INVESTIGATION OF MATERIAL AND THERAPEUTIC STRATEGIES TO REDUCE THE INFLAMMATORY RESPONSE TO INTRACORTICAL IMPLANTS
by
JESSICA KIMBERLY NGUYEN
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy
Dissertation Advisor: Dr. Jeffrey R. Capadona
Department of Biomedical Engineering CASE WESTERN RESERVE UNIVERSITY
August 2015 CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES
We hereby approve the dissertation of Jessica Nguyen, candidate for the degree of Doctor of Philosophy*.
Jeffrey R. Capadona Committee Chair
Dustin J. Tyler Committee Member
Stuart J. Rowan Committee Member
Horst von Recum Committee Member
Nicholas P. Ziats Committee Member
4/29/2015 Date of Defense
*We also certify that written approval has been obtained for any proprietary material contained therein.
ii
To Mom and Dad, for always believing in me
iii Table of Contents List of Tables ...... x
List of Figures ...... xi
Acknowledgements ...... xv
Abstract ...... 1
Chapter 1 Specific Aims ...... 3
Aim 1: Characterize the neural tissue properties surrounding compliant implants ...... 4
Sub-Aim 1A: Determine the time course of neuroinflammatory events surrounding compliant neural implants ...... 5
Sub-Aim 1B: Quantify the strain profiles surrounding acute neural implants of varying compliance ...... 5
Aim 2: Investigate the utility of local antioxidant therapy from compliant implants on improving neuronal health ...... 6
Sub-Aim 2A: Characterize the effects of antioxidant-releasing mechanically compliant probes on the neuroinflammatory response ...... 7
Sub-Aim 2B: Develop a method to surface conjugate a synthetic antioxidant to neural implants ...... 7
Chapter 2 Introduction...... 9
2.1 Neural Recordings ...... 9
2.1.1 Intracortical Microelectrodes ...... 10
2.1.1.1 Types of Intracortical Microelectrodes: ...... 12
2.1.1.2 Neuronal Recording Signals: Single-Unit Activity (SUA), Multi-Unit Activity (MUA), and Local Field Potentials (LFP) ...... 15
2.2 Challenges to Intracortical Microelectrode Use ...... 17
2.2.1 Classes of Microelectrode Failure ...... 19
2.2.1.1 Biological Failure Mode ...... 20
2.2.1.2 Material Failure Mode ...... 21
2.2.1.3 Mechanical Failure Mode ...... 22
iv 2.3 The Neuroinflammatory Response ...... 23
2.3.1 The Acute Inflammatory Phase ...... 23
2.3.2 The Chronic Inflammatory Phase ...... 25
2.3.2.1 Microglia/Macrophages ...... 25
2.3.2.2 Astrocytes ...... 28
2.3.2.3 Neurodegeneration ...... 30
2.3.2.4 BBB Permeability ...... 31
2.3.2.5 Anti-inflammatory Strategies ...... 33
2.4 Strategies to Improve BBB integrity ...... 35
2.4.1 Effect of Microelectrode Compliance ...... 36
2.4.2 Strategies to Reduce Microelectrode-induced Tissue Strain ...... 39
2.4.2.1 Polymers ...... 39
2.4.2.2 Insertion Shuttles ...... 42
2.4.2.3 Mechanically-adaptive materials ...... 44
2.4.2.4 Architecture Modifications ...... 46
2.4.2.5 Wireless Systems ...... 48
2.4.3 Effect of Oxidative Stress ...... 50
2.4.3.1 Reactive Oxygen Species (ROS) ...... 50
2.4.3.2 Electrode Corrosion ...... 52
2.4.3.3 Strategies to Reduce Oxidative Stress ...... 55
2.5 Summary ...... 56
Chapter 3 Mechanically-Compliant Intracortical Implants Reduce the Neuroinflammatory Response ...... 58
3.1 Abstract ...... 58
3.2 Introduction ...... 59
3.2.1 Materials and Methods ...... 61
v 3.2.2 Intracortical Implants ...... 61
3.2.3 Strain Field Modeling ...... 62
3.2.4 Animal Surgery ...... 63
3.2.5 Tissue Processing ...... 65
3.2.6 Immunohistochemistry ...... 65
3.2.7 Quantitative Analysis ...... 67
3.2.8 Statistical Analysis ...... 67
3.3 Results ...... 68
3.3.1 Finite Element Analysis of Tissue Strain ...... 69
3.3.2 Neuronal Nuclei ...... 70
3.3.3 Glial Cell Markers ...... 72
Astrocytes ...... 72
Microglia/Macrophages ...... 74
Blood Brain Barrier Integrity (IgG) ...... 77
3.4 Discussion ...... 79
3.5 Conclusions ...... 90
3.6 Acknowledgements ...... 90
Chapter 4 Compliant Intracortical Implants Reduce Strains and Strain Rates in Brain Tissue In Vivo ...... 92
4.1 Abstract ...... 92
4.2 Introduction ...... 93
4.3 Materials and Methods ...... 96
4.3.1 Electrode fabrication & characterization ...... 96
4.3.2 Animal surgery and force measurements ...... 97
4.3.3 Force Measurement ...... 98
4.3.4 Estimation of stress and strain from force measurements ...... 101
vi 4.3.5 Analysis of Viscoelastic Parameters and Estimating Shear Moduli ...... 102
4.3.6 Analysis of stresses due to brain micromotion ...... 104
4.4 Results ...... 104
4.4.1 Comparison of forces exerted by non-compliant and compliant shanks at the brain-implant interface ...... 104
4.4.2 Coated, non-compliant and compliant shanks impose less strain at brain- implant interface ...... 106
4.4.3 Compliant shanks slow interfacial brain-tissue relaxation properties ...... 109
4.4.4 Coated, non-compliant and compliant shanks minimize mechanical stresses due to tissue micromotion at the interface ...... 112
4.5 Discussion ...... 113
4.6 Conclusion ...... 117
Chapter 5 The effect of antioxidant-releasing mechanically-adaptive implants on modulating the neural tissue response ...... 119
5.1 Abstract ...... 119
5.2 Introduction ...... 120
5.3 Methods...... 123
5.3.1 Chemicals and reagents ...... 123
5.3.2 Preparation of antioxidant-releasing nanocomposite films ...... 124
5.3.3 Mechanical characterization ...... 125
5.3.4 Finite Elemental Modeling (FEM) of antioxidant release ...... 125
5.3.5 In vitro antioxidant release ...... 126
5.3.6 Cell culture ...... 127
5.3.6.1 Glial cell viability ...... 128
5.3.6.2 BV-2 and NSC-34 co-culture: DHE and LIVE/DEAD® ...... 128
5.3.7 Measurement of antioxidant activity of films ...... 129
5.3.8 Animal surgery ...... 130
5.3.9 Tissue processing ...... 132
vii 5.3.10 Immunohistochemistry ...... 132
5.3.11 Quantitative analysis ...... 133
5.3.12 Statistical analysis ...... 134
5.4 Results ...... 135
5.4.1 Material Characterization ...... 135
5.4.2 Antioxidant release profile ...... 136
5.4.3 In vitro validation ...... 139
5.4.3.1 Glial Cell Cytotoxicity ...... 140
5.4.3.2 Antioxidant Activity ...... 142
5.4.3.3 Intracellular ROS and Neuron Cytotoxicity ...... 143
5.4.4 In vivo characterization ...... 144
5.4.4.1 Three Day Time Point ...... 145
5.4.4.2 Two Week Time Point ...... 147
5.4.4.3 Sixteen Week Time Point ...... 149
5.5 Discussion ...... 150
5.6 Conclusions ...... 154
5.7 Supplementary Data ...... 155
5.7.1 Materials and Methods ...... 155
5.7.1.1 Microscopy Studies...... 155
5.7.1.2 Swelling Behavior ...... 156
5.7.1.3 Thermogravimetric Analysis ...... 156
Chapter 6 Development of Superoxide Dismutase Mimetic Surfaces to Reduce Accumulation of Reactive Oxygen Species Surrounding Intracortical Microelectrodes ...... 161
6.1 Abstract ...... 161
6.2 Introduction ...... 162
6.3 Experimental Methods ...... 164
viii 6.3.1 Chemicals and Reagents ...... 164
6.3.2 MnTBAP Substrate Modification ...... 164
6.3.3 Surface Characterization ...... 167
6.3.3.1 Contact Angle Measurement: Determination of Surface Hydrophobicity of Substrates ...... 167
6.3.3.2 X-ray Photoelectron Spectroscopy (XPS): Determination of Atomic Composition of Substrates ...... 168
6.3.3.3 Quartz Crystal Microbalance: Determination of MnTBAP Surface Density ...... 168
6.3.4 Activity of MnTBAP-modified Surfaces ...... 169
6.3.5 Activity of MnTBAP in Solution ...... 170
6.3.6 Intracortical Microelectrode Implantation and Resveratrol Measurement .. 171
6.3.7 In vitro Assessment ...... 172
6.3.7.1 Intracellular superoxide anion accumulation ...... 173
6.3.7.2 Soluble nitric oxide and superoxide anion ...... 173
6.3.7.3 Live/Dead Assay ...... 174
6.3.8 Statistical Analysis ...... 175
6.4 Results and Discussion ...... 175
6.4.1 Design of Synthetic Anti-oxidative Modified Substrates ...... 175
6.4.2 Anti-oxidative Activity of MnTBAP-modified Surfaces ...... 178
6.4.3 Effect of MnTBAP-modified Surfaces on Reactive Oxygen Species and Nitric Oxide Accumulation and Release ...... 183
6.4.4 Effect of MnTBAP modified surfaces on cellular viability ...... 187
6.5 Conclusions ...... 190
6.6 Acknowledgements ...... 191
Chapter 7 Conclusions and Future Directions ...... 192
Appendix: Protocols...... 205
References ...... 214
ix List of Tables
Table 1. Strategies to reduce microelectrode compliance with various types of polymers.
...... 39
Table 2. Contact surface areas of Si-Bare, PVAc, and NC based on SolidWorks CAD models ...... 103
Table 3. Storage moduli (E') of dry and ACSF-swollen NCs determined by DMA...... 136
Table 4. Estimated amount of antioxidant contained within an in vivo implant...... 138
Table 5. Contact Angle Measurements on Modified Surfaces...... 177
Table 6. XPS Analysis on Modified Surfaces ...... 177
x List of Figures
Figure 1. Examples of recording neural electrodes for brain machine interface devices ...... 11
Figure 2. Types of traditional intracortical microelectrodes for neural recordings...... 12
Figure 3 Summary of recording performance for animals which had isolated units on at least two recording sessions...... 18
Figure 4. Failure modes for intracortical microelectrodes ...... 20
Figure 5 Activity states of microglia...... 26
Figure 6. Self-perpetuating neuroinflammatory pathways in response to microelectrode implantation ...... 32
Figure 7. Finite element analysis of tissue strain surrounding tethered microelectrodes ...... 38
Figure 8. Second generation mechanically-adaptive nanocomposite (NC) ...... 45
Figure 9. Examples of microelectrodes with architectural modifications ...... 47
Figure 10. The level of oxidative stress is dictated by the ROS levels ...... 51
Figure 11. Example SEMs of microelectrodes pre- and post-implant ...... 53
Figure 12. Images of Michigan-style silicon probes before (A) and after (B) PVAc coating...... 62
Figure 13. Predicted strain profiles induced by a tangential tethering force ...... 70
Figure 14. Immunohistochemical analysis of neuronal nuclei (NeuN) around the implant site ...... 72
Figure 15. Immunohistochemical analysis of the astrocytic scar...... 73
xi Figure 16. Immunohistochemical analysis of total microglia/macrophages populations (Iba1) ...... 75
Figure 17. Immunohistochemical analysis of CD68, a cellular marker for activated microglia/macrophages...... 77
Figure 18. Immunohistochemical analysis of IgG staining shows that the blood brain barrier integrity is improved for compliant implants, compared to stiff implants ...... 79
Figure 19. Schematic representation of the tissue response around stiff (A) and compliant (B) implants ...... 89
Figure 20. The experimental setup to measure force-displacement curves in brain tissue in vivo is illustrated ...... 97
Figure 21. Compliant nanocomposite shanks withstand indentation test ...... 100
Figure 22. For estimation of stress from complex geometries, CAD models in Solidworks™ were used to derive contact surface areas with surrounding brain tissue for stress calculations for each type of shank ...... 102
Figure 23. Average force-displacement curves (black) ± standard deviation (gray)in (A) & (D) for bare silicon (non-compliant) shanks in phases 1 and 2 respectively (n=4 shanks), (B) & (E) for PVAc-coated silicon (coated, non-compliant) shanks in phases 1 and 2 respectively(n=3 shanks), and (C) & (F) for nanocomposite (compliant) shanks in phases 1 (n=6 shanks) and 2 respectively (n=5 shanks)...... 105
Figure 24. (A) Dynamic stresses for non-compliant Si-bare (n=4 measurements), coated, non-compliant PVAc-coated Si (n=3 measurements), and compliant nanocomposite microelectrodes post-swelling (n=6 measurements) in phase 2 as the microelectrodes were moved from 1.0 to 1.5 mm are shown as a function of tissue depth...... 108
Figure 25. Representative relaxation stresses for non-compliant bare silicon (A,D), PVAc- coated non-compliant silicon (B,E), and compliant nanocomposite (C,F) microelectrodes...... 110
xii Figure 26. A viscoelastic material model was derived from the relaxation stresses such as those in Fig.6 ...... 111
Figure 27. Micromotion induced stress (y-axis on the left side) and strain (y-axis on the right side) amplitudes were pooled for non-compliant, bare silicon (n=4), PVAc-coated non-compliant silicon (n=3), and compliant nanocomposite (n=6) shanks from steady steady stresses at 1.0 mm and 1.5 mm tissue depth ...... 113
Figure 28. FEM estimation of in vivo drug release ...... 137
Figure 29. Normalized drug release percentage of curcumin (Cur)-loaded NC and resveratrol (Res)-loaded NC in ACSF at 37 °C ...... 139
Figure 30. Cytotoxicity of antioxidant-releasing NC with microglia cells ...... 141
Figure 31. Antioxidative activity of 0.005% and 0.01% antioxidant-releasing NC ...... 142
Figure 32. In vitro characterization of NC films containing 0.01% Res and Cur ...... 144
Figure 33. Assessment of the initial inflammatory response at three days post-implantation ...... 146
Figure 34. Assessment of the inflammatory response at two weeks post-implantation ...... 148
Figure 35. Assessment of the inflammatory response at sixteen weeks post-implantation...... 150
Figure 36. Reaction scheme for synthesis of MnTBAP modified glass surfaces...... 166
Figure 37. Activity of MnTBAP composite surfaces over time ...... 179
Figure 38. Bio-distribution of resveratrol around implanted microelectrodes up to 48 hours after administration ...... 181
Figure 39. Accumulation of intracellular superoxide anion (dihydroethidium; DHE) from activated microglia cells (BV2s) 48 hours after seeding onto modified surfaces ...... 184
xiii Figure 40. Release of reactive oxygen species (ROS) from activated microglia cells (BV2s) 48 hours after seeding onto surfaces ...... 185
Figure 41. Cytotoxicity of modified surfaces on activated microglia cells (BV2s) after 48 hours ...... 188
Figure 42. Proposed mechanism of action for MnTBAP composite surfaces ...... 189
Figure 43. In vivo implantation of SOD-coated Michigan-style silicon implants ...... 197
Figure 44. In vitro characterization and validation of SOD-coated NC surfaces ...... 198
xiv Acknowledgements
I would not have been able to get through the last five years without the support,
help, and kindness of countless wonderful people. I cannot thank you all enough for
keeping me sane, making me laugh, and reminding what this is all for.
First, I would like to thank my advisor Jeff for his unwavering support. I came to
Case to work for him and not once did I regret my decision. Thank you for always believing
in me. Whenever I ran into your office with nerves or worries, you never failed to talk me
off the ledge and tell me that you had no doubt I’d do great. You taught me how to be a
better scientist, but also the value of kindness and patience as a mentor. Through
everything, you always were looking out for my best interest in my career and life. I feel
lucky to have you as an advisor and look forward to our continued friendship.
Thank you also to my dissertation committee for all their guidance: Dustin Tyler,
Stuart Rowan, Nicholas Ziats, and Horst von Recum. Your valuable insights, comments
and advice shaped my research and further developed my scientific and professional skills.
Thank you so much for challenging me to think like a PhD—to be critical, push the limits
of my knowledge, and think outside the box.
Additionally, I would like to thank Stephen Selkirk and the Miller lab, with whom
I spent the first two years of my PhD. Steve Selkirk, thank you for making my transition to
graduate school easier. I really appreciate your persistent guidance through a very
challenging project. Thank you to Jen Dunger for all her help with my projects and teaching
me protocols. Mary Petrulis and Anita Zaremba, it was a blast sharing a lab with you. I always looked forward to our daily chats and sanity breaks. To the rest of the Miller lab,
xv thank you for sharing your space and helping me learn everything I needed to about neuroscience.
A huge thank you my Capadona lab mates, past and present, for putting up with me all these years. Kelsey Potter, thank you for being so welcoming, friendly, and patient— you taught me everything I know! I really enjoyed our desk chats and you showed me what it’s like to have a real passion for science. To my lab husband, John Hermann (aka Hermie) thank you for constantly making me laugh (with your dad puns or random television references) even when things got crazy or stressful. Kyle Kovach, thank you for being a reliable and considerate co-worker, always being happy to help no matter what. Evon
Ereifej, thank you for being someone I can talk to about lab and life. I’m glad we got this past year to hang out. Hillary Bedell, it’s been fun working with you this past year, bumping into you at Starbucks, going out for trivia nights and happy hours. Jen Keene, thank you for your help with the SOD project and all the scientific discussions. To my mentee and friend, Priya Srivastava, you inspire me by your strength and humility. Thank you for always being a positive person. And to my incredible undergrads, without you all the work here would not have been possible. Kelly Buchanan, you are awesome. As my first undergrad, I could not have asked for a more responsible, kind, intelligent student. Thank you for all your immeasurable hard work, positive attitude, and always being down for whatever I needed help with—even on weekends! Dan Park, thank you for being my speed demon. If I needed something done well and quickly, I could always rely on you. Thank you for putting up with having to re-analyze the data a million different ways, and always having a good sense of humor about it too. Martin Gitomer and Colin Uthe, thank you for all your help with surgeries, SOD optimizations, and my last set of experiments.
xvi To all the friends I’ve made during my time here at Case. I would not have survived
Cleveland without any of you. Phuong Dang, I can’t even begin to put into words how
happy I am we survived this together, and I could not imagine going through it with anyone
else. Despite you tricking me into coming here (“Winters aren’t so bad!”), I’m grateful for
everything we’ve gone through these years together: our random nights at ghetto bell, mahjong game nights, complaining about the cold, road trips, long talks/venting sessions about graduate school. Where are we going next??? Jenny Bastijanic, my first Case friend, who I remember vividly meeting on the bus during Open House week. I remember thinking
“This girl is cool, we should be friends.” Thank you for always being a listening ear; I definitely would not have been able to get through this crazy ride without our talks and knowing that we were going through all the emotions together. And thank you for making me an unofficial member of the Bastijanic family Julia Samorezov, thank you for being a work buddy, motivating me and also giving me much needed chat breaks. I will also never cease to be amazed by your eating! Sonia Merritt, thank you for always giving me an honest and valuable point of view. Punkaj Ahuja, thank you for being a great friend and
always feeding me delicious food. Thank you to all the NEC faculty and students for
scientific discussions, their support getting through NEC talks, and camaraderie in the bull
pen. To the rest of my BME grad friends, even though I am not naming each of you
individually, please know that I am so grateful for all the fun times that gave me a much
needed break from the stress of graduate school.
Last but not least, a huge thank you to my family and friends back home. Without
all of you, I would not be where I am today. To all my uncles, aunts, cousins, nieces,
nephews thank you for always reminding me how supportive and proud you are of me.
xvii Coming home to your enthusiasm and kind words always reminded me that I could get
through anything because I have the support of my amazing family. To all of my SV
friends, high school friends, UCLA friends, despite being so far away, you’ve always made
me feel like I’m missed and welcomed with open arms when I come home. Thank you for
always checking in on me and reminding me that there’s more to life than work. I’d like to
specifically thank Christina Lui, Nina Hoang, Frank Garcia, Samantha Delamere, Christine
Cardema, and Chia-wen Chang for actually taking a trip to Cleveland to visit me! To my
step-dad Tom, thank you for treating me like your own and always making sure I’m taken
care of, from making sure I’m eating to welcoming me into your family. To my sister
Vanessa and my new brother-in-law Erik, thank you for all your support and kind words.
Most importantly, thank you to my amazingly supportive parents. I do what I do to
make you proud. To my dad, thank you for instilling in me the importance of education
and hard-work. Your thirst for knowledge has always inspired me to be curious and open to learning new things. I would not be here without all the life skills you taught me growing
up and continue to teach me to this day. To my mom, thank you for being the most loving, caring and considerate mom I could ask for. Through everything I’ve gone through, you have always been my rock. I aspire to be as kind, generous, and strong as you. I would not be here without your unconditional support and love.
xviii Investigation of Material and Therapeutic Strategies to Reduce the Inflammatory Response to Intracortical Implants
by JESSICA KIMBERLY NGUYEN
Abstract
Intracortical microelectrodes have provided researchers with a tool to understand
the neural system and restore motor control in patients with tetraplegia. However,
recording instability and variability throughout implantation lifetime have limited the
potential of microelectrodes. The biological response to implanted microelectrodes is a
major contributor to device failure. The overall goal of this work was to explore material and therapeutic strategies to reduce the inflammatory response to intracortical implants. In this dissertation, we investigated the effect of material compliance on the neuroinflammatory response. Use of soft modulus, compliant materials that more closely match the brain modulus significantly reduced the neuroinflammatory response compared to traditional, stiff materials. Additionally, we report on acute force measurements of tissue strain from implanted microelectrodes. Our results indicated that compliant implants reduced the dynamic stress relaxation rate of the brain tissue and micromotion-induced tissue stresses compared to stiff implants. To further modulate the neuroinflammatory response, we also report on antioxidant-releasing compliant implants. The neuroinflammatory response is complex and adaptive, and we hypothesized that coupling of compliant materials and antioxidant therapy may further improve microelectrode integration. Antioxidant release from compliant materials demonstrated the short-term
1 benefits of antioxidant treatment and long-term reduction in neuroinflammation with compliant materials. Furthermore, we developed a method to chemically-conjugate an antioxidant mimetic on the microelectrode surface to create a more sustained antioxidative effect. The results of this work provide support for the use of compliant materials and antioxidant therapies to improve microelectrode integration. Further, the work here additionally substantiates that combining multiple strategies may better enhance modulation of the neuroinflammatory response and improve reliability of intracortical microelectrodes.
2 Chapter 1
Specific Aims
The goals of the work in this dissertation are to (1) investigate the use of compliant
materials to reduce the neuroinflammatory response to neural implants and (2) determine
if a combined strategy utilizing antioxidant therapy and mechanically compliant materials
can improve the proximity of neurons surrounding intracortical microelectrodes.
Intracortical microelectrodes create an intimate connection between neural tissue and external devices to restore function in patients with movement disorders.[1] However, wide clinical use is limited, in large part, by unreliability and instability of electrodes to record over long periods of time.[2] The leading cause of microelectrode dysfunction is
still being debated. However, there is agreement that proximity of neuronal cell bodies
and the characteristics of the surrounding tissue play an important role in microelectrode
function.[2, 3]
The mechanical mismatch between microelectrode materials and brain tissue has
been a long-debated hypothesis with little supporting evidence. The inherent stiffness of
traditional intracortical microelectrodes coupled with brain micromotion has been
suggested to induce strain on the surrounding tissue and perpetuate continual blood brain
barrier breach. Since Goldstein and Salcman introduced the issue of mechanical mismatch
in 1973, in silico studies have supported the hypothesis that micromotion-induced tissue
strain is increased around stiff implants.[4, 5] Additionally, Muthuswamy et al. reported
on quantification of micromotion-induced tissue stresses surrounding microelectrodes,
which indicated remodeling of the tissue weeks after implantation.[6] Many groups have
3 utilized approaches such as modifying implant architecture, bioactive coatings, and
alternative materials to improve mechanical integrity but have only reported short-term
improvements in neuroinflammation and recording quality.[7-10]
More recently, literature has supported the hypothesis that the integrity of the
blood-brain barrier (BBB) may be a major factor in the neuroinflammatory response to
microelectrodes, which was correlated with the long-term stability of functional
recordings.[11] When examining the tissue response surrounding implanted microwire
and Michigan-style microelectrodes, the Tresco group found that the local BBB integrity
immediately surrounding the implantation site was compromised.[12, 13] Our lab has
previously verified that the administration of antioxidants is able to improve BBB integrity
and reduce neuroinflammation at acute time points.[14, 15]
Given evidence that mechanics and BBB integrity can influence microelectrode
success, targeting multiple components of the inflammatory response may be a better
approach to reduce neuroinflammation and improve microelectrode function throughout
the entire time of implantation. Therefore, we hypothesize that an approach combining
local antioxidant delivery from mechanically compliant microelectrodes can additively
reduce neuroinflammation at the brain-electrode interface and potentially enhance
functional neural recordings. To this end, this dissertation explores the following aims:
Aim 1: Characterize the neural tissue properties surrounding compliant implants
Previous efforts to investigate the role of mechanical mismatch in neural implant integration have focused predominantly on in silico and in vitro experiments. Studies utilizing soft intracortical implants still involved polymer materials that are 2-3 orders of
4 magnitude stiffer than brain tissue, as implantation of softer materials proved difficult.[7,
16-18] With this in mind, our group of researchers at Case Western Reserve University developed a mechanically-adaptive material to explore the effects of material compliance
on the in vivo neuroinflammatory response and single unit recordings.[19-22] Our
materials soften from 5 GPa to 12 MPa upon in vivo implantation—the stiff state allowing
for easy insertion and the soft state reducing tissue strain.[23]
Sub-Aim 1A: Determine the time course of neuroinflammatory events surrounding
compliant neural implants
Preliminary in vivo experiments suggested an initial advantage for compliant
implants at four weeks, but saw no differences at eight weeks post-implantation.[20] Our
lab previously reported on a biphasic response to intracortical Michigan-style
microelectrodes, suggesting the presence of a secondary neurodegenerative state at sixteen
weeks post-implantation.[24] Therefore, in this sub-aim, we implanted non-functional
compliant implants for various time points, up to sixteen weeks, that exhibit
neurodegenerative effects to compare with traditional, stiff silicon implants (Chapter
3).[19]
Sub-Aim 1B: Quantify the strain profiles surrounding acute neural implants of varying
compliance
While researchers have acknowledged the effects of microelectrode-induced strain
on surrounding tissue, to date there has been no evidence measuring the actual strain
induced on the tissue from neural implants. Studies on strain have predominantly focused
5 on modeling studies, in vitro studies, or ex vivo slice cultures.[4, 5, 25-29] In order to quantify the strain in vivo, Muthuswamy et al. developed a method to measure the mechanical properties of the tissue surrounding stainless steel microwires in rat brain tissue.[6] The data indicated changes in the micromotion-induced stresses and remodeling of the tissue surrounding the implants over time.
Utilizing the method developed by Muthuswamy et al., we endeavored to compare the mechanical properties of tissue surrounding our compliant neural probes and traditional stiff probes, and investigate the strain profile at the brain-electrode interface. In this sub- aim, we report on acute measurements of compliant nanocomposite and stiff Michigan-
style silicon probes to quantify strain, strain-rate, and micromotion induced stresses in vivo
(Chapter 4).[30]
Aim 2: Investigate the utility of local antioxidant therapy from compliant implants on
improving neuronal health
Aim 1 in this dissertation explored the efficacy of compliant materials on
neuroinflammation. However, the neuroinflammatory response is multi-faceted and
complex, and a single solution may not be sufficient to adequately mediate the damaging
elements. Material compliance and BBB integrity have been widely regarded as major
influencers of neuroinflammation. In order to combat two major inflammatory mediators,
this aim utilized an additive approach involving soft material implants and antioxidant
delivery to reduce BBB breakdown. Here, we investigated methods to facilitate local
antioxidant release from compliant implants and circumvent potential adverse effects of
systemic therapeutic administration.
6 Sub-Aim 2A: Characterize the effects of antioxidant-releasing mechanically compliant
probes on the neuroinflammatory response
Natural antioxidants, resveratrol and curcumin, have been widely studied for their
anti-cancer and anti-inflammatory properties. Our lab has specifically demonstrated
efficacy of resveratrol and curcumin in reducing acute neuroinflammatory events
surrounding intracortical implants.[14, 15] Here, we determined proper doping concentration to locally release antioxidants from our compliant nanocomposite materials.
Using the chosen antioxidant concentration, we examined the effects of antioxidant- releasing mechanically compliant probes on the neuroinflammatory response in vivo
(Chapter 5).
Sub-Aim 2B: Develop a method to surface conjugate a synthetic antioxidant to neural
implants
While administration of therapeutics has been promising in reducing
neuroinflammation, the effects are not long-lasting and can often lead to undesirable side-
effects. A class of synthetic antioxidants that act as superoxide dismutase mimetics has
been validated as an effective superoxide scavenger during oxidative stress events.[31-33]
Specifically, Mn(III)tetrakis(4-benzoic acid)porphyrin (MnTBAP) can reduce reactive
oxygen species stably and repeatedly in biological conditions.[34, 35] In this sub-aim, we
developed a method to chemically-conjugate MnTBAP to the microelectrode surface to provide a long-term approach to scavenge free radicals.[36] Then, we characterized and
validated our surface modification and evaluated the antioxidant activity in vitro with
microglia and neuron cell cultures (Chapter 6). Additionally, the surface conjugation
7 method of MnTBAP was applied to our mechanically-compliant nanocomposite to explore the potential benefits of a dual system (Chapter 7).
8 Chapter 2
Introduction
2.1 Neural Recordings
Since the discovery of electrical activity in the nervous system in the 1790s, the
inner workings of the human nervous system has been an exciting source of interest.[37]
Over the next hundred years, neuroscience was revolutionized as researchers developed
techniques to record electrical signals generated by neurons, discovered the mechanism of
action potentials, and attempted to decode meaning from neural signals. Recording of
action potentials, the basic electrical unit derived from a single neuron, has permitted
intimate understanding of complex neuron activities and invaluable insight into the human
condition.
Information from central nervous system (CNS) recordings has been used for a
variety of applications, such as basic science knowledge, disease diagnosis, and
rehabilitation. Electrophysiology studies measure currents in biological systems to
understand a wide array of phenomena, from single ion channel proteins to whole organs.
Understanding of ion channel function has provided mechanisms for cellular
communications, normal body functions, and especially disease states. Neuron recordings
have also provided information in regards to cell-cell connections and how neurons network with each other and other types of cells. Moving into a larger scale, researchers have utilized cortical mapping of the brain to understand connectivity of the billions of neurons within the brain.[38, 39] Studies have isolated precise regions of the brain to be important for specific functions such as speech, touch, learning, and memory.
9 Additionally, recording techniques can provide useful information on neurological disease
states. Electroencephalography (EEG) recordings have been used in diagnosis of disease
states such as epilepsy, stroke, and cancer. More recently, neural recordings have been used to recover functional movement in patients with movement disorders. Brain machine interfaces (BMI) record neural signals and relay these signals to an external device to decode intent and perform an output response.[40]
2.1.1 Intracortical Microelectrodes
Electrodes for CNS recording come in a variety of classes with increasing invasivity (Figure 1) . As the electrode invasivity increases, the resolution of neural signals that can be recorded also increases.[41, 42] The type of electrode used is determined by desired output and recording resolution. For instance, recording of large populations of neurons can be sufficiently done with less invasive methods, such as with electroencephalography (EEG) electrodes that are placed directly on the scalp. More invasive electrocorticography (ECoG) electrodes are placed either above (epidural) or below (subdural) the dura mater surrounding the brain to record neural activity at the cortical surface. However, EEG and ECoG mainly sample from the superficial layers of the cortex. More invasive electrodes can explore deeper locations and provide better spatial selectivity. Single cell resolution is often desired to understand very specific neuronal properties or pathways, requiring the use of invasive intracortical recording electrodes.[43]
10
Figure 1. Examples of recording neural electrodes for brain machine interface devices. (A) EEG activity is recorded non-invasively with electrodes placed on the scalp. (B) ECoG electrodes are placed either outside the dura mater (epidural ECoG) or under the dura mater (subdural ECoG) and can record neural activity on the cortical surface. (C) Intracortical microelectrodes penetrate the cortex and can record action potentials from individual or small populations of neurons within the cortex. Image and caption published with permission from [41].
Intracortical microelectrodes have the benefits of spatio-temporal resolution,
neuron selectivity, high SNR, and increased data transfer rate.[44] Smaller electrode sizes
detect from a smaller area to allow for spatial selectivity. By implanting directly into the
cortex, the output signal does not get filtered from traveling through biological tissue,
which increases the signal amplitude. The information transfer rate for single unit
recordings can also be much higher than EEG and ECoG recordings. Researchers have
suggested that in order to control more complex systems or perform very intricate
movements, a higher rate of information transfer is needed than the non-invasive methods
can provide.[45, 46]
11 2.1.1.1 Types of Intracortical Microelectrodes:
Intracortical microelectrodes vary predominantly in electrode material, electrode
placement, electrode size, shank size, and number of shanks, and. While they can be
fabricated into various styles, the most commonly used will be discussed here (Figure 2).
Glass micropipettes are traditionally used in basic research to observe single ion channel
function and changes to membrane potentials because they allow for very sensitive
intracellular, low frequency recordings (not shown in Figure 2D). However, for
extracellular recordings, polarized metal electrodes better handle high frequency responses
with lower noise, but requires sacrificing input impedance and drift.[47, 48]
Figure 2. Types of traditional intracortical microelectrodes for neural recordings. Microwires (A) and Michigan-style (B) electrodes can be fabricated as a single shank or array. Utah electrode arrays (UEA) (C) are available in uniform or slanted architecture for detection at different depths. Cone (glass micropipette) electrodes (D) are utilized mostly for intracellular recordings. Image reprinted with permission from [41].
12 One of the first types of metal microelectrodes used is the insulated microwire.
Typically a cylindrical wire of stainless steel, platinum-iridium or tungsten, electrode sites
are at the tip of probe.[49] Electrode sizes typically range from 20-500 µm in diameter
with an impedance between 0.5-20 MΩ at 1 kHz. Smaller electrode sizes record single
unit activity better but have higher impedance, while larger sites better record field
potentials.[50, 51] Microwires can be fabricated as a single insulated wire or in a two or
three-dimensional array.
A majority of studies continue to use microwire electrodes (Figure 2A) due to ease
of fabrication and low price. [52-57] Of note, the Nicolelis group works almost exclusively
with microwire in BCI studies with non-human primates and humans, both acutely and chronically.[54, 58-60] Corrosion and impedance changes with microwire electrodes have been more thoroughly studied than with other electrode types.[57, 61-63] Microwire also may produce a lower tissue response compared to other types of microelectrodes, attributed to the cylindrical shape and relatively small size of the electrodes.[64] However, a major drawback to microwire arrays is difficulty of reproducible fabrication into dense arrays.
Additionally, they still exhibit recording instability and failure over time similar to other electrode types.[2, 8, 9, 61, 64]
Michigan arrays are planar, silicon arrays with electrode sites along the shank of the probe (Figure 2B).[65, 66] Placing electrode sites along the shank allows for a higher density of metalized electrode sites (typically iridium, platinum, or gold) and increased spatial resolution, by recording at different depths within the tissue. [67] Fabrication of
Michigan-style arrays requires access to expensive fabrication equipment. However,
NeuroNexus®, a subsidiary of GreatBatch Inc.®, provides commercial custom-fabrication
13 of Michigan-style arrays, making Michigan arrays a popular microelectrode choice.
Studies have incorporated a variety of specialty modifications to the original design for
increased recording performance, including conducting polymers, drug-release coatings,
microfluidics, and optogenetics.[68] The Kipke group has reported on long-term recording with Michigan arrays, with the longest array lasting up to 127 days.[67, 69] However, the vast majority of electrodes failed prior than two months. Unfortunately, Michigan-style arrays appear to have increased tissue response and chronic instability in comparison to other electrode types, potentially due to electrode architecture.[8, 70] To date, Michigan-
style electrodes have not been utilized in human studies.
Utah electrode arrays (UEA) are typically 10x10 arrays made of silicon with electrodes at the tip, with uniform probe length or a slanted architecture (Figure 2C). Utah
electrode arrays have the most clinical success and are the only arrays currently used in
long-term human studies. Developed by Normann and colleagues, UEAs were initially
intended for stimulation applications.[71, 72] The first studies with recording UEAs
demonstrated chronic visual and auditory recording in cats.[73-76] Subsequent BMI
studies in primates showed promise in various control paradigms, such as cursor control
and arm movements.[77, 78] Most notably, as part of the BrainGate project, Hochberg and
colleagues have successfully implanted UEAs into the primary cortex of human
subjects.[79-83] Excitingly, the team was able to demonstrate point-and-click cursor
control and robotic arm control in patients with tetraplegia. While the success in humans
is very promising, UEAs still suffer from recording instability, with researchers commonly
reporting >60% loss of function over a six-month period.[76] Also, due to the architecture and weight of the implants, UEAs must be inserted pneumatically to reduce tissue damage
14 from dimpling if inserted too slowly.[84] Customization of the architecture is also more
difficult; therefore incorporation of other technologies or sensors (which can be done with
Michigan arrays) is seldom done. Histological studies have indicated significant fibrous
encapsulation and “settling” or sinking within the tissue, causing tissue indentation.[76,
85] The electrode “settling” may affect the final resting position of the recording tips,
causing them to be deeper than the initial recording position. Additionally, it is unknown
if the structural changes cause any functional and behavioral deficits.
2.1.1.2 Neuronal Recording Signals: Single-Unit Activity (SUA), Multi-Unit Activity
(MUA), and Local Field Potentials (LFP)
Neuronal recordings are made possible due to dielectric current flow between the neuron and electrode. Insertion of the electrode into ionic solutions causes charge distribution across the electrode. Depending on the type of microelectrode, this potential can be measured against a reference electrode in various ways.[47, 48] With non-
polarizable electrodes such as glass micropipettes, ions in the solution are charged and
discharged creating a current flow through the electrode. This generates a change in
voltage through the electrode with respect to time. With respect to polarized metal
electrodes, the ions and electrons at the metal surface become polarized with respect to the
solution potential. No steady current passes, so the metal acts as a capacitor as charges
orient at the interface to create an electrical double layer. The change in capacitance with
respect to time can be converted to voltage. After a neuron fires, changes in voltage can
be measured intracellularly or extracellularly as ionic current flows in and out of excitable
15 regions, creating potential fields around the neuron. Most actual electrodes function as a combination of both of these types.
There are three main types of signals isolated from intracortical microelectrodes: single-unit activity (SUA), multi-unit activity (MUA), and local field potentials (LFP).
SUA and MUAs measure high frequency spikes (>1kHz) from the neuron soma.[86] In contrast, LFPs constitute the low frequency (<500 Hz) portion of a raw recording that is generated by changes in the membrane currents at dendritic synapses of local neurons.[87,
88] A majority of BMI studies utilizing intracortical microelectrodes use SUAs. SUAs typically record from neurons within 50-140 um from the electrode.[89]. However, it is difficult to maintain recording from the same neuron for more than a few weeks. Signals recorded within the cortex have high fidelity, but the stability of intracortical recordings can be variable and decay with time.[90] Serious signal degradation occurs due to various reasons (See Section 2.2 for complete description), resulting in a decline in decoding performance. MUAs are extracted, but unsorted spikes defined as threshold-crossings, originating from a handful of neurons. Only a few studies have demonstrated that MUAs can decode kinematic information, and it has become evident that MUAs undergo similar signal degradation over time as SUAs. [91]
Local field potentials generally originate from tens of neurons within 250 µm of the recording electrode.[88] Several studies have suggested that LFPs provide many advantages over SUAs, including longevity, lower sampling rate, less power requirements, and eliminated need for spike sorting.[92] LFP signals can also be decoded without daily retraining or adaptations of the control algorithm.[93] While LFPs appear to be a reasonable alternative to SUAs, studies have not been done to find out if LFPs can produce
16 selective and complex decoding information as well as SUAs. Besides, regardless of the
type of isolated signal, decay in number of channels recording and decoding quality is
apparent over time.
2.2 Challenges to Intracortical Microelectrode Use
Stability and longevity of neural recordings are major hurdles preventing use of
intracortical microelectrodes to their full potential (Figure 3). Many studies have shown
that recording quality decays over time, characterized by reduction in signal to noise ratio,
increased impedance, and loss of neuronal discrimination.[53, 69, 94] A handful of groups
have reported successful long-term neural recording over several years, but there are
conflicting reports when discussing microelectrode recording quality over time. Chestek
et al. reported a rather slow action potential amplitude decline of 2.4% per month (when analyzing the largest action potential on each channel) during a long-term study on rhesus macaques.[95] Simeral et al. assessed neural control in a BrainGate system 1000 days after implantation and reported spiking activity on 41 of 96 channels over 5 days.[81] Suner et
al. recorded primary cortex neural signals in Macaque monkeys for up to 1.5 years,
reporting signals from 65-85 out of 100 electrodes with over 80% showing high quality
signals.[77] However, the number of studies that report significant signal degradation over
time greatly outweighs the successful reports.
17
Figure 3 Summary of recording performance for animals which had isolated units on at least two recording sessions. (Left) Number of single units isolated at each recording session for each animal over time. (Right) First session and last session averages for each microelectrode position as viewed from the top of the 4X4 UEA. Both the average number of units and the average SNR decreased significantly from the first to the last recording sessions (p = 0.002 and 0.01, respectively). Image and caption reprinted with permission from [96].
In 1974, Burns et al. reported a progressive decline in number of single unit
recordings from unrestrained cats, with only 8% probability of recording quality spikes by
5 months post-implantation.[97] Studies since then have indicated gradual decline in SNR from Michigan arrays[69] and significant decrease in firing rate and number of recorded neurons 6 months after implantation.[58] Issues with neural recordings is not limited to the ability to record neurons over long periods of time but also stability of the recordings over multiple sessions. Liu et al. reported instability of implanted microelectrodes during the acute tissue remodeling phase.[98, 99] In a study by Williams et al., on average it took
~33 days before unit activity was continuously observed on all functioning microwire electrodes. However, half of the implanted arrays failed before two months and for the long-term implants they observed a rapid decrease in continuous unit activity from
18 individual electrodes.[53] Instability of neural activities, which may result from functional
reorganization of neural populations[55] or temporal changes in neuronal tuning over time[100], also impacts decoding stability. In a recent 2014 study examining stability of
Utah microelectrode array recordings for up to 690 days, Wang et al. showed variations in preferred directions when evaluating decoding performance. They also observed a serious decay in number of channels recording (~9.7% decline in SUA per month) and mean neuron firing rate.[91] Therefore, forty years later, recording inconsistency is still a commonly reported problem.
2.2.1 Classes of Microelectrode Failure
Microelectrode failure can be defined as complete loss of the ability to isolate neuronal signals through background noise. There is a wide variation in the amount of time that intracortical microelectrodes typically fail, ranging from a few months to a few years. Additionally, the causes of electrode failure are varied and complex. Below the
major sources of failure are defined (Figure 4). However, there is typically not a single
cause of failure as many of the potential sources are intertwined.
19
Figure 4. Failure modes for intracortical microelectrodes. An ideal electrode remains intact and in close proximity to neuronal cell bodies for the entire implantation period. Three common modes of failure are defined: biological, material and mechanical. In many cases, material and mechanical failures can be attributed, in part, to biological events. Image reprinted with permission from [3]
2.2.1.1 Biological Failure Mode
There has been overwhelming evidence that the neuroinflammatory response resulting from device implantation is the primary obstacle preventing wide-spread use of
microelectrode technologies.[101-104] In the Barrese study mentioned above, biological
failure was the primary method of chronic failure, causing slow, progressive loss of signals
over time.[3] Hundreds of studies have been performed describing features of the tissue
response to microelectrode implantation. Efforts have been made to determine how specific characteristics of microelectrode implantation, such as type of implant, insertion method, or material composition, impact various aspects of the inflammatory response and recording quality.
20 In a comprehensive histological study of biocompatibility of tungsten
microelectrode arrays, Freire et al. noted a progressive drop of neuronal activity which they directly associated with glial encapsulation.[58] Interestingly, Vetter et al. noted
significant variability of SNR between probes rather than within probes in an array,
indicating that tissue surrounding a probe affects the entire probe, not just individual
electrodes.[69] The inflammatory response is also non-uniform and depth-dependent
around the probe, showing an increased response at the surface of the brain, which may
have implications on recording electrode position.[105] The shape and size of implanted
microelectrodes has been shown to have a large impact on the extent of tissue response.[8]
Small, cylindrical-shaped microelectrodes produce a much lower inflammatory response
than large, planar microelectrodes. Studies have also reported on a late onset
neurodegenerative state with silicon implants, presumably from continuous BBB
disruption.[24, 106] A more in-depth description of the neuroinflammatory response to
intracortical microelectrodes and existing strategies used to combat it will be discussed in
Section 2.3.
2.2.1.2 Material Failure Mode
Microelectrode failure may also occur due to material degradation of the electrode.
In the Barrese study, they noted material failures due to connector shorting in five of the
implanted arrays. However, their report did not include visual analysis of electrodes post-
implantation for effects of material degradation.[3] Additional material issues that may
occur are corrosion of electrical contacts, degradation of passivation layers and insulating
coatings, and probe breakage. Corrosion of electrical contacts and delamination of
21 insulating coatings can occur over time from cracking due to mechanical stress, film
defects, and chemical or electrochemical reactions. Physical changes to the electrodes can
impact electrical recording properties by incurring changes in electrode surface area,
electrode impedance, and charge transfer.[61-63] Interestingly, as an example of how two failure modes become intertwined, accumulation of highly reactive free radicals during inflammation (a biological failure mode) results in a state of oxidative stress and direct corrosion of electrode sites (a material failure mode). Further details on the effect of oxidative stress on material integrity and microelectrode-induced inflammation will be discussed in Section 2.4.3.
2.2.1.3 Mechanical Failure Mode
A large portion of microelectrode failures occur due to damage to the device hardware. Currently, the cables for each electrode must be routed to a percutaneous connector that has to be connected to a preamplifier system and then to a computer- controlled signal acquisition system. In the study by Barrese et al., they reported that 30 out of 78 arrays failed due to acute mechanical issues.[3] These issues mostly were related to damage to the connector or wire bundle. In a study in rats, Vetter et al. also reported headcap and connector failure in 4 out of 10 animals.[69] Ward et al. noted mechanical failure in 7 out of 19 implanted electrodes taken out to 31 days.[2] Only one of these failures was due to breakage of the probe itself. In the BrainGate study taken out to 1000 days, it was noted that eighteen electrodes had inconsistent recordings attributed to poor pin-to-pad coupling in the connector.[81] With mechanical damage appearing to make up
30-40% of microelectrode failures, improvements to the external hardware and methods to
22 more effectively and stably attach the connector bundle to the skull are needed. Groups
have developed connector caps to protect external portions of the connectors and also more
compact, high-density connectors.[3] Alternatively, the development of wireless systems would circumvent the need for wires and connectors. There has been some progress in
development of wireless systems, which will be discussed in Section 2.4.2.5.
2.3 The Neuroinflammatory Response
From the previous section, it is clear that the failure modes are intertwined. In particular, the neuroinflammatory response can play a role in initiating or worsening material or mechanical failures. Therefore, it has become critical to obtain a more comprehensive understanding of the neuroinflammatory response to microelectrode implants in order to devise strategies to ameliorate its impact and enhance stability of
neural recordings. In the next section, we will provide an overview of each of the major components of the neuroinflammatory response, describe its impact on intracortical microelectrodes, and briefly discuss strategies to mitigate the response. For a more in- depth analysis, the author recommends [107], [101], and [41].
2.3.1 The Acute Inflammatory Phase
The initial acute reaction to tissue injury is similar in CNS tissue as with other vascularized tissue.[108] Early inflammatory events occur within minutes to days of the injury and are targeted towards wound closure and tissue remodeling from damage.
Insertion of invasive microelectrodes induces vascular damage and microhemorrhaging, releasing blood plasma and peripheral blood cells into the wound. This initiates protein
23 adsorption to the implant then platelet aggregation leading to activation of the coagulation
cascade. Clot formation prevents excessive bleeding, while platelet activation within the clot also leads to cytokine release, signaling for chemotaxis of inflammatory cells to the wound site. Additionally, initial insertion can cause damage to neurons and glial cells at the site and activation of the alternative pathway of the complement cascade which further signals chemotaxis and opsonization.
