IMAGINE A WORLD WITHOUT BIOSECURITY.

Transcription

1 IMAGINE A WORLD WITHOUT BIOSECURITY.

2 Using Biosensors to Detect Emerging Infectious Diseases Prepared for The Australian Biosecurity Cooperative Research Centre Final Report July 2005 Prepared by Gordon Parkinson & Bobby Pejcic Nanochemistry Research Institute Curtin University of Technology Perth, Western Australia

3 Table of Contents 1 EXECUTIVE SUMMARY INTRODUCTION OBJECTIVES SEARCH STRATEGY DEFINITION OF A BIOSENSOR BIOSENSORS FUNDAMENTAL ASPECTS OF A BIOSENSOR Biological Recognition Element Transducer IMMOBILISATION TECHNIQUES ENZYME-BASED SENSORS IMMUNOSENSORS DNA / NUCLEIC ACID SENSORS ELECTROCHEMICAL-BASED BIOSENSORS Amperometric Potentiometric Electrical Impedance Spectroscopy Conductometric Electrode Materials Microelectrodes OPTICAL-BASED BIOSENSORS Absorption & Reflectance Spectroscopy Chemiluminescence Fluorescence & Phosphorescence Optical Fibers PIEZOELECTRIC-BASED BIOSENSORS CALORIMETRIC-BASED BIOSENSORS OTHERS LAB-ON-A-CHIP DEVICES IN VIVO & IMPLANTABLE BIOSENSORS SOME COMMERCIAL BIOSENSORS BIOSENSOR ACTIVITIES AROUND THE GLOBE UNIVERSITIES & LABORATORIES Cranfield University (Silsoe, England) Lawrence Livermore National Laboratory (Livermore, USA) Lund University (Sweden) Naval Research Laboratory (Washington, USA) New Mexico State University (Las Cruces, USA) Oak Ridge National Laboratory (USA) Swiss Federal Institute of Technology (Zurich, Switzerland) Tokyo University of Technology & University of Tokyo (Japan) University of Connecticut (Farmington, USA) University of Florence (Italy) University of Michigan (Ann Arbor, USA) COMPANIES Abbott Laboratories ABTECH Scientific Inc Affymetrix, Inc Bayer AG (Diagnostics Division) Biacore International AB Cygnus, Inc

4 5.2.7 DiagnoSwiss Lifescan Neogen Corporation Panbio Pelikan Technologies, Inc Roche Diagnostics AG BIOSENSOR OUTLOOK MARKET FORECASTS AN OVERVIEW OF TRANSDUCER TECHNOLOGIES SUMMARY BIOSENSORS IN THE CONTEXT OF THE AB-CRC GLOBAL STATUS OF BIOSENSOR DEVELOPMENT NON BIOSENSOR STRATEGIES BIOSENSOR DEVELOPMENT WITHIN THE AB-CRC CONCLUSIONS REFERENCES

5 1 Executive Summary The work contained in this report is the result of a scoping study on biosensors. This report provides an overview of the various types of biosensor technologies, which have been developed for the detection of various diseases. Since well over 6000 articles have been published in the area of biosensors just from the years 1997 to 2004 alone (English journals), this document is not meant to be a comprehensive review, but rather a critical review, presenting a selection of the most significant technologies and advances in this field. It has been shown that biosensors have recently undergone significant improvements in terms of their achievable selectivity and detection limits. Although there is a host of biosensor technologies available, either commercially or in the scientific literature, electrochemical-based sensors appear more suited for field monitoring applications (e.g. hand-held). Likewise, the amperometric-based technology is the most appropriate platform for the development of an implantable biosensor. However, in terms of screening a large number of samples simultaneously, optical biosensors are more suited for this type of application. A search of the scientific literature (e.g., journal articles, conference proceedings, patents) along with information obtained from companies and personal communications with various world authorities on biosensors has revealed that biosensors play a significant role in medicine, agriculture / food, environmental and industrial monitoring. Will biosensors continue to play an important role in the future? Which biosensor / sensor platform is suitable for the real-time detection of infectious diseases? Which biosensor is geared up for the development of an implantable device? These and many other questions are addressed in this report. The purpose of this document is to present an impartial forecast on the future of biosensors particularly in relation to the detection of emerging infectious diseases. It describes some of the technologies behind the various types of biosensors in terms of transduction and bioreceptor techniques. This will provide a technical basis on which decisions about the selection and application of biosensors can be made. 2 Introduction In early 2004, the Australian Biosecurity Cooperative Research Centre (AB-CRC) commissioned a study to identify and assess opportunities in the field of implantable and handheld biosensors, for the detection of emerging infectious diseases. The Nanochemistry Research Institute has carried out a scoping study and this report presents some of the initial outcomes. A search of the scientific literature (i.e., journal 3

6 articles, conference proceedings) has provided most of the references. This scoping study also contains information obtained from various companies, the patent literature, and personal communications with several world authorities on biosensors. 2.1 Objectives The aims of this study are to: a) Provide an overview of the most common and widely used biosensor technologies, which have been reported recently in the literature; b) Address the development of implantable and/or handheld biosensors to detect pathogens and diseases; c) Highlight some recent advances in biosensor technology; d) Identify market opportunities for existing biosensor technologies; e) Determine the potential role of biosensor development / application within the AB-CRC. 2.2 Search strategy A number of electronic databases were searched to locate the relevant studies. The following databases were searched: American Chemical Society Publications Australian Medical Index Current Contents ISI Web of Knowledge MEDLINE / PubMed Science Direct SciFinder (1970 to 2005) Web-based using Google search engine Wiley InterScience 3 Definition of a Biosensor Before the various types of biosensor technologies are discussed, it is first necessary to define what is a biosensor. Unfortunately, the term biosensor has been loosely applied in the literature, noting that on various occasions it has been used to describe an analytical device that also incorporates an additional separation step. According to 4

