Mechanically Adaptive Intracortical Implants Improve the Proximity of Neuronal Cell Bodies

Home Search Collections Journals About Contact us My IOPscience

Mechanically adaptive intracortical implants improve the proximity of neuronal cell bodies

This content has been downloaded from IOPscience. Please scroll down to see the full text. 2011 J. Neural Eng. 8 066011 (http://iopscience.iop.org/1741-2552/8/6/066011)

View the table of contents for this issue, or go to the journal homepage for more

Download details:

IP Address: 137.99.31.134 This content was downloaded on 22/05/2015 at 14:26

Please note that terms and conditions apply. IOP PUBLISHING JOURNAL OF NEURAL ENGINEERING J. Neural Eng. 8 (2011) 066011 (13pp) doi:10.1088/1741-2560/8/6/066011 Mechanically adaptive intracortical implants improve the proximity of neuronal cell bodies

J P Harris1,2, J R Capadona1,2, R H Miller3,BCHealy4,5, K Shanmuganathan6, S J Rowan1,2,6, C Weder6,7 and D J Tyler1,2,8

1 Department of Biomedical Engineering, CWRU, 2071 Martin Luther King Jr Drive, Wickenden Bldg, Cleveland, OH 44106, USA 2 Rehabilitation Research and Development, L Stokes Cleveland VA Medical Center, 10701 East Blvd Mail Stop 151 AW/APT, Cleveland, OH 44106-1702, USA 3 Department of Neurosciences, CWRU, 2109 Adelbert Road, Cleveland, OH 44106, USA 4 Department of Neurology, Brigham and Women’s Hospital, Partners MS Center, Harvard Medical School, 1 Brookline Place West, Suite 225, Boston, MA 02445, USA 5 Biostatistics Center, Massachusetts General Hospital, 50 Staniford St Suite 560, Boston, MA 02114, USA 6 Department of Macromolecular Science and Engineering, 2100 Adelbert Road, Kent Hale Smith Bldg, CWRU, Cleveland, OH 44106, USA 7 Adolphe Merkle Institute and Fribourg Center for Nanomaterials, University of Fribourg, Rte de l’Ancienne Papeterie/PO Box 209, CH-1723 Marly 1, Switzerland E-mail: [email protected]

Received 6 March 2011 Accepted for publication 27 September 2011 Published 2 November 2011 Online at stacks.iop.org/JNE/8/066011

Abstract The hypothesis is that the mechanical mismatch between brain tissue and microelectrodes influences the inflammatory response. Our unique, mechanically adaptive polymer nanocomposite enabled this study within the cerebral cortex of rats. The initial tensile storage modulus of 5 GPa decreases to 12 MPa within 15 min under physiological conditions. The response to the nanocomposite was compared to surface-matched, stiffer implants of traditional wires (411 GPa) coated with the identical polymer substrate and implanted on the contralateral side. Both implants were tethered. Fluorescent immunohistochemistry labeling examined neurons, intermediate filaments, macrophages, microglia and proteoglycans. We demonstrate, for the first time, a system that decouples the mechanical and surface chemistry components of the neural response. The neuronal nuclei density within 100 μm of the device at four weeks post-implantation was greater for the compliant nanocomposite compared to the stiff wire. At eight weeks post-implantation, the neuronal nuclei density around the nanocomposite was maintained, but the density around the wire recovered to match that of the nanocomposite. The glial scar response to the compliant nanocomposite was less vigorous than it was to the stiffer wire. The results suggest that mechanically associated factors such as proteoglycans and intermediate filaments are important modulators of the response of the compliant nanocomposite. (Some figures in this article are in colour only in the electronic version)

8 Author to whom any correspondence should be addressed.

1741-2560/11/066011+13$33.00 1 © 2011 IOP Publishing Ltd Printed in the UK J. Neural Eng. 8 (2011) 066011 J P Harris et al 1. Introduction reactivity, measured through the expression of glial fibrillary acidic protein (GFAP) (Lu et al 2009). Despite increasing evidence for the importance of cellular Given the evidence for the contribution of mechanical mechanotransduction in tissue repair and homeostasis (Clark force to the glial scar from in vitro, in silico and in vivo et al 2007, Ingber 2006), the role of mechanical mismatch has experiments, we hypothesize that the in vivo glial scar not been fully elucidated in an in vivo study to explain the response is influenced by a chronic stiffness mismatch between interplay of chemical and mechanical factors that contribute the soft brain tissue and stiff intracortical implants. To to glial scarring surrounding intracortical implants. Previous create an in vivo device to characterize the importance of research in vitro and in silico has supported the importance mechanical compliance of intracortical implants in both the of mechanical signaling in several cell types in the brain. chronic inflammatory reaction and the long-term electrode Cultured astrocytes have been shown to respond to mechanical performance, we used our recently developed mechanically stimuli via calcium signaling (Ostrow and Sachs 2005). adaptive polymer nanocomposite (NC), previously described Investigations have shown that higher strain rates for cultured in detail (Capadona et al 2007, 2008, 2009, Shanmuganathan astrocytes lead to an increased reactivity (Cullen et al 2007). et al 2010a, 2010b, 2009). Our previous work has Cell types have responded differently to substrate stiffness as demonstrated that the polymer NC can reduce its tensile well. Notably, the rate of astrocyte and neuron proliferation as storage modulus from 5 GPa to 12 MPa within 15 min upon well as oligodendrocyte spreading and neuronal branching was exposure to simulated physiological conditions in artificial influenced by substrate stiffness (Kippert et al 2009, Flanagan cerebral spinal fluid at 37 ◦C (Capadona et al 2007, 2008, 2009, et al 2002, Georges et al 2006). Substrate stiffness is also Shanmuganathan et al 2010a, 2010b, 2009). The polymer NC known to shift cell differentiation in mesenchymal stem cells has modest aqueous swelling of 60–75%, in comparison to the to be neurogenic, myogenic or osteogenic (Engler et al 2006). several hundred per cent exhibited by hydrogel approaches. In addition to the effect of substrate stiffness, in silico Here, we have created model electrodes from the adaptive modeling studies indicate that indwelling electrodes exert polymer NC material. We have demonstrated that the NC forces on local populations of cells (Lee et al 2005, Subbaroyan material performs similarly in vitro and in vivo (Harris et al 2005, McConnell et al 2007). Over time, the implanted et al 2011). While the previous work performed modest electrode is anchored to the tissue via the extracellular matrix histological analysis, this is a more complete histological (ECM) and neural inflammatory cells (including microglia and report on their use as cortical implants, demonstrating that astrocytes), resulting in cellular attachments to the electrode there are both foreign body elements and mechanical effects modifying the forces exerted on the brain tissue (McConnell contributing to the overall tissue response at the biotic–abiotic et al 2007). Micromotion associated with electrode movement interface. These components of the inflammatory response within the tissue and the mechanical properties of the electrode appear to be separate but intricately coupled to the neuronal dynamically change the level of exerted forces on the cortical response. tissue during scar maturation. This phenomenon can also be exacerbated as a result of mechanically tethering the electrodes 2. Materials and methods to the skull. In vivo studies which focus on the effects of electrode tethering have shown that untethered implants reduce 2.1. Implant fabrication and imaging the extent of the glial scar (Biran et al 2007,Kimet al 2004, Subbaroyan 2007). It has been suggested that a reduced glial Implants consisted of two types, referred to as nanocomposite scar results from the reduction in forces applied to the tissue. (NC) and wire. NC implants were created by casting Alternatively, it has been proposed that less scarring is the films from a dimethylformamide solution of poly(vinyl result of limited meningeal ingrowth in response to untethered acetate) (PVAc) and tunicate whiskers, as reported previously implants (Subbaroyan 2007). (Capadona et al 2008). The NC had a cellulose whisker To investigate the effects of mechanical mismatch content of 15% w/w. The resulting films were used to create between the electrodes and the cortical tissue, several groups sheets via a custom mold in a hot press (Carver, Wabash, IN). have developed electrode substrates and substrate coatings Sheets were used to create 3 mm long probes that measured from materials such as polyimide, polydimethylsiloxane 100 ± 5 μm in thickness and 203 ± 19 μm in width with and parylene, which are more compliant than materials an approximately 45 ± 3◦ angle tip (figure 1(a)) via a cutting traditionally used to create electrodes. However, these process. The cutting process created a NC with two types materials have had limited success in attenuating glial scar of surfaces: two pressed sides (figure 1(c)) and two cut sides formation. A limitation to many of these studies is that (figure 1(d)). Both implant types were single shank probes such materials still have moduli six orders of magnitude with nearly the same cross-sectional area, refer to Harris et al larger than those of the brain, while also introducing different 2011 for a characterization of implant mechanical properties surface chemistries that could confound the tissue–material of the NC material. interaction (Lee et al 2004, Nikles et al 2003, Takeuchi Wire implants consisted of 50 μm diameter tungsten et al 2004,Westeret al 2009, Takeuchi et al 2005) and wires (AM-Systems, Sequim, WA) from which the Teflon complicate the interpretation of these results. Of interest is insulation was mechanically stripped using cutting tweezers. that surface treatments of electrode materials with a poly(vinyl Wire implants (figure 1(b)) were dip-coated in a solution of alcohol)/poly(acrylic acid) hydrogel have reduced astrocyte PVAc (MW 113 000 g mol−1, density 1.19 g cm−3,Sigma

