Repairing peripheral nerve damages caused by nerve trauma, compression or tumor is still a challenge. For severe nerve defects, appropriate devices for bridging and guiding axons are necessary. Neural autograft segments can be used for such purposes but limited availability of donor nerves and function loss of donor sites have hampered the use of autografts. Tissue engineering nerve guides are a promising alternative. A variety of techniques can produce 3D scaffolds with various architectures and properties but electrospinning and electrospun fibrous scaffolds have attracted great attention owing to many advantages. In nerve tissue repair, nanofibrous scaffolds can provide favourable conditions for Schwann cell migration and for guiding neurite outgrowth. Delivering locally growth factors such as nerve growth factor (NGF) and glial cell line-derived growth factor (GDNF) can promote a range of neural responses and affect cell fate. In this investigation, for dual delivery and sustained release of growth factors, bicomponent nanofibrous scaffolds containing NGF and GDNF were fabricated via dual-source dual-power electrospinning and then evaluated. NGF and GDNF were encapsulated in PDLLA fibers and PLGA fibers, respectively, through emulsion electrospinning. The mass ratio of NGF/PDLLA fibers to GDNF/PLGA fibers in bicomponent scaffolds could be varied to modulate growth factor release. Through optimization, uniform fibers were made and both types of fibers were evenly distributed in bicomponent scaffolds. Using various techniques, the structure and properties of bicomponent scaffolds were studied. Core-shell structures were formed for both types of fibers, with the growth factor-containing water phase being the core. All fibrous scaffolds exhibited good wettability. In vitro degradation of scaffolds and in vitro release of both growth factors were investigated for up to 42 days. The degradation rate of NGF/PDLLA fibers was much lower than that of GDNF/PLGA fibers due to slower degradation of PDLLA. The degradation rate of bicomponent scaffolds increased with an increasing amount of GDNF/PLGA fibers. Scaffolds with fibers of different diameters exhibited different degradation rates. NGF and GDNF were both successfully incorporated in bicomponent scaffolds with relatively high encapsulation efficiency. Sustained release of both growth factors was observed. The release of GDNF from mono- or bicomponent scaffolds was much faster than that of NGF. The cumulative release amount of NGF or GDNF increased with an increasing amount of NGF/PDLLA fibers or GDNF/PLGA fibers in bicomponent scaffolds, and increased with a decrease in fiber diameter. The cellular response, including cytocompatibility, cell migration and neurite outgrowth, to the scaffolds was also assessed.

We report the development of a photo-reversible neural cell culture substrate. We demonstrate the capacity to modify the adhesivity of the substrate using light, altering its capacity to support cell growth. Polyelectrolyte multilayers (PEMs) were used to produce tunable substrates of different thickness and matrix stiffness, which have different intrinsic capacities to support cell adhesion and survival. Surfaces were top-coated with a poly(acrylic acid)-poly(allylamine hydrochloride) polyelectrolyte bilayer functionalized with a small fraction (<1%) of an azobenzene based photo-switchable sidegroup, which included the cell adhesive three amino acid peptide RGD. Irradiation with light induced geometric switching of the azo bond, resulting in changes to RGD exposure and consequently to cell adhesion and survival, which was investigated on a variety of surfaces of different thickness and stiffness. Substrate stiffness, as modified by the thickness, had a significant influence on the adhesion of NIH 3T3 cells, consistent with previous studies. However, by disrupting the isomerization state of the azobenzene-linked RGD and exposing it to the surface, cell adhesion and survival could be enhanced up to 40% when the positioning of the RGD peptide was manipulated on the softest substrates. These findings identify permissive, yet less-than-optimal, cell culture substrate conditions that can be substantially enhanced using non-invasive modification of the substrate triggered by light. Indeed, where cell adhesion was tuned to be suboptimal in baseline conditions, the light-induced triggers displayed the most enhanced effect, and identification of this ‘Goldilocks zone&’ was key to enabling light triggering.

Electrically conducting polymers (ECPs) are promising biomaterials because they satisfy electrical, biochemical and mechanical requirements for cell and tissue engineering. ECPs were frequently used in the bioelectronic devices to electrically couple with cells in vitro and in vivo. The previous researches well demonstrated long term stability of ECPs to electrically stimulate the cell and record cell signal for tissue regeneration and neural interface devices. Unfortunately, as ECPs are not biodegradable, the ECP-based implants have to be removed after used, or replaced when their function lost. The implant removal would definitely damage the cells and tissues surrounding implants due to their strong interaction. To solve this issue, we developed a dynamic conducting polymer material based on 3,4-ethylenedioxythiophene bearing electrochemically switchable hydroquinone groups (PEDOT-HQ). Hydroquinone could be electrochemically oxidized to form benzoquinone, and further react with oxyamine-terminated biomolecules to switch on the biofunction of the polymers. On the other hand, the biofunction conjugated benzoquinone could be electrochemically reduced to switch off the biofunction. Firstly, we investigated the electrochemical and optical properties of PEDOT-HQ films and demonstrated stability of electrochemical switching hydroquinone groups in aqueous solution. Moreover, we further demonstrated its spatially selective immobilization and release of fluorescent dyes on patterned PEDOT-HQ microelectrode arrays. Finally, we demonstrated its superior performance in electrochemically stimulating cells and releasing cells after stimulation. The cell viability of the detached cells was as high as 94 % and indicating very small damage to cells. In summary, we successfully developed a new dynamic biointerface based on conducting polymer and this platform has great potential for tissue engineering.

Cortical microelectrodes can provide intimate contacts with neural cells and show promise as electrical interfaces to the brain, which permit the treatment of a range of neurological deficits. The functionality of current electrodes, however, decreases over time, which is to a significant extent caused by neuron degeneration and foreign body encapsulation. Many factors contribute to neuroinflammation following device implantation, including the mechanical-mismatch between the implant and the brain tissue, and oxidative stresses that result from acute inflammation. It is thought that each of these pathways plays a unique role in different phases after implantation. In separate studies the tissue response to mechanically-adaptive materials, designed to minimize mechanical-mismatch between the brain and implant, and drug-releasing implants, designed to exert anti-inflammatory activity on macrophages, have been studied. With the objective to minimize acute and chronic responses, mechanically-adaptive materials were now modified to also release antioxidant supplements. Thus, a new series of biologically-inspired, physiologically-responsive, mechanically-adaptive nanocomposites based on poly(vinyl alcohol) (PVOH) and cellulose nanocrystals (CNCs) was studied, which are also capable of releasing curcumin or resveratrol; these anti-oxidants are known to reduce oxidative stress and promote blood-brain barrier (BBB) stability. In the dry state, the PVOH nanocomposites exhibit a high stiffness to facilitate implant insertion, but they soften considerably (modulus reduction up to 1000-fold) upon exposure to physiological conditions. It was shown that variation of the CNC type, content, and the processing conditions allows one to tailor the mechanical properties of these materials over a broad range. The ability of these materials to release the antioxidant drugs was studied in vitro, using artificial cerebrospinal fluid (ACSF) at body temperature. The most prominent result of an in vivo study was that the antioxidant drug-releasing materials offered significant neuro- and BBB protection two weeks post-implantation. These results support the hypothesis that reduced oxidative stress helps maintain a more stable BBB and provides localized neuroprotection at the electrode-brain interface at acute time-points.