7 IUPAC recommendations 1999, a biosensor is a self-contained integrated receptortransducer device, which is capable of providing selective quantitative or semiquantitative analytical information using a biological recognition element [1]. Essentially it is an analytical device, which incorporates a biological recognition element in close proximity or integrated with a signal transducer to provide a sensing system specific for the target analyte [2]. Ideally, a biosensor should be a reagentless analytical device; however, in most studies reported in the literature a co-substrate is normally used in the detection process. The purpose of a biosensor is to provide rapid, real-time, and reliable information about the biochemical composition of its surrounding environment. Ideally, it is a device that is capable of responding continuously, reversibly, and does not perturb the sample. 4 Biosensors 4.1 Fundamental Aspects of a Biosensor A biosensor consists of three main components, a biological detection system, a transducer and an output system. Figure 1 shows a schematic diagram of the typical components in a biosensor. Many different types of biosensors are presently available [2-4]; however, all of them essentially comprise a biological recognition element or bioreceptor, which interacts with the analyte and responds in some manner that can be detected by a transducer. BIORECEPTOR TRANSDUCER Signal TARGET ANALYTE X No Signal SAMPLE Figure 1 Schematic diagram of a typical biosensor. 5

8 4.1.1 Biological Recognition Element The biological recognition element or bioreceptor is the most crucial component of the biosensor device. The bioreceptor is the key to specificity, and can be classified according to several different groups as shown in Figure 2. Its function is to impart selectivity so that the biosensor responds only to a particular analyte or molecule of interest, hence avoiding interferences from other substances. Generally, there are three principal classes of biosensors. The three groups are distinguished from one another by the nature of the process and in terms of their biochemical or biological component, e.g. biocatalytic (i.e., enzyme), immunological (i.e., antibody) and nucleic acid (i.e., DNA). It is important to note that some biosensors have been developed which use either biomimetic or cell bioreceptors [5]. However, these bioreceptors will not be discussed in this report. Biosensors Bioreceptor Transducer Antibody DNA Biomimetic Optical Mass Enzyme Cell Electrochemical Other Cellular System Non-Enzymatic Proteins Figure 2 Different categories of a biosensor [5] Transducer The transducer is another component of the biosensor, which plays an important role in the detection process. Biosensors are normally categorised according to the transduction 6

9 method they employ (see Figure 2). A wide variety of transducer methods have been developed in the past decade; however, a recent literature review has shown that the most popular and common methods presently available are: a) electrochemical; b) optical; c) piezoelectric; d) thermal or calorimetric [2-4, 6]. It is also important to note that these groups can be further divided into general categories: nonlabeled or label-free types, which are based on the direct measurement of a phenomena occurring during the biochemical reactions on a transducer surface; and labeled, which relies on the detection of a specific label. Research into label-free biosensors continues to grow [7]; however labeled ones are more common and are extremely successful in a multitude of platforms. 4.2 Immobilisation Techniques For a biosensor to be highly successful it is somewhat necessary that the biorecognition molecule, the molecule that is responsible for biological recognition, remains attached irreversibly to the transducer. Immobilisation plays a major role in determining the overall performance of a biosensor [2]. There are significant challenges in immobilising antibodies, proteins or nucleic acids on selected sites while still retaining the activity of the biological moiety. In the case of a redox enzyme, the transduced signal is the electron transfer process in the enzyme-substrate reaction, and efficient electron transfer is needed between the enzyme and transducer. This cannot be achieved by simply ensuring intimate contact between the electrode surface and the redox protein, but can be accomplished by employing a mediator to shuttle the electrons between the enzyme and electrode [8]. Controlling the surface chemistry and coverage is paramount in ensuring high reactivity, stability, orientation and accessibility as well as minimizing non-specific adsorption processes. Immobilisation of the biomolecule at the sensor surface can be accomplished in various ways, and several different methods have been developed to ensure that the biorecognition molecule remains attached to the transducer. A number of papers have recently appeared in the literature, which review various immobilisation techniques [8-14]. Most workers have covalently attached the bioreceptors onto solid surfaces via a self-assembled monolayer [9, 10, 12, 15, 16]. It is reported that the formation of a selfassembled monolayer is both a simple and reliable method of binding biorecognition 7

10 molecules [10, 12]. Some reports suggest that protein adsorption is a more simplistic and reproducible method [17]. One of the main problems when immobilising a biomolecule such as an enzyme onto a solid platform is that the activity of the enzyme may be compromised. However, the work of Reynolds and coworkers has shown that enzymes can be attached to porous silicon using an organic linker, and still retain biomolecule activity [18, 19]. Consequently, this method of biomolecule attachment represents a major breakthrough, and opens the way for a new class of silicon-based biosensors. Recent cutting-edge research by Cheung et al. (2003) has shown that scanning probe nanolithography can be used in conjunction with chemoselective protein-to-surface linkers to create templates for fabricating virus arrays [20]. This work has important applications in proteomic analysis and genomic analysis. In summary, the proper functioning of the enzyme biosensor is mainly determined by the immobilisation procedure [8]. 4.3 Enzyme-Based Sensors Enzymes have the longest tradition in the field of biosensors. Since 1997, there have been over 2000 articles published in the literature on enzyme-based biosensors. These biosensors primarily rely on two operational mechanisms. They can be used as a bioreceptor based on their specific binding capabilities or according to their catalytic transformation of a species into a detectable form [2, 4]. In most reports in the literature they have been employed according to their catalytic activity [5, 21], noting that horseradish peroxidase (HRP), alkaline phosphatase (AP) and glucose oxidase (GOD) are three enzymes that have been employed in most biosensor studies [4, 22, 23]. The detection limit of these biosensors is mainly determined by the enzyme s activity, which can be described by the Michaelis-Menten equation [2]. However, the major limitation with enzyme-based biosensors is the stability of the enzyme, which depends on various conditions such as the temperature, ph, etc [3, 21, 22]. The ability to maintain enzyme activity for a long period of time still remains a major obstacle [3, 21, 24]. Another issue that governs the success of an enzyme-based sensor depends primarily on the contact between the enzyme and electrode surface [4]. Despite these pitfalls the enzyme-based sensor is still the most commonly used biosensor, and this is largely due to the need for monitoring glucose in blood [21, 24]. Some recent studies 8