2 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(b)

(a)

Figure 1. Materials bilaterally implanted in rodent cortex. (a) Microscopic picture of the PVAc-NC polymer nanocomposite implant, 3 mm in length, 200 μm wide, 100 μm thick. (b) Microscopic picture of the PVAc-coated tungsten wire. The tungsten wire is 50 μm in diameter before coating. The diameter after coating is ∼160 μm. Light white lines outline the wire within the PVAc coating. (c)–(e) SEM images of implants. (c) SEM image of the pressed side of the PVAc-NC implant. (d) SEM image showing the cut side of the PVAc-NC implant that is rough. The image also shows the pressed side of the PVAc-NC implant at the bottom of the image. Inset: a higher magniﬁcation SEM image shows the cellulose whiskers on the cut side of the PVAc-NC implant. (e) SEM image of the PVAc-coated tungsten wire.

Aldrich) in toluene at a concentration of 10% w/wat70◦C. A 2.3. Surgical procedures droplet of PVAc was placed on the edge of a glass dish and the All surgeries were performed under sterile conditions with tungsten wire was steadily dragged through the drop repeatedly autoclaved instruments. Both wire and NC implants were about 20 times or until the coated wire had a diameter sterilized via an autoclave before implantation. The modulus ∼ of 160 μm. Samples were then cut into 3 mm long sections, of NC implants was measured and verified to be unaltered and with the aid of a microscope, sections of uniform diameter by the autoclave process (data not shown). The same surgeon of 160 μm were selected for implantation. performed all surgeries while an assistant monitored anesthesia A scanning electron microscope (SEM) imaged the NC levels and manipulated equipment. All procedures were and wire implants at high magnification. Both the NC and approved by the Case Western Reserve University Institutional wire implants were covered with a thin layer of palladium, Animal Care and Use Committee (CWRU IACUC) and approximately 7–10 nm. The samples were then attached to a minimized the pain and discomfort of the animal. chuck with Teflon tape and placed inside the S-4500 Hitachi Male Sprague–Dawley rats from 206 to 335 g were SEM. Samples were imaged at various magnifications and implanted bilaterally with a wire and a NC probe. The implants angles (figures 1(c)–(e)). were affixed to sterile ceramic forceps tips (Fine Science Tools, Foster City, CA) with glucose. Anesthesia was induced via an intraperitoneal injection of ketamine (80 mg kg−1) and xylazine (10 mg kg−1). After the animal reached the surgical 2.2. Water contact angle plane, as determined by lack of response to toe pinch, the To analyze the surface hydrophilicity of the two implants, animal’s head was shaved and eyes protected via sterile ocular lubricant. A circulating water mat was placed underneath the water contact angle measurements were performed on sheets animal while the animal was fitted with a thermometer and of the NC and neat PVAc. Sheets of NC were made as pulse oximeter (Surgivet, Waukesha, WI) to monitor vitals. described above, and the fabrication of neat polymer sheets The animal was secured on a stereotaxic frame (Stoelting Co., was performed as described in the previous literature (Harris Wood Dale, IL) fitted with a gas mask. The animal was et al 2011). The FTA 200 (First Ten Angstroms, Portsmouth, ventilated (Engler, Hialeah, FL) with oxygen; once the animal VA) instrument dispensed and imaged droplets on the sheets. started to come out of the initial ketamine/xylazine induction Three to five samples from sheets were imaged, and the FTA32 (determined by whisking, reacting to toe pinch, raised heart software program analyzed the water contact angle. On the beat), isoflurane (1–3%) was introduced to oxygen flow to image, the user set the base of the droplet and one-fourth of maintain a surgical level of anesthesia. the envelope of the drop. The program calculated the angle After securing the animal, marcaine (∼0.1 mL, diluted 1:5 using the base and envelope of water droplet. with saline) was injected before incision as a local anesthetic,

3 J. Neural Eng. 8 (2011) 066011 J P Harris et al Table 1. Summary of primary antibodies. Antibody Vendor (Part) Dilution Isotype Cell type/labeling

GFAP Chemicon (AB5804) 1:1000 Rb IgG Astrocytes NeuN Chemicon (MAB377) 1:500 Ms IgG1 Neurons CD68 (ED1) Chemicon (MAB1435) 1:250 Ms IgG1 Activated macrophages, microglia IBA1 Wako (019-19741) 1:615 Rb IgG Microglia CS56 Sigma (C8035) 1:500 Ms IgM Chondroitin sulfate proteoglycans Vimentin Sigma (V6389) 1:200 Ms IgG1 Varied- immature/reactive astrocytes, microglia, endothelial cells and ﬁbroblasts

