www.advmat.de www.MaterialsViews.com
REVIEW
Graphite Oxide to Graphene. Biomaterials to Bionics Brianna C. Thompson,* Eoin Murray, and Gordon G. Wallace* synthetic scaffolds to make tissue engineered structures is shown in Figure 1. The engineering of different kinds of tissues has specific challenges and requirements, from the form factor of the implant (e.g., blocks, tubes, membranes, sponges etc.) and the physical properties of the materials (e.g., stiffness, softness, porosity, degradability) to the desired biological outcomes (e.g., promoting cell proliferation, differentiation, cell invasion, cell repulsion). All tissue engineering targets require different materials and fabrication strategies, and so the fields of tissue engineering and biomaterials development require a diverse array of expertise. The form factors of implants can vary tremendously between applications – a variety of which are illustrated in Figure 2. Some applications require the replacement of large blocks of tissue (such as muscle or fat), and in these applications, it is often necessary to devise a strategy to allow for transport of nutrients and waste into/out of the implant. Other applications need a membrane (to replace skin in wound healing or to replace internal membranes), or fibrous structures (in the engineering of nerve or muscle fibers), or something much more complicated and specific (in the engineering the complicated structure of an artificial kidney or heart). As fabrication technologies become more advanced, more tools to enable the production of these more complicated structures such as 3D printing, electrospinning, wet-spinning, extrusion, or lithography become available. Each fabrication strategy imparts its own requirements on the material used, so the ultimate material properties must be balanced between the requirements for the tissue development needed, processability, and being amenable to fabrication protocols that enable the ideal form factor to be realized. However, despite the need for tailoring material and implant properties for different parts of the body, one property is required in all tissue engineering – biocompatibility. The concept of biocompatibility is often poorly defined, but is generally taken to mean that the material or implant lacks toxicity and immunogenicity and will have no adverse effect on the biological system into which it is being introduced. Cells from the patient are used to seed cell scaffolds in order to reduce the potential for an immune¡ reaction against transplants of tissues or stem cells from donors, or xenotransplantation, reducing the risk of implant/organ rejection. The aim is to allow a “normal” tissue to develop that has the same genetic and epigenetic cues as the other tissues in the patient.
The advent of implantable biomaterials has revolutionized medical treatment, allowing the development of the fields of tissue engineering and medical bionic devices (e.g., cochlea implants to restore hearing, vagus nerve stimulators to control Parkinson’s disease, and cardiac pace makers). Similarly, future materials developments are likely to continue to drive development in treatment of disease and disability, or even enhancing human potential. The material requirements for implantable devices are stringent. In all cases they must be nontoxic and provide appropriate mechanical integrity for the application at hand. In the case of scaffolds for tissue regeneration, biodegradability in an appropriate time frame may be required, and for medical bionics electronic conductivity is essential. The emergence of graphene and graphene-family composites has resulted in materials and structures highly relevant to the expansion of the biomaterials inventory available for implantable medical devices. The rich chemistries available are able to ensure properties uncovered in the nanodomain are conveyed into the world of macroscopic devices. Here, the inherent properties of graphene, along with how graphene or structures containing it interface with living cells and the effect of electrical stimulation on nerves and cells, are reviewed.
1. Tissue Engineering The goal of a tissue engineer is to fix or replace tissues or organs that have been damaged or destroyed by accident or disease. In order to achieve this, one approach is to combine a fabricated scaffold composed of one or more materials having the desired properties with cells (either sourced and differentiated externally or internally to be populated with endogenous cells in situ), chemical cues (drugs, growth factors or other endogenous signals), and physical cues to encourage cells to develop into functional tissues. The general process of combining cells and
Dr. B. C. Thompson School of Mechanical and Aerospace Engineering Nanyang Technological University 639798, Singapore E-mail: [email protected] Dr. E. Murray Institute for Sports Research Nanyang Technological University 639798, Singapore Prof. G. G. Wallace Intelligent Polymer Research Institute ARC Center of Excellence for Electromaterials Science University of Wollongong 2500, Australia E-mail: [email protected]
DOI: 10.1002/adma.201500411
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1
www.advmat.de
REVIEW
www.MaterialsViews.com
While the concept of biocompatibility is loosely defined in the literature, the regulatory requirements to test the biocompatibility of devices for implantation are not. The Food and Drugs Administration (USFDA), the US-based consumer protection and regulatory body, set out a comprehensive set of requirements and tests for the biocompatibility of a device, as summarized in Table 1 and documented in full in ref. [2]. Any device to be implanted into humans must pass all of these tests before it can be released to market in the USA. The guidelines provided by the USFDA are based on ref. [3], however, with 2009–2010 revisions to the ISO10993 guidelines (particularly ISO10993–1, dealing with “evaluation and testing in the risk management process”), the USFDA released a draft revised version in April 2013 (UCM348890), which while similar to #G95–1, gives more detailed guidelines for the expected test outcomes, follow-up tests that researchers should consider in special cases, increased detail on how previously approved materials had been treated, and updated definitions for more novel materials, including nanostructured materials such as nanoparticles, carbon nanotubes, or graphene, or materials that are polymerized in situ.[4] At the time of writing this article, these guidelines remained in draft form, retrievable from the USFDA website.[5] Most other large markets around the world base their product testing requirements on either IS10993 or more directly from the guidelines laid down by the USFDA.