Few studies have done histological analysis of the tissue response at acute time points (<1 week post-implantation). At Day 1 following silicon microelectrodes implantation, immunoreactivity for GFAP+ and vimentin+ astrocytes and ED1+
microglia/macrophages indicated diffuse activation of these inflammatory cells. [109]
Saxena et al. found no acute differences in BBB breach between Michigan and microwire
arrays, which they attributed to excessive tissue edema at 3 days post-implantation.[11,
110] We have also looked at 3 day time points in attempts to examine the acute effects of
material compliance (Chapter 3) and antioxidant delivery (Chapter 6) and observed no
significant differences between control and experimental groups. We suggest that this data
supports the idea that wound healing and tissue remodeling events that occur during acute
time points are affected predominantly by trauma associated with device implantation.
Strategies to reduce initial acute damage may improve electrode stability. A study
by the Shain group suggested that there is an early cellular response lasting 7-10 days
triggered by device insertion, and a separate sustained chronic response at 2+ weeks.[109]
The early cellular response is proportional to device size and may be associated with the
amount of damage generated during insertion. They subsequently reported on the effects
of various insertion speeds and implant tip shapes. [27, 84, 111, 112] It was found that
24 blunt tip insertion increased shear stress and tissue damage, but these effects could be
avoided with faster insertion speeds. Similarly, Utah electrode arrays are inserted
pneumatically, in part to, to completely insert arrays with minimal tissue trauma.[84]
Researchers have also attempted to reduce the probe size to avoid vascular damage and
reduce tissue trauma.[113] However, reduction of shank size may produce other issues
such as breakage from brittleness and buckling during insertion (see Section 2.4.2.4). An
in vivo imaging study showed that blood cell and serum infiltration was greatly affected by
proximity of the inserted probe to large vasculature.[114] In order to avoid excessive vascular disruption, Kozai et al. utilized two-photon mapping to provide in vivo image guidance during insertion.[115] Results indicated that outgrowth of microglial processes occurred immediately following implantation. However, the proposed technique only allowed examination of ~500 µm depth and the study did not examine if reduced initial neurovascular damage improved electrode function.
2.3.2 The Chronic Inflammatory Phase
2.3.2.1 Microglia/Macrophages
After the acute inflammatory response, the body’s response transitions into a chronic inflammatory response to the foreign implant. The first cells to respond are resident brain microglia and blood borne macrophages. Microglia constitute 5-20% of
brain cells and are the immune effectors cells of the CNS (similar to macrophages in the
rest of the body). In a normal, resting state microglia maintain central homeostasis and
constantly monitor the brain tissue for potential threats (Figure 5, left).[116-120]
25
Figure 5 Activity states of microglia. On the left, in healthy tissue, resting microglia survey for disturbances in homeostasis and neurons deliver signals that inform microglia that they are functioning normally. On the right, a large insult, such as microelectrode implantation, triggers activation, proliferation, and migration of microglia. Activated microglial cells are neurotoxic, releasing reactive oxygen species, nitric oxide, and inflammatory cytokines. Sustained responses lead to substantial impairment of neurons and glia. Image reprinted with permission from [117]
Significant tissue injury from microelectrode implantation triggers activation of
microglia, characterized by a change from a ramified to amoeboid shape and rapid changes
in gene expression and functional behavior (Figure 5, right). Local microglial populations may proliferate to induce more cells for defense and protection. Microglia also become highly phagocytic to clear tissue debris, damaged cells or microbes.[121] Changes in gene expression leads to altered receptor expression and release of proinflammatory and immunoregulatory cytokine and chemokines. Accumulation of proinflammatory molecules, such as reactive oxygen species, also creates a state of oxidative stress that can further activate inflammatory pathways (See Section 2.4.3).[122] Additionally, release of chemoattractive factors recruits infiltrating leukocytes to the site of injury.[118]
Upon detection of threat, microglia transition into a reactive state. There are varying classes of microglia polarization. Activated microglia and macrophages are predominantly classified into two types: classically activated M1 microglia and
26 alternatively activated M2 microglia.[117] The classification of activated microglia/macrophages is determined by functional properties such as activators, receptor expression, and cytokines released. M1 microglia/macrophages are associated with the inflammatory response and release of pro-inflammatory cytokines, chemokines, and reactive free radicals. M2 microglia/macrophages are generally characterized by low production of proinflammatory cytokines and promote tissue remodeling and immunoregulation. However, recent studies indicate that these two activation states are extremes of a spectrum of possible forms of microglia/macrophages activation, and upon activation microglia/macrophages may express varying combinations of characteristics from each activation class.[119, 120]
A key feature of the brain’s response to microelectrode devices is persistent inflammation involving activation of resident microglia and recruitment of blood borne macrophages.[70, 101, 102, 109] The presence of insoluble materials in the brain may lead to a state of “frustrated phagocytosis” when microglia are unable to remove the foreign body, resulting in persistent release of neurotoxic substances.[101] Szarowski et al. suggested that the microglial response would continue as long as the material is implanted[109], and additional studies have shown that continued microglial activation correlates to neurodegeneration.[102] Biran et al. similarly found that nearly complete tissue repair was possible in a stab wound in contrast to chronic implants.[70] In a long term study, Freire et al. reported varying levels of microglia over time. ED1+ microglia were upregulated at 1-2 months and decreased by 4 months, but ED1+ levels rose again after >6 months.[58] There also appears to be a depth related effect to the amount of glial
scarring. Expression of Iba1+ microglia along an electrode shank was highest within the
27 first 1-400 um at the base of the electrode (close to the skull) and decreased near the tip.[105]
2.3.2.2 Astrocytes
Astrocytes make up about 30-65% of brain cells and play a key role in healthy brain homeostasis. They are vital for specific neuroprotective functions such as glutamate uptake, K+ buffering, and elimination of free radicals.[123] Additionally astrocytes contain receptors to most neurotransmitters, providing key mechanisms for intercellular communication.[124] Astrocytes have ~11 distinct phenotypes, and 8 of those phenotypes involve specific interactions with blood vessels. Evidence suggests that astrocytes upregulate many BBB features, including creating stronger tight junctions, increasing expression and localization of transporters, and activating specialized enzyme systems.[125]
After traumatic CNS injury, astrocytes undergo a characteristic change to a reactive state, which can be distinguished by increased expression of glial fibrillary acidic protein
(GFAP). The presence of these reactive astrocytes has been shown to have both beneficial and detrimental effects during traumatic injury.[124] Implantation of a microelectrode leads to the formation of an astroglial scar at the site of injury and encapsulation of the device. The hypertrophic astrocytes secrete extracellular matrix proteins, including laminin, proteoglycans, and fibronectin, which make up the glial scar. It is largely unknown if the detrimental effects of astrogliosis is due to decreased astrocytic support activities or from astrocytic production of cytotoxic factors. For instance, damage to astrocytes may reduce normal glutamate uptake and contribute to excitotoxic cell death of
28 neurons. Furthermore, after injury, astrocytes can produce pro-inflammatory cytokines
such as TNFα which facilitate leukocyte extravasation into the CNS.[126] The glial scar
also serves as a physical barrier to regenerating axons. Upon transition into “reactive” astrocytes, astrocytes increase expression of chondroitin sulfate proteoglycans, which are
shown to inhibit neurite outgrowth.[127] Astrogliosis can also significantly affect the diffusion properties of neural tissue. Cellular swelling and changes to ECM volume can affect neuron-glia communications, synaptic transmission, and soluble factor diffusion.[128-130] Alternatively, microelectrode encapsulation may be beneficial in mechanically shielding surrounding tissue from micromotion-induced electrode movement.[5, 20, 23]
Many microelectrode studies focus on characterizing the effects of glial encapsulation surrounding the implant on electrode impedance, suggesting that astrogliosis increases tissue impedance and creates a barrier to electrical transmission. Impedance spectroscopy to quantify electrode functionality has been widely utilized, and groups have suggested that these recordings at frequencies above 1 kHz can provide information on the characteristics of the tissue surrounding the electrode.[94, 131] However, data has been inconsistent in directly correlating increase in impedance to reduced recording quality.
Furthermore, no studies have directly shown that an increase in electrode impedance directly relates to tissue composition. Impedance increases may occur due to electrode corrosion or damage to insulting layers.[132] Additional studies need to be done to correlate impedance changes with tissue changes and recording ability.
29 2.3.2.3 Neurodegeneration
The success of microelectrodes is contingent on obtaining and maintaining recordings from neurons within the brain. Neurons produce electrical signals to encode and control every bodily function. Neurodegeneration, defined as the progressive loss of structure or function of neurons potentially leading to neuron death, is implicated in many neurodegenerative diseases such as amyotrophic lateral sclerosis, Parkinson’s, Alzheimer’s and Huntington’s. In the context of microelectrodes, neurodegeneration has been prominently described when assessing the tissue response to implanted electrodes. [70, 98,
102] Buzsaki et al. suggested that neurons must be at least 50-140 µm from the electrode to obtain reliable and stable recordings. Studies have generally seen a 40-60% reduction in neuronal density at the interface.[41, 133] However, no studies have explored the functionality of the neurons at the interface so in reality, the neuron viability may be even lower. When exploring the inflammatory events following implantation, several groups have indicated an exponential loss of neurons immediately after insertion.[64, 70, 134]
McConnell et al. and Potter et al. both observed a multiphasic response, characterized by an early and late onset of neurodegeneration.[24, 102]
The health and proximity of neurons has been strongly linked to the extent of neuroinflammation. Insertion damage can shear and induce neuronal cell death. Extensive studies have linked microglial activation and neurodegeneration.[135-141] Production of pro-inflammatory cytokines from reactive astrocytes and microglia during inflammation have neurotoxic effects when overproduced and unregulated. Astrocytes may also play a role in neurodegeneration, by creating a physical barrier for growth of neural processes
30 close to the electrode sites or failing to modulate glutamate levels leading to excitotoxic
neuronal death (Section 2.3.2.2).
2.3.2.4 BBB Permeability
The blood brain barrier (BBB) is comprised of brain endothelial cells that create a physical barrier lining cerebral microvessels. Specifically, the BBB is formed by capillary endothelial cells surrounded by basal lamina, pericytes, and astrocytic endfeet. The BBB supplies the brain with essential nutrients and mediates clearance of waste products.
Additionally, a complex network of tight junctions between adjacent endothelial cells prevents molecular traffic paracellularly and restricts entry of hydrophilic drugs, large molecules, and even movement of small ions such as Na+ and Cl-. Specific ion transporters and channels regulate ionic traffic between the blood and the brain.[125] Due to reliance on the BBB to allow transport of vital nutrients, neuron cell bodies are typically less than
10 µm from the nearest capillary.[125]
31
Figure 6. Self-perpetuating neuroinflammatory pathways in response to microelectrode implantation. Blood- brain barrier integrity is compromised after microelectrode implantation. In response, infiltrating blood cells and resident microglia release pro-inflammatory molecules which further activate pro-inflammatory cells and aggravate BBB breakdown. Image reprinted with permission from [15]
Disruption of the BBB is a common feature of many neurodegenerative disorders,
due to infiltration of pathogens or cytotoxic immune proteins and cells.[142] Blood brain
barrier trauma induces significant vasogenic edema three days after injury and even sustained BBB injury after 2 months.[110] Therefore, trauma from microelectrode
implantation causes direct mechanical damage, an efflux of detrimental elements into the
surrounding tissue, and activation of glial cells (Figure 6). Many of the cytokines
produced in response to tissue injury also contribute to BBB dysfunction, such as TGFβ,
IL-6, MCP-1, and TNFα.[125, 143] The mechanical mismatch between the electrode and
brain tissue may also induce perpetual damage to the BBB, exacerbating continual
inflammation and neurodegeneration.[4, 5, 26] Numerous studies have examined the
relationship between material compliance and neuroinflammation. Computational
32 modeling suggests that softer materials reduce tissue strain on surrounding tissue.
Researchers have compared the immune response to tethered and untethered probes,
finding that untethered probes have reduced neuronal loss and inflammatory response.
[144, 145] Together, the above mentioned studies suggest minimizing BBB mechanical
trauma can reduce neuroinflammation. Further, in Chapter 3, compliant materials are
shown to reduce the chronic inflammatory response compared to stiff implants. A more
comprehensive description of the effects of material compliance on microelectrode
implantation can be found in Section 2.4.1.
2.3.2.5 Anti-inflammatory Strategies
Strategies to reduce inflammation surrounding intracortical microelectrodes have
mainly focused on probe coatings and administration of anti-inflammatory drugs. With varied success, dexamethasone, an anti-inflammatory glucocorticoid, has been administrated via systemic injection [146, 147] and released from hydrogel coatings around neural probes [148-151]. Most studies report reduction in reactive astrocytes but little effect on activated microglia/macrophages. Rennaker et al. found that systemic administration of the antibiotic minocycline for up to 4 weeks post-implantation improved the SNR and % of active channels, attributed to reduced reactive cells and increased neuronal survival.[9] However, minocycline failed to stabilize neural recordings at early time points and chronic administration of an antibiotic is not a viable solution. A promising study by Bellamkonda et al. found that immobilization of neuropeptide α-melanocyte stimulating hormone (α-MSH) exhibited anti-inflammatory effects at 1 week and 4 weeks post-implantation, suppressing microglial activation and reducing production of
33 proinflammatory cytokines.[152] Another alternative explored by our lab is administration of natural antioxidants, specifically resveratrol and curcumin, to combat oxidative stress, described in detail in Section 2.4.3.3.
Another set of studies have focused on enhancing neural adhesion or neural regeneration at the implant site.[153, 154] Coating of probes with laminin, a neurite promoting ECM protein, showed neurite outgrowth in vitro, but no observable effect on neuron growth in vivo.[155, 156] The Cui group immobilized L1, a neural adhesion molecule, to promote neurite outgrowth and neuronal survival surrounding implanted microelectrodes.[157] The results indicated no loss of neuronal cell bodies and a significant increase in axonal density at the interface.[158, 159] Studies have also considered coating electrodes with neural progenitor cells to re-establish neuronal populations, but the feasibility of delivering live cells is questionable.[160, 161]
As this section has highlighted, the characteristics of the tissue surrounding microelectrodes play a crucial role in the stability and integration of implanted devices.
The neuroinflammatory response is multi-faceted and complex, therefore a single solution may not be sufficient to adequately mediate the damaging elements. However, there is still debate on the true effects of neuroinflammation on neural recording. Studies have indicated successful recordings for months to years, and histological studies have failed to directly correlate neural recording function to the neuroinflammatory time course of events. However, recent literature has supported the hypothesis that the integrity of the blood-brain barrier (BBB) may be a major factor in the neuroinflammatory response to microelectrodes. When examining the tissue response surrounding implanted microwire and Michigan-style microelectrodes, the Tresco group found that the local BBB integrity
34 immediately surrounding the implantation site was compromised.[12, 13] Rousche et al.
suggested that the extent of neuroinflammation is related to proximity of the implant to
large vessels.[76] A study by Kozai et al. also implicated that increased bleeding intensity can cause a reduction in SNR.[162] Our lab has previously verified that the administration of antioxidants is able to improve BBB integrity and neuronal proximity at acute time points.[14, 15] Additionally, Saxena et al. correlated the long-term stability of functional
single unit recordings in rats with changes in BBB permeability.[11] This was the first study to directly relate neuroinflammatory events to the quality of neural recordings.
Further, a recent study by the Tresco group reported a correlation between tissue loss and recording performance with UEAs.[96] Microelectrodes located in the center of the array had significantly higher levels of GFAP and IgG immunoreactivity correlating to reduced
SNR compared to microelectrodes on the edge. Therefore, given the fresh evidence, improving BBB integrity appears to be a reasonable target and many strategies to reduce the damaging effects of BBB dysfunction are currently being explored.
2.4 Strategies to Improve BBB integrity
As mentioned in Section 2.3.2.4, BBB integrity is compromised following microelectrode implantation by a variety of factors. The insertion process itself induces mechanical breakage and damage to the BBB. Strategies to avoid vasculature and reduce insertion damage were previously discussed (Section 2.3.1). Furthermore, the presence of the microelectrode can induce tissue strain and micromotion can cause microelectrode drift.[98, 163] Blood brain barrier integrity is also heavily influenced by pro-inflammatory molecules that are released from activated glia during the inflammatory response. These
35 include accumulation of cytokines, chemokines, and reactive oxygen species (ROS) which
can create a neurotoxic environment while also damaging the microelectrode. Here, we
will discuss two of the major strategies to reduce BBB leakiness: (1) Alterations in
microelectrode compliance to reduce the effects of micromotion-induced tissue strain and
(2) Antioxidant therapies to reduce oxidative stress events.
2.4.1 Effect of Microelectrode Compliance
In the 1970s, Goldstein and Salcman first discussed the issue of microelectrode drift due to brain micromotion.[163-166] They hypothesized that pulsation of the cortical
surface and mechanical mismatch between implant and brain tissue cause continual
mechanical disruption of tissue and may play a role in poor neural recording performance.
In their work, they also analyzed the tradeoff between small diameter implants to
minimizing tissue damage and large diameters to avoid electrode buckling. This
hypothesis went relatively unnoticed until the late 1990s when Loeb et al. referenced
Goldstein and Salcman in a review describing that the ultimate metal microelectrode “must
minimize disruption of tissue along its insertion path but it must be stiff enough so that the
tip will not be deformed and the shank will follow a straight track even when inserted
through mechanically tough materials such as epinerium and dura mater.”[167] Following this, the impact of micromotion on microelectrode function became a new study interest.
In 2000, Maynard et al. reported on implantation of a Teflon barrier between the electrode array and the dura in an attempt to reduce fibrous adhesion and subsequent mechanical displacement between the electrode and cortical tissue. While there was an improvement in number of viable electrodes, no data was presented on the SNR of the
36 recordings.[85] Similarly, studies indicated that wireless, untethered microelectrode systems would elicit a lower neuroinflammatory response.[8, 145]
In silico studies have supported the hypothesis that micromotion-induced tissue
strain is increased around stiff implants.[4, 5, 29] The model results indicated that soft substrates in the MPa range reduced strain at the probe-tissue interface (Figure 7).
Polymers commonly under investigation as alternative microelectrodes are typically in the low GPa modulus range and are not soft enough to reduce strain in soft brain tissue
(E = ~7 kPa). Additionally, good adhesion and tissue integration at the probe-tissue interface is desirable to reduce slip and shear stresses due to micromotion forces.
Molecular analysis of astrocytes and microglia that underwent low magnitude cyclic strain showed upregulation of IL-36Ra gene expression, a change not observed in LPS activated glial cells.[26] These results suggest that IL-36Ra may be a marker for strain-induced inflammation and contribute to neurotoxic events. In vitro studies by Moshayedi et al. also recently reported that surface stiffness has mechanoresponsive effects, causing upregulation of various microglial inflammatory receptors such as TLR4.[25]
37
Figure 7. Finite element analysis of tissue strain surrounding tethered microelectrodes. Displacement from tangential tethering forces on a traditional stiff probes (E=100 GPa) (A) induces strain along the entire shank of the electrode. The strain profile surrounding a polyimide probe (E=1 GPa) (B) has less strain at the tip, but the hypothetical soft implant (E=1 MPa) (C) has negligible strain at the tip and along the shank. Image reprinted with permission from [4]
In an effort to better understand the magnitude of brain tissue micromotion, the
Muthuswamy group characterized the surface micromotion in the rat cortex, observing
brain displacement of 10-30 µm due to respiration and 2-4 µm due to vascular
pulsatility.[168] To compensate for electrode drift from brain micromotion, they developed
microactuated neural probes, capable of moving in 1 µm steps.[169] Additionally,
Muthuswamy et al. developed a custom-made force measurement system to quantify the micromotion-induced tissue stresses and strains surrounding microelectrodes. The results indicated remodeling of the tissue surrounding probes for weeks after implantation.[6]
Unfortunately, little research has been done to directly measure the strains at the probe-tissue interface. Most research cites indirect relationships extrapolated from acute data. Short term force measurements from various types of electrodes inform on acute insertion stresses and buckling forces.[23, 27, 111, 112, 170] However, efforts must be
38 made to quantify long-term strains and provide compelling evidence that micromotion- induced strain is a valid concern. Current efforts between our group and the Muthuswamy group to quantify and compare changes in tissue remodeling around traditional stiff probes and soft probes is discussed in Chapter 4. Despite the lack of hard evidence, many groups have devised strategies to reduce the potential effects of mechanical mismatch.
2.4.2 Strategies to Reduce Microelectrode-induced Tissue Strain
2.4.2.1 Polymers
Table 1. Strategies to reduce microelectrode compliance with various types of polymers. Type Coating Elastic Modulus (MPa) References Alginate N/A* [171, 172] Polymer PEG N/A* [173, 174] Coatings Fibrin N/A* [175] Gelatin N/A* [176, 177] Polyimide 2500 [7, 178-181] Parylene C 2800 [17, 182-185] Polymer Substrates SU-8 2200 [18, 177, 186-188] PDMS 1.81 [173, 189, 190] Benzocyclobutene (BCB) 3000 [191] Polypyrrole (PPy) 300 [192-199] Conductive Polymers PEDOT 900 [200-202] EDOT Unknown [203] PVAc-CNC 5000 (stiff) to 12 (softened) [19-23, 204-211] Mechanically- Thiolene/Acrylate 1000 (stiff) to 18 (softened) [10, 212-214] Adaptive Polymers PVA 8800 (stiff) to 7 (softened) [14] Agarose/CNT 1000 (stiff) to 220 (softened) [215] *Elastic modulus varies because polymer coatings were applied to a variety of substrates
Initial attempts to reduce microelectrode-induced strain involved thick polymer coatings of traditional implant (Table 1). Biologically inert polymers such as polyimide,
Parylene C, and PDMS, are commonly used insulating coatings. A study by Winslow et al. also showed that there was no difference in in the tissue response to Parylene C coated
39 and uncoated silicon microelectrodes.[12] Research has since moved towards hydrogel coatings, including alginate, fibrin, PEG, and agarose. The goal of many of these studies were to reduce glial adhesions, or enhance neurite outgrowth, but a secondary consequence of the coatings was a change in the probe mechanics. Also, hydrogel coatings, because of their swelling, can function as a “cytokine sink” to passively entrap soluble factors until they degrade.[171] In a study by Gutowski et al., they found reduced astrocytic scarring around microgel coated silicon electrodes all the way up to 24 weeks, but no effect on microglia or neuron density compared to controls.[216] This suggests that hydrogel coatings alone may not be sufficient to reduce the tissue response. A number of studies also use hydrogel coatings as a drug delivery platform. A majority of studies report on release of dexamethasone, showing varying degrees of improvement in astrocyte encapsulation. Recent studies have investigated live cell delivery, glial and/or neural cells, in attempts to improve electrode tolerance and induce neural regeneration.[161] Initial studies show reduced inflammation and improved neurite outgrowth as well as acute recordings, but chronic effects of cell delivery have not been determined.
An alternative and popular approach to reduce microelectrode compliance is the use of polymer substrates (Table 1). While polymer substrates are appealing due to their low modulus and ease of fabrication, a major disadvantage is the need for insertion aid.
The low compliance of the fabricated electrodes requires the electrodes to be larger than traditional implants or employ an insertion shuttle to prevent buckling during insertion.
Polyimide is a popular polymer substrate due to mechanical flexibility, good biocompatibility, insulation resistance, and dielectric strength.[217] Fabrication with polyimide is also easy and many papers have developed methods to fabricate
40 microelectrodes of various materials on polyimide.[7, 178-181] Results from multiple groups report on good quality acute and chronic recordings with minimal tissue response.
However, these studies fail to directly compare polyimide microelectrode function to traditional microelectrodes. Another popular polymer substrate, Parylene C has demonstrated long-term biocompatibility and low fluid uptake.[217] Fabrications with
Parylene C substrates typically incorporate PEG or a silicon backbone to enhance stiffness.[180, 182, 184] Multiple groups have also created microfluidic delivery probes of Parylene C.[17, 183] However, a majority of these studies report on electrode fabrication, but present limited recording data and no data on the inflammatory response to these implants. Other polymers such as SU-8, PDMS, and benzocyclobutene, have been similarly used but also have no evidence of long term in vivo recording and biocompatibility.[173, 186, 189, 191]
Conductive polymers have also gained attention in the last decade. Conductive polymers benefit from reduced impedance which improves charge transfer, high surface area to increase conductivity, and typically are made of softer materials which can reduce mechanical mismatch.[218] Extensive studies by the Martin group have explored polypyrrole (PPy) electrodes with bioactive molecules as a means for increasing tissue integration with electrode sites.[192-194, 219] Polypyrrole on hydrogel probes have been shown to induce tissue integration and improve compliance of electrode against micromotion.[192] Unfortunately, PPy is highly unstable due to its poorly defined chemical structure. Recently, a more electrochemically stable, alternative conductive polymer, poly(3,4-ehtylenedioxythiophene) (PEDOT), has been developed.[200]
Bioactive molecules incorporated into conductive polymers include laminin peptides,
41 hyaluronic acid, polysaccharides, and nerve growth factor. More recently, neural stem
cells have been embedded in conductive hydrogel coatings, made of PEDOT and PVA, on
platinum electrodes.[161] The feasibility study indicated that the charge injection limit of
these electrodes was eight times higher than conventional platinum electrodes and the
stiffness was four orders of magnitude lower. Future studies are still needed to determine
in vivo efficacy and long term recording potential.
2.4.2.2 Insertion Shuttles
The increased compliance of microelectrodes increases the likelihood of buckling
upon insertion into the brain. As researchers continue to develop softer and smaller devices, there has been a need for development of microelectrode insertion aids. These are typically classified as two types: biodegradable coatings and insertion deployment systems.
A vast majority of insertion aids involve biodegradable coatings, including carboxymethyl cellulose (CMC), PLGA, sucrose, silk, and a tyrosine-derived terpolymer.[217, 220-224] Biodegradable coatings have been applied to a number of substrate materials to enhance microelectrode stiffness. CMC, a natural, water-soluble polysaccharide, has been demonstrated to be non-toxic and is extensively used for dissolvable shuttle coatings.[221] A long-term study comparing CMCs shuttles of various sizes indicates that large CMC shuttles had no benefit on the tissue response in comparison to traditional microwire implants, whereas small CMC shuttles displayed improved neuron density chronically.[220] Silk fibroin, biocompatible and biodegradable, has garnered attention because it can be programmed to control dissolution rate as a stiffening material or for extended drug release.[223, 225] Silk-coated polyimide electrodes (E=1.8 GPa)
42 enhanced penetration mechanics and reduced gliosis and demonstrated sustained drug delivery for up seven days in vitro. Functional recording and in vivo characterization are still unknown. One of the most promising biodegradable coatings utilizes tyrosine-derived polycarbonate.[226] The ultrafast degradation and resorption properties enable pial penetration and enhanced brain tissue recovery in comparison to PLGA coatings at four weeks.[227] However, no differences in the tissue reaction were seen between coated and uncoated microwire probes and results varied for electrophysiological recording.[224]
Dexamethasone release from tyrosine-derived polycarbonate coatings reduced the reactive astrocyte response, but no functional recording data was reported. While biodegradable coatings successfully enhance insertion capabilities of compliant devices, the potential toxicity of degradative by-products needs to be thoroughly assessed long-term.
Microelectrode deployment systems have also been developed. Kozai et al. employed self-assembled monolayer (SAM) coated silicon shuttles to deliver PDMS.
When implanted into the brain, the hydrophilic SAM attracts water at the probe-shuttle interface causing the polymer probe to detach from the shuttle.[190] Felix et al. employed a similar system utilizing PEG as a dissolvable adhesive between the probe and shuttle.[228] Neither of these studies investigated the effects on the tissue response or neural recording. A novel strategy involves magnetically-guided insertion of ultrasmall ferromagnetic microelectrodes.[229] Dryg et al. demonstrated successful implantation of
25 µm diameter microwires, high fidelity recordings up to 31 days, and reduced tissue response around the ultrasmall implants. However, the % of active channels recording was still <40% for the entire duration of the study.
43 2.4.2.3 Mechanically-adaptive materials
Another approach to avoid issues with buckling of compliant materials is to develop an entirely new class of materials. Inspired by the sea cucumber, which changes from a soft to stiff state when distressed, a group of researchers at Case Western developed a mechanically-adaptive nanocomposite (NC) material that softens upon implantation.[19-
22] Comprised of cellulose nanocrystals (CNC) and polyvinyl acetate (PVAc), the dry material is rigid (E’=5 GPa) due to the glassy state of the PVAc and percolating network of the CNC. Upon in vivo implantation, the NC undergoes a phase transition due to water plasticization and disassembly of the CNC network, resulting in material softening to ~12
MPa over 15 minutes. The initial dry, stiff state allows for easy penetration through the brain tissue and the soft state reduces tissue strain.[23, 211] An initial study exploring the impact of the NC implants on neuron viability indicated advantages over microwire controls up to four weeks post-implantation, which wore off by eight weeks.[20] However, previous research suggests that there is a secondary, later onset neuroinflammatory event.[24, 58] Therefore, non-functional compliant implants were implanted for longer time points (see Chapter 3). Here, we report that compliant materials can reduce neuroinflammation and improve neuronal viability long-term compared to stiff controls.[19] Similar to the previous study by Harris et al., the compliant material was less effective at acute time points. Therefore, an additional approach is needed to produce short-term improvements in neuronal viability (see Chapter 5). Fabrication of functional electrodes from the NC has proved challenging due to fluid uptake by the material (70% swelling by weight), but recently acute neural recordings have been demonstrated from
Au/Pt electrodes on the second generation of NC substrate(Figure 8).[230]
44 A
B
Figure 8. Second generation mechanically-adaptive nanocomposite (NC). (A) The mechanically-adaptive material comprised of a composite of polyvinyl(acetate) and cotton cellulose nanocrystals (PVAc-cCNC) demonstrates a change in Young’s modulus over time as it is immersed in deionized water at t=0. (B) A sample microprobe fabricated from PVAc-cCNC (15%) with a Parylene C insulated Ti/Au contact at the tip on the left and a connection contact pad on the left. Images reprinted with permissions from [230]
A second type of mechanically-adaptive polymer has since been developed by a
team at UT Dallas.[10, 212, 213] The first generation shape memory polymer utilized a
cross-linked acrylic network which softens from 700 MPa to 300 kPa within 24 hours,
whereby the modulus drop is triggered by water plasticization and thermal activation when
heated above glass transition temperature (Tg). However, electrode fabrication proved difficult because of rapid fluid uptake by the polymer. The second generation materials are thiolene/acrylate based. The thiol-ene “click” reaction was added to increase material
45 rigidity, reduce swelling, and better support microelectrode fabrication. The thiol-ene reaction polymerizes thiol and –ene monomers through a free-radical step-growth mechanism to yield a highly uniform, crosslinked polymer network. The addition of hydrophobic diacrylate monomers allows further tuning of the Tg. The resulting materials
soften from 1 GPa to 18 MPa after one week of implantation, and in contrast to previous
materials only swell <3%. The second generation materials were successfully fabricated
into microelectrodes and produced acute recordings with average SNR=3. However, long-
term recording quality or tissue response to traditional microelectrodes have not been
assessed.
2.4.2.4 Architecture Modifications
An alternative approach to reduce material compliance is to change the electrode
architecture. A clear strategy would be to decrease the size of microelectrodes. The size
of implanted devices appears to have a considerable impact on activation of inflammation.
[231] Not only does it minimize insertion damage, but direct studies of the microelectrode size strongly suggest that the inflammatory response is reduced around ultrasmall electrodes.[113] To date, three studies have reported on successful recording from ultrasmall microelectrodes (Figure 9E) in vivo for up to five weeks post-implantation.[113,
215, 229] One study utilized a carbon nanotube/agarose fiber hybrid to create ultrasmall implants.[215] While CNTs have been deemed safe in vitro, there have been mixed results
when applied in vivo. Free-floating CNTs are linked with brain, lung, and skin
cytotoxicity, which may be a potential issue if the CNT fiber degrades.[231] Additionally,
46 insertion is an issue with these ultrasmall implants and both studies required implantation
assistance due to extreme fragility of the probes.
Figure 9. Examples of microelectrodes with architectural modifications. Changes to the structural design are aimed to improve tissue integration (A,B), reduce compliance (C,D), and reduce vascular tissue damage (E). Images reprinted with permission from [13, 113, 134, 223, 229, 232-234]
Other structural design modifications have also been attempted with positive
results. Investigations by Seymour et al.[134] and Skousen et al.[13] found that a lattice structure reduced tissue encapsulation, including decreased macrophage activation and reduced neuronal cell loss (Figure 9B). Additionally, lattice and fish-bone shaped implants allow for tissue integration and better proximity to neurons.[232] A sinusoidal microelectrode (Figure 9D) is designed to minimize tethering forces and a spheroid tip
47 anchors with the brain. Special insertion with a needle and PEG adhesive is required for insertion.[234] Sinusoidal microelectrodes had similar SNR to microwire electrodes but increased signal stability over 678 days. Histological analysis also indicated reduction in microglial activation and improvement in neurofilament staining at 12- and 24-months.
Another interesting study utilized braided electrodes (Figure 9C), 24 ultra-fine wires interwoven into a tubular braid, creating a final diameter of 150-200 µm.[233] Mechanical bending tests indicated that the braided implants had 4-21 times better mechanical compliance than a single 50 um wire. Preliminary recordings from implantation in the spinal cord had SNR from 4 to 7, but histological analysis has not been reported.
2.4.2.5 Wireless Systems
The ultimate solution to reduce micromotion-induced strain is to implant untethered, wireless implants. In Biran et al.’s study of tethered and untethered implants, there was a significant decrease in the tissue response around untethered implants.[145]
FEM modeling of a flexible interconnect indicated reduced interfacial strain with polyimide E=2 GPa (66%) and PDMS E=6 MPa (2 orders of magnitude) connectors.[16]
Furthermore, micromotion-induced tissue strain would not be an issue with wireless implants since microelectrodes would be able to freely-float on the brain surface.[4, 8]
Wireless systems would also be more portable and suitable for real world use.
Additionally, complete healing of the scalp wound would reduce the risk of potential infections. Wireless transmission signals can be transmitted by electromagnetic radio frequency (RF), infrared, or acoustic energy.[65] The main challenges to wireless systems include power limitations (~10mW) to avoid tissue damage from heating and the ensuring
48 the capability to transfer large, continuously streaming data sets with little latency. The
power limitations also mean the implanted device tend to only perform minimum required
amplification, data compression and telemetry. [235]
One of the longest and most successful wireless systems has been used by the
Kennedy group since the late 1980s, the neurotrophic cone electrode.[154, 236-238] A
glass cone filled with matrigel, neural growth factor (NGF), and an insulated gold wire
induces cortical neurite growth into the cone. Signal transmission is achieved via FM
radio. The neurotrophic cone electrode is the only wireless, implantable BMI system that
has been successfully implanted in humans to restore speech for 4+ years.[239, 240]
However, the resolution and information transmission is still poor in comparison to UEAs.
A neurotrophic probe only contains a single wire channel and only 4-5 probes can be
implanted in a single subject due to size constraints. Studies are underway to increase the
number of wires per electrode to increase the number of recordable units.[235]
With regards to other microelectrode types, the development of wireless arrays are
currently underway.[65, 241, 242] Sodagar et al. reported on the most recent design of a
64 channel wireless Michigan array.[243] Compared to previous wireless systems, this
system was lighter (275 mg), smaller (2.3 mm2), dissipated 14.4 mW of power, and transferred at 2 Mbps. Spontaneous and evoked neural responses were successfully recorded from guinea pig auditory cortex for up to 100 days. Harrison et al. developed a low-power 100 electrode wireless UEA which receives power and commands at 6.5 kb/s wirelessly and transmits neural data back at a rate of 330 kb/s using a frequency-shift- keying (FSK) transmitter.[244, 245] The initial prototype was able to record from the cat cortex for over five weeks. In another study, a sixteen-channel wireless UEA array
49 transmits digitized neural signals via infrared light pulses and showed robust spikes in
nonhuman primates for up to one month.[246] While there have been great strides towards wireless systems, there are still many challenges including improved probe design to minimize tissue reaction, packaging techniques for long-term use, increased data transmission for data flow, spike recognition and signal processing.[65]
2.4.3 Effect of Oxidative Stress
Oxidative stress is characterized as harmful and damaging biological effects caused by accumulation of free radicals, chemical species with unpaired electrons. Studies have indicated oxidative stress affects both the biotic and abiotic response to intracortical microelectrodes.[247] In this section, the role of free radicals during inflammation, the effects on microelectrode materials, and strategies targeted to mediate oxidative stress events are discussed.
2.4.3.1 Reactive Oxygen Species (ROS)
The most biologically relevant free radicals are predominantly derived from oxygen, reactive oxygen species (ROS), but also include reactive nitrogen species (RNS).
Reactive oxygen species are generated from multiple sources within a cell, including the mitochondrial electron transport chain, ionizing radiation, and enzyme production
(NADPH oxidases, lipoxygenases, and cyclooxygenases).[122] The extent of ROS accumulation plays a key role in the occurrence of oxidative stress. At low levels, ROS is beneficial as a defense to microbial pathogens and leads to production of antioxidative enzyme, creating a nice balance of ROS and antioxidant.(Figure 10) An increase of ROS
50 triggers the activation of the NF-kB and AP-1 pathways, key inflammatory pathways that lead to production of inflammatory cytokines.[248] Further, very high ROS levels cause disruption of the electron transport chain causing apoptosis and necrosis.[249]
Figure 10. The level of oxidative stress is dictated by the ROS levels. While a low level of ROS can be beneficial, increased ROS induces inflammation and cell death. Imaged reprinted with permission from [122]
During the microelectrode-induced inflammatory response, the intermediate and high oxidative stress states are in play. Significant ROS accumulation around intracortical microelectrodes, as seen by dihydroethidium immunostaining, has been noted for up to four weeks.[15] ROS activation of NF-kB pathway, propagates production of inflammatory cytokines, which further induces release of ROS, activates microglia, and recruits blood-borne inflammatory cells. Additionally, ROS plays a role in cytokine-
induced NF-kB activation via specific IL1β, TNFα, and LPS pathways. High levels of
ROS induces cell apoptosis through mitochondrial events and mediates direct damage to
cellular components sensitive to oxidation, including DNA, phospholipids, and proteins.
Indiscriminate breakdown of local cellular components can also further contribute to BBB
51 leakiness. Overproduction of ROS also leads to reduction of intracellular O2, damaging
normal cell function and causing cell toxicity.
Release of ROS is predominantly by activated microglia and from infiltrating blood
cells (neutrophils, eosinophils, and macrophages). Accumulating evidence points to
activated microglia as a chronic source of neurotoxic factors, including ROS, and ROS is
implicated as a key mechanism of chronic microglial activation.[137] Oxidative stress is
also highly implicated as a major contributor to neurodegenerative disease, such as
Alzheimer’s and Parkinson’s disease. While oxidative stress is not an initiating event for
neurodegeneration, it is a propagating event of cellular injury and inhibition of oxidative
stress may break the cycle of neurodegeneration.[250]
2.4.3.2 Electrode Corrosion
A direct impact of free radical accumulation is oxidation of electrical contacts.
(Figure 11A). Electrode corrosion can impact electrical recording properties by incurring
changes in electrode surface area, electrode impedance, and charge transfer.[63, 132]
Unfortunately, there are only a handful of studies that have analyzed electrode deformation
after exposure to physiological conditions. The Sanchez group has reported severe
structural changes (corrosion, changes to the surface area, damaged insulation) to electrode
recording sites over time surrounding tungsten microwires.[61-63] Furthermore, electrode corrosion may also lead to generation of toxic species. In this study of tungsten microelectrodes, they noted the dominant electrochemical breakdown of tungsten involved generation of tungstic ions, which are moderately toxic.[62] However, the extent of corrosion is material specific. Platinum-based electrodes have shown increased resistance
52 to oxidative corrosion. They exhibit stability in saline/H2O2 environments and have the
antioxidative ability to convert H2O2 species to water.[62] Implanted Pt/Ir electrodes exhibited very little corrosion and surface area changes for up to 6 months of implantation.[57] There should be further exploration of materials that are more resistant to the corrosive physiological environment and can be used in microelectrode device systems. Bench-top methods are currently being developed to enable easier corrosion testing that better mimics in vivo environments.[132]
A
B
Figure 11. Example SEMs of microelectrodes pre- and post-implant. (A) Six month implants show changes in electrode recording site from corrosion. Of note, a crack can be seen forming in implant F2(#5). (B) Delamination and cracks of the insulation layer is apparent in implanted electrodes regardless of time implanted (7 days - 6 months). Image reprinted with permission from [57]
The physiological environment can also cause degradation of passivation layers and
insulating coatings applied to microelectrodes (Figure 11B). Passivation layers, such as
53 silicon dioxide, provide protection from corrosion and serve as a dielectric. However,
traditional passivation layers perform poorly as a barrier in physiological conditions as they
were originally developed for clean, noncorrosive environments. Failures of
microelectrode passivation layers most commonly result from cracking due to mechanical
stress, film defects, and chemical or electrochemical reactions.[251] Therefore, to create
another layer of protection, microelectrodes are more commonly being additionally coated
with insulating polymers, such as Parylene C and polyimide.[217]
Studies have suggested that Parylene C coated microelectrodes had improved
chronic stability and recording longevity[252] and did not adversely affect the neuroinflammatory response.[12] However, damage to the insulating layers is largely
reported. Schmidt et al. reported longitudinal cracks and surface craters in insulating
Parylene C layers which caused electrode failure.[253] They also noted that proper deposition of insulating coatings affects function, with smooth insulating coatings having less failures than electrodes with a rough coating. Therefore, fabrication procedures should be optimized to reduce film defects on electrodes. Insulating polyimide layers have also shown swelling, buckling, and delamination of polyimide layers after 12 days of testing.[251] Prasad et al. reported cracks and peeling of polyimide insulation layers from tungsten microwires as early as 42 days after implantation, with insulation damage being common in chronic implants.[61] These studies postulate that diffusion of water thru the insulating polymer causes fluid accumulation below the layers.
More recently studies have emphasized the need for in vitro methods to test corrosion and degradation of insulating layers. Phosphate buffered saline alone does not mimic the inflammatory environment, which contains a high concentration of degradative
54 enzymes and oxidative species.[254, 255] Researchers have utilized elevated temperatures or hydrogen peroxide to simulate accelerated aging. [62, 256] Takmakov et al. took this a step further and developed a reactive accelerated aging method with reactive oxygen species (ROS) to recreate a harsh aging environment for electrodes. They found that these environments are especially damaging to polymer insulating layers, causing structural damage and changing electrical properties in all types of electrodes.[132] Therefore, strategies to reduce accumulation of toxic levels of ROS surrounding implanted microelectrodes have been developed.
2.4.3.3 Strategies to Reduce Oxidative Stress
Our group was the first to indicate a major role of oxidative stress in microelectrode inflammatory propagation and propose antioxidative therapies to reduce the effects.[14,
15, 36, 247] Administration of antioxidative molecules has proven to be a promising strategy. Most antioxidative molecules have multiple mechanisms of action with which they can affect oxidative stress. They can directly neutralize ROS, increase antioxidative enzyme synthesis, and reduce microglial activation. The first antioxidative strategy demonstrated that two-dose administration of resveratrol, a natural antioxidant derived from red wine, resulted in an improvement in neuronal survival surrounding intracortical microelectrodes for up to four weeks.[15] Further studies are underway to explore the utility of continuous, long-term administration.
Later studies by our group have focused on local delivery of antioxidants to circumvent potential adverse side-effects. Delivery of another natural antioxdaint, curcumin, from PVA polymer probes showed improvement over neat PVA probes at short-
55 time points.[14] However, degradation of the PVA polymer within the brain tissue resulted in increased chronic inflammation. In this dissertation, the effects of local delivery of antioxidants, resveratrol and curcumin, (Chapter 5) from PVAc nanocomposite materials will be discussed. Additionally, for potential long-term local effects, immobilization of a synthetic antioxidant on microelectrode surfaces is presented in Chapter 6.
While antioxidant therapy appears to be beneficial, the mechanism of action needs to be further elucidated. A better understanding of the inflammatory events that are mediated may allow for specific strategies to reduce oxidative stress events. Additionally, evidence suggests that antioxidative therapy alone may not be enough to mediate the inflammatory response. Various other strategies to reduce oxidative stress should also be explored. Antioxidant therapy can be coupled with new materials that are resistant to oxidation to reduce the abiotic effects of ROS. Another strategy to reduce oxidative stress includes reducing inflammatory activation of macrophages/microglia. This can be done by preventing activation of microglia/macrophages or promoting switching of microglia/macrophages into the M2a tissue remodeling phenotype.
2.5 Summary
Intracortical microelectrodes have played a vital role in basic neuroscience and disease-state knowledge. Further, it is a promising tool for rehabilitation and functional control of motor systems. However, current challenges to the technology prevent intracortical microelectrodes from being used to their full potential. A combination of abiotic and biotic events contribute to mechanical, material, and biological failure modes.
Of particular interest, intracortical microelectrode failure due to biological events
56 contributes directly to mechanical and material failures, indicating that better
understanding and mediation of biological events will greatly improve microelectrode
performance.
The neuroinflammatory response to intracortical microelectrodes involves a
complex assortment of cellular and molecular events. Permeability of the BBB has been directly implicated to affect the recording quality of intracortical microelectrodes. There are a variety of strategies currently being investigated to improve BBB integrity. BBB leakiness is connected to a variety of neuroinflammatory events including initial tissue trauma, microglia/macrophage activation, oxidative stress, and neurodegeneration.
Therefore, a single solution may not exist to mediate neuroinflammation and combined approaches are required to target multiple mechanisms at once. Work presented in this dissertation suggests methods to combine strategies to reduce micromotion-induced tissue strain using compliant materials and oxidative stress events using various antioxidative approaches.
57 Chapter 3
Mechanically-Compliant Intracortical Implants Reduce the Neuroinflammatory Response*
*The following chapter is reproduced, with permission by The Royal Society of Chemistry, from: Nguyen JK, Park DJ, Skousen JL, Hess-Dunning AE, Tyler DJ, Rowan SJ, Weder C, Capadona JR. Journal of Neural Engineering, (2014) 11 056014 DOI: 10.1088/1741-2560/11/5/056014
3.1 Abstract
Objective. The mechanisms underlying intracortical microelectrode encapsulation and failure are not well understood. A leading hypothesis implicates the role of the mechanical mismatch between rigid implant materials and the much softer brain tissue.
Previous work has established the benefits of compliant materials on reducing early neuroinflammatory events. However, recent studies established late onset of a disease-like
neurodegenerative state. Approach. In this study, we implanted mechanically-adaptive
materials, which are initially rigid but become compliant after implantation, to investigate
the long-term chronic neuroinflammatory response to compliant intracortical
microelectrodes. Main results. Three days after implantation, during the acute healing
phase of the response, the tissue response to the compliant implants was statistically similar
to that of chemically matched stiff implants with much higher rigidity. However, at two,
eight, and sixteen weeks post-implantation in the rat cortex, the compliant implants
demonstrated a significantly reduced neuroinflammatory response when compared to stiff
reference materials. Chronically implanted compliant materials also exhibited a more
stable blood-brain barrier than the stiff reference materials. Significance. Overall, the data
show strikingly that mechanically-compliant intracortical implants can reduce the
neuroinflammatory response in comparison to stiffer systems.
58 3.2 Introduction
Intracortical microelectrodes that provide an electrical interface to the brain are potentially useful to achieve command, control, and feedback in numerous clinical applications. However, the clinical use of such devices is limited due to their inability to consistently record neural signals for extended periods [2]. One of the main reasons for device failure is the body’s natural inflammatory response directed against the implanted foreign object. In order to prevent inflammation and subsequent implant failure, engineers have designed devices with new architectures, bioactive coatings, or from alternative materials [103]. Each of these approaches has demonstrated a clear short-term improvement in the neuroinflammatory response to implanted intracortical microelectrodes.
However, a disconnect remains between the time course in the neuroinflammatory response and failure in the ability to record neuronal action potentials [2, 61, 257]. Two primary hypotheses have been presented to explain this temporal disconnect. The first is that recording inconsistencies are due to mechanical failure of the devices themselves [2,
3]. However, a number of mechanical failure modes can be attributed to the reactive inflammatory environment, including oxidation and corrosion of the device [62, 63].
Additionally, a majority of the studies examining the neuroinflammatory response to intracortical microelectrodes have been limited to timeframes of at most eight to twelve weeks post-implantation.
In an attempt to find a direct correlation between neural inflammation and device failure, several groups have begun exploring longer time points (for example, 16 weeks).
Longer studies have resulted in the second leading hypothesis – the development of a late-
59 onset, secondary neurodegenerative, disease-like state at the neural tissue-device interface
[24]. Specifically, some of us have observed that neuronal populations around traditional, rigid silicon implants were greatly reduced early after implantation, show a mid-term recovery, which is however followed by another decline [24]. To date, it is unclear if, or when, the response stabilizes.