11 have shown that enzyme-based biosensors can be used to detect very low levels (i.e., ~10 16 M) of pesticides [25]. 4.4 Immunosensors Immunosensors are a type of biosensor, which use antibodies as the biospecific sensing element, and are based on the ability of an antibody to form complexes with the corresponding antigen [2]. The antibody-antigen reaction is highly selective, and is analogous to a lock and key fit [2]. Some reports suggest that the lock-and-keymechanism is much more complicated, noting that both antigens and antibodies are flexible and can undergo mutual adaptation [26]. Immunoassays have become a standard tool in clinical chemistry, noting that these are highly sophisticated automated instruments used to analyze a number of samples in a short time frame [27]. Since 1997, there have been at least 800 papers reported in the literature on immunosensors. The development of immunosensors for the detection of diseases has received a great deal of attention lately and this has largely been driven by the need to develop hand-held devices for point of care measurements [27-29]. It is important to note that immunosensor technologies have been derived from the standard immunoassay approach [27, 30]. A tracer either labels the occupied sites of the antibody or the free ones [30]. Immunosensors can incorporate either the antigen or the antibody onto the sensor surface, although the latter approach has been used most often [28]. Optical and electrochemical detection methods are most frequently used in immunosensors [28]. A number of papers have been recently published on various immunosensors, noting that these have employed either an electrochemical [31, 32] or optical [33-36] transduction method. Detection by electrochemical immunosensors is generally achieved by using either electroactive labels or enzyme labelling [21]. A common challenge facing immunosensors is that they are not completely reversible, so that only a single immunoassay can be performed [3]. Subsequently, some research effort has been directed towards the development of renewable antibody surfaces [37]. 4.5 DNA / Nucleic Acid Sensors Traditional techniques for DNA sequencing are based on the coupling of electrophoretic separations and radio-isotopic ( 32 P) detection [38]. These methods are known to be labour intensive, time consuming, high cost, hazardous, have disposal problems 9

12 associated with radioactive waste, and are not well suited for routine and rapid environmental analysis [38, 39]. Subsequently, various promising alternative methods of DNA detection, which use a non-radioactive labelled probe, have been developed. The detection of specific DNA sequences provides the fundamental basis for detecting a wide variety of microbial and viral pathogens [9]. Several reviews have been published on the development and application of DNA sensors for the testing of virus infections [9, 11, 38-40], noting that viruses appear to be almost uniquely DNA or RNA composed within an outer coat or capsid of protein [2]. In essence, the technology relies on the immobilisation of a short (20 40mer) synthetic oligomer [the single-stranded DNA (ssdna)], whose sequence is complementary to the target of interest [38]. Exposure of the sensor to a sample containing the target results in the formation of the hybrid on the surface, and various transduction methods (i.e., optical, electrochemical and piezoelectric) have been used to detect duplex formation [38, 41]. Gooding (2002) revealed that relatively few DNA biosensor studies have been carried out in real complex biological samples [11]. Well over 700 papers have appeared in the literature, since 1997, on the development of nucleic acid biosensors. Almost all papers that have dealt with the DNA biosensor have used relatively short synthetic oligodeoxynucleotides for detecting target DNAs of about the same length [42]. Most reports have immobilised DNA in the form of a selfassembled monolayer onto a gold surface using thiol chemistry [11, 15, 41]. However, in some cases binding of the oligonucleotide probe to the sensing surface is achieved by using the biotin / avidin interaction [17]. In DNA sensors, the recognition is based on the formation of stable hydrogen bonds between the two nucleic acid strands. The bonding between nucleic acids takes place at regular (nucleotide) intervals along the length of the nucleic acid duplex [39]. The specificity of nucleic acid probes relies on the ability of different nucleotides to form bonds only with an appropriate counterpart. An important property of DNA is that the nucleic acid ligands can be denatured to reverse binding and then regenerated by controlling buffer-ion concentrations [39]. It is important to note that some workers have employed peptide nucleic acid as the biorecognition element [5]. The peptide nucleic acid is an artificial oligo-amide that is capable of binding very strongly to complementary oligonucleotide sequences [5]. 10

13 Many papers that have appeared in the literature on the DNA biosensor have used either optical [17, 43-45] or electrochemical [46-51] detection. Electrochemical transduction is becoming a popular method, and a recent review has shown that this method is ideal for studying DNA damage and interactions [40]. Recently, much of the work has focused on improving the detection methods for DNA hybridization [42]. However, further research is needed to develop methods for directly targeting natural DNA, which is present in organisms and in human blood [42]. Considerable work has been undertaken in improving the sensitivity of electrochemical DNA biosensors [16, 52-54]. In general, the base-pairing recognition event has been detected via electroactive or redox indicators such as metal coordination complexes or intercalating organic compounds [15, 16, 52-54]. Wong et al. (2004) recently studied the influence of various intercalators and found that the cationic ones (i.e., methylene blue, and rhodium metal complexes) undergo non-specific binding with the electrode surface [16]. It has been shown recently that DNA hybridization can be detected using a magnetically induced solid-state electrochemical sensor [55]. This process involves hybridization of a target oligonucleotide to probe-coated magnetic beads, followed by binding of a streptavidin-coated gold nanoparticle to the captured target [55]. 4.6 Electrochemical-Based Biosensors It has been stated that more than half of the biosensors reported in the literature are based on electrochemical transducers [56]. A recent survey of the literature has revealed that the electrochemical-based sensor platform is the most common and in many cases the most frequently cited in the literature [21, 24, 57, 58]. A review by Stefan et al. (2000) has revealed that electrochemical immunosensors are gradually increasing in popularity in clinical analysis and this is partly due to improved sensor design [28]. Similarly, Warsinke et al. (2000) demonstrated that the electrochemical immunosensor is a promising alternative compared to existing laboratory-based immunochemical assays [29]. Wang (2002) suggests in his review of nucleic acid biosensors that the electrochemical-based device will be responsible for achieving future large-scale genetic testing [59]. This may not be surprising considering that electrochemical transduction possesses the following advantages: low cost; high sensitivity; independence from solution turbidity; easily miniaturised / well suited to microfabrication; low power requirements; and relatively simple instrumentation [11, 29]. These characteristics make electrochemical transduction methods highly 11