2. Bionics One often-overlooked but significant influence over the behavior of cells, tissue, or biological systems is electromagnetic signalling. A wide range of cell processes are influenced by naturally occurring electrical signals that modify biological behaviors, in the form of electromagnetic fields, ionic electrical charges, or even direct electrical charges (Figure 3).[6] Medical bionics involves the interfacing of biological and electronic systems, in order to externally replicate or influence these electronic signals. Controlling the electrical field or providing voltages, currents, or charge to stimulate cells via electrodes can be used to communicate information to cells (stimulation, discussed in Section 2.2.1), while measuring ionic fluxes associated with action potentials or muscle contraction allows communication from tissue to electronic systems (recording systems are discussed in more detail in Section 2.2.2).
2.1. Excitable Cells Cellular systems that generate electrical signals, such as nerves and muscle, obviously respond to electrical signals provided to them. Other cellular systems are also known to respond to electrical stimulation.
2.1.1. Nerves In the nervous system, electrical signals are communicated around the body in the form of controlled membrane potential (as shown in Figure 3C) changes resulting in charge gradients
2
wileyonlinelibrary.com
Brianna Thompson is a Research Fellow working on developing novel ophthalmological prostheses within the School of Aerospace Engineering, Nanyang Technological University. She has previously worked on neural interfacing and regeneration, bionic interfacing, and drug delivery at University of Wollongong, Australia, and biological applications of ionic liquids at Monash University, Melbourne, Australia. Her main research interests lie in finding uses for novel materials and fabrication process in biological systems. Eoin Murray is a Senior Research Fellow/R+D Project Manager at the Institute for Sports Research at Nanyang Technological University, Singapore. He has previously worked on the development of materials for bionics as an Australian Research Council Superscience Fellow at the University of Wollongong, Australia, nanoparticle synthesis and nano-structure formation at the LeibnizInstitut für Neue Materialien, Germany, and nanoparticles and conducting materials as an IRCSET Embark scholar at Dublin City University. His main research interests lie in the development and real-world application of multifunctional materials. Gordon Wallace is currently the Executive Research Director at the ARC Center of Excellence for Electromaterials Science and Director of the Intelligent Polymer Research Institute. He previously held an ARC Federation Fellowship and currently holds an ARC Laureate Fellowship. Professor Wallace’s research interests include organic conductors, nanomaterials and electrochemical probe methods of analysis, and the use of these in the development of Intelligent Polymer Systems. A current focus involves the use of these tools and materials in developing bio-communications from the molecular to skeletal domains in order to improve human performance via medical Bionics.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
REVIEW Figure 1. The tissue engineering process. 1) Cells are isolated from the patient or derived from established stem cell cultures. 2) A scaffold is constructed with the correct physical, mechanical, chemical and structural cues to encourage the development of cells into tissues. 3) Cells are seeded onto the scaffold in vitro and transferred into a bioreactor. The scaffold encourages cell proliferation and/or differentiation, the construct is incubated until the nascent tissue is ready for implantation. 4) The tissue-engineered scaffold with cells is re-implanted into the patient to develop into an integrated tissue. Adapted with permission.[1] Copyright 2007, Elsevier.
across cell membranes and action potentials along nerves, where the membrane potential may fluctuate from −70 mV to +35 mV briefly before returning to the −70 mV resting potential (a schematic of the voltage changes in membrane potential during action potentials is shown in Figure 4). Electrical fields or electrical stimulation may interact with the distribution of the gated ion channels or may modulate the function of voltage-influenced channels to modulate the firing of nerves at this physiological level.
Additionally, electrical fields or direct electrical stimulation can influence nerves at the molecular level, influencing proliferation, migration, axonal regeneration, and function of nerve cells.[9,10] Migration, axonal regeneration, and function are of the greatest interest from a tissue-engineering perspective, where axons and glial cells need to be encouraged to regrow along or repopulate damaged peripheral or spinal nerve gaps. Electrical fields and electrical stimulation have been used to enhance or direct axonal growth in vitro and in vivo and to
Figure 2. Examples of potential form factors for tissue-engineered structures for different applications. Many form factors require different fabrication techniques, which can impart their own restrictions and requirements on material properties.
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
3
www.advmat.de
REVIEW
www.MaterialsViews.com Table 1. Biocompatibility testing requirements outlined by USFDA #G95–1(Use of International Standard ISO-10993, “Biological Evaluation of Medical Devices Part 1: Evaluation and Testing”,1995)[2] and 2013 draft amendments[4] for approval of implanted medical devices (and materials) with a lifespan of greater than 30 days in humans for implantation into bone/tissue, or into areas in direct contact with the circulatory system (blood). Blooda)
Bone/Tissuea)
Cytotoxicity
䊉
䊉
Sensitization
䊉
䊉
Genotoxicity
䊉
䊉
Implantation
䊉
䊉
Irritation or intracutaneous reactivity
䊉
+
Systemic toxicity (acute)
䊉
+
Subchronic toxicity
䊉
+
Chronic toxicity
䊊
䊊
Carcinogenicity
䊊
䊊
Test
a) 䊉 – required for initial evaluation; 䊊 – may be required for further evaluation; + – are included in the Draft proposal for future FDA guidelines (UCM348890) (April 2013).