In previous studies, we have demonstrated that the mechanical mismatch between current neural devices and neural tissue is a significant contributor to the neuroinflammatory response [20, 23]. We employed a new family of mechanically- adaptive nanocomposites, which change their mechanical properties from a rigid to a compliant state in less than 5 minutes after implantation [22, 210]. The mechanically- adaptive nanocomposites were designed to support functional microelectrode components and enable implantation of highly compliant microelectrodes [208]. For the application at hand, we developed materials that are initially stiff (tensile storage modulus E’ ~5 GPa) to enable insertion into the brain tissue and then soften after implantation upon exposure to physiological conditions (E’~12 MPa) to more closely match the mechanical characteristics of the brain tissue [21, 22]. Initial histological evaluations of these materials demonstrated that at four weeks post-implantation, compliant implants more rapidly stabilize neural cell populations at the device interface than rigid, non-dynamic microwire implants [20]. However, in the light of our recent findings that the neuroinflammatory response to intracortical microelectrodes fluctuates with time, we considered it important to extend our understanding of the neuroinflammatory response to compliant intracortical microelectrodes, particularly looking at time points corresponding to both early and late- onset neurodegeneration. Thus, we report herein on the acute and chronic inflammatory
60 response of mechanically-adaptive neural interfaces and show significant differences to
traditional rigid materials.
3.2.1 Materials and Methods
3.2.2 Intracortical Implants
For in vivo experiments, two different implants were used: (1) stiff poly(vinyl acetate)-coated silicon implants; and (2) mechanically-adaptive poly(vinyl acetate) /
tunicate cellulose nanocrystal (tCNC) nanocomposites (NC), which become compliant
after implantation. Single shank “Michigan” type silicon probes were fabricated in-house
to a thickness of 15 μm, a length of 2 mm, and a shank width of 123 μm. PVAc-coated
silicon implants (1) were created by dip-coating such silicon implants in a solution of PVAc
in hot toluene (10% w/w at 70°C). Implants were dipped in succession twice, allowed to
dry for 30 minutes to deposit a PVAc surface layer with a thickness of ~15 +/- 5 μm
(FIGURE 12). The thickness of the PVAc surface layer was determined using stylus
profilometry (KLA-Tencor P-2 Long Scan Profilometer) at a scan rate of 50 µm/sec and
stylus force of 12 mN. NC implants (2) were created by casting films from a solution of
poly(vinyl acetate) (PVAc) and tCNC in dimethylformamide, as previously reported [21,
22]. The NC had a tCNC content of 15% w/w. The resulting films were custom molded
between sheets in a hot press (Carver, Wabash, IN). NC implants were then fabricated by
laser-micromachining with a direct-write CO2 laser (VLS 3.5, VersaLaser) to a thickness
of ~63 μm, a length of 2 mm, and a shank width of 130-140 μm [208]. All implants were ethylene oxide sterilized with the exception of sixteen week implants, which were UV
61 sterilized before implantation. Previous work has established that sterilization method does
not affect the inflammatory response at chronic time points [258].
Figure 12. Images of Michigan-style silicon probes before (A) and after (B) PVAc coating. Dip-coating deposited a PVAc-surface layer of 15 ± 5µm. Scale bar = 50 µm.
3.2.3 Strain Field Modeling
To further understand the impact of varying implant stiffness and structural
compliance on the surrounding brain tissue we developed a 3D finite-element model to
simulate interfacial strains induced by brain micromotion as described previously [4, 5].
All models were developed using COMSOL Multiphysics (COMSOL, Inc. Burlington,
MA). In brief, the developed models consisted of two components: a single probe shank
(modeled after each of the implants described above) and a surrounding brain tissue block.
Taking advantage of symmetries in the model architecture only one quarter of the implanted region was modeled to facilitate visualization of the induced primary strains in the surrounding tissue. Brain tissue was approximated as linearly elastic with isotropic material properties as described by Taylor and Miller [259]. The impact of a tangential
62 tethering force on an implant fixed to the skull was simulated by applying a 20 μm displacement to the upper surface of the probe’s shank.
The tensile storage modulus of silicon is between 130 and 185 GPa, while the tensile storage modulus of the NC is 5.2 GPa pre-insertion, and 12.7 MPa after insertion
[21, 22]. The NC swells ~70% by volume under physiological conditions. Neat PVAc exhibits a tensile storage modulus of ~1.8 GPa pre-insertion, and ~1-12 MPa after insertion, and swells ~ 3-5% by volume when exposed to physiological conditions [21]. Therefore, based on the law of mixtures for thin film composites [260], the composite modulus for the
PVAc-coated silicon implants is between 49 and 78 GPa, i.e., significantly higher than that of the mechanically-adaptive nanocomposite in the compliant (post-insertion) state.
3.2.4 Animal Surgery
Surgical procedures for device implantation closely followed an established protocol, with only minor changes [24, 261, 262]. Briefly, twenty-six male Sprague
Dawley rats (250 – 300 grams) (Charles River, Spencerville, OH) received implants and were euthanized after three days or two, eight, or sixteen weeks. Eight additional animals were used as age-matched sham controls and did not undergo implantation surgery. Prior to surgery, animals were anesthetized with ketamine (80 mg/kg) and xylazine (10 mg/kg) administered intraperitoneally (IP). Once anesthetized, the surgical area was shaved, and then the animal was mounted on a stereotaxic frame and maintained on isofluorane (0.5-
2%). A single injection of local anesthetic, Marcaine (0.5%), was administered below the incision site subcutaneously (SQ), then the surgical area was scrubbed with betadine and
70% isopropanol for sterilization. Animal body temperature was maintained on a
63 circulating water pad and vitals were monitored using a blood-oxygen and heart rate measurement system (PulseSense, Nonin Medical, Inc,).
For implantation, a one-inch incision was made at midline using a scalpel and the skull was exposed. The surrounding tissue was retracted and a 3 mm hole was created in the skull, manually, using a biopsy punch (P/N #536, PSS Select), approximately 3 mm lateral to midline and 4 mm caudal to bregma. Then the dura was reflected using a 45 degree dura pick. Animals received a PVAc-coated silicon implant in one hemisphere and
a NC implant in the contralateral hemisphere (n=5-7 for each time point), randomizing
which implants were on each side. Implants are assumed to be independent of each other,
as glial scarring has been demonstrated to extend up to 600 µm from the
microelectrode/cortical tissue interface, without affecting the contralateral hemisphere
[263, 264]. All implants were inserted approximately 2 mm deep into the cortical tissue by
hand. Implants were implanted perpendicular to the surface of the brain, to minimize the
footprint of tissue damage, while avoiding larger vasculature. Following implantation,
implants were tethered to the skull using Kwik-sil (World Precision Instruments) and UV-
cured dental acrylic (Fusio- liquid dentin, Pentron Clinical) over the surgical area and skull.
The incision was then closed with 5-0 monofilament polypropylene suture (Henry Schein)
and a triple antibiotic ointment was applied to the incision. Once the animal woke from
anesthesia, meloxicam (5 mg/kg, SQ) and cefazolin (16 mg/kg, SQ) were administered for
potential pain and to prevent infection. Surgical procedures and animal care practices were
performed in accordance with the Louis Stokes Cleveland Department of Veterans Affairs
and Case Western Reserve University Institutional Animal Care and Use Committees
(IACUC).
64 3.2.5 Tissue Processing
At three days, and two, eight and sixteen weeks post-implantation, animals were anesthetized using a mixture of ketamine (80 mg/kg) and xylazine (10 mg/kg) given IP.
Each animal was perfused transcardially with 1X Phosphate Buffer Saline (PBS,
Invitrogen) until the exudate was clear and then fixed with 10% formalin (~300 mL, VWR).
The brain was carefully removed, placed in fresh 10% formalin and stored at 4° C until sectioning. Prior to sectioning, the tissue was cryoprotected in a step-wise gradient of 10%-
20%-30% sucrose (Sigma) in 1X PBS at 4° C, until equilibrium was reached at each step.
After equilibration in 30% sucrose, implants were removed from the brain tissue with forceps, and the tissue was frozen at -80° C in optimal cutting temperature compound
(Tissue-Tek), sliced axially in 20 μm sections at -25°C, and mounted on slides to be stored at -80° C until immunohistochemical labeling.
3.2.6 Immunohistochemistry
Prior to immunolabeling, the tissue was removed from -80° C and equilibrated to room temperature in a humidity chamber to prevent tissue-drying. Optimal cutting temperature compound was removed with three washes in 1X PBS and then re-hydrated in
1X PBS containing 0.1% Triton-X 100 (Sigma) (1X PBS-T) for 15 minutes. Tissue was blocked for one hour at room temperature in goat serum blocking buffer (4% v/v serum
(Invitrogen), 0.3% v/v Triton-X 100, 0.1% w/v sodium azide (Sigma)). Then, the tissue was incubated in primary antibody diluted in goat serum blocking buffer overnight at 4°C.
Primary antibodies used were: mouse anti-neuronal nuclei (NeuN) (1:250, Millipore) for neurons, mouse anti-glial fibrillary acidic protein (GFAP) (1:500, Invitrogen) for
65 astrocytes, rabbit anti-Iba1 (1:250, Wako) for all microglia/macrophages, mouse anti-
CD68 (ED1) (1:100, Chemicon) for activated microglia/macrophages, and rabbit anti- immunoglobulin G (IgG) (1:100, AbD Serotec) for blood brain barrier permeability. After eighteen hours, the tissue was washed six times (5 minutes per wash) in 1X PBS-T, then stained for two hours at room temperature with appropriate Alexa Fluor conjugated secondary antibodies (diluted 1:1000 in blocking buffer) and counterstained with 4',6- diamidino-2-phenylindole (DAPI) to mark total cell nuclei. The tissue was washed another six times (5 minutes per wash) in 1X PBS-T, then tissue autofluorescence was removed with a ten minute incubation in 0.5 mM copper sulfate buffer (50 mM Ammonium Acetate, pH 5.0) (Sigma) [261]. Samples were rinsed with distilled water and mounted with
Fluoromount-G (Southern Biotech).
Immunolabeled slides were imaged using a 10X objective on an AxioObserver Z1
(Zeiss, Inc.) and AxioCam MRm (Zeiss Inc.). For a larger field of view, without compromising resolution, mosaic images were obtained by stitching sixteen 10X images together using MosaiX software (Zeiss, Inc.). Exposure times were held constant for each respective cellular marker analyzed. Following acquisition, unaltered, linearized images were converted to 16-bit tagged imaging files (TIFs) using Axiovision LE (Zeiss Inc.) for minimal image compression to allow for optimal intensity analysis. For clarity in presentation only, images in this report have been pseudo-colored and slightly enhanced to improve visual display.
66 3.2.7 Quantitative Analysis
Following image acquisition, the images were converted to tagged imaging files
(TIFs) for intensity analysis. All images except for NeuN were analyzed using MINUTE,
a custom MATLAB program developed to analyze fluorescent intensity profiles around
the implant [261]. Briefly, the implant hole was manually defined for each image from the
bright field image. Note, this method does not account for any small amount of tissue that
may be removed with the explanted implant. The program then quantified the fluorescent
intensity in 2 μm bins of expanding concentric circles around the implant hole up to 1500
μm away (0 μm is defined as the edge of the hole). Raw fluorescent intensities were
normalized to background, defined as average intensity in the radii between 1000 to 1050
µm from the interface. Normalized fluorescent intensity profiles were obtained for each tissue section. After normalization, for statistical comparisons of each stain for the different implants, the area under the curve from 0 to 50, 50 to 100, 100 to 150, and 150 to 200 μm were obtained using MATLAB.
To quantify neuron populations around the implant site, the implant hole was
manually defined and then expanding concentric circles up to 600 m from the interface
were defined for each NeuN image in Adobe Photoshop. The number of neurons in each
ring was manually counted to obtain the number of neurons per area for each radial
distance. Neuron counts were converted to percent to sham background by normalizing to
neuron counts from age-matched sham animals at each respective time point.
3.2.8 Statistical Analysis
For statistical analysis, the area under the curve for all stains (except for NeuN) was
used. For NeuN, the number of neurons per area was used for analysis. Statistical analyses
67 were run using a general linear ANOVA model in Minitab 16 (Minitab Inc.) to allow for
comparisons between the different implants. For significance, Tukey post-hoc tests were
run for pairwise comparison and significance defined as p<0.05. For each animal, 6 images
across the entire length of the implant were analyzed per stain; two from the top of the
implant, two from the middle of the implant, and two from the tip of the implant. Therefore,
with 5-7 animals per time point, 30-42 slices were imaged per stain. All images were then treated as an independent measurement and averaged with the entire group of images from a single cohort.
3.3 Results
In a comparative study, two implant types that serve as models for intracortical microelectrodes were implanted into the cortex of age-matched male rats. In order to
investigate the role of material stiffness on tissue compliance-induced neuroinflammation,
two types of implants were employed: 1) a polymer-coated silicon implant (PVAc-coated),
designed to have a bulk stiffness close to the silicon implant and a surface chemistry that
is matched to the nanocomposite; and 2) a compliant mechanically-adaptive polymer
nanocomposite. Since the two implants are chemically matched, we will focus our
nomenclature on the defining properties of each implant. Therefore, PVAc-coated implants
will be referred to as “stiff” and NC implants will be referred to as “compliant.” It is
important to note that stress on surrounding tissue can be affected by various implant
design parameters to reduce material compliance, such as modulus, size, or porosity.
Therefore, here, we consciously utilize the term “compliant” instead of “soft” implant.
68 3.3.1 Finite Element Analysis of Tissue Strain
To better understand the influence of the varying stiffness within our two types of implants, we used a 3D finite element model to predict micromotion-induced strains surrounding our objects. The impact of a tangential tethering force on an implant fixed to the skull was simulated by applying a 20 μm displacement to the upper surface of the probes shank. The predictions made on this basis are shown in FIGURE 13. Simulated
micromotion surrounding tethered silicon devices with and without PVAc coating resulted
in elevated strains both at the tip and at the sharp edges of the probe track in the tissue. The
comparison predicts that nearly identical strain fields are exerted on the cortical tissue for
both the bare silicon and silicon implants with a thin PVAc coating. Unlike previous
models that examined a smaller total deflection [4], our model predicts that the greatest induced strain would be near the surface of the cortex rather than at the implants tip. When simulating the nanocomposite implant in the stiff state, it behaves similarly to the bare silicon and PVAc-coated silicon implants. By contrast, the model predicts that the nanocomposite implant in the compliant soft state behaves more like a (hypothetical) implant whose modulus matches that of the brain and induces considerably less strain on the surrounding tissue throughout all depths of the cortex.
69
Figure 13. (A-E) Predicted strain profiles induced by a tangential tethering force on (A) a bare silicon implant (E’ = 200 GPa), (B) a silicon implant with a 15.5µm PVAc coating (E’ = 78 GPa), (C) an implant created entirely of the PVAc nanocomposite material (63µm x 130µm) in the stiff state (E’ = 5 GPa), (D) an implant created entirely of the PVAc nanocomposite material (63µm x 130µm) in the compliant soft state (E’ = 12 MPa), and (E) a hypothetical implant that matches brain modulus (E’ = 6 kPa). Strain profiles are normalized to the maximum induced strain surrounding the uncoated silicon probe. (F-H) Normalized strain profiles extending in the positive y-direction are taken from the brain surface, the implant’s midpoint and implant’s tip (levels are shown as white dashed lines in (A)). No significant differences were seen between the bare and coated silicon implants and stiff NC. Despite the nanocomposite implant having larger dimensions, the model demonstrated that the compliant nanocomposite implant would induce less strain on the surrounding tissue than both the uncoated and coated Si implants along their entire length. Additionally, no significant differences were seen between the compliant nanocomposite and a (hypothetical) implant with a modulus that matches the one of the cortical tissue at the midpoint and tip levels.
3.3.2 Neuronal Nuclei
Theoretical modeling has suggested that neuronal cell bodies must be within the first 50 to 140 µm of the intracortical microelectrode in order to maintain recordings of action potentials from individual neurons [89]. To quantify the number of neurons around
70 the implants, we utilized the NeuN antibody, which selectively stains for neuronal nuclei
[265]. Representative fluorescence microscopy images of NeuN stained tissue can be found in Figure 14. Each image corresponds to either the stiff (Figure 14A,C,E,G) or the compliant (Figure 14B,D,F,H) implant. The number of neurons per area was determined at regular intervals up to 600 µm from the material/tissue interface. The percent to sham was reported after normalizing the neuron counts to those of age-matched sham animals.
At three days post-implantation, both implants exhibited significant neuronal loss
compared to age-matched sham animals, up to 300 μm and 600 μm away from the implant
for stiff and compliant implants, respectively. Further, there were no significant
differences seen between the stiff and compliant implants (Figure 14A-B,I). By two
weeks, loss of neurons was indicated up to 200 μm from stiff implants compared to sham
animals. However, neuronal densities were indistinguishable from sham tissue within only
50 μm from the surface of the compliant implant. Comparing the two implants, significant
improvements in neuronal populations surrounding the compliant implant were seen
between 50 and 200 μm from the implant surfaces (Figure 14C-D,J). However, the neuron densities around the compliant implants remained comparable to the two week time point at eight weeks post-implantation. Significant difference between implant types at two weeks shifted to tissue volumes further from the implant surface (from 100-300 μm from implant surface at eight weeks post-implantation, Figure 14E-F,K). Compared to age- matched sham animals, the stiff implants demonstrated significant neuron loss up to 200
μm from the implant-tissue interface, while the compliant implants fully recovered to native levels by 100 μm. Most importantly, by sixteen weeks post-implantation, no neuronal loss was detected for the tissue surrounding the compliant implant, regardless of
71 the distance from the implant. However, the stiff implants revealed appreciable neuronal loss in the most critical area within 50 µm from the implant surface, compared to both sham tissue and tissue surrounding compliant implants (Figure 14G-H, L).
Figure 14. Immunohistochemical analysis of neuronal nuclei (NeuN) around the implant site. Representative fluorescence microscopy images of stained tissue show that neuronal dieback around the stiff PVAc-coated silicon implant was significantly higher than in case of the compliant nanocomposite implant at various time points (3 days (A-B), 2 weeks (C-D), 8 weeks (E-F), 16 weeks (G-H)). The bar graphs show quantification of neuron densities. Statistical analysis identified several regions with significantly different neuron populations, which varied between time points. * Denotes significance between stiff and compliant samples; # Denotes significance between noted implant and age-matched sham control (p < 0.05). Scale bar = 100 µm. The horizontal dashed line represents the 100% neuron level as determined by quantification of age-matched sham animals. Error bars represent standard error.
3.3.3 Glial Cell Markers
Astrocytes
Immunolabelling for GFAP was used to monitor both immature and mature resting or activated astrocytes [266]. At three days post-implantation no appreciable glial scarring can be discerned (Figure 15A-B,I), regardless of the type of implant. However, at all
72 subsequent time points, immunostaining for GFAP showed a dense astrocytic scar around
both the stiff and compliant implants (Figure 15C-H, J-L). At two and eight weeks post-
implantation, elevated levels of GFAP+ scarring around the stiff implant extended to a
distance of ~500 µm away from the implant before the background levels were reached
(Figure 15C-F,J-K). By contrast, scarring around the compliant implant was contained to
an area within a radius of 50 µm surrounding the implant. Significantly less GFAP
expression was detected beyond 50 µm from the implant surface for compliant implants,
compared to chemically matched stiff implants (Figure 15C-F,J-K). By sixteen weeks,
GFAP expression decreased for both implants (Figure 15G-H,L). However, the
normalized fluorescence intensity around the compliant implant still remained significantly
lower than around the stiff implant in the area 0-100 µm away from the implantation site.
Figure 15. Immunohistochemical analysis of the astrocytic scar. Representative fluorescence microscopy images of stained tissue show the formation of a more compact scar surrounding mechanically compliant implants (B, D, F, H), compared to the chemically-matched stiff implants (A, C, E, G) beginning at 2 weeks post-implantation. IHC staining for GFAP+ astrocytes after 2 weeks were seen at higher densities with broader distribution following implantation of the stiff implants compared to the mechanically compliant implants (I-L) (p < 0.05). Scale bar = 100 µm. Error bars represent standard error.
73 Microglia/Macrophages
Microglia/macrophages are a major component of the innate immune response in
the CNS. Microglia/macrophage-released inflammatory factors sustain the innate
immune/inflammatory response and recruit additional cell types. Therefore, we also
investigated the spatial expression of all microglia and macrophages at the implant/tissue
interface. The ionized calcium binding adapter molecule (Iba1) is a cell-surface marker for both resting and activated microglia/macrophages [267].
At three days post-implantation, increased Iba1+ immunoreactivity was observed out to a distance of 1 mm away from the implantation site (Figure 16A-B,I). This could be due to either increased cell number or cellular activation causing up-regulation of Iba1
[267]. The microglia/macrophages condensed around the implants by two weeks post- implantation (Figure 16C-D,J). For both 3 days and 2 weeks post-implantation, no significant differences were seen between the stiff and compliant implants. However, eight weeks post implantation, the accumulation of microglia/macrophages around the compliant implants has become significantly more compact, as noted by a significant decrease in
Iba1+ staining in the area between 50 to 150 µm away from the implant (Figure 16E-F,K).
Sixteen weeks post-implantation, an intense, circular zone of upregulated Iba1 immunoreactivity was observed around the stiff implant, whereas implant-mediated up- regulation of IBA1+ cells subsided to native levels and very little accumulation of microglia and macrophages were present around the compliant implant (Figure 16G-H).
Normalized average intensity profiles depict a higher peak intensity and broader distribution of Iba1+ immunoreactivity for the stiff implants compared to the minimally up-regulated expression surrounding the mechanically compliant implant (Figure 16L).
74 Notably, after sixteen weeks, quantification of the Iba1 expression levels for the two
conditions were significantly different over a distance of 75 μm from the implant, where a
much larger distribution was seen in stiff implants than for compliant implants.
Figure 16. Immunohistochemical analysis of total microglia/macrophages populations (Iba1). Representative fluorescence microscopy images of stained tissue show no effect of implant compliance at acute time points (3 days (A-B,I), 2 weeks (C-D,J)), but a significant increase in Iba1 immunoreactivity around stiff implants at chronic time points (8 weeks (E-F,K), 16 weeks (G-H,L)) (p < 0.05). Scale bar = 100 µm. Error bars represent standard error.
While Iba1 labels both resting and activated microglia/macrophages, CD68 is more
regularly used to identify activated microglia/macrophages. CD68 is a cytoplasmic antigen
found in the lysosomal compartment of activated microglia and macrophages [268]. As seen with the Iba1 staining, at three days post-implantation there was activation of microglia/macrophages around both implants up to a distance of 1 mm away from the implantation site with no significant difference between implant types (Figure 17A-B, I).
75 However, after two weeks, the distribution of CD68+ immunoreactivity was more compact
for both implants. Furthermore, there was a significant decrease in CD68+ staining
surrounding the compliant implants in the area 0-100 µm away from the implant (Figure
17C-D,J). At eight weeks post-implantation, there was a large drop in CD68+ staining surrounding both implants (Figure 17E-F,K). However, there was significantly less
CD68+ staining surrounding the compliant implant in the range 50-200 µm from the
implantation site, suggesting a broader CD68+ cell distribution around the stiff implants.
Representative images of CD68+ staining after sixteen weeks show very little
microglia/macrophage activation at either interface (Figure 17G-H). However, normalized
intensity profiles do exhibit significantly more CD68 immunostaining at distances 0 to 50
µm away from the implant for the stiff material compared to the compliant implant (Figure
17L). Further, the mechanically compliant implant lacked any appreciable presence of
activated microglia or macrophages at the implant/tissue interface. Contrary, the stiff
implant was accompanied by persistently activated microglia and macrophages at sixteen
weeks post-implantation.
76
Figure 17. Immunohistochemical analysis of CD68, a cellular marker for activated microglia/macrophages. Representative fluorescence microscopy images of stained tissue show a distinct benefit of mechanically compliant implants (B, D, F, H), compared to the chemically matched stiff implants (A, C, E, G). Starting at two weeks post implantation, IHC staining and fluorescent quantification showed increased expression of CD68+ tissue surrounding the stiff implants (I-L) (p < 0.05). Scale bar = 100 µm. Error bars represent standard error.
Blood Brain Barrier Integrity (IgG)
Insertion of the implant into the cortex causes blood brain barrier (BBB) damage and leakage of blood-derived cells and serum proteins into the cortical tissue. Additionally, inflammatory events, resulting from neurological disease or device implantation, can cause the BBB to remain open [269]. The integrity of the BBB can be correlated to the amount of IgG present within the surrounding tissue [13, 102]. Note that sodium fluorescein or
Evans Blue albumin staining were not considered for this study as alternatives to IgG staining. While both have been shown to be valuable markers for blood-brain barrier permeability [270], either method limits the utility of the tissue for further histological evaluation, and typically requires additional dedicated animals. Therefore, to investigate
77 BBB integrity, the amount of IgG present at the implant site was examined. Three days
post implantation, the IgG accumulation around both implants is comparable, extending up
to 1 mm away from the implantation site (Figure 18A-B,I). After two weeks, there is a
large increase in IgG present around the implant for both compliant and stiff implant types.
However, there is significantly more accumulation of IgG around the stiff implant in the
range of 100-500 µm away from the implantation site (Figure 18C-D,J), compared to the
compliant implant. After eight weeks, the integrity of the BBB is partially reestablished.
IgG infiltration is reduced for both implants with the stiff implants still maintaining higher
IgG levels in the range 100-500 µm away from the implantation site (Figure 18E-F,K).
Representative fluorescence microscopy images of stained tissue showing IgG infiltration
sixteen weeks post implantation reveal that the stiff implants prevented the complete healing of the BBB, as indicated by the diffuse, intense IgG staining of the tissue surrounding the implant surface (Figure 18G). On the other hand, the mechanically
compliant implants show significantly less deterioration of the BBB integrity (reduced IgG
detection) around the implant site sixteen weeks post-implantation (Figure 18H). Further,
the stiff implants had a much higher peak intensity and broader infiltration of IgG into the
brain tissue (300 μm) in comparison to the much softer compliant implant (p<0.001)
(Figure 18L).
78
Figure 18. Immunohistochemical analysis of IgG staining shows that the blood brain barrier integrity is improved for compliant implants, compared to stiff implants. Representative fluorescence microscopy images of IgG immunoreactivity show no difference three days post implantation (A, B). At all subsequent time points, a marked increase in IgG intensity was observed around the stiff implant (D, F, H) compared to the compliant implant (C, E, G). Quantification of the fluorescent intensity indicates significant differences between implants at 2, 8, and 16 weeks post implantation (J-L). (p < 0.05). Scale bar = 100 µm. Error bars represent standard error.
3.4 Discussion
The base materials used for traditional microelectrodes are significantly stiffer and
less compliant than cortical tissue (Figure 13). Starting with Goldstein and Salcman’s work
in 1973, several groups have suggested that motion of the brain with respect to the
microelectrode may induce damage to the surrounding tissue [4, 10, 20, 23, 163, 192, 205].
Recently, Moshayedi et al. reported that astrocytes and microglia respond to surface
stiffness and under acute conditions show enhanced foreign body response to stiff implants
[25]. Therefore, this study was designed to investigate the temporal effects of the stiffness of the microelectrode implant on the neurodegenerative inflammatory response. In order to keep the study as simple as possible, experiments were not conducted with functional
79 microelectrodes, but with simple shanks made from base materials with different
properties.
Utilizing the same nanocomposite materials as we present here, we previously
demonstrated in an initial study that at four weeks post-implantation, neuronal densities were higher surrounding compliant implants than stiff microwires [20]. However, the beneficial effects of material modulus appeared to depreciate by eight weeks post-
implantation. While these results may indicate that mechanical mismatch has limited
effects on the chronic inflammatory response, the Tresco lab has observed that microwire implants produce a reduced inflammatory response compared to Michigan-style implants
[64]. Previous work also has indicated that mechanics and architecture can each independently manipulate the inflammatory response [8, 11-13]. Additionally, we have previously reported that rigid silicon implants show a mid-term recovery phase in neuron densities [24]. Therefore, the lack of difference at eight weeks may be due to differences in implant architecture, or a recovery phase.
To verify this interpretation, we here undertook a comprehensive evaluation of the neuroinflammatory response to nanocomposite implants using Michigan-style controls. In order to investigate the role that implant compliance has on the neuroinflammatory response, implants fabricated from a previously reported mechanically adaptive material, which becomes compliant upon exposure to physiological conditions, were implanted into the cortex of rats for three days, two weeks, eight weeks or sixteen weeks. PVAc-coated silicon microelectrodes were implanted in the contralateral hemisphere to serve as a stiff control with a chemically-matched surface. Findings from this study demonstrate that mechanically compliant intracortical implants elicit a reduced neuroinflammatory
80 response, resulting in significantly less neurodegeneration and a more robust BBB than
chemically matched stiff implants at two, eight, and sixteen weeks, but not 3 days post- implantation. Interestingly, in contrast to our previous initial study, few statistically significant differences were observed between compliant and stiff PVAc implants at early time points. Additionally, various differences in glial scar immunoreactivity between this study and the initial study were observed at eight weeks post-implantation. These discrepancies could be due to differences in the controls, surgical procedures, analysis methods, or other factors. Several groups have been working to overcome such potential inconsistencies by creating standardized techniques. Unfortunately, a consensus on methodology has not been reached by the field. Therefore, the discussion here will focus on placing the current results into context. The authors also note that the compliant implants had larger dimensions than the stiff controls due to fabrication limitations of the polymer nanocomposite materials. However, despite this difference in size, the compliant implants still significantly reduced neuroinflammatory events.
A leading hypothesis for the loss of electrical contact with implanted microelectrodes is the decrease in viability of neuron populations around the implant.
Therefore, we first investigated the neuron density around each implant to determine if there was a benefit to using a softer, more compliant implant on neuronal survival. Glial scarring has been demonstrated to extend beyond 600 µm from the microelectrode/cortical tissue interface [263]. However, neuronal ‘die back’ is typically understood to extend only a few hundred micrometers or less away from the implant [105, 145, 271]. Compared to age-matched sham animals, stiff implants exhibit significant reduction in neuron densities at all of the time points investigated (Figure 14, #p<0.05). The loss of neurons surrounding
81 the implant may indicate neurodegeneration or mechanical movement of the cells. At three
days post-implantation, both the compliant and stiff fail to demonstrate statistically
different neuronal populations when compared to each other. However, neuronal dieback
extends out 300 µm further (to 600 µm) for the stiff implants than for the compliant
implants. The increased tissue volume in which significant neurons densities are lost for
stiff implants suggests that the compliance of the implant plays a role in neuronal loss, even
during the acute phases of wound healing. However, our results further suggest that at such
acute time points, the mechanics of the implant may not be able to completely abolish
neuronal loss around the devices.
Starting at 2 weeks post-implantation, the compliant implants exhibited increased neuronal populations compared to their stiff counterparts. While at two and eights week neuronal populations gradually increased surrounding both implants, the number of neurons around the compliant implants improved significantly compared to the stiff implants, starting at two weeks. Additionally, after sixteen weeks, examination of neuron populations surrounding the two probes indicated a significant advantage of the mechanically compliant implant within the first 50 µm from the implant. Only the compliant implant supported full recovery of neuronal populations near the implant surface, which is critical to obtain single unit recordings [89].
Next, we investigated the primary markers of the glial scar, known to rapidly encapsulate implanted microelectrodes [2, 103, 109, 257]. Astrocytes form a major component of the glial scar around the electrode. This cellular encapsulation can result in increased impedance around the electrode, making recordings more difficult to maintain
[94]. However, some glial scar formation is desirable to stabilize and anchor the electrode
82 in the brain [5]. Additionally, microglia/macrophages mediate the inflammatory and
immune response to minimize bacterial/viral invaders [272], as well as infiltrating blood
proteins within the CNS [273]. Therefore, many studies of the tissue response to
intracortical electrodes have focused on microglia/macrophage activation in response to
indwelling implants [103].
Like NeuN staining, the quantified intensity of GFAP, Iba1, and CD68 stained
tissue around the implant was the same for both implants at three days post-implantation
(Figure 15-Figure 17A,B,I). Similarities in the acute inflammatory responses to both
implants suggest that wound healing events may dominate the tissue reaction at this early
time point. Much is known about the role of early wound healing events in injury and other
device implantation models [274, 275]. However, when considering intracortical microelectrodes, comparably few studies have explored the early wound healing events after implantation [276]. The majority of what is known about the brain’s response to implanted microelectrodes comes from end-point histological studies focused on later time points that range from ~1-24 weeks post-implantation. However, Biran et al. compared various markers of neuroinflammation in chronically implanted animals to animals that received only a stab wound injury [70]. The authors found that chronic neuroinflammation does not accompany stab wound injuries made with identical recording devices. Therefore, the lack of difference in the neurodegenerative and neuroinflammatory response at three days post-implantation is somewhat expected.
By two weeks, the GFAP+ staining surrounding compliant implants is significantly less dense than that displayed in the tissue surrounding stiff control implants. Increased astrocytic reactivity around stiff implants, which are predicted to have increased tissue
83 strain, are consistent with findings from Biran et al. Biran controlled tissue strain through
the comparison of a tethered versus untethered electrode and established the connection
between increased tissue strain and increased tissue reactivity [145]. Therefore, our results also suggest that decreased prevalence of an astrocyte-rich glial scar may be a consequence of the mechanically compliant implants producing less strain on the tissue. Similarly, at later time points, the glial scarring around the compliant implant has lower peak intensity and is more compact. Interestingly, there appears to be a spatial relationship between the ranges of increased neuronal populations and decreased glial scarring surrounding the compliant implants. Specifically, the glial scar peaks until about 100 µm away from the compliant at two and eight weeks post-implantation, which corresponds to regions where the compliant implant also preserves significantly higher neuronal densities. Together, the decreased glial scarring and increased neuronal populations surrounding compliant implants promises to prevent increased impedance often associated with more stiff implants, thus allowing for improved neural recordings from the increased population of remaining neurons.
Our assessment of microglia/macrophages around the compliant implant showed reduced Iba1 (all microglia/macrophages) (Figure 16) and CD68 (activated microglia/macrophages) immunoreactivity (Figure 17) compared to the stiff implants beginning two weeks post-implantation. While at two weeks there is no difference in total microglia/macrophages, there is a significant increase in the activation of microglia/macrophages (Figure 17C,D) surrounding the stiff implant. This suggests that material modulus may affect the activation of microglia/macrophages but does not prevent accumulation of cells at the site of injury. After sixteen weeks, activated
84 microglia/macrophage populations surrounding compliant implants were equivalent to
background sham controls, regardless of the distance from the implant (Figure 17H,L).
Interestingly, CD68 staining for activated microglia/macrophages was in excellent
agreement with NeuN staining, suggesting a correlation between microglia activation and
neuron viability as an effect of implant stiffness. Several studies have indicated a large
number of activated microglia cells on the implant surface long after the initial wound
healing response is complete. Activated microglia and macrophages release a plethora of
pro-inflammatory/cytotoxic soluble factors that can damage healthy bystander cells and
the surrounding tissue [117, 277-280]. Pro-inflammatory cytokines can have direct toxic
effects on neurons and oligodendrocytes, while reactive oxygen species and chemokines
are involved in opening the BBB and recruiting new macrophages to sites of injury and
inflammation [15]. Several recent studies have also provided further support for the
predominant role of macrophage-released factors on the recording function and the
neuroinflammatory response. For example, Karumbaiah et al. have shown that gene
expression for various pro-inflammatory soluble factors is up-regulated in tissue
surrounding poorly performing microelectrodes [26]. In addition, Skousen et al. have
suggested that macrophage-secreted soluble factors may be critical in both propagating as
well as shaping the response to traditional microelectrode designs.[13]
Gilletti and Muthuswamy have demonstrated that implants in the cortex of rats can move on the order of 10-30 µm due to pressure changes during respiration, and 2-4 µm due to vascular pulsatility.[168] They concluded that such significant micromotion can impact a wide variety of acute and chronic procedures involving any brain implant. Since the brain is highly vascularized [281], continual pulsatile activity of implants may also lead to the
85 repetitive trauma to micro-vessels. Vascular damage, combined with increased release of
reactive oxygen species and chemokines, can cause a constant influx of blood-derived cells
and neurotoxic serum proteins [15]. Irregularities in the stability of the BBB functionality
are increasingly investigated in implant-associated disease-like states, due to the
neurodegenerative role of several plasma proteins.
Therefore, to visualize BBB permeability, we stained for IgG, a common blood
derived protein not typically found in neural tissue [282]. Interestingly, maximal IgG+
staining was not apparent until two weeks post-implantation. Saxena et al. also investigated a three day time point and saw minimal IgG staining [11], which was attributed to brain edema found to be maximal at three days post-trauma. Starting at two weeks post- implantation, both compliant and stiff implants exhibited a significant increase in accumulated IgG within the cortical tissue. However, the stiff implant facilitated a much broader distribution of IgG+ tissue. Notably, at sixteen weeks post-implantation, there were still considerable amounts of IgG present at the injury site for both compliant and stiff implants. Here, staining for IgG was in agreement with the same volume of tissue supporting increased levels of CD68+ cells and the lowest levels of NeuN+ cells. However, it remains unclear if the compliant material facilitates less initial damage to the BBB, or if the softer material has the same initial damage but allows for a more complete repair due to reduction in micromotion induced damage.
As indicated previously, the compliant material swells considerably while becoming compliant, creating an additional factor to consider in this study. Skousen et al. investigated the utility of hydrogel coatings on microelectrodes and noted that they can create “cytokine sinks” [283]. While absorption of pro-inflammatory molecules is possible,
86 the porosity of our material most likely impedes the uptake of proteins during swelling, and was not investigated here. Additionally, aqueous uptake creates challenges for engineering functional microelectrodes for neural recordings. The Tyler and Zorman labs are currently developing microsystems with our mechanically dynamic materials, in which the compliant material serves as a substrate to replace rigid silicon. Their design utilizes insulating materials to isolate the water soaked compliant material from the electrodes
[208].
From the above described experiments, we have demonstrated that the more compliant implants demonstrate a significant reduction in several markers of glial scarring,
BBB stability, and an increased neuron density (Figure 19) compared to the stiff, chemically matched control implants. To date, there have been limited efforts to directly measure the strain on cortical neural tissue following device implantation. While LaPlaca and colleagues have demonstrated in vitro that strain can cause reactive gliosis and neuron loss [284], experiments are difficult to reproduce in vivo. Interestingly, Bellamkonda and colleagues have reported that low magnitude strain on astrocytes and microglia induced up-regulation of IL-36Ra as early as three days post-implantation, which was correlated with increased neuronal apoptosis [26]. In agreement with both LaPlaca and
Bellamkonda’s work, the acute study by Moshayedi et al. reported that astrocytes and microglia both respond to surface stiffness, facilitating differences in the foreign body response [25]. However, the effect seen by Bellamkonda was less pronounced by twelve weeks post-implantation, presumably due to strain shielding of cortical tissue by more robust glial scars. Supporting this hypothesis, Muthuswamy and coworkers [6] recently reported on changes in the mechanical properties in the brain-electrode interface during
87 chronic implantation. Using a customized jig, a microelectrode was immobilized onto the
skull and chronic force measurements obtained with a tension/compression load cell at
various time points. In vivo measurements of strain and micromotion stresses surrounding microelectrode implants indicated changes in estimated elastic modulus of the surrounding brain tissue.
The data presented here suggest that the effects of mechanical mismatch between compliant neural tissue and the implanted microelectrode may play a role in neuroinflammation and indicate a dominant role of the mechanical stiffness of the implant during chronic conditions. It is critical that future studies exploit cutting edge methods developed by groups such as the Muthuswamy Lab, using novel materials, such as those used here, in order to better understand how device stiffness/compliance results in changes in the duration and magnitude of strain and micromotion stresses surrounding microelectrode implants.
88
Figure 19. Schematic representation of the tissue response around stiff (A) and compliant (B) implants. Stiff implants induce increased gliosis, BBB permeation, and neurodegeneration in comparison to compliant materials. (C) Graphical representation of the peak intensity of specific inflammatory markers over time, which shows a distinct advantage of compliant materials at chronic time points.
89 3.5 Conclusions
The results presented in this study support the hypothesis that late-onset neuron loss
may be attributed to continuous mechanical damage from the mechanical mismatch of stiff implants. While previous studies have indicated the advantage of compliant implants in reducing acute neuroinflammatory responses, here we gathered new insights on the long- term effects of compliant implants. Specifically, we found that while wound healing events dominated acute inflammatory events, tissue surrounding the compliant implants at later time points demonstrated a significant reduction in the expression of glial scar formation
markers, and a significant increase in neuron density compared to stiffer implants. Of note,
neuronal viability appeared to be related to low microglial activity and reduced BBB
permeability. Further work is required to pinpoint the exact mechanism and causal
relationship between mechanical, chemical, inflammatory events, and the role of BBB
stability. However, we have demonstrated that at extended periods post-implantation,
mechanically compliant materials present a significant advantage in stabilizing the neural
interface. It will be critical to next explore how these changes in neuroinflammatory events
directly relate to tissue strain and how functional neural electrodes derived from compliant
substrates mediate the stability of neural recordings and/or the function of other neural
implants that suffer from decreased device performance due to an encapsulating glial scar.
3.6 Acknowledgements
This work was supported by the Department of Biomedical Engineering at Case
Western Reserve University through laboratory start-up funds (Capadona). Additional
funding for this research program was provided by Rehabilitation Research and
90 Development, Department of Veterans Affairs: Career Development Award (B6344W,
Capadona), Merit Review (B7122R, Capadona), the Advanced Platform Technology
Center (C3819C), and Presidential Early Career Award for Scientists and Engineers
(PECASE, Capadona); the National Institutes of Health: National Institute of Biomedical
Imaging and Bioengineering Integrated Engineering and Rehabilitation Training Program
(5T32EB004314-15), National Institute of Neurological Disorders and Stroke (R01-
NS082404-01A1, Capadona and R21-NS053798, Tyler); National Science Foundation:
(CBET-0828155, DMR-0804874 and DMR-1204948, Weder and Rowan); the Swiss
National Science Foundation (NRP 62: Smart Materials, Nr. 406240_126046, Weder); and
the Adolphe Merkle Foundation (Weder). The authors also thank L. Hsu for providing the
dynamic materials. None of the funding sources aided in the collection, analysis and
interpretation of data, in the writing of the report, or in the decision to submit the paper for
publication. The authors have no conflicts of interest related to this work to disclose.
91 Chapter 4
Compliant Intracortical Implants Reduce Strains and Strain Rates in Brain Tissue In Vivo*
*The following chapter is reproduced, with permission, from: Sridharan A, Nguyen JK, Capadona JR, Muthuswamy J. Journal of Neural Engineering, (2015) 12 036002 doi:10.1088/1741-2560/12/3/036002
4.1 Abstract
Objective. The objective of this research is to characterize the mechanical interactions of
(1) soft, compliant and (2) non-compliant implants with the surrounding brain tissue in a
rodent brain. Understanding such interactions will enable the engineering of novel
materials that will improve stability and reliability of brain implants.
Approach. Acute force measurements were made using a load cell in n=3 live rats, each
with 4 craniotomies. Using an indentation method, brain tissue was tested for changes in
force using established protocols. A total of 4 non-compliant, bare silicon microshanks, 3
non-compliant polyvinyl acetate (PVAc)-coated silicon microshanks, and 6 compliant,
nanocomposite microshanks were tested. Stress values were calculated by dividing the
force by surface area and strain was estimated using a linear stress-strain relationship.
Micromotion effects from breathing and vascular pulsatility on tissue stress were estimated
from a 5 sec interval of steady-state measurements. Viscoelastic properties were estimated
using a second-order Prony series expansion of stress-displacement curves for each shank.
Main results. The distribution of strain values imposed on brain tissue for both compliant
nanocomposite microshanks and PVAc-coated, non-compliant silicon microshanks were
significantly lower compared to non-compliant bare silicon shanks. Interestingly, step-
indentation experiments also showed that compliant, nanocomposite materials
92 significantly decreased stress relaxation rates in the brain tissue at the interface (p<0.05) compared to non-compliant silicon and PVAc-coated silicon materials. Further, both
PVAc-coated non-compliant silicon and compliant nanocomposite shanks showed significantly reduced (by 4-5 fold) stresses due to tissue micromotion at the interface.
Significance. The results of this study showed that soft, adaptive materials reduce strains and strain rates and micromotion induced stresses in the surrounding brain tissue.
Understanding the material behavior at the site of tissue contact will help to improve neural implant design.
4.2 Introduction
Intracortical microelectrodes are used typically to record the electrical activity of individual or ensembles of neurons [285, 286]. Neuroscientists implant microelectrodes into both healthy and diseased animal models to gain an understanding of neuronal connectivity and the progression of neurodegenerative diseases [287-290].
Microelectrodes can also be used as the biotic-abiotic interface for brain computer interfaces in rehabilitative applications [80-83]. Unfortunately, intracortical microelectrode technology is limited by the inability to record high-quality signals reliably over time [97, 99, 201].
Increasing evidence suggests a dominant role of neuroinflammatory events in mediating mechanical, material and biological failure modes in intracortical microelectrodes [3, 57, 61]. Several groups have suggested a major role of microglia and infiltrating blood derived macrophages in mediating neuroinflammatory events following microelectrode implantation, due to many stimulating factors [11, 70, 102, 291]. A leading
93 hypothesis also revolves around the mechanical mismatch surrounding implanted devices
[85, 163]. Studies suggest that micromotion and tethering forces can cause damage of
neural tissue surrounding implants [70, 292, 293]. For example, cortical tissue in the rat brain can displace 2-4 µm due to vascular pulsation and 10-30 µm due to breathing [168].
Self-propagation of either the cortical tissue or endothelial barrier damage can exacerbate
the foreign body response to the microelectrodes and prevent wound healing events [15,
24]. Additionally, in silico studies have predicted that non-compliant microelectrodes
composed of high modulus materials induce strain on the surrounding soft brain tissue
while materials more closely matching the brain modulus significantly reduced strain on
surrounding tissue [4]. In vitro studies have also indicated that induced strain can cause
activation of pro-inflammatory pathways [26, 284]. Recently, Moshayedi et al. showed that CNS glial cells become activated and release pro-inflammatory molecules in response to stiff substrates [25]. Therefore, substantial efforts from several groups have gone into the development of techniques to minimize both device micromotion and the mechanical mismatch between non-compliant implants and compliant brain tissue [16, 145, 185, 190,
294, 295]. Multiple groups, including the Capadona group, have demonstrated decreased neuroinflammatory responses to compliant implants [10, 19, 20, 296, 297]. However, experimental evidence confirming that compliant microelectrodes reduce strain and micromotion induced forces in brain tissue in vivo is lacking.
A recently reported method from our group measures the mechanical properties of the biotic component of the brain-electrode interface, surrounding non-compliant stainless steel microelectrode implants [6]. Specifically, we found that the estimated elastic modulus in the surrounding brain tissue fluctuates and evolves over implantation time ranging from
94 a median value of 7.5 kPa on day 1 to 35.2 kPa after 4 weeks, indicating a progressively
increasing stiffness in the brain tissue at the microelectrode-tissue interface likely due to ongoing repair and inflammatory responses in the brain tissue. Beyond 4 weeks, the estimated elastic modulus of the surrounding tissue decreases to median value of 19.6 kPa.The estimated shear modulus decreases from 28.7 kPa (4 weeks) to 7.9 kPa at 6-8 weeks, suggesting replacement or compaction of the stiffer tissue at the interface [44].
Mechanical stresses due to tissue micromotion are also reported to change significantly with implantation time along with the dynamic material property changes in brain tissue.
Typical micromotion induced effective stress amplitudes range from 0.2-0.4 kPa at day 1 of implantation to 2.6 kPa observed 4 weeks after implantation and decreasing to 0.07-0.29 kPa beyond 6 weeks of implantation (comparable to stress levels at day 1). Considering the continuously active imposition of mechanical stress on the tissue due to the mechanical mismatch between the implant and brain tissue, development of mechanically compliant substrates and coatings that also have the ability to absorb micromotion-induced stresses may mitigate the severity of scarring around the microelectrode. In this current study, we utilized the above reported novel method for mechanical characterization of brain tissue
[6] to carefully assess, for the first time the mechanical stresses and strains induced in the brain tissue at the interface of a novel compliant, nanocomposite implant developed by our collaborator (Capadona lab) [22]. We then performed a quantitative comparison of the above mechanical stresses and strains with two conventional non-compliant shanks – a silicon shank and a polymer-coated silicon shank, to control for the difference in the surface chemistries of the implants.