Long-term, multisite recording and stimulation of neurons by microelectrode arrays to restore and enhance impaired sensory, motor and cognitive functions has been the subject of intense studies for decades. Similar micro- and nano-technologies are also used to analyze normal and pathological neural circuits for basic brain research. Nonetheless, to date these efforts have failed to provide satisfactory concepts or technologies that can lead to the interfacing of individual neurons with electronic devices for long-term recording of the entire neuronal repertoire of synaptic- and action-potentials.
In this presentation I describe how we have begun to better integrate manmade electronic devices with individual excitable cells. Our approach involves the use of energy resources of the living cell and its structural and functional plasticity to actively identify and engulf microelectrodes. Specifically, we "fool" neurons into “believing” that the electrodes are “brother cells” rather than foreign elements. Thus, rather than constructing classical flat substrate-integrated sensing pads, we fabricate protruding microelectrodes that mimic the geometry and dimensions of dendritic spines (gold mushroom-shaped microelectrodes-gMµE). The gMµE are then functionalized by a multiple RGD repeat peptide. The localized presentation of the RGD-peptide by the gMµE “lures” the cell into using its internal energy resources to actively engulf the electrodes by reconstructing its cytoskeleton. The gMµE engulfment nevertheless maintains the cell&’s plasma membrane intact. The tight physical contact formed between the neuron and the engulfed gMµEs increases what is known as seal resistance. In addition, the electrical conductance of the membrane patch facing the electrode increases, probably by recruitment of voltage independent ion channels in response to the gMµE curvature and the RGD-peptide. The newly created cell/gMµE configuration provides unprecedented multisite, long-term, noninvasive recording and stimulation of action potentials and subthreshold synaptic potentials from individual neurons with qualities equivalent to intracellular microelectrodes that operate by breaking the neuron plasma membrane. The prospects of using this approach for cell-biological research and clinical BMI applications will be discussed.
Supported by: EU FP7 MERIDIAN Grant No. 280778. EU FP7 Marie Curie ITG, Grant No. 264872. EU FP7 Brainleap Grant No.306502 and the Charles E. Smith and Prof. Elkes Laboratory for Collaborative Research in Psychobiology

High-count microelectrode arrays have shown a tremendous potential, particularly in the basic neuroscience research, neuropathology research and for clinical applications. In the past two decades, lot of research work has been focused on the characterization of a good bondable metallization stack and understanding of right electrode materials for providing lower impedances during recording/stimulation of neurons. The current versions of Utah Electrode Array (UEA) make use of Ti/Pt/TiW/Pt (100/200/480/520nm) as metallization stack for wire bonding and Ti/IrO2 (100/650nm) stack as electrode-tissue materials. During the in-vitro and in-vivo studies of UEAs, these stacks have shown bad metal adhesion to the silicon substrate, poor ohmic contacts and non-reliable impedances. In this study, we present a new bondable metallization stack (Pt/Ir/Pt) and a new electrode-tissue stack (Pt/Ir/IrOx) with optimized annealing conditions that provides reliable ohmic contacts, good metal adhesion and very low and reliable impedances for recording and stimulation. The UEAs that were used in this study consisted of high electrode-density (25-electrodes/mm2), graded electrode lengths (200-1000µm) on a smaller footprint (2mm2 for a 10×10 array). The substrate material was boron doped p-type <100> silicon work piece (2mm thick) with 0.005Omega;cm resistivity. UEAs were fabricated using manufacturing techniques that were previously described in the literature. Each pad size for wire bonding was 150×150µm. The bondable metallization stack consisted of sputter deposited Pt/Ir/Pt thin films in thicknesses of 200/400/500nm, respectively. The first Pt layer was used as seed layer that formed a silicon-platinum silicide when the stack was annealed at 375C for 45 minutes, in an ambient of Ar/H2 at a later process step. Ir was used as a diffusion barrier while Pt on the top was used for bonding purposes. For the electrode-tissue interface, Pt/Ir/IrO2 with thicknesses of 200/200/500 nm was sputter deposited on the electrode tips of the UEA. To enhance the metal-substrate adhesion, the UEAs were annealed again at 475C for 30 minutes in an ambient of O2 environment. Two-point I-V measurements determined that the backside metallization formed a Schottky contact as deposited, which become Ohmic when the stack was annealed at 375C for 45 mins. Tape tests and SEM microscopy analysis was done to evaluate the adhesion strength of the bondable metallization stack and the electrode-tissue stack. Three 10×5 UEA array was wire-bonded to a TDT connector. In-vitro soak tests were done in saline solution for 5 days. The mean impedance for 48 channels was 40.9Kohms with a median impedance of 17.2Kohms. Initial in-vivo experiments were done through acute implants of three wired UEAs into the sciatic nerve of three rats. In each experiment, there were > 25 electrodes out of the 48 wired electrodes that were able to record either single unit or multiunit waveforms.

Multi Electrode Arrays (MEAs) as electronic devices coupling to biological systems has emerged as a promising technology for future biosensors. It has been shown [1] previously, that electrogenic cells such as cardiac cells or neurons engulf 3 dimensional (3D) micro and nano structures similar to the uptake of a solid particle by cell via phagocytosis/endocytosis. We characterise the interface between the device and cell and try to gain mechanistic insights to how the shape/dimensions of the structures, cell membrane properties, and relative positioning between the cells to the structures affect the coupling between the two.
Modification of planar electrodes by 3D structures can result in improved improves the recording or stimulation of electrically active cells [2]. We present novel fabrication techniques [3] of cylindrical and mushroom shaped 3D microstructures. Next, we investigate the deformation profile of the cell membrane engulfing them when interacting with 3D structures using a focused ion beam for transversally sectioning the cell and the micropillar. Membrane deformation is characterised for 2 dimensionless geometric parameters - (a) Aspect ratio between the height and the diameter of the stalk (ARcylinder) (b) Aspect ratio between the major and minor axes of the mushroom cap (ARcap). We find that the cell membrane attains different curvatures depending upon the pillar geometry. In the case of a cylindrical pillar, the membrane detaches above the pillar while the cell engulfs the ‘mushroom&’ shaped pillar almost completely.
We also find that the relative position of the micropillar to the cell plays a significant role in the deformation profile. Hence, from an application point of view of such 3D electrodes for sensing and stimulation, we characterise the deformation profile (via. membrane curvature and the wrapped fraction). In addition to the above mentioned dimensionless parameters (ARcylinder & ARcap), we take into account the relative position between the micropillar and the cell by considering a fit of the HL-1 cell periphery by an ellipsoid. We find that in the center of the cell the cytoskeleton in particular with the actin filament network are present plays a dominant role in determining the deformation profile, due to the positive curvature values. At the edge of the cell, the membrane curvature is less affected by the contribution of the cortex and so has a linear deformation with zero curvature. At the very edge of the cell we find almost membrane without cortex like behavior with negative curvature and here the membrane deformation parameters shall be compared with theoretical models [4].
[1] M.E. Spira , A. Hai, Nature Nanotech.8, 83-94, (2013).
[2] A. Hai, J. Shappir, M.E. Spira, Nature Methods 7, 200 - 202, (2010).
[3] G. Panaitov, S. Thiery, B. Hoffmann, A. Offenhäusser, Microel. Eng. 88, 8, 1840-1844,(2011).
[4] S. Dasgupta, T. Auth, G. Gompper, Soft Matter, 9, 5473-5482, (2013).

We have recently demonstrated a carbon-based array of high aspect-ratio microelectrodes on a flexible substrate for applications in neural sensing and simulations. The microelectrodes are made of pyrlozyed carbon derived from polymer pre-cursor using C-MEMS (Carbon MEMS) process [1-2] and are capable of interrogating a wider area. The ensuing 'microelectrode array fabric' could be used for both implantable as well as wearable applications. The other key potential application includes a microelectrode system as part of tactile sensor in a prosthetic socket. The 3-dimensional aspect of the electrodes offer a significantly higher area of interaction with nerve cells as compared to traditional 2D metal electrodes resulting in higher signal-to-noise ratio. Our electrical characterizations have shown that the polymer-derived graphitic electrode arrays have better electrical signal/noise ratio than traditional 2-dimensional thin-film electrodes.
On the other hand, due to their wide electrochemical window in ionic solution as well as reduced tendency to bio-foul, polymer-derived graphitic electrodes have recently caught the attention of researchers for in vivo electrochemical detection of biological species [3]. In particular, microfabricated microelectrode arrays could offer a fast and reliable simultaneous, decoupled detection of neurotransmitters like dopamine and serotonin. In this research, we report on investigation of our high aspect-ratio microelectrodes on a flexible substrate for applications in in-vivo electrochemical sensing of neurotransmitters sensing. The multi-site microelectrode array will have separate electrodes for sensing electrical and electrochemical signals - but all integrated on a single chip. This simultaneous sensing of ECoG electrical signals as well as electrochemical detection of neurotransmitters like dopamine and serotonin could find powerful use in DBS (deep brain stimulation) where the effect of application of electrical input (voltage) on the response of neurons could be monitored real-time. The array nature of the electrodes will also provide a key advantage of simultaneous and de-coupled detection of several neurotransmitters.