2.2. Electrical Stimulation and Recording
control the migration of neuronal, astrocyte, and Schwann cells.[9,11]
The interfacing of biology and electronics provides two-way communication – the ability to record from cellular systems and the ability to stimulate them.
2.1.2. Muscle
2.2.1. Stimulation Mechanisms
As with nerves, muscles have the ability to undergo depolarization and repolarization in an action potential (Figure 4). However, the end result of an action potential in muscle is not the transmission of information (as in nerves), but the contraction of the muscle cell. Muscles also typically have a more-negative membrane potential than neurons, generally sitting at closer to −90 mV than the approximately −50 to −70 mV across neural membranes. The electrical effects on muscle development have received less attention than neural cells and tissues. However, several papers have shown interesting changes in cultured muscle cells that have received electrical stimulation,[12] and on a wholebody level, electrical stimulation is applied to muscle tissue in the form of neuromuscular electric stimulation (NMES) for training and rehabilitation purposes in sports science.[13] Chronic low-frequency stimulation (1–10 Hz) of in vitro and in vivo muscle has been used to induce contractions in order to study the impact of muscle use on its development. These studies have found that electrical stimulation that results in contraction of muscle cells induces changes in the transcription of genes and the metabolism of the cells, with the pulse frequency determining the muscle phenotype.[12,14–16]
Whilst it is known that electrical fields or electrical stimulation are important in wound healing, nerve and muscle development and function, and cell migration,[8,19] there are many unknowns about the electrical parameters required for electrical control of behavior and about the mechanisms of cell response to electromagnetic cues. Molecular and cell level mechanisms for electromagnetic influence on cell behavior have been speculated upon, however no definitive answers have been given. In fact, it is likely that there are combinations of the many proposed mechanisms, including direct effects on specific voltage-gated receptors,[20] electrophoretic asymmetric distributions of membrane receptors,[8] membrane lipid redistribution, changes in fluidity of the plasma membrane, and the influence on lipid rafts, or modification of cell signalling cascades.[21] McCaig et al. proposed a model in which the electrical and chemical gradients that drive neural migration and axonal regeneration may be combined, wherein the receptors for neurotrophins may be electrophoretically concentrated on one side of the cell or another, leading to asymmetric reception of the chemical cues.[8] However, this alone would not account for cell responses to AC electrical stimulation, which has been shown to have positive effects on axonal regeneration. With incremental advances in imaging techniques, molecular markers, and spectroscopic techniques, the tools to study these influences should allow more-detailed characterization of cell response in the near future. In addition, as traditional metal electrodes have always been separated from the tissue undergoing stimulation, the differences between electric-field stimulation, ionic currents, and the provision of current/voltage directly at the membrane of the
2.1.3. Other Tissues In addition to the excitable tissues above, which generate their own electrical signals (and are therefore an obvious target for electrical stimulation to change behavior), several other tissues and processes have been shown to be greatly influenced
4
by electrical fields or electrical stimulation. Bones and dental tissues are influenced by the application of small electrical fields,[17] meaning that bionic interference with these systems may be applied to increase injury repair, or increase implant integration with the body. Recently, application of electric vfields to the dental field has generated much media attention with the promise to remove the need for dental fillings after a recent highly publicized discovery by King’s College London researchers claiming to increase remineralization of dental caries with application of a small electric current to accelerate the movement of calcium and phosphate into the area, allowing the body to regenerate its own tissues in a technique they have called electrically accelerated and enhanced remineralization.[18] In addition to these hard-tissue orthopaedic applications, electrical signals have been shown to greatly influence and direct soft-tissue wound healing (for more details on electrical influences on wound healing, readers are directed to the excellently titled reviews written by McCaig and co-workers).[8,19]
wileyonlinelibrary.com
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
REVIEW Figure 3. Naturally occurring electrical influences in biological systems. From A) Galvani's frog leg experiments with touching (1), electrically stimulating (2) or coupling (3) sciatic nerves to stimulate muscle contraction, to B) Borgens et al.'s work with transected lamprey axons,[7] there is a long history of electrical stimulation of, and electrical recordings from, biological tissues. At the cellular level, membrane potentials (Vm) can be established across intact plasma membranes (C) or transepithelial potentials (VTEP) across epithelial layers (D) due to the actions of ion-selective membrane channels, with cells having a net negative charge compared with their environment. Disruption of VTEP in wound injuries leads to a flow of current, as positive ions can now flow into the cell (E) or across the epithelium (F), demonstrating an electrical signal that mediates a biological response in non-excitable tissues. Reproduced with permission.[8] Copyright 2005, American Physiological Society.
cell remain unstudied and are completely unknown. However, with the advent of processable, potentially soluble conductors, such as chemically polymerized conducting polymer nanoparticles, the work of Martin and co-workers in growing conducting polymers around living cells,[22,23] or some of the graphene hydrogels discussed below, there is the opportunity to separate these modes of electrical stimulation and to study differences in the responses of cells.