and the head area was sterilized with betadine and alcohol. 2.4. Fixation, tissue preparation and immunohistochemistry An incision along the midline exposed the skull. Skin was After the designated timeframe, four or eight weeks, animals retracted using an eye speculum. Bilaterally, a 3 mm biopsy were euthanized by transcardial perfusion. Animals were punch (Miltex, York, PA) was used to drill two implant sites first anesthetized to a surgical plane as described above. centered 4.5 mm caudal to the bregma and 3 mm lateral The perfusion rate was controlled via a mechanical pump to the midline. The biopsy punch prevented heat damage (VWR, West Chester, PA) with 500 mL of D-PBS (Invitrogen, caused by motorized drills and provided a well-defined 3 mm Carlsbad, CA) or when no trace of blood was seen exiting opening. Three screws were implanted for later fixation of the right atrium. Approximately 250 mL of 10% buffered dental cement. For both openings, one screw was placed formalin (Fisher, Pittsburgh, PA) fixed the tissue of the animal. 1 mm caudal to the biopsy punch-produced openings. The The brain and probes were placed in formalin for 24 h, after remaining screw was placed 4 mm rostral to the biopsy hole which the brain was then transferred to fresh formalin. The on the animal’s left side. After three 0–80 × 3/32 screws brain remained in fresh formalin for up to one week, until it were attached to the skull, a fine 45◦ angle microprobe (Fine was cryoprotected in 30% sucrose solution. Science Tools, Foster City, CA) was used to create an opening After removing the probes, brains were frozen in in the dura (approximately 1 mm diameter) for both implant embedding molds (Electron Microscopy Products, Hatfield, sites. After manually lowering the implant to the surface PA) containing OCT compound (Sakura, Tokyo, Japan) and of the brain, an automated mechanical inserter controlled by mounted and sectioned horizontally by a cryostat (Thermo LabView 7.1 (National Instruments, Austin, TX) inserted the Fisher Scientific, Waltham, MA) creating 30 μm thick slices. implants 2 mm into the tissue at a rate of 2 mm s−1.A1mm Slices chosen were spread over the whole length of the implant, tab was left exposed above the brain tissue to enable tethering while remaining slices were preserved in a cryoprotectant solution (Watson et al 1986). Following procedures outlined via Kwik-Sil (WPI Inc., Sarasota, FL). Kwik-Sil sealed the previously (Biran et al 2005), slices were labeled via a free- implant sites, and dental cement (Stoelting, Wood Dale, IL) floating protocol. was applied to cover the implant sites and screws. The skin Briefly, slices were placed in a blocking solution overnight was sutured with 5-0 polypro sutures (Henry Schein, 1016409, at 4 ◦C. The blocking solution was composed of 4% (v/v) Melville, NY) enclosing the dental cement headcap. Each normal goat serum (Sigma-Aldrich, St Louis, MO), 0.3% animal received one wire and one NC implant in random order (v/v) Triton-X-100 (Sigma-Aldrich) and 0.1% (w/v) sodium and side placement. azide (Sigma-Aldrich) in 1× D-PBS (Invitrogen, Carlsbad, After surgery, an animal was isolated in a clean cage on a CA). Slices were then placed in a fresh blocking solution warm water blanket and monitored every 30 min until sternal. with primary antibodies (table 1) overnight at 4 ◦C. On the One antibiotic shot of cefazolin (20 mg kg−1) and one analgesic third day, slices were placed in the blocking solution with shot of buprenorphine (0.01 mg kg−1) was given on the day secondary antibodies, goat anti-rabbit IgG or anti-mouse IgG1 of surgery. Two cefazolin shots, of the same dose, were given or IgM, Alexa Fluor 594, 488 and DAPI dilactate (Invitrogen, the day following surgery, while buprenorphine was given as Carlsbad, CA) for 1 h at room temperature. After primary needed. No steroids were applied at any time throughout and secondary application periods three 15 min washes in the implant period. Each rat was monitored regularly after PBS were performed on a shaker. Slices were mounted surgery for signs of distress, weight loss and dehydration. In and cover slipped using Fluoromount-G (Southern Biotech, Birmingham, AL). rare cases, complications with infections and sutures caused early euthanasia and these animals were removed from the study. In the remaining animals, some animals exhibited a 2.5. Image acquisition and analysis large void around the implant. The response was analyzed Images of slices were taken by an Axio Imager.Z1M and determined to be a result of excessive bleeding after Microscope and AxioCam MRm CCD camera (Zeiss, implantation. These hemispheres were excluded from analysis Thornwood, NY). Implant sites were centered in the field, since the response was due to insertion trauma rather than the and exposure settings were set to prevent overexposure. The response to the implant. settings were then maintained the same for the whole set of

4 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(a) (b)

Figure 2. Quantification of fluorescent immunohistochemistry staining via two methods. Horizontal slices of representative brains are shown. (a) Representative image indicating the 100 radial lines drawn to compute the pixel intensity radiating outward from electrode–tissue interface. (b) Representative image showing computer- and user-selected cell bodies to compute the distance of cell bodies from the electrode–tissue interface. Inset: close-up of marked cell bodies. stained images. Four automated exposures were performed of the 100 rays were averaged to measure the intensity of on every site: one bright field and three fluorescent colors. the pressed sides versus the cut sides of the NC. Based on Seventeen animals with bilateral implants were analyzed at the cutting technique used to create the probes, the pressed four weeks. After removing hemispheres with excessive side was the longer of the two sides. The majority of images insertion trauma, there were 11 trials for the wire and 16 analyzed had clearly discernable long and short sides. The for the NC. Seven animals were analyzed after eight weeks. images with indistinguishable sides were left out of analysis. Statistical analysis based on prior work predicted that about Approximately 10 of the 100 rays were used to determine the ten animals were needed for analysis. On average, three slices image intensity based on a single side. For each slice, the were used per animal for a given label. two long sides were averaged together and the two short sides A custom Matlab R2009b (MathWorks, Natick, MA) were averaged together. Therefore, approximately 20 of the script was developed to measure the intensity around the 100 rays determined the reaction to the cut sides, and a separate implant. The script automatically detected the border and area 20 of the rays determined the reaction for the pressed sides. of the implant based on the bright field image. Bright field The intensity over the first 20 and 50 μm was examined for images were used for border detection since it was difficult DAPI and ED1 to examine the effect of the surface roughness to see the actual border in fluorescent images. Therefore, on the tissue response. A mean intensity was computed over some fluorescent images appeared to have a larger implant the distance as well as the 95% confidence interval of the area than actual. This effect and the variable shrinkage of mean intensity. DAPI analysis was performed to analyze the tissue during histology contribute to the apparent variation in total cell density around the implants. Analysis of ED1 was size of implants pictured in fluorescent images. Calculating performed since the previous literature has indicated that ED1 the size of the implants in images, the average sizes of the positive cells are often closest to the implant (Biran et al 2005). implants were similar: 23 329 ± 3279 μm2 for the NC and The automatic detection removed human error across 24 513 ± 5264 μm2 for the wire. samples, but could result in the inclusion of up to 3–5 μm One hundred, equally spaced, radial lines emanated from of the image without tissue, which could slightly distort data the centroid of the implant area. For each of the 100 rays, in this range. The distance between the implant and the peak the intensity (pixel values) over the first 200 μm immediately external to the border was measured around the implant site tissue response error is within 5 μm, but is consistent between (figure 2(a)). The average intensity of the 100 rays was divided the experimental conditions and does not affect the results of by the background measurement of the image to account this work. for differences between images due to photobleaching and Another custom Matlab script used the same border other effects inherent in fluorescent immunohistochemistry. detection algorithm but computationally isolated and Background intensity values were measured per image at the highlighted nuclei (figure 2(b)). After the automated marking, edge of the image, at least 300 μm from the implant and outside an investigator manually reviewed the identification and added the region measured for the response. Several measurements or subtracted nuclei missed by the algorithm. After marking, were made for each image including peak intensity, near Matlab computed the distance from each nucleus to the border integral intensity (0–50 μm) and far integral intensity (50– along the ray emanating from the centroid of the implant 100 μm) as well as certain intervals dictated by the data and site. These distances were binned into 10 μm increments, highlighted in the text. and a histogram was created for the number of nuclei. The For certain immunohistochemistry (IHC) labels, an counts were scaled by dividing by the area of the corresponding additional evaluation of IHC intensity was performed to 10 μm wide concentric polygonal donut to provide the neuron determine the effect of the NC side on the response. A subset density.