95 4.3 Materials and Methods
4.3.1 Electrode fabrication & characterization
For in vivo experiments, three different implants were used: (1) bare silicon
implants (non-compliant); (2) poly(vinyl acetate)-coated (PVAc) silicon implants (PVAc-
coated non-compliant); and (3) mechanically-adaptive poly(vinyl acetate)/tunicate
cellulose nanocrystal (tCNC) nanocomposites (compliant) implants, which become
compliant after implantation. Single shank “Michigan” type silicon probes (1) were
fabricated to a thickness of 15 μm, a length of 2 mm, and a shank width of 123 μm. PVAc-
coated silicon implants (2) were created by dip-coating silicon implants in a solution of
PVAc in hot toluene (10% w/w at 70°C). Implants were dipped in succession twice, allowed to dry for 30 minutes to deposit a PVAc surface layer with a thickness of ~15 +/-
5 μm [19]. Compliant implants (3) were created by casting films from a solution of PVAc and tCNC in dimethylformamide, as previously reported [21, 22]. The nanocomposite had a tCNC content of 15% w/w. The resulting films were custom molded between sheets in a hot press (Carver, Wabash, IN). Compliant implants were then fabricated by laser- micromachining with a direct-write CO2 laser (VLS 3.5, VersaLaser) to a thickness of
~127 μm, a length of 2 mm, and a shank width of 130-140 μm [208]. All implants were ethylene oxide sterilized.
96
Figure 20. (A) The experimental setup to measure force-displacement curves in brain tissue in vivo is illustrated. A load cell (10g, Futek, Irvine, CA) is attached to a stereotactic frame perpendicular to the craniotomy of an animal immobilized in the stereotactic frame. The attached shank (bare silicon, PVAc - coated silicon, or nanocomposite) with load cell is lowered into the craniotomy using an FHC microdrive (Bowdoin, ME) at 100 µm/sec. (B) Illustration shows the approximate placement of 4 craniotomies in the animal. Craniotomies were covered in gelfoam wetted with 0.9% saline to maintain hydration in between force measurements. In this representative picture, a PVAc-coated silicon shank is being placed at 1 mm depth in craniotomy #4. (C) Shown is a closeup of a craniotomy with dura (left) and after reflection of dura (right) showing minimal damage to vasculature. (1) marks the spot on brain tissue where minimal superficial vasculature is present and a bare silicon probe shank is inserted for force measurement. Note the silicon shank is in reality 90° (perpendicular) to the brain surface. The slanted view is due to the camera angle. Scale bar in (B) and (C) represents 1 mm.
4.3.2 Animal surgery and force measurements
All animal procedures were carried out with the approval of the Institute of Animal
Care and Use Committee (IACUC) of Arizona State University, Tempe. The experiments were performed in accordance with the National Institute of Health (NIH) guide for the care and use of laboratory animals (1996). All efforts were made to minimize animal suffering and to use only the number of animals necessary to produce reliable scientific
97 data. The adult Sprague-Dawley rats (250-300 g) were induced using a mixture of (50
mg/ml) ketamine, (5 mg/ml) xylazine, and (1 mg/ml) acepromazine administered intramuscularly with an initial dosage of 0.1 ml/100 g body weight. The maintenance dose contained a mixture of 50 mg/ml ketamine and 5 mg/ml xylazine. Updates were given at a dose of 0.05 ml/100 g body weight based on the toe-pinch test. The rat was attached to a stereotaxic frame (Kopf Instruments, Tujunga, CA, USA). After the skull was exposed, 4 craniotomies ( 2.0 mm diameter) were drilled for the microelectrode probes with (a) 2 center points being∼ 2.5 mm lateral and medial to the midline and 3 mm posterior to the bregma and (b) 2 center points being 2.5 mm lateral and medial to midline and 5.0 mm posterior to the bregma (Figure 20a,b). After the bone chips were carefully removed with a micro-dissection scissor, the dura was carefully incised to allow for microelectrode insertion with minimal disturbance to the observed vasculature (Figure 20c). All craniotomies were made prior to the force measurement and the exposed brain tissue was hydrated with 0.9% saline and maintained free of debris by keeping it covered with wet gelfoam. The probe-shanks were pre-mounted onto the load cell and inserted into the brain at a rate of 100 μm/s using a FHC Microdrive (Bowdoin,ME). The microelectrode probe was implanted to a depth of 1 mm in the cortex for all experiments and subsequently moved an additional 0.5 mm after 30 min. Forces experienced by the microelectrode during the insertion/penetration process were monitored using a precision load cell.
4.3.3 Force Measurement
Dynamic and equilibrium forces on selected microelectrode shank-probes were measured in vivo in adult Sprague-Dawley rats (250-300 g) using a 10g load cell (Futek,
98 LSB210, Irvine, CA). A total of 3 rats were used for all acute measurements. To test
individual microshank indentation and force response, a total of 4 craniotomies for each
rat were made. Force measurements for bare silicon (Si-bare) (n=4), PVAc-coated silicon
(n=3), and nanocomposite (n=6) shanks were tested in randomized order with each craniotomy used for a single shank, with the exception of two bare silicon shanks that shared the same craniotomy but the measurements were spaced at 2.0 mm apart.
Microelectrode shanks were glued onto a custom mount with Loctite™ control gel super glue. Forces on the microelectrode were measured during the following sequence of steps— 1) implant microelectrodes to 1 mm depth at 100 µm/sec (phase 1), 2) hold
microelectrode in place for 30 min to reach equilibrium, 3) move the microelectrode a
further 0.5 mm down at 100 µm/sec (phase 2). Forces were typically registered as negative
values by the load cell due to compression and tissue resistance during movement of the
microelectrode into the brain. Force measurements were sampled at a frequency of 54 Hz.
In this study, we utilized a loading rate of 100 µm/sec for consistent dynamic force
measurements from small dimension (~100 µm diameter) nanocomposite shanks. To test
whether compliant nanocomposite shanks withstand force-indentation test methodology, nanocomposite shanks were tested using the same protocol in 0.5% (w/v) agarose gels
(elastic modulus of ~8 kPa). Agarose gels have been extensively used in literature as mimics of brain tissue [298]. The force curve of a compliant nanocomposite microelectrode as it penetrates a 0.5% agarose gel mimic at 100 µm/sec is shown in Figure 21. Based on
the test protocol, the nanocomposite shank is held stationary for 30 min prior to further
movement in phase 2. The force curve during movement of the nanocomposite shank over
a further 500 µm in depth at 100 µm/sec is also shown. The endpoint forces at 1 mm
99 obtained using 10 µm/sec were 500-800 µN compared to 1500-2500 µN at 100 µm/sec.
Agarose based gels were used to study if the nanocomposites 1) can penetrate brain like materials and 2) if the electrodes bent inside the gel after hydration. Additional tests showed that the force curves obtained from 100 µm/sec movement protocols were more consistent compared to the 10 µm/sec, especially after penetration (data not shown). Forces on the nanocomposite microelectrode exhibited less sample-to-sample variability, especially in the relaxation process, using the 100 µm/sec penetration speed (data not shown). Nanocomposite microelectrodes were also able to penetrate without bending in the gel and remained intact after removal from gel, suggesting that a softened and more compliant microelectrode shank maintains shape and form during the indentation test and after 30-45 min exposure to hydrated brain-like conditions.
Figure 21. Compliant nanocomposite shanks withstand indentation test. (A) In phase 1, the force- displacement curve of a compliant nanocomposite shank while it penetrates a 0.5% agarose gel mimic (elastic modulus of ~8kPa) at 100 µm/sec. First arrow at approximately 4.5 sec (450 µm depth) indicates penetration into the gel. Inset shows a closeup of nanocomposite shank prior to movement (left) and after placement at 1 mm depth (right). The nanocomposite shank is held stationary for 30 min prior to further movement in phase 2. (B) In phase 2, the force-displacement curve of the compliant nanocomposite shank as it moved an additional 500 µm in depth at the rate of 100 µm/sec is shown. Inset shows an intact nanocomposite shank after removal from gel, suggesting that a softened and more compliant shank maintains shape and form during the indentation test and after 30-45 min exposure to hydrated brain-like conditions.
100 4.3.4 Estimation of stress and strain from force measurements
To estimate the effective stress on the microelectrode as a force per unit contact
area, the contact areas of the microelectrodes in brain tissue were estimated for each
individual shank geometry using SolidWorks™ (Figure 22). Since the loading rate was
constant, the change in effective surface area could be calculated for each point in the force-
displacement curve based on the sampling frequency of the load cell. To further calculate
the effective stress on the microelectrode, the measured force was divided by the area
calculated at the corresponding time point. Effective strain was calculated using a linear
model:
(1) σ= Eε,
where σ is effective stress, ε represents effective strain, and E is brain tissue elastic
modulus taken to be 7.5 kPa based on previous studies [6]. Maximum strain is expected to occur at a depth of 200-300 µm in the tissue, which correspond to previous observations of dimpling and points of penetration into the tissue [6]. Since motional artifacts and surface variations due to local vasculature are expected to contribute to noise in dynamic strain measurements (Figure 24), significance was assessed using an α of 0.1 for these measurements.
101 Figure 22. For estimation of stress from complex A geometries, CAD models in Solidworks™ were used to derive contact surface areas with surrounding brain tissue for stress calculations for each type of shank– (A) non-compliant bare silicon, (B) PVAc coated (15 μm thick) non- compliant silicon, (C) compliant nanocomposite with an average roughness peak height (Rp) of 7 µm occurring every ~130 µm along the shank (based on bright-field images). The units for dimensions shown in the models are in (mm). Inset shows the bright-field image of each representative shank. B
C
4.3.5 Analysis of Viscoelastic Parameters and Estimating Shear Moduli
After termination of movement, the forces on the microelectrode relax in a time-
dependent, viscoelastic manner. The stress relaxation curves at 1 mm and 1.5 mm tissue
depth (Figure 25) could be treated as a small, stepped strain experiment and characterized using conventional viscoelastic models to characterize the shear stresses on the microelectrode. Effective stresses were calculated using the terminal depth achieved by the microelectrode (Table 2). The relaxation stress curves were smoothed using moving
102 average method to remove effects of tissue micromotion. Due to small strains involved in
the relaxation phase, a viscoelastic model described by a relaxation modulus with a 2nd
order Prony series expansion was found to be the best fit (R2 >0.95):
/ / (4.2) G(t) = Gl + + −𝑡𝑡 𝜏𝜏1 −𝑡𝑡 𝜏𝜏2 1 2 where G(t) is the relaxation𝐺𝐺 𝑒𝑒 modulus𝐺𝐺 𝑒𝑒 as a function of time, Gl is the long term shear
modulus calculated 100 sec after termination of microelectrode movement, G1 and G2
represent the parametric constants describing the prony series expansion, and and are
1 2 the characteristic short and long term time constant respectively for the tissue.𝜏𝜏 This model𝜏𝜏
with a Maxwell type element was also extensively described by Gefen et al. [299]. The
instantaneous shear modulus was calculated at t = 0, using
(4.3) G(0) = G1+ G2+ Gl\
Statistical means and standard error (SE) were determined excluding outliers and
plotted as mean ± SE for G(0), and . Significance was assessed using one way
1 2 ANOVA for G(0), and values,𝜏𝜏 with an𝜏𝜏 α criterion of 0.05. Student’s t-test was used
1 2 to assess significance𝜏𝜏 between𝜏𝜏 individual groups for , with an α criterion of 0.05 in
1 Figure 26. 𝜏𝜏
Table 2. Contact surface areas of Si-Bare, PVAc, and NC based on SolidWorks CAD models
**The nanocomposite samples were assumed to have an isotropic increase in volume of 11.2% after hydration (unpublished experimental data). This converged to a 6.5-7.0 µm increase in thickness in each dimension compared to the initially implanted shank.
103 4.3.6 Analysis of stresses due to brain micromotion
To characterize the amplitude of periodic stress variations due to tissue
micromotion which correlated with breathing and vascular pulsatility, the average peak-to-
peak force amplitudes was determined over a 5 sec interval during steady-state (typically
>100 sec after terminating microelectrode movement) of the relaxation phase. Tukey-box
plots of average stresses due to micromotion in each case were plotted to determine any outliers as marked by a red dot on relevant figures. Outliers were determined as values that exceed 1.5 times the inter-quartile distance (25% and 75% percentile markers). Student’s
t-tests with an α criterion of 0.05 was used to compare stresses due to tissue micromotion between any two of the following three groups - non-compliant, bare silicon shanks (n=4), non-compliant, PVAc-coated silicon shanks (n=3), and compliant nanocomposite shanks
(n=6).
4.4 Results
4.4.1 Comparison of forces exerted by non-compliant and compliant shanks at the brain- implant interface
Average force-displacement curves with standard deviation (gray) during penetration of tissue as the microelectrode moves 1 mm down into the tissue (in phase 1)
at 100 µm/sec are shown in Figure 23a-c. Subsequent 0.5 mm downward movement (in phase 2) at 100 µm/sec after a 30 min wait are shown in Figure 23d-f for bare silicon,
PVAc-coated silicon, and nanocomposite shanks. In all, n=4 pairs (corresponding to phases
1 & 2) of force curves for bare-silicon shanks, n=3 pairs for PVAc-coated silicon shanks,
and n=6 pairs for compliant nanocomposites in three different animals are measured. The
104 initial portions of the force curves in Figure 23a-c (corresponding to initial displacements of 200-600 µm) reflect the dimpling/compression of the brain tissue against the electrode before penetration. After the shank is stopped at a depth of 1 mm, mainly shear relaxation forces dominate till a steady state is reached corresponding to a full relaxation of brain tissue around the shank. The steady-state forces of ~200-500 µN after relaxation were comparable in all 3 types of microelectrodes.
Figure 23. Average force-displacement curves (black) ± standard deviation (gray)in (A) & (D) for bare silicon (non-compliant) shanks in phases 1 and 2 respectively (n=4 shanks), (B) & (E) for PVAc-coated silicon (coated, non-compliant) shanks in phases 1 and 2 respectively(n=3 shanks), and (C) & (F) for nanocomposite (compliant) shanks in phases 1 (n=6 shanks) and 2 respectively (n=5 shanks). One force curve was determined to be an outlier and was not included for nanocomposite phase 2 curves. Shanks are inserted at a speed of 100 µm/sec in both phase 1 (penetration phase (A-C)) and phase 2 (further downward movement by 0.5 mm (D-F)).
The range of maximum forces at a depth of 1 mm for non-compliant, bare silicon
shanks are 600-1200 µN (Figure 23a). After zeroing the starting point of force curves corresponding to phase 2, the range of maximum forces at 1.5 mm depth ranged from 600-
1750 µN. For non-compliant, PVAc-coated silicon shanks, the range of maximum forces at the end of phase 1, at a depth of 1 mm, was 300-500 µN, which is smaller than those of
105 the bare silicon samples. At a depth of 1.5 mm, maximum forces ranged from 200-1700
µN for the PVAc-coated silicon shanks. Figure 23c,f shows the force curves generated by the compliant nanocomposite shank during phases 1 and 2. Due to the temporal changes in the effective elastic modulus of the nanocomposite material, variability was higher in the observed force curves. The range of maximum forces at a depth of 1 mm was 500-3000
µN and that corresponding to a depth of 1.5 mm was 200-2500 µN with an outlier at 10000
µN (data not shown). It is possible that the outlier shank sample might have touched against an underlying vasculature.
The force curves for the compliant nanocomposite shanks before (‘stiff’ state) and after hydration (‘soft’ state) are shown in Figure 23c,f respectively. The main difference between the ‘stiff’ state and the ‘soft’ state is observed in the relaxation curves where curves corresponding to the ‘stiff’ state electrodes were generally faster to relax and those corresponding to the ‘soft’ state generally took longer to relax.
4.4.2 Coated, non-compliant and compliant shanks impose less strain at brain-implant interface
Estimated dynamic stresses were calculated based on surface areas derived from
CAD-based geometric models for shanks made of non-compliant, bare silicon, PVAc- coated, non-compliant silicon, and compliant nanocomposites. The CAD based models are used to derive tissue contact surface areas at any given depth (Figure 22) and a comparison of contact surface areas at a depth of 1.0 mm & 1.5 mm inside the brain tissue for each of the shanks is shown in Table 2. The nanocomposite shanks have ~98% more surface area than that of bare silicon and the PVAc-coated silicon surface areas are ~45% more than
106 that of bare silicon at a depth of 1.0 mm. Changes in surface area account for swelling of nanocomposite material under hydration conditions at 1.5 mm. The dynamic stresses in cortical tissue induced by bare silicon, PVAc-coated silicon, and compliant nanocomposite shanks during their movement from 1.0-1.5 mm (phase 2) are shown in Figure 24a.
Penetration stresses and stresses from meningeal layers are not expected to be dominant
during this phase.
A linear approximation is used to estimate dynamic strain at 200 µm tissue depth.
The rationale to use strain as a point of comparison at 200 µm tissue depth is based on prior
experimental observations and mechanical modeling which show that microelectrode
movement beyond this depth is typically characterized by tissue tearing and shear forces
beginning to dominate the observed forces [6]. Therefore, maximum tissue strain is
expected to occur at this depth. Tukey-box plots show the estimated strain at 200 µm tissue
depth for non-compliant, bare silicon (n=4 shanks), PVAc-coated non-compliant silicon
(n=3 shanks), and compliant nanocomposite shanks (n=6 shanks) in Figure 24b. The range
of strain values was 0.10 -0.44 for bare silicon, 0.03 - 0.26 for PVAc-coated silicon, and
0.01 - 0.21 for compliant nanocomposite excluding an outlier (0.45). The distribution of
strain values show that 75% of the strain values corresponding to PVAc-coated silicon and
nanocomposite shanks are below the median strain value for non-compliant, bare silicon
shanks with one outlier seen for nanocomposite shanks. Strain values for non-compliant,
bare silicon (n=4 shanks) were significantly different from compliant nanocomposites (n=6
shanks) (p<0.09). Variability in strain values corresponding to each shank type are likely
due to variations in local vasculature and modulations due to related tissue micromotion.
107 A
B
Figure 24. (A) Dynamic stresses for non-compliant Si-bare (n=4 measurements), coated, non-compliant PVAc-coated Si (n=3 measurements), and compliant nanocomposite microelectrodes post-swelling (n=6 measurements) in phase 2 as the microelectrodes were moved from 1.0 to 1.5 mm are shown as a function of tissue depth. A total of n=3 animals were used. Microelectrodes were moved at a constant speed of 100 µm/sec for a net change of 500 µm depth. (B) Strain was calculated using a linear model (σ= Eε), where σ =compressive stress, ε=compressive strain and the elastic modulus (E) of the brain is based on the median value for acute animals (7.5 kPa) [6]. Estimated strain which is expected to be maximum at a depth of 1200 µm corresponding to a net change of 200 µm from the beginning of microelectrode movement in phase 2 is plotted as a Tukey-Box plot for all three types of shanks: Si-bare (n=4 measurements), coated, non-compliant PVAc-coated Si (n=3 measurements), and compliant nanocomposite microelectrodes post-swelling (n=6 measurements). Dotted line highlights that 75% of strains in both PVAc-coated Si and nanocomposite samples are below the median strain value for bare Si. Red dot marks outliers in nanocomposite samples.
108 4.4.3 Compliant shanks slow interfacial brain-tissue relaxation properties
After stopping movement of the electrode at 1.0 mm and 1.5 mm depth, the brain tissue undergoes a relaxation process that is reflected in the measured forces (Figure 23).
The corresponding effective relaxation stresses were determined using estimated surface areas reported in Table 2. Representative relaxation curves for non-compliant, bare silicon,
PVAc-coated non-compliant silicon, and compliant nanocomposite shanks are plotted in
Figure 25. At t=0, the brain tissue begins to relax exponentially over time after the shanks reach cortical tissue depths of either 1 mm or 1.5 mm. For non-compliant, bare silicon shanks (Figure 25a,d), the family of relaxation force curves at a depth of 1.0 mm is similar in characteristics to the one at a depth of 1.5 mm. The relaxation force curves for PVAc- coated silicon shanks are shown in Figure 25b,e corresponding to depths of 1 mm and 1.5 mm respectively. The initial stresses at t=0 are the least in value for PVAc-coated silicon shanks compared using student’s t-test with bonferroni correction to both bare silicon and compliant nanocomposite samples (p<0.01). The relaxation force curves for nanocomposite samples are shown in Figure 25c,f corresponding to depths of 1 mm and
1.5 mm respectively.
109
Figure 25. Representative relaxation stresses for non-compliant bare silicon (A,D), PVAc-coated non- compliant silicon (B,E), and compliant nanocomposite (C,F) microelectrodes. (A-C) represent relaxation stresses taken after microelectrode movement stops at a depth of 1.0 mm. (D-F) represent relaxation stresses taken after movement stops at a depth of 1.5 mm. Insets show representative micromotion related stresses due to breathing and vascular pulsatility over a 5 sec interval.
To better quantitate the relaxation characteristics, a Maxwell type model with a 2nd
order Prony series expansion (Equation 2) was used to fit the relaxation force curves for
all shanks. Viscoelastic parameters determined using the above model under step
indentation-like conditions have been previously shown to be consistent with brain tissue
material properties [6]. Instantaneous shear moduli calculations based on Equation 3
showed no significant differences among measurements made by the different shanks as
shown in Figure 26a. All the average shear modulus values (1.2-3.5 kPa) were within the same range of values as previously reported [6], confirming the consistency in material properties of the brain tissue surrounding the shanks of different materials. However, the characteristic short-term time constant (τ1) for compliant nanocomposite shanks was
significantly smaller (p<0.05) for relaxation force measurements at a depth of 1.0 mm (2.2
± 2.1 sec) compared to that at a depth of 1.5 mm (5.6 ± 2.0 sec), suggesting that softened
110 interfaces slow the rate at which brain tissue relaxes (Figure 26b). The short-term time
constants for non-compliant, bare silicon shanks (4.7 ± 1.5 sec at 1.0 mm and 3.0 ± 2.0 sec
at 1.5 mm) and PVAc-coated silicon shanks (2.5 ± 3.5 sec at 1.0 mm and 2.6 ± 0.9 sec at
1.5 mm) show no significant differences between relaxation forces at depths of 1 mm and
1.5 mm. However, the mean short-term relaxation (τ1) is significantly (p<0.05) smaller for
PVAc-coated silicon shanks compared to that of nanocomposites at 1.5 mm depth suggesting that the interaction between the softened nanocomposite and brain tissue leads to a slower rate of relaxation in stress. No significant differences are seen in the long-term time constants (τ2) for all shanks, which ranged from 52.6-95.5 sec based on one way
ANOVA. (Figure 26c)
Figure 26. A viscoelastic material model was derived from the relaxation stresses such as those in Fig.6. nd Parameter estimation was performed for a 2 order prony series model G(t) = G1exp(t/τ1) + G2exp(t/τ2) + Gl where G(t) is the shear stress over time (t), G1 ,G2, and Gl (sometimes refered to as G∞) are stress materials parameters. τ1 and τ2 are time constants describing the short term and long-term relaxation of the brain tissue. (A) The instantaneous shear modulus (G(0)) which is calculated as: G(0) = G1+ G2+ Gl, shows no significant differences among the different samples, indicating acute brain material properties do not change due to interaction with shanks. (B) and (C) shows the comparative τ1 and τ2 time constants representative of the short-term and long-term relaxation properties of the brain tissue, respectively. In (B), τ1 for the nanocomposite shank after hydration at 1.5 mm is significantly larger than those of the same shank before hydration at 1.0 mm and those corresponding to PVAc-coated silicon at 1.5 mm (p < 0.05 for both comparisons, indicated by *). (C) τ2 shows no significant differences among the different samples. Significance was assessed using one-way ANOVA for G(0), and values, with α=0.05. Further tests among values using Student’s t-test using α=0.05 showed significant differences as indicated by the *. 𝜏𝜏1 𝜏𝜏2 𝜏𝜏1
111 4.4.4 Coated, non-compliant and compliant shanks minimize mechanical stresses due to
tissue micromotion at the interface
Tissue micromotion induced stresses were observed in all types of shanks under
steady-state conditions. Tissue micromotion due to breathing and heart rate pulsatility
imposed a periodic, dynamic stress on the shank. Figure 27 shows the distribution of
micromotion induced stress amplitudes for non-compliant, bare silicon (n=4 measurements
for each depth), PVAc-coated, non-compliant silicon (n=3 measurements for each depth),
and compliant, nanocomposite shanks (n=6 measurements for each depth). The distribution
of stresses (and corresponding strains) at depths of 1.0 mm and 1.5 mm are shown for each
shank as Tukey-Box plots in Figure 27. The average micromotion induced stress
amplitudes for non-compliant, bare silicon were significantly larger (221.5 ± 27.8 Pa at 1.0
mm and 170.6 ± 52.9 Pa at 1.5 mm) compared to those of either PVAc-coated silicon (82.2
± 15.0 Pa at 1.0 mm and 68.3 ± 27.2 Pa at 1.5 mm) or compliant nanocomposite shanks
(99.1 ± 44.3 Pa at 1.0 mm and 49.8 ± 13.4 Pa at 1.5 mm). The changing mechanical
properties of ‘stiff’ nanocomposite before hydration at a depth of 1.0 mm compared to
‘soft’ and compliant states after hydration at a depth of 1.5 mm are also shown to
significantly reduce (p<0.05) interfacial micromotion induced stresses on the shank. The
comparative measurements at depths of 1.0 mm and 1.5 mm for both non-compliant, bare silicon and PVAc –coated silicon shanks were not significantly different.
112
Figure 27. Micromotion induced stress (y-axis on the left side) and strain (y-axis on the right side) amplitudes were pooled for non-compliant, bare silicon (n=4), PVAc-coated non-compliant silicon (n=3), and compliant nanocomposite (n=6) shanks from steady steady stresses at 1.0 mm and 1.5 mm tissue depth. Strain was estimated using a linear model (Equation 1) as described in the methods section. * indicates a significant difference (p<0.05) based on Student’s t-test (α=0.05). Red dot indicates an outlier.
4.5 Discussion
The mechanical mismatch between brain tissue and implants has long been
hypothesized to induce strain in brain tissue in the immediate vicinity of the implants
exacerbating glial scarring and neurodegeneration at the interface, leading to progressive
signal loss over long implantation periods. Typical materials used for electrically active
implants and electrodes include tungsten (400 GPa elastic modulus), stainless steel (200
GPa), and silicon (180-200 GPa) [300, 301]. Comparatively, the elastic modulus of brain
tissue is at least 6 orders of magnitude less (3-15 kPa) [302] with individual neurons and astrocytes reportedly on the order of 1-100 Pa in elastic modulus [303]. In this study, we compared the mechanical behavior of the interfacial brain tissue in response to indentation and penetration by shanks of differing material properties. Here, non-compliant silicon
(elastic modulus of ~200 GPa), PVAc-coated non-compliant, bare silicon (~49-78 GPa),
113 and a novel, tunably compliant nanocomposite material in its ‘stiff’ state (5.2 GPa) and finally in its hydrated, ‘soft’ state (12 MPa) were tested. [19]
Despite the larger contact surface area of compliant nanocomposites, a key result of the experiments show that softer materials, especially nanocomposites, affect the viscoelastic properties of brain tissue. Based on viscoelastic models of interfacial brain tissue (Figure 26), the ‘soft’ state of nanocomposite materials are shown to significantly retard the dynamic stress relaxation rate of the brain tissue. In addition, interfacial micromotion induced stress amplitudes are 30-50% lower with softer, compliant shanks compared to non-compliant, bare silicon shanks. This compares well with previously reported modeling results where a soft interface with an elastic modulus of 6 MPa was predicted to result in ~30-40% reduction in micromotion related stress/strain [304].
Interestingly, PVAc-coated silicon shanks (with an elastic modulus of 49-70 GPa) also exhibited significantly (p<0.05) lower micromotion induced stresses (by 37-40%) compared to bare silicon shanks. Clearly, alternative mechanisms, such as surface adhesion properties, must be considered to explain similar reduction in micromotion induced stresses by two materials (PVAc-coated silicon and nanocomposites) with significantly different elastic moduli. PVAc is known to have strong tissue adhesive properties and therefore it may reduce the micromotion by strongly adhering to the tissue along all axial and normal/shear components, resulting in an extremely rigid, negative stiffness-like material
[305].
In addition, since micromotion induces a continuous, periodic stress on the shank, the rate at which brain tissue relaxes may play a significant role under chronic implantation conditions. The mean short term time constant (τ1) for PVAc-coated shanks is smaller (2.6
114 sec) than hydrated, ‘soft’ nanocomposites (5.6 sec), suggesting that the rate at which the
brain tissue material returns to the original relaxed state is lower for the latter compared to
the former. Since stress rate is directly related to strain rate, it is possible therefore that
softer, compliant nanocomposite materials result in significantly lower strain rates on the
brain tissue. There is evidence that high strain rates lead to increased injury in traumatic
brain injury scenarios [306, 307]. Further, in vitro tests by Karumbaiah et al. showed that
an imposition of 1-3% strain on neuronal, astrocyte, and microglia did not cause immediate
cell death as compared to results at 5% strain [26]. Further experiments conducted at 3%
cyclical strain to emulate micromotion conditions at 1.3 Hz resulted in cortical neuronal
death after 4 hours of exposure, microglial cell death after 8 hours of exposure, and
astrocyte death after 24 hours of exposure [26]. The estimated dynamic strain in the brain tissue in the current study based on measured micromotion stresses in vivo is 2.5-3.5% for bare silicon shanks (Figure 27). Comparatively, PVAc-coated shanks and soft nanocomposite shanks imposed less strain on the interfacial tissue (up to 1.5%) which when compared to work by Karumbaiah et al. suggest that the above shanks are expected to cause minimal cell death under long-term implantation conditions. Recent histological evidence indicate that the softer, nanocomposite implants reduced infiltration of IgG into the brain tissue as well as microglial activation and neurodegeneration compared to the much stiffer PVAc-coated implants at chronic time points (p<0.001) [19] However, in the context of the acute measurements made here, it is interesting to note that immunohistological analysis indicated no difference in neuroinflammatory events between the above two implants at short time points. The above data raises the interesting possibility that soft substrates may not reduce initial tissue injury from implantation but, as seen with
115 the nanocomposites used here, result in significant reduction in both chronic strain rates
and chronic strain levels (due to tissue micromotion) in the surrounding brain tissue.
Therefore, at least two possible explanations exist for the discrepancy in the immediate
reduction in tissue strain, but only a chronic effect on neuroinflammation. First, differences
in the maturation of the glial scar seen in tissue surrounding compliant implants versus stiff
implants [19] may result in increasing differences in the tissue strain. We have previously
shown that steady state stresses remain as high as 10% (at 6-8 weeks) following the implantation of stainless steel microelectrodes [6]. Therefore, the long-term changes in the
material properties of brain tissue at the implant-tissue interface, for soft/compliant
implants, remain critical to our understanding for chronic microelectrode design. While
beyond the scope of this study, tissue anisotropy in the healing cortex will need to be
accounted for in understanding the long-term changes in material properties. Nguyen et al.
show using finite element modeling that the spatial variation in tissue strain extends up to
500µm, which corresponds well to results from prior studies [19]. However, the complexity
of the glial scar formation suggests that an isotropic assumption may not accurately
represent the brain tissue material at the interface. Aside from anisotropy from grey matter/
white matter transition areas and white matter tracts in the healing cortex, anisotropy is
also expected to arise from the glial scar formation. Recent findings using diffusion tensor
based imaging show that a cortical impact method that induced TBI in rodents led to
significant and increased anisotropic contribution from reactive astrocytes in injured areas
in the cortex [308]. Future mechanical models of brain tissue interfaces may benefit from
incorporating data from diffusion tensor anisotropy and tractography for evaluation of
novel electrode designs.
116 Additionally, while beyond the scope of this study, further understanding of
whether negative-stiffness materials like PVAc [305] are a suitable class of coating materials for neural implants needs to be further elucidated. Second, it is equally as likely that different and distinct inflammatory pathways dominate various temporal aspects of tissue healing and remodeling following microelectrode implantation. Potter et al. has previously shown that curcumin-releasing softening polymer implants cause minimal microelectrodes-mediated neuroinflammation at acute time points [14]. However, in their study, Potter et al. used poly(vinyl alcohol) as the polymer, which resulted in increased chronic inflammation. Therefore, future studies may also investigate the complementary mechanism of drug-releasing, strain-reducing poly(vinyl acetate) polymer microelectrodes
(used here), at acute and chronic time points. Further, the strains and strain rates estimated in this study for bare silicon is in the same range that was reported earlier [26] to cause cell death in a culture of neurons, astrocytes and microglia. However, the strains and strain rates estimated for the nanocomposite and PVAc-coated bare silicon electrodes in this study were significantly lower than those reported to cause cell death [26].
4.6 Conclusion
The mechanical properties of underlying substrate have been extensively shown to play a significant role in synaptic strength and neuronal function and morphology in numerous in vitro studies [309, 310]. Microglia and astrocytes, in addition to subcellular structures such as synapses in neurons have been demonstrated to be mechanosensitive [25,
311, 312]. Harris et al. [20] and Nguyen et al. [19] have shown that the presence of neuronal cell bodies increases near softer implants. Numerous modeling studies suggest
117 that softer material impart less strain on tissue [19, 304]. In this study, nanocomposites and
PVAc-coated, non-compliant materials were shown to impose the least strain on the interfacial brain tissue (~2-10 fold less) compared to non-compliant, bare silicon shanks.
The micromotion induced stresses and viscoelastic parameters of the brain tissue derived
surrounding bare silicon in this current study are comparable to literature values for brain
tissue viscoelastic properties [6]. Although, the effect of soft interfaces on neural signal
quality remains unknown, electrode designs in the future could benefit from soft interfaces
to control and compensate for chronic mechanical stresses on the brain tissue.
Acknowledgments – The authors want to thank NIH for supporting this research through R01NS055312-S1 and Ruth-Kirchstein post-doctoral fellowship (NIH F32
NS073422-02) for AS. Research was further supported by NIH Neuroengineering
Training Grant (5T32EB004314-14), Dept. of Veterans Affairs Merit Review (MR-208
B7122R). The authors acknowledge C. Weder for supplying the dynamic materials, and A.
Hess for micromachining of the materials into neural probes.
118 Chapter 5
The effect of antioxidant-releasing mechanically-adaptive implants on modulating the neural tissue response
5.1 Abstract
The stability and longevity of recordings obtained from intracortical microelectrodes continues to remain an area of concern for neural interfacing applications.
Microelectrode performance has increasingly been associated with the integrity of the blood brain barrier (BBB) and the neuroinflammatory response to the microelectrode.
Here, we investigate an additive approach to target both mechanical and chemical factors believed to contribute to chronic BBB instability surrounding implanted intracortical microelectrodes. Our approach utilized our novel mechanically adaptive, compliant nanocomposites (NC) to simultaneously reduce the tissue response and tissue strain, with an imbibed antioxidant for an added local delivery. Various concentrations of antioxidants, curcumin (Cur) or resveratrol (Res), were doped into NC films. In vitro analysis of drug release, antioxidant activity, and cytotoxicity determined that 0.01% antioxidant- containing films were optimal for in vivo assessment. Consequently, NC, NC Res, and NC
Cur probes were implanted into the cortical tissue of rats for up to sixteen weeks. Our results suggested that at three days post-implantation, neither materials nor therapeutic approaches (independently or in combination) could alter the initial wound healing response. However, at two weeks post-implantation, NC Res probes displayed a reduction in activated microglia and improvement in neuron density at the tissue-probe interface.
When the antioxidants were expected to be exhausted, NC Res and NC alone exhibited
119 similar tissue responses sixteen weeks post-implantation. The NC Res demonstrates the short-term, short-lived benefits of antioxidant treatment, and the long-term reductions in neuroinflammation exhibited by compliant implants. Together, these results demonstrate that local delivery of antioxidants can provide an additive advantage to further improve the tissue response to compliant implants.
5.2 Introduction
Chronic recordings from intracortical microelectrodes provide a platform to enable neural control for motor restoration.[40] Reports of motor restoration have described a variety of successful control tasks in patients with tetraplegia, including control of a computer cursor, control of a prosthetic limb (reach and grasp), point-and-click communication, and speech synthesis.[81, 82, 238, 313] Although top performing arrays have been implanted for 4+ years, a majority of studies still display daily signal instability and decreases in viable recording channels over time.[3, 41, 91]
While there are several noted failure modes for intracortical microelectrodes, it is widely regarded that the proximity of neurons and characteristics of the tissue surrounding microelectrodes play a crucial role in the stability and integration of implanted microelectrodes.[42, 70, 101-103] Recent work by the Bellamkonda group has demonstrated that maintaining the integrity of the blood-brain barrier influences recording stability.[11] Several mechanisms are believed to contribute to prolonged BBB dysfunction surrounding implanted microelectrodes.
First, traditional intracortical electrode materials are stiff compared to the brain tissue. In silico studies suggest that the mechanical mismatch induces substantial strain on
120 surrounding tissue, exacerbating the inflammatory response and neurodegeneration.[4, 5,
70, 101, 102] These studies further suggest that reduction of micromotion-induced strain by using compliant microelectrode materials can reduce the tissue response. With development of mechanically-adaptive NC materials, a class of intracortical implant materials have been generated that better match the brain modulus after implantation while also sustaining insertion without the need of an assistive device.[22, 23, 207, 211]
Comprised of cellulose nanocrystals (CNC) and polyvinyl acetate (PVAc), these NC materials soften from 5 GPa to 12 MPa upon in vivo implantation.[23, 211] Chronic implantation of the compliant NC have shown an improvement in neuronal density surrounding microelectrodes.[19, 20, 23]
Second, continual BBB leakiness may exacerbate the inflammatory response and accumulation of cytotoxic molecules. Delivery of anti-inflammatory therapies can mediate the tissue response to implanted microelectrodes. Dexamethasone, an anti-inflammatory glucocorticoid, has been administrated via systemic injection [146, 147] and released from hydrogel coatings around neural probes [148-151]. Most studies reported reduction in reactive astrocytes but little effect on activated microglia/macrophages. Rennaker et al. found that systemic administration of the antibiotic minocycline for up to 4 weeks post- implantation improved the SNR and % of channels recording units, attributed to reduced reactive cells and increased neuronal survival.[9] However, minocycline failed to stabilize neural recordings at early time points and chronic antibiotic administration is not a viable solution. Additionally, groups have investigated local delivery of a variety of bioactive modulators to enhance neural integration.[147, 158, 161, 222, 314] The Cui group immobilized L1, a neural adhesion molecule, to promote neurite outgrowth and neuronal
121 survival surrounding implanted microelectrodes.[157] The results indicated no loss of neuronal cell bodies and a significant increase in axonal density at the interface.[158, 159]
Studies have also considered coating electrodes with neural progenitor cells to re-establish
neuronal populations, but the feasibility of delivering live cells is questionable. [160, 161]
Our lab previously reported on successful neuroinflammatory modulation with
antioxidant delivery of resveratrol and curcumin.[14, 15] Inflammatory-mediated oxidative stress events, resulting from dangerously high levels of reactive oxygen species (ROS), can induce corrosion of the electrode site and compromise neuronal health.[15, 57, 61, 62, 247]
An increase in ROS triggers activation of the NF-kB and AP-1 pathways, key inflammatory pathways that lead to production of inflammatory cytokines.[122] Therefore, antioxidative therapies have the capability to reduce reactive oxygen species accumulation and consequently inflammatory events. However, the neuroinflammatory response is multi- faceted and complex, and thus far single solutions have not been sufficient to adequately mediate the damaging elements.
While compliant electrode materials can reduce long-term inflammatory events[19], addition of local antioxidant delivery may enable a cooperative solution to reduce BBB-mediated damage throughout the implant lifetime. Curcumin-releasing polyvinyl alcohol (PVA) polymer implants demonstrated potential for antioxidant- releasing compliant materials, reducing the acute inflammatory response.[14] However, long-term exposure to the inflammatory environment may have increased the solubility of
PVA implants, causing a loss of effectiveness chronically. Our PVAc NC has been implanted for up to sixteen weeks with demonstrated long-term material stability and biocompatibility.[19] Additionally, an alternative antioxidant, resveratrol, has
122 demonstrated more potent antioxidant effects than curcumin, improving neuronal
populations surrounding intracortical implants for up to four weeks, with a single
injection.[15] In this study, we propose a system to release curcumin or resveratrol from
PVAc NC implants to further stabilize neuron populations around compliant implants.
5.3 Methods
5.3.1 Chemicals and reagents
Pharmaceutical grade, 99% pure trans-resveratrol (Res) powder was purchased
from Mega Resveratrol (Danbury, CT) and 99% pure curcumin (Cur) was obtained from
® ChromaDex (Figure 5-1). Poly(vinyl acetate) (PVAc, Mw = 100,000), 2,2-diphenyl-1- picrylhydrazyl (DPPH), Triton-X 100, and all other reagents were purchased from Sigma
Aldrich. Cellulose nanocrystals (CNCs) in this study were isolated from tunicate (Styela clava) collected from floating docks in Point View Marina (Narragansett, RI), and prepared by sulfuric acid hydrolysis of the cellulose pulp, according to established protocols, as previously reported.[315, 316] To simulate the ionic composition of endogenous brain fluid, artificial cerebrospinal fluid (ACSF) was prepared following an established protocol[22] by dissolving the following materials in one liter of deionized water: sodium chloride (NaCl) = 7.25 g, potassium chloride (KCl) = 0.22 g, sodium bicarbonate
. (NaHCO3) = 2.18 g, calcium chloride dihydrate (CaCl2 2H2O) = 0.29 g, monopotassium
. phosphate (KH2PO4) = 0.17 g, magnesium sulfate heptahydrate (MgSO4 7H2O) = 0.25 g,
and D-glucose = 1.80 g. Cell culture reagents were purchased from Life Technologies.
123 5.3.2 Preparation of antioxidant-releasing nanocomposite films
A poly(vinyl acetate) (PVAc) stock solution was prepared by dissolving PVAc in dimethylformamide (DMF) at a concentration of 50 mg/mL by stirring for 3 hours at room temperature (RT). Similarly, stock solutions containing curcumin or resveratrol were made by separately dissolving the antioxidants in DMF (10 mg/mL) by stirring for 1 hour at RT.
Lyophilized CNCs were dispersed in DMF at a concentration of 5 mg/mL by sonicating for 10 hours at room temperature using BANDELIN SONOREX TECHNIK RL 70 UH sonicator operating at 40 kHz. Films containing either 0%, 0.005%, 0.01%, 1.0% or 3.0% w/w of the antioxidant (i.e. curcumin or resveratrol) and 15% w/w CNCs were made by combining appropriate amounts of the above stock solutions and CNC dispersion, and stirring the mixture for 30 minutes at RT. The mixtures were then cast into a Teflon® Petri
dish, dried at 70 ºC for 5 days and further dried at 120 ºC under high vacuum in an oven
for an additional 24 hours to evaporate all of the solvent. After drying, nanocomposite (NC)
films were compression-molded between spacers in a Carver® laboratory press (1000 psi
for 2 minutes, followed by an increase of pressure to 3000 psi for 10 minutes) at 90 ºC, to
yield ~100 μm thin films. The antioxidant-releasing NC films thus produced were stored
in a desiccator at ambient temperature. NC implants for in vivo experiments were fabricated
by laser-micromachining with a 355 nm wavelength picosecond laser (Model A-355-pico,
Oxford Lasers) to a thickness of ~70-110 μm, a length of 2 mm, and a shank width of 50
μm. All materials were ethylene oxide sterilized before both in vitro and in vivo
experiments.
124 5.3.3 Mechanical characterization
The mechanical properties of the antioxidant-releasing NC films were
characterized by dynamic mechanical analysis (DMA, TA Instruments, Model Q800).
Tests were conducted in tensile mode, sweeping the temperature between 23-80 °C at a fixed frequency of 1 Hz, and using a strain amplitude of 15 µm, a heating rate of 5 °C/min.
Samples for mechanical testing were prepared by cutting strips (~30 mm × ~6 mm × ~100
μm) from the films. To determine the mechanical properties of the films in the wet state, samples were first swelled in ACSF at 37 °C for one week. After the degree of swelling had been measured, DMA experiments were conducted in submersion clamp using tensile mode, which allowed mechanical measurements while the NC films were immersed in
ACSF. In submersion measurements, the temperature sweeps in the range of 25-50 °C with
a heating rate of 1 °C/min, a constant frequency of 1 Hz, and strain amplitude of 15 µm.
These experiments were repeated five times, and the result was expressed as mean ±
standard deviation.
5.3.4 Finite Elemental Modeling (FEM) of antioxidant release
To estimate antioxidant release from doped NC films, a 2D finite-element model
was developed to predict the drug release profile at the probe-tissue interface. The model
consisted of a cross-section of a drug-containing NC implant within a tissue block.
Boundary conditions were defined as no flux at the tissue borders and open boundary at
the polymer-tissue interface. Drug delivery through the tissue was modelled as Fickian
diffusion. The effect of NC swelling on drug diffusion was not taken into account. Since
diffusion properties of resveratrol and curcumin through brain tissue have not been
125 specifically determined, diffusion parameters were assumed to be similar to
dexamethasone because the molecular weight, solubility, and density of dexamethasone
closely matches that of resveratrol and curcumin.[317] The system was defined as a no convection system with isotropic diffusion and first-order elimination kinetics (C=C0*e^(-
kt), k_e = 0.4 hr-1) (assuming non-enzymatic reactions). The diffusion coefficient through
brain tissue was defined as D_brain = 2e-6 cm2/s and through a porous polymer as
D_polymer = 2.04e-10 cm2/s.[317] The initial concentration of drug within the probe at t=0 was determined assuming a polymer probe of 100 µm x 50 µm x 2 mm doped with
0.005%, 0.01%, 1% or 3% w/w drug. Percent drug release at the probe-tissue interface with respect to time was determined.
5.3.5 In vitro antioxidant release
To determine the release rates of resveratrol or curcumin from the antioxidant- loaded NC films, samples (~30 mm × ~6 mm × ~100 μm) were incubated at 37 °C in a mixture of 20 mL of 99.5% v/v ACSF and 0.5% v/v Tween-80, which was added to increase the solubility of the drug in the ACSF. In set time intervals (t=1, 2, 4, 6, 8, 10, 24,
29, 34, 48, 55, and 72 hours), 1 mL aliquots of the solutions were withdrawn, diluted with
1.5 mL of the neat solvent (99.5% v/v ACSF and 0.5% v/v Tween-80), and the amount of antioxidant released was detected spectrophotometrically (UV-vis 2401-PC spectrophotometer, Shimadzu). The concentration of antioxidant released was calculated using a calibration curve established by measuring the absorbance at 425 nm and 308 nm of solutions series of curcumin and resveratrol (5-30 μg/mL) in ACSF/Tween 80, respectively. The amount of antioxidant released (relative to the amount of antioxidant
126 originally present in the NCs) was calculated using Equation 4, where [Antioxidant]UV is
the concentration of the antioxidant measured by UV (mg/mL), VSample the volume of the
sample in mL, WFilm the weight of the film in mg, and [Antioxidant]Film is the nominal concentration of the antioxidant in the film (mg/mg). This experiment was repeated four times, and the result was expressed as mean ± standard deviation.
[ ] × ( ) Antioxidant Release (%) = × 100 4 ×[ ] Antioxidant 𝑈𝑈𝑈𝑈 𝑉𝑉𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑊𝑊𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 Antioxidant 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹
5.3.6 Cell culture
BV-2 microglia cells were graciously donated by Dr. Stephen Selkirk and NSC-34
neural stem cells were graciously donated by Dr. Horst von Recum. Both cell types were
maintained in DMEM Dulbecco's Modified Eagle Medium (ATCC) supplemented with
10% fetal bovine serum (Invitrogen) and 1% penicillin–streptomycin (ATCC). For all
experiments, BV-2 and NSC-34 cells were used between passages 5 to 25. To differentiate
NSC-34 cells into neuron-like cells, the NSC-34 cells were incubated in differentiation media (DMEM/F12 supplemented with 1% fetal bovine serum, 0.5% penicillin- streptomycin, non-essential Amino Acids (NEAA), and 1 µM retinoic acid) for 4 days, changing the differentiation media every 2 days. For all in vitro tests, BV-2s were seeded at a density of 8000 cells/cm2, and NSC-34s were seeded at a density of 5000 cells/cm2.
All representative images from the assays detailed below were acquired fluorescently using
an inverted AxioObserver Z1 (Zeiss) equipped with an AxioCam MRm. For each assay,
exposure times were held constant between conditions.
127 5.3.6.1 Glial cell viability
To analyze the effects of the doped NC films on microglia cell viability, a slightly
modified version of the LIVE/DEAD® Viability/Cytotoxicity Kit (Life Technologies) was
used. BV-2 cells were tested with films with several concentrations of dopant: 0.005%,
0.01%, 1% and 3% w/w Res or Cur. BV-2 cells were seeded in a 12-well plate and film samples (1 cm × 1 cm × 70-110 μm) were suspended in the media using an insert and adhesive (Kwik-Sil). Positive (all live) and negative (all dead) controls were performed using the insert and adhesive only, no film. The negative control was established by incubating cells with 70% methanol for 10 minutes. After 48 hours, the media was removed and the cells were incubated in 8 µM ethidium homodimer-1 (EthD-1) and 0.1 mM calcein-
AM in complete phosphate buffered saline (PBS) for 15 min. The samples were then
washed once with PBS. Fluorescent images were taken (488nm green/live, 594 nm
red/dead) to qualitatively determine the cell viability and cytotoxicity of the different
antioxidant concentrations.
