratio, transferred to each well, and incubated at 37 °C for 1 h. Subsequently, the supernatant was transferred to 96-well plates, and the absorbance was measured at 450 nm using a microplate reader.

2.9. In vivo subcutaneous implantation of the electrodes

All animal experiments were performed in accordance with the permission of the Committee on Animal Research and Ethics at the Gwangju Institute of Science and Technology (GIST), Republic of Korea (approval number: GIST-2023-005). Gold-coated epoxy electrodes (11 × 2 mm) were fabricated by spin-coating SU-8100 onto a glass substrate at 1500 rpm, followed by sequential precuring (30 min at 65 °C and 30 min at 95 °C), ultraviolet light exposure (1 min), post-baking (90 min at 95 °C), and development in isopropyl alcohol. A 50-nm-thick gold layer was then coated onto the cured epoxy probe by sputtering, and the electrodes were peeled off the substrate. For implantation, only the tapered tip region (6.25 mm × 1 mm) of the gold-coated epoxy electrodes was inserted into the subcutaneous tissue. The PPy/HA and PPy/cHA electrodes were synthesized via electrochemical deposition under the conditions described in Sections 2.2, 2.3). Prior to implantation, the samples were sterilized by immersion in 70% ethanol for 1 h, followed by washing with sterile DPBS. The dorsal region of six-week-old male BALB/c mice was shaved, and the shaved area was disinfected with 70% ethanol and 10% povidone–iodine solution. The mice were anesthetized with 2% isoflurane, and subcutaneous pockets were created on the left and right sides of the dorsal midline incision site. Gold, PPy/HA, or PPy/cHA electrodes were implanted into these pockets. The skin was then sutured using silk sutures (6-0). Five mice were used for each experimental group.

2.10. In vivo tissue compatibility

Twenty-one days after implantation, the mice were sacrificed. Skin and muscle tissues surrounding the implanted electrodes were harvested. The harvested tissues were fixed in 3.7% formaldehyde for 30 min, washed with DPBS, and embedded in paraffin. The paraffin-embedded samples were sectioned into 10-μm-thick slices using a microtome (RM2235, Leica, Germany). Masson's trichrome and H&E staining were performed according to the manufacturer's protocols (BBC Biochemical). Optical microscopic images of stained tissue slides were obtained using an optical microscope (Leica DM 2500). The collagen-stained (blue) areas surrounding the samples in the Masson's trichrome-stained images were analyzed using ImageJ software.

2.11. In vivo ECG measurements

Gold-coated polyimide film electrodes were fabricated to record in vivo ECG signals. The flexible polyimide films were cleaned by thorough washing with isopropyl alcohol, acetone, and DDIW. Chromium (10 nm) and gold (100 nm) layers were successively deposited onto the cleaned polyimide films using a DC sputtering system. The fabricated electrodes were encapsulated with polyimide tape, in which the contact electrode area (1.25 × 1.25 mm) was uncovered for exposure to the tissue. PPy/HA films were electrochemically polymerized onto gold-coated polyimide electrodes at a current density of 1 mA/cm² to a final charge density of 50 mC/cm², which were further crosslinked to the PPy/cHA electrodes, as described in Section 2.3. The mice were anesthetized with 2% isoflurane. The dorsal region of each mouse was shaved, and the shaved area was disinfected with 70% ethanol and 10% povidone-iodine solution. Two identical electrodes from each experimental group (gold, PPy/HA, and PPy/cHA) were subcutaneously implanted into each side of the dorsal region. The skin was sutured using silk sutures (6-0). A needle-type ground electrode was inserted into the tail of each mouse as a ground electrode during the ECG recording. ECG signals were recorded for up to 21 days using an MP36 data acquisition system (Biopac Systems, Inc., CA, USA) and analyzed using Biopac Student Lab 4.1 software. The signal-to-noise ratio (SNR) of the recorded ECG signal was calculated using the following equation:

$$\text{SNR}\left(\text{dB}\right)=20\times {\log }_{10}\left(\frac{\text{Peak}‐\text{to}‐\text{peak}\text{amplitude}\left(\mathrm{V}\right)}{\text{Noise}\text{amplitude}\left(\mathrm{V}\right)}\right)$$

2.12. Statistical analysis

All experiments were performed at least three times, unless otherwise noted. Statistical significance was evaluated using one-way analysis of variance (ANOVA), followed by Tukey's post hoc test for multiple comparisons at a significance level of 0.05 using Origin Pro software (version 2024, OriginLab Corporation, Northampton, MA, USA).