5 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(b) (a)

Figure 3. Analysis of NeuN immunohistochemistry. (a) Quantiﬁcation of NeuN based on the distance from the tissue–implant border in response to implants: four week NC (square), four week wire (asterisk), eight week NC (cross), eight week wire (circle). The points represent histogram counts in 10 μm intervals. Counts have been scaled based on area as well as background count per image. (b) Average count of NeuN as a function of distance from the tissue–implant border ± standard error. Four week NC (white), four week wire (black), eight week NC (gray), eight week wire (striped), ∗ = p < 0.05 for intraweek comparisons, ∗∗ = p < 0.05 for interweek comparisons. Representative images of NeuN for (c) four week NC, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm.

2.6. Statistics weeks, however, the neuronal density within 200 μmofthe NC implant is significantly greater than it is around the wire In order to estimate the effect of the implant type on the (figures 3(a), (b)). At eight weeks, the neuronal density around outcome, the differences between measured values (e.g. the both implant types is no longer significantly different at any peak intensity, integrals, cell count, etc) were modeled using distance from the implant (figures 3(a), (b)). For the wire, the a multilevel model. The model (equation (1)) included a neuronal density within a radius of 20 μm from the interface baseline level for the outcome (β0), a fixed effect for the is significantly greater at eight weeks than at four weeks difference in the outcome based on the implant type (β1), (p = 0.04). The data at other intervals are not statistically a random effect for each animal (b0,i), a random effect for significant, but trends indicate that the neuronal density around each slice within each animal (c0,ij) and residual error (eij k ): rigid wire implants increases between four and eight weeks to a

Outcomeij k = β0 + β1 ∗ IMPLANT TYPEij k + b0,i density similar to the NC at both four and eight weeks. Further, + c + e , (1) there are no trends in decreasing NeuN density between four 0,ij ij k and eight weeks with the NC. where i indexes the rat, j indexes the slice and k indexes the side. If β1 was significantly different than zero, the 3.3. Astrocytes implant had a significant effect on the outcome of interest. All statistical analysis was completed in the statistical package R Analysis of GFAP labeling indicates the reactive astrocytic (R Foundation for Statistical Computing, Vienna, Austria) (R response. The intensity of GFAP labeling peaks approximately Development Core Team 2009, Pinheiro et al 2009). 10–20 μm from the tissue–implant border and gradually decreases further away (figures 4(a), (b)). For both four and eight week data, the 0–50 μm region from the implant 3. Results exhibits hypertrophied astrocytes with a reactive morphology (figures 4(c)–(f), (c) inset). Astrocytes outside of 200 μm 3.1. Water contact angle do not exhibit GFAP labeling (figures 4(c)–(f)). At four Water contact angle measures show that there is no significant weeks, the maximum intensity and the 0–50 μmintegral difference between the hydrophobicity of the PVAc wire of intensity are not significantly different between the tissue coating and the PVAc-NC material being studied. The NC around the wire and NC implants, though the 50–100 μm had a contact angle of 70.3 ± 2.2◦, and the neat material had a intensity integral is significantly greater (p = 0.05) for wire contact angle of 64.0 ± 3.1◦. A two-sample t-test did not show samples compared to NC at four weeks. At eight weeks, the any significant differences in the contact angle (p = 0.22). intensity maximum is significantly greater around the wire implants than that around the NC implants (p = 0.04), and the 3.2. Neuronal nuclei intensity integral from 25 to 100 μm around wire implants is significantly less than that around the NC (p = 0.03). There Neuronal nuclei (NeuN)-labeled cells showed NeuN around are no statistical differences for GFAP intensity with wire the probe (figure 3). There is a reduction of neuronal density samples between four and eight weeks, but the labeling for toward the tissue–implant border for both implants. At four GFAP around the NC at eight weeks is significantly greater

6 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(a) (b)

Figure 4. Analysis of GFAP-reactive astrocytes as a function of distance from the tissue–implant border. (a) Relative intensity ± standard error as a function of distance from the border of four week NC (gray) and wire (black). The brackets indicate an integral of intensity. (b) Relative intensity ± standard error as a function of distance from border of eight week NC (gray) and wire (black). Representative images of GFAP for (c) four week NC; inset: 50 μm wide close-up of reactive astrocyte, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm. ∗ = p < 0.05.

(a) (b)

Figure 5. Analysis of CS56-CSPGs as a function of distance. (a) Relative intensity ± standard error as a function of distance from the border of four week NC (gray) and wire (black). The brackets indicate an interval of integral of intensity calculation. (b) Relative intensity ± standard error as a function of distance from the border of eight week NC (gray) and wire (black). Representative images of CS56 for (c) four week NC, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm.∗ = p < 0.05. over the 20–80 μm interval than at the four week time point from the tissue–implant border (figure 5(b), (e), (f)). CS56 (p = 0.03). Additionally, GFAP labeling tends to be more labeling at four weeks was significantly greater around the diffuse or less compact around the NC than that around the wire than that around the NC starting at 10 μm and extending wire at eight weeks (figures 4(b), (e), (f)). out to 150 μm(p < 0.01) (figures 5(a), (c), (d)). At eight weeks, the peaks and integrals were similar for the tissue 3.4. Proteoglycan around the NC and wire implants. Chondroitin sulfate proteoglycan (CSPG) is a structural 3.5. Vimentin component of ECM and its upregulation within the glial scar has been linked to inhibition of axonal outgrowth (Silver and Vimentin labeling is associated with immature astrocytes, Miller 2004). Intensity for CSPGs, via labeling for CS56 ECM components and developing axons (Alonso 2005, Pekny antibodies (table 1), peaks within 15 μm from the border 2003,Levinet al 2009). At both time points, vimentin (figures 5(a)–(f)). The peak around the NC implant returns expression is upregulated with a peak at 10 μmfromthe to baseline by 50 μm for the four and eight week samples implant (figures 6(a)–(f)) and decreases to baseline levels implanted with NC; levels remain elevated around the four by 100–150 μm from the implant border. At four weeks week wire implant up to 150 μm. The wire implants at eight (figures 6(a), (c), (d)), the vimentin intensity maximum and the weeks feature a response that returns to baseline by 100 μm 0–32 μm intensity integral for NC implants are significantly

7 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(a) (b)

Figure 6. Analysis of vimentin as a function of distance. (a) Relative intensity ± standard error as a function of distance from the border of four week NC (gray) and wire (black). The brackets indicate an integral of intensity. (b) Relative intensity ± standard error as a function of distance from the border of eight week NC (gray) and wire (black). Representative images of vimentin for (c) four week NC, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm.∗ = p < 0.05.