5.3.6.2 BV-2 and NSC-34 co-culture: DHE and LIVE/DEAD®
BV-2 cells and NSC-34 cells were co-cultured to analyze their response to exposure
to NC films. Here, we loaded with 0.01% w/w Res or Cur, since 1% and 3% films indicated
cytotoxicity with BV2 microglia. Neat NC was used as a control. In all experiments, NSC-
34 cells were first differentiated on 12-well plates for 4 days; then, BV-2 cells were seeded
on tissue-culture treated 0.4 µm pore polycarbonate membrane inserts (Sigma-Aldrich).
The NSC-34 and BV-2 cells were incubated with the experimental film (1 cm × 1 cm × 70-
110 μm) suspended in the media for 48 hours. Following 48 hours, the BV-2 cells were
128 tested for intracellular superoxide accumulation and NSC-34 cells were analysed for cell
viability.
Dihydroethidium (DHE) labeling was utilized to measure intracellular superoxide anion accumulation in BV-2 cells.[318] Prior to performing the staining, brightfield images were taken of the cells to ensure cell viability. Cells were washed one time with PBS, and then incubated with 3 µM DHE in PBS for 30 minutes at room temperature. After one wash in PBS, fluorescent images were taken at 555nM to qualitatively visualize the DHE within the cells. To quantify DHE staining, total fluorescent intensity of DHE was determined for each image and normalized to total fluorescent intensity of cells only control. Although total cell count varied for the different conditions, data was not normalized to total cell count because antioxidants can affect microglial proliferation. To determine viability of neurons with NC films, cell morphology was assessed. Differentiated neurons were sensitive to repeated wash steps, therefore Live/Dead staining was not possible.
5.3.7 Measurement of antioxidant activity of films
Antioxidant activity of NC films loaded with resveratrol and curcumin was
determined by measuring the reduction of 2,2-diphenyl-1-picrylhydrazyl (DPPH), a stable
free radical. Samples (3 cm × 0.5 mm × 70-110 μm) were incubated at 37 °C in 1 mL of
complete PBS with 0.5% v/v Tween-80; the Tween-80 was added to improve the solubility
of the antioxidants in PBS. Neat NC controls and NC films loaded with 0.005% or 0.01%
w/w Cur or Res were tested. To indicate the targeted therapeutic range, resveratrol
solutions of 5 µM and 25 µM were also tested. Previous analysis of resveratrol
administration suggested that from 5 to 25 µM was an effective dosage to modulate
129 inflammation in neural tissue.[15, 36] At set time intervals (t = 1, 3, 24, 48, 72 hours), the incubation solution was removed and replaced with fresh complete PBS with 0.5% v/v
Tween-80. The removed incubation solution was mixed 1:1 with DPPH (100 µM diluted in 95% ethanol) in a 96 well plate. The plate was incubated at room temperature for 30 minutes, and then the absorbance of the solution was measured spectrophotometrically at
516 nm. The radical scavenging activity of the loaded films was expressed using Equation
5.
(5) % = 𝐴𝐴−𝐵𝐵 100 𝐴𝐴 𝐶𝐶 where A was the absorbance 𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷values at𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 time𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 t of the𝑎𝑎𝑎𝑎 DPPH𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 solution with∗ neat NC and B was the absorbance values at time ‘t’ of the DPPH solution with Res or Cur loaded NC. Values were normalized to C, the average absorbance of DPPH with 25 µM resveratrol, the upper limit of the targeted therapeutic range. All samples were tested three times per trial, and three trials were performed to ensure consistency and repeatability of the measurements.
5.3.8 Animal surgery
Surgical procedures for probe implantation closely followed established protocols, with minor changes [24, 261, 262]. Briefly, seventy-two male Sprague Dawley rats (250 –
300 grams) (Charles River, Spencerville, OH) received implants and were euthanized after three days or two or sixteen weeks. Prior to surgery, animals were anesthetized with ketamine (80 mg/kg) and xylazine (10 mg/kg) administered intraperitoneally (IP). Once anesthetized, the surgical area was shaved, and the animal was mounted on a stereotaxic frame and maintained on isofluorane (0.5-2%). A single injection of local anesthetic,
Marcaine (0.5%), was administered below the incision site subcutaneously (SQ), then the
130 surgical area was scrubbed with betadine and 70% isopropanol for sterilization. Animal
body temperature was maintained on a circulating water pad and vitals were monitored
using a blood-oxygen and heart rate measurement system (PulseSense, Nonin Medical,
Inc,).
First, a one-inch incision was made at midline to expose the skull, using a scalpel.
The surrounding tissue was retracted and a 3 mm hole was created in the skull, manually,
using a biopsy punch (P/N #536, PSS Select), approximately 3 mm lateral to midline and
4 mm caudal to bregma. Then the dura was reflected using a 45 degree dura pick. Animals
received either a NC only, 0.01% resveratrol-releasing NC (NC Res), or 0.01% curcumin- releasing NC (NC Cur) implant (n=6-8 for each time point), randomizing which hemisphere was implanted. All implants were inserted approximately 2 mm deep into the cortical tissue by hand, by the same skilled surgeon. Implants were implanted perpendicular to the surface of the brain, to minimize the footprint of tissue damage, while avoiding larger vasculature. Following implantation, implants were tethered to the skull using Kwik-sil (World Precision Instruments) and UV-cured dental acrylic (Fusio-liquid
dentin, Pentron Clinical) over the surgical area and skull. The incision was then closed with
5-0 monofilament polypropylene suture (Henry Schein) and a triple antibiotic ointment
was applied to the incision. Once the animal woke from anesthesia, meloxicam (5 mg/kg,
SQ, once a day for 2 days) and cefazolin (16 mg/kg, SQ, twice a day for 2 days) were
administered for potential pain and to prevent infection. Surgical procedures and animal
care practices were performed in accordance with the Case Western Reserve University
Institutional Animal Care and Use Committees (IACUC).
131 5.3.9 Tissue processing
At three days, two weeks, or sixteen weeks post-implantation, animals were anesthetized using an intraperitoneal injection of ketamine (80 mg/kg) and xylazine (10 mg/kg). Each animal was perfused transcardially with 1X PBS (Invitrogen) until the exudate was clear. The brain was carefully removed, placed in fresh 10% sucrose and stored at 4° C for up to one week. Prior to sectioning, the tissue was cryoprotected in a step-wise gradient of 10%-20%-30% sucrose (Sigma) in 1X PBS at 4° C, until equilibrium was reached at each step. After equilibration in 30% sucrose, the tissue was frozen at -80°
C in optimal cutting temperature compound (Tissue-Tek), sliced axially in 20 μm sections at -25° C, and mounted on SuperfrostTM Plus microscope slides (Fisher Scientific) to be stored at -80° C until immunohistochemical labelling.
5.3.10 Immunohistochemistry
In preparation for immunostaining, the tissue was removed from -80 °C and placed in room temperature in a humidity chamber to prevent dehydration of the tissue. The tissue was fixed with 4% paraformaldehyde for 15 minutes and then washed three times with 1X
PBS, followed by conditioning in 1X PBS with 0.1% Triton-X 100 (Sigma) (1X PBS-T) for 15 minutes. Tissue was then blocked with goat serum blocking buffer (4% v/v serum
(Invitrogen), 0.3% v/v Triton-X 100, 0.1% w/v sodium azide (Sigma)) to prevent nonspecific binding. The tissue was incubated in primary antibody diluted in goat serum blocking buffer at 4 °C overnight. The following primary antibodies were used: mouse anti-glial fibrillary acidic protein (GFAP) (1:500, Invitrogen) to stain astrocytes, rabbit anti-immunoglobulin G (IgG) (1:100, AbDSerotec) to quantify blood brain barrier
132 stability, mouse anti-CD68 (ED1) (1:100, Chemicon) to mark activated microglia/macrophages, and mouse anti-neuronal nuclei (NeuN) (1:250, Millipore) to mark neurons. On the following day, the tissue was washed six times for 5 minutes each
with 1X PBS-T, then incubated for two hours at room temperature with Alexa Fluor
conjugated secondary antibodies (1:1000 in goat serum blocking buffer) and
counterstained with 4’,6-diamidino-2-phenylindole (DAPI) to stain total cell nuclei. The
tissue was again washed six times with 1X PBS-T for 5 minutes each. To remove tissue autofluorescence, the tissue was placed in 0.5 mM copper sulfate buffer (50 mM
Ammonium Acetate, pH 5.0) (Sigma) for 10 minutes.[261] Finally, the tissue was rinsed with distilled water and mounted with Fluoromount-G (Southern Biotech) on microscope slides for imaging.
5.3.11 Quantitative analysis
Following immunohistochemistry, slides were imaged using a 10X objective on
AxioObserver Z1 (Zeiss) equipped with an AxioCamMRm (Zeiss). MosaiX software
(Zeiss) was used to capture and stitch 16 separate images to provide a larger field of view without compromising resolution. Exposure times were held constant for each cellular marker. The acquired images were then linearized and exported as 16-bit tagged image file
(TIF) with minimal compression to optimize intensity analysis. Representative images have been slightly enhanced only for presentation purposes to accurately reflect our experimental data.
For all images except NeuN, a custom MATLAB program (MINUTE) was used to analyze the fluorescent intensity profiles as a function of distance from the implant site.
133 For each image, the implant hole was manually defined. The program quantified the intensity of cellular marker expression for each 2 µm bin up to 1500 µm away from the edge of the implant hole. The quantified fluorescent intensities were then normalized to background which was defined as the average intensity 1000 to 1050 µm away from the interface. The fluorescent intensity profiles as a function of distance from the implant site were acquired for each image. For each stain, six slices per animal were analysed and averaged by animal. The area under the curve (AUC) for each profile was calculated using
MATLAB and averaged for each stain, time point, and implant type at 0 to 50, 50 to 100,
100 to 150, and 150 to 200 µm from the implant-tissue interface.
To quantify the neuronal population around the site of implantation, the implant hole was manually defined in Adobe Photoshop. Concentric rings up to 200 µm from the implant were created and the number of neurons within each ring was manually counted.
The neuronal areal density was calculated and converted to percent to sham background by normalizing to neuronal density from age-matched sham animals for each time point.
5.3.12 Statistical analysis
For all in vitro experiments, a minimum of three trials (N=3, n=9) were used for statistical evaluation (N≥3). Statistical analysis was determined with Minitab 16. For the
Live/Dead assays a one-way analysis of variance (ANOVA) with a Tukey’s post-hoc test was used to find significance. For all other assays, a standard student t-test was utilized to determine significance. To compare the fluorescent intensity profiles between different implant types in vivo, the AUCs for each stain were averaged per animal (N=6-8) and analysed by using a general linear ANOVA model. For NeuN, the number of neurons per
134 area was used for statistical analysis. In all cases statistical significance was defined as
p<0.05.
5.4 Results
5.4.1 Material Characterization
Unloaded, dry NC implants readily insert through the pia mater into the cerebral
cortex of a rat without the need for assistive devices, and dynamically soften in brain
tissue.[23, 211] Therefore, it is important to ensure that the addition of Res or Cur did not
alter the material properties, and inhibit either the ability to implant the polymer devices,
or the ability to dynamically soften within cortical tissue. Here, the mechanical properties
of both Cur-loaded and Res-loaded NC films were investigated using dynamic mechanical
analysis (DMA). Table 3 summarizes the tensile storage moduli (E’) of the antioxidant-
loaded NC as well as the neat NC control film in the dry and ACSF-swollen state as a
function of temperature. All of the materials studied undergo a pronounced modulus
reduction from ~6000 MPa to ~10 MPa upon exposure to (emulated) physiological
conditions, due to matrix plasticization and/or decoupling of CNCs on account of
competitive hydrogen bonding with water. For example, the 3% w/w Cur NC films display
a reduction of E’ from 6346 MPa (dry, RT) to 14 MPa (ACSF-swollen, 37 °C), and the 3%
w/w Res NC films exhibit a similar reduction of E’ from 6340 MPa (dry, RT) to 14 MPa
(ACSF-swollen, 37 °C) (Table 3, Figure S2). This is consistent with previous studies in
our labs in which NCs with 15% v/v CNCs exhibited adequate mechanical switching upon
exposure to physiological conditions and upon implantation into cortical tissue (E' = 5.1
GPa for a dry/pre-insertion and 12 MPa for ACSF-swollen NC).[22, 23, 207] Further,
135 Table 3 shows that in the compositional range studied, neither the amount nor the nature of the antioxidant had a significant influence on the mechanical properties. Additional characterization of antioxidant-loaded NC indicated no changes in swelling behaviour
(Figure S3) and thermogravimetric properties (Figure S4) from addition of either
antioxidant at all concentrations.
Table 3. Storage moduli (E') of dry and ACSF-swollen NCs determined by DMA. Data represent averages (n=5) and are shown for the neat NC as well as for curcumin (Cur) and resveratrol (Res) releasing NCs loaded with different contents of these antioxidants. Swollen Antioxidant Dry Nanocomposites Nanocomposites Nanocomposite Content at 25 °C at 70 at 37 °C after (% w/w) ′(MPa) °C′ (MPa) 1 week′ in ACSF (MPa) Neat NC 6300𝐸𝐸 ± 190 720𝐸𝐸 ± 180 𝐸𝐸 13 ± 1.8 NC Cur 0.005 6300 ± 270 930 ± 98 9.0 ± 2.5 0.01 6000 ± 300 810 ± 75 7.2 ± 3.2 1.0 6100 ± 27 680 ± 45 10 ± 1.0 3.0 6346 ± 100 740 ± 99 14 ± 1.4 NC Res 0.005 6100 ± 350 970 ± 72 7.0 ± 3.6 0.01 6050 ± 140 900 ± 85 6.0 ± 1.5 1.0 6100 ± 197 851 ± 67 13 ± 2.0 3.0 6340 ± 135 840 ± 205 14 ± 2.5
5.4.2 Antioxidant release profile
Systemic administration of resveratrol demonstrated improvement in neuronal
populations for up to four weeks after the initial dose in rats.[15] Subsequent biodistribution analysis revealed anti-inflammatory efficacy of 5-25 µM resveratrol delivered to the brain. Therefore, the therapeutic target range in the surrounding tissue was
~25 µM. Here, the optimal drug concentration for the resveratrol and curcumin within loaded films was determined using the results from finite element analysis, in vitro antioxidant release characterization, and cell culture.
136 Finite element analysis indicated that drug was released up to 5 µm away from the
implant surface (Figure 28). Additionally, the analysis showed that 42% of the antioxidant
was released by 2.7 hours and 100% released by approximately 24 hours. Initial amounts
of antioxidant within in vivo implants (50 µm x 100 µm x 2mm) were estimated as %
weight of the film using Equation 6 and 7.
(6) = ( ) (% )
𝑊𝑊𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑊𝑊𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑥𝑥 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 (7) = 𝑊𝑊𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 𝑉𝑉𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑉𝑉𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 where Wantiox is the weight of antioxidant in a film sample, Wfilm is the total weight of a
film sample, Vsample is the volume of the film sample, Wimplant is the unknown weight of
antioxidant in an in vivo implant, and Vimplant is the volume of an in vivo implant. Calculated values are indicated in Table 4. Assuming a diffusion volume of 5 µm from the implant surface and the starting concentration of the films, the 0.005% film fell within the target range. However, given that the FEM model did not account for metabolic breakdown, chemical reactions, or swelling properties of the NC, it is assumed that the release may be much lower than the model indicates and thus, higher concentration films were created.
Figure 28. FEM estimation of in vivo drug release. A model of a 0.01% antioxidant-loaded NC implanted within the brain indicates infiltration up to 5 µm from the tissue-probe interface and release within the first 24 hours.
137 Table 4. Estimated amount of antioxidant contained within an in vivo implant. Amount of antioxidant in implant (ng)
0.005% 0.01% 1% 3% Resveratrol 0.91 1.82 182.4 547.2 Curcumin 1.04 2.08 208 624
In vitro antioxidant release studies were performed to establish the timeframe in which the antioxidants are released from the NC under (emulated) physiological conditions. The release studies were carried out by submerging antioxidant-loaded NC films in ACSF at 37 °C and monitoring the UV absorbance of curcumin or resveratrol in the supernatant solution as a function of time. The cumulative release percentage, relative to the total amount of antioxidants nominally included in the composition, is shown in
Table S1. After the drug release experiments were carried out, the drug-loaded films were found to be colourless, indicating that the majority of the antioxidants loaded in NC films were released (100% release) into the medium within a timeframe of ~70 hours. However, only about ~45% w/w of the curcumin and ~60% w/w of the resveratrol comprised in the
NC was detected by UV-vis after 72 hours (See Table S.1).
The fact that the percentage of antioxidants released from the NC films was lower than 100% w/w could be due to a number of factors. For example, decomposition of fractional amount of the antioxidants during processing conditions, which can cause undesirable hydrolysis reactions in the active antioxidant compounds. Several factors have been reported in the literature regarding to the degradation of either curcumin or resveratrol such as light, solvents or pH conditions.[319, 320] We found (data not shown) that antioxidants studied here are decomposed upon exposure of the antioxidant solutions
(dissolved in DMF) to oxygen at room temperature. Based on these results, the data presented in Table S.1 were normalized as antioxidant released percentage versus time
138 (Figure 29.). Additionally, as expected, the absolute amount released from the films scaled
with the drug content; the NC + 3% Cur released ca. 3.0 times more curcumin than the 1% curcumin-loaded PVAc/CNC films and the NC + 3% Res films released ca. 3.5 times more resveratrol than the NC + 1% Res films after 72 hours incubation in ACSF at 37 °C.
Unfortunately, it was not possible to detect the release of the antioxidants photometrically in low dose nanocomposites (0.005 and 0.01% w/w) due to the limited sensitivity of the technique. However, the relative percentage of antioxidant released was assumed to hold for the low dose nanocomposites.
Figure 29. Normalized drug release percentage of curcumin (Cur)-loaded NC and resveratrol (Res)-loaded NC in ACSF at 37 °C. All NC contain 15% w/w CNCs and 1% or 3% Cur or Res, as indicated in the figure. Values represent means ± standard deviations of N = 4 experiments.
5.4.3 In vitro validation
In order to determine the optimal antioxidant concentration for in vivo implants, it
was important to investigate the potential cytotoxicity and antioxidative activity of the
various concentrations. To this end, in vitro cytotoxicity assays were performed to ensure
safety of released antioxidant levels. Further, antioxidative activity assays (DPPH and DHE
139 staining), were utilized to assess the efficacy of released antioxidants in the reduction of
accumulated ROS.
5.4.3.1 Glial Cell Cytotoxicity
Cell viability assays were initially performed by exposing BV2 microglia to neat
NC samples or NC films containing 0.005%, 0.01%, 1%, or 3% Res or Cur (Figure 30).
Exposure to neat NC films had no effect on cell viability when compared to a cells only control. Additionally, exposure to 0.005% and 0.01% Res or Cur containing films demonstrated no substantial cell death. However, 1% and 3% Res or Cur NC films induced considerable cell death compared to controls. NC with 1% or 3% w/w Cur or Res films were not used in further experiments due to the exhibited cytotoxicity.
140
Figure 30. Cytotoxicity of antioxidant-releasing NC with microglia cells. BV-2 microglia exposed to 0.005% and 0.01% Res (B,C) and Cur (G,H) NC do not induce cell death, and look similar to antioxidant-free controls without materials (A) or with neat NC (B). 1% and 3% Res (D,E) and Cur (I,J) NC induce significant cytotoxicity. Green indicates live cells and red indicates dead cells. Scale bar = 100 µm.
141 5.4.3.2 Antioxidant Activity
Next, the DPPH assay was utilized to investigate the antioxidative activity of
0.005% and 0.01% resveratrol and curcumin containing NC films. Reduction of DPPH by
resveratrol and curcumin indicates the ability to reduce ROS species. In Figure 31, the
targeted therapeutic range of 5- 25 µM is denoted by the grayed region and values were normalized such that reduction by 25 µM resveratrol is defined as 100% DPPH reduction.
Assessment of 0.005% and 0.01% Res or Cur containing NC films shows that 0.005%
PVAc/CNC films do not exhibit antioxidative efficacy within the targeted range. For both
0.01% Res and Cur containing NC, antioxidative activity was maximal by 24 hours (76% for 0.01% Res, 67% for 0.01% Cur) and within the therapeutic range for up to 48 hours.
Figure 31. Antioxidative activity of 0.005% and 0.01% antioxidant-releasing NC. DPPH assay in PBS shows that 0.01% Res and Cur exhibit antioxidative activity within the targeted therapeutic range of 5-25 µM for up to 48 hours.
142 5.4.3.3 Intracellular ROS and Neuron Cytotoxicity
Films with 0.01% w/w antioxidants were found to have no cytotoxic effects on glial cells and exhibited antioxidant activity within the target therapeutic range. Further in vitro experiments were performed to confirm neuron viability and ROS reduction with 0.01% antioxidant containing films. Since neurons exhibit higher cytotoxic sensitivity, 0.01% containing films were also exposed to neuron cultures for 48 hours (Figure 32a). Both
resveratrol and curcumin NCs showed no effect on neuronal morphology or processes. To
determine potential effects on intracellular ROS production, DHE fluorescence was
detected on microglia exposed to 0.01% containing films for 48 hours (Figure 32b).
Exposure of microglia to NC surfaces alone did not affect intracellular ROS. Quantification of DHE fluorescence showed that microglia incubated with NC Res and NC Cur had significantly less intracellular ROS than neat NC controls. Therefore, probes for implantation were fabricated from 0.01% resveratrol-containing NC (NC Res) and 0.01% curcumin-containing NC (NC Cur) films for in vivo characterization.
143
Figure 32. In vitro characterization of NC films containing 0.01% Res and Cur. (A) Neuronal outgrowth and viability are unaffected by NC or 0.01% antioxidant-releasing NC films. Scale bar = 100 µm. (B) Intracellular ROS accumulation measured by DHE fluorescent dye. Microglia exposed to NC Res and NC Cur had significantly less DHE fluorescence compared to NC only. *p<0.05 compared to NC only
5.4.4 In vivo characterization
Following validation of in vitro cytocompatibility and antioxidative activity, neural implants were analysed for their ability to reduce the neuroinflammatory response compared to non-loaded NC. Immunostaining of brain slices for various neuroinflammatory markers at the implant-tissue interface was performed for several time points spanning acute and chronic inflammation. Insertion of intracortical microelectrodes induces tissue and vascular damage, allowing infiltration of blood cells and proteins into the brain tissue. Immunostaining for immunoglobulin G (IgG) indicated damage and permeability of the blood brain barrier (BBB).[102] Microglia/macrophages play a major role in the brain’s immune response to foreign implants, inducing recruitment of
144 inflammatory cell types and releasing cytotoxic factors such as reactive oxygen species
(ROS). Up-regulation of activated microglia/macrophages around the implant was
indicated by CD68+ immunostaining.[268] The extent of astrocytic encapsulation
following electrode implantation was measured with GFAP+ immunostaining.[266] Most
importantly, to ensure high quality neural recordings, neurons should be in close proximity
of the electrode.[89] In order to quantify the neuronal density around the implant, neuronal
nuclei (NeuN) was counted as a function of distance from each type of implant.
5.4.4.1 Three Day Time Point
At three days post- implantation, no significant differences were apparent for any
of the inflammatory markers (Figure 33). No substantial glial scarring was visible around
any of the implant types, regardless of antioxidant loading (Figure 33A-C,M). A small but diffuse accumulation of serum proteins were observed up to 800 µm away from the interface for all implant types (Figure 33D-F,N). Similarly, the population of activated microglia/macrophages (CD68+) was detected as far as approximately 600 µm away from all types of implants (Figure 33G-I,O). However, no significant difference was observed between any implant types. All implant conditions also caused significant neuronal dieback
(Figure 33J-L,P), reducing neuronal populations to ~40% near the implants (0 to 50 µm), compared to age-matched sham animals. There was no significant difference between the different types of implants.
145
Figure 33. Assessment of the initial inflammatory response at three days post-implantation. No significant difference between types of implants was seen three days after implantation. All implants induced significant neuronal loss at 0 to 50 µm from the interface. #p<0.05 compared to age-matched sham controls (dotted line). White line denotes electrode location. Scale bar = 100 µm.
146 5.4.4.2 Two Week Time Point
The most significant differences between implant conditions were seen at the acute neuroinflammatory time point (two weeks post-implantation) (Figure 34). NC Res implants exhibited significantly improved neuronal survival around the implant at 0 to 50
µm compared to other implant types (Figure 34J-L,P). The neuron density around NC Res was statistically similar to that of the age-matched sham at 0 to 50 µm away from the implant site. However, NC and NC Cur implants had significant neuronal dieback at 0 to
50 µm away compared to age-matched sham animals. At distances greater than 50 µm from the interface, all implant types exhibited full recovery of neurons compared to age-matched sham animals. Additionally, expression of CD68+ microglia/macrophages was significantly reduced around NC Res implants (Figure 34G-I,O). The CD68+ expression extended to ~300-400 µm for all implant types. However, NC Res implants exhibited significantly less CD68+ immunostaining at 0 to 50 µm away from the implant.
While improved neuronal populations directly correlated with reduced CD68+ cell populations, astrocytic scaring and BBB instability did not. For all conditions, glial scarring extended up to 600 µm from the implant, and all conditions were statistically similar
(Figure 34A-C,M). Further, IgG+ immunostaining showed statistically similar profiles between all three conditions, with staining extending up to ~500 µm from the implant
(Figure 34D-F,N).
147
Figure 34. Assessment of the inflammatory response at two weeks post-implantation. NC Res implants exhibit a significant reduction in CD68+ staining and improvement in neuron density from 0 to 50 µm from the interface. *p<0.05 between NC vs NC Res and NC Res vs NC Cur. #p<0.05 compared to age-matched sham controls (dotted line). White line denotes electrode location. Scale bar = 100 µm.
148 5.4.4.3 Sixteen Week Time Point
NC Cur implants did not exhibit any benefit at either three day or two week time points, compared to no treatment controls. Therefore, NC Cur implants were excluded from the chronic study. At sixteen weeks post-implantation, there was no significant difference between NC and NC Res implants for any of the inflammatory markers (Figure 35). For both NC and NC Res, the glial scarring was more compacted, extending up to 300-400 µm from the implant site (Figure 35A-B,I). IgG+ immunostaining was low at sixteen weeks, and similar to earlier time points, there was no significance between the different implant types (Figure 35C-D,J). There was no significance between the expression of CD68 around NC and NC Res implants at sixteen weeks post-implantation (Figure 35E-F,K).
Further, no significant difference in neuronal density was observed between NC and NC
Res at any distance from the implant (Figure 35G-H,L). Significant reduction in neuron density (~50%) was evident from 0 to 50 µm from the interface for NC Res compared to age-matched sham animals. However, at further distances, full recovery to native levels was evident for both implants.
149
Figure 35. Assessment of the inflammatory response at sixteen weeks post-implantation. NC Cur implants were not tested up to sixteen weeks, due to lack of efficacy at earlier time points. For all markers tested, no significant difference was seen between NC and NC Res implants. #p<0.05 compared to age-matched sham controls. White line denotes electrode location. Scale bar = 100 µm.
5.5 Discussion
The characteristics of the tissue surrounding microelectrodes play a crucial role in the stability and integration of implanted devices. Recent literature has supported the
150 hypothesis that the integrity of the blood-brain barrier (BBB) may be a major factor in the
neuroinflammatory response to microelectrodes. Specifically examination of the tissue
response surrounding implanted microwire and Michigan-style microelectrodes revealed that the BBB integrity immediately surrounding the implantation site was compromised.[12, 13] Additionally, the extent of neuroinflammation has been correlated to the proximity of the implant to large vessels.[76] Most important to device function, several studies have begun to correlate microelectrode performance the stability of the
BBB. For example, Kozai et al. demonstrated that increased bleeding intensity can cause
a reduction in SNR[162], while Saxena et al. correlated the long-term stability of functional
single unit recordings with changes in BBB permeability.[11] The study by Saxena et al.
was the first to directly relate increased BBB leakiness to reduced quality of neural
recordings.
Since the integrity of the BBB appears to play a critical role in intracortical
microelectrode performance, we have explored several mechanisms to inhibit potential
causes for BBB breakdown in parallel. The administration of antioxidants were able to
improve BBB integrity and neuronal proximity at acute time points.[14, 15] However,
long-term delivery methods are still under investigation, to minimize potential undesirable
consequences. Alternatively, our initial NC implants designed to reduce tissue strain were
able to improve neuronal proximity for up to sixteen weeks post-implantation.[19, 20]
However, fluctuations in BBB integrity and microglial activation were noted during the
implantation period, which may affect neuron recording capabilities throughout device
lifetime. This study describes a strategy to combine mechanically-adaptive NC materials
151 with therapeutic antioxidant delivery as a cooperative system to improve BBB healing and
implant integration.
At both three days (Figure 33) and sixteen weeks (Figure 35) post-implantation,
the NC Cur and NC Res implants had no discernible difference from NC controls. At three
days, the insertion damage may be overwhelming any potential actions of the antioxidants.
This suggests that wound healing events in response to insertion damage may be the
predominant issue immediately following implantation. In this case, reduction of insertion-
mediated damage should be prevented. Several groups have investigated the effects of
insertion speed, tip geometry, and probe size during implantation. [27, 84, 111, 112] .
Efforts by Kozai et al. to reduce insertion damage have explored utilizing ultrasmall probe or avoiding large vasculature with in vivo two-photon microscopy.[28, 113-115] At sixteen weeks post-implantation, it was expected that the effects of the antioxidant delivery would wear off, due to complete release of antioxidant, and there would be no difference between
NC and NC Res implants (as seen in Figure 35). This result was actually a goal in the design of the combinatory antioxidant and compliant implant approach. We set out to maximize the short-term advantages of antioxidant therapy, without the possible long-term dosing complications; while also benefiting from the long-term neuroinflammatory improvements provided by our more compliant NC implants.
Characterization of antioxidant-containing films used in this study indicated no changes in material properties due to incorporation of either resveratrol or curcumin.
(Table 1, Figure S2-S4) While in vitro release and antioxidative activity were noted up to
48-72 hours (Figure 29, Figure 31), in vivo results exhibited benefits from NC Res (0.01% resveratrol-containing NC) for at least two weeks post-implantation (Figure 34).
152 Interestingly, no difference in the inflammatory response surrounding neat NC and NC Res
implants was seen at three days post-implantation. Delayed reaction to resveratrol delivery may be due to the mechanism by which resveratrol affects inflammation. Resveratrol can directly neutralize and scavenge free radicals. However, it can also inhibit mRNA production of pro-inflammatory cytokines and inhibit NF-kB and NOS expression.[321]
Reduction in pro-inflammatory molecules and expression of inflammatory genes may induce changes in inflammatory cell phenotype and lead to a delayed change in environmental neurotoxicity.
A previous report of curcumin delivery from PVA implants showed higher neuronal survival and a more stable BBB.[14] However, in the current study, curcumin release provided no benefit compared to NC only. A major difference between studies is that the
PVA study utilized 3% curcumin containing films, which in our present study induced glial cell cytotoxicity. Additionally, 3% Cur/PVA implants still exhibited significant neuronal loss compared to age-matched sham animals. Solubility of curcumin may have affected the diffusion kinetics in this study. Curcumin bioavailability, due to its poor aqueous solubility, is a widely-reported issue and many groups have made attempts to improve delivery.[322]
Table S1 also shows that only ~50% of curcumin was released after 72 hours. Therefore, diffusion of curcumin through the polymer and tissue may be prevented due to severe hydrophobicity of curcumin.
Delivery of resveratrol has proven more successful than curcumin in improving the inflammatory response to intracortical implants (Figure 34). Here, NC Res implants reduce activation of microglia compared to neat NC implants. Significant reduction in microglial activation can influence the inflammatory environment and allow for increased neuronal
153 survival, which is seen in Figure 34K,P. Further, the neuron viability from NC Res
implants indicated a similar improvement to a single systemic dose of resveratrol, showing
~75% neuronal nuclei density at two weeks. However, the systemic resveratrol study
indicated reduction in BBB permeability at two weeks with resveratrol dosed animals,
which is not seen in the present study (Figure 34N). Systemic antioxidative delivery may have directly affected blood-derived cells and factors involved in inflammation, which
have been found to be the predominant cells involved in mediating
neurodegeneration.[323]. However, resveratrol still exhibits neuroprotection regardless of
type of delivery.
Our lab continues to investigate antioxidative strategies. Experiments are underway
to improve resveratrol delivery to the brain and investigate feasibility of continual
injections for a more prolonged effect. Alternatively, our lab has explored the use of
surface-conjugated superoxide dismutase (SOD) mimetics to provide sustained ROS
scavenging around silicon implants.[36] SOD mimetics could potentially be conjugated
and released from the NC for another combined approach to modulate neuroinflammatory
events.
5.6 Conclusions
In summary, the results of this study demonstrate that the addition of local
antioxidant delivery can further improve neuron density surrounding already beneficial
compliant implants. Incorporation of drug into the NC had no effect on material properties
and antioxidant release was expected for up to 72 hours. With the in vivo studies, at three
days, insertion damage and wound healing events dominated the response, showing no
154 difference in neuroinflammatory events surrounding neat NC and with antioxidant
releasing NC implants. However, beneficial effects of local resveratrol delivery were seen
at two weeks in vivo, with significant reduction in microglial activation and improvement
in neuron density. NC only and NC Res exhibited similar tissue responses sixteen weeks
post-implantation when the antioxidants were exhausted. Therefore, NC Res implants
exhibited short-term benefits of antioxidant treatment and the long-term reduction in
neuroinflammation by compliant implants. Together, these results demonstrate that local
delivery of antioxidants can provide an additive advantage to enhance attenuation of the
tissue response to compliant implants.
5.7 Supplementary Data
5.7.1 Materials and Methods
5.7.1.1 Microscopy Studies.
Atomic force microscopy (AFM) and transmission electron microscopy (TEM) were used to characterize the morphology and physical dimensions of the isolated CNCs.
Atomic force microscopy of CNCs was carried out on a NanoWizard II (JPK Instruments) microscope. 5 µL of the CNCs re-dispersed in DMF (0.01 mg/mL, 30 min sonication) was placed on freshly cleaved mica (SPI Supplies Division of Structure Probe, Inc.) and allowed to dry. The scans were performed in tapping mode in air using silicon cantilevers
(NANO WORLD, TESPA-50) with a scan rate of 1 line/second.
A Hitachi H-1700 microscope operating at an accelerating voltage of 75 kV was used to examine the dimensions of the CNCs and the homogeneity of the CNC dispersion by TEM micrographs. To assess the CNCs dimensions, samples were prepared by dropping
155 3 µL of the aqueous CNC dispersions (0.01 mg/mL) on carbon-coated grids (Electron
Microscopy Sciences) and allowed to dry in an oven at 70 °C for 2 h.
5.7.1.2 Swelling Behavior
The swelling behavior of the PVAc/CNC control nanocomposite and the
antioxidant-loaded nanocomposite films was investigated by cutting the films into ~30 mm
× ~6 mm × ~100 μm rectangular strips and immersing the samples in artificial
cerebrospinal fluid (ACSF) at the physiological temperature of 37 ºC for one week. After
one week incubation in ACSF, the degree of swelling was calculated by comparing the
weight of the films pre- and post-swelling:
(Ws - Wd) (S1) Degree of swelling (%) = × 100 Wd
where Wd is the weight of the dry film prior to swelling and Ws is the weight of the swollen
film. After the swollen films were removed from the ACSF, the samples were briefly
placed on a small piece of tissue paper to wick any excess ACSF from the surface, and the
samples were immediately weighed. This experiment was repeated four times, and the
results are expressed as mean ± standard deviation.
5.7.1.3 Thermogravimetric Analysis
Thermogravimetric (TGA) analysis of the PVAc/CNC reference materials and the antioxidant releasing PVAc/CNC nanocomposites were performed with a thermogravimetric analyzer (Mettler Toledo STAR). Each sample (~10 mg) was heated in an aluminum pan from 30 to 600 °C at a flow rate of 10 °C/min under nitrogen.
156 Table S1. Cumulative release (%) data of curcumin (Cur)-loaded PVAc/CNC nanocomposites and resveratrol (Res)-loaded PVAc/CNC nanocomposites in ACSF at 37 °C determined by UV-vis. All PVAc/CNC nanocomposites contain 15% w/w CNCs and 1 or 3% Cur or Res, as indicated in the Table. Values represent means ± standard deviations of N = 4 experiments, and relative to the original nominal amount of drug loaded in the nanocomposites. Time 1% w/w 3% w/w 1% w/w 3% w/w (hours) Cur/PVAc/CNCs Cur/PVAc/CNCs Res/PVAc/CNCs Res/PVAc/CNCs 1 5.4 ± 1.6 3.0 ± 0.9 10.9 ± 2.2 6.7 ± 0.4 2 7.5 ± 2.0 4.1 ± 0.9 13.3 ± 3.3 10.1 ± 0.7 4 9.6 ± 2.6 8.4 ± 0.4 19.8 ± 0.8 15.1 ± 2.2 6 16.5 ± 2.6 12.4 ± 0.6 24.0 ± 0.8 20.2 ± 1.8 8 17.9 ± 1.5 14.9 ± 0.8 27.8 ± 1.6 23.2 ± 3.4 10 19.9 ± 2.9 18.0 ± 0.3 30.7 ± 4.5 27.3 ± 4.4 24 31.3 ± 2.4 32.5 ± 2.7 45.8 ± 5.5 44.0 ± 5.1 29 32.4 ± 3.0 34.5 ± 2.8 47.0 ± 5.2 47.8 ± 2.4 34 34.7 ± 3.5 37.5 ± 1.8 50.1 ± 3.5 50.0 ± 4.8 48 43.4 ± 3.7 42.0 ± 2.0 59.1 ± 4.1 55.4 ± 4.6 55 44.9 ± 4.2 43.8 ± 2.4 61.2 ± 4.2 56.8 ± 5.2 72 48.7 ± 4.8 46.1 ± 1.9 64.3 ± 4.0 59.4 ± 4.6
Figure S1. Chemical structures of materials used in this study.
157 Figure S2. Representative dynamic mechanical analysis (DMA) traces showing the storage moduli E’ of (A) dry Cur/PVAc/CNC, (B) dry Res/PVAc/CNC, (C) ACSF-swollen Cur/PVAc/CNC, and (D) ACSF-swollen Res/PVAc/CNC nanocomposites as a function of temperature and drug content (0.005, 0.01, 1.0 or 3.0% w/w). The neat PVAc/CNC nanocomposite was also studied for reference purposes and all PVAc/CNC nanocomposites contain 15% w/w CNCs. Average data of multiple experiments are compiled in Table 5.1.
158 Figure S3. Swelling behavior of PVAc/CNC nanocomposites (15% w/w CNCs) with different amounts of antioxidants after the films were immersed in ACSF at 37 °C for 1 week. (N) neat PVAc/CNC nanocomposite; (C1) 0.005% w/w curcumin (Cur); (C2) 0.01% w/w Cur; (C3) 1.0% w/w Cur; (C4) 3.0% w/w Cur; (R1) 0.005% w/w resveratrol (Res); (R2) 0.01% w/w Res; (R3) 1.0% w/w Res; and (R4) 3.0% w/w Res-loaded nanocomposites. Data represent averages of N = 3 measurements ± standard deviations.
159 Figure S4. Thermogravimetric analysis traces of (A) curcumin (Cur) and (B) resveratrol (Res) loaded PVAc/CNC nanocomposites (15% w/w CNCs) as a function of temperature.
160 Chapter 6
Development of Superoxide Dismutase Mimetic Surfaces to Reduce Accumulation of Reactive Oxygen Species Surrounding Intracortical Microelectrodes*
*The following chapter is reproduced, with permission, from: Potter-Baker KA*, Nguyen JK*, Kovach KM, Gitomer MM, Srail TW, Stewart WG, Skousen JL, Capadona JR. Journal of Materials Chemistry B, (2014) 2, 2248-2258. DOI: 10.1039/C4TB00125G
6.1 Abstract
Despite successful initial recording, neuroinflammatory-mediated oxidative stress products can contribute to microelectrode failure by a variety of mechanisms including: inducing microelectrode corrosion, degrading insulating/passivating materials, promoting blood-brain barrier breakdown, and directly damaging surrounding neurons. We have shown that a variety of antioxidant treatments can reduce intracortical microelectrode- mediated oxidative stress, and preserve neuronal viability. Unfortunately, short-term
soluble delivery of antioxidant therapies may be unable to provide sustained therapeutic
benefits due to low bio-availability and fast clearance rates. In order to develop a system
to provide sustained neuroprotection, we investigated modifying the microelectrode
surface with an anti-oxidative coating. For initial proof of concept, we chose the superoxide
dismutase (SOD) mimetic Mn(III)tetrakis(4-benzoic acid)porphyrin (MnTBAP). Our
system utilizes a composite coating of adsorbed and immobilized MnTBAP designed to
provide an initial release followed by sustained presentation of an immobilized layer of the
antioxidant. Surface modification was confirmed by XPS and QCMB-D analysis.
Antioxidant activity of composite surfaces was determined using a Riboflavin/NitroBlue
Tetrazolium (RF/NBT) assay. Our results indicate that the hybrid modified surfaces
161 provide sustained anti-oxidative activity. Additionally, in vitro studies with BV-2
microglia cells indicated a significant reduction of intracellular and extracellular reactive
oxygen species when cultured on composite MnTBAP surfaces.
6.2 Introduction
Intracortical microelectrodes are implanted into the cerebral cortex to record
changes in neural activity which can be directly related to a variety of essential behavioural
and motor-based states.[324] Microelectrode-mediated recordings in animals have advanced our fundamental understanding of brain function in both normal and diseased states.[95, 290, 325] In paralyzed individuals, chronic microelectrode recordings promise
a way to provide control of various assistive devices.[1, 285] Unfortunately, the implementation of intracortical microelectrodes for brain computer interface applications has been severely hindered by inconsistent recording and premature microelectrode failure.[81]
The ability of intracortical microelectrodes to record ‘usable’ activity from single neurons is directly related to the proximity of viable neurons to functional recording sites.[89] Therefore, the most widely accepted theories regarding microelectrode failure focus on changes in the viability and function of neurons near the microelectrode recording sites[105] and damage to the electrode itself, including both the recording sites as well as
insulating and passivating coatings.[3, 61] Changes in both viable neuron populations and degradation of the microelectrode itself can be largely attributed to the neuroinflammatory response to the implanted microelectrodes.[286]
162 Consequently, efforts have been made to minimize the reactive tissue response to
intracortical microelectrodes. The most promising strategies have targeted inhibition of
microglia and macrophage activation, or stabilization of the blood-brain barrier through
various materials-based[13, 14, 20, 213] and therapeutic strategies.[9, 15, 152] Of note, we recently identified a key role for oxidative stress-mediated events following microelectrode implantation in the cerebral cortex.[14, 15, 121, 258] We have found that short-term systemic or localized delivery of natural antioxidants, resveratrol or curcumin, can significantly improve neuronal viability and attenuate neuroinflammation encompassing implanted intracortical microelectrodes.[14, 15] However, short term antioxidant administration was unable to provide sustained neuroprotection around implanted devices.
We hypothesized that the lack of sustained neuroprotection was based on fast clearance rates and low bioavailability.[322, 326, 327]
In an attempt to combat the limitations of systemic and/or local antioxidant delivery, multiple groups have demonstrated the success of immobilized anti-oxidative approaches in mitigating inflammatory pathways following device implantation.[32, 295]
For example, Cheung et al. demonstrated that immobilization of a custom superoxide dismutase (SOD) mimetic into a hydrogel system could reduce the formation of reactive oxygen species and improve cell viability.[32, 328]
Therefore, the goal of this study was to develop a sustained anti-oxidative coating for intracortical microelectrode applications, based on the immobilization of mimetic SOD.
Here, we focused on the characterization and in vitro evaluation of a composite coating of the SOD mimetic Mn(III)tetrakis(4-benzoic acid)porphyrin (MnTBAP). Our coating was designed to provide a synergistic initial and sustained anti-oxidative effect. An initial
163 release is provided to minimize neuron loss following device implantation, while a covalently immobilized layer of MnTBAP is designed to regulate chronic neuroinflammation, once deployed under in vivo conditions.
6.3 Experimental Methods
6.3.1 Chemicals and Reagents
Mn(III)tetrakis(4-benzoic acid)porphyrin (MnTBAP) was purchased directly from
EMB Millipore (Billerica, MA). Glass coverslips (12 mm, No. 1.5), ethanol (EtOH) and hydrochloric acid (HCl) were purchased from Fisher Scientific. All other utilized reagents and solutions were purchased from Sigma Aldrich.
6.3.2 MnTBAP Substrate Modification
Immobilization of MnTBAP onto silicon dioxide (glass coverslips) was performed using established protocols for biomolecule immobilization[329], with slight modifications. A summary of the total reaction scheme used here has been shown in Figure
36.
First, glass coverslips were cleaned by submersion in hot (70°C) 2M HCl under constant agitation with a stir bar for 1 hour. Care was taken to ensure that glass coverslips were not shattered during agitation. Following acid treatment, coverslips were washed using an EtOH gradient (95%-70%-50%) and then thoroughly rinsed with deionized water.
Following cleaning, coverslips were dried under a stream of nitrogen and placed in a
Nordson MARCH PX-250 Plasma Cleaning System (Concord, CA) powered by an MKS
ACG-3B RF Plasma Generator (Andover, MA). The coverslips were then exposed to
164 oxygen plasma under vacuum (900 mTorr O2, 25 W, 25 sec.), in order to remove any
remaining contaminants and fully oxidize the glass surface. Following plasma treatment,
the coverslips were stored in deionized water until use to preserve the cleaned and oxidized
surfaces. Plasma treated surfaces were used a maximum of 10 days after initial treatment.
Next, oxidized glass coverslips were placed into a 2% solution of
(3-aminopropyl)triethoxysilane (APTES) diluted in toluene (100° C for 1 hour). The glass coverslips were then thoroughly washed, under sonication, two times each, with toluene,
95% ethanol and deionized water for five minutes each. Following washing, coverslips were dried under a stream of nitrogen and cured (110° C for 15 minutes), and stored at 37°
C in deionized water for up to 2 days, until use.
165
. Reaction scheme for synthesis of MnTBAP modified glass surfaces. glass modified MnTBAP of synthesis for scheme Reaction . 36 Figure
166 In parallel, commercially available MnTBAP (see Section 6.3.1) was prepared for covalent immobilization to modified glass coverslips. To functionalize MnTBAP, 3 mL of a 4.7 mM aqueous solution of MnTBAP was diluted to 2.4 mM in 4 mM EDC (1-ethyl-3-
(3-dimethyl-aminopropyl) carbondiimide hydrochloride), 10 mM NHS (N- hydroxysuccinimide) in 0.1 M 2-(N-morpholino)-ethanesulfonic acid and 0.5 M NaCl, pH
6.0. After 30 minutes at room temperature, the reaction was quenched with 2- mercaptoethanol (20 mM for 5 minutes).
Finally, APTES modified substrates were incubated in functionalized MnTBAP (22
to 24 hours at room temperature). Following incubation, coverslips were thoroughly
washed three times with deionized water, 95% EtOH, and then deionized water. For
assessment of covalent attachment, a subset of MnTBAP functionalized surfaces were
washed three times with deionized water under sonication. After washing, unreacted
succinimidyl esters were quenched in 20 mM glycine for 10 minutes. MnTBAP conjugated
coverslips were then thoroughly washed in deionized water and stored at 37° C in deionized
water until use.
For the purpose of this work, Day 0 was defined as the time point immediately
following the final washing.
6.3.3 Surface Characterization
6.3.3.1 Contact Angle Measurement: Determination of Surface Hydrophobicity of
Substrates
Surface hydrophobicity was measured to track the progression of surface
modification reactions. Briefly, the contact angle of glass coverslips at each major stage of
167 the coating process was measured using the sessile drop method on a custom-built goniometer (deionized water, 8 µl drop size). In the case of APTES and MnTBAP coated coverslips, measurements were done within 5 days of modification. All surfaces were thoroughly dried under a stream of nitrogen before measurement. Images of the water droplets were taken with a Tucsen TCA 1.31C camera and the droplet angles were later
measured using a custom Matlab program. Three separate batches of each surface were
fabricated for analysis (N=3, n=9). The contact angle was measured on three material
samples from each batch and averaged.
6.3.3.2 X-ray Photoelectron Spectroscopy (XPS): Determination of Atomic Composition of Substrates
Coated glass surfaces were analyzed for elemental composition using a PHI
VersaProbe XPS Microprobe (Chanhassen, MN) at the Swagelok Center for Surface
Analysis of Materials, Case Western Reserve University. Each sample was scanned using a 15° incident angle, averaging the signal over a 0.1x1.4 mm area (100 µm ion beam size).
Survey scans were performed over 0 - 1,100 eV with a pass energy of 93.90 eV.