3. Results and discussion

3.1. Synthesis and characterization of PPy/cHA

PPy/HA was electrochemically synthesized on gold electrodes in a chronopotentiometry mode at a constant current density of 1 mA/cm² for 500 s (Fig. S1). PPy/cHA was further prepared by incubating PPy/HA in 10% BDDE solution at 37 °C for 24 h (Fig. 2a). The epoxy groups of BDDE preferentially react with the hydroxyl residues of the HA backbone under alkaline conditions, resulting in the formation of stable ether bonds between the HA chains and BDDE [25]. BDDE has been widely used to crosslink HA, and several BDDE-crosslinked HA hydrogels have been approved by the FDA for biomedical and cosmetic applications [26]. While BDDE is traditionally utilized to form bulk 3D hydrogel networks, in our study, we established an enzymatically stable protective layer by selectively crosslinking the surface-exposed HA dopants for the specific interfacial engineering of PPy/HA.

Fig. 2.

Surface characterization of PPy/HA and PPy/cHA electrodes. (a) Schematic illustration of electrode surface modifications to fabricate PPy/cHA. (b) Atomic force micrographs of PPy/HA and PPy/cHA. Scale bar = 1 μm. (c) Surface roughness (R_a) of PPy/HA and PPy/cHA. (d) ToF-SIMS analysis of PPy/HA and PPy/cHA. (e) Relative peak intensities of fragment ions (m/z= 45, 59, 113, and 131). (f) WCA measurement of gold, PPy/HA, and PPy/cHA. (g) Surface carboxyl group quantification of PPy/HA and PPy/cHA. (h) Electrochemical impedance spectra (Bode plots) of gold, PPy/HA, and PPy/cHA. An asterisk (∗) denotes a statistically significant difference (p < 0.05). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

The surface morphology and roughness of PPy/HA and PPy/cHA were analyzed using atomic force microscopy (AFM) (Fig. 2b). Both PPy/HA and PPy/cHA exhibited typical granular morphologies with no noticeable differences. A quantitative analysis of the average surface roughness (R_a) indicated no structural changes in PPy/HA after BDDE crosslinking (Fig. 2c). These results suggest that the crosslinking of surface HA does not alter the intrinsic nano-/microstructure of the PPy/HA surface and does not introduce substantial structural irregularities. The preservation of surface roughness is especially important because the electrochemical properties and non-cell-adhesive characteristics of PPy-based electrodes can be substantially influenced by the interfacial topography [[27], [28], [29]]. In addition, maintaining an appropriate surface roughness is critical for ensuring the appropriate functions of the implanted electrodes. For example, moderate roughness has been reported to support stable charge transfer and attenuate cell adhesion by reducing the number of favorable sites for nonspecific protein adsorption [30]. In contrast, excessively rough surfaces can increase potential trapping sites and promote protein adsorption, which impedes charge transport and induces immune responses at the tissue–electrode interface [31].