References
1. Wang, C., Taherabadi, L., Jia, G., Kassegne, S., Zoval, J., and Madou, M., “Carbon-MEMS Architecture for 3D Microbatteries”, Proceedings of SPIE Photonics Europe, France, April 2004.
2. Wang, C., Madou, M., “From MEMS to NEMS with carbon”, Short communication, Biosensors and Bioelectronics 20 (2005) 2181-2187.
3. Zachek et al. “Simultaneous Decoupled Detection of Dopamine and Oxygen using Pyloyzed Carbon Microarrays and FSCV”, Anal. Chem, 2009.

Increased life span due to improved life-style and advances in medicine have brought the spectrum of neuro-degenerative diseases to the attention of scientific community and general public alike. Generation of neuronal circuits “in vitro” represents a fundamental step in advancing our understanding in number of areas, including: disease modelling, drug testing and, the study of brain activity.
Diamond as a substrate, presents outstanding properties such as: extreme mechanical hardness and resistance to chemical corrosion. Furthermore, diamond is biologically inert and it has been shown not to induce an immune response when implanted. Diamond crystalline films can be produced relatively inexpensively by the method of chemical vapour deposition. Diamond films are highly electrically insulating, but can be made semiconducting by doping, making them ideal for the study of electrically active cells like neurons.
Embryonic Stem (ES) cells are pluripotent cells which present great promise for the fields of regenerative medicine and drug discovery. Recently, robust protocols have been made available for the differentiation of specific human neuronal cell types from ES cells, including cortical neurons. Furthermore, it is now possible to induce formation of pluripotent cells from mature cell types of adult individuals (Induced Pluripotent StemCells - IPS). IPS are considered key to understanding the mechanism of diseases, as they can be generated from patients with a known mutation, the effect of which can be studied in vitro.
Here we show that human and rodent cortical neurons differentiated from ES and IPS cells can be cultivated on diamond films. We present a method of patterning a neuronal network and propose that this method can be used for the study of neuronal circuit activity.

Micro-fabricated neural electrode arrays, placed in the nervous system to directly interface with neurons, have tremendous clinical and research significance. Intra-cortical arrays experience chronic recording or stimulation failure due to the persistent inflammatory tissue responses characterized by BBB leakage, neuronal loss and glial scarring around the implant. The tissue responses are caused by several factors, and one of which is the mechanical mismatch between the stiff device and soft brain tissue. Current arrays are mostly made of materials that are 3-6 orders of magnitude stiffer than the brain tissue. To overcome this issue, an elastomeric electrically conductive polymer blend is synthesized that has the mechanical modules similar to that of brain tissue. This conducting elastomer can be fabricated into soft (young&’s modulus of 130 kPa) and stretchable wires (elongation at break of 216%), which is then insulated except for the tip. In vitro culture assays showed that soft wires made of the new materials recruited and activated less microglia in culture than the stiff microwires, typically used to assemble chronic neural arrays. Furthermore, surface immobilization with biomolecules was done to the soft wires to control their interactions with neurons and glia. In vivo experiment has shown that the soft microwires can be successfully implanted into the rat visual cortex and record neural signals. One challenge with implanting any soft intra-cortical devices is how to insert them in a minimally invasive manner. We examined the use of carboxymethylcellulose (CMC) as a bio-dissolvable delivery vehicle that can provide the stiffness to facilitate the penetration. With a micro-molding process, the CMC can be made into needles of various size and geometry and encapsulate the electrode wires. Implantation and histology studies have found that the dissolution of pure CMC is considerabley slower in the brain than in vitro. Early time points show the CMC shuttle expanded after insertion as it absorbed moisture from the brain and slowly dissolved. However, at later time points neuronal cell bodies re-populated regions within the original probe tract and no activated microglia was found at 12 weeks. Composite needles made of CMC and small molecule fillers showed much quicker dissolution and tissue recovery. These new materials and designs show great potential for the development of next generation flexible neural implant devices that can seamlessly integrate with the host tissue.

Organotypic cultures are able to fill the gap between performing in vivo type of studies in vitro, viz. within the preclinical phase. Currently, however, very limited survival times of only a few days for adult tissue - especially complex neuronal structures - due to substrate issues often severely limit their application. We propose a novel biotechnological concept, which allows for unprecedented long culture times even in absence of biochemical growth factors, as demonstrated for adult retinal and brain tissue [1]. As tissue distortion is commonly preceded by migration of individual cells from the tissue to the substrate, tailoring the substrate properties unfavorable for adhesion or migration of individual cells while ensuring good adhesion of the overall tissue constitutes the key for successful long-term organotypic tissue culture. Employing substrates composed of TiO₂ nanotubes arrays which exhibit easily tunable surface parameters and nanotube topography, we show for the first time that different adult neuronal explants can be successfully cultured organotypically longer than 14 days with no indications of degeneration. Our results show that the interaction of the tissue with the nanotube array is the major determinant for tissue and cell survival. In fact, the ability of the explant to adhere to the individual nanotubes is a key property for long-term culture which is tuned by adequate nanotube diameter, wall thickness and surface roughness. Interestingly, we obtained that these parameters vary for successful long-term culture of adult mammalian retinae and brain slices. While adult mammalian retinae maintain organotypic on smooth and rough nanotube arrays with tube diameters ranging from 30-85 nm, adult mammalian brain slices of neocortex need smooth surfaces with tube diameters of at least 100 nm. Moreover, our new biotechnological approach for long-term culture is employed without the need of perfusion systems for medium supply. Since the nanostructure of our arrays provides an intrinsic super-hydrophilicity of the substrates, a continuous supply of fresh medium from a reservoir below the substrates is automatically maintained by wetting effects without exposing the explants to the bulk liquid. To conclude, we show for the first time that tissue preservation is maintained because the chosen nanotube parameters of the nanostructured substrates suppress single cells migration out of the explant but support entire tissue adhesion and ensures proper medium supply. These findings additionally pave the way for in vitro drug testing, as well as retina and brain tissue regeneration.
[1] Dallacasagrande et al., Adv. Mater. 24, 2399 (2012)

Silicon has been considered a primary substrate for micro-machined intracortical neural implants (INI). The presence of various ions and cell activity within the brain establish what has been considered a harsh, corrosive environment for the implant, and as such the implants material must be able to resist these environments. We are examining if this environment contributes to changes in the surface of the material, which in turn could be one of the reasons for decreased long-term reliability in INI optimal neural recordings, which have plagued these devices for the last 4 decades.
In our previous investigation, we indicated that Si had surface damage after cell culture, but the exact source of this damage was unknown. In this experiment, we attempt to identify the cause of these surface modifications. The surface morphology of (100) Si was examined using samples cultured with H4 neuroglioma cells (ATCC HTB-148) against samples soaked in Dulbecco's modified eagle medium (DMEM), phosphate buffer saline solution (PBS) and artificial cerebrospinal fluid (ACSF). All of the samples were evaluated in the various mediums for 96 hours at 37°C. The samples were extracted and rinsed with flowing deionized water for 10 minutes. We characterized the surface of each set of samples at each stage of the cleaning procedure. Cleaning consisted of three stages: solvent cleaning using acetone and isopropanol (degreasing agents), piranha solution cleaning (3:1, H₂SO₄: H₂O₂) (removal of carbon containing materials), and hydrofluoric acid (1:4, HF: H₂O) (removal of silicon dioxide). Surface characterization consisted of scanning electron microscopy (SEM), atomic force microscopy (AFM), optical profilometry and optical microscopy.
We observed that the samples soaked in PBS did not change significantly, while the samples immersed in DMEM and ACSF show only minor surface alterations. The major finding is that samples cultured with H4 cells exhibited drastic surface morphological changes. The SEM micrographs show the presence of pyramid shaped pits. Further characterization with AFM and optical profilometry confirmed the SEM quantified severe changes in the surface roughness of these samples. At this initial stage of the investigation, we are endeavoring to quantify the cause of these changes to the Si surface, but based on our observations believe that the corrosion could be result of cell chemical products released into the surrounding environment.