2.2.2. Recording from Tissues As mentioned in Section 2.1.1 and 2.1.2, nerve and muscle tissues generate their own electrical signals, in the form of action potentials (illustrated Figure 4). The nervous system uses these electrical signals to pass and process large amounts information
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
around the body. Understanding how that information is coded, passed on, and integrated could be of great use to many fields of medicine, and to our understanding of the human brain, in addition to having use in future control of artificial prosthetics. Electrical recording of these action potentials with electrodes has allowed some information on neural networking to be gathered both in vitro and in vivo, but materials available for the production of recording electrodes have, until recently, limited how much information can be recorded from the complex and information-rich circuitry of the mammalian nervous system. Improvements in the flexibility, compliance, and electrode density available have been needed to get sufficient information to understand the systems. For recording action potentials from single excitable cells, intracellular recordings or patch-clamping are typically used. Microelectrodes are either inserted into the cell (in intracellular
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
5
www.advmat.de
REVIEW
www.MaterialsViews.com
Figure 4. Action potentials in neural, skeletal, and cardiac muscles, showing typical membrane potentials, action potential lengths and refractory periods for the depolarization events in different cells.
recordings) or sealed onto a small area of the membrane of the cell (typically via a micrometer-sized glass pipette), and variations in the membrane potential of the cell are monitored. These labor-intensive, specialized measurements have produced useful information about cell physiology and changes in cell behavior in response to applied stimuli, but, however, can only be applied to in vitro situations, and the limitation to recording a single cell at once limits the use of these techniques. Most recordings for multiple cells (named neural works when the cells being measured are neural in origin) or tissues are done through microelectrode array (MEA) systems, with the most popular in vivo arrays being the Michigan and Utah arrays shown in Figure 5. These metallic probes are inserted into neural tissues, where the uncoated tips (of micrometer dimension) record the neural-network activity. The size and impedance of the electrodes are important to allow recording with sufficiently high signal to noise ratio, to record with the high sensitivity required for the monitoring of the numerous and fast signals transferred in a neural network.
2.2.3. Requirements for Electrical Stimulation and Recording Electrodes The requirements for electrodes are the same as those for any implanted materials or device – they must be biocompatible and non-immunogenic, and have the ability to transmit highfidelity electronic information across the electrode–cellular interface.[25] The composition and structure of this interfacial region is determined by the cascade of molecular and cellular events initiated by the trauma of implantation, with both stimulating and recording electrodes often showing a decreased effectiveness over time, due to the formation of a fibrotic scar around the electrode. One major reason for this scar formation is the difference in compliance between the electrode and tissue, with the hard electrode often undergoing microscale movement within the soft tissue as the body moves. Not only does the chemistry of the electrodes need to be optimized to
6
wileyonlinelibrary.com
Figure 5. Microelectrode arrays that have been used for in vivo recordings of neural activity. a) Silicon 100 electrode array with 400 µm spacing between the electrodes, developed at the University of Utah. b) Polyamide “bioactive electrode” array. c) Michigan thin film 256-shank array of 1024 multiplexed sites with mounted signal processing electronics (arrow). Adapted with permission.[24] Copyright 2002, Nature Publishing Group.
avoid toxicity or immunogenicity issues, but the mechanical properties should be flexible or soft enough to match those of the tissue that the electrode is being implanted into in order to avoid this shear-induced inflammation. As well as the flexibility to move with the body, the electrode also needs to have
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
3. Graphene, Graphene Oxide, and Reduced Graphene Oxide “Graphene” can be used to refer to a broad range of carbonbased materials, with widely varying properties due to large differences on a molecular level from synthesis or functionalization. Similarly, nomenclature varies widely between research groups and fields, and this review will use the nomenclature proposed by Sanchez et al. (Figure 6).[27] Briefly, “pristine” or “monolayer” graphene is a single twodimensional sheet of defect-free, polycyclic, hexagonally arranged, sp2-bonded aromatic carbon, and is generally produced either by mechanical exfoliation of graphite, epitaxial growth, or chemical vapor deposition. The unique, exemplary
REVIEW
sufficient toughness to hold together under surgical and physiological conditions. Depending on the application, the electrode may need to degrade away to make room for re-developing tissues, or may need the durability to last and function in the body for many years. The degradation or durability of the electrode needs to be considered for each proposed application, and the materials used as a conductor chosen accordingly. There are very few physiologically degradable materials that will have sufficient electrical properties to be used as effective electrodes. A final, vital property of an electrode is the electrical properties. The impedance of the electrode must be sufficiently low to enable safe stimulation or efficient recording in physiological conditions. This means that the electrode must have sufficient conductivity in a slightly salty, aqueous, and 37 °C environment to allow fine control over current, voltage, or electric field, or the recording of tiny fluctuations in any of these parameters. For the recording of electrical signals, field potentials (electroencephalograms and electrocortiograms) are low frequency (1–200 Hz) and low current (5–300 µV for electroencephalograms measured on top of the skull, or 0.01–5 mV for electrocortiograms measured on the surface of the brain, or can be measured locally with sharp, low-surface-area penetrating electrodes (
REVIEW