To investigate the functional groups of the fabricated electrodes, FT-IR and Raman spectroscopy were first conducted (Fig. S2 and S3). The FT-IR spectra of both PPy/HA and PPy/cHA exhibited characteristic peaks at 3320 cm⁻¹ (O-H and N-H stretching), 2920 cm⁻¹ (symmetric C-H stretching), 1537 cm⁻¹ (C=C and C-C stretching of the pyrrole ring), 1402 cm⁻¹ (symmetric stretching of COO⁻ in HA), and 1021 cm⁻¹ (C-O-C stretching) [32,33]. Similarly, Raman spectroscopy confirmed the structural presence of the PPy and HA components. Notably, no distinct spectral differences were observed between PPy/HA and PPy/cHA in both bulk analyses. Because the BDDE crosslinking is strictly localized to the surface-exposed HA moieties, its relatively small chemical footprint may be masked by the dominant signals of the underlying bulk PPy/HA network. This confirms that the crosslinking strategy preserves the bulk chemical integrity of the bioelectrodes. Consequently, to precisely verify the chemical changes at the extreme surface, highly surface-sensitive ToF-SIMS analysis was employed for the PPy/HA and PPy/cHA samples (Fig. 2d). The SIMS spectra revealed distinct peaks corresponding to the fragments derived from HA, BDDE, and PPy. For the PPy/HA and PPy/cHA samples, prominent peaks were observed at m/z values of 45, 59, 113, and 131. The peak at m/z 45 was assigned to a carboxylic fragment (COOH⁺) originating from the HA- and BDDE-related fragments (C₂H₅O⁺ and CH₃CHO⁺). The peak at m/z 59 appeared to be attributed to BDDE fragments (C₃H₇O⁺) and HA-derived fragments (CH₃O₂⁺). The peaks at m/z 113 and 131 were presumed to originate from HA- and HA-BDDE-derived fragments, such as C₆H₇O₂⁺ and C₇H₉O₂⁺. After BDDE crosslinking (PPy/cHA), the intensities of these peaks significantly increased (Fig. 2e). For example, the C₂H₅O₂⁻ signal increased by approximately 2.5-fold. This increase in signal intensity suggests successful BDDE crosslinking of the HA moieties on PPy. Moreover, highly surface-sensitive X-ray photoelectron spectroscopy (XPS) was performed (Fig. S4). The high-resolution C_1s spectra revealed noticeable changes in the surface composition after crosslinking. For PPy/cHA, the peak area associated with C-N and C-O bonds (∼286.3-286.5 eV) increased to 21.0 ± 0.7% compared to 12.3 ± 0.9% in PPy/HA, whereas the C=O peak (∼287.7-287.8 eV) decreased from 19.1 ± 1.2% to 11.3 ± 1.2%. As a result, the area ratio of (C-N, C-O) to C=O increased by approximately 2.9-fold after crosslinking. Because the BDDE crosslinker introduces numerous ether (C-O-C) linkages, this significant increase in the C-O component provides supporting evidence of successful BDDE-mediated crosslinking of the HA moieties at the electrode surface.

The surface hydrophilicity of the electrodes was evaluated by measuring the WCAs of the gold, PPy/HA, and PPy/cHA surfaces (Fig. 2f and Fig. S5). The gold electrode exhibited a relatively high WCA of approximately 53.3 ± 2.9°, while both PPy/HA (24.9 ± 2.9°) and PPy/cHA (24.5 ± 3.7°) showed significantly lower WCAs, confirming the highly hydrophilic nature of HA-doped PPy. The difference between PPy/HA and PPy/cHA was not statistically significant, suggesting that the BDDE crosslinking process did not compromise the inherent hydrophilicity of HA on the electrode surface. The quantification of the surface carboxyl groups indicated no significant differences between PPy/HA and PPy/cHA (Fig. 2g).

EIS was conducted to assess the electrical and electrochemical properties relevant to bioelectronic performance (Fig. 2h). The PPy/HA and PPy/cHA electrodes exhibited significantly lower impedances in the low-frequency region (<10 Hz) than the gold electrode. Moreover, the impedance of the PPy/cHA electrode was similar to that of the PPy/HA electrode. For example, impedance values at 1 Hz were 22693 ± 4335, 1777 ± 166, and 2057 ± 257 Ω for the gold, PPy/HA, and PPy/cHA electrodes, respectively (Fig. S6). These results suggest that PPy/HA deposition significantly improves charge transport at the interface and that crosslinking does not significantly alter the electrochemical properties of PPy/HA [34].