6.3.3.3 Quartz Crystal Microbalance: Determination of MnTBAP Surface Density
Surface density of conjugated and unconjugated MnTBAP was determined using a quartz crystal microbalance with dissipation (QCMB-D; Qsense E1). SiO2 sensors were first cleaned using a 30 minute wash in 2% sodium dodecyl sulphate. Sensors were then thoroughly washed in water, dried under nitrogen, plasma treated and APTES coated as outlined above.
168 APTES coated sensors were then placed in the flow module and a baseline measurement was established by flowing deionized water over the sensor at a rate of 150
µL per minute. All frequency and dissipation measurements were recorded and monitored using the QSOFT software system (Qsense). After a baseline was found, functionalized
MnTBAP was flowed over the sensor at the same rate until a plateau was found (~500 µL).
Upon plateau, the fluid pump was turned off and the MnTBAP was allowed to interact with the surface of the sensor for 22 to 24 hours. Sensors were then washed as described above, until a baseline frequency and dissipation curve was established. Similar to glass coverslips, the sensor surface was then exposed to 20 mM glycine, and washed with water.
The frequency and dissipation baseline following the final water wash was used for quantification.
Data was modelled using the QTOOLS software. Since our system demonstrated viscoelastic properties, surface densities of conjugated MnTBAP were calculated using a
Voigt model[330]. A minimum of three harmonics (overtones 3, 5 and 7) was required for computational modelling. Four separate sensors were tested for analysis (n=4).
6.3.4 Activity of MnTBAP-modified Surfaces
On day 0, 1, and 2 following surface modification, superoxide dismutase activity of MnTBAP-modified surfaces was confirmed using a nitroblue tetrazolium
(NBT)/riboflavin (RF) assay.[331] Within the NBT/RF method, RF generates superoxide anion under illumination which reduces NBT, producing a blue dye.[331] In our system, conjugated MnTBAP competes with NBT, directly inhibiting the formation of reduced
NBT.[34] Therefore we monitored the inhibition of NBT reduction.
169 The NBT/RF assay in this work followed previously outlined methods, with minor
alterations.[332] Briefly, surfaces were placed into 1.5 mL of a reaction mixture containing
phosphate buffer, (50 mM KH2PO4, 0.1 mM EDTA, pH 7.8) 2 mM riboflavin, and 57 µM
NBT. Samples were incubated in the dark at 4° C for 30 minutes. Following incubation,
samples were subsequently illuminated for 15 minutes by two fluorescence tubes
(Sylvania, 32 watts) in a foil lined container. Following incubation, the reaction was
stopped using a 1:1 addition of chloroform under vortexing. Finally, 100 µL of the organic
layer from each sample was measured for absorbance at 560 nm on a SpectraMax M2e
spectrophotometer. The percent activity of MnTBAP-modified surfaces was calculated using Equation 8.
( )
(8) % = 𝑨𝑨−𝑩𝑩 × 𝑨𝑨 𝑪𝑪 here, A is the absorbance 𝑵𝑵𝑵𝑵𝑵𝑵of solutions𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨𝑨 incubated with𝟏𝟏𝟏𝟏𝟏𝟏 plasma treated glass, B is the
absorbance of solutions incubated with the MnTBAP coated surface and C was the
absorbance of the amount of soluble MnTBAP on composite surfaces (see Section 6.3.5).
Assays were run a minimum of three times in triplicate (N = 3, n=9).
6.3.5 Activity of MnTBAP in Solution
To distinguish the impact of the soluble vs immobilized MnTBAP, additional tests
were conducted to determine the activity of soluble MnTBAP in solution. Here, the amount
of mimetic, as determined using the surface densities of MnTBAP measured using QCMB,
was assessed under a NBT/RF assay. Specifically, since our coverslips had an area of 1.13
cm2, and the determined surface density using QCMB was 21.9 µg/cm2, our soluble stock
solution was made by combining 24.75 µg of MnTBAP into 100 µL of 50% ethanol.
170 The NBT assay was run using the same procedure as modified surfaces (See
Section 6.3.4) with one modification. Here, 100 μl of soluble MnTBAP solution (0.2475
µg/µL) was added to reach the final volume of 1.5 mL reaction buffer. Following the assay,
a chloroform extraction was performed as outlined above to ensure MnTBAP absorbance
would not confound the measured NBT absorbance. The chloroform extract was read and
560 nm, and the absorbance was reported according to Equation 8. Where here, A was the
absorbance of ethanol only and B was the absorbance of the MnTBAP containing solution.
6.3.6 Intracortical Microelectrode Implantation and Resveratrol Measurement
To better understand the utility of our surfaces, we compared the anti-oxidative
activity of MnTBAP-modified surfaces with the anti-oxidative activity of detected levels
of resveratrol, a known antioxidant, measured around an implanted microelectrode
following systemic resveratrol administration. All procedures complied with the Case
Western Reserve University Institutional Animal Care and Use Committee (IACUC).
Surgical procedures followed our previously reported methods.[15] Here,
immediately following device implantation, a subset of animals received an intraperitoneal
injection of 30 mg/kg resveratrol, as previously described.[15] At both 1 and 18 hours after surgical implantation, animals were heavily anesthetized using a mixture of ketamine (80 mg/kg) and xylazine (10 mg/kg) and underwent a cardiac perfusion with 1X PBS. The brain tissue was carefully extracted and a 3-mm biopsy punch was used to extract cortical tissue around the implanted intracortical microelectrode. For this extraction, the electrode site was placed in the middle of the punch allowing for an approximate 1.5 mm radius of
171 tissue to be extracted around the implanted device. All samples were placed in 50% ethanol
to extract the resveratrol into the organic phase.
Tissue samples were then thoroughly homogenized and centrifuged to remove tissue and cell debris. The supernatant was measured in a fluorescent plate reader
(excitation: 315 nm, emission: 385 nm) to determine the amount of resveratrol that accumulates in the cortical tissue adjacent to the implanted microelectrode. Absolute values were established for resveratrol using a standard curve generated from stock ethanolic solutions. Data are reported as a concentration based on the extracted tissue volume.
Based on the found concentration range of resveratrol found around implanted electrodes, and as a control, the NBT activity of 5 to 25 µM resveratrol was also tested.
NBT activity was determined following methodology to soluble MnTBAP (See Section
6.3.5). For all assays, a stock solution of 100 µM resveratrol in 50% ethanol was serially diluted until the tested concentration range was obtained. Where, for our outlined assay, concentrations below 5 µM could not be tested due to sensitivity limitations.
6.3.7 In vitro Assessment
Glass coverslips and MnTBAP composite coverslips were used for in vitro
assessment. Before exposure to cells, coverslips were sterilized under ultraviolet light for
15-30 minutes. An immortalized murine microglial cell line (BV-2) was generously donated from Dr. Stephen Selkirk, L. Stokes Cleveland Department of Veterans Affairs,
for in vitro experiments. BV-2 cells were maintained in DMEM Dulbecco’s Modified
Eagle Medium (ATCC) supplemented with 10% fetal bovine serum (Invitrogen) and 1%
172 penicillin-streptomycin (ATCC). For all experiments, BV-2 cells were used between passages 5 to 15. For all in vitro tests, BV2s were seeded at a density of 250 cells/mm2 directly on each coverslip.
6.3.7.1 Intracellular superoxide anion accumulation
Dihydroethidium (DHE) labelling was utilized to measure intracellular superoxide anion accumulation.[318] BV-2 cells were cultured on silicon substrates for 48 hours, washed three times with 1X PBS, and then incubated 3 µM DHE in PBS (30 minutes at room temperature). After three subsequent washes in 1X PBS, samples were transferred to a black well plate and read for total fluorescence (excitation 480 nm, emission 586 nm).
Data were reported as a total fluorescent intensity and was conducted a minimum of three times in triplicate (N = 3, n=9).
6.3.7.2 Soluble nitric oxide and superoxide anion
The levels of soluble nitric oxide were measured from the media of BV-2 cells cultured on sample surfaces for 48 hours using a Griess Reagent Kit (Life Technologies).
Briefly, equal parts of N-(1-naphthyl)ethylenediamine and sulfanilic acid were mixed together to form the Griess reagent. Griess reagent was added to our samples and incubated for 30 minutes at room temperature. The absorbance of the samples was measured using a microplate reader at 548 nm. Using a standard curve of nitrite concentrations, data was reported as total nitrite concentration.
Additionally, superoxide dismutase activity was measured for samples incubated with cells for 48 hours. Here, culture media was collected for analysis following methods
173 described above in Section 6.3.4, with slight modification. Specifically, addition of
riboflavin was not performed, as the amount of reduced NBT from the media was used to
indicate the amount of ROS. In addition, a chloroform extraction was not performed since
extraction methods precipitated serum proteins in the media, complicating signal analysis.
6.3.7.3 Live/Dead Assay
Cell viability was assessed to ensure the surface modifications were not inducing
adverse cytotoxic effects. The LIVE/DEAD® Viability/Cytotoxicity Kit (Life
Technologies) was utilized with slight modifications. Here, glass and MnTBAP treated coverslips were tested. Positive (all live) and negative (all dead) controls were performed utilizing clean glass coverslips. The negative control was established by incubating cells with 70% methanol for 10 minutes. Cells were seeded for 48 hours prior to removal of media and incubated in 8 µM ethidium homodimer-1 (EthD-1) and 0.1 µM calcein-AM in complete PBS for 15 min. The samples were then washed once with 1xPBS. For quantification, cell-seeded samples were transferred to a black well plate for a total fluorescence read on the plate reader of EthD-1 activity (excitation 528 nm, emission 617 nm). Percentage of dead cells was calculated by normalizing total fluorescent intensity to the negative controls (N=3, n=9).
Representative images from live/dead and DHE assays were acquired fluorescently using an inverted AxioObserver Z1 (Zeiss) equipped with an AxioCam MRm prior to detection on the plate reader, to confirm measured values. For each assay, exposure times were held constant between conditions.
174 6.3.8 Statistical Analysis
For all conducted experiments, a minimum of three trials (N=3, n=9) were used for
statistical evaluation (N≥3). For the Live/Dead assays a one-way analysis of variance
(ANOVA) with a Tukey’s post-hoc test was used to find significance. For all other assays,
a standard student t-test was utilized to determine significance. In all cases statistical
significance was defined as p<0.05.
6.4 Results and Discussion
6.4.1 Design of Synthetic Anti-oxidative Modified Substrates
Building on our previous work with the antioxidant resveratrol, we sought to
engineer a surface modification that would provide an initial release of anti-oxidative
activity to mimic results found in our previous work. In addition, our goal was to also
ensure that our engineered substrates would improve the longevity of local anti-oxidative
effects. Therefore, we designed an anti-oxidative surface treatment that provided an initial
release of antioxidant, followed by prolonged anti-oxidative activity. To achieve our desired surface modifications, we focused on the characterization and in vitro evaluation
of a composite coating of the SOD mimetic Mn(III)tetrakis(4-benzoic acid) porphyrin
(MnTBAP) (Figure 36).
A key advantage of MnTBAP and other metalloporphyrins are their ability to repeatedly scavenge reactive oxygen species.[333, 334] The two half reactions enabling
MnTBAP to repeatedly scavenge superoxide anion are shown in Equation 9.
2+ - + 3+ (9i) Mn + O2 + 2H Mn + H2O2 3+ - 2+ (9ii) Mn + O2 Mn + O2
175 A number of studies have shown MnTBAP to be stable in aqueous media for
extended periods.[35, 335, 336] Due to the small size and chemical structure, MnTBAP is also capable of crossing biological membranes, enabling MnTBAP to act within the cell, as well as extracellularly.[31, 337, 338]
The ability of MnTBAP to stably and repeatedly scavenge reactive oxygen species
(ROS) is of critical importance for the application of intracortical microelectrodes.
Specifically, multiple groups have suggested a key role of ROS in mediating corrosion of electrical contacts ultimately resulting in microelectrode failure.[3, 61, 62] In fact, a recent study by Barrese et al. suggested that the majority of chronic Utah electrode array failures could be attributed to corrosion and/or degradation of electrode insulation.[3] Further, it has also been suggested that the high biocompatibility of platinum microelectrodes in the brain, versus other metals such as tungsten, may be due to platinum’s ability to directly act as an antioxidant.[62] In addition, previously we have shown a correlation between neuronal health and intracellular accumulation of ROS around implanted microelectrodes.[15] Therefore, given the versatility of MnTBAP, and the impact of ROS on intracortical microelectrode stability, our approach promises to become an excellent candidate for in vivo applications of intracortical microelectrodes.
Composite coatings of the MnTBAP SOD mimetic were created by first functionalizing plasma treated glass (SiO2) coverslips with (3-aminopropyl)-
triethoxysilane. In order to confirm amine functionalization of the plasma treated silicon
substrates, both the hydrophobicity and atomic composition were characterized with
goniometry (Table 5) and XPS (Table 6), respectively. Consistent with literature
values[339], surface amination caused an increase in contact angle from < 5° ± 0 to 55.2°
176 ± 2.8°. Additionally, XPS analysis confirmed grafting to the anime terminated silane.
Specifically, XPS analysis demonstrated a reduction in the composition of both oxygen
(68.8% to 39.0%, Table 6) and silicon (27.3% to 16.9%), and an increase in both carbon
(3.8% to 38.1%) and nitrogen (<0.1% to 6.0%), consistent with published values.[340] The amine modified substrate was designed for both covalent immobilization of the MnTBAP through succinimidyl ester coupling[329, 341], and to facilitate more natural conformations of non-specifically adsorbed protein in future in vivo studies (Figure
36).[342]
Table 5. Contact Angle Measurements on Modified Surfaces. Surface Contact Angle a Plasma treated SiO2 < 5° ± 0 Amine functionalized 55.2° ± 2.8° MnTBAP SOD mimetic composites 24.3° ± 1.4° aSurface wetting was < 5°, which was below the detectable region
Table 6. XPS Analysis on Modified Surfaces
Sulfo-NHS ester functionalized MnTBAP SOD mimetics were subsequently exposed to the anime functionalized substrates to create a composite coating. The composite consisted of both covalent immobilized and non-specifically adsorbed
177 MnTBAP (Figure 36). Both contact angle measurements and XPS analysis confirmed the
surface coating of MnTBAP SOD mimetics. Goniometry indicated a reduction in contact
angle from 55.2° ± 2.8° to 24.3° ± 1.4° (Table 5). Additionally, XPS analysis confirmed a reduction in the composition of nitrogen (6.0% to 3.1%, Table 6), and an increase in both carbon (38.1% to 47.1%) and Mn (<0.1% to 0.5%). Passively adsorbed MnTBAP could be mechanically removed with cotton swabs, or additional washing with EtOH. Removal of the adsorbed MnTBAP and subsequent analysis with XPS indicated no significant difference in the atomic composition of the substrates, compared to MnTBAP SOD mimetic composites
In order to determine the surface density of our composite MnTBAP substrates, we utilized a quartz crystal microbalance (QCMB). QCMB allowed for the real-time evaluation of the evolution of the MnTBAP SOD composite. Using a Voigt model, the change in frequency after the finalized chemical reaction indicated a mass increase of 21.9
µg/cm2, for the MnTBAP composite surfaces. However, the chemical compatibility limits
of our QCMB did not allow for additional steps that would have given us a definitive
evaluation of surfaces with only covalently attached MnTBAP.
6.4.2 Anti-oxidative Activity of MnTBAP-modified Surfaces
In order to determine the anti-oxidative activity of the SOD mimetic modified
surfaces, we utilized the Riboflavin/NitroBlue Tetrazolium (RF/NBT) assay. Under
- illumination, RF generates a superoxide O2 radical that will reduce the NBT. In the
presence of an antioxidant, the reduction of NBT is inhibited. In this assay, SOD mimetics
(or resveratrol in the case of control experiments) compete with NBT for the superoxide
178 - O2 radical. Therefore, Figure 37 presents the anti-oxidative activity of each surface treatment as the % Inhibition of NBT Reduction by RF, normalized to the activity of plasma treated silicon substrates. As a control, the amount of soluble MnTBAP on the composite surfaces was run prior to surface testing. Specifically, given the QCMB determined surface density and area of coverslip, the anti-oxidative activity of 24.75 µg of soluble MnTBAP was utilized as a reference and noted as 100%. Therefore, 100% inhibition of NBT reduction was defined as the reduction demonstrated by soluble MnTBAP equivalent to the amount immobilized on the composite surface (See Equation 8).
Figure 37. Activity of MnTBAP composite surfaces over time. The ability of anti-oxidative surfaces to prevent NBT reduction was measured up to two days after synthesis and incubation in PBS. Hybrid (composite) surfaces (red square) demonstrated an initial decline in NBT reduction activity, followed by low- level sustained activity. In contrast, surfaces with only conjugated MnTBAP maintained activity over time (blue circle). Data was normalized to soluble MnTBAP (0.24 M). Grey region represents the working therapeutic range for resveratrol. Data is shown as an average ± s.e.m. N=3, n=9.
179 As an additional control, we compared the activity of our modified surfaces to the
natural antioxidant resveratrol. We have previously demonstrated that administration of
resveratrol the day before and the day of microelectrode implantation provides
neuroprotection at the implant site for up to four weeks.[15] Therefore to ensure that our
surfaces had similar anti-oxidative properties to our previous study, here we report on: (1)
the concentration of accumulated resveratrol around the cortical tissue adjacent the
microelectrode up to 48 hours after electrode implantation and (2) the relative NBT activity
of the measured resveratrol concentration in comparison to soluble MnTBAP and our modified surfaces.
Resveratrol is a conjugated bi-phenol, and naturally fluoresces at 385 nm.
Therefore, following our two-dose administration of resveratrol, cortical tissue adjacent to the microelectrode was retrieved from euthanized animals. Resveratrol was extracted from tissue as described above, and the concentration of resveratrol within the tissue was determined spectophotometrically against a standard curve. We found that our dosing scheme resulted in a concentration range of resveratrol of between ~0.5 to 25 µM for up to
48 hours after injection/implantation (Figure 38). Interestingly, several groups, including ours, have also shown high efficacy and low toxicity of resveratrol from 5 to 25 µM in vitro. [121, 343]
Most importantly, we found that within this concentration range, resveratrol
provided significant levels of anti-oxidative activity. Specifically, a concentration of 25
µM of resveratrol had the ability to inhibit approximately 65 ± 9.7% of NBT reduction capabilities of soluble MnTBAP.
180
Figure 38. Bio-distribution of resveratrol around implanted microelectrodes up to 48 hours after administration. A therapeutic range (5 to 25 µM) was found around implanted devices for up to 18 hours after implantation. Concentration is reported as an average ± s.e.m. (n≥3).
After immobilization, MnTBAP composite surfaces maintained nearly 60% of the
activity of the soluble MnTBAP (Figure 37). Further, at early time points, MnTBAP
composite surfaces were capable of similar anti-oxidative activity to the levels facilitated by the upper limits of resveratrol detected at the microelectrode surface 1-18 hours after administration (Figure 38). Specifically, we found that the NBT inhibition ability of our composite MnTBAP surfaces was statistically similar to the peak tested concentration of resveratrol (25 µM) at day 0. Despite a decline in activity after day 0, at both days 1 and 2, the anti-oxidative activity of MnTBAP composite surfaces was statistically similar to the lower end of functional resveratrol concentrations (Figure 37, 5 µM). Although our
181 MnTBAP composite surfaces displayed a sustained anti-oxidative property, further testing would be required to determine the effect of prolonged activity on neuroprotection.
In order to confirm the source of the decline in anti-oxidative activity of composite surfaces, immobilized only surfaces were created. At time 0, approximately 45% of the anti-oxidative activity demonstrated from the composite surfaces was generated from the surface immobilized MnTBAP. Upon testing at subsequent time points, we found no statistical differences between the immobilized only and composite surfaces (FIGURE 37).
In comparison to the hybrid composite surfaces, it would appear that the adsorbed layer of
MnTBAP disassociates from the surface within 24 hours from exposure to physiologically
relevant systems. Most importantly, the activity of the immobilized MnTBAP remained
consistent during the extent of our initial testing, and provided statistically similar anti-
oxidative activity to therapeutically relevant levels of resveratrol (5 µM[121],Figure 37)
Further, our results also demonstrate, for the first time to our knowledge, that covalent immobilization of MnTBAP is capable of physiologically relevant anti-oxidative activity
(Figure 37).
It is important to note that without the adsorbed layer of MnTBAP, our composite surfaces would not be fully capable of similar high anti-oxidative activity provided by resveratrol early after administration. Specifically, for all investigated time points, surfaces with only immobilized MnTBAP had significantly less NBT reduction activity than 25 µM of resveratrol. However, hybrid systems had similar activity to resveratrol at Day 0 post synthesis. Further, maintained activity was noted in hybrid surfaces, and by day 1 and 2, no significant differences from 5 µM of resveratrol were noted. Thus, the hybrid composite
182 MnTBAP surface provides distinct advantages to both soluble resveratrol delivery and
surface immobilization only of MnTBAP, without the deliverable adsorbed layer.
6.4.3 Effect of MnTBAP-modified Surfaces on Reactive Oxygen Species and Nitric Oxide
Accumulation and Release
Following validation of active MnTBAP composite surfaces, we utilized in vitro
evaluation to determine the effects of our engineered systems on activated microglia cells.
Specifically, for our analysis we utilized an immortalized murine microglia cell line that
we have found is constitutively in the M1 pro-inflammatory phenotype as shown by CD86+
and CCR7+ labelling (data not shown). Other groups have suggested the key role of
microglia and macrophages in the neuroinflammatory response that exists around
intracortical microelectrodes.[101, 109, 291] It has been suggested that the formation of
neurotoxic regions around the implant is likely the result of an accumulation of M1
phenotypic microglia/macrophage population.[344, 345] Further, given that M1 microglia
and macrophages primarily release high amounts of ROS[346], it is also probable that microglia and macrophage cells are involved in the corrosion of the actual microelectrode.
Therefore, we investigated the ability of the anti-oxidative MnTBAP composite surfaces to prevent ROS accumulation and release from activated M1 phenotypic microglia cells seeded on our modified materials.
183
Figure 39. Accumulation of intracellular superoxide anion (dihydroethidium; DHE) from activated microglia cells (BV2s) 48 hours after seeding onto modified surfaces. Cells seeded on MnTBAP modified surfaces demonstrated a significant reduction (*p<0.05) in DHE labelling. N=3, n=9.
First we investigated the accumulation of ROS in microglia cells. Specifically, to label intracellular accumulation of ROS, we utilized dihydroethidium (DHE), a dye that selectively immobilizes inside the cell upon reaction with superoxide anion. Thus, increased DHE positive labelling would infer more intracellular superoxide anion within the cells. Here we found that microglia cells seeded on MnTBAP composite surfaces demonstrated significant decreases in intracellular superoxide anion compared to unmodified glass substrates (Figure 39).
Changes in intracellular ROS accumulation can directly affect the anti-oxidative properties within an inflammatory cell.[347, 348] Further, increases in intracellular ROS can also result in the accelerated release of ROS into the local extracellular environment.[349, 350] In the case of intracortical microelectrodes, increases in extracellular ROS in the local environment could result in both (1) neuronal cell death[15]
184 and/or (2) corrosion of the insulating and conducting layers on the electrode.[3, 61, 62]
Therefore, it is also critical that we evaluated the effects of our MnTBAP hybrid surfaces
on extracellular ROS released from activated M1 microglia cells.
The effect of MnTBAP modified surfaces on the accumulation of cell-released ROS
was investigated using a modified NBT assay. Specifically, neat NBT was added to tissue
culture medium and the amount of NBT dye reduction was quantified. The amount of ROS
in solution is therefore directly proportional to the amount of NBT dye reduction; a surface
modification that demonstrated more NBT reduction would indicate higher levels of ROS accumulation. Notably we found that our MnTBAP composite surfaces were capable of significant decreases in soluble ROS accumulation from activated M1 microglia cells
(Figure 40).
Figure 40. Release of reactive oxygen species (ROS) from activated microglia cells (BV2s) 48 hours after seeding onto surfaces. Cells seeded onto MnTBAP modified surfaces released a significantly smaller amount of ROS in comparison to glass controls (*p<0.04). N=3, n=9.
185 We have previously shown that reduction of ROS around implanted microelectrodes directly correlates with a more stable blood-brain barrier and more viable neurons.[15] Thus, we wanted to ensure that the mimetic employed here would be capable of similar protection. Based on this input criteria, MnTBAP, was chosen due to its ability to selectively alter the release and accumulation of reactive oxygen species (ROS) in contrast to nitric oxide. The selectivity of MnTBAP was critical given the role of nitric oxide in cortical tissue. Specifically, in homeostasis, nitric oxide, being freely able to pass the blood-brain barrier, has been identified as a key messenger molecule in brain tissue.[351] The synthesis and release of nitric oxide has been correlated with neuronal activity, cerebral blood flow and glial cell homeostatic function.[352, 353] In contrast, during inflammatory states, nitric oxide can be pro- and anti-inflammatory.[354]
Particularly, only in the presence of ROS can nitric oxide be reacted to form reactive peroxynitrite (ONOO-)[351, 355], a molecule known to be in high amounts during oxidative stress and inflammatory states. In our model, we aimed to specifically reduce
ROS to prevent direct neurotoxicity and to prevent the accumulation of nitrite anion.
Therefore, to confirm the specificity of MnTBAP in our model, we tested the effect of our MnTBAP composites on nitric oxide (NO) release and accumulation from cultured activated microglia. After two days of incubation on our surfaces, we found that MnTBAP modifications did not result in any significant changes in NO release from microglia cells in comparison to glass controls, as expected (data not shown). In addition, it has been previously demonstrated that only concentrations above 25 µM of resveratrol have been shown to substantially alter nitric oxide concentrations.[356]
186 6.4.4 Effect of MnTBAP modified surfaces on cellular viability
Taken collectively, our results suggested that our MnTBAP SOD mimetic composite surfaces were capable of reducing both intracellular and extracellular reactive oxygen species. However, to ensure that the molecular changes that were noted in activated microglia cells were not the result of changes in cell viability, we conducted a Live/Dead cell cytotoxicity assay on seeded microglia cells. Notably, we found that MnTBAP modified surfaces had statistically similar levels of live and dead cells in comparison to glass controls (Figure 41). Therefore, our results suggested that our reported changes in cellular accumulation and release of ROS from MnTBAP modified surfaces were likely the result of our engineered surfaces, and not the result of changes in cell viability after seeding.
187
Figure 41. Cytotoxicity of modified surfaces on activated microglia cells (BV2s) after 48 hours. No significant levels of cellular cytotoxicity were noted for investigated modifications. Green denotes live cells and red denotes dead cells. N=3, n=9.
In addition, the results presented in Figure 41 are desirable given the critical role
microglia play in the inflammatory cascade following device implantation.[103]
Specifically, microglia cells have been shown to directly adhere to the surface of the
electrode.[70] After adhesion, microglia cells begin to release soluble factors that can (1)
further activate surrounding cells, (2) affect local neuronal homeostasis and (3) affect the
functionality of the actual electrode.[345] Elimination of adherent microglia may result in
disruption of the inflammatory response and impede the wound healing process.
188
Figure 42. Proposed mechanism of action for MnTBAP composite surfaces. Left: For an unmodified electrode, microglia cells adhere to the surface of the device and release high amounts of reactive oxygen species (ROS). Right: In the presence of our MnTBAP composite surfaces, three possible mechanisms of action could result in decreases in localized ROS released from activated microglia. (A) Immobilized MnTBAP could directly convert extracellular ROS to water. (B) Microglia can internalize released MnTBAP, which could directly decrease the amount of intracellular ROS accumulation. (C) Non- internalized released MnTBAP freely diffuses into the tissue space and can neutralize ROS in the extracellular space.
Given that we observe reductions in both intra- and extracellular ROS accumulation as a result of immobilized and released MnTBAP, we hypothesize that our hybrid composite surfaces have the ability to interact with microglia cells adherent to the microelectrode, as well as locally activated microglia cells that are responding to the tissue damage and the change in local inflammatory state Figure 42. Specifically, adsorbed
MnTBAP, once released, could directly pass through the microglia cell membrane and result in changes in intracellular ROS production and accumulation, similar to those we noted in our study (Figure 39). Released MnTBAP could also reduce extracellular ROS accumulation. In addition, covalently immobilized MnTBAP should facilitate the
189 repetitive conversion of extracellular ROS into oxygen in the localized environment (see
Equation 9 and Figure 40).
We have previously shown that a stimulant-initiated threshold-dependent response in microglia activation can result in differences in neuronal survival at the microelectrode- tissue interface.[258] Therefore, utilizing redundant mechanisms to decrease of ROS accumulation around the implanted microelectrode increases the probability that our designed surfaces could be a viable option for the prevention of both corrosive damage to the electrodes and inflammatory-mediated neurodegeneration; both leading to improvements to the chronic stability of intracortical microelectrodes following implantation.
6.5 Conclusions
Neuroinflammation and oxidative stress events have been shown to result in the loss of neuronal bodies around implanted microelectrodes.[70, 101] Therefore, in the present study, we sought to develop anti-oxidative surface modifications to intracortical microelectrodes. Here, we investigated the feasibility of MnTBAP SOD mimetic conjugated surfaces that were capable of an initial release to combat initial neuronal loss and sustained anti-oxidative activity to regulate chronic neuroinflammation. Notably, we found that our engineered systems were capable of anti-oxidative activity up to two days after exposure to physiologically relevant conditions. Further, MnTBAP modified surfaces were able to reduce intracellular and extracellular ROS release from activated microglia cells in vitro.
190 It is important to note that the results presented here are of particular interest, since as to date, no group has successfully immobilized a commercially available SOD mimetic to a material surface. Therefore, we anticipate that the modification scheme that has been developed here has the potential to be deployed in several biomedical applications, such as cortical/peripheral neural electrodes, stents, hip implants and cardiac leads. Further, given the success of our modified surfaces in vitro, future studies will investigate the use of
MnTBAP modified surfaces in reducing neuronal cell death and corrosion of intracortical microelectrodes in an in vivo model.
6.6 Acknowledgements
This work was supported by the Department of Biomedical Engineering and Case
School of Engineering at Case Western Reserve University through lab start-up funds, the
GAANN Training Grant for Neural Engineering (K. Potter), and the NIH Neural
Engineering and Rehabilitation Training Grant (J. Nguyen), (5T32EB004314-14).
Additional funding on this research was supported in part by the Department of Veterans
Affairs Merit Review (B7122R), Presidential Early Career Award for Scientist and
Engineers (PECASE). The authors have no conflicts of interest related to this work to disclose. None of the funding sources aided in the collection, analysis and interpretation of data, in the writing of the report, or in the decision to submit the paper for publication.
The authors also acknowledge the support of Archana Jaiswal from Biolin
Scientific for help in processing quartz crystal microbalance data sets. In addition, the author’s also acknowledge the support of S. Selkirk and K. Buchanan.
191 Chapter 7
Conclusions and Future Directions
The work in this dissertation examined material and therapeutic strategies to reduce
the neuroinflammatory response and improve intracortical microelectrode integration.
Specifically, the effects of compliant microelectrode substrates and antioxidative therapies
were discussed. While each strategy alone demonstrated specific beneficial outcomes for attenuating the neuroinflammatory response, the work here supports the idea that combining strategies may better enhance modulation of the complex neuroinflammatory response.
Instability and quality of neural recordings with intracortical microelectrodes are a common concern, resulting in limited usage of intracortical microelectrodes. The biological response is a major contributing factor in chronic microelectrode stability.
Numerous theories and strategies for ways to reduce the tissue response exist. Reports from Saxena et al. and Nolta et al. have asserted that BBB integrity can profoundly influence the tissue response and microelectrode recording quality.[11, 96]
A long-standing hypothesis proposes that the mechanical mismatch between microelectrode material stiffness and brain modulus contributes to chronic BBB damage and neuroinflammation. In silico studies support the idea that utilizing low modulus, compliant implants reduce forces on surrounding brain tissue.[4, 5] However, attempts to utilize soft polymer substrates suffer from implantation difficulty due to material buckling when trying to penetrate through the pial surface. In response, mechanically-adaptive
192 nanocomposite (NC) materials have been developed that are stiff enough to easily implant
and soften when exposed to the physiological environment.[22, 207]
Mechanical characterization of NC materials indicated a relatively quick switch
(~15 minutes) from 5 GPa to 12 MPa upon implantation into the brain.[23, 211] Initial investigation of the inflammatory response found that the NC was able to improve neuronal proximity at the implant-tissue interface at four weeks, but not at eight weeks.[20]
However, work by Potter et al. and McConnell et al. indicated the presence of a late onset neurodegenerative state at sixteen weeks post implantation.[24, 102] Additionally, in the initial study by Harris et al. microwire controls were used as a comparison. Studies have since established that cylindrical microwire implants generally result in a reduced inflammatory response compared to planar electrodes.[8, 64]
Therefore, in Chapter 3, compliant, NC implants were compared to chemically- matched, polymer-coated Michigan style silicon implants at various time points that exhibit significant neuronal death.[19] Tissue analysis at three days post-implantation showed no difference between stiff and compliant implants. These results support previous data that suggest wound healing events dominate at early time points.[109] To reduce the acute response, methods to lessen insertion damage and, in particular, vascular damage are being explored.[27, 84, 111, 112, 114] A more compact astrocytic scar and reduced microglial activation was seen surrounding compliant implants compared to stiff implants at 2 and 8 weeks post-implantation. However, at 16 weeks post-implantation, a complete recovery of neuronal populations around compliant implants was seen at 0 to 50 µm from the electrode interface. Significant reduction in microglial activation was also seen from 0 to
50 µm. Further, with the exception of three days post-implantation, compliant implants
193 significantly reduced BBB leakiness compared to stiff implants. These results verified that compliant implants can reduce BBB leakiness and thus, may reduce self-perpetuating inflammatory activation resulting in improved neuronal survival. Additionally, it is possible that utilizing the compliant NC implant may also be speeding up the wound healing process.
A common question arises when investigating compliant implants: “What is the actual, measured strain elicited on the surrounding tissue?” While in silico and in vitro studies suggest effects of material compliance on tissue strain and glial cell activation[4, 5,
25, 26], there have been limited efforts to quantify the tissue strain surrounding implanted microelectrodes. In Chapter 4, in collaboration with the Muthuswamy group, assessment of tissue strain surrounding compliant versus stiff implants was explored.[30] Acute force measurements were made utilizing a custom jig from a previous study by the Muthuswamy group.[6] Results showed that softer NC materials affected the viscoelastic properties of the brain, reducing the dynamic stress relaxation rate of the brain tissue and lowering micromotion-induced stress amplitudes (~4-5 fold) compared to bare silicon implants. The results suggest that reduction of micromotion-induced stress on surrounding tissue by compliant implants may lessen BBB disruption and strain-mediated inflammation.
Interestingly, PVAc-coated microelectrodes exhibited a similar reduction in micromotion- induced stress amplitudes, possibly due to surface adhesion properties, as modeling studies have indicated that adhesion plays a role in the extent of tissue strain.[4, 5]
The neuroinflammatory response is complex and dynamic; therefore, a single targeted strategy may not be enough to provide continuous neuroprotection. While NC implants show a remarkable improvement in the chronic tissue response, issues with
194 instability in neural recordings are reported throughout implant lifetime. Researchers have reported high variability in recording quality from day to day. Therefore, it would be beneficial to create a microelectrode recording platform that has improved short-term stability as well as longevity. Our lab has lead the way with antioxidant therapies to reduce microelectrode-mediated neuroinflammatory events. Specifically, delivery of resveratrol and curcumin have been shown to provide neuroprotection and reduce the acute inflammatory response.[14, 15] Therefore, coupling antioxidant release and compliant materials may improve microelectrode stability throughout device implantation time.
In Chapter 5, a strategy to combine the effects of compliant materials and antioxidative therapy was investigated. In this study, curcumin-releasing NC and resveratrol-releasing NC implants had no discernible effect on neuroinflammation in comparison to NC control implants at three days or sixteen weeks post-implantation. At three days, the insertion damage may be overwhelming any potential benefits of the antioxidants. The acute results further substantiate the lack of improvement seen with compliant implants at three days in Chapter 3, suggesting that the wound healing events in response to insertion damage may be the overriding issue immediately following implantation. At sixteen weeks post-implantation, the antioxidant is presumed to be completely released. Thus, it was expected to see no differences in neuroinflammation between NC and NC Res.
Curcumin did not appear to provide an additive effect to NC implants for any of the investigated time points. Curcumin has very low bioavailability, which may have reduced the diffusion-mediated delivery. A higher dosage or improved delivery system could improve curcumin solubility and release. However, at two weeks post-implantation,
195 beneficial effects of resveratrol-releasing NC implants were apparent with decreased
microglial activation and neurodegeneration at 0 to 50 µm from the interface.
Neuroprotection by resveratrol may be mediated by a microglial-driven pathway. A major anti-inflammatory mechanism of resveratrol is targeting of microglia, causing reduction of pro-inflammatory molecules production, ROS production, and activation of signaling pathways.[357]
Short-term resveratrol delivery was able to provide an additive benefit to compliant implants and stabilize neuronal populations at early time points. An alternative antioxidant strategy may be able to provide a more sustained effect. Superoxide dismutase (SOD) mimetics are able to reversibly and continually scavenge oxygen free radicals. In Chapter
6, a method was developed to surface conjugate SOD mimetics on the microelectrode surface. By applying a direct surface modification, the antioxidant had the potential to scavenge ROS and protect the electrode surface for up to 48 hours. Characterization of surfaces and in vitro results indicated successful immobilization of SOD mimetics on
surfaces. Further, reduction in intracellular and extracellular ROS production was seen with SOD-coated surfaces. Preliminary in vivo studies with SOD-coated silicon electrodes
indicated complete neuron recovery at the tissue interface at two weeks (Unpublished,
Figure 43). The significant increase in neuron populations around SOD-coated implants
suggests that ROS-mediated neurodegeneration is a major factor in neuronal health at the
interface. Currently, experiments are underway with implanted SOD-coated functional electrodes to investigate the effects on neural recording quality.
196
Figure 43. In vivo implantation of SOD-coated Michigan-style silicon implants. Neuronal nuclei staining indicates significant improvement neuron populations around SOD-coated Si implants compared to uncoated Si. **p<0.001, *p<0.05
Additionally, SOD-coatings have been applied to NC implants as an alternative to
drug delivery. Due to the fluid uptake of NC and wet chemistry involved in SOD surface
conjugation, SOD mimetic activity with NC implants can be attributed to both SOD
conjugated to the NC and SOD released from the NC. In vitro assessment of antioxidative
activity of SOD-coated NC surfaces via NBT assay are shown in Figure 44A. High
absorbance indicates increased ROS levels. Absorbance values are normalized to glass
control surfaces. SOD-coated NC samples displayed a significant reduction of ROS
accumulation, similar to SOD-coated glass controls. Cell studies also showed that SOD-
197 releasing NC reduced intracellular ROS accumulation in BV-2 microglia cells (as seen with
DHE staining), and the SOD coating did not affect neuron viability (Figure 44B,C). With the significant improvement of neuron density around SOD-coated stiff silicon implants, future in vivo studies should assess if the addition of SOD mimetic on NC implants will show enhanced neuroprotection.
Figure 44. In vitro characterization and validation of SOD-coated NC surfaces. (A) ROS accumulation was measured via NBT assay. Higher absorbance values represent increased ROS levels, while lower values represent reduction in ROS. SOD conjugation to NC significantly reduces ROS similar to SOD-coated glass samples. *p<0.002 (B) Measurement of intracellular ROS accumulation with DHE staining indicated significant decrease in microglia incubated with NC SOD surfaces for 48 hours. *p<0.02 (C) SOD-coatings of NC films did not affect cell viability. Green indicates live cells and red indicates dead cells.
198 The work in this dissertation demonstrated that material compliance and
antioxidant delivery can affect the neuroinflammatory response to intracortical implants.
However, moving forward, a better understanding of the mechanisms and pathways that
are affected by material compliance and antioxidant therapy are needed. In terms of material compliance, the threshold modulus that is required to reduce inflammation is unknown. While the NC materials utilized in Chapter 3 soften to become significantly more compliant than traditional metal implants, they are still an order of magnitude stiffer than brain tissue. It should be investigated if softer materials provide further benefits, or even if slightly stiffer materials provide similar benefits.
While studies have shown that compliant materials can influence the neuroinflammatory response, the direct impact of compliant materials on functional recordings is still debated. Reports have shown variable changes in acute neural recordings from compliant microelectrodes.[7, 179, 186, 191, 213] There are many challenges to fabrication of functional microelectrodes on soft substrates that have hindered long-term studies. With the discovery of a late onset neurodegenerative state, the chronic effects of soft substrates on recording quality will be more telling for potential long-term use.
Studies have already suggested that the modulus of the substrate material can affect glial cell response. Specific up-regulation of IL-36Ra expression has been attributed to glial strain[26]. Additionally, growth of glial cells on substrates of varying stiffness also exhibit different gene expression profiles.[25] While direct damage to the tissue is a clear result of microelectrode micromotion, it would be of interest to investigate if and how mechanical stresses can directly affect specific cell morphology, proliferation, activation, receptor expression, or cytokine production. For example, tissue strain may activate
199 mechanoreceptors to influence the surrounding tissue. Mechanoreceptors sensitive to mechanical pressure or distortion have been described in skin and hair follicles, and baroreceptors are a type of sensory mechanoreceptor that is excited by stretching of blood vessels. Focal adhesions kinase (FAK), a mechanosensor located at focal adhesions, has also been shown to play a role in cell migration. FAK-null cells were found to have impaired responses to substrate flexibility and migrate away from soft substrates.[358]
Additionally, the Wang group demonstrated that myosin II is involved in fibroblasts
rigidity sensing, wherein fibroblasts preferentially migrate to rigid substrates and retract
from soft substrates.[359] Application of molecular biology techniques, such as western
blotting, RT-PCR, gene or cytokine arrays, would be able to identify specific factors.
Elucidating what factors are involved in the response to compliant implants will allow for
specific targeting strategies.
Another interesting option to consider is adjusting the architecture of currently used
polymer implants. Due to ease of fabrication, polymer implants are generally formed as
planar electrodes with recording electrodes along the shank. However, cylindrical shaped
implants produce significantly less inflammation than planar implants. It would be
interesting to see if cylindrical polymer implants have a similar effect, although it may
prove challenging to fabricate and could restrict the size and location of electrode sites.
Another question to investigate is: what has a bigger effect—probe size or modulus?
Effectively a reduction in probe size improves the flexibility within the tissue but issues
with fragility and breakage become an issue. However, a reduced implant size will
significantly reduce initial vascular tissue damage, which has been extensively associated
to be a major factor during implantation and been shown to affect neural recording quality.
200 While the acute measurements of tissue stresses around implants reported a benefit of compliant materials, of particular interest is the long-term effect. Histological studies demonstrated a more compact glial scar profile surrounding compliant implants. Future studies on chronic, long-term implants would be able to provide detailed quantitation information on the mechanical properties of glial scar maturation around compliant and stiff implants. A previous report by Muthuswamy indicated dynamic changes in glial scar elastic modulus surrounding microwire implants for up to eight weeks.[6] It is likely that the glial scar surrounding a compliant implant has a reduced elastic modulus compared to stiff implants. Moreover, the elastic modulus of the glial scar is unknown and would be an informative tissue property for modeling parameters. It is important to note that the force measurement system used here had limitations. The measured force was a compressive strain from implantation in the axial direction, hence the predominant forces were from shear stresses and adhesion properties on the probe. Most in silico studies have indicated a larger effect of tangential strain from tethering forces. A new technique to monitor in vivo implants in real-time would be a great tool. In vivo monitoring with optical coherence tomography (OCT) showed an increase in optical attenuation surrounding the fiber optic probe, suggesting formation of a dense glial scar.[360] Quantification of the scar mechanical properties was not reported, but OCT would be informative to directly monitor and compare changes in glial scar formation surrounding compliant and stiff implants.
Incorporation of strain gauges on the electrode surface could also inform on the strain that is being elicited on the microelectrode in real time.
The importance of BBB integrity on recording quality has been increasingly investigated. However, BBB integrity, microglial activation, and neurodegeneration are
201 intimately related. A detailed understanding of the interplay between different factors and
the evolution of the inflammatory response to microelectrodes would provide information
on how to better modulate the system. Intravital two-photon microscopy, involving live
imaging of fluorescently labeled cells in transgenic mice through a cranial window, could
allow for direct visualization and measurement of microelectrode movement and
displacement over time.[361] Real-time in vivo imaging would provide countless benefits
for visualization of neuron function and neuroinflammatory events. Real-time monitoring could elucidate the progression of glial cell migration, activation, and scar formation surrounding an intracortical implant. Labeling of firing neurons could inform on which neurons are being directly recorded from. Fluorescent calcium sensors are widely used to image neural activity.[362] Calcium indicator proteins including a fluorophore sensor can be delivered genetically to detect rapid changes in intracellular free calcium. Optogenetics allows for light-mediated control of neuron firing. A system coupling in vivo visualization and optogenetic stimulation may enable researchers to observe if implanted microelectrodes can detect neural signals from various specifically-targeted locations.
Additionally, with in vivo imaging, it may be possible to monitor if neurons are migrating
to and from the interface, or what neuronal remodeling is occurring around the implant.
Future studies should also be done to better understand resveratrol delivery and the
mechanism of action. Experiments to enhance resveratrol bioavailability and delivery for
microelectrode applications are currently underway. Initial studies investigating long-term resveratrol administration showed reduction in degenerating neurons at two and sixteen weeks with daily resveratrol dosage.[363] However, thread-like adhesions were found between the liver and diaphragm in chronically dosed resveratrol animals. Studies must be
202 done to determine how long resveratrol should be administered and reduce adverse effects
from long-term administration. Resveratrol can affect many inflammatory events,
including activation of NF-kB, mitogen-activated protein kinase (MAPK), and activator protein 1 (AP-1) inflammatory pathways. Therefore, isolating and understanding specific effects from various pathways may allow researchers to design a more targeted approach to modulate the tissue response. For instance, resveratrol induces expression of heme
oxygenase-1 (HO-1), a stress-inducible protein with potential anti-inflammatory effect via
the MAPK pathway.[364, 365] Molecular biology techniques, such as Western blots,
laser-capture microdissection, RT-PCR, cytokine arrays or DNA microarrays, should be
used to investigate changes in gene and protein expression. Conditional gene knock-out
animals may could also be used to knock out cell types at specific times and observe what
changes occur in the system. Understanding individual effects can inform on what aspects
of the tissue response are important for microelectrode function.
Further, alternative anti-inflammatory strategies should also be explored. Many different anti-inflammatory therapeutics, such as dexamethasone and minocycline, have demonstrated positive results in modulating the inflammatory response. There are many options and understanding what to target will inform on what therapeutics may be beneficial. Quercetin, a flavonoid found in many fruits, vegetables, and grains, has
demonstrated antihistamine and anti-inflammatory properties.[366] Administration of quercetin has shown neuroprotective effects, particularly in cerebrovascular insults.[367,
368] Non-steroidal anti-inflammatory drugs (NSAIDs), such as aspirin and indomethacin,
also demonstrate neuroprotective effects in neurodegenerative diseases such as Parkinson’s
and Alzheimer’s disease.[369, 370]
203 Reduction of microglial activation has been a widely discussed target to attenuate
neuroinflammation, but microglia play a positive role in cleaning up cellular debris and
enabling tissue remodeling. Instead of completely reducing microglial activation, an
interesting approach would be to modulate activation. It has been shown that different microglial phenotypes can be pro- or anti-inflammatory. Studies of other inflammatory diseases have begun to look at the potential of pushing microglia/macrophages to their anti- inflammatory M2 phenotype. Specifically, M2 polarization with IL-10/TGF-β has shown protection again renal inflammation and diabetes.[371-374] Polarization to M2 phenotype may enhance and speed up the wound healing process to more quickly stabilize the damaged tissue after microelectrode implantation.
In conclusion, neural microelectrode technology has immense potential to enhance scientific research knowledge and biomedical device function. While strategies to improve microelectrode integration are increasingly reported, further understanding of microelectrode failures modes and the influence of the inflammatory process is essential.
The work in this dissertation emphasized the impact of material compliance on the neuroinflammatory response. Additionally, given the intricacy of inflammatory events, approaches should utilize multiple mechanisms to cooperatively attenuate neuroinflammation. Thorough investigation of the mechanism of effective inflammatory modulators, on a molecular and cellular level, will allow for specific targets for engineering strategies to improve microelectrode stability and longevity.
204 Appendix
Protocols
Dip-coating Electrodes:
Clean electrodes: Rinse Si electrodes in ethanol, water, water, water, water. Dry in heating oven for ~ 30 min (until completely dry). Put in eppendorf tube.
