Surface modification of neural electrodes with pyrrole-hyaluronic acid conjugate to attenuate reactive astrogliosis in vivo

Abstract

Surface of neural probes were electrochemically modified with a non-cell adhesive and biocompatible conjugate, pyrrole-hyaluronic acid (PyHA), to reduce reactive astrogliosis. Poly(PyHA)-modified wire electrodes were implanted into rat motor cortices for three weeks and were found to markedly reduce the expression of glial fibrillary acidic protein compared to uncoated electrodes.

Neural prosthetic probes have been developed to electrically stimulate and/or record neural activity in the brain¹. For patients with central nervous system (CNS) disorders, electrical stimulation via implanted neural probes has been reported to help restore normal neural functions^{2, 3}. For example, some symptoms of Parkinson’s disease are alleviated by deep brain electrical stimulation in patients⁴. In addition, neural prosthetic probes allow the recording of neuronal firing potentials and their patterns adjacent to the electrodes^{5, 6}. However, implanted electrodes often lose their electrical connectivity within weeks to months and become unable to electrically communicate with neuronal cells due to high impedance between the implanted electrode and neural tissues¹. The impaired electrode performance results from a thick glial scar tissue that serves as both a physical and electrical barrier between the brain tissue and the electrodes⁷. Scar tissue formation around the injury/implant is known as brain tissue response to foreign electrode materials, in which reactive astrogliosis plays critical roles⁸. Neural electrode implantation causes astrocytes to become activated and secrete cytokines and extracellular matrix molecules, of which responses are considered to isolate the foreign materials and thus to protect native brain tissue by forming to glial scar tissue^7–9. Therefore, it is essential to reduce the reactive astrogliosis to increase the connectivity and longevity of implanted electrodes.

Various factors of a neural probe influence the inflammatory astrogliosis, including probe geometry¹⁰, micro-motion between a probe and surrounding tissue^{11, 12}, and surface properties of implanted probes^{13, 14}. Among them, coating the electrode surfaces with biocompatible or bio-inert polymers can be an effective strategy to shield the foreign electrode materials and reduce astrocyte activation. Several attempts have been made to coat neural probes with poly(2-hydroxyethyl methacrylate) (pHEMA), alginate, polyethylene glycol, and silicone polymers^15–19. However, these coatings resulted in substantial increase in electrical impedance after the surface modification and/or did not reduce the glial response.

In this study, we developed a method of biomimetic surface modification of neural electrodes using HA as a component of a conductive coating. HA is a polyanionic polysaccharide that naturally exists in the extracellular matrix of the brain, and is non-immunogenic, biocompatible, and generally non cell-adhesive to a variety of different cell types^{20, 21}. Furthermore, high molecular weight HA has anti-inflammatory activity in CNS tissues, especially to the astroglial cells^{21, 22}. Thus, HA was selected as the backbone for electrically conductive graft copolymers used to coat neural electrodes in this study. We previously reported the synthesis of such graft copolymers, pyrrole-hyaluronic acid conjugates (PyHA), and their electro-polymerization on the surface of electrically conducting materials (e.g., indium tin oxide, platinum, polypyrrole)²³. Importantly, this electrode surface modification with poly(PyHA) could prevent the adhesion of fibroblasts and astrocytes in vitro without impairment of electrical properties of the electrodes. Accordingly, we investigate the potential of such coatings to reduce glial cell attachment, activation, and scarring in vivo in this report (Figure 1).

Fig. 1.

Schematic illustration of the poly(PyHA)-coated neural electrode for attenuation of reactive astrogliosis. (a) Electrochemical coating of electrode surfaces using PyHA conjugate. (b) Biomimetic modification of a neural electrode surface with poly(PyHA) to reduce astrocyte activation by masking the foreign material.

Two types of neural probes (i.e., silicon microelectrodes, iridium microwire electrodes) were tested for modification and brain implantation. The microwire electrode consists of a conductive portion only on its surface, whereas the silicon microelectrode consists of conductive and non-conductive portions. Both the iridium microwire probes and silicon microelectrode probes were electrochemically coated with poly(PyHA). The electrode surface turned hydrophilic after the poly(PyHA) coating process because HA is highly negatively charged and PPy is positively charged. This electrochemical polymerization parameter resulted in a 30 nm thick layer poly(PyHA)²³. As shown in Figure 2a and 2b, positive staining of the poly(PyHA)-coated electrodes with HA binding protein confirmed the successful surface immobilization of HA, and modified iridium microwires fluoresced brightly, as did the conductive microelectrodes (circular dots) on the unstained insulating silicon nitride pad.

Fig. 2.

Characterization of poly(PyHA)-coated neural probes. Fluorescence images of (a) an iridium microwire electrode and (b) the silicon microelectrode probe after electrochemical modification with poly(PyHA). The probes were stained with biotinylated HABP, followed by incubation with streptavidin-PE. For the silicon microelectrode probe, a fluorescence intensity profile was plotted (bottom) from the line of the top image. Impedance spectra of neural probes before and after the poly(PyHA) coating: (c) the microwire and (d) the silicon microelectrode (1250 μm²).