3.2. Enzymatic stability of PPy/cHA electrodes in HAase environments

The enzymatic stability of the surface-functionalized electrodes was evaluated by treating gold, PPy/HA, and PPy/cHA electrodes with HAase (Fig. 3a). As described earlier, the untreated gold electrode exhibited a high WCA (53.3 ± 2.9°), and the HA-incorporated PPy (PPy/HA (24.9 ± 2.9°) and PPy/cHA (24.5 ± 3.7°)) exhibited markedly lower WCAs. Upon HAase treatment, a clear difference in the WCAs was observed between the PPy/HA and PPy/cHA electrodes. The WCA of PPy/HA increased significantly to 34.7 ± 0.9°, suggesting substantial degradation of surface-bound HA and the subsequent loss of hydrophilicity. In contrast, the WCA of PPy/cHA remained similar (27.2 ± 1.4°), indicating that BDDE crosslinking effectively protected HA from enzymatic cleavage. Hence, BDDE-crosslinked HA is expected to maintain its surface hydration and anti-biofouling properties under enzymatic physiological conditions. We have previously reported that the incorporation of glycyrrhizin, a hyaluronidase inhibitor, into PPy/HA electrodes supported the preservation of surface functionality against HAase [9]. Kim et al. demonstrated that BDDE-crosslinked HA hydrogels retained their hydration and functional group integrity after exposure to HAase [35].

Fig. 3.

Surface characterization of PPy/HA and PPy/cHA after hyaluronidase (HAase) treatment. (a) WCAs of gold, PPy/HA, and PPy/cHA before and after HAase treatment. (b) Quantification of surface carboxyl groups on PPy/HA and PPy/cHA after HAase treatment. (c) Electrochemical impedance spectra of gold, PPy/HA, and PPy/cHA electrodes with or without HAase treatment. (d) Relative impedance changes (%) at 1 Hz for PPy/HA and PPy/cHA electrodes after HAase treatment. The impedance value at 1 Hz of each electrode was normalized to that of HAase non-treated PPy/HA. (e) Representative optical micrographs of NIH-3T3 fibroblasts adhered on gold, PPy/HA, and PPy/cHA with and without HAase treatment. Scale bar = 200 μm. (f) Quantification of adhered cell numbers. An asterisk (∗) denotes a statistically significant difference (p < 0.05). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

We further quantified the surface carboxyl groups to assess the chemical stability of HA on the electrode against HAase (Fig. 3b). Upon exposure to HAase, the carboxyl group density of PPy/HA significantly decreased from 1.71 ± 0.12 μmol/cm² to 1.56 ± 0.10 μmol/cm² (p < 0.05). In contrast, PPy/cHA maintained its carboxyl content even after HAase treatment (1.58 ± 0.05 μmol/cm² after treatment vs. 1.64 ± 0.07 μmol/cm² initially). This indicates that BDDE crosslinking effectively protects HA chains from enzymatic hydrolysis [25].