Making 10% PVAc in Toluene: 10% w/v = 10g in 100 mL
Dip-coating electrodes: Materials • 10% PVAC in Toluene • Cuvee/petri dish • Electrode
1. Heat up 10% PVAC in Toluene to 60-80 deg C a. Play around with temp. Want a little tacky, not too liquidy and not too sticky b. Keep under evaporation temp of toluene 2. Place a drop on the petri dish 3. Run through drop multiple times (2-3) making sure to stop when gets sticky or else will get a long strip of polymer 4. Put in clean petri dish to dry for at least 10-15 min 5. Take pictures on scope of each probe and number/label each
205 Acid Wash Protocol
**Wear appropriate PPE**
For glass cover slips:
1. Heat 1M HCl to 50-60° C with stir bar in hood. 2. Add coverslips and stir for 1 hour. Maintain temperature at 50-60°C. 3. Remove acid. 4. Wash with 70% EtOH for 5 minutes on bench (no sonication). Repeat with 95% EtOh and dH2O 5. Store in dH2O.
For PVAc:
1. Heat 1M HCl to 50°C in hood. Do not heat above 50°C because PVAc melts at 60°C. 2. Add coverslips and heat for 1 hour. Do not stir- PVAc will break up if stir bar is used. 3. Remove acid. 4. Wash with dH2O for 5 minutes on bench (no sonication). Repeat 6 times. 5. Store in dH2O.
206 APTES Vapor Deposition Protocol
For glass cover slips: 1. Acid wash then plasma treat glass slides (see appropriate protocols). 2. Dry plasma treated slides in oven at until dry (~10-15 minutes). 3. Place up to 12 slides in ceramic coverslip holder. 4. Add 1 mL of 2% APTES into slide holder with lid. a. 1 mL of 2% APTES from APTES stock : 20 uL of APTES and 980 uL toluene b. **After using APTES, flash solution with a stream of N2 gas.** 5. Place coverslip holder in slide holder and seal with autoclave tape. Place foil over lid and seal again with autoclave tape. 6. Use one of the following methods to heat: c. Oven Method: Heat in oven at 150°C for 2 hours. d. Hot Plate Method: Heat on hot plate for 2 hours. Place hot plate on high heat until solution boils and remains boiling for 2 hours. 7. Wash for 5 minutes in sonicator 1x for the following solutions e. Toluene f. 95% EtOh g. dH2O 8. Dry slides with N2 gas. 9. Anneal at 110°C in oven for 15 minutes. 10. Store in dH2O for up to 1 week.
For PVAc: 1. Acid wash then plasma treat glass slides (see appropriate protocols). 2. Dry plasma treated PVAc in oven at until dry at no higher than 40°C (~15-30 minutes). 3. Place up to 12 PVAc pieces in ceramic coverslip holder. 4. Add 1 mL of 2% APTES into slide holder with lid. a. 1 mL of 2% APTES from APTES stock : 20 uL of APTES and 980 uL toluene b. **After using APTES, flash solution with a stream of N2 gas.** 5. Place coverslip holder in slide holder and seal with autoclave tape. Place foil over lid and seal again with autoclave tape. 6. Heat on hot plate for 2 hours. Place hot plate on high heat until solution boils and remains boiling for 2 hours. 7. Wash in dH2O for 5 minutes 3x. *DO NOT SONICATE* 8. Dry slides with N2 gas. 9. Anneal at 40°C in oven for 15 minutes. 10. Store in dH2O for up to 1 week.
207 SOD Mimetic (MnTBAP) Coating Protocol
For glass cover slips (from SOD Paper):
1. Reconstitute MnTBAP in 6 mL of dH2O. Store for up to 1 week at 4°C. 2. Make SOD mimetic solution following the excel sheet. Wrap in foil to protect from light. h. Incubate solution 1 and 2 together for 30 minutes on rocker at RT. i. Add solution 3 to 1+2 and incubate for 5 minutes on rocker at RT. 3. Add 500 uL SOD solution to APTES coated glass coverslips in 12-well plate. Incubate for 24 hrs on rocker at RT. 4. Wash 3x with dH2O in sonicator for 5 min each. 5. Wash 3x with 95% EtOh in sonicator for 5 min each. 6. Wash 3x with dH2O in sonicator for 5 min each. 7. Add 20 mM glycine to each well. Incubate coverslips for 10 min on rocker at RT. 8. Wash 3x with dH2O in sonicator for 5 min each. 9. Keep coated coverlips submerged in dH20 at 37°C.
For PVAc:
1. Reconstitute MnTBAP in 6 mL of dH2O. Store for up to 1 week at 4°C. 2. Make SOD mimetic solution following the excel sheet. Wrap in foil to protect from light. a. Incubate solution 1 and 2 together for 30 minutes on rocker at RT. b. Add solution 3 to 1+2 and incubate for 5 minutes on rocker at RT. 3. Add 500 uL SOD solution to APTES coated glass coverslips in 12-well plate. Incubate for 24 hrs on rocker at RT. 4. Wash 9x with dH2O on bench for 5 min at RT. *DO NOT SONICATE.* 5. Add 20mM glycine to each well. Incubate coverslips for 10 min on rocker at RT. 6. Wash 3x with dH2O on bench for 5 min each at RT. *DO NOT SONICATE.* 7. Keep coated coverslips submerged in dH2O at 37°C.
208 NBT Assay:
Stock solutions
NBT: 4.6 mg/mL in dH2O. Make 1 mL at a time. Store @ 4°C for 1 week Riboflavin: 0.752 mg/mL in dH2O. Add 37.6 mg in 50 mL dH2O. Store @ RT in the dark. KH2PO4 buffer (50 mM KH2PO4, 0.1 mM EDTA): For 250 mL total in dH2O, 1.7 g KH2PO4 and 7.31 mg EDTA, pH 7.0* *Make sure you pH the buffer, the NBT assay will not work if it is not the correct pH
For SOD coated coverslips, 1. Put coverslips in 12 well plate 2. Add 1.5 mL solution per well (see below). Mix well. a. 15 ul NBT b. 5 ul Riboflavin (Mix well!) c. 1.48 mL KH2PO4 solution 3. Cover with aluminum foil. Incubate at 4°C for 30 min 4. Expose to light for 15 min 5. Separate in chloroform. Add 500 uL chloroform (in 4°C fridge) to glass tube. Add 500 uL sample. Vortex. Collect 100 uL of bottom layer and put in 96 well plate. 6. Read absorbance at 560 nm in plate reader
For solution samples, 1. Add 1.5 mL solution per well (see below). Mix well. a. 100 uL sample solution b. 15 ul NBT c. 5 ul Riboflavin (Mix well!) d. 1.38 mL KH2PO4 solution 2. Cover with aluminum foil. Incubate at 4°C for 30 min 3. Expose to light for 15 min 4. Separate in chloroform. Add 500 uL chloroform (in 4°C fridge) to glass tube. Add 500 uL sample. Vortex. Collect 100 uL of bottom layer and put in 96 well plate. 5. Read absorbance at 560 nm in plate reader
209 DPPH Protocol
For Res-releasing PVAc:
1. Incubate 1cm x 3cm PVAc piece in 1mL 1xPBS + 0.5% Tween for desired amount of time at 37°C (in incubator). Use 2mL tubes to hold solution and PVAc. 2. After desired incubation time, add 150uL of solution to 150uL of 100 uM DPPH solution in 96 well plate. a. 100 uM DPPH is in 95% EtOh. 3. Direct read absorbance at 517 nm using plate reader.
210 NSC-34 Neuron Cell line:
To culture: • DMEM, 10% FBS, 1% P/S, 1% Glutamine or Glutamax, no HEPES, no Sodium Pyruvate • Subculture at 60-70% confluency... (usually around 7 million cells) • Splitting usually 1:4 or 1:5 • Freeze around 2 Million cells/vial
To Differentiate: • Seed 5000 cells/sq cm (50/sq mm). Add Differentiation Media • Differentiation Media: DMEM/F12, 1% FBS, 1% Non Essential AA, 0.5% P/S, 1% Glutamine or Glutamax, no phenol red,HEPES, no Sodium Pyruvate and 1micromolar Retinoic Acid • Takes 4 days to differentiate... too long and they will start to die. Change Diff. Media every other day. • Also, the RA is light sensitive and will degrade over time. So you need to add the RA fresh every time. Also turn off the lights when handling RA. • When differentiating, seed neurons to differentiate with normal DMEM and change to differentiation media after 3-4 hrs (Day 0 is day the cells are seeded). Takes 4 full days for neurons to differentiate, but will last about 6 days.
Differentiation media recipe: 25 mL DMEM/F12 250 uL FBS 125 uL Pen/Strep 2.5 uL 100x Non-essential Amino Acids (NEAA) 2.5 uL retinoic acid 0.01M stock (final concentration 1 µM; **light sensitive so turn off lights when using**)
211 BV2 and NSC34 Co-Culture Protocol
To test antioxidant-releasing NC 1. Grow BV-2 and NSC-34 cells in separate flasks in DMEM Dulbecco's Modified Eagle Medium (ATCC) supplemented with 10% fetal bovine serum (Invitrogen) and 1% penicillin–streptomycin (ATCC). 2. Four days prior to intended experimentation, differentiate neurons in a 12-well plate: a. To remove neurons from flask, incubate flask for no more than 5 minutes with 1% trypsin. Wash with at least 5mL media and spin down neurons in conical tube to get pellet. Prior to spin down, count cells using hemocytometer. b. Reconstitute pellet and plate cells in DMEM media i. 5000 cells/cm2, 1mL per well ii. Plate remaining NSC-34 cells in a flask in DMEM media c. After 3-4 hours on Day 0, change media for NSC-34 cells to differentiation media d. After 2 days, change media to fresh differentiation media. e. After 4 days, neurons are ready. 3. After 4 days, seed BV-2 cells on tissue-culture treated 0.4 µm pore polycarbonate membrane inserts (Sigma-Aldrich). a. To remove neurons from flask, incubate flask for no more than 5 minutes with 1% trypsin. Wash with at least 5mL media and spin down neurons in conical tube to get pellet. Prior to spin down, count cells using hemocytometer. b. Reconstitute pellet and plate cells in NSC-34 differentiation media. i. 8000 cells/cm2, 750uL per insert c. Allow cells to seed for at least 15 minutes. d. Meanwhile, plate remaining BV-2 cells in a flask in DMEM Dulbecco's Modified Eagle Medium (ATCC) supplemented with 10% fetal bovine serum (Invitrogen) and 1% penicillin–streptomycin (ATCC). 4. Add inserts to plate with neurons, let incubate for 30 min for cells to adhere. 5. Add surfaces samples into insert. 6. Incubate for 48 hours. 7. Perform desired assays (LiveDead and DHE).
To test SOD coated-glass: Utilize above protocol, except BV2 microglia are attached to SOD coverslip on the bottom well and NSC-34 neurons are added to the insert. 1) Differentiate NSC-34 on 0.4 um pore inserts as above (5000/cm2) for 4 days 2) Adhere BV-2 microglia to control or SOD-coated glass coverslip a. 8000 cells/cm2, bubble on 125 ul per coverslip, NSC-34 differentiation media b. Allow to adhere for at least 30 min c. Add 900 uL media to fill well
212
Live/Dead Assay: NSC-34 cells are very sensitive to repeated washes, so washes were minimized in these assays. 1) Take brightfield images of all wells before performing the assay to ensure the cells are not getting washed away after the assay. 2) Use the Life Technologies LIVE/DEAD® Viability/Cytotoxicity Kit (Life Technologies) 3) Incubate cells with 70% methanol for 10 minutes for negative (all dead) control 4) Add 500 ul of Live/Dead solution into well directly (do not remove media) a. 16 µM ethidium homodimer-1 (EthD-1) and 2 µM calcein-AM in complete phosphate buffered saline (PBS) 5) Incubate for 30 min. 6) Wash once with PBS 7) Immediately, take fluorescent images at 488 nm green/live, 594 nm red/dead
DHE Staining: 1) Take brightfield images of all wells before performing the assay to ensure the cells are not getting washed away after the assay. 2) Wash cells one time with PBS 3) Incubate with 3 um DHE in PBS for 30 min at room temperature 4) Wash one time with PBS 5) Take fluorescent images at 555 nm on microscope 6) To quantify: Use ImageJ to quantify total fluorescent intensity of the image and normalize to control samples 7) For plate reader: Transfer coverslips to black plate. Read on plate reader at 480 nm excitation, 586 emission
213 References: [1] M. A. L. Nicolelis, "Brain–machine interfaces to restore motor function and probe neural circuits," Nature Reviews Neuroscience, vol. 4, pp. 417-422, 2003. [2] M. P. Ward, P. Rajdev, C. Ellison, and P. P. Irazoqui, "Toward a comparison of microelectrodes for acute and chronic recordings," Brain Research, vol. 1282, pp. 183-200, 2009. [3] J. C. Barrese, N. Rao, K. Paroo, et al., "Failure mode analysis of silicon-based intracortical microelectrode arrays in non-human primates," J Neural Eng, vol. 10, p. 066014, 2013. [4] J. Subbaroyan, D. C. Martin, and D. R. Kipke, "A finite-element model of the mechanical effects of implantable microelectrodes in the cerebral cortex," Journal of Neural Engineering, vol. 2, pp. 103-113, 2005. [5] H. Lee, R. V. Bellamkonda, W. Sun, and M. E. Levenston, "Biomechanical analysis of silicon microelectrode-induced strain in the brain," Journal of Neural Engineering, vol. 2, pp. 81-9, 2005. [6] A. Sridharan, S. D. Rajan, and J. Muthuswamy, "Long-term changes in the material properties of brain tissue at the implant–tissue interface," Journal of Neural Engineering, vol. 10, p. 066001, 2013. [7] P. J. Rousche, D. S. Pellinen, D. P. Pivin, et al., "Flexible polyimide-based intracortical electrode arrays with bioactive capability," IEEE Trans Biomed Eng, vol. 48, pp. 361-371, 2001. [8] L. Karumbaiah, T. Saxena, D. Carlson, et al., "Relationship between intracortical electrode design and chronic recording function," Biomaterials, vol. 34, pp. 8061- 8074, 2013. [9] R. L. Rennaker, J. Miller, H. Tang, and D. A. Wilson, "Minocycline increases quality and longevity of chronic neural recordings," J Neural Eng, vol. 4, pp. L1- 5, Jun 2007. [10] T. Ware, D. Simon, D. E. Arreaga-Salas, et al., "Fabrication of responsive, softening neural interfaces," Advanced Functional Materials, vol. 22, pp. 3470- 3479, 2012. [11] T. Saxena, L. Karumbaiah, E. A. Gaupp, et al., "The impact of chronic blood–brain barrier breach on intracortical electrode function," Biomaterials, vol. 34, pp. 4703- 4713, 2013. [12] B. D. Winslow, M. B. Christensen, W. K. Yang, F. Solzbacher, and P. A. Tresco, "A comparison of the tissue response to chronically implanted Parylene-C-coated and uncoated planar silicon microelectrode arrays in rat cortex," Biomaterials, vol. 31, pp. 9163-9172, 2010. [13] J. Skousen, S. Merriam, O. Srivannavit, et al., "Reducing surface area while maintaining implant penetrating profile lowers the brain foreign body response to chronically implanted planar silicon microelectrode arrays," Prog Brain Res, vol. 194C, pp. 167-180, 2011. [14] K. A. Potter, M. Jorfi, K. T. Householder, et al., "Curcumin-releasing mechanically-adaptive intracortical implants improve the proximal neuronal density and blood-brain barrier stability," Acta Biomaterialia, vol. 10, pp. 2209- 2222, 2014.
214 [15] K. A. Potter, A. C. Buck, W. K. Self, et al., "The effect of resveratrol on neurodegeneration and blood brain barrier stability surrounding intracortical microelectrodes," Biomaterials, vol. 34, pp. 7001-7015, 2013. [16] J. Subbaroyan and D. R. Kipke, "The role of flexible polymer interconnects in chronic tissue response induced by intracortical microelectrodes--a modeling and an in vivo study," in 28th Annual International Conference of the Engineering in Medicine and Biology Society (IEEE-EMBS), 2006, pp. 3588-91. [17] S. Takeuchi, D. Ziegler, Y. Yoshida, K. Mabuchi, and T. Suzuki, "Parylene flexible neural probes integrated with microfluidic channels," Lab on a Chip, vol. 5, pp. 519-523, 2005. [18] L. J. Fernandez, A. Altuna, M. Tijero, et al., "Study of functional viability of SU- 8-based microneedles for neural applications," Journal of Micromechanics and Microengineering, vol. 19, p. 025007, 2009. [19] J. K. Nguyen, D. J. Park, J. L. Skousen, et al., "Mechanically-compliant intracortical implants reduce the neuroinflammatory response," Journal of Neural Engineering, vol. 11, p. 056014, 2014. [20] J. P. Harris, J. R. Capadona, R. H. Miller, et al., "Mechanically adaptive intracortical implants improve the proximity of neuronal cell bodies," Journal of Neural Engineering, vol. 8, p. 066011, 2011. [21] K. Shanmuganathan, J. R. Capadona, S. J. Rowan, and C. Weder, "Stimuli- responsive mechanically adaptive polymer nanocomposites," ACS Applied Materials & Interfaces, vol. 2, pp. 165-174, 2010. [22] J. R. Capadona, K. Shanmuganathan, D. J. Tyler, S. J. Rowan, and C. Weder, "Stimuli-responsive polymer nanocomposites inspired by the sea cucumber dermis," Science, vol. 319, p. 1370, 2008. [23] J. P. Harris, A. E. Hess, S. J. Rowan, et al., "In vivo deployment of mechanically adaptive nanocomposites for intracortical microelectrodes," Journal of Neural Engineering, vol. 8, p. 046010, 2011. [24] K. A. Potter, A. C. Buck, W. K. Self, and J. R. Capadona, "Stab injury and device implantation within the brain results in inversely multiphasic neuroinflammatory and neurodegenerative responses," J Neural Eng, vol. 9, p. 046020, 2012. [25] P. Moshayedi, G. Ng, J. C. F. Kwok, et al., "The relationship between glial cell mechanosensitivity and foreign body reactions in the central nervous system," Biomaterials, vol. 35, pp. 3919-3925, 2014. [26] L. Karumbaiah, S. E. Norman, N. B. Rajan, et al., "The upregulation of specific interleukin (IL) receptor antagonists and paradoxical enhancement of neuronal apoptosis due to electrode induced strain and brain micromotion," Biomaterials, vol. 33, pp. 5983-5996, 2012. [27] A. A. Sharp, A. M. Ortega, D. Restrepo, D. Curran-Everett, and K. Gall, "In vivo penetration mechanics and mechanical properties of mouse brain tissue at micrometer scales," IEEE Trans Biomed Eng, vol. 56, pp. 45-53, Jan 2009. [28] T. D. Y. Kozai, K. Catt, X. Li, et al., "Mechanical failure modes of chronically implanted planar silicon-based neural probes for laminar recording," Biomaterials, vol. 37, pp. 25-39, 2015. [29] R. Zhu, G. L. Huang, H. Yoon, C. S. Smith, and V. K. Varadan, "Biomechanical Strain Analysis at the Interface of Brain and Nanowire Electrodes on a Neural
215 Probe," Journal of Nanotechnology in Engineering and Medicine, vol. 2, p. 031001, 2011. [30] A. Sridharan, J. K. Nguyen, J. R. Capadona, and J. Muthuswamy, "Compliant intracortical implants reduce strains and strain rates in brain tissuein vivo," Journal of Neural Engineering, vol. 12, p. 036002, 2015. [31] D. Salvemini, D. P. Riley, and S. Cuzzocrea, "SOD mimetics are coming of age," Nature Reviews, vol. 1, pp. 367-374, 2002. [32] P. S. Hume and K. S. Anseth, "Polymerizable superoxide dismutase mimetic protects cells encapsulated in poly(ethylene glycol) hydrogels from reactive oxygen species-mediated damage," Journal of biomedical materials research. Part A, vol. 99, pp. 29-37, Oct 2011. [33] J. Peng, F. F. Stevenson, S. R. Doctrow, and J. K. Andersen, "Superoxide dismutase/catalase mimetics are neuroprotective against selective paraquat- mediated dopaminergic neuron death in the substantial nigra: implications for Parkinson disease," J Biol Chem, vol. 280, pp. 29194-8, Aug 12 2005. [34] I. Batinic-Haberle, S. Cuzzocrea, J. S. Reboucas, et al., "Pure MnTBAP selectively scavenges peroxynitrite over superoxide: comparison of pure and commercial MnTBAP samples to MnTE-2-PyP in two models of oxidative stress injury, an SOD-specific Escherichia coli model and carrageenan-induced pleurisy," Free Radical Biology & Medicine, vol. 46, pp. 192-201, Jan 15 2009. [35] M. Zahmatkesh, M. Kadkhodaee, S. M. S. Moosavi, et al., "Beneficial effects of MnTBAP, a broad-spectrum reactive species scavenger, in rat renal ischemia/reperfusion injury," Clinical and Experimental Nephrology, vol. 9, pp. 212-218, 2005. [36] K. A. Potter-Baker, J. K. Nguyen, K. M. Kovach, et al., "Development of superoxide dismutase mimetic surfaces to reduce accumulation of reactive oxygen species surrounding intracortical microelectrodes " Journal of Materials Chemistry B, vol. 2, pp. 2248-2258, 2014. [37] M. Piccolino, "Luigi Galvani and animal electricity: two centuries after the foundation of electrophysiology," Trends in Neurosciences, vol. 20, pp. 443-448, 1997. [38] M. M. Merzenich, R. J. Nelson, M. P. Stryker, et al., "Somatosensory cortical map changes following digit amputation in adult monkeys," J Comp Neurol, vol. 224, pp. 591-605, Apr 20 1984. [39] D. V. Buonomano and M. M. Merzenich, "Cortical plasticity: from synapses to maps," Annual Review of Neuroscience, vol. 21, pp. 149-86, 1998. [40] J. R. Wolpaw, N. Birbaumer, W. J. Heetderks, et al., "Brain-computer interface technology: a review of the first international meeting," IEEE transactions on rehabilitation engineering, vol. 8, pp. 164-173, 2000. [41] M. Jorfi, J. L. Skousen, C. Weder, and J. R. Capadona, "Progress towards biocompatible intracortical microelectrodes for neural interfacing applications," Journal of Neural Engineering, vol. 12, p. 011001, 2015. [42] B. Gunasekera, T. Saxena, R. Bellamkonda, and L. Karumbaiah, "Intracortical recording interfaces: current challenges to chronic recording function," ACS Chemical Neuroscience, vol. 6, pp. 68-83, 2015.
216 [43] M. A. Lebedev and M. A. Nicolelis, "Brain-machine interfaces: past, present and future," Trends in Neuroscience, vol. 29, pp. 536-46, Sep 2006. [44] L. F. Nicolas-Alonso and J. Gomez-Gil, "Brain Computer Interfaces, a review," Sensors, vol. 12, pp. 1211-1279, 2012. [45] E. J. Tehovnik, L. C. Woods, and W. M. Slocum, "Transfer of information by BMI," Neuroscience, vol. 255, pp. 134-146, 2013. [46] G. Baranauskas, "What limits the performance of current invasive brain machine interfaces?," Frontiers in Systems Neuroscience, vol. 8, 2014. [47] R. C. Gesteland, B. Howland, J. Y. Lettvin, and W. H. Pitts, "Comments on Microelectrodes," Proceedings of the IRE, vol. 47, pp. 1856-1862, 1959. [48] D. A. Robinson, "The electrical properties of metal microelectrodes," Proceedings of the IEEE, vol. 56, pp. 1065-1071, 1968. [49] T. D. Parker, D. D. Strachan, and W. I. Welker, "Tungsten ball microelectrode for extracellular single-unit recording," Electroencephalography and Clinical Neurophysiology, vol. 35, pp. 647-651, 1973. [50] S. L. BeMent, K. D. Wise, D. J. Anderson, K. Najafi, and K. L. Drake, "Solid-state electrodes for multichannel multiplexed intracortical neuronal recording," IEEE Trans Biomed Eng, vol. 33, pp. 230-41, Feb 1986. [51] J. D. Green, "A simple microelectrode for recording from the central nervous system," Nature, vol. 182, pp. 962-962, 1958. [52] J. P. Donoghue, J. N. Sanes, N. G. Hatsopoulos, and G. Gaal, "Neural discharge and local field potential oscillations in primate motor cortex during voluntary movements," J Neurophysiol, vol. 79, pp. 159-73, Jan 1998. [53] J. C. Williams, R. L. Rennaker, and D. R. Kipke, "Long-term neural recording characteristics of wire microelectrode arrays implanted in cerebral cortex," Brain Research, vol. 4, pp. 303-13, Dec 1999. [54] M. A. Nicolelis, D. Dimitrov, J. M. Carmena, et al., "Chronic, multisite, multielectrode recordings in macaque monkeys," Proc Natl Acad Sci USA, vol. 100, pp. 11041-6, Sep 16 2003. [55] J. M. Carmena, M. A. Lebedev, R. E. Crist, et al., "Learning to control a brain- machine interface for reaching and grasping by primates," PLoS Biol, vol. 1, p. E42, Nov 2003. [56] R. L. Rennaker, A. M. Ruyle, S. E. Street, and A. M. Sloan, "An economical multi- channel cortical electrode array for extended periods of recording during behavior," Journal of Neuroscience Methods, vol. 142, pp. 97-105, 2005. [57] A. Prasad, Q.-S. Xue, R. Dieme, et al., "Abiotic-biotic characterization of Pt/Ir microelectrode arrays in chronic implants," Frontiers in Neuroengineering, vol. 7, 2014. [58] M. A. Freire, E. Morya, J. Faber, et al., "Comprehensive analysis of tissue preservation and recording quality from chronic multielectrode implants," PLoS One, vol. 6, p. e27554, 2011. [59] J. Wessberg, C. R. Stambaugh, J. D. Kralik, et al., "Real-time prediction of hand trajectory by ensembles of cortical neurons in primates," Nature, vol. 408, pp. 361- 5, Nov 16 2000.
217 [60] J. K. Chapin, K. A. Moxon, R. S. Markowitz, and M. A. Nicolelis, "Real-time control of a robot arm using simultaneously recorded neurons in the motor cortex," Nat Neurosci, vol. 2, pp. 664-70, Jul 1999. [61] A. Prasad, Q.-S. Xue, V. Sankar, et al., "Comprehensive characterization and failure modes of tungsten microwire arrays in chronic neural implants," Journal of Neural Engineering, vol. 9, p. 056015, 2012. [62] E. Patrick, M. E. Orazem, J. C. Sanchez, and T. Nishida, "Corrosion of tungsten microelectrodes used in neural recording applications," Journal of Neuroscience Methods, vol. 198, pp. 158-71, 2011. [63] J. C. Sanchez, N. Alba, T. Nishida, C. Batich, and P. R. Carney, "Structural modifications in chronic microwire electrodes for cortical neuroprosthetics: a case study," IEEE Trans Neural Syst Rehabil Eng, vol. 14, pp. 217-221, 2006. [64] B. D. Winslow and P. A. Tresco, "Quantitative analysis of the tissue response to chronically implanted microwire electrodes in rat cortex," Biomaterials, vol. 31, pp. 1558-67, Mar 2010. [65] K. D. Wise, D. J. Anderson, J. F. Hetke, D. R. Kipke, and K. Najafi, "Wireless implantable microsystems: high-density electronic interfaces to the nervous system," Proceedings of the IEEE, vol. 92, pp. 76-97, 2004. [66] K. D. Wise and J. B. Angell, "A low-capacitance multielectrode probe for use in extracellular neurophysiology," IEEE Trans Biomed Eng, vol. 22, pp. 212-9, May 1975. [67] D. R. Kipke, R. J. Vetter, J. C. Williams, and J. F. Hetke, "Silicon-substrate intracortical microelectrode arrays for long-term recording of neuronal spike activity in cerebral cortex," IEEE Trans Neural Syst Rehabil Eng, vol. 11, pp. 151- 5, Jun 2003. [68] I. NeuroNexus Technolgies. (2015, March 30, 2015). NeuroNexus 2015 Research Product Catalog (1 ed.). Available: http://neuronexus.com/images/Catalogs/2015_NNxCatalog_WebRes.p df [69] R. J. Vetter, J. C. Williams, E. A. Nunamaker, and D. R. Kipke, "Chronic neural recording using silicon-substrate microelectrode arrays implanted in cerebral cortex," IEEE Transactions in Biomedical Engineering, vol. 51, pp. 896-904, 2004. [70] R. Biran, D. C. Martin, and P. A. Tresco, "Neuronal cell loss accompanies the brain tissue response to chronically implanted silicon microelectrode arrays," Experimental Neurology, vol. 195, pp. 115-126, 2005. [71] C. T. Nordhausen, E. M. Maynard, and R. A. Normann, "Single unit recording capabilities of a 100 microelectrode array," Brain Research, vol. 726, pp. 129-140, 1996. [72] E. M. Maynard, C. T. Nordhausen, and R. A. Normann, "The Utah Intracortical Electrode Array: A recording structure for potential brain-computer interfaces," Electroencephalography and Clinical Neurophysiology, vol. 102, pp. 228-239, 1997. [73] S. J. Kim, S. C. Manyam, D. J. Warren, and R. A. Normann, "Electrophysiological mapping of cat primary auditory cortex with multielectrode arrays," Annals of Biomedical Engineering, vol. 34, pp. 300-9, Feb 2006.
218 [74] D. J. Warren, E. Fernandez, and R. A. Normann, "High-resolution two-dimensional spatial mapping of cat striate cortex using a 100-microelectrode array," Neuroscience, vol. 105, pp. 19-31, 2001. [75] R. A. Normann, E. M. Maynard, P. J. Rousche, and D. J. Warren, "A neural interface for a cortical vision prosthesis," Vision Res, vol. 39, pp. 2577-87, Jul 1999. [76] P. J. Rousche and R. A. Normann, "Chronic recording capability of the Utah Intracortical Electrode Array in cat sensory cortex," Journal of Neuroscience Methods, vol. 82, pp. 1-15, 1998. [77] S. Suner, M. R. Fellows, C. Vargas-Irwin, G. K. Nakata, and J. P. Donoghue, "Reliability of signals from a chronically implanted, silicon-based electrode array in non-human primate primary motor cortex," IEEE Trans. on Rehabilitation Engineering, vol. 13, pp. 524-541, 2005. [78] J. Egan, J. Baker, P. House, and B. Greger, "Detection and classification of multiple finger movements using a chronically implanted Utah Electrode Array," in Engineering in Medicine and Biology Society, EMBC, 2011 Annual International Conference of the IEEE, 2011, pp. 7320-7323. [79] L. R. Hochberg, M. D. Serruya, G. M. Friehs, et al., "Neuronal ensemble control of prosthetic devices by a human with tetraplegia," Nature, vol. 442, pp. 164-71, Jul 13 2006. [80] J. P. Donoghue, A. Nurmikko, M. Black, and L. R. Hochberg, "Assistive technology and robotic control using motor cortex ensemble-based neural interface systems in humans with tetraplegia," Journal of Physiology, vol. 579, pp. 603-611, 2007. [81] J. D. Simeral, S.-P. Kim, M. J. Black, J. P. Donoghue, and L. R. Hochberg, "Neural control of cursor trajectory and click by a human with tetraplegia 1000 days after implant of an intracortical microelectrode array," Journal of Neural Engineering, vol. 8, p. 025027, 2011. [82] L. R. Hochberg, D. Bacher, B. Jarosiewicz, et al., "Reach and grasp by people with tetraplegia using a neurally controlled robotic arm," Nature, vol. 485, pp. 372-5, May 17 2012. [83] S.-P. Kim, J. D. Simeral, L. R. Hochberg, J. P. Donoghue, and M. J. Black, "Neural control of computer cursor velocity by decoding motor cortical spiking activity in humans with tetraplegia," Journal of Neural Engineering, vol. 5, pp. 455-476, 2008. [84] P. J. Rousche and R. A. Normann, "A method for pneumatically inserting an array of penetrating electrodes into cortical tissue," Annals of Biomedical Engineering, vol. 20, pp. 413-22, 1992. [85] E. M. Maynard, E. Fernandez, and R. A. Normann, "A technique to prevent dural adhesions to chronically implanted microelectrode arrays," J Neurosci Methods, vol. 97, pp. 93-101, Apr 15 2000. [86] S. P. Burns, D. Xing, and R. M. Shapley, "Comparisons of the dynamics of local field potential and multiunit activity signals in macaque visual cortex," The Journal of Neuroscience, vol. 30, pp. 13739-13749, October 13, 2010 2010. [87] U. Mitzdorf, "Current source-density method and application in cat cerebral cortex: investigation of evoked potentials and EEG phenomena," Physiol Rev, vol. 65, pp. 37-100, Jan 1985.
219 [88] S. Katzner, I. Nauhaus, A. Benucci, et al., "Local origin of field potentials in visual cortex," Neuron, vol. 61, pp. 35-41, 2009. [89] G. Buzsáki, "Large-scale recording of neuronal ensembles," Nat Neurosci, vol. 7, pp. 446-451, 2004. [90] G. Santhanam, M. D. Linderman, V. Gilja, et al., "HermesB: a continuous neural recording system for freely behaving primates," IEEE Trans Biomed Eng, vol. 54, pp. 2037-50, Nov 2007. [91] D. Wang, Q. Zhang, Y. Li, et al., "Long-term decoding stability of local field potentials from silicon arrays in primate motor cortex during a 2D center out task," Journal of Neural Engineering, vol. 11, p. 036009, 2014. [92] R. D. Flint, Z. A. Wright, and M. W. Slutzky, "Control of a biomimetic brain machine interface with local field potentials: Performance and stability of a static decoder over 200 days," in 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2012, pp. 6719-6722. [93] R. D. Flint, Z. A. Wright, M. R. Scheid, and M. W. Slutzky, "Long term, stable brain machine interface performance using local field potentials and multiunit spikes," Journal of Neural Engineering, vol. 10, p. 056005, 2013. [94] J. C. Williams, J. A. Hippensteel, J. Dilgen, W. Shain, and D. R. Kipke, "Complex impedance spectroscopy for monitoring tissue responses to inserted neural implants," Journal of Neural Engineering, vol. 4, pp. 410-23, Dec 2007. [95] C. A. Chestek, V. Gilja, P. Nuyujukian, et al., "Long-term stability of neural prosthetic control signals from silicon cortical arrays in rhesus macaque motor cortex," Journal of Neural Engineering, vol. 8, p. 045005, Aug 2011. [96] N. F. Nolta, M. B. Christensen, P. D. Crane, J. L. Skousen, and P. A. Tresco, "BBB leakage, astrogliosis, and tissue loss correlate with silicon microelectrode array recording performance," Biomaterials, vol. 53, pp. 753-762, 2015. [97] B. D. Burns, J. P. Stean, and A. C. Webb, "Recording for several days from single cortical neurons in completely unrestrained cats," Electroencephalography and Clinical Neurophysiology, vol. 36, pp. 314-8, Mar 1974. [98] X. Liu, D. B. McCreery, R. R. Carter, et al., "Stability of the interface between neural tissue and chronically implanted intracortical microelectrodes," IEEE Trans Rehabil Eng, vol. 7, pp. 315-26, Sep 1999. [99] X. Liu, D. B. McCreery, L. A. Bullara, and W. F. Agnew, "Evaluation of the stability of intracortical microelectrode arrays," IEEE Trans Neural Syst Rehabil Eng, vol. 14, pp. 91-100, Mar 2006. [100] J. N. Sanes and J. P. Donoghue, "Plasticity and primary motor cortex," Annual Review of Neuroscience, vol. 23, pp. 393-415, 2000. [101] V. S. Polikov, P. A. Tresco, and W. M. Reichert, "Response of brain tissue to chronically implanted neural electrodes," J Neurosci Methods, vol. 148, pp. 1-18, 2005. [102] G. C. McConnell, H. D. Rees, A. I. Levey, et al., "Implanted neural electrodes cause chronic, local inflammation that is correlated with local neurodegeneration," J Neural Eng, vol. 6, p. 056003, 2009. [103] P. A. Tresco and B. D. Winslow, "The challenge of integrating devices into the central nervous system," Critical Reviews in Biomedical Engineering, vol. 39, pp. 29-44, 2011.
220 [104] Y. Zhong and R. V. Bellamkonda, "Biomaterials for the central nervous system," Journal of the Royal Society Interface, vol. 5, pp. 957-975, 2008. [105] A. J. Woolley, H. A. Desai, and K. J. Otto, "Chronic intracortical microelectrode arrays induce non-uniform, depth-related tissue responses," Journal of Neural Engineering, vol. 10, p. 026007, 2013. [106] G. C. McConnell, "Chronic inflammation surrounding intra-cortical emectrodes is correlated with a local neurodegenerative state," Ph.D, Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, 2008. [107] D. M. D. Landis, "The early reactions of non-neuronal cells to brain injury," Annual Review of Neuroscience, vol. 17, pp. 133-151, 1994. [108] J. D. Stroncek and W. M. Reichert, Eds., Overview of wound dealing in different tissue types (Frontiers in Neuroengineering: Indwelling Neural Implants-Strategies for Contending with the in vivo Environment. CRC Press, 2008, p.^pp. Pages. [109] D. H. Szarowski, M. D. Andersen, S. Retterer, et al., "Brain responses to micro- machined silicon devices," Brain Research, vol. 983, pp. 23-35, 2003. [110] E. Preston, J. Webster, and D. Small, "Characteristics of sustained blood-brain barrier opening and tissue injury in a model for focal trauma in the rat," Journal of Neurotrauma, vol. 18, pp. 83-92, Jan 2001. [111] C. S. Bjornsson, S. J. Oh, Y. A. Al-Kofahi, et al., "Effects of insertion conditions on tissue strain and vascular damage during neuroprosthetic device insertion," J Neural Eng, vol. 3, pp. 196-207, Sep 2006. [112] C. Tian and J. He, "Monitoring Insertion Force and Electrode Impedance during Implantation of Microwire Electrodes," 27th Annual International Conference of the Engineering in Medicine and Biology Society (IEEE-EMBS), vol. 7, pp. 7333- 6, 2005. [113] T. D. Y. Kozai, N. B. Langhals, P. R. Patel, et al., "Ultrasmall implantable composite microelectrodes with bioactive surfaces for chronic neural interfaces," Nature Materials, vol. 11, pp. 1065-1073, 2012. [114] T. D. Y. Kozai, T. C. Marzullo, F. Hooi, et al., "Reduction of neurovascular damage resulting from microelectrode insertion into the cerebral cortex using in vivo two- photon mapping," Journal of Neural Engineering, vol. 7, p. 046011, 2010. [115] T. D. Y. Kozai, A. L. Vazquez, C. L. Weaver, S.-G. Kim, and X. T. Cui, "In vivo two-photon microscopy reveals immediate microglial reaction to implantation of microelectrode through extension of processes," Journal of Neural Engineering, vol. 9, p. 066001, 2012. [116] H. Kettenmann, U.-K. Hanisch, M. Noda, and A. Verkhratsky, "Physiology of microglia," Physiol Rev, vol. 91, pp. 461-553, 2011. [117] U. K. Hanisch and H. Kettenmann, "Microglia: active sensor and versatile effector cells in the normal and pathologic brain," Nature neuroscience, vol. 10, pp. 1387- 94, 2007. [118] F. Aloisi, "Immune function of microglia," Glia, vol. 36, pp. 165-179, 2001. [119] A. Mantovani, A. Sica, S. Sozzani, et al., "The chemokine system in diverse forms of macrophage activation and polarization," Trends in Immunology, vol. 25, pp. 677-686, 2004. [120] F. O. Martinez, A. Sica, A. Mantovani, and M. Locati, "Macrophage activation and polarization," Frontiers in Bioscience, vol. 13, pp. 453-61, 2008.
221 [121] M. Ravikumar, S. Jain, R. H. Miller, J. R. Capadona, and S. M. Selkirk, "An organotypic spinal cord slice culture model to quantify neurodegeneration," Journal of Neuroscience Methods, vol. 211, pp. 280-288, 2012. [122] G. Gloire, S. Legrand-Poels, and J. Piette, "NF-κB activation by reactive oxygen species: Fifteen years later," Biochemical Pharmacology, vol. 72, pp. 1493-1505, 2006. [123] M. Pekny and M. Nilsson, "Astrocyte activation and reactive gliosis," Glia, vol. 50, pp. 427-34, Jun 2005. [124] M. D. Norenberg, "Astrocyte responses to CNS injury," Journal of Neuropathology & Experimental Neurology, vol. 53, pp. 213-20, May 1994. [125] N. J. Abbott, L. Rönnbäck, and E. Hansson, "Astrocyte–endothelial interactions at the blood–brain barrier," Nature Reviews Neuroscience, vol. 7, pp. 41-53, 2006. [126] M. J. Carson, J. C. Thrash, and B. Walter, "The cellular response in neuroinflammation: The role of leukocytes, microglia and astrocytes in neuronal death and survival," Clinical Neuroscience Research, vol. 6, pp. 237-245, Dec 2006. [127] R. J. McKeon, M. J. Jurynec, and C. R. Buck, "The chondroitin sulfate proteoglycans neurocan and phosphacan are expressed by reactive astrocytes in the chronic CNS glial scar," Journal of Neuroscience, vol. 19, pp. 10778-88, Dec 15 1999. [128] T. Roitbak and E. Sykova, "Diffusion barriers evoked in the rat cortex by reactive astrogliosis," Glia, vol. 28, pp. 40-8, Oct 1999. [129] E. Sykova, "Glia and volume transmission during physiological and pathological states," J Neural Transm, vol. 112, pp. 137-47, Jan 2005. [130] Y. T. Kim, M. J. Bridge, and P. A. Tresco, "The influence of the foreign body response evoked by fibroblast transplantation on soluble factor diffusion in surrounding brain tissue," Journal of controlled release : official journal of the Controlled Release Society, vol. 118, pp. 340-7, Apr 23 2007. [131] G. C. McConnell, R. J. Butera, and R. V. Bellamkonda, "Bioimpedance modeling to monitor astrocytic response to chronically implanted electrodes," Journal of Neural Engineering, vol. 6, p. 055005, 2009. [132] P. Takmakov, K. Ruda, K. Scott Phillips, et al., "Rapid evaluation of the durability of cortical neural implants using accelerated aging with reactive oxygen species," Journal of Neural Engineering, vol. 12, p. 026003, 2015. [133] Y. K. Hirshler, U. Polat, and A. Biegon, "Intracranial electrode implantation produces regional neuroinflammation and memory deficits in rats," Experimental Neurology, vol. 222, pp. 42-50, Mar 2010. [134] J. P. Seymour and D. R. Kipke, "Neural probe design for reduced tissue encapsulation in CNS," Biomaterials, Apr 5 2007. [135] U. Neniskyte, J. J. Neher, and G. C. Brown, "Neuronal death induced by nanomolar amyloid beta is mediated by primary phagocytosis of neurons by microglia," J Biol Chem, vol. 286, pp. 39904-13, Nov 18 2011. [136] J. J. Neher, U. Neniskyte, J. W. Zhao, et al., "Inhibition of microglial phagocytosis is sufficient to prevent inflammatory neuronal death," Journal of immunology, vol. 186, pp. 4973-83, Apr 15 2011.
222 [137] M. E. Lull and M. L. Block, "Microglial activation and chronic neurodegeneration," Neurotherapeutics, vol. 7, pp. 354-65, Oct 2010. [138] A. D. Kraft and G. J. Harry, "Features of microglia and neuroinflammation relevant to environmental exposure and neurotoxicity," International journal of environmental research and public health, vol. 8, pp. 2980-3018, Jul 2011. [139] E. R. Hascup, S. af Bjerken, K. N. Hascup, et al., "Histological studies of the effects of chronic implantation of ceramic-based microelectrode arrays and microdialysis probes in rat prefrontal cortex," Brain Research, vol. 1291, pp. 12-20, Sep 29 2009. [140] A. E. Cardona, E. P. Pioro, M. E. Sasse, et al., "Control of microglial neurotoxicity by the fractalkine receptor," Nature Neuroscience, vol. 9, pp. 917-24, Jul 2006. [141] G. C. Brown and J. J. Neher, "Inflammatory neurodegeneration and mechanisms of microglial killing of neurons," Mol Neurobiol, vol. 41, pp. 242-7, Jun 2010. [142] A. M. Palmer, "The role of the blood-CNS barrier in CNS disorders and their treatment," Neurobiology of Disease, vol. 37, pp. 3-12, Jan 2010. [143] N. J. Abbott, "Inflammatory mediators and modulation of blood-brain barrier permeability," Cellular and molecular neurobiology, vol. 20, pp. 131-47, Apr 2000. [144] J. Subbaroyan, "Investigations of tethering induced injury response in brain tissue by intracortical implants through modeling and in vivo experiments," Ph.D., Biomedical Engineering, University of Michigan, Ann Arbor, MI, 2007. [145] R. Biran, D. C. Martin, and P. A. Tresco, "The brain tissue response to implanted silicon microelectrode arrays is increased when the device is tethered to the skull," Journal of Biomedical Materials Research Part A, vol. 82, pp. 169-78, 2007. [146] L. Spataro, J. Dilgen, S. Retterer, et al., "Dexamethasone treatment reduces astroglia responses to inserted neuroprosthetic devices in rat neocortex," Experimental Neurology, vol. 194, pp. 289-300, 2005. [147] W. Shain, L. Spataro, J. Dilgen, et al., "Controlling cellular reactive responses around neural prosthetic devices using peripheral and local intervention strategies," IEEE Trans Neural Syst Rehabil Eng, vol. 11, pp. 186-188, 2003. [148] Y. Zhong, G. C. McConnell, J. D. Ross, S. P. DeWeerth, and R. V. Bellamkonda, "A novel dexamethasone-releasing, anti-inflammatory coating for neural implants," in Neural Engineering, 2005. Conference Proceedings. 2nd International IEEE EMBS Conference on, 2005, pp. 522-525. [149] Y. Zhong and R. V. Bellamkonda, "Dexamethasone-coated neural probes elicit attenuated inflammatory response and neuronal loss compared to uncoated neural probes," Brain Research, vol. 1148, pp. 15-27, May 7 2007. [150] X. Luo, C. Matranga, S. Tan, N. Alba, and X. T. Cui, "Carbon nanotube nanoreservior for controlled release of anti-inflammatory dexamethasone," Biomaterials, 2011. [151] A. Mercanzini, S. T. Reddy, D. Velluto, et al., "Controlled release nanoparticle- embedded coatings reduce the tissue reaction to neuroprostheses," Journal of Controlled Release, vol. 145, pp. 196-202, Aug 3 2010. [152] Y. Zhong and R. V. Bellamkonda, "Controlled release of anti-inflammatory agent α-MSH from neural implants," Journal of Controlled Release, vol. 106, pp. 309- 318, 2005. [153] B. J. Kim, J. T. Kuo, S. A. Hara, et al., "3D Parylene sheath neural probe for chronic recordings," Journal of Neural Engineering, vol. 10, p. 045002, Aug 2013.
223 [154] P. R. Kennedy, "The cone electrode: a long-term electrode that records from neurites grown onto its recording surface," Journal of Neuroscience Methods, vol. 29, pp. 181-93, Sep 1989. [155] W. He, G. C. McConnell, and R. V. Bellamkonda, "Nanoscale laminin coating modulates cortical scarring response around implanted silicon microelectrode arrays," J Neural Eng, vol. 3, pp. 316-26, Dec 2006. [156] H. S. Koh, T. Yong, C. K. Chan, and S. Ramakrishna, "Enhancement of neurite outgrowth using nano-structured scaffolds coupled with laminin," Biomaterials, vol. 29, pp. 3574-3582, 2008. [157] E. Azemi, W. R. Stauffer, M. S. Gostock, C. F. Lagenaur, and X. T. Cui, "Surface immobilization of neural adhesion molecule L1 for improving the biocompatibility of chronic neural probes: In vitro characterization," Acta Biomaterialia, Mar 20 2008. [158] E. Azemi, C. F. Lagenaur, and X. T. Cui, "The surface immobilization of the neural adhesion molecule L1 on neural probes and its effect on neuronal density and gliosis at the probe/tissue interface," Biomaterials, vol. 32, pp. 681-692, 2011. [159] C. L. Kolarcik, D. Bourbeau, E. Azemi, et al., "In vivo effects of L1 coating on inflammation and neuronal health at the electrode–tissue interface in rat spinal cord and dorsal root ganglion," Acta Biomaterialia, vol. 8, pp. 3561-3575, 10// 2012. [160] E. Azemi, G. T. Gobbel, and X. T. Cui, "Seeding neural progenitor cells on silicon- based neural probes," Journal of Neurosurgery, vol. 113, pp. 673-681, 2010. [161] R. A. Green, K. S. Lim, W. C. Henderson, et al., "Living electrodes: Tissue engineering the neural interface," in Engineering in Medicine and Biology Society (EMBC), 2013 35th Annual International Conference of the IEEE, 2013, pp. 6957- 6960. [162] T. D. Y. Kozai, A. S. Jaquins-Gerstl, A. L. Vazquez, A. C. Michael, and X. T. Cui, "Brain tissue responses to neural implants impact signal sensitivity and intervention strategies," ACS Chemical Neuroscience, vol. 6, pp. 48-67, 2015. [163] S. R. Goldstein and M. Salcman, "Mechanical factors in the design of chronic recording intracortical microelectrodes," IEEE Trans Biomed Eng, vol. 20, pp. 260- 9, 1973. [164] M. Salcman and M. J. Bak, "A new chronic recording intracortical microelectrode," Med Biol Eng, vol. 14, pp. 42-50, Jan 1976. [165] M. Salcman and M. J. Bak, "Design, fabrication, and in vivo behavior of chronic recording intracortical microelectrodes," IEEE Trans Biomed Eng, vol. 20, pp. 253- 60, Jul 1973. [166] M. Salcman, "Letter: Trauma after electrode implantation," Archives of Neurology, vol. 33, p. 215, Mar 1976. [167] G. E. Loeb, R. A. Peck, and J. Martyniuk, "Toward the ultimate metal microelectrode," Journal of Neuroscience Methods, vol. 63, pp. 175-183, 1995. [168] A. Gilletti and J. Muthuswamy, "Brain micromotion around implants in the rodent somatosensory cortex," Journal of Neural Engineering, vol. 3, pp. 189-95, 2006. [169] J. Muthuswamy, A. Gilletti, T. Jain, and M. Okandan, "Microactuated neural probes to compensate for brain micromotion," in Engineering in Medicine and Biology Society, 2003. Proceedings of the 25th Annual International Conference of the IEEE, 2003, pp. 1941-1943 Vol.2.