(a) (b)

Figure 7. Analysis of IBA1-microglia immunohistochemistry as a function of distance from the tissue–implant border. (a) Relative intensity ± standard error as a function of distance from the border of four week NC (gray) and wire (black). The brackets indicate an integral of intensity. (b) Relative intensity ± standard error as a function of distance from the border of eight week NC (gray) and wire (black). Representative images of IBA1 for (c) four week NC, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm. ∗ = p < 0.05. greater than for the wire implants (p < 0.01). Note that at decreased to near-background levels at more than 100 μm from 32 μm, there is an intersection of the two implant profiles. the border. ED1 labeling (figures 8(a)–(f)) exhibited similar At eight weeks (figures 6(b), (e), (f)), measures of vimentin features but was limited mainly to amoeboid cells close to the labeling are not significantly different between the NC and the implant interface. wire. In the interval of 10–150 μm from the tissue–implant The four and eight week IBA1 maximum intensity border, the intensity integral around the wire significantly (figures 7(a)–(f)) from the tissue around the NC is significantly decreased from four to eight weeks (p = 0.03). greater than the intensity from the tissue with wire implants at the same time point (p = 0.02, p < 0.01). The 3.6. Macrophages/microglia 0–150 μm intensity integral is significantly greater from the samples around the eight week NC than from those around Analysis of IBA1 and ED1 labeling quantified the density the eight week wire (figures 7(b), (e), (f)). Response changes of activated microglia and macrophages. IBA1 labels all for a particular implant type are not significant between four microglia (figures 7(a)–(f)) and ED1 labels (figures 8(a)–(f)) and eight weeks. activated microglia and activated macrophages. Microglia The peak intensities of ED1 at four and eight weeks as shown by IBA1 labeling appeared punctate or amoeboid are not significantly different between the NC and the wire near the border and show a ramified morphology away from (figures 8(a)–(f)), but the 17–150 μm intensity integral is the implant site. The microglia response was characterized significantly greater for wire implants compared to the NC at by intense IBA1 labeling 10 μm from the border that rapidly four weeks (p = 0.03). The 0–150 μm ED1 intensity integral

8 J. Neural Eng. 8 (2011) 066011 J P Harris et al

(a) (b)

Figure 8. Analysis of ED1-activated macrophage and microglia immunohistochemistry as a function of distance from the tissue–implant border. (a) Relative intensity ± standard error as a function of distance from the border of four week NC (gray) and wire (black). The brackets indicate an integral of intensity. (b) Relative intensity ± standard error as a function of distance from the border of eight week NC (gray) and wire (black). Representative images of vimentin for (c) four week NC, (d) four week wire, (e) eight week NC and (f) eight week wire. Scale bars 100 μm. ∗ = p < 0.05.

Table 2. Summary of IHC analysis when subdivided based on side. The number in the table represents the number of sections analyzed that showed a measurable difference in normalized stain intensity for one side or the other. The results are further divided into measures within the first 50 μm and the first 20 μm. These correspond to the distance related to unit recordings and approximately the first 3–4 cell layers, respectively. It does not appear that there is any particular bias in the results to one side or the other, or that surface roughness is contributing significantly to the response. Label DAPI ED1 Greater side Long side Short side No difference Long side Short side No difference

Within 50 μm 13 12 13 6 13 12 Within 20 μm1091971212 significantly decreases for the wire from four to eight weeks density than the wire implants at four weeks. These results (p = 0.04). The 0–50 μm ED1 intensity integral is significantly aligned with previous studies, which reported that the NeuN greater for the NC compared to wire implants at eight weeks density was reduced around the implant in conjunction with (p = 0.03) (figures 8(b), (e), (f)). similar upregulation of identical glial scar factors (Winslow and Tresco 2009,Biranet al 2005, Zhong and Bellamkonda 3.7. IHC intensity based on NC side 2007). In this work, we implanted NC probes and neat PVAc- To examine the effect of surface roughness on the IHC coated tungsten wires. To confirm that surface roughness and response, we examined DAPI and ED1 labeling across many morphology were not factors in the response, we analyzed the animals. Several different examinations of intensity were NC and neat PVAc. The examination via SEM showed that performed, but no examination yielded significant effects the neat polymer had a smooth surface (figure 1(e)), and the based on side. The complete numbers are shown in table 2. NC had two sides of different roughness. The pressed side Similar intensity between sides indicates that the difference of of the NC had a smooth surface with only minor indentations mean intensities between sides is less than the 95% confidence (figure 1(c)). The cut side of the NC had a rough surface interval. A sign test for the paired data was performed to (figure 1(d)). Under 50 000× magnification (figure 1(d)), determine if there was a statistically significant difference inset), the tunicate whiskers of the NC were clearly evident based on side. The results indicate that side does not have and protruded on the cut edge. a significant effect on the tissue response. Next, we examined whether differences in surface roughness were contributing to the tissue response. There 4. Discussion are two distinct sides with different surface roughness, and we investigated whether the side affected the tissue response. The results presented here support the hypothesis that The examination of the response based on side indicated that the mechanical mismatch between brain tissue and the response was similar between the two sides of the NC. microelectrodes influences the inflammatory response. These results support the fact that surface roughness did not Specifically, we found that the implant stiffness affects the affect the tissue response. From SEM images, it is clear that neuronal response and composition of the glial scar, as the the surface roughness differences between the wire and the NC-implanted animals had a significantly greater neuronal NC were much less than the differences between the sides