The influence of enzymatic degradation on the electrochemical properties of the PPy/HA electrodes was examined by EIS measurements (Fig. 3c and Fig. S7) as, in aqueous physiological environments, the interfacial impedance and charge transfer resistance effectively represent the functional electrical conductivity of the bioelectrodes. Bode plot analysis revealed that the gold electrode exhibited the highest impedance in the low-frequency region (<10 Hz). Preservation of low impedance, especially at low frequencies, is critical for bioelectrodes operating in enzymatically active environments because increases in interfacial impedance degrade slow-signal components (e.g., local field potentials or signal baseline) and reduce the SNR of signals [36]. For PPy/HA, HAase treatment led to a marked increase in impedance at all tested frequencies, likely because enzymatic degradation of surface-bound HA impaired the charge transfer efficiency, resulting in increased charge transfer resistance at the electrode interfaces (Table S1). Enzymatic cleavage of the surface HA chains leads to disruption of the hydration layer and an increase in surface resistance [9]. In contrast, PPy/cHA demonstrated much greater electrical signal transduction stability than PPy/HA. For example, no significant changes in the impedance were observed after HAase treatment of the PPy/cHA electrodes, whereas the PPy/HA electrodes exhibited a 1.8-fold increase in the impedance at 1 Hz after HAase treatment(Fig. 3d). The crosslinked HA layer on PPy protects HA from enzymatic degradation and preserves the hydrated polymeric interface and efficient electrical transmission at the interface. This stabilization is beneficial for maintaining a low interfacial resistance, especially at low frequencies, which are sensitive to the ionic mobility and capacitive behavior at the biointerface. Hence, the excellent resistance of the PPy/cHA electrodes to enzymatic degradation was anticipated to offer consistent signal quality, minimize drift, and extend device longevity. For deeper insight into the underlying charge transfer mechanisms, we analyzed the equivalent circuit fitting parameters (Table S1). The preservation of low impedance in PPy/cHA is mechanistically driven by the stabilization of interfacial charge density, double-layer capacitance, and ion mobility. First, the polyanionic nature of HA provides a high interfacial charge density via its abundant carboxyl groups, which attract cations and facilitate charge transfer. Because BDDE crosslinking preserves these carboxyl groups, PPy/cHA maintains its charge density even after HAase treatment. Second, the crosslinked HA network protects the double-layer capacitance. The Constant Phase Element (CPE-T) representing capacitance dropped drastically for PPy/HA (from 1.44 × 10⁻⁵ to 8.33 × 10⁻⁶) upon enzymatic cleavage due to the loss of the hydrated layer, whereas PPy/cHA retained its capacitive behavior (1.70 × 10⁻⁵). Finally, the stabilized hydration layer acts as an effective aqueous pathway that sustains high ion mobility. Consequently, the Warburg resistance (W-R), which relates to ion diffusion, increased by nearly two-fold for PPy/HA after HAase treatment, while PPy/cHA successfully resisted such severe increases in diffusion resistance. Collectively, these results confirm that the BDDE-crosslinked HA layer mechanically and chemically secure the charge transfer pathways at the biointerface.

The antifouling ability of the electrodes was examined using NIH-3T3 fibroblasts with and without HAase treatment. The gold electrode showed extensive nonspecific fibroblast attachment, forming a dense confluent monolayer in 1 day of culture. In contrast, both the PPy/HA and PPy/cHA surfaces showed substantially reduced cell adhesion under HAase-free culture conditions, demonstrating excellent nonbiofouling properties originating from HA. Importantly, a clear difference in the cell attachment behavior between PPy/HA and PPy/cHA was observed after HAase treatment (Fig. 3e, f). The number of fibroblasts adhered to PPy/HA dramatically increased from 41.0 ± 3.4 × 10⁴ cells/mm² to 170.5 ± 27.1 × 10⁴ cells/mm² (p < 0.001), indicating substantial enzymatic degradation of surface HA moieties and loss of a non-cell-adhesive character. In contrast, the PPy/cHA surfaces retained their ability to resist cell adhesion even after HAase treatment. The number of adherent cells (8.5 ± 3.3 × 10⁴ cells/mm²) remained minimal and was not statistically different from that in the untreated PPy/cHA group (3.8 ± 1.4 × 10⁴ cells/mm²). These results clearly suggest that BDDE crosslinking enhances the anti-adhesive surface properties.

3.3. In vitro cytocompatibility test

The cytocompatibility of the electrode materials was investigated by culturing NIH-3T3 cells in their individual extract solutions. Live/dead staining revealed that fibroblasts treated with extracts from all three electrode types exhibited a predominantly green fluorescent signal, indicating high cell viability (Fig. 4a). Quantitative analysis indicated that the cell viability exceeded 99% in all groups, with no statistically significant differences between them (p > 0.05) (Fig. 4b). Similarly, the WST-1 assay showed high metabolic activity levels in all extract-treated groups (Fig. 4c). These results confirm that both PPy/HA and PPy/cHA are cytocompatible and do not adversely affect mitochondrial function. The excellent cytocompatibility of PPy/HA and PPy/cHA can be attributed to their inherent biocompatibility (PPy and HA) and appropriate production procedures. Notably, BDDE-crosslinked HA hydrogels have been widely shown to be noncytotoxic [[37], [38], [39]].

Fig. 4.