We are investigating the design and performance of hybrid interfaces between the residual peripheral nerves of an amputee and a prosthetic limb. We have developed a Regenerative Peripheral Nerve Interface (RPNI) in which a unit of free muscle has been neurotized by a transected peripheral nerve. The RPNI device contains stainless steel electrodes that are electrochemically coated with a biocompatible conjugated polymer such as poly(3,4-ethylenedioxythiophene) (PEDOT) to improve the signal transport between the rigid, inorganic, solid metallic electrode and the soft, wet, ionically-conductive living tissue. We are examining the performance of the RPNI in a rat model using the extensor digitorum longus (EDL) muscle. Our results indicate that the transplanted EDL muscle remains healthy and reinnervated, as evidenced by compound muscle action potentials (CMAP). Histology reveals axonal sprouting, elongation, and synaptogenesis within the RPNI. The PEDOT-coated electrodes had a significant increase in signal amplitude at all time points, and the increase in the recorded signal was seen throughout the course of the study. The RPNI may provide a means for the direct, reliable connection of advanced prosthetic devices with the motor nerves and sensory nerves of patients with limb loss.

We present experimental and theoretical results on biased axonal growth and interconnectivity aimed at elucidating some of the basic rules that neuronal cells use for establishing functional connections. We demonstrate that a nanofilm surface composed of unidirectional angled nanorods can bias axonal growth, and that this bias can be controlled through precise alterations to the nanorod tilt angles and orientations. We perform a systematic investigation of neuronal growth on these tunable surfaces and quantify the effect of biomechanical surface cues on growth cone guidance. These results give new insight into engineering directed axonal growth for neuro-regeneration studies.

After injury to peripheral nerves, axons must navigate the injury and connect to distal targets for functional recovery. Exogenous electrical stimulation (ES) to promote nerve and tissue repair has been used with limited success in animal models and the clinic. It is possible that ES parameters are not optimal or that signal attenuation within the injury site results in a non-reproducible stimulation. The inclusion of an electrically conductive biomaterial may mitigate attenuation in the wound leading to reproducible applications of the stimulus. In this work, conductive single walled carbon nanotubes (SWNT) were utilized to manipulate electrical properties of a hydrogel to investigate the effects of the nanofiller on primary neurite extension. We have shown SWNT hydrogels are supportive of Schwann cells (SC) within this conductive biomaterial (Behan JMBR 2009). DC ES (50 mV/mm, 1 mA, 8 hrs) also supports increased neurite outgrowth (Koppes JNE 2011). In this work, neurite extension from rat dorsal root ganglia (DRG) encapsulated in hydrogels (SWNT or nanofiller-free) with or without ES was characterized to determine 1) if the nanofiller-hydrogels support neurite extension and 2) if ES through the conductive hydrogel enhances neurite outgrowth.
To examine these phenomena, DRG were encased in 3D collagen-MatrigelTM composite hydrogels containing 0, 10, 20, 50, or 100 µg/mL of acid-treated SWNTs, and allowed to incubate 12 hrs in medium with 50 ng/mL nerve growth factor (NGF). Samples were exposed to 50 mV/mm (1 mA, 8 hrs) or 0 mV/mm (control) ES and moved to fresh medium for an added 48 hours of growth. Samples were fixed, stained, imaged, and analyzed with NIH ImageJ to quantify neurite outgrowth and directionality. The elastic moduli and conductivity of the hydrogels were determined via rheology and a 3-point probe test. A Two-tailed ANOVA was conducted in Excel with significance as p<0.05; n=3.
In the absence of ES, hydrogels containing 10-100 µg/mL SWNT support neurite extension with a maximal 3.3x increase at 20 µg/mL compared to SWNT-free controls. ES (50 mV/mm) or 20 µg/mL SWNT alone increased neurite outgrowth 2.9x and 3.3x compared to unstimulated, SWNT-free controls, respectively. The concurrent presentation of SWNT with ES enhanced outgrowth 7.0x compared to unstimulated nanofiller-free controls. The conductivity of SWNT hydrogels was increased 1.7x with no changes to the elastic modulus. Electrically induced changes to the glial cells may contribute to the observed increases in outgrowth (Koppes JNE 2011); but it is unknown what features of the biomaterial support extension. Results described herein indicate that the inclusion of SWNT impacts neurite outgrowth, but which nanofiller features, such as size or dispersion, maximize neurite extension remain unknown. Characterizing neuronal behavior in model systems, such as these, will aid the development of conductive biomaterials and electrical stimulus parameters for nerve repair.

Introduction:
We are developing a regenerative peripheral nerve interface (RPNI) to achieve high fidelity, intuitive prosthetic control. Multiaxial control of individual actuators within advanced neuroprostheses requires implantation of multiple adjacent yet electrically independent RPNIs, necessitating effective biomaterial strategies for signal isolation. This study investigates the acute and long-term efficacy of silicone insulator implantation to reduce extraneous signal in an RPNI.
Methods:
Using a rat hindlimb, acute evaluations (n=8) were performed using non-transferred extensor digitorum longus (EDL) muscle with intact neurovascular supply, whereas chronic RPNI implantations (n=3) were performed by free-muscle transfer of the EDL muscle to the ipsilateral thigh. RPNI neurotization was performed by transecting and implanting the common peroneal nerve into the transferred muscle. A recording electrode with an overlying 7x5x1mm silicone layer (experimental), and an uninsulated recording electrode (control), were affixed to the EDL muscle. The entire construct was then encircled with small intestinal submucosa. The peroneal nerve was directly stimulated to activate the muscle of interest, whereas the tibial nerve was stimulated to emit extraneous signal from the posterior compartment. Electromyography was performed immediately for acute evaluations, and 6 weeks postoperatively for chronic implantations. Data were analyzed using paired samples T-test.
Results:
Acute evaluation: At stimulation threshold, the experimental electrode recorded 47.5% extraneous signal whereas the control electrode recorded 55.9%, revealing an 8.4% improvement in signal isolation by utilizing the silicone insulator (p=0.030). At maximum stimulation, the experimental electrode recorded 53.2% extraneous signal versus the 62.7% recorded from the control electrode, demonstrating an improvement of 9.5% (p=0.001).
Chronic RPNI implantation: Peroneal nerve stimulation showed RPNI performance recorded by the experimental electrode was similar to control for tests of stimulation threshold, latency, and noise as expected, yet marginally reduced for compound muscle action potential (CMAP) amplitude and area. Tibial nerve stimulation revealed isolation of extraneously derived EMG signal was improved in the experimental electrode, reducing approximately 4%-11% extraneous signal compared to control at both minimum and maximum stimulation levels.
Conclusion:
While signal acquisition is marginally impaired by silicone insulation, electrophysiological characteristics follow a pattern of early regeneration similar to uninsulated muscle. Although additional strategies will likely be required to optimize signal separation, utilization of a silicone insulator results in improved signal isolation and is a feasible option for incorporation into an RPNI.

In modern neuroscience, therapeutic regenerative strategies (i.e., brain repair after damage) aim to guide and enhance the intrinsic capacity of the brain to reorganize by promoting plasticity mechanisms in a controlled fashion. Direct and specific interactions between synthetic materials and biological cell membranes may play a central role in this process and nerve tissue engineering has increasingly involved nanotechnology for the development of super-molecular architectures to sustain and promote neural regeneration following injury. The interaction between neurons and nano-structured materials is increasingly attracting interest, because it holds the potential of unexpected openings towards novel concepts for the design of smart devices based on nano(bio)materials properties. Ongoing efforts in this arena require the development of synthetic extracellular scaffolds able to provide unique microenvironments to tissue-specific cell types. We used a multidisciplinary approach to investigate the impact of interfacing synthetic nano-materials (carbon nanotubes) to neuronal networks.