14 compatible for implantable and/or portable hand-held devices. In general, there are several approaches that can be used to detect electrochemical changes during a biorecognition event and these can be classified as follows: amperometric; potentiometric; impedance; and conductometric Amperometric In the amperometric approach, the signal transduction process is accomplished by controlling the potential of the working electrode (i.e., usually an inert metal) at a fixed value relative to a reference electrode (usually silver / silver chloride), and monitoring the current as a function of time. The applied potential serves as the driving force for the electron transfer reaction, and the current produced is a direct measure of the rate of electron transfer. Amperometric biosensors take advantage of the fact that certain molecules can be oxidised or reduced at the working electrode (i.e., gold, carbon, platinum, etc). If the working electrode is driven to a positive potential an oxidation reaction occurs, and the current flow depends on the concentration of the electroactive species (analyte) diffusing to the surface of the working electrode. Similarly, if the working electrode is driven to a negative potential then a reduction reaction occurs. A third electrode called the counter (or auxiliary) electrode is often used to help measure the current flow. In most cases the bioreceptor molecule is immobilized on the working electrode, and as the analyte diffuses to the electrode surface the current generated reflects the reaction occurring between the bioreceptor molecule and analyte. A recent review by Habermuller et al. (2000) discusses various electron transfer mechanisms [8]. It is important to note that at least 150 articles have been published over the past decade on improving the electron transfer mechanism in amperometric biosensors [8]. The amperometric sensor for glucose is the most studied of all biosensors, noting that it employs an enzyme (glucose oxidase) to catalyse the conversion of glucose to gluconic acid [24, 60-64]. Similarly, the amperometric approach has become widely used for the detection of nucleic acid and antigens for disease identification / diagnosis [29, 54, 65, 66]. In fact, amperometric transduction is the most suitable and common electrochemical detection method in immunosensors [28]. Another important application of the amperometric biosensor has been in environmental monitoring of pesticides [67]. These biosensors are highly sensitive, rapid and inexpensive [39]. In addition, they display a high degree of reproducibility, which removes the need for 12

15 repeated calibration [22]. A possible limitation with amperometric transduction is the interferences that arise from electroactive compounds / species, and this can sometimes generate a false current reading [22]. However, these problems have been largely eliminated by the use of electrodes coated with various polymers [4, 68] Potentiometric In this method the analytical information is obtained by converting the biorecognition process into a potential signal. A permselective ion-conductive membrane is normally used to measure the potential signal, which occurs when the analyte molecule interacts with the surface. A high impedance voltmeter is used to measure the electrical potential difference or electromotive force (EMF) between two electrodes as shown in Figure 3, noting that potential measurements are made at near zero current. One of the electrodes develops a change in potential as a function of analyte activity or concentration in solution and this electrode is known as the indicator electrode or sometimes called an ion-selective electrode (ISE). The potential response of an ISE is described by the Nernst equation (i.e., the potential is proportional to the logarithm of the concentration of the substance being measured). The second electrode is the reference and is used to complete the electrochemical cell by providing a constant half-cell potential, which is independent of the analyte concentration. ISEs are chemical sensors with the longest history and with the largest number of applications [57, 58, 69]. In fact, billions of measurements are performed each year in nearly every hospital all over the world [69]. This comes as no surprise considering that these devices are well known for providing direct, rapid, maintenance-free and non-expensive measurements [22, 69]. Most of the work that is reported in the literature on the potentiometric sensor for antigen and DNA detection has employed the indirect approach. This involves measuring a change in either the ph or changes in the ionic concentration of an elemental species, which occur during a biorecognition event. A common strategy that has been employed is the use of enzymes to catalyse the consumption or production of protons and/or charged elemental species [22]. Uithoven et al. (2000) demonstrated that this detection platform can rapidly (<15 mins) monitor biological warfare (BW) agents using an enzyme-immunoassay approach [70]. In this process, a BW agent forms an immunocomplex with both a fluorescein-labelled antibody and a biotin-streptavidinlabelled antibody [70]. The fluorescein-labelled immunocomplex undergoes a further 13

16 complexation reaction with an anti-fluorescein urease conjugated antibody and the enzymatic breakdown of urea causes a change in ph, which is detected potentiometrically [70]. More importantly, the biosensor employs an eight-channel instrument, and has been designed to assay up to eight BW agents simultaneously [70]. A similar but much simpler approach developed recently by Purvis et al. (2003) involves the formation of an enzyme labeled immuno-complex at the surface of a polypyrrole-coated gold electrode [71]. Detection is achieved by a secondary reaction that produces charged products (i.e., changes in the redox state, ph and/or ionic strength), and the potential shift is measured at the sensor surface [71]. Figure 3 A typical experimental set-up using potentiometric-based sensing. Some reports suggest that potentiometric transducers cannot provide the required sensitivity for the detection of antibody-antigen reactions [28]. However, a light addressable potentiometric sensor (LAPS) based on field effect transistor (FET) technology has proved to be highly successful for immunoassay of various pathogens. A LAPS device consists of n-type silicon doped with phosphorus and an insulating layer. The FET is used to detect changes in the potential at the silicon-insulator surface [39, 72]. A LAPS measures an alternating photocurrent generated when a light source, such 14