In vitro cytotoxicity tests. (a) Representative live/dead fluorescence images of NIH-3T3 fibroblasts cultured in extract media obtained from gold, PPy/HA, and PPy/cHA electrodes after 24 h of incubation. Live and dead cells are stained green and red, respectively. Scale bars = 200 μm. (b) Live cell percentages (cell viability). (c) Metabolic activity of cells assessed by WST-1 assay after 24 h incubation with extract media from gold, PPy/HA, and PPy/cHA electrodes. An asterisk (∗) denotes a statistically significant difference (p < 0.05). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.4. Tissue compatibility of implanted bioelectrodes

Inflammatory tissue responses to the electrodes (gold, PPy/HA, and PPy/cHA), including fibrotic encapsulation, were examined by subcutaneous implantation. Three weeks post-implantation, the surrounding tissues were harvested and stained with H&E and Masson's trichrome (Fig. 5a). H&E staining revealed fibrous tissue formation surrounding all the implants. The gold electrodes exhibited a pronounced inflammatory response, as indicated by the dense infiltration of inflammatory cells at the electrode–tissue interface and the disorganized surrounding tissue architecture. Metals such as platinum, gold, and silicon are much stiffer and mechanically dissimilar to soft biological tissues; therefore, they often provoke a foreign body reaction characterized by chronic inflammation, fibrotic capsule formation, and loss of interfacial function [14,40]. In contrast, the PPy/HA electrodes demonstrated a moderate degree of inflammatory cell infiltration. Remarkably, the PPy/cHA electrodes showed less inflammatory cell infiltration and better preserved tissue architecture at the implantation site than the other tested electrodes. Masson's trichrome staining revealed that the gold and PPy/HA electrodes induced thick collagenous scar tissue, whereas the PPy/cHA implants demonstrated substantially less collagen deposition. Quantitative measurement of the fibrotic layer thickness revealed that the gold group induced the thickest fibrotic capsule (39 ± 6 μm), followed by PPy/HA (38 ± 4 μm). In contrast, PPy/cHA induced a significantly thinner capsule (23 ± 1 μm) than gold and PPy/HA, indicating a reduced foreign body response due to the stable crosslinked HA surface (Fig. 5b). These results suggest that surface HA not only reduces initial protein and cell adhesion but also modulates the host response in vivo. Consistent with our findings, HA-modified surfaces have been shown to mitigate fibrosis [41,42]. Altogether, our PPy/cHA electrodes, displaying their stable resistance to enzymatic HA degradation, demonstrated reduced fibrotic thickness and improved tissue integration, making them suitable for long-term implantation.

Fig. 5.

Histological analysis of tissue responses and fibrotic tissue formation at 3 weeks after subcutaneous implantation. (a) Representative H&E and Masson's trichrome staining images of tissues surrounding the implants. (b) Fibrotic capsule thickness based on Masson's trichrome-stained images. Scale bar = 100 μm. An asterisk (∗) denotes a statistically significant difference (p < 0.05).