224 [170] F. Casanova, P. R. Carney, and M. Sarntinoranont, "In vivo evaluation of needle force and friction stress during insertion at varying insertion speed into the brain," Journal of Neuroscience Methods, vol. 237, pp. 79-89, 2014. [171] J. L. Skousen, M. J. Bridge, and P. A. Tresco, "A strategy to passively reduce neuroinflammation surrounding devices implanted chronically in brain tissue by manipulating device surface permeability," Biomaterials, vol. 36, pp. 33-43, 2015. [172] D. H. Kim and D. C. Martin, "Sustained release of dexamethasone from hydrophilic matrices using PLGA nanoparticles for neural drug delivery," Biomaterials, vol. 27, pp. 3031-7, May 2006. [173] L. Rao, H. Zhou, T. Li, C. Li, and Y. Y. Duan, "Polyethylene glycol-containing polyurethane hydrogel coatings for improving the biocompatibility of neural electrodes," Acta Biomaterialia, vol. 8, pp. 2233-2242, 2012. [174] L. He, D. Lin, Y. Wang, Y. Xiao, and J. Che, "Electroactive SWNT/PEGDA hybrid hydrogel coating for bio-electrode interface," Colloids and surfaces. B, Biointerfaces, vol. 87, pp. 273-9, Oct 15 2011. [175] S. De Faveri, E. Maggiolini, E. Miele, et al., "Bio-inspired hybrid microelectrodes: a hybrid solution to improve long-term performance of chronic intracortical implants," Frontiers in Neuroengineering, vol. 7, p. 7, 2014. [176] G. Lind, C. E. Linsmeier, J. Thelin, and J. Schouenborg, "Gelatine-embedded electrodes—a novel biocompatible vehicle allowing implantation of highly flexible microelectrodes," Journal of Neural Engineering, vol. 7, p. 046005, 2010. [177] P. Köhler, A. Wolff, F. Ejserholm, et al., "Influence of probe flexibility and gelatin embedding on neuronal density and glial responses to brain implants," PLoS One, vol. 10, p. e0119340, 2015. [178] K.-K. Lee, J. He, A. Singh, et al., "Polyimide-based intracortical neural implant with improved structural stiffness," Journal of Micromechanics and Microengineering, vol. 14, pp. 32-37, 2004. [179] K. C. Cheung, P. Renaud, H. Tanila, and K. Djupsund, "Flexible polyimide microelectrode array for in vivo recordings and current source density analysis," Biosensors and Bioelectronics, vol. 22, pp. 1783-1790, 2007. [180] S. Takeuchi, T. Suzuki, K. Mabuchi, and H. Fujita, "3D flexible multichannel neural probe array," Journal of Micromechanics and Microengineering, vol. 14, pp. 104-107, 2004. [181] A. Mercanzini, K. Cheung, D. L. Buhl, et al., "Demonstration of cortical recording using novel flexible polymer neural probes," Sensors and Actuators A: Physical, vol. 143, pp. 90-96, 2008. [182] T. Suzuki, K. Mabuchi, and S. Takeuchi, "A 3D flexible parylene probe array for multichannel neural recording," in First International IEEE EMBS Conference on Neural Engineering, 2003, pp. 154-156. [183] D. S. Pellinen, T. Moon, R. J. Vetter, R. Miriani, and D. R. Kipke, "Multifunctional flexible parylene-based intracortical microelectrodes," in Engineering in Medicine and Biology Society, 2005. IEEE-EMBS 2005. 27th Annual International Conference of the, 2005, pp. 5272-5275. [184] E. G. R. Kim, J. K. John, H. Tu, et al., "A hybrid silicon–parylene neural probe with locally flexible regions," Sensors and Actuators B: Chemical, vol. 195, pp. 416-422, 2014.
225 [185] B. A. Wester, R. H. Lee, and M. C. LaPlaca, "Development and characterization of in vivo flexible electrodes compatible with large tissue displacements," J Neural Eng, vol. 6, p. 024002, Apr 2009. [186] A. Altuna, E. Bellistri, E. Cid, et al., "SU-8 based microprobes for simultaneous neural depth recording and drug delivery in the brain," Lab on a Chip, vol. 13, pp. 1422-1430, 2013. [187] A. Altuna, L. Menendez de la Prida, E. Bellistri, et al., "SU-8 based microprobes with integrated planar electrodes for enhanced neural depth recording," Biosensors and Bioelectronics, vol. 37, pp. 1-5, 2012. [188] V. N. Vernekar, D. K. Cullen, N. Fogleman, et al., "SU-8 2000 rendered cytocompatible for neuronal bioMEMS applications," Journal of Biomedical Materials Research Part A, 2008. [189] M. A. McClain, I. P. Clements, R. H. Shafer, et al., "Highly-compliant, microcable neuroelectrodes fabricated from thin-film gold and PDMS," Biomedical Microdevices, vol. 13, pp. 361-373, 2011. [190] T. D. Kozai and D. R. Kipke, "Insertion shuttle with carboxyl terminated self- assembled monolayer coatings for implanting flexible polymer neural probes in the brain," Journal of Neuroscience Methods, Aug 8 2009. [191] K. Lee, J. He, R. Clement, S. Massia, and B. Kim, "Biocompatible benzocyclobutene (BCB)-based neural implants with micro-fluidic channel," Biosensors and Bioelectronics, vol. 20, pp. 404-407, 2004. [192] D. H. Kim, M. Abidian, and D. C. Martin, "Conducting polymers grown in hydrogel scaffolds coated on neural prosthetic devices," Journal of Biomedical Materials Research Part A, vol. 71, pp. 577-85, 2004. [193] X. Cui, V. A. Lee, Y. Raphael, et al., "Surface modification of neural recording electrodes with conducting polymer/biomolecule blends," J Biomed Mater Res, vol. 56, pp. 261-272, 2001. [194] X. Cui, L. J. A. Wiler, M. Dzaman, R. A. Altschluer, and D. C. Martin, "In vivo studies of polypyrrole/peptide coated neural probes," Biomaterials, vol. 24, pp. 777-787, 2003. [195] N. Gomez and C. E. Schmidt, "Nerve growth factor-immobilized polypyrrole: Bioactive electrically conducting polymer for enhanced neurite extension," Journal of Biomedical Materials Research Part A, vol. 81A, pp. 135-149, 2007. [196] R. Wadhwa, C. F. Lagenaur, and X. T. Cui, "Electrochemically controlled release of dexamethasone from conducting polymer polypyrrole coated electrode," J Control Release, vol. 110, pp. 531-41, Feb 21 2006. [197] V. Finkenstadt, "Natural polysaccharides as electroactive polymers," Applied Microbiology and Biotechnology, vol. 67, pp. 735-745, 2005/06/01 2005. [198] J. H. Collier, J. P. Camp, T. W. Hudson, and C. E. Schmidt, "Synthesis and characterization of polypyrrole–hyaluronic acid composite biomaterials for tissue engineering applications," Journal of Biomedical Materials Research, vol. 50, pp. 574-584, 2000. [199] J. Y. Lee, C. A. Bashur, A. S. Goldstein, and C. E. Schmidt, "Polypyrrole-coated electrospun PLGA nanofibers for neural tissue applications," Biomaterials, vol. 30, pp. 4325-4335, 2009.
226 [200] X. Cui and D. C. Martin, "Electrochemical deposition and characterization of poly(3,4-ethylenedioxythiophene) on neural microelectrode arrays," Sensors and Actuators B, vol. 89, pp. 92-102, 2003. [201] K. A. Ludwig, J. D. Uram, J. Yang, D. C. Martin, and D. R. Kipke, "Chronic neural recordings using silicon microelectrode arrays electrochemically deposited with a poly(3,4-ethylenedioxythiophene) (PEDOT) film," Journal of Neural Engineering, vol. 3, pp. 59-70, 2006. [202] R. A. Green, N. H. Lovell, and L. A. Poole-Warren, "Cell attachment functionality of bioactive conducting polymers for neural interfaces," Biomaterials, vol. 30, pp. 3637-3644, 2009. [203] B. Zhu, S.-C. Luo, H. Zhao, et al., "Large enhancement in neurite outgrowth on a cell membrane-mimicking conducting polymer," Nat Commun, vol. 5, 2014. [204] J. R. Capadona, K. Shanmuganathan, S. Trittschuh, et al., "Polymer nanocomposites with nanowhiskers isolated from microcrystalline cellulose," Biomacromolecules, vol. 10, pp. 712-6, Apr 13 2009. [205] A. Hess, J. Dunning, J. Harris, et al., "A bio-inspired, chemo-responsive polymer nanocomposite for mechanically dynamic microsystems," presented at the Solid- State Sensors, Actuators and Microsystems Conference, 2009. TRANSDUCERS 2009. International, Denver, CO, 2009. [206] K. Shanmuganathan, J. R. Capadona, S. J. Rowan, and C. Weder, "Bio-inspired mechanically-adaptive nanocomposites derived from cotton cellulose whiskers," Journal of Materials Chemistry, vol. 20, p. 180, 2010. [207] K. Shanmuganathan, J. R. Capadona, S. J. Rowan, and C. Weder, "Biomimetic mechanically adaptive nanocomposites," Progress in Polymer Science, vol. 35, pp. 212-222, 2010. [208] A. Hess, J. Capadona, K. Shanmuganathan, et al., "Development of a stimuli- responsive polymer nanocomposite toward biologically-optimized, MEMS-based neural probes," Journal of Micromechanics and Microengineering, vol. 21, pp. 54009 -17, 2011. [209] A. E. Hess, K. Shanmuganathan, J. R. Capadona, et al., "Mechanical behavior of microstructures from a chemo-responsive polymer nanocomposite based on cotton cellulose nanofibers " presented at the Micro Electro Mechanical Systems (MEMS), 2011 IEEE 24th International Conference Cancun, MX, 2011. [210] J. R. Capadona, D. J. Tyler, C. A. Zorman, S. J. Rowan, and C. Weder, "Mechanically adaptive nanocomposites for neural interfacing," Materials Research Society Bulletin, vol. 37, pp. 581-589, 2012. [211] A. E. Hess, K. Potter, D. J. Tyler, C. A. Zorman, and J. R. Capadona, "Environmentally-controlled microtensile testing of mechanically-adaptive polymer nanocomposites for ex vivo characterization," J. Vis. Exp., vol. 78, p. e50078, 2013. [212] T. Ware, D. Simon, R. L. Rennaker, and W. Voit, "Smart polymers for neural interfaces," Polymer Reviews, vol. 53, pp. 108-129, 2013. [213] T. Ware, D. Simon, C. Liu, et al., "Thiol-ene/acrylate substrates for softening intracortical electrodes," Journal of Biomedical Materials Research Part B: Applied Biomaterials, p. 10.1002/jbmb.32946, 2013.
227 [214] T. Ware, D. Simon, K. Hearon, et al., "Three-dimensional flexible electronics enabled by shape memory polymer substrates for responsive neural interfaces," Macromolecular Materials and Engineering, vol. 297, pp. 1193-1202, 2012. [215] D. Y. Lewitus, J. Landers, J. Branch, et al., "Biohybrid carbon nanotube/agarose fibers for neural tissue engineering," Advanced Functional Materials, vol. 21, pp. 2624-2632, 2011. [216] S. M. Gutowski, K. L. Templeman, A. B. South, et al., "Host response to microgel coatings on neural electrodes implanted in the brain," Journal of Biomedical Materials Research Part A, vol. 102, pp. 1486-99, 2013. [217] C. Hassler, T. Boretius, and T. Stieglitz, "Polymers for neural implants," Journal of Polymer Science: Part B Polymer Physics, vol. 49, pp. 18-30, 2011. [218] R. A. Green, N. H. Lovell, G. G. Wallace, and L. A. Poole-Warren, "Conducting polymers for neural interfaces: Challenges in developing an effective long-term implant," Biomaterials, vol. 29, pp. 3393-3399, 2008. [219] X. Cui, J. F. Hetke, J. A. Wiler, D. J. Anderson, and D. C. Martin, "Electrochemical deposition and characterization of conducting polymer polypyrrole/PPS on multichannel nueral probes," Sensors and Actuators A, vol. 93, pp. 8-18, 2001. [220] T. D. Y. Kozai, Z. Gugel, X. Li, et al., "Chronic tissue response to carboxymethyl cellulose based dissolvable insertion needle for ultra-small neural probes," Biomaterials, vol. 35, pp. 9255-9268, 2014. [221] P. J. Gilgunn, R. Khilwani, T. D. Y. Kozai, et al., "An ultra-compliant, scalable neural probe with molded biodissolvable delivery vehicle," in IEEE 25th International Conference on Micro Electro Mechanical Systems (MEMS), 2012, pp. 56-59. [222] Y. Kato, I. Saito, T. Hoshino, T. Suzuki, and K. Mabuchi, "Preliminary study of multichannel flexible neural probes coated with hybrid biodegradable polymer," 2006 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Vols 1-15, pp. 5444-5447, 2006. [223] L. W. Tien, F. Wu, M. D. Tang-Schomer, et al., "Silk as a multifunctional biomaterial substrate for reduced glial scarring around brain-penetrating electrodes," Advanced Functional Materials, pp. n/a-n/a, 2013. [224] D. Y. Lewitus, K. L. Smith, W. Shain, and J. Kohn, "Ultrafast resorbing polymers for use as carriers for cortical neural probes," Acta Biomaterialia, vol. 7, pp. 2483- 91, Jun 2011. [225] D.-H. Kim, J. Viventi, J. J. Amsden, et al., "Dissolvable films of silk fibroin for ultrathin conformal bio-integrated electronics," Nat Mater, vol. advance online publication, 2010. [226] D. Lewitus, R. J. Vogelstein, G. H. Zhen, et al., "Designing tyrosine-derived polycarbonate polymers for biodegradable regenerative type neural interface capable of neural recording," IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 19, pp. 204-212, Apr 2011. [227] D. Y. Lewitus, K. L. Smith, W. Shain, D. Bolikal, and J. Kohn, "The fate of ultrafast degrading polymeric implants in the brain," Biomaterials, vol. 32, pp. 5543-5550, Aug 2011.
228 [228] S. H. Felix, K. G. Shah, V. M. Tolosa, et al., "Insertion of flexible neural probes using rigid stiffeners attached with biodissolvable adhesive," Journal of Visualized Experiments, p. e50609, 2013/09/27/ 2013. [229] I. Dryg, M. Ward, K. Qing, et al., "Magnetically inserted neural electrodes: tissue response and functional lifetime," IEEE Transactions on Neural Systems and Rehabilitation Engineering, pp. 1-1, 2015. [230] A. E. Hess-Dunning, D. J. Tyler, J. P. Harris, et al., "Microscale characterization of a mechanically adaptive polymer nanocomposite with cotton-derived cellulose nanocrystals for implantable bioMEMS," JMEMS, vol. 23, p. 774, 2014. [231] N. A. Kotov, J. O. Winter, I. P. Clements, et al., "Nanomaterials for neural interfaces," Advanced Materials, vol. 21, pp. 3970-4004, 2009. [232] W. Fan, I. Maesoon, and Y. Euisik, "A flexible fish-bone-shaped neural probe strengthened by biodegradable silk coating for enhanced biocompatibility," in Solid-State Sensors, Actuators and Microsystems Conference (TRANSDUCERS), 2011 16th International, 2011, pp. 966-969. [233] T. Kim, A. Branner, T. Gulati, and S. F. Giszter, "Braided multi-electrode probes: mechanical compliance characteristics and recordings from spinal cords," Journal of Neural Engineering, vol. 10, p. 045001, 2013. [234] H. S. Sohal, A. Jackson, R. Jackson, et al., "The sinusoidal probe: a new approach to improve electrode longevity," Frontiers in Neuroengineering, vol. 7, p. 10, 2014. [235] J. S. Brumberg, A. Nieto-Castanon, P. R. Kennedy, and F. H. Guenther, "Brain– computer interfaces for speech communication," Speech Communication, vol. 52, pp. 367-379, 2010. [236] P. R. Kennedy and R. A. Bakay, "Activity of single action potentials in monkey motor cortex during long-term task learning," Brain Research, vol. 760, pp. 251-4, Jun 20 1997. [237] J. Bartels, D. Andreasen, P. Ehirim, et al., "Neurotrophic electrode: method of assembly and implantation into human motor speech cortex," J Neurosci Methods, vol. 174, pp. 168-76, Sep 30 2008. [238] F. H. Guenther, J. S. Brumberg, E. J. Wright, et al., "A wireless brain-machine interface for real-time speech synthesis," PLoS One, vol. 4, p. e8218, 2009. [239] P. R. Kennedy, M. T. Kirby, M. M. Moore, B. King, and A. Mallory, "Computer control using human intracortical local field potentials," IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 12, pp. 339-344, Sep 2004. [240] P. R. Kennedy, R. A. Bakay, K. Adams, and J. Goldwaithe, "Direct control of a computer from the human central nervous system," IEEE Trans Rehabil Eng, vol. 8, pp. 198-202, 2000. [241] P. Mohseni, K. Najafi, S. J. Eliades, and W. Xiaoqin, "Wireless multichannel biopotential recording using an integrated FM telemetry circuit," Neural Systems and Rehabilitation Engineering, IEEE Transactions on [see also IEEE Trans. on Rehabilitation Engineering], vol. 13, pp. 263-271, 2005. [242] S. Farshchi, P. H. Nuyujukian, A. Pesterev, I. Mody, and J. W. Judy, "A TinyOS- enabled MICA2-based wireless neural interface," IEEE Trans Biomed Eng, vol. 53, pp. 1416-24, Jul 2006.
229 [243] A. M. Sodagar, G. E. Perlin, Y. Ying, K. Najafi, and K. D. Wise, "An implantable 64-channel wireless microsystem for single-unit neural recording," Solid-State Circuits, IEEE Journal of, vol. 44, pp. 2591-2604, 2009. [244] R. R. Harrison, P. T. Watkins, R. J. Kier, et al., "A low-power integrated circuit for a wireless 100-electrode neural recording system," Solid-State Circuits, IEEE Journal of, vol. 42, pp. 123-133, 2007. [245] R. R. Harrison, R. J. Kier, C. A. Chestek, et al., "Wireless neural recording with single low-power integrated circuit," IEEE Trans Neural Syst Rehabil Eng, vol. 17, pp. 322-9, Aug 2009. [246] Y. K. Song, D. A. Borton, S. Park, et al., "Active microelectronic neurosensor arrays for implantable Brain Communication Interfaces," IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 17, pp. 339-345, 2009. [247] K. A. Potter-Baker and J. R. Capadona, "Reducing the “Stress”: Antioxidative therapeutic and material approaches may prevent intracortical microelectrode failure," ACS Macro Letters, pp. 275-279, 2015. [248] N. Khansari, Y. Shakiba, and M. Mahmoudi, "Chronic inflammation and oxidative stress as a major cause of age-related diseases and cancer," Recent Patents on Inflammation & Allergy Drug Discovery, vol. 3, pp. 73-80, Jan 2009. [249] H. U. Simon, A. Haj-Yehia, and F. Levi-Schaffer, "Role of reactive oxygen species (ROS) in apoptosis induction," Apoptosis, vol. 5, pp. 415-8, Nov 2000. [250] M. Valko, D. Leibfritz, J. Moncol, et al., "Free radicals and antioxidants in normal physiological functions and human disease," The International Journal of Biochemistry & Cell Biology, vol. 39, pp. 44-84, 2007. [251] J.-W. S. G Schmitt, F Faßbender, G Buß, H Lüth, M.J Schöning, "Passivation and corrosion of microelectrode arrays," Electrochimica Acta, vol. 44, pp. 3865-3883, 1999. [252] G. E. Loeb, M. J. Bak, M. Salcman, and E. M. Schmidt, "Parylene as a chronically stable, reproducible microelectrode insulator," IEEE Trans Biomed Eng, vol. 24, pp. 121-8, Mar 1977. [253] E. M. Schmidt, J. S. McIntosh, and M. J. Bak, "Long-term implants of Parylene-C coated microelectrodes," Med Biol Eng Comput, vol. 26, pp. 96-101, Jan 1988. [254] S. A. Ali, S. P. Zhong, P. J. Doherty, and D. F. Williams, "Mechanisms of polymer degradation in implantable devices. I. Poly(caprolactone)," Biomaterials, vol. 14, pp. 648-56, Jul 1993. [255] Q. Zhao, J. Casas-Bejar, P. Urbanski, and K. Stokes, "Glass wool-H2O2/CoCl2 test system for in vitro evaluation of biodegradative stress cracking in polyurethane elastomers," J Biomed Mater Res, vol. 29, pp. 467-75, Apr 1995. [256] D. W. L. Hukins, A. Mahomed, and S. N. Kukureka, "Accelerated aging for testing polymeric biomaterials and medical devices," Medical Engineering & Physics, vol. 30, pp. 1270-1274, 2008. [257] W. He and R. V. Bellamkonda, "A molecular perspective on understanding and modulating the performance of chronic central nervous system (CNS) recording electrodes," in Indwelling Neural Implants: Strategies for Contending with the In Vivo Environment W. M. Reichert, Ed., ed Boca Raton (FL): CRC Press, 2008.
230 [258] M. Ravikumar, D. J. Hageman, W. H. Tomaszewski, et al., "The effect of residual endotoxin contamination on the neuroinflammatory response to sterilized intracortical microelectrodes," J Mater Chem B, vol. 2, pp. 2517-2529, 2014. [259] L. Zhao, J. Y. Lee, and D. H. Hwang, "Inhibition of pattern recognition receptor- mediated inflammation by bioactive phytochemicals," Nutrition Reviews, vol. 69, pp. 310-20, 2011. [260] R. J. Parro, M. C. Scardelletti, N. C. Varaljay, S. Zimmerman, and C. A. Zorman, "Amorphous SiC as a structural layer in microbridge-based RF MEMS switches for use in software-defined radio," Solid-State Electronics, vol. 52, pp. 1647-1651, 2008. [261] K. A. Potter, J. S. Simon, B. Velagapudi, and J. R. Capadona, "Reduction of autofluorescence at the microelectrode-cortical tissue interface improves antibody detection," J Neurosci Methods, vol. 203, pp. 96-105, 2012. [262] K. A. Potter-Baker, M. Ravikumar, A. A. Burke, et al., "A comparison of neuroinflammation to implanted microelectrodes in rat and mouse models," Biomaterials, vol. 34, pp. 5637-5646, 2014. [263] J. N. Turner, W. Shain, D. H. Szarowski, et al., "Cerebral astrocyte response to micromachined silicon implants," Experimental Neurology, vol. 156, pp. 33-49, 1999. [264] G. Lind, L. Gällentoft, N. Danielsen, J. Schouenborg, and L. M. Pettersson, "Multiple implants do not aggravate the tissue reaction in rat brain," PLoS ONE, vol. 7, p. e47509, 2012. [265] R. J. Mullen, C. R. Buck, and A. M. Smith, "NeuN, a neuronal specific nuclear protein in vertebrates," Development, vol. 116, pp. 201-211, 1992. [266] M. Eddleston and L. Mucke, "Molecular profile of reactive astrocytes- implications for their role in neurologic disease," Neuroscience, vol. 54, pp. 15-36, 1993. [267] D. Ito, Y. Imai, K. Ohsawa, et al., "Microglia-specific localisation of a novel calcium binding protein, Iba1," Molecular Brain Research, vol. 57, pp. 1-9, 1998. [268] C. Dijkstra, E. Dopp, P. Joling, and G. Kraal, "The heterogeneity of mononuclear phagocytes in lymphoid organs: distinct macrophage subpopulations in the rat recognized by monoclonal antibodies ED1, ED2 and ED3," Immunology, vol. 54, pp. 589-599, 1985. [269] H. E. d. Vries, M. C. M. Blom-Roosemalen, M. v. Oosten, et al., "The influence of cytokines on the integrity of the blood-brain barrier in vitro," Journal of Neuroimmunology, vol. 64, pp. 37-43, 1996. [270] B. T. Hawkins and R. D. Egleton, "Fluorescence imaging of blood–brain barrier disruption," Journal of Neuroscience Methods, vol. 151, pp. 262-267, 2005. [271] G. C. McConnell, T. M. Schneider, and R. V. Bellamkonda, "Acute spatiotemporal changes in neuronal density surrounding microelectrode arrays implanted in rat motor cortex," in 3rd International IEEE/EMBS Conference on Neural Engineering, Kohala Coast, HI, 2007, pp. 137-140. [272] G. W. Kreutzberg, "Microglia: a sensor for pathological events in the CNS," Trends in Neurosciences, vol. 19, pp. 312-8, 1996. [273] D. S. Arroyo, J. A. Soria, E. A. Gaviglio, M. C. Rodriguez-Galan, and P. Iribarren, "Toll-like receptors are key players in neurodegeneration," International Immunopharmacology, vol. 11, pp. 1415-21, 2011.
231 [274] J. M. Anderson, "Biological responses to materials," Annual Review of Materials Research, vol. 31, pp. 81-110, 2001. [275] J. M. Anderson, A. Rodriguez, and D. T. Chang, "Foreign body reaction to biomaterials," Seminars in Immunology, vol. 20, pp. 86-100, 2008. [276] A. L. Pollock, Early brain tissue reaction surrounding the Michigan-style microelectrode arrays indicates that extensive remodeling of glial and neuronal populations occurs during the foreign body response. Salt Lake City: Department of Bioengineering, University of Utah, 2007. [277] C. C. Chao, S. Hu, and P. K. Peterson, "Glia, cytokines, and neurotoxicity," Critical reviews in neurobiology, vol. 9, pp. 189-205, 1995. [278] A. Mantovani, S. Sozzani, M. Locati, P. Allavena, and A. Sica, "Macrophage polarization: tumor-associated macrophages as a paradigm for polarized M2 mononuclear phagocytes," Trends Immunol, vol. 23, pp. 549-55, 2002. [279] V. J. Quagliarello, B. Wispelwey, W. J. Long, Jr., and W. M. Scheld, "Recombinant human interleukin-1 induces meningitis and blood-brain barrier injury in the rat. Characterization and comparison with tumor necrosis factor," J Clin Invest, vol. 87, pp. 1360-6, 1991. [280] M. L. Block, L. Zecca, and J.-S. Hong, "Microglia-mediated neurotoxicity: uncovering the molecular mechanisms," Nature Reviews Neuroscience, vol. 8, pp. 57-69, 2007. [281] M. T. T. Wong-Riley, "Cytochrome oxidase: an endogenous metabolic marker for neuronal activity," Trends in Neurosciences, vol. 12, pp. 94-101, 1989. [282] N. Aihara, H. Tanno, J. J. Hall, L. H. Pitts, and L. J. Noble, "Immunocytochemical localization of immunoglobulins in the rat brain: relationship to the blood-brain barrier," Journal of Comparative Neurology, vol. 342, pp. 481-496, 1994. [283] J. L. Skousen, "Novel design strategies to reduce the foreign body response to central nervous system implants," PhD PhD, Department of Biomedical Engineering, University of Utah, Salt Lake City, 2013. [284] D. K. Cullen, C. M. Simon, and M. C. LaPlaca, "Strain rate-dependent induction of reactive astrogliosis and cell death in three-dimensional neuronal-astrocytic co- cultures," Brain Research, vol. 1158, pp. 103-15, Jul 16 2007. [285] A. B. Schwartz, "Cortical neural prosthetics," Annual Review of Neuroscience, vol. 27, pp. 487-507, 2004. [286] S. F. Cogan, "Neural stimulation and recording electrodes," Annual Review of Biomedical Engineering, vol. 10, pp. 275-309, 2008. [287] M. L. Anderson, "Neural reuse: A fundamental organizational principle of the brain," Behavioral and Brain Sciences, vol. 33, pp. 245-266, 2010. [288] T. R. Oakes, D. A. Pizzagalli, A. M. Hendrick, et al., "Functional coupling of simultaneous electrical and metabolic activity in the human brain," Human Brain Mapping, vol. 21, pp. 257-70, Apr 2004. [289] H. Duffau, "Brain mapping in tumors: Intraoperative or extraoperative?," Epilepsia, vol. 54, pp. 79-83, 2013. [290] K. S. Carlson, C. Z. Xia, and D. W. Wesson, "Encoding and representation of intranasal CO2 in the mouse olfactory cortex," J Neurosci, vol. 33, pp. 13873- 13881, 2013.
232 [291] B. K. Leung, R. Biran, C. J. Underwood, and P. A. Tresco, "Characterization of microglial attachment and cytokine release on biomaterials of differing surface chemistry," Biomaterials, vol. 29, pp. 3289-97, Aug 2008. [292] J. U. Meyer, D. J. Edell, W. L. C. Rutten, and J. Pine, "Chronically implantable neural information transducers," presented at the 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Amsterdam, 1996. [293] Y. T. Kim, R. W. Hitchcock, M. J. Bridge, and P. A. Tresco, "Chronic response of adult rat brain tissue to implants anchored to the skull," Biomaterials, vol. 25, pp. 2229-37, May 2004. [294] L. Kee-Keun and et al., "Polyimide-based intracortical neural implant with improved structural stiffness," Journal of Micromechanics and Microengineering, vol. 14, p. 32, 2004. [295] W. He, G. C. McConnell, T. M. Schneider, and R. V. Bellamkonda, "A novel anti- inflammatory surface for neural electrodes," Advanced Materials, vol. 19, pp. 3529-3533, 2007. [296] A. A. Sharp, H. V. Panchawagh, A. Ortega, et al., "Toward a self-deploying shape memory polymer neuronal electrode," J Neural Eng, vol. 3, pp. L23-30, Dec 2006. [297] R. Langer and D. A. Tirrell, "Designing materials for biology and medicine," Nature, vol. 428, pp. 487-493, 2004. [298] R. Das, D. Gandhi, S. Krishnan, L. Saggere, and P. J. Rousche, "A benchtop system to assess cortical neural interface micromechanics," IEEE Transactions on Biomedical Engineering, vol. 54, pp. 1089-96, Jun 2007. [299] A. Gefen, N. Gefen, Q. Zhu, R. Raghupathi, and S. S. Margulies, "Age-dependent changes in material properties of the brain and braincase of the rat," Journal of Neurotrauma, vol. 20, pp. 1163-77, Nov 2003. [300] W. C. Oliver and G. M. Pharr, "An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments," Journal of Materials Research, vol. 7, pp. 1564-1583, 1992. [301] L. Qian, M. Li, Z. Zhou, H. Yang, and X. Shi, "Comparison of nano-indentation hardness to microhardness," Surface and Coatings Technology, vol. 195, pp. 264- 271, 2005. [302] S. J. Bensmaia and L. E. Miller, "Restoring sensorimotor function through intracortical interfaces: progress and looming challenges," Nat Rev Neurosci, vol. 15, pp. 313-25, May 2014. [303] Y. B. Lu, K. Franze, G. Seifert, et al., "Viscoelastic properties of individual glial cells and neurons in the CNS," Proc Natl Acad Sci U S A, vol. 103, pp. 17759-64, Nov 21 2006. [304] M. Polanco, H. Yoon, and S. Bawab, "Micromotion-induced dynamic effects from a neural probe and brain tissue interface," Journal of Micro/Nanolithography, MEMS, and MOEMS, vol. 13, pp. 023009-023009, 2014. [305] Y. Kucuk, "Modeling nonlinear viscoelastic nanoindentation of PVAc at different unloading rates," Journal of Mechanical Engineering, vol. 59, 2013. [306] M. C. LaPlaca, D. K. Cullen, J. J. McLoughlin, and R. S. Cargill, 2nd, "High rate shear strain of three-dimensional neural cell cultures: a new in vitro traumatic brain injury model," J Biomech, vol. 38, pp. 1093-105, May 2005.
233 [307] H. Ahmadzadeh, D. H. Smith, and V. B. Shenoy, "Viscoelasticity of tau proteins leads to strain rate-dependent breaking of microtubules during axonal stretch injury: predictions from a mathematical model," Biophysical Journal, vol. 106, pp. 1123-33, Mar 4 2014. [308] M. D. Budde, L. Janes, E. Gold, L. C. Turtzo, and J. A. Frank, "The contribution of gliosis to diffusion tensor anisotropy and tractography following traumatic brain injury: validation in the rat using Fourier analysis of stained tissue sections," Brain, vol. 134, pp. 2248-60, Aug 2011. [309] M. L. Previtera, C. G. Langhammer, N. A. Langrana, and B. L. Firestein, "Regulation of dendrite arborization by substrate stiffness is mediated by glutamate receptors," Annals of Biomedical Engineering, vol. 38, pp. 3733-43, Dec 2010. [310] P. C. Georges, W. J. Miller, D. F. Meaney, E. S. Sawyer, and P. A. Janmey, "Matrices with compliance comparable to that of brain tissue select neuronal over glial growth in mixed cortical cultures," Biophysical Journal, vol. 90, pp. 3012-8, Apr 15 2006. [311] P. Paoletti and P. Ascher, "Mechanosensitivity of NMDA receptors in cultured mouse central neurons," Neuron, vol. 13, pp. 645-55, Sep 1994. [312] W. J. Tyler, "The mechanobiology of brain function," Nat Rev Neurosci, vol. 13, pp. 867-78, Dec 2012. [313] D. Bacher, B. Jarosiewicz, N. Y. Masse, et al., "Neural point-and-click communication by a person with incomplete locked-in syndrome," Neurorehabilitation & Neural Repair, Nov 10 2014. [314] D.-H. Kim, S. M. Richardson-Burns, J. L. Hendricks, C. Sequera, and D. C. Martin, "Effect of immobilized nerve growth factor on conductive polymers: electrical properties and cellular response," Advanced Functional Materials, vol. 17, pp. 79- 86, 2007. [315] M. Jorfi, M. N. Roberts, E. J. Foster, and C. Weder, "Physiologically responsive, mechanically adaptive bio-nanocomposites for biomedical applications," ACS Applied Materials & Interfaces, vol. 5, pp. 1517-1526, 2013. [316] V. Favier, H. Chanzy, and J. Y. Cavaille, "Polymer nanocomposites reinforced by cellulose whiskers," Macromolecules, vol. 28, pp. 6365-6367, 1995. [317] W. M. Saltzman and M. L. Radomsky, "Drugs released from polymers: diffusion and elimination in brain tissue," Chemical Engineering Science, vol. 46, pp. 2429- 2444, 1991. [318] M. Zanetti, L. V. d’Uscio, T. E. Peterson, Z. S. Katusic, and T. O’Brien, "Analysis of superoxide anion production in tissue," in Methods in Molecular Medicine vol. 108, J. P. F. a. A. H. Baker, Ed., ed. Totowa, NJ: Humana Press Inc., 2005, pp. 65- 72. [319] J. S. Jensen, C. F. Wertz, and V. A. O'Neill, "Preformulation stability of trans- resveratrol and trans-resveratrol glucoside (Piceid)," J Agric Food Chem, vol. 58, pp. 1685-90, Feb 10 2010. [320] S. K. Mallapragada and N. A. Peppas, "Crystal dissolution-controlled release systems: I. Physical characteristics and modeling analysis," Journal of Controlled Release, vol. 45, pp. 87-94, 1997.
234 [321] R. Pangeni, J. K. Sahni, J. Ali, S. Sharma, and S. Baboota, "Resveratrol: review on therapeutic potential and recent advances in drug delivery," Expert Opinion on Drug Delivery, vol. 11, pp. 1285-98, Aug 2014. [322] P. Anand, A. B. Kunnumakkara, R. A. Newman, and B. B. Aggarwal, "Bioavailability of curcumin: problems and promises," Molecular Pharmaceutics, vol. 4, pp. 807-818, 2007/12/01 2007. [323] M. Ravikumar, S. Sunil, J. Black, et al., "The roles of blood-derived macrophages and resident microglia in the neuroinflammatory response to implanted intracortical microelectrodes," Biomaterials, vol. S0142-9612, pp. 8049-8064, 2014. [324] M. Bortole, M. Controzzi, I. Pisotta, and A. Úbeda, "BMIs for motor rehabilitation: key concepts and challenges," in Emerging Therapies in Neurorehabilitation. vol. 4, J. L. Pons and D. Torricelli, Eds., ed: Springer Berlin Heidelberg, 2014, pp. 235- 247. [325] D. M. Taylor, S. I. Tillery, and A. B. Schwartz, "Direct cortical control of 3D neuroprosthetic devices," Science, vol. 296, pp. 1829-32, Jun 7 2002. [326] T. Walle, "Bioavailability of resveratrol," Annals of the New York Academy of Sciences, vol. 1215, pp. 9-15, 2011. [327] M. Asensi, I. Medina, A. Ortega, et al., "Inhibition of cancer growth by resveratrol is related to its low bioavailability," Free Radical Biology and Medicine, vol. 33, pp. 387-398, 2002. [328] C. Y. Cheung, S. J. McCartney, and K. S. Anseth, "Synthesis of polymerizable superoxide dismutase mimetics to reduce reactive oxygen species damage in transplanted biomedical devices," Advanced Functional Materials, vol. 18, pp. 3119-3126, 2008. [329] J. R. Capadona, T. A. Petrie, K. P. Fears, et al., "Surface-nucleated assembly of fibrillar extracellular matrices," Advanced Materials, vol. 17, pp. 2604-2608, 2005. [330] M. V. Voinova, M. Rodahl, M. Jonson, and B. Kasemo, "Viscoelastic acoustic response of layered polymer films at fluid-solid interfaces: continuum mechanics approach," Physica Scripta, vol. 59, p. 391, 1999. [331] M. V. Berridge, P. M. Herst, and A. S. Tan, "Tetrazolium dyes as tools in cell biology: New insights into their cellular reduction," in Biotechnology Annual Review. vol. Volume 11, M. R. El-Gewely, Ed., ed: Elsevier, 2005, pp. 127-152. [332] P. J. Janknegt, J. W. Rijstenbil, W. H. van de Poll, T. S. Gechev, and A. G. J. Buma, "A comparison of quantitative and qualitative superoxide dismutase assays for application to low temperature microalgae," Journal of Photochemistry and Photobiology B: Biology, vol. 87, pp. 218-226, 2007. [333] I. Batinić-Haberle, J. S. Rebouças, and I. Spasojević, "Superoxide dismutase mimics: chemistry, pharmacology, and therapeutic potential," Antioxidants & Redox Signaling, vol. 13, pp. 877-918, 2010/09/15 2010. [334] C. Li and H.-M. Zhou, "The role of manganese superoxide dismutase in inflammation defense," Enzyme research, vol. 2011, p. 387176, 2011 2011. [335] B. J. Day, S. Shawen, S. I. Liochev, and J. D. Crapo, "A metalloporphyrin superoxide dismutase mimetic protects against paraquat-induced endothelial cell injury, in vitro," Journal of Pharmacology and Experimental Therapeutics, vol. 275, pp. 1227-1232, December 1, 1995 1995.
235 [336] J. S. L. Mok, K. Paisley, and W. Martin, "Inhibition of nitrergic neurotransmission in the bovine retractor penis muscle by an oxidant stress: effects of superoxide dismutase mimetics," British Journal of Pharmacology, vol. 124, pp. 111-118, 1998. [337] D. Salvemini, C. Muscoli, D. P. Riley, and S. Cuzzocrea, "Superoxide dismutase mimetics," Pulmonary Pharmacology and Therapeutics, vol. 15, pp. 439-447, 2002. [338] P. J. F. Gauuan, M. P. Trova, L. Gregor-Boros, et al., "Superoxide dismutase mimetics: synthesis and structure–activity relationship study of MnTBAP analogues," Bioorganic and Medicinal Chemistry, vol. 10, pp. 3013-3021, 2002. [339] L.-S. Jang and H.-J. Liu, "Fabrication of protein chips based on 3- aminopropyltriethoxysilane as a monolayer," Biomedical Microdevices, vol. 11, pp. 331-338, 2009/04/01 2009. [340] Y. Han, D. Mayer, A. Offenhäusser, and S. Ingebrandt, "Surface activation of thin silicon oxides by wet cleaning and silanization," Thin Solid Films, vol. 510, pp. 175-180, 2006. [341] J. E. Raynor, J. R. Capadona, D. M. Collard, T. A. Petrie, and A. J. Garcia, "Polymer brushes and self-assembled monolayers: Versatile platforms to control cell adhesion to biomaterials (Review)," Biointerphases, vol. 4, pp. FA3-16, Jun 2009. [342] B. G. Keselowsky, D. M. Collard, and A. J. Garcia, "Surface chemistry modulates fibronectin conformation and directs integrin binding and specificity to control cell adhesion," Journal of Biomedical Materials Research, vol. 66A, pp. 247-259, 2003. [343] B. Turan, E. Tuncay, and G. Vassort, "Resveratrol and diabetic cardiac function: focus on recent in vitro and in vivo studies," Journal of Bioenergetics and Biomembranes, vol. 44, pp. 281-296, 2012/04/01 2012. [344] N. Mokarram, A. Merchant, V. Mukhatyar, G. Patel, and R. V. Bellamkonda, "Effect of modulating macrophage phenotype on peripheral nerve repair," Biomaterials, vol. 33, pp. 8793-8801, 2012. [345] N. Mokarram and R. V. Bellamkonda, "Overcoming endogenous constraints on neuronal regeneration," IEEE Trans Biomed Eng. , vol. 58, pp. 1900-1906, 2011. [346] K. A. Kigerl, J. C. Gensel, D. P. Ankeny, et al., "Identification of two distinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord," Journal of Neuroscience, vol. 29, pp. 13435-44, Oct 28 2009. [347] V. Shukla, S. K. Mishra, and H. C. Pant, "Oxidative stress in neurodegeneration," Advances in Pharmacological Sciences, vol. 2011, p. 572634, 2011. [348] A. A. Starkov, "Measurement of mitochondrial ROS production," Methods in molecular biology, vol. 648, pp. 245-55, 2010. [349] K. Apel and H. Hirt, "Reactive oxygen species: metabolism, oxidative stress, and signal transduction," Annual Review of Plant Biology, vol. 55, pp. 373-99, 2004. [350] B. D'Autreaux and M. B. Toledano, "ROS as signalling molecules: mechanisms that generate specificity in ROS homeostasis," Nature reviews. Molecular cell biology, vol. 8, pp. 813-24, Oct 2007. [351] J. Garthwaite and C. Boulton, "Nitric oxide signaling in the central nervous system," Annual Review of Physiology, vol. 57, pp. 683-706, 1995.
236 [352] N. Akgören, M. Fabricius, and M. Lauritzen, "Importance of nitric oxide for local increases of blood flow in rat cerebellar cortex during electrical stimulation," Proceedings of the National Academy of Sciences, vol. 91, pp. 5903-5907, 1994. [353] R. P. Patel, J. McAndrew, H. Sellak, et al., "Biological aspects of reactive nitrogen species," Biochimica et Biophysica Acta (BBA) - Bioenergetics, vol. 1411, pp. 385- 400, 1999. [354] D. A. Drechsel, A. G. Estévez, L. Barbeito, and J. S. Beckman, "Nitric oxide- mediated oxidative damage and the progressive demise of motor neurons in ALS," Neurotoxicity research, vol. 22, pp. 251-264, 2012. [355] T. Mizuno, J. Kawanokuchi, K. Numata, and A. Suzumura, "Production and neuroprotective functions of fractalkine in the central nervous system," Brain Research, vol. 979, pp. 65-70, 2003. [356] X. Lu, L. Ma, L. Ruan, et al., "Resveratrol differentially modulates inflammatory responses of microglia and astrocytes," J Neuroinflammation, vol. 7, p. 46, 2010. [357] F. Zhang, J. Liu, and J.-S. Shi, "Anti-inflammatory activities of resveratrol in the brain: Role of resveratrol in microglial activation," European Journal of Pharmacology, vol. 636, pp. 1-7, 2010. [358] H.-B. Wang, M. Dembo, S. K. Hanks, and Y.-l. Wang, "Focal adhesion kinase is involved in mechanosensing during fibroblast migration," Proceedings of the National Academy of Sciences of the United States of America, vol. 98, pp. 11295- 11300, 2001. [359] S. Wong, W.-H. Guo, and Y.-L. Wang, "Fibroblasts probe substrate rigidity with filopodia extensions before occupying an area," Proceedings of the National Academy of Sciences, vol. 111, pp. 17176-17181, December 2, 2014 2014. [360] Y. Xie, N. Martini, C. Hassler, et al., "In vivo monitoring of glial scar proliferation on chronically implanted neural electrodes by fiber optical coherence tomography," Frontiers in Neuroengineering, vol. 7, p. 34, 2014. [361] T. A. Evans, D. S. Barkauskas, J. T. Myers, et al., "High-resolution intravital imaging reveals that blood-derived macrophages but not resident microglia facilitate secondary axonal dieback in traumatic spinal cord injury," Experimental Neurology, vol. 254C, pp. 109-120, Jan 24 2014. [362] T.-W. Chen, T. J. Wardill, Y. Sun, et al., "Ultrasensitive fluorescent proteins for imaging neuronal activity," Nature, vol. 499, pp. 295-300, 2013. [363] K. Potter-Baker, W. Stewart, W. Tomaszewski, et al., "Implications of chronic daily anti-oxidant administration on the inflammatory response to intracortical microelectrodes," Journal of Neural Engineering, vol. In Press, 2015. [364] H. Zhuang, Y.-S. Kim, R. C. Koehler, and S. DorÉ, "Potential mechanism by which resveratrol, a red wine constituent, protects neurons," Annals of the New York Academy of Sciences, vol. 993, pp. 276-286, 2003. [365] T. S. Lee and L. Y. Chau, "Heme oxygenase-1 mediates the anti-inflammatory effect of interleukin-10 in mice," Nat Med, vol. 8, pp. 240-6, Mar 2002. [366] A. W. Filho, V. C. Filho, L. Olinger, and M. M. Souza, "Quercetin: Further investigation of its antinociceptive properties and mechanisms of action," Archives of Pharmacal Research, vol. 31, pp. 713-721, 2008. [367] B. Ossola, T. M. Kääriäinen, and P. T. Männistö, "The multiple faces of quercetin in neuroprotection," Expert Opinion on Drug Safety, vol. 8, pp. 397-409, 2009.
237 [368] G. Bureau, F. Longpré, and M. G. Martinoli, "Resveratrol and quercetin, two natural polyphenols, reduce apoptotic neuronal cell death induced by neuroinflammation," Journal of Neuroscience Research, vol. 86, pp. 403-410, 2008. [369] M. Asanuma, I. Miyazaki, and N. Ogawa, "Neuroprotective effects of nonsteroidal anti-inflammatory drugs on neurodegenerative diseases," Current Pharmaceutical Design, vol. 10, pp. 695-700, 2004. [370] B. P. Imbimbo, V. Solfrizzi, and F. Panza, "Are NSAIDs useful to treat Alzheimer's Disease or mild cognitive impairment?," Frontiers in Aging Neuroscience, vol. 2, p. 19, 2010. [371] Q. Cao, Y. Wang, D. Zheng, et al., "IL-10/TGF-beta-modified macrophages induce regulatory T cells and protect against adriamycin nephrosis," J Am Soc Nephrol, vol. 21, pp. 933-42, Jun 2010. [372] F. Al-Mulla, S. J. Leibovich, I. M. Francis, and M. S. Bitar, "Impaired TGF-beta signaling and a defect in resolution of inflammation contribute to delayed wound healing in a female rat model of type 2 diabetes," Mol Biosyst, vol. 7, pp. 3006-20, Nov 2011. [373] R. Parsa, P. Andresen, A. Gillett, et al., "Adoptive transfer of immunomodulatory M2 macrophages prevents type 1 diabetes in NOD mice," Diabetes, vol. 61, pp. 2881-92, Nov 2012. [374] C. Bogdan and C. Nathan, "Modulation of macrophage function by transforming growth factor beta, interleukin-4, and interleukin-10," Annals of the New York Academy of Sciences, vol. 685, pp. 713-39, Jun 23 1993.
238