17 as a light emitting diode (LED), flashes rapidly. Lee et al. (2000) developed a LAPS biosensor to detect Newcastle disease virus and report a detection limit of ~2 ng/ml [31]. Likewise, Ercole and coworkers (2003) used LAPS for the detection of Escherichia coli in various foods [73]. It is reported that Escherichia coli can be detected down to 10 cells/ml, which represents a significant improvement in sensitivity compared to conventional methods [73]. More recently, LAPS was used to identify the virus Venezuelan equine encephalitis [74]. In this study, an immunofiltration enzyme assay was used in conjunction with the LAPS device, and a limit of detection of ~30 ng/ml was achieved [74] Electrical Impedance Spectroscopy Electrochemical impedance spectroscopy (EIS) is a method that has only recently become a popular tool for bioreceptor transduction [75-79]. A review by Katz and Willner (2003), which has cited almost 200 references, suggests that this technique has played an important role in biosensor development over the past decade and will continue to play a significant role in the future [80]. Impedance spectroscopy has been widely used by many research groups to detect DNA hybridisation [81-86], antibodyantigen reactions [87, 88], and enzyme reactions [79, 80, 89]. In EIS measurements, a controlled AC electrical stimulus of between 5-10 mv is applied over a range of frequencies, and this causes a current to flow through the biosensor, which depends on various processes. During a biorecognition event the interfacial characteristics (i.e., capacitance and resistance) of the biosensor change, and the application of a voltage perturbation allows the interfacial capacitance and resistance to be evaluated. Typically, a conventional three-electrode system (i.e., counter, reference and the working electrode) is used to monitor the current variations, noting that a potentiostat / galvanostat and a lock-in-amplifier or a frequency response analyzer (FRA) are used in the detection process. The role of the lock-in-amplifier or FRA is to supply the excitation waveform over a range of frequencies and measure the AC current and voltage waveforms. A potentiostat is normally incorporated into to provide high input impedance, and is also used when precise control of the electrode potential is required. An important feature of EIS is that it is able to provide reagentless or label-free sensing, which makes it highly attractive for real-time monitoring. 15

18 4.6.4 Conductometric Conductometric-based biosensors harness the relationship between conductance and a biorecognition event. Most reactions involve a change in the ionic species concentration and this can lead to a change in the solution electrical conductivity or current flow [22]. Essentially, a conductometric biosensor consists of two metal electrodes (usually platinum or silver) separated by a certain distance. Normally an AC (alternating current) voltage is applied across the electrodes, which causes a current flow to be sustained between them. During a biorecognition event the ionic composition changes and an Ohmmeter (or multimeter) is used to measure the change in conductance between the metal electrodes. Some recent studies have shown that this technique is capable of rapidly detecting (<10 mins) various food borne pathogens (i.e., Escherichia coli O157:H7, Salmonella) [90-92]. Alocilja and coworkers used a conductive polyaniline label in the sandwich immunoassay scheme, which significantly improved the sensitivity via the formation of a conductive molecular bridge between the two electrodes [91, 92]. Unfortunately, one of the major issues with this technique is that the sensitivity is generally inferior compared to other electrochemical methods [22] Electrode Materials Gold and carbon are the most common materials used to carry current / charge during an electrochemical event [13]. However, the development of nanomaterials as electrodes for electrochemical-based detectors represents an exciting area of research. The ability of carbon-nanotube modified electrodes to promote electron-transfer reactions of important biomolecules has been recently reported by Wang and coworkers [93-95]. Carbon nanotubes represent a new class of materials, which are composed of graphitic carbon with one or several concentric tubules, and have shown promising results in DNA- and enzyme-based biosensors [83, 93-95]. Likewise, gold and alumina nanotubules have been exploited as membranes for biomimetic ion channels and sensing applications [96, 97]. Consequently, the development of nanomaterial sensors has the potential to revolutionise the bioanalytical, biomedical and pharmaceutical fields [97]. There has also been considerable interest in the development of biosensors that use conductive polymers (e.g. polyaniline, polypyrrole) as an electrochemical transducer [71, 91, 98, 99]. The conductive polymer is usually prepared by electropolymerisation 16

19 of the monomer onto a metal surface such as gold. The growth in the use of conductive polymers has primarily been stimulated by improved response characteristics such as increased sensitivity, stability, and reproducibility. The response characteristics (i.e., sensitivity, stability, reproducibility, etc) of polymer-based sensors are greatly dependent on the mode of polymerization, the monomer concentration, and the counterions used during polymerization [71]. By varying these parameters, it allows the surface properties of the biosensor to be modified, and this feature can be used to optimise the transduction signal. Screen printed electrodes have also attracted a great deal of attention recently [32, 46-49, 71, 94, 99, 100]. This technology, as shown in Figure 4, is a particularly attractive procedure for the mass production of disposable electrodes [23]. It is well known that memory effects and membrane fouling, which are sometimes observed with electrochemical-based biosensors can be alleviated when using disposable sensors. Disposable biosensors prepared by screen-printing technology are characterised by high reproducibility, low cost and require no calibration. More importantly, this technology has been widely used as a platform in DNA-, immuno and enzyme-based biosensors [23, 32, 46-48, 71, 94, 99, 100]. Figure 4 Disposable screen-printed electrodes Microelectrodes Microelectrodes represent a major area of biosensor research and development [101]. The use of microelectrodes offers many advantages such as imparting stir independent 17