Graphite Oxide to Graphene. Biomaterials to Bionics Brianna C. Thompson,* Eoin Murray, and Gordon G. Wallace* synthetic scaffolds to make tissue engineered structures is shown in Figure 1. The engineering of different kinds of tissues has specific challenges and requirements, from the form factor of the implant (e.g., blocks, tubes, membranes, sponges etc.) and the physical properties of the materials (e.g., stiffness, softness, porosity, degradability) to the desired biological outcomes (e.g., promoting cell proliferation, differentiation, cell invasion, cell repulsion). All tissue engineering targets require different materials and fabrication strategies, and so the fields of tissue engineering and biomaterials development require a diverse array of expertise. The form factors of implants can vary tremendously between applications – a variety of which are illustrated in Figure 2. Some applications require the replacement of large blocks of tissue (such as muscle or fat), and in these applications, it is often necessary to devise a strategy to allow for transport of nutrients and waste into/out of the implant. Other applications need a membrane (to replace skin in wound healing or to replace internal membranes), or fibrous structures (in the engineering of nerve or muscle fibers), or something much more complicated and specific (in the engineering the complicated structure of an artificial kidney or heart). As fabrication technologies become more advanced, more tools to enable the production of these more complicated structures such as 3D printing, electrospinning, wet-spinning, extrusion, or lithography become available. Each fabrication strategy imparts its own requirements on the material used, so the ultimate material properties must be balanced between the requirements for the tissue development needed, processability, and being amenable to fabrication protocols that enable the ideal form factor to be realized. However, despite the need for tailoring material and implant properties for different parts of the body, one property is required in all tissue engineering – biocompatibility. The concept of biocompatibility is often poorly defined, but is generally taken to mean that the material or implant lacks toxicity and immunogenicity and will have no adverse effect on the biological system into which it is being introduced. Cells from the patient are used to seed cell scaffolds in order to reduce the potential for an immune¡ reaction against transplants of tissues or stem cells from donors, or xenotransplantation, reducing the risk of implant/organ rejection. The aim is to allow a “normal” tissue to develop that has the same genetic and epigenetic cues as the other tissues in the patient.
The advent of implantable biomaterials has revolutionized medical treatment, allowing the development of the fields of tissue engineering and medical bionic devices (e.g., cochlea implants to restore hearing, vagus nerve stimulators to control Parkinson’s disease, and cardiac pace makers). Similarly, future materials developments are likely to continue to drive development in treatment of disease and disability, or even enhancing human potential. The material requirements for implantable devices are stringent. In all cases they must be nontoxic and provide appropriate mechanical integrity for the application at hand. In the case of scaffolds for tissue regeneration, biodegradability in an appropriate time frame may be required, and for medical bionics electronic conductivity is essential. The emergence of graphene and graphene-family composites has resulted in materials and structures highly relevant to the expansion of the biomaterials inventory available for implantable medical devices. The rich chemistries available are able to ensure properties uncovered in the nanodomain are conveyed into the world of macroscopic devices. Here, the inherent properties of graphene, along with how graphene or structures containing it interface with living cells and the effect of electrical stimulation on nerves and cells, are reviewed.
1. Tissue Engineering The goal of a tissue engineer is to fix or replace tissues or organs that have been damaged or destroyed by accident or disease. In order to achieve this, one approach is to combine a fabricated scaffold composed of one or more materials having the desired properties with cells (either sourced and differentiated externally or internally to be populated with endogenous cells in situ), chemical cues (drugs, growth factors or other endogenous signals), and physical cues to encourage cells to develop into functional tissues. The general process of combining cells and
Dr. B. C. Thompson School of Mechanical and Aerospace Engineering Nanyang Technological University 639798, Singapore E-mail: [email protected] Dr. E. Murray Institute for Sports Research Nanyang Technological University 639798, Singapore Prof. G. G. Wallace Intelligent Polymer Research Institute ARC Center of Excellence for Electromaterials Science University of Wollongong 2500, Australia E-mail: [email protected]
DOI: 10.1002/adma.201500411
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1
www.advmat.de
REVIEW
www.MaterialsViews.com
While the concept of biocompatibility is loosely defined in the literature, the regulatory requirements to test the biocompatibility of devices for implantation are not. The Food and Drugs Administration (USFDA), the US-based consumer protection and regulatory body, set out a comprehensive set of requirements and tests for the biocompatibility of a device, as summarized in Table 1 and documented in full in ref. [2]. Any device to be implanted into humans must pass all of these tests before it can be released to market in the USA. The guidelines provided by the USFDA are based on ref. [3], however, with 2009–2010 revisions to the ISO10993 guidelines (particularly ISO10993–1, dealing with “evaluation and testing in the risk management process”), the USFDA released a draft revised version in April 2013 (UCM348890), which while similar to #G95–1, gives more detailed guidelines for the expected test outcomes, follow-up tests that researchers should consider in special cases, increased detail on how previously approved materials had been treated, and updated definitions for more novel materials, including nanostructured materials such as nanoparticles, carbon nanotubes, or graphene, or materials that are polymerized in situ.[4] At the time of writing this article, these guidelines remained in draft form, retrievable from the USFDA website.[5] Most other large markets around the world base their product testing requirements on either IS10993 or more directly from the guidelines laid down by the USFDA.