3.5. In vivo signal recording of implanted electrodes

The reduced scar thickness around the PPy/cHA electrodes suggests better preservation of the electrical interface and reduced signal attenuation over time. Hence, the functional performance was assessed by monitoring the ECG signals in vivo (Fig. S8). Gold, PPy/HA, and PPy/cHA electrodes were subcutaneously implanted in the back region of the mice, and ECG signal recordings were obtained over a 3-week period (Fig. 6a). The gold and PPy/HA electrodes exhibited recognizable ECG peaks. However, their baseline noise progressively increased over time. In contrast, the PPy/cHA electrodes consistently exhibited distinct ECG signals with minimal changes in baseline noise throughout the 3-week period. A quantitative analysis of the SNR confirmed the superior long-term performance of the PPy/cHA electrodes (Fig. 6b). After implantation, the PPy/cHA (40 ± 4) electrodes showed significantly higher SNR values than gold (31 ± 3) and PPy/HA (34 ± 3) electrodes, likely due to the integrity of the crosslinked HA layer. The PPy/HA electrode showed a slightly higher signal sensitivity than the gold electrode; however, the difference was not statistically significant. Interestingly, at week 1, the SNR values increased for all the implanted electrodes, which may be associated with the initial wetting of the electrode surface and the resulting enhancement in electrical sensitivity. However, after week 1, the SNR values of the gold and PPy/HA electrodes gradually decreased, indicating a progressive loss of recording sensitivity. In contrast, the PPy/cHA electrodes maintained significantly higher SNR values, enabling stable and high-fidelity signal acquisition throughout the implantation period. For example, at week 3, the SNR values of gold, PPy/HA, and PPy/cHA electrodes were 31 ± 6, 31 ± 3, and 45 ± 7, respectively, which corresponded to 85.2, 89.7, and 99.9% of their initial values at week 1. The impairment of the SNR in the gold and PPy/HA groups is likely attributable to progressive fibrotic encapsulation (Fig. 5), which increases the interfacial impedance and attenuates the signal quality. On the other hand, the superior performance of the PPy/cHA electrodes appears to be attributed to the long-lasting non-biofouling, hydrophilic, and enzymatically stable surfaces of crosslinked HA, which minimize fibrotic layer formation and maintain low-impedance signal interfaces. Overall, the ability of the PPy/cHA electrodes to preserve high signal fidelity is beneficial for use in implantable electrophysiological devices, particularly in long-term cardiac, neural, or muscular recording applications. Although the present study demonstrated the stable in vivo performance and reduced fibrotic encapsulation of PPy/cHA electrodes over a 3-week period, extended in vivo evaluations spanning beyond 3 weeks (e.g., 6–12 weeks) are necessary to fully validate their long-term stability and sustained anti-fibrotic efficacy for chronic implantation. Future studies with long-term animal models will be necessary to comprehensively assess the structural and functional longevity of the PPy/cHA.

Fig. 6.

In vivo ECG signal monitoring using subcutaneously implanted gold, PPy/HA, and PPy/cHA electrodes for 21 days. (a) Representative ECG signals recorded from gold, PPy/HA, and PPy/cHA electrodes on days 1, 7, 14, and 21. (b) Signal-to-noise ratio (SNR) calculated from ECG signals. An asterisk (∗) and a hashtag (#) indicate statistically significant differences compared with the gold and PPy/HA electrodes, respectively (p < 0.05). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

4. Conclusion

In this study, we developed and evaluated BDDE-crosslinked HA-doped PPy (PPy/cHA) electrodes as stable, nonbiofouling, and implantable bioelectrodes for long-term in vivo signal monitoring. Although PPy/HA electrodes exhibit excellent hydrophilicity and antibiofouling properties, their functional stability is significantly compromised by the enzymatic degradation of HA after implantation. To address this problem, we introduced a mild post-polymerization crosslinking strategy using BDDE, which covalently stabilized the HA chains on PPy. Importantly, the PPy/cHA electrodes retained their surface morphology, surface carboxyl groups, hydrophilicity, and low impedance, even after HAase treatment. PPy/cHA effectively resisted enzymatic hydrolysis and preserved non-cell-adhesive properties, leading to significantly reduced fibroblast adhesion in vitro. In vivo subcutaneous implantation in mice demonstrated reduced fibrotic encapsulation around the PPy/cHA electrodes, with significantly thinner collagenous capsules than in gold and PPy/HA controls. Furthermore, the PPy/cHA electrodes exhibited superior electrical performance over a three-week implantation period, maintaining a higher SNR in ECG signal recordings. These findings suggest that the enzymatic stability and nonfouling characteristics of crosslinked HA surfaces are critical for minimizing the interfacial impedance and preserving long-term bioelectronic functionality. Overall, our study successfully demonstrates that PPy/cHA is a promising platform for next-generation implantable bioelectrodes that require long-term stability, biocompatibility, and signal fidelity for a wide range of biomedical applications, including neural interfaces, biosensors, and cardiac monitoring devices.

CRediT authorship contribution statement

Jongdarm Yi: Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – review & editing. Gaeun Kim: Formal analysis, Investigation, Methodology, Visualization. Sanghun Lee: Formal analysis, Methodology. Jae Young Lee: Conceptualization, Funding acquisition, Project administration, Supervision, Writing – original draft, 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.

Appendix A. Supplementary data

The following is/are the supplementary data to this article:

Data availability

Data will be made available on request.