20 response characteristics, lower limits of detection, and increased sensitivity [4, 102]. These advantages make microelectrodes very attractive for in vivo biosensor studies. Recently, Higson et al. (2004) developed a novel sonochemical approach in the fabrication of a microelectrode array enzyme-based glucose biosensor [103]. When a polymer-modified electrode is sonochemically ablated it exposes localized areas on the electrode surface, which act as individual microelectrodes and collectively as a microelectrode array [103]. It was shown that this approach generates a biosensor with significantly improved response characteristics [103]. 4.7 Optical-Based Biosensors Optical transducers represent another major family of biosensors that have been exploited commercially. Optical biosensors, which are sometimes called optodes, have received considerable interest for disease / pathogen detection. The optical biosensor format may involve direct detection of the analyte of interest or indirect detection through optically labeled probes. In general, there are at least four types of biosensors using the principles of optical technology. These are as follows: absorption / reflection; chemiluminescence; fluorescence; and phosphorescence [3, 6]. To a large extent, these all require some type of spectrophotometer to record the spectrochemical properties of the analyte [5]. It appears that sensors based on surface plasmon resonance and fluorescence principles are the two most common techniques of optical detection employed in biosensor studies [5]. Recently, sufficient progress has been made in fiberoptic technology, laser miniaturisation and the reproducible fabrication of prisms / waveguides that optical biosensors may become a powerful tool in the imminent future Absorption & Reflectance Spectroscopy When light (usually monochromatic) is passed through a sample several things can transpire. The light can either be reflected back or it can be transmitted through the sample. The process that occurs will depend on the wavelength of light, the composition (i.e., the type and concentration of molecules, etc) and thickness of the sample. The energy from the electromagnetic spectrum can be used to provide information about the changes in the local environment surrounding the analyte. Absorption spectroscopy is one technique that can be used to monitor the transmitted light intensity according to the Beer-Lambert Law. This is achieved by using a spectrophotometer to collect the absorption spectrum of the sample. The basic components of a spectrophotometer are as 18

21 follows: light source (i.e., deuterium laser, laser, etc); wavelength selector; radiation transducer; and signal processor / readout device. The wavelength selector comprises various slits, lenses, mirrors, windows, gratings or prisms to isolate the radiation of interest. The radiation transducer, which is usually a semiconductor material, converts the photon energy into an electrical signal. However, in the case of an infrared spectrometer (IR) a dielectric material is sometimes employed to convert the heat energy into an electrical signal. Infrared spectroscopy has recently become a popular tool for the detection of microorganisms such as Escherichia coli, Staphylococcus, Candida albicans, Mycobacterium fortuitum, etc [56, ]. An infrared spectrometer (IR) is able to provide a fingerprint of the microorganism by revealing changes in the molecular bond vibrations, noting that IR is based on the principles of absorption spectroscopy. Raman spectroscopy, another form of vibrational spectroscopy, has also been employed to identify a number of microorganisms [105, ]. In Raman spectroscopy the sample is irradiated with a powerful laser source (i.e., argon ion, helium / neon, etc) and the scattered radiation is measured by the spectrometer. Both infrared and Raman spectroscopy are able to provide rapid information on the molecular composition of a sample [104]; however, these techniques are not sensitive enough to directly detect pathogens in a complex matrix such as blood. In fact, the sample generally needs to be subjected to a lengthy concentration and/or culturing step before it can be measured by IR or Raman spectroscopy. Nonetheless, a recent article has appeared in the literature on the use of Raman spectroscopy as a glucose biosensor [111]. Alternatively, reflectance spectroscopy can be used, which measures the light reflected back from a surface. A commonly used method of reflectance spectroscopy for disease detection is surface plasmon resonance (SPR). This detector is able to detect subtle changes in the refractive index, which occur when cells binds to receptors immobilized on the transducer surface. SPR is a phenomenon that has been known for over 25 years and occurs during optical illumination of a metal surface (usually gold). This method is particularly attractive for direct label-free detection of pathogens and arises when light is reflected under certain conditions from a conducting film at the interface between two media of different refractive index. A thin layer of gold (~600 Å thick) is used to ensure that a quasi-electron cloud is established [56]. The generation of plasmons, which 19

22 represent the excited free electron portion of the surface metal layer, causes a reduction in the intensity of reflected light at a specific angle of reflection. An evanescent wave is created at the interface of two transparent materials, which have different refractive indices (or dielectric constants), and the coupling of the electromagnetic energy of this wave with a plasma wave travelling along the surface of a thin metal film. The penetration depth of the evanescent wave is roughly nm. Figure 5 shows a schematic diagram of a typical SPR detector. When molecules in the sample bind to the sensor surface, the refractive index at the surface changes and the SPR signal is monitored. Evanescent wave Antibody Y Y Y Y Y Y Antigen Gold layer Glass layer Prism Plane polarized monochromatic light Detector Photodiode Array Figure 5 A surface plasmon resonance biosensor. SPR is a simple technique that provides non-invasive real-time kinetic data on association and dissociation rates along with equilibrium binding constants for receptor / ligand systems [56]. It has the advantage that it can measure complex formation without labelling the reactants, and it can analyse samples from crude preparations [37]. However, some reports suggest that the SPR signal is very sensitive to non-specific physical binding on the surface, and this can severely limit this approach [56]. In addition, the technique is not sensitive enough when the molecular weight of the compound / biochemical species is less than 5000 Daltons. Notwithstanding, SPR 20