2. Bionics One often-overlooked but significant influence over the behavior of cells, tissue, or biological systems is electromagnetic signalling. A wide range of cell processes are influenced by naturally occurring electrical signals that modify biological behaviors, in the form of electromagnetic fields, ionic electrical charges, or even direct electrical charges (Figure 3).[6] Medical bionics involves the interfacing of biological and electronic systems, in order to externally replicate or influence these electronic signals. Controlling the electrical field or providing voltages, currents, or charge to stimulate cells via electrodes can be used to communicate information to cells (stimulation, discussed in Section 2.2.1), while measuring ionic fluxes associated with action potentials or muscle contraction allows communication from tissue to electronic systems (recording systems are discussed in more detail in Section 2.2.2).
2.1. Excitable Cells Cellular systems that generate electrical signals, such as nerves and muscle, obviously respond to electrical signals provided to them. Other cellular systems are also known to respond to electrical stimulation.
2.1.1. Nerves In the nervous system, electrical signals are communicated around the body in the form of controlled membrane potential (as shown in Figure 3C) changes resulting in charge gradients
2
wileyonlinelibrary.com
Brianna Thompson is a Research Fellow working on developing novel ophthalmological prostheses within the School of Aerospace Engineering, Nanyang Technological University. She has previously worked on neural interfacing and regeneration, bionic interfacing, and drug delivery at University of Wollongong, Australia, and biological applications of ionic liquids at Monash University, Melbourne, Australia. Her main research interests lie in finding uses for novel materials and fabrication process in biological systems. Eoin Murray is a Senior Research Fellow/R+D Project Manager at the Institute for Sports Research at Nanyang Technological University, Singapore. He has previously worked on the development of materials for bionics as an Australian Research Council Superscience Fellow at the University of Wollongong, Australia, nanoparticle synthesis and nano-structure formation at the LeibnizInstitut für Neue Materialien, Germany, and nanoparticles and conducting materials as an IRCSET Embark scholar at Dublin City University. His main research interests lie in the development and real-world application of multifunctional materials. Gordon Wallace is currently the Executive Research Director at the ARC Center of Excellence for Electromaterials Science and Director of the Intelligent Polymer Research Institute. He previously held an ARC Federation Fellowship and currently holds an ARC Laureate Fellowship. Professor Wallace’s research interests include organic conductors, nanomaterials and electrochemical probe methods of analysis, and the use of these in the development of Intelligent Polymer Systems. A current focus involves the use of these tools and materials in developing bio-communications from the molecular to skeletal domains in order to improve human performance via medical Bionics.
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
REVIEW Figure 1. The tissue engineering process. 1) Cells are isolated from the patient or derived from established stem cell cultures. 2) A scaffold is constructed with the correct physical, mechanical, chemical and structural cues to encourage the development of cells into tissues. 3) Cells are seeded onto the scaffold in vitro and transferred into a bioreactor. The scaffold encourages cell proliferation and/or differentiation, the construct is incubated until the nascent tissue is ready for implantation. 4) The tissue-engineered scaffold with cells is re-implanted into the patient to develop into an integrated tissue. Adapted with permission.[1] Copyright 2007, Elsevier.
across cell membranes and action potentials along nerves, where the membrane potential may fluctuate from −70 mV to +35 mV briefly before returning to the −70 mV resting potential (a schematic of the voltage changes in membrane potential during action potentials is shown in Figure 4). Electrical fields or electrical stimulation may interact with the distribution of the gated ion channels or may modulate the function of voltage-influenced channels to modulate the firing of nerves at this physiological level.
Additionally, electrical fields or direct electrical stimulation can influence nerves at the molecular level, influencing proliferation, migration, axonal regeneration, and function of nerve cells.[9,10] Migration, axonal regeneration, and function are of the greatest interest from a tissue-engineering perspective, where axons and glial cells need to be encouraged to regrow along or repopulate damaged peripheral or spinal nerve gaps. Electrical fields and electrical stimulation have been used to enhance or direct axonal growth in vitro and in vivo and to
Figure 2. Examples of potential form factors for tissue-engineered structures for different applications. Many form factors require different fabrication techniques, which can impart their own restrictions and requirements on material properties.
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
3
www.advmat.de
REVIEW
www.MaterialsViews.com Table 1. Biocompatibility testing requirements outlined by USFDA #G95–1(Use of International Standard ISO-10993, “Biological Evaluation of Medical Devices Part 1: Evaluation and Testing”,1995)[2] and 2013 draft amendments[4] for approval of implanted medical devices (and materials) with a lifespan of greater than 30 days in humans for implantation into bone/tissue, or into areas in direct contact with the circulatory system (blood). Blooda)
Bone/Tissuea)
Cytotoxicity
䊉
䊉
Sensitization
䊉
䊉
Genotoxicity
䊉
䊉
Implantation
䊉
䊉
Irritation or intracutaneous reactivity
䊉
+
Systemic toxicity (acute)
䊉
+
Subchronic toxicity
䊉
+
Chronic toxicity
䊊
䊊
Carcinogenicity
䊊
䊊
Test
a) 䊉 – required for initial evaluation; 䊊 – may be required for further evaluation; + – are included in the Draft proposal for future FDA guidelines (UCM348890) (April 2013).