9 J. Neural Eng. 8 (2011) 066011 J P Harris et al of the NC. Collectively, the data suggest that our observed a significant difference over the 50–100 μm intensity integral differences in tissue response are not due to differences in (p = 0.05) at four weeks (figures 4(a), (c), (d)). Further, the surface roughness. Our findings agree with previous research peak intensity of GFAP expression was larger for the wire (p = that has shown that the surface roughness of an implant does 0.04), yet the distribution of GFAP expression was broader for not affect the tissue response past four weeks (Szarowski et al the NC at eight weeks (25–100 μm, p = 0.03, figures 4(b), (e), 2003). (f)). From four to eight weeks, the neuronal response (NeuN) To further examine the NC and the neat polymer, we and GFAP labeling around the NC implant were unchanged. investigated the surface chemistry via water contact angle. This suggests that the NC formed a static scar. Yet, with wire The NC and neat polymer had similar water contact angles. implants, the GFAP expression at eight weeks became more Since the two materials had a similar hydrophobicity, the data compact, and NeuN positive cells in the same region of tissue suggest that surface chemistry was not drastically modified appeared to increase around wire implants. The results from by the insertion of tunicate whiskers into the neat polymer. wire implants suggest that the scar is continuing to compact Therefore, the analysis of the NC and neat polymer suggests and remodel in response to the greater mechanical stiffness. that any effect of surface roughness and surface chemistry Cytoskeletal components play a major role in the ability should be contributing similarly, if at all, to the different tissue of a cell to interface with and respond to the extracellular responses to the materials. environment. The type and quantity of the filament are The tissue response in our results was similar to previous important in determining the ability of the cell to respond to the reports (Winslow et al 2010, Winslow and Tresco 2009). mechanical environment and move through the extracellular However, in our case, the classical markers of the glial scar space (Alberts 1994). In addition to GFAP, vimentin is (GFAP and ED1) do not fully explain the differences seen an intermediate filament expressed in astrocytes and was in the neuronal density around each of the implant types at investigated here (see Pekny (2003) for a review). In contrast four weeks. Though the softer NC implant had a significantly to GFAP, the peak intensity and the 0–32 μm intensity integral greater neuronal density around the probe than that around the for vimentin labeling were significantly greater at four weeks rigid wire after four weeks, there was no significant difference for the NC, compared to the wire. Historically, vimentin has in GFAP or ED1 labeling (peak or 0–50 μm interval). In the been tied to immature astrocytes, but recent research has linked 0–50 μm interval nearest to the device, considered critical for vimentin expression to rapid neurite extension in response unit recording, the brain tissue around the NC implants had to damage (Levin et al 2009). Likewise, NG2+ cells that nearly twice the neuronal density than that around the wire express vimentin have been proposed to support repair of implants at four weeks that may improve recording of single central nervous system damage and stabilize axons in response units from neurons located close to the implant at that time to dieback from ED1+ cells (Alonso 2005, Nishiyama 2007, period (Buzsaki´ 2004). Busch et al 2010). Here, the increase in vimentin labeling Examining the correspondence of GFAP and neural around four week implants was associated with more neurons density in our results, GFAP did not account for all the around NC implants than around wire implants (figures 6(a), differences seen between neuronal densities around the (c), (d), 3(a), (c), (d)). This is consistent with potential implant types. Specifically at four weeks, the GFAP peak neuroprotective implications for vimentin positive cells. Thus, intensity and the 0–22 μm intensity integrals were similar for we believe that the correlation is due to the positive effects both implant types, while there was a significant difference associated with the intermediate filament vimentin. At eight over the same range for neuronal density. This may be due weeks, the contrast in vimentin levels was no longer evident, to the wide assortment of roles, both positive and negative, which also correlates with similarity in neuronal densities associated with astrocytes. Astrocytes have been implicated around each implant type, further supporting the role of the in several different functions associated with the cytoskeletal intermediate filament, vimentin, in supporting the neuronal structure, including neurotransmitter maintenance and blood– density around the cortical implant. We believe that these brain barrier function (Simard et al 2003, Ortinski et al results suggest that the expression of intermediate filaments in 2010). GFAP positive cell functions can be neuroprotective, the tissue surrounding our implants is related to the mechanical as transgenic animals without GFAP show a larger functional properties of the electrode material. deficit after injury (Faulkner et al 2004). Detrimental to neural To further examine the role of mechanics in the wound protection, astrocytes have been implicated as a diffusion healing response, we examined another component of the barrier (Roitbak and Sykova 1999), remyelination limiter glial scar, proteoglycans. Proteoglycans are negatively (Fawcett and Asher 1999) and producer of cytokines and charged polysaccharide chains and are an important structural ECM components such as fibronectin (Shearer et al 2003) constituent of the ECM (Alberts 1994). When found in and CSPGs (McKeon et al 1999). Since astrocytes have elevated concentrations within tissue, the high density of been tied to many structural components including cytoskeletal negative charges found across their backbone has been shown arrangement and secretion of ECM components, it is believed to sequester cations, causing large amounts of water to diffuse that GFAP would be a key component in mechanical signaling. out of adjacent tissue into the proteoglycan-rich areas. This Therefore, it is notable that the rigid wire created a wider radius phenomenon creates a swelling pressure that enables the tissue of GFAP activation at four weeks but a more compact scar at to withstand compressive forces. A subclass of proteoglycans eight weeks. A wider radius is suggested by a nearly significant secreted by astrocytes, CSPG, has been shown to be deposited difference over the 22–150 μm intensity interval (p = 0.06) and by astrocytes in gradients of increasing concentration toward

10 J. Neural Eng. 8 (2011) 066011 J P Harris et al the injury site (Silver and Miller 2004). Interestingly, our Indicators of the glial scar components suggest that results indicated that the area with elevated CS56 labeling mechanics continues to modify the glial scar at eight weeks. (label for CSPGs,) is much larger around the wire implants Although the mechanical-associated labels, GFAP and CS56, than around the NC at four weeks (figures 5(a), (c), (d)). This no longer show a similar increased radius of activation around is in agreement with previous research demonstrating that wire implants at eight weeks, GFAP labeling (figures 4(b), CSPGs inhibit neuronal outgrowth (Silver and Miller 2004, (e), (f)) around wire implants exhibit a more compact, higher Busch and Silver 2007, Fawcett and Asher 1999), further peak than that around the compliant NC implants. Prior work supported by our four week NeuN staining (figures 3(a)– (Frampton et al 2010) hypothesizes that a more compact GFAP (d)). Additionally, in vitro experiments have shown increased response increases the impedance of an electrode, which may CSPG labeling in response to increased strain (Cullen et al decrease the quality of electrode recordings. Aligning with 2007), further supporting our hypothesis that the mechanical previous research (Fujita et al 1998, Nisbet et al 2009), both mismatch between brain tissue and microelectrodes influences implant types appear to lower ED1 activation from four to the response. eight weeks (figures 8(a)–(f)) suggesting an ongoing reaction While our results provide interesting insights into the NC, that lessens over time. The 0–50 μm intensity integral for ED1 reduced mechanical forces and improved neuronal densities, and IBA1 was significantly greater around NC implants than we cannot ignore the confounding effects of the inflammation that around the wire implants at eight weeks (figures 7(b), (e), mediating microglia cells. Microglia and macrophage cells (f) and 8(b), (e), (f)), but further work is needed to determine mediate the inflammatory and immune response to minimize whether the difference is neuroprotective, neurotoxic or neither bacterial/viral invaders (Kreutzberg 1996). Macrophages at longer time points. Regardless of the similar neural densities have also been implicated in axonal dieback (Horn et al at eight weeks, the NC continues to modify the overall tissue 2008). Many studies of the tissue response to intracortical response up to eight weeks. electrodes have focused on macrophage activation in response Another possible interpretation of the data is that softer to indwelling implants (Biran et al 2005). Activated implants are affecting the time-course of the response rather macrophages and microglia, as labeled by ED1, were similar than final results, and hence, similar neuronal densities at eight for peak intensity and the 0–17 μm (border to divergence weeks. It is known that the insertion event causes trauma, but it point) intensity integral at four weeks (figures 8(a), (c), (d)). is unknown how stiffness might affect the repair of this trauma. Similar to the GFAP and CSPG responses, there is a greater Previous research has shown that the response past four weeks radius of ED1 activation in the tissue around the wire implants is independent of size, shape and surface roughness (Szarowski at four weeks; the 17–150 μm intensity integral was greater et al 2003), but the softer NC may allow for quicker repair of for wire implants at four weeks. At four weeks, increased insertion damage. The increased levels of vimentin and IBA1 ED1 activation can be loosely associated with decreased at four weeks could indicate a more robust healing process. NeuN labeling over a similar distance from the electrode– Additionally, the similar neuronal densities between the NC tissue interface. At eight weeks, ED1 activity was less and the wire at eight weeks, coupled with the recovery of the pronounced around wire implants (figures 8(b), (e), (f)), which neuronal density around wire implants to NC levels, could corresponded to an apparent recovery in NeuN positive tissue. suggest a quicker repair around NC implants. In this paper, While ED1 labels only activated macrophages and our goal was to demonstrate that there is an effect based on microglia, IBA1 marks all microglia and macrophages. The probe stiffness. Further experiments are necessary to clearly four and eight weeks IBA1 maximum intensity (figures 7(a)– describe the time-course of the response and the specific effects (f)) from the tissue around the NC was significantly greater of probe mechanics on that time course. than the intensity from the tissue with wire implants at In summary, the four and eight week results show that the same time point (p = 0.02, p < 0.01). Differences the NC material modifies the glial scar and neuronal density between IBA1 and ED1 labeling can therefore be attributed around implants in comparison to stiffer wire implants. Four to the presence of either resting microglia or a subset of week NC implants feature greater neuronal densities than wire beneficial, ‘alternatively activated’ (M2), macrophages (Kigerl implants, and several glial scar labels such as GFAP,CSPG and et al 2009). The ED1 peak intensity was similar between ED1 show a larger radius of activation from the implant–tissue the two implant types at four weeks, but the IBA1 peak for border around wire implants than that around NC implants. the NC was 18% greater than that for the wire, indicating Notably GFAP and CSPGs, structural components, suggest the the presence of microglia cells independent of the levels of involvement of mechanics in the tissue response. Eight week activated cells (ED1) (figures 7(a), (c), (d), 8(a), (c), (d)). For results provide further evidence that the response to implants the NC implants, the presence of additional non-ED1 reactive is ongoing, dynamic and modified by the NC. Results show microglia and higher neuronal density at four weeks may that the NC can maintain its neuronal density for at least eight indicate a beneficial role of microglia. Regardless of the state weeks despite macrophage/microglia activation. of the macrophages/microglia, the results suggest an ongoing reaction for both implant types since activated macrophages 5. Conclusions should return to baseline 21 days after a stab injury to cortex (Fujita et al 1998, Nisbet et al 2009). Our results are consistent By utilizing our mechanically adaptive polymer NC materials, with the research showing a sustained response to indwelling it is possible to separate the foreign body response to the implants (Winslow et al 2010, McConnell et al 2009). implanted electrodes from the effect of material stiffness.