23 technology has been widely applied and embraced for immunochemical sensing in the environmental, pharmaceutical, food and medical fields [6, 112, 113] Similarly, a number of papers have shown that SPR is a powerful tool for pathogen / disease identification [17, 34, 36, 43, ]. Gomara et al. (2000) developed an immunosensor, which uses synthetic peptides for the detection of the hepatitis A virus in human serum [34], whereas Koubova et al. (2001) used an SPR-based immunosensor to detect various bacteria (i.e., Salmonella enteritidis and Listeria monocytogenes) [36]. More importantly, these reports suggest that the sensitivity of SPR is comparable to ELISA [34, 36]. Gomes et al. used SPR to study the effects of combining multiple amino acid replacements within the sequence of the antigenic GH loop of foot-andmouth disease virus [26]. Recently, Canziani and coworkers (2004) demonstrated that it is possible to screen many antibodies from hybridoma culture samples using the same SPR sensor surface [37]. Various other groups have developed SPR biosensors for the detection of DNA [17, 43] Chemiluminescence Luminescence is a term normally applied when discussing emission of radiation. It occurs when light is emitted from atoms or molecules due to an electronic transition from an excited state to a lower energy state or the ground state [3]. Chemiluminescence is a technique that can be used to detect specific biochemical reactions. It involves the generation of light as a result of a chemical reaction between the analyte and the chemiluminescent species (e.g. Rhodamine B). In fact, some chemical reactions can create an excited intermediate and in the process emit light, noting that no external light source is needed to instigate the reaction. A photomultiplier tube is used to detect the emitted light. Some reports suggest that the chemiluminescence transducer is the most suitable optical method for detecting antigen-antibody reactions [28], whereas recent studies by Chen and coworkers have used this transduction method to detect DNA from calf thymus and herring sperm [117, 118]. More importantly, it was revealed that chemiluminescence is a very sensitive method, noting that a detection limit of ~6.5x10 6 μg/ml 1 was obtained for calf thymus DNA and ~4.3x10 8 μg/ml 1 for herring sperm DNA [117, 118]. However, very few studies have been undertaken using chemiluminescence detection compared to other optical methods. In addition, this technique is time consuming and is not amenable for real time monitoring in blood. 21

24 4.7.3 Fluorescence & Phosphorescence Fluorescence is slightly different from chemiluminescence, in that it requires an external light source to initiate electronic transitions in an atom or molecule. In this process light is used to excite electrons to a higher energy state and as the electrons return to a lower energy level this causes light to be emitted at a longer wavelength. On a similar note, when a phosphorescent material is illuminated with light, some light is absorbed, exciting it into a higher energy state [3]. The energy is then transferred (normally called quenching) either by light emission or some other process. It is important to note that phosphorescence is generally a slower process and is not as popular compared to fluorescence detection. Since, neither antigens nor antibodies exhibit any fluorescence properties, a conjugate is used to generate a fluorescent signal. In fluorescent immunoassays, fluorochrome molecules are used to label immunoglobulins. The fluorochrome absorbs short-wavelength light and then emits light at a higher wavelength, which can be detected using fluorescent microscopy or a spectrophotometer (i.e., Fluorometer). Fluorescein isothiocyanate and rhodamine isothiocyanate-bovine serum albumin are the most common fluorochromes used to tag antibodies [39]. Fluorescence measurements are of particular interest in biochemical systems due to their high sensitivity. Today, most of the DNA chips are used with fluorescent markers [119]. A recent review has shown that this method is very appealing for nucleic acid detection [120]. Similarly, immunosensors that involve fluorescence transduction are probably one of the most sensitive methods of detection [28]. Rowe-Taitt and coworkers have developed a fluorescence-based multianalyte immunosensor array to simultaneously detect various microorganisms / toxins (i.e., Bacillus anthracis, Francisella tularensis, Cholera toxin, Ricin, Staphylococcal enterotoxin, Botulinum toxoids, etc) [121, 122]. It is reported that this technology exhibits comparable sensitivity to the standard ELISA method [121]. In addition, it has led to the fabrication of a portable biosensor, which essentially consists of a patterned array of biological recognition elements (i.e., antibodies, receptors) immobilised on the surface of a planar waveguide, and a fluorescence assay is performed on the patterned surface, which yields an array of fluorescent spots [123]. Signal transduction is achieved by using a diode laser for fluorescence excitation and a CCD camera to capture the image [123]. Recently, other workers have used the fluorescence detection platform for Bacillus anthracis and 22

25 Escherichia coli [124]. Lichlyter et al. (2003) developed a novel immunosensor technique based on fluorescence resonance energy transfer [125]. Moschou et al. (2004) report that fluorescence-based glucose detection is a growing class of sensors [126]. However, a major drawback of fluorescence technology is that it is relatively expensive, often gives rise to a reaction that is time consuming, exhibits lower selectivity and is not amenable for real time monitoring [28, 120] Optical Fibers The use of optical fibers is beginning to play an important role in optical detection, particularly in remote and real-time sensing applications [127]. Several articles have appeared in the literature reviewing the use of optical fibers in various biosensors [128, 129]. More importantly, it was highlighted that the detection limits of optical fiberbased biosensors are comparable to the sophisticated large bench-top instruments [129]. Recently, a fiber optic biosensor has been successfully used to detect Escherichia coli O157:H7 in food and water [130]. In this study, the polymerase chain reaction (PCR) and immunological technique were integrated into the biosensor [130]. In practice, fiber optics can be coupled with all optical techniques, thus increasing their versatility [22]. In the simplest form of measurement, the optical fibers are employed as waveguides to transport light to and from a solution to be analysed. It is important to note that the core of the optical fibre is usually silica, plastic or glass surrounded by an optical insulator (cladding), which has a lower refractive index than the core. In essence, total internal reflection is the basic mechanism that occurs in the optical fiber and this is achieved when the refractive index of the core is greater than that of the cladding and also when the incident light enters within the cone of acceptance (i.e., the critical angle for reflection at the core / cladding interface). Optical biosensors when used in conjunction with optical fibers have the following advantages over electrochemical-based biosensors: a) bioreceptor does not have to be in intimate contact with the optical fiber; b) no reference electrode is needed and by using multiwavelength measurements one can correct for any drift in the optical components; c) interferences from electrical noise does not occur; and d) they permit sample analysis to be done over long distances [56]. The main drawbacks are as follows: a) ambient light can cause high levels of background noise unless various techniques are used (e.g. lock-in-amplification); b) there are relatively expensive; c) they have a smaller dynamic 23