Neural implants are already used for retinal prosthesis applications, to temper the symptoms of Parkinson's disease, or help locked-in patients to communicate, and are expected to play an increasingly important role in disease management in the future. Most of the electrodes currently used for these applications have a smooth surface and elicit a tissue response, which leads to the formation of an isolating layer around the electrodes, composed mostly of glial cells. Recently, it was shown that nanostructured surfaces can lower the tissue response to the implant. In order to understand and manipulate the way nanostructures or nanodevices interact with neural cells, we must study both the biostability and the biocompatibility of these materials.
We report here an in vitro model where retinal cells, which comprise glial cells and neuronal cells, are cultured on different nanowire (NW) surfaces. Using this system, we have analyzed the attachment and survival of the different retinal cell types over time, as well as the degree of neurite extension of retinal cells cultured on gallium phosphide (GaP) NWs of different densities, lengths and diameters, and on (Si) NWs of different topographies and surface chemistries.
On GaP NWs, photoreceptors, ganglion cells and bipolar cells survive on the substrates for at least 18 days in vitro (DIV). Neurons extend numerous long and branched neurites that express the synaptic vesicle marker synaptophysin. We found also a direct correlation between nanowire length and cell attachment as well as neurite elongation and our analysis suggests that neurons sense the vertical dimension of the NW and/or the biomolecules adsorbed on the nanowires. Glial cells survive until at least 18 DIV as well, but are few in number and present a different morphology than that observed on flat control substrates, suggesting that they may be less proliferative or less mature. By choosing the position of the NWs on the substrate, it is possible, in addition, to guide ganglion cell axonal growth and to confine glial cells in specific areas.
When cultured on Si NWs, however, retinal cells do not exhibit neurite outgrowth and glial cells do not show any cytoplasmic extension. Only a few retinal cell markers are expressed or weakly expressed at 3DIV or 18 DIV. This decreased survival may be due to the degradation of the silicon nanowires over time and/or to the presence of chemical residues from the nanowire fabrication.

Organic based transistors for biochemical sensing and cellular applications are becoming an alternative approach to standard technology because of their attractive features (e.g., low-cost, mechanical flexibility, enhanced biocompatibility). A particular configuration of organic FET, namely Organic Charge Modulated FET (OCMFET) is here proposed as a sensor for both electrical and metabolic activity of electroactive cells. Its peculiar structure allows sensing any local charge variation occurring in the sensing area without any chemical modification of the surface, making it a very interesting tool in electrophysiology and biosensing applications.
We will focus on interfacing an OCMFET device specifically tailored for in-vitro electrophysiological applications with living cells cultures and we will propose an innovative approach for the simultaneous detection of both the electrical and the metabolic activity of living cells.
Low operating voltages and high operating frequencies, which are mandatory requirements for the application of the device to electrophysiological measurements, were obtained thanks to the combination of an ultra-thin dielectric and a self-aligned structure. The feasibility of exploiting the interesting properties of the OCMFET for the detection of different bio-related effects (ranging from pH variations related to the cells metabolism, to action potentials and field effect potentials) has been investigated. The capability of the device to correctly transduce both electrical and metabolic signals coming from cells cultured onto the sensing area was demonstrated. The recorded performances are comparable with the ones of standard, reference methods, thus opening up interesting applications in electrophysiology and biosensing.

Real-time stimulation and recording of neural cell bioelectrical activity could provide an unprecedented insight in understanding the functions of the nervous system, and it is crucial for developing advanced in vitro drug screening approaches. Among organic materials, suitable candidates for cell interfacing can be found that combine long-term biocompatibility and mechanical flexibility. Here,we report on transparent organic cell stimulating and sensing transistors (O-CSTs), which provide bidirectional stimulation and recording of primary neurons.We demonstrate that the device enables depolarization and hyperpolarization of the primary neuron membrane potential. The transparency of the device also allows the optical imaging of the modulation of the neuron bioelectrical activity. The maximal amplitude-to-noise ratio of the extracellular recording achieved by the O-CST device exceeds that of a microelectrode array system on the same neuronal preparation by a factor of 16. Our organic cell stimulating and sensing device paves theway to a newgeneration of devices for stimulation, manipulation and recording of cell bioelectrical activity in vitro and in vivo.
Nature Materials doi:10.1038/nmat3630

The ability of chronic microelectrodes to record resolvable neuronal activities in the cerebral cortex is often reduced or completely lost over time. By integrating several commonly-used intracortical devices into a single hybrid array, the performance of the different devices can be analyzed more objectively. We have assembled and implanted a hybrid array and recorded neural activity up to 18 months in cats in-vivo. This report focuses on array assembly, in-vivo neural recording, and immunohistochemistry. A long-term objective of the study is to better understand why microelectrodes implanted chronically in the brain gradually lose the ability to record well-resolved neuronal action potentials.

Understanding and treatment of debilitating neurological conditions, such as Parkinson&’s disease, require the ability to reliably record activity from single neurons. Currently available neural probes used for such recordings, however, suffer from deterioration of the conductive interfaces due to damage to the neural tissue, along with scarring around the implanted probe leading to the decreased quality of the recordings over time, making reliable long term recordings difficult or even impossible.
To overcome these limitations, neural interfaces should be minimally invasive and, ideally, “stealthy” to the tissue surrounding them, which can be achieved by fabrication of very thin and flexible probes. Using a consecutive thermal drawing process we have fabricated flexible, thin polymer-metal composite microelectrode arrays with individual electrode diameters as low as 1 mu;m, and controlled geometry and pitch. Our probes employ several polymers with different chemical resistance, which allows for selective etching in order to change the geometry of the probe or to reduce its overall size. Applying our method, we have successfully fabricated flexible electrode arrays of less than 100 mu;m in overall diameter. Furthermore, we have demonstrated the utility of our devices for the recording of single-neuron and population electrophysiological activity in vivo.

Neurological implantable medical devices such as retinal prostheses, deep brain stimulators, spinal cord stimulators and cochlear implants communicate with the brain via microelectrodes by transferring electrical signals to targeted nerve cells. For many current and future neuromodulation devices, smaller electrode size is desired because it allows more precise communication with smaller cell populations thus enabling higher resolution for both stimulation and recording. Additionally, smaller electrodes enable smaller, less invasive device designs. Charge injection limits present a major barrier to further decreasing stimulating microelectrodes size.
Previously we have reported on a novel electrochemically deposited 60:40% platinum-iridium (PtIr) electrode material that helps expand the microelectrode design space by increasing maximum charge injection capacity to 12.7 mC/cm2.
This study compares power consumption of an electrochemically deposited PtIr microelectrode vs. a standard PtIr probe microelectrode (FHC, Inc.) produced using conventional techniques. Both electrodes were tested using in vitro (PBS solution) and in vivo (live rat) models. Charge-balanced, cathodic first, biphasic current were delivered to the epiretinal surface on rats across 4 pulse durations (0.3, 0.5, 1 and 2 ms) through the stimulating electrode. For each pulse duration, a current-controlled pulse train at four different charge levels was delivered, resulting in 16 stimulus conditions (4 charge levels x 4 pulse durations). The amplitude of the pulses was chosen such that the charge in the cathodic phase was between 10 and 60 nC. Stimulus pulses were supra-threshold. Power consumption was measured by multiplying the biphasic current amplitude by the polarization voltage measured at the electrode-fluid interface over the stimulus time duration. In all scenarios tested, the PtIr coated electrode demonstrated statistically significant lower power consumption compared to the conventional electrode. Greater power savings were realized with longer pulse widths. For example, for a 50 nC pulse, a 0.3 ms duration consumed 18% less power while a 2 ms duration consumed 51% less power, compared to standard PtIr microelectrode. Thus, the exact amount of power savings will depend on a number of factors. These results were corroborated by the in vitro study.