2.2. Electrical Stimulation and Recording
control the migration of neuronal, astrocyte, and Schwann cells.[9,11]
The interfacing of biology and electronics provides two-way communication – the ability to record from cellular systems and the ability to stimulate them.
2.1.2. Muscle
2.2.1. Stimulation Mechanisms
As with nerves, muscles have the ability to undergo depolarization and repolarization in an action potential (Figure 4). However, the end result of an action potential in muscle is not the transmission of information (as in nerves), but the contraction of the muscle cell. Muscles also typically have a more-negative membrane potential than neurons, generally sitting at closer to −90 mV than the approximately −50 to −70 mV across neural membranes. The electrical effects on muscle development have received less attention than neural cells and tissues. However, several papers have shown interesting changes in cultured muscle cells that have received electrical stimulation,[12] and on a wholebody level, electrical stimulation is applied to muscle tissue in the form of neuromuscular electric stimulation (NMES) for training and rehabilitation purposes in sports science.[13] Chronic low-frequency stimulation (1–10 Hz) of in vitro and in vivo muscle has been used to induce contractions in order to study the impact of muscle use on its development. These studies have found that electrical stimulation that results in contraction of muscle cells induces changes in the transcription of genes and the metabolism of the cells, with the pulse frequency determining the muscle phenotype.[12,14–16]
Whilst it is known that electrical fields or electrical stimulation are important in wound healing, nerve and muscle development and function, and cell migration,[8,19] there are many unknowns about the electrical parameters required for electrical control of behavior and about the mechanisms of cell response to electromagnetic cues. Molecular and cell level mechanisms for electromagnetic influence on cell behavior have been speculated upon, however no definitive answers have been given. In fact, it is likely that there are combinations of the many proposed mechanisms, including direct effects on specific voltage-gated receptors,[20] electrophoretic asymmetric distributions of membrane receptors,[8] membrane lipid redistribution, changes in fluidity of the plasma membrane, and the influence on lipid rafts, or modification of cell signalling cascades.[21] McCaig et al. proposed a model in which the electrical and chemical gradients that drive neural migration and axonal regeneration may be combined, wherein the receptors for neurotrophins may be electrophoretically concentrated on one side of the cell or another, leading to asymmetric reception of the chemical cues.[8] However, this alone would not account for cell responses to AC electrical stimulation, which has been shown to have positive effects on axonal regeneration. With incremental advances in imaging techniques, molecular markers, and spectroscopic techniques, the tools to study these influences should allow more-detailed characterization of cell response in the near future. In addition, as traditional metal electrodes have always been separated from the tissue undergoing stimulation, the differences between electric-field stimulation, ionic currents, and the provision of current/voltage directly at the membrane of the
2.1.3. Other Tissues In addition to the excitable tissues above, which generate their own electrical signals (and are therefore an obvious target for electrical stimulation to change behavior), several other tissues and processes have been shown to be greatly influenced
4
by electrical fields or electrical stimulation. Bones and dental tissues are influenced by the application of small electrical fields,[17] meaning that bionic interference with these systems may be applied to increase injury repair, or increase implant integration with the body. Recently, application of electric vfields to the dental field has generated much media attention with the promise to remove the need for dental fillings after a recent highly publicized discovery by King’s College London researchers claiming to increase remineralization of dental caries with application of a small electric current to accelerate the movement of calcium and phosphate into the area, allowing the body to regenerate its own tissues in a technique they have called electrically accelerated and enhanced remineralization.[18] In addition to these hard-tissue orthopaedic applications, electrical signals have been shown to greatly influence and direct soft-tissue wound healing (for more details on electrical influences on wound healing, readers are directed to the excellently titled reviews written by McCaig and co-workers).[8,19]
wileyonlinelibrary.com
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
REVIEW Figure 3. Naturally occurring electrical influences in biological systems. From A) Galvani's frog leg experiments with touching (1), electrically stimulating (2) or coupling (3) sciatic nerves to stimulate muscle contraction, to B) Borgens et al.'s work with transected lamprey axons,[7] there is a long history of electrical stimulation of, and electrical recordings from, biological tissues. At the cellular level, membrane potentials (Vm) can be established across intact plasma membranes (C) or transepithelial potentials (VTEP) across epithelial layers (D) due to the actions of ion-selective membrane channels, with cells having a net negative charge compared with their environment. Disruption of VTEP in wound injuries leads to a flow of current, as positive ions can now flow into the cell (E) or across the epithelium (F), demonstrating an electrical signal that mediates a biological response in non-excitable tissues. Reproduced with permission.[8] Copyright 2005, American Physiological Society.
cell remain unstudied and are completely unknown. However, with the advent of processable, potentially soluble conductors, such as chemically polymerized conducting polymer nanoparticles, the work of Martin and co-workers in growing conducting polymers around living cells,[22,23] or some of the graphene hydrogels discussed below, there is the opportunity to separate these modes of electrical stimulation and to study differences in the responses of cells.