In addition, electrochemical impedance spectra (EIS) of the modified electrodes were obtained and compared with those of unmodified controls to study the electrical performance of the neural probes. Figure 2c and 2d indicate that our modified neural electrodes exhibit similar impedance spectra to those of unmodified electrodes in a range of 1 Hz - 100 kHz. As results, the poly(PyHA) coating could introduce biocompatible HA moieties onto conducting surfaces while maintaining the original electrical properties of the electrodes.

We examined the effects of poly(PyHA) coatings on the surfaces of the electrodes in an in vivo study. Two groups of animals were implanted with either poly(PyHA)-coated probes (n=5) or unmodified probe controls (n=5). The probes were implanted into the motor cortices of rats and allowed to remain there for three weeks. Astrocytes are known to become activated a few days after implantation, proliferate near the implanted site, and begin to form organized scar tissue around the implanted probes since three weeks²⁴. Thus, we selected a three week time point to evaluate brain tissue responses. Staining of the retrieved modified wires revealed the presence of poly(PyHA) on the surface, whereas no distinct fluorescence was detected from the retrieved unmodified wire electrodes (controls) (Supplementary Information Fig S1). The results suggest the stability of poly(PyHA) layers on the probes during surgical probe insertion and within the brain tissue for at least three weeks. The poly(PyHA) coating appears to be stable against hyaluronidase enzyme in the brain tissue during the experimental period (3 weeks). Previous studies indicated that the poly(PyHA) coatings could maintain their ability to prevent cell adhesion at a concentration of hyaluronidase up to 5 U/mL²³, which is significantly higher than the hyaluronidase activity in brain tissue (approximately 1 mU/mL)²⁵.

To assess reactive astrogliosis in the brain tissue, we examined the glia fibrillar acidic protein (GFAP) expression level as an activated astrocyte marker^{26, 27}. Figure 3a illustrates representative immunofluorescence images of brain tissue slices stained for GFAP from animals that received modified and unmodified probes. Tissues implanted with the poly(PyHA)-coated microwires displayed approximately 60% lower fluorescence intensity compared to those implanted with un-coated controls probes (p=0.098), indicating substantial reduction in GFAP expression with the poly(PyHA) modification. However, implantation of the silicon microelectrode probes, whether modified with poly(PyHA) or left unmodified into the brain, did not show significant differences in GFAP-positive fluorescence intensity (p=0.831).

Fig. 3.

Astrogliosis in brain tissues around the neural probes implanted after three weeks. (a) Representative immunofluorescence images of brain sections obtained from different animals, which were implanted with the microwire probes and the silicon microelectrode probes. Brain slices were stained for GFAP, which is an astrocyte marker. (b) Analysis of GFAP expression from the brains implanted with probes. The fluorescence images were processed to binary images with the same threshold value. Area positively covered by GFAP was measured and normalized to the control (areas from unmodified probes). From each group (poly(PyHA)-coated or uncoated), averages and SEM were calculated and reported.

Effects of the poly(PyHA) coating on reactive astrogliosis was different depending on the neural electrode type. This different observation may be explained by several reasons. First, for brain samples implanted with silicon microelectrode probes, it was difficult to locate the precise position of the poly(PyHA)-coated areas on the probes in brain tissue sections because there were abundant insulating surfaces not coated with poly(PyHA). Second, the presence of non-conducting portions (and therefore non-poly(PyHA)-coated areas) might be playing a greater role in inducing reactive astrogliosis than conductive electrode (poly(PyHA)-coated) sites of the silicon microelectrode probes. The silicon microelectrode probes had only small electrodes covered with poly(PyHA) with the larger insulating surface unmodified. Regarding the possible roles of non-conducting components of probes interacting with brain tissue, to the best of our knowledge, there are no other previous studies examining this interaction specifically. Our results suggest that the coating/modification of the silicon microelectrode probes is not sufficient to reduce astrogliosis in vivo and/or the unmodified insulating surface plays major roles in activating astrocytes. Hence, future work should include i) a more precise immunohistological analysis of neuronal and glial cell interactions with electrode coatings and insulating pads in vivo, ii) functional studies of the poly(PyHA)-coated neural probes with respect to electrical connectivity after long-term implants, and iii) systematic studies on biomimetic coatings of probes, for instance, selective coating versus entire coating of neural probes.

Conclusions

In this study, we electrochemically coated electrode surfaces with a non-cell adhesive and biocompatible poly(PyHA) to improve brain tissue compatibility. The effectiveness of the poly(PyHA) coating to minimize astrogliosis was tested in vivo with two different neural electrodes for at least three weeks. The modified electrodes presented HA moieties selectively on electrode surfaces and exhibited intact electrical impedances after the coating. Histological responses performed three weeks after implantation into rat cortices revealed attenuated GFAP expression from the poly(PyHA) coated microwires compared to unmodified microwires. However, no significant differences were found between the poly(PyHA)-coated silicon microelectrode probes and uncoated control microelectrode probes, necessitating further studies on functional electrical sensitivity and the roles of non-conductive pad components in tissue interactions. This novel technique for surface modification of metallic and non-metallic conducting substances can also be extended for use in other applications such as stents and biosensors.