11 J. Neural Eng. 8 (2011) 066011 J P Harris et al While mechanical properties of the materials correlate with Capadona J R, Van Den Berg O, Capadona L A, Schroeter M, differences between neuronal densities at four weeks, we Rowan S J, Tyler D J and Weder C 2007 A versatile approach also demonstrate that the NC parallels the response to the for the processing of polymer nanocomposites with self-assembled nanofibre templates Nat. Nanotechnol. field-leading wire at eight weeks (Winslow and Tresco 2009, 2 765–9 Winslow et al 2010). Our data also suggest the importance of Clark K, Langeslag M, Figdor C G and van Leeuwen F N 2007 controlling vimentin and CSPG response to improve neuronal Myosin II and mechanotransduction: a balancing act Trends density. Through the use of the mechanically adaptive Cell Biol. 17 178–86 nanocomposite, we demonstrate a system that can decouple Cullen D K, Simon C M and LaPlaca M C 2007 Strain rate-dependent induction of reactive astrogliosis and cell death the mechanical and surface chemistry components of the in three-dimensional neuronal-astrocytic co-cultures Brain Res. neural response. We expect this to be a valuable tool 1158 103–5 in future research regarding cortical tissue. Our findings Engler A J, Sen S, Sweeney H L and Discher D E 2006 Matrix indicate two parallel tracks to capitalize on our work: (1) elasticity directs stem cell lineage specification Cell further development of advanced materials to modulate the 126 677–89 Faulkner J R, Herrmann J E, Woo M J, Tansey K E, Doan N B response to mechanical and molecular factors and (2) research and Sofroniew M V 2004 Reactive astrocytes protect tissue and into mechanisms and neuroprotective pathways, specifically preserve function after spinal cord injury J. Neurosci. M2 macrophage/microglia presence, CSPG reduction and 24 2143–55 vimentin upregulation. Fawcett J W and Asher R A 1999 The glial scar and central nervous system repair Brain Res. Bull. 49 377–91 Flanagan L A, Ju Y E, Marg B, Osterfield M and Janmey P A 2002 Acknowledgments Neurite branching on deformable substrates Neuroreport 13 2411–5 This work was supported by grant numbers R21- Frampton J P, Hynd M R, Shuler M L and Shain W 2010 Effects of NS053798 and F31-NS063640 from the National Institute glial cells on electrode impedance recorded from neural prosthetic devices in vitro Ann. Biomed. Eng. 38 1031–47 of Neurological Disorders and Stroke and T32-EB004314- Fujita T, Yoshimine T, Maruno M and Hayakawa T 1998 Cellular 06 for the National Institute of Biomedical Imaging and dynamics of macrophages and microglial cells in reaction to Bioengineering. Additional support was from Veteran’s stab wounds in rat cerebral cortex Acta Neurochir. 140 275–9 Affairs grants C3819C, F4827H and B6344W. The authors Georges P C, Miller W J, Meaney D F, Sawyer E S and Janmey P A acknowledge the support of Anne DeChant and Dr Horst von 2006 Matrices with compliance comparable to that of brain tissue select neuronal over glial growth in mixed cortical Recum. None of the funding sources aided in the collection, cultures Biophys. J. 90 3012–8 analysis and interpretation of data, in the writing of the report, Harris J, Hess A E, Rowan S, Weder C, Zorman C, Tyler D or in the decision to submit the paper for publication. The and Capadona J 2011 In vivo deployment of mechanically authors have no conflicts of interest related to this work to adaptive nanocomposites for intracortical microelectrodes disclose. J. Neural Eng. 8 046010 Horn K P, Busch S A, Hawthorne A L, van Rooijen N and Silver J 2008 Another barrier to regeneration in the CNS: activated References macrophages induce extensive retraction of dystrophic axons through direct physical interactions J. Neurosci. 28 9330–41 Alberts B 1994 Molecular Biology of the Cell 3rd edn ed B Alberts Ingber D E 2006 Cellular mechanotransduction: putting all the et al (New York: Garland) chapter 19, pp 948–1009 pieces together again Faseb J. 20 811–27 Alonso G 2005 NG2 proteoglycan-expressing cells of the adult rat Kigerl K A, Gensel J C, Ankeny D P, Alexander J K, Donnelly D J brain: possible involvement in the formation of glial scar and Popovich P G 2009 Identification of two distinct astrocytes following stab wound Glia 49 318–38 macrophage subsets with divergent effects causing either Biran R, Martin D C and Tresco P A 2005 Neuronal cell loss neurotoxicity or regeneration in the injured mouse spinal cord accompanies the brain tissue response to chronically implanted J. Neurosci. 29 13435–44 silicon microelectrode arrays Exp. Neurol. 195 115–26 Kim Y T, Hitchcock R W, Bridge M J and Tresco P A 2004 Chronic Biran R, Martin D C and Tresco P A 2007 The brain tissue response response of adult rat brain tissue to implants anchored to the to implanted silicon microelectrode arrays is increased when skull Biomaterials 25 2229–37 the device is tethered to the skull J. Biomed. Mater. Res. A Kippert A, Fitzner D, Helenius J and Simons M 2009 Actomyosin 82 169–78 contractility controls cell surface area of oligodendrocytes Busch S A, Horn K P, Cuascut F X, Hawthorne A L, Bai L, Miller R BMC Cell Biol. 10 71 H and Silver J 2010 Adult NG2+ cells are permissive to neurite Kreutzberg G W 1996 Microglia: a sensor for pathological events in outgrowth and stabilize sensory axons during the CNS Trends Neurosci. 19 312–8 macrophage-induced axonal dieback after spinal cord injury Lee H, Bellamkonda R V, Sun W and Levenston M E 2005 J. Neurosci. 30 255–65 Biomechanical analysis of silicon microelectrode-induced Busch S A and Silver J 2007 The role of extracellular matrix in CNS strain in the brain JNeuralEng.2 81–9 regeneration Curr. Opin. Neurobiol. 17 120–7 Lee K, He J, Clement R, Massia S and Kim B 2004 Biocompatible Buzsaki´ G 2004 Large-scale recording of neuronal ensembles Nat. benzocyclobutene (BCB)-based neural implants with Neurosci. 7 446–51 micro-fluidic channel Biosens. Bioelectron. 20 404–7 Capadona J R, Shanmuganathan K, Trittschuh S, Seidel S, Levin E C, Acharya N K, Sedeyn J C, Venkataraman V, D’Andrea Rowan S J and Weder C 2009 Polymer nanocomposites with M R, Wang H Y and Nagele R G 2009 Neuronal expression of nanowhiskers isolated from microcrystalline cellulose vimentin in the Alzheimer’s disease brain may be part of a Biomacromolecules 10 712–6 generalized dendritic damage-response mechanism Brain Res. Capadona J R, Shanmuganathan K, Tyler D J, Rowan S J and 1298 194–207 Weder C 2008 Stimuli-responsive polymer nanocomposites Lu Y, Wang D, Li T, Zhao X, Cao Y, Yang H and Duan Y Y 2009 inspired by the sea cucumber dermis Science 319 1370 Poly(vinyl alcohol)/poly(acrylic acid) hydrogel coatings for