Copolymer films of ethylenedioxythiophene (EDOT) with a carboxylic-acid functional EDOT (EDOTacid) were electrochemically deposited and characterized as a function of the EDOT/EDOTacid comonomer feed ratio. Copolymer films with systematic variations in EDOTacid content were electrochemically deposited from propylene carbonate solutions onto Au/Pd electrodes. Chemical characterization of the films, via Fourier transform infrared (FTIR) and X-ray photoelectron spectroscopy (XPS), confirmed the presence of both EDOT and EDOTacid units. Toluidene blue assays showed that the surface concentration of the carboxylic acid groups increased as the monomer feed ratio of EDOTacid increased, and contact angle measurements confirmed the increased hydrophilicity of the films with increasing EDOTacid content. Electrochemical impedance spectroscopy (EIS) and cyclic voltammetry showed that the films had comparable charge storage capacities regardless of their composition, indicating that the charge transport process was dominated by the conjugated EDOT backbone. The morphology of the films varied depending on the monomer feed ratio. These methods provide a facile means for synthesizing electrically active carboxylic-acid functional PEDOT copolymer films with tunable surface morphologies and hydrophilicity. Future work will include functionalizing the PEDOT copolymer films with laminin-based and other peptides, to provide enhanced neural interfaces.

Long-term functionality and stability of neural interfaces with complex geometries is one of the major challenges for chronic clinic applications due to lack of effective encapsulation. We report a bi-layer encapsulation that consists of atomic layer deposited (ALD) Al₂O₃ and Parylene C for biomedical implantable devices. ALD Al₂O₃ functions as a highly effective moisture barrier, while Parylene C is a barrier to many ions and also prevents corrosion of Al₂O₃ by liquid water.
Utah Electrode Array (UEA) based neural interfaces with different configurations were used to evaluate the performance of bi-layer encapsulation from different aspects using accelerated lifetime testing. Wired UEAs were used to evaluate the long-term impedance stability, fully integrated wireless devices were used to assess the long-term wireless signal strength and frequency stability, and active devices with ASIC chips were used to monitor current draw over time. Devices were coated with 52 nm of Al₂O₃ deposited by plasma-assisted (PA) ALD, followed by a 6-µm thick Parylene-C layer deposited by CVD using Gorman process. A-174 (Momentive Performance Materials), an organosilane, was used as adhesion promoter between the Al₂O₃ and Parylene C.
A three step self-masked process was developed to selective etch the coating materials on the tips of UEA in order to expose the iridium oxide underneath for neural recording/stimulation. Laser ablation was used to remove the Parylene C layer, followed by 2 minutes of oxygen plasma for removing the carbon residual. Then buffered oxide etch was used for Al₂O₃ removal. The devices were then put into saline solution for soak testing.
Impedance for wired array was measured at 1 kHz. Median impedance increased from 61 kOmega; to 160 kOmega; over 960 equivalent soaking days at 37 °C. Compared with the typical tip impedance drop of Parylene coated UEA, the increase of impedance of bi-layer encapsulated UEA is a combination of excellent coating insulation and PBS etching of silicon. For the wireless neural interfaces, the power-up frequency was constantly ~ 910 MHz and the RF signal strength was stably around -73 dBm during equivalent soaking time of 1000 days at 37 °C (still under soak testing). The current draw of flip-chip bonded ASIC chip was stably about 3 mA during 228 equivalent days of soak testing at 37 °C.
In summary, we have demonstrated 3 years of lifetime on Al₂O₃ and Parylene C bi-layer coated UEA based neural interfaces. The bi-layer coated neural interfaces showed relatively constant impedance and stable RF frequency and signal strength over 3 years. This bi-layer encapsulation can be easily applied to other implantable devices with minor or no modifications.

Precise monitoring of electrical activity in interconnected neurons is critical to probing the behavior of neural networks and decoding the processes behind sensation, movement, and learning. There is increasing demand for new neuronal recording devices that offer high signal to noise measurements, is suitable for continuous long-term recording, and can be parallelized for large-scale brain mapping efforts. However, current patch-clamping and multielectrode arrays provide an incomplete snapshot of cell activity, forcing a choice between high quality intracellular recordings and nonperturbing long-term experiments.
One of the most critical features of intracellular patch clamps is the quality and robustness of the membrane to electrode interface and the corresponding seal resistance. Here we describe our efforts to re-create the quality of transmembrane protein to lipid interfaces by mimicking membrane protein&’s nanoscale hydrophobicity on an inorganic electrode. The key feature of this structure is a narrow (~5nm thick) hydrophobic band that mimics the structure of the lipid bilayer, and is capable of forming a tight seal upon membrane penetration. The junction formed by stealth probes is both mechanically stable and electrically tight, providing gigaohm-level leak resistance.
Here we discuss developing “Stealth electrodes” toward harnessing this nanostructure in a multi-electrode device that has the electrical characteristics necessary for sensitive electrophysiological recordings. In addition to the seal resistance, several other key elements must be addressed, including electrical impedance, electrode-membrane contact, and membrane rupture of soft cells. Once fully realized, this and other new electrical interfaces will pave the way for deeper understanding and assessment of cell behavior and network function.

Stem cells have shown immense potential in the field of regenerative medicine, owing to their ability to self-renew and differentiate into specialized cells. However, realization of this potential at the clinical level is severely hampered by our current inability to deliver genetic materials efficiently to stem cells for modulating their fate. Although virus-based delivery vectors have shown high efficiency, their inherent limitations such as mutagenicity and tumorigenicity has led to development of non-viral alternatives for delivering genetic material to stem cells. To this end, inorganic nanoparticle (NP)-based gene delivery is an attractive approach as these NPs can also offer imaging and therapeutic capabilities owing to their unique morphological, optical, chemical and physical properties, in addition to being efficient delivery vehicles. Among different inorganic NPs, different types of magnetic nanoparticles (MNPs), including metal or metal oxides, metal alloys and more recently, doped MNPs have been utilized for biological applications. These MNPs can also be modified with a biocompatible material (for example, SiO2, polymer, gold), to form a core-shell structure as well as act as a platform for surface functionalization of the MNPs. These resultant core-shell MNPs can have multifunctionalities, resulting from the combination of their innate magnetism, the properties of the shell and the surface functionalization. Such multifunctional core-shell nanoparticles can have several advantages such as enhanced delivery of genetic materials as well as near-infrared absorption and photon scattering which can be utilized for non-invasive monitoring of the stem cells. Herein we describe the synthesis of well-defined magnetic core-shell nanoparticles [MCNPs], composed of a highly magnetic core surrounded by a thin uniform gold shell and their application for magnetically facilitated delivery of genetic materials (siRNA and plasmid DNA) into stem cells. To this end, as a proof-of-concept experiment, we chose neural stem cells as they are known to be sensitive to exogenous transfection reagents as well as difficult-to-transfect. We then delivered siRNAs against neural switch genes (SOX9 and Caveolin-1) and demonstrated that using our MCNP-based delivery approach, we were able to attain high efficiency of differentiation of stem cells into neurons and oligodendrocytes, without compromising stem cell viability and biological functions. Additionally, we also demonstrated the gold shell afforded the capability of dark-field imaging, in addition to improving the solubility and stability of the MCNPs. Thus, the MCNP-based genetic manipulation can potentially be a powerful tool for stem cell applications.

Heat generated via hysteretic power dissipation by ferrite magnetic nanoparticles (MNPs) in the presence of an alternating magnetic field (AMF) was used to remotely control intracellular calcium levels in vitro and in vivo by coupling it to heat-actuated calcium ion channels. While calcium influx has lead to action potential firing in neurons, the timescale required - tens of seconds - suggests that further optimization to this magnetothermal approach is required to shorten the actuation time to biologically relevant timescales.
To explore whether remote firing in neurons can be achieved within milliseconds, we calculated the specific power losses (SLPs) of a range of ferrite MNPs to identify MNPs with the most heat dissipation in our AMF field conditions. To correlate experimentally measured SLPs to qualitative predictions made by our model a comprehensive palette of ferrite MNPs of varying size and composition was synthesized via thermal decomposition of organometallic precursors. We found that iron oxide MNPs ~22 nm in diameter have some of the highest measured SLP values for this material after high-temperature annealing and phase transfer into water. Furthermore, carboxylic functional groups introduced during the phase transfer step allow for easy functionalization of the MNPs for cell specific targeting.
In conjunction with materials optimization, a genetic toolkit is developed to sensitize the neurons to heat as well as introduce functional groups to target MNPs to the cell membrane. We demonstrate how coupling materials with different heat dissipative abilities to genetically modified neurons has implications for minimally invasive deep brain stimulation therapies.