2.2.2. Recording from Tissues As mentioned in Section 2.1.1 and 2.1.2, nerve and muscle tissues generate their own electrical signals, in the form of action potentials (illustrated Figure 4). The nervous system uses these electrical signals to pass and process large amounts information
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
around the body. Understanding how that information is coded, passed on, and integrated could be of great use to many fields of medicine, and to our understanding of the human brain, in addition to having use in future control of artificial prosthetics. Electrical recording of these action potentials with electrodes has allowed some information on neural networking to be gathered both in vitro and in vivo, but materials available for the production of recording electrodes have, until recently, limited how much information can be recorded from the complex and information-rich circuitry of the mammalian nervous system. Improvements in the flexibility, compliance, and electrode density available have been needed to get sufficient information to understand the systems. For recording action potentials from single excitable cells, intracellular recordings or patch-clamping are typically used. Microelectrodes are either inserted into the cell (in intracellular
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
5
www.advmat.de
REVIEW
www.MaterialsViews.com
Figure 4. Action potentials in neural, skeletal, and cardiac muscles, showing typical membrane potentials, action potential lengths and refractory periods for the depolarization events in different cells.
recordings) or sealed onto a small area of the membrane of the cell (typically via a micrometer-sized glass pipette), and variations in the membrane potential of the cell are monitored. These labor-intensive, specialized measurements have produced useful information about cell physiology and changes in cell behavior in response to applied stimuli, but, however, can only be applied to in vitro situations, and the limitation to recording a single cell at once limits the use of these techniques. Most recordings for multiple cells (named neural works when the cells being measured are neural in origin) or tissues are done through microelectrode array (MEA) systems, with the most popular in vivo arrays being the Michigan and Utah arrays shown in Figure 5. These metallic probes are inserted into neural tissues, where the uncoated tips (of micrometer dimension) record the neural-network activity. The size and impedance of the electrodes are important to allow recording with sufficiently high signal to noise ratio, to record with the high sensitivity required for the monitoring of the numerous and fast signals transferred in a neural network.
2.2.3. Requirements for Electrical Stimulation and Recording Electrodes The requirements for electrodes are the same as those for any implanted materials or device – they must be biocompatible and non-immunogenic, and have the ability to transmit highfidelity electronic information across the electrode–cellular interface.[25] The composition and structure of this interfacial region is determined by the cascade of molecular and cellular events initiated by the trauma of implantation, with both stimulating and recording electrodes often showing a decreased effectiveness over time, due to the formation of a fibrotic scar around the electrode. One major reason for this scar formation is the difference in compliance between the electrode and tissue, with the hard electrode often undergoing microscale movement within the soft tissue as the body moves. Not only does the chemistry of the electrodes need to be optimized to
6
wileyonlinelibrary.com
Figure 5. Microelectrode arrays that have been used for in vivo recordings of neural activity. a) Silicon 100 electrode array with 400 µm spacing between the electrodes, developed at the University of Utah. b) Polyamide “bioactive electrode” array. c) Michigan thin film 256-shank array of 1024 multiplexed sites with mounted signal processing electronics (arrow). Adapted with permission.[24] Copyright 2002, Nature Publishing Group.
avoid toxicity or immunogenicity issues, but the mechanical properties should be flexible or soft enough to match those of the tissue that the electrode is being implanted into in order to avoid this shear-induced inflammation. As well as the flexibility to move with the body, the electrode also needs to have
© 2015 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2015, DOI: 10.1002/adma.201500411
www.advmat.de www.MaterialsViews.com
3. Graphene, Graphene Oxide, and Reduced Graphene Oxide “Graphene” can be used to refer to a broad range of carbonbased materials, with widely varying properties due to large differences on a molecular level from synthesis or functionalization. Similarly, nomenclature varies widely between research groups and fields, and this review will use the nomenclature proposed by Sanchez et al. (Figure 6).[27] Briefly, “pristine” or “monolayer” graphene is a single twodimensional sheet of defect-free, polycyclic, hexagonally arranged, sp2-bonded aromatic carbon, and is generally produced either by mechanical exfoliation of graphite, epitaxial growth, or chemical vapor deposition. The unique, exemplary
REVIEW
sufficient toughness to hold together under surgical and physiological conditions. Depending on the application, the electrode may need to degrade away to make room for re-developing tissues, or may need the durability to last and function in the body for many years. The degradation or durability of the electrode needs to be considered for each proposed application, and the materials used as a conductor chosen accordingly. There are very few physiologically degradable materials that will have sufficient electrical properties to be used as effective electrodes. A final, vital property of an electrode is the electrical properties. The impedance of the electrode must be sufficiently low to enable safe stimulation or efficient recording in physiological conditions. This means that the electrode must have sufficient conductivity in a slightly salty, aqueous, and 37 °C environment to allow fine control over current, voltage, or electric field, or the recording of tiny fluctuations in any of these parameters. For the recording of electrical signals, field potentials (electroencephalograms and electrocortiograms) are low frequency (1–200 Hz) and low current (5–300 µV for electroencephalograms measured on top of the skull, or 0.01–5 mV for electrocortiograms measured on the surface of the brain, or can be measured locally with sharp, low-surface-area penetrating electrodes (