12 J. Neural Eng. 8 (2011) 066011 J P Harris et al improving electrode-neural tissue interface Biomaterials Shanmuganathan K, Capadona J R, Rowan S J and Weder C 2010a 30 4143–51 Bio-inspired mechanically-adaptive nanocomposites derived McConnell G C, Rees H D, Levey A I, Gutekunst C A, Gross R E from cotton cellulose whiskers J. Mater. Chem. 20 180 and Bellamkonda R V 2009 Implanted neural electrodes cause Shanmuganathan K, Capadona J R, Rowan S J and Weder C 2010b chronic, local inflammation that is correlated with local Biomimetic mechanically adaptive nanocomposites Prog. neurodegeneration J. Neural Eng. 6 56003 Polym. Sci. 35 212–22 McConnell G C, Schneider T M, Owens D J and Bellamkonda R V Shearer M C, Niclou S P, Brown D, Asher R A, Holtmaat A J, 2007 Extraction force and cortical tissue reaction of silicon Levine J M, Verhaagen J and Fawcett J W 2003 The microelectrode arrays implanted in the rat brain IEEE Trans. astrocyte/meningeal cell interface is a barrier to neurite Biomed. Eng. 54 1097–107 outgrowth which can be overcome by manipulation of McKeon R J, Jurynec M J and Buck C R 1999 The chondroitin inhibitory molecules or axonal signalling pathways Mol. Cell. sulfate proteoglycans neurocan and phosphacan are expressed Neurosci. 24 913–25 by reactive astrocytes in the chronic CNS glial scar J. Neurosci. Silver J and Miller J H 2004 Regeneration beyond the glial scar Nat. 15 10778–88 Rev. 5 146–56 Nikles S A, Pellinen D S, Kitagawa J, Bradley R M, Kipke D R Simard M, Arcuino G, Takano T, Liu Q S and Nedergaard M 2003 and Najafi K 2003 Long term in vitro monitoring of polyimide Signaling at the gliovascular interface J. Neurosci. 23 9254–62 microprobe electrical properties Proc. 25th Annual Int. Conf. of Subbaroyan J 2007 PhD Thesis University of Michigan the IEEE Engineering in Medicine and Biology Society, Subbaroyan J, Martin D C and Kipke D R 2005 A finite-element 2003 4 3340–3 model of the mechanical effects of implantable microelectrodes Nisbet D R, Rodda A E, Horne M K, Forsythe J S and Finkelstein D in the cerebral cortex J. Neural Eng. 2 103–13 I 2009 Neurite infiltration and cellular response to electrospun Szarowski D H, Andersen M D, Retterer S, Spence A J, Isaacson M, polycaprolactone scaffolds implanted into the brain Craighead H G, Turner J N and Shain W 2003 Brain responses Biomaterials 30 4573–80 to micro-machined silicon devices Brain Res. 983 23–35 Nishiyama A 2007 Polydendrocytes: NG2 cells with many roles in Takeuchi S, Suzuki T, Mabuchi K and Fujita H 2004 3D flexible development and repair of the CNS Neuroscientist multichannel neural probe array J. Micromech. Microeng. 13 62–76 14 104–7 Ortinski P I, Dong J, Mungenast A, Yue C, Takano H, Watson D J, Takeuchi S, Ziegler D, Yoshida Y, Mabuchi K and Suzuki T 2005 Haydon P G and Coulter D A 2010 Selective induction of Parylene flexible neural probes integrated with microfluidic astrocytic gliosis generates deficits in neuronal inhibition Nat. channels Lab Chip 5 519–23 Neurosci. 13 584–91 Watson R E Jr, Wiegand S J, Clough R W and Hoffman G E 1986 Ostrow L W and Sachs F 2005 Mechanosensation and endothelin in Use of cryoprotectant to maintain long-term peptide astrocytes—hypothetical roles in CNS pathophysiology Brain immunoreactivity and tissue morphology Peptides 7 155–9 Res. Brain Res. Rev. 48 488–508 Wester B A, Lee R H and LaPlaca M C 2009 Development and Pekny M 2003 Glial Cell Function edBCLopez´ and M characterization of in vivo flexible electrodes compatible with Nieto-Sampedro (Amsterdam: Elsevier) chapter 3, pp 23–30 large tissue displacements J. Neural Eng. 6 024002 Pinheiro J, Bates D, DebRoy S, Sarkar D and the R Core team 2009 Winslow B D, Christensen M B, Yang W K, Solzbacher F nlme: Linear and Nonlinear Mixed Effects Models and Tresco P A 2010 A comparison of the tissue response to (version 3.1–96) chronically implanted parylene-C-coated and uncoated planar R Development Core Team 2009 R: A Language And Environment silicon microelectrode arrays in rat cortex Biomaterials For Statistical Computing (Vienna: R Foundation for 31 9163–72 Statistical Computing) Winslow B D and Tresco P A 2009 Quantitative analysis of the Roitbak T and Sykova E 1999 Diffusion barriers evoked in the rat tissue response to chronically implanted microwire electrodes cortex by reactive astrogliosis Glia 28 40–8 in rat cortex Biomaterials 31 1558–67 Shanmuganathan K, Capadona J R, Rowan S J and Weder C 2009 Zhong Y and Bellamkonda R V 2007 Dexamethasone-coated neural Stimuli-responsive mechanically adaptive polymer probes elicit attenuated inflammatory response and neuronal nanocomposites ACS Appl. Mater. Interfaces 2 165–74 loss compared to uncoated neural probes Brain Res. 1148 15–27