Carbon nanotubes (CNTs), with their intriguing chemical and physical properties, can be engineered and integrated into biological systems to form a new class of hybrid organic/inorganic material at the membrane interface. Such fortunate synergy led to unexpected and intriguing improvements of cell behavior, especially at central nervous system (CNS) level.
Recent work shows that CNTs anchored on planar substrates can promote cell attachment, growth, differentiation and long-term survival of neurons. Furthermore, neurons grown on a CNT meshwork always display more efficient neuronal signal transmission ascribable to the intimate physical interactions between the nanotubes and the cell membrane.
Explanation of these astonishing results is not trivial, but the solution could be find in specific membrane microdomains called lipid rafts (LRs).
In these enriched in cholesterol and sphingolipids domains are confined specific proteins leading to the proposal that LRs are involved in multiple fundamental cellular functions including signaling, cell polarity, protein sorting and trafficking. These consideration and the localization of LRs with the hybrid CNT/cell-membrane interface promote LRs to a central role in the explanation of CNT induced neuronal potentiation.
Unfortunately, due to their sub-micrometrical size and peculiar metastable character, LR fine structure and functionality at molecular level are still poorly understood and remain a difficult system to study.
To address the issue we chose to study LRs at the hybrid CNT/membrane interface combining atomic force microscopy (AFM) investigation with electrophysiology and immunostaining analysis.
AFM was used to characterize at the nanoscale LR-like domains embedded in artificial lipid bilayer deposited on a CNT carpet. This study revealed CNTs ability to pierce the lipid bilayer highlighting their capability to stabilize LR-like domains when a specific LR disrupting agent (methyl-β-cyclodextrin, MBCD) was added.
From the other side, whole-cell patch-clamp recordings on dissociated hippocampal neurons without and in the presence of MBCD were performed. This electrophysiology technique revealed a higher ratio between excitatory and inhibitory synapses on neurons plated on CNT than on control glass when both were treated with MBCD. Such result may be explained having in mind that, in dissociated neuronal cultures, excitatory synapses normally localize closer to the substrate than inhibitory ones. CNT substrate presumably stabilizes closer protein densities, assimilable to LRs, of excitatory synapses.
In the same conditions immunostaining experiments with cholera toxin, a specific LR marker, pointed out higher LR survival and larger dimensions on CNT neurons when treated with MBCD.
In conclusion, the hybrid CNT/membrane layer seems characterized by stabilized LRs. We made the hypothesis that CNTs, acting as “pins”, pinch LRs domains reducing their mobility and inducing enlargement.

Strong coupling between the plasma membrane and the recording electrodes is crucial for sensitive measurement of electrical activities of a cell. We and others have shown that vertical nano- and micro- electrodes promote spontaneous formation of such strong coupling. In particular, vertical nano-electrodes deform plasma membrane inwards and induce negative curvature when the cell engulfs them, leading to a reduction of the membrane-electrode gap distance and a higher sealing resistance. The 3D topology of the nanopillar electrodes is crucial for its enhanced signal detection. Here, we show a nanoelectrode of a new topology, namely nanotubes of iridium oxide, further enhances membrane-electrode coupling and records larger intracellular potentials than nanopillar electrodes. The new topology also significantly increases the time duration of intracellular access. In addition to electrophysiology recording, we investigated how the presence of high membrane curvature induced by vertical nanostructures, affects protein distributions and cellular processes. In particular, we found that the vertical nanostructure induced drastic accumulation of specific proteins, enhances the activity of signaling pathways during neuronal development, and significantly increases the endocytosis process in its vicinity. Those results show the strong interplay between biological cells and nanosized electrode, which is an essential consideration for future development of interfacing devices.

Action potentials play a central role in neurons and cardiomyocytes. Traditionally, patch clamp pipettes and microelectrode arrays are used to record action potentials. However, patch clamping suffers drawback of invasiveness and low throughput whereas microelectrode arrays recording gives low signal recording that are also difficult to interpret. Nanoelectronic devices hold unique advantages to solve these problems in cell electrophysiology measurement. Previously we have demonstrated intracellular recording of rat cardiomyocyte action potentials by noble metal nanopillar electrodes (Nat. Nanotechnology 7, 185-190 (2012)). Here we present nanoelectrodes of a new topology, namely nanotubes of iridium oxide, are capable of action potential recording with increased amplitudes and access durations 1-2 orders of magnitude longer than those recorded by solid gold nanopillars of the same surface area. We demonstrate that these nanotube electrodes are capable of parallel recording of single cell action potentials as well as monitoring the action potentials of single cells throughout their maturation and aging. Finally, we apply our nanoelectrode recording method to study the electrophysiology of human embryonic and induced pluripotent stem cell derived cardiomyocytes. In addition to reliably distinguishing different subtypes of cardiomyocytes, we also monitor with high precision the evolution of their action potentials as differentiation progresses over a period of several weeks. Nanopillar and nanotube electrodes are therefore powerful tools that allow high throughput, minimally invasive intracellular recording of excitable cells such as cardiomyocytes and neurons and the studying of their development.

Softening neural interfaces attempt to solve one the major challenges for the chronic electronic interaction with nervous system: a large mechanical mismatch between the neural tissue (10 kPa) and conventional neural interfaces (150 GPa). Smart softening substrates are stiff prior and during the insertion process (10 GPa) and gradually soften after implantation to less than 7 MPa. This mechanical adaptation has initially shown a reduction in the immune response during a 12 week period in lab rats. In previous work, softening neural interfaces were integrated with standard electrode materials such as Pt, IrOx and TiN. Softening devices showed charge injection capacity (CIC), electrochemical impedance and long term in vitro stability similar to those fabricated on silicon substrates. Nevertheless the electrochemical capabilities of the mentioned materials do not meet the requirements of several neuroprosthetic applications: high charge injection capacity (1 mC/cm2) and chronic electrochemical stability (> 1 year).
In recent years, nano materials for neural electrodes have caught the attention of several research groups. The intrinsic morphology of these kinds of materials offers a high ratio of electrochemical surface area (ESA) to geometric surface area. However in most cases, the growth process requires high temperatures (> 600°C), not compatible with polymeric substrates. In order to overcome this limitation, researchers have adopted electrodeposition of nanomaterials such as carbon nanotubes and nanoparticles. However, the adhesion between electroplated materials and the contact electrode jeopardizes the stability of such neural devices. Delamination of the coating film has been reported; compromising its electrochemical interaction with nearby cells and its use in stimulating chronic applications. Nanostructured electrodes with stable electrochemical performance and well-established biocompatibility are of major interest for the development of high charge injection capacity neural electrodes.
In this work, TiN nanorods are fully integrated with smart softening substrates. Neural electrodes are fabricated using a full photolithography, low-temperature process. AFM and SEM are used to reveal the nanotopography of these novel electrodes. The electrochemical properties are characterized by cyclic voltammetry, and A. C. impedance spectroscopy. CICs greater than 1 mC/cm2 were measured through a transient voltage in a three-electrode setup. Electrochemical stability and other properties are directly compared with softening devices with gold electrodes coated with PEDOT through electropolymerization and electrodes coated with carbon nanotubes. Through the development and understanding of TiN nanorod electrodes, applications that demand a higher CIC will be able to take full advantage of the new generation of softening neural interfaces and perhaps enable chronic interface with the nervous system.

Nanoscale materials enable unique opportunities at the interface between the physical and life sciences, and the interfaces between nanoelectronic devices and cells, cell networks, and tissue makes possible communication between these systems at the length scale relevant to biological function. In this presentation, the development of nanowire nanoelectronic devices and their application as powerful tools for the recording and stimulation from level of single cells to tissue will be discussed. First, a brief introduction to nanowire nanoelectronic devices as well as comparisons to other tools will be presented to illuminate the unique strengths and opportunities enabled by active electronic devices. Second, opportunities for the creation of powerful new probes based on controlled synthesis and/or bottom-up assembly of nanomaterials will be described with an emphasis on the creation of nanowire probes capable of intracellular recording and stimulation at scales heretofore not possible with existing electrophysiology techniques. Third, we will take an ‘out-of-the-box&’ look and consider what the future might hold in terms of merging nanoelectronics with cell networks in three-dimensions to ‘synthesize&’ ‘cyborg&’ tissues, as well as novel tools for in-vivo recording. The prospects for blurring the distinction between electronic and living systems in the future will be highlighted.

Stable long-term intracortical recordings have the potential to revolutionize our understanding of the human brain and pave the way for a new generation of closed-loop neuroprosthetics. However, while the feasibility of long-term recordings from intra-cortical electrodes has been demonstrated, their reliability remains unsatisfactory. We report that the extent and duration of compromise of the blood brain barrier is predictive of long term intra-cortical electrode function. We establish a mechanistic link between myeloid cell presence, cytokine expression, local neurodegeneration and intra-cortical electrode failure in the adult CNS.

The development of the high-resolution cell-specific map of the neural activity associated with a particular behavior presents one of the major challenges in modern neuroscience. This dynamic electrophysiological mapping is particularly difficult in deep brain regions and within the spinal cord. It is further aggravated by the limited long-term stability of the existing high-density electrophysiological platforms and the inability to uniquely identify cell types during recording.
Inspired by optical fiber industry, we employ thermal drawing process to create a new generation of multifunctional neural probes, which combine electrophysiological recording, light and drug delivery, while minimizing tissue damage. Our devices can be fabricated solely of polymers, which provide unprecedented flexibility of the waveguides and recording electrodes and thus are compatible with applications in peripheral nerves and spinal cord. Alternatively, by including low-melting temperature metal alloys into the polymer matrix, we can achieve single-neuron resolution with electrode diameters as low as 1 mu;m, virtually unlimited number of channels and controlled geometry and pitch. Furthermore we apply our technology to the design of optoelectronic neural scaffolds that may provide an active repair strategy for damaged nerves.
To date we have evaluated our fiber-inspired technology in vivo and demonstrated the utility of these devices for simultaneous neural recording, optical and pharmacological interrogation within the mouse brain and spinal cord. Based on our preliminary data we believe that our fabrication strategy may provide a pathway towards the dynamic neural activity mapping as well as customizable high-resolution neural prosthetics.

The continuous appearance of new medical treatments based on neuromodulation and neurostimulation is providing a strong impulse in the development of new electrodes and leads for both recording and stimulation of neural tissue having complex geometry, high number of electrodes, reduced invasivity, low production costs, easy surgical insertion and reduced risk of failure due to breakage or migration.
WISE has developed a technique for the realization of electrodes for neural stimulation addressing all these issues. This is obtained by means of the Supersonic Cluster Beam Implantation (SCBI) of neutral metal cluster in a elastomeric substrate. The clusters are produced in the form of a supersonic beam and are implanted at Room Temperature in the polymer substrate forming a metal-polymer nanocomposite layer. The technique is extremely versatile in terms of both the substrate (material, hardness, thickness) and the metals (platinum, iridium, gold, etc.). This process avoids both sample heating and sample charging enabling the metallization of soft polymeric materials with a thickness that can be lower than 10 µm. The metal films obtained show excellent electrical conductivity, excellent stretchability and foldability and strong adhesion to the substrates.
We have measured platinum electrodes both in-vitro and in-vivo. The high surface area of the metal prepared by SCBI technique makes the electrode impedance in the frequency range useful for practical purposes (10^-2 - 10³ Hz) significantly lower than the flat bulky metal surface commonly used in commercial electrodes.
The electrodes obtained by this technique adapt intimately to the tissues surfaces of either the central or the peripheral nervous system, thus making them excellent candidates to perform stimulation and nerve signal recording. We have performed neural signal recording and stimulation on guinea pig brain showing excellent electrode sensitivity and very good stability over time.
Finally the electrodes placed under 100.000 repetitive stretching cycles up to 50% elongation shows no degradation of the electrical properties, demonstrating to be suitable for long term electrode implantation in regions of the body continuously subjected to stress.

Traditional neural prosthetic devices induce chronic inflammation because of the mismatch in mechanical properties with neural tissue and relatively large size of the electrode. The new generation of neural devices requires flexible materials with excellent electrochemical performance which are not available in the toolbox of classical neurotechnology. In addition, the development of such materials must be combined with their integration with micromanufacturing techniques and innovative methods of implantation. In this study, we demonstrate the possibility of mechanical flexible devices from a purposefully developed CNT nanocomposite. The first nanocomposite based flexible neural electrodes were microfabricated using MEMS technology and implanted into rat motor cortex. The devices were successfully visualized with ex vivo MRI and photoacoustic imaging. In vivo evaluation demonstrated their functionality by successful registration of brain activity. This study opens the door for different nanocomposites to be used for long-term brain-machine interface.

Researchers have made significant progress in the neural stimulation and recording technologies in the past decades. The majority of the devices available to date, however, have been developed for treatment of brain diseases and brain-machine interfaces, not for recordings or intervention in the spinal cord. This is likely because of the high flexibility and fibrous geometry of the spinal cord, impeding the development of such probes. Recently, epidural electrical stimulation has shown promise in the treatment of spinal cord injury; however, it ignores cellular organization of the spinal cord and hence cannot yet be applied to restoring complex voluntary motor functions. With the recent advances in optogenetic neural stimulation technologies it became possible to control specific neural populations, while simultaneously recording the evoked neural activity. Thus it is highly advantageous to create flexible multifunctional neural probes that can conform to spinal cord geometry, while providing optical stimulation and neural recording.
We have fabricated fiber-inspired devices which consist exclusively of polymers, including electrodes and waveguides. Our devices maintain light transmission in the visible range even at large bending deformations, and, unlike conventional brittle silica fibers, can conform to the mobile spinal cord. Moreover, our embedded conductive polymer electrodes have elastic moduli ~3 orders of magnitude lower than the traditional electrode materials (metals and doped silicon). We demonstrate the utility of our devices for light transmission and neural recording in transgenic Thy1-ChR2-YFP mice expressing light-sensitive protein channelrhodopsin 2 in neurons. To our knowledge, this is the first demonstration of an optically evoked electrophysiological and behavioral response in vivo during optogenetic stimulation of a mouse spinal cord.

Probing and stimulating neural tissue are important to the study of information processing as well as pathologies of the nervous system. While various types of neural probes have been developed in recent years, seamless integration of optical, electrical and chemical stimulation functionalities in a form factor that allows for chronic in-vivo measurements remains a challenge. Here we demonstrate fiber neural probes fabricated using a thermal drawing process. This process allows for high throughput (kilometer-long fibers within one draw), integrated functionality (electrical recording, optical guidance, and drug delivery), and the incorporation of multiple disparate materials into functional architectures. We have produced hollow-core and flexible all-polymer fibers with four conductive polymer electrodes and a wave-guiding layer. The hollow core is used as a drug delivery channel. This flexible fiber has a minimum wall thickness of ~60 µm after etching away the outer layer, and maintains low damage to the brain during in vivo measurements. Stimulation and recording measurements in transgenic Thy1-ChR2-YFP mice have been demonstrated. This technology paves the way for more complex flexible fiber neural probes for simultaneous neural stimulations and recordings.

In neuroscience generally, and in optogenetics in particular, an ability to insert light sources, detectors, sensors and other active semiconductor components into precise locations of the deep brain provides important capabilities. Here, we summarize our recent work on injectable classes of cellular-scale optoelectronics that offer such features, with examples of wireless behavioral control over freely moving and socially interacting animals. These ultrathin, mechanically compliant, biocompatible devices afford minimally invasive operation, with broad utility in both biomedical science and engineering.