26 range; and d) these instruments are generally large and are not practical for on-site measurements; e) they have miniaturisation problems i.e., reducing the sample volume reduces the signal intensity and hence the sensitivity [22, 56]. 4.8 Piezoelectric-Based Biosensors These are mass sensitive detectors, which work on the principle that an oscillating crystal resonates at a fundamental frequency (sometimes called the natural resonance frequency). Piezoelectric materials have the ability to generate and transmit acoustic waves in a frequency-dependent manner. Quartz (i.e., SiO 2 ) is the most commonly used piezoelectric material; however, there is a report that lithium niobate (LiTaO 3 ) can also been used [131]. When a piezoelectric sensor surface, which has been coated with a biological substance (i.e., antibody), is placed in a solution containing the virus / bacteria, the attachment of the agent to the antibody coated surface results in an increase in the crystal mass, and this gives rise to a corresponding frequency shift. The physical dimensions and properties of the piezoelectric material influence the optimal resonant frequency for transmission of the acoustic wave. This transduction method is relatively easy to use, cost effective, and offers direct label-free analysis [131]. In addition, it is able to provide the option of several immunoassay formats for increased sensitivity and specificity [131]. Unfortunately, problems such as crystal regeneration, relatively long incubation times, lower selectivity, non-specific binding of proteins or other biomaterials, loss of material coating after washing, and difficulties in coating / immobilisation are well known limitations of this technique [28, 39, 131]. There is also some evidence that each crystal needs to be calibrated separately, since its frequency depends on the crystal geometry and the immobilisation technique [22]. It is important to note that the piezoelectric sensor has presently received much less attention for disease detection relative to electrochemical and optical-based biosensors. There are two main types of mass sensors: (a) bulk wave (BW) or quartz crystal microbalance (QCM) and (b) surface acoustic wave (SAW) [72]. BW or QCM devices consist of parallel circular electrodes placed on both sides of a thin cut piece of crystal. When an electric field is applied, it gives rise to a potential difference between the electrodes, and this causes shear deformation of the crystal [131]. The crystals are made to vibrate at the natural resonance frequency with the application of an electrical signal, and this generates an electric current. The frequency of oscillation is dependent on the 24

27 electrical frequency applied to the crystal, the physical properties of the crystal (i.e., size, density) and the properties of the crystal surface in contact with the solution phase. When the mass of the crystal increases as a material binds or becomes adsorbed to the surface it causes a corresponding change in the oscillation frequency, which is measured electrically [72]. The frequency shift can be related to changes in the reaction. In the case of the SAW, the sensor consists of a cut crystal with two electrodes on the same crystal face and the transducer can act as both the transmitter and receiver [131]. The excited wave travels across the crystal face, and this sensor has the ability to directly sense changes in the mass and mechanical properties. It is reported that the SAW sensor is more sensitive than the BW sensor [131]. The use of the piezoelectric biosensor format for the detection of various pathogens in food and environmental samples has been reviewed by various authors [39, 131]. Ivnitski et al. (1999) evaluated the limit of detection for various biosensor platforms and concluded that the piezoelectric method is inferior compared to the electrochemical and optical detectors [39]. However, a number of recent articles have appeared in the literature, which use the piezoelectric biosensor to detect hepatitis, the bacterium Escherichia coli, tumour necrosis factor-α, and Pseudomonas aeruginosa [ ]. It is apparent that these studies have overcome some of the challenges, which face this technique. In addition, it was shown that the sensitivity and reliability of the piezoelectric biosensor is comparable to the conventional ELISA method [135]. 4.9 Calorimetric-Based Biosensors Most chemical and biochemical processes involve the generation and absorption of heat. This heat change can be measured using either a thermistor or thermopile and is related to the amount of substance present. A thermistor consists of a metal oxide whereas a thermopile comprises a silicon-gold material. The device is coated with the bioreceptor (i.e., enzyme) and when this comes into contact with the analyte it generates an exothermic reaction, which is registered as a heat change [72]. A recent study by Zhang and Tadigadapa (2004) demonstrated that it is possible to detect enzymatic reactions of various molecules (i.e., glucose, urea) using a calorimetric-based biosensor [137]. An obvious advantage of this technology is that it can be used in turbid samples, can be easily miniaturised and is a label-free approach [7]. However, a recent survey of the 25

28 literature suggests that very little work has been undertaken using calorimetric transduction to detect DNA hybridisation and antibody-antigen reactions Others Although not technically a biosensor, the mass spectrometer (MS) has been extensively employed for pathogen identification and characterisation. A mass spectrometer is a very sensitive technique, which is based on the detection of molecules as a function of their molecular weight. A review by Mandrell and Wachtel (1999) revealed that pathogen identification in poultry using MS can be an attractive alternative compared to existing technologies [138]. A study by Fergenson et al. (2004) demonstrated that mass spectrometry can rapidly characterise (<1 min) in real-time two different species of Bacillus spores [139]. These workers performed a reagentless characterisation of the individual airborne cells without any sample preparation, and report that the method is accurate with no false positives [139]. In another study, Ruelle et al. (2004) used a timeof-flight mass spectrometer (TOFMS) to identify various bacterial strains (i.e., Escherichia coli, Salmonella, Acinetobacter) in wastewater samples [140]. It is important to note that mass spectrometer instruments are currently expensive, require skilled operators to run and maintain, and cannot be deployed as a hand held device. Nevertheless, they have the potential to be employed in the future in microbiology and clinical diagnostic laboratories Lab-on-a-Chip Devices There appears to be a lot of interest in the development of integrated biosensors for the detection of multiple biologically relevant species. A miniaturised biosensor device that comprises the probe, sampler, detector, amplifier and logic circuitry for monitoring infectious pathogens is an attractive alternative to existing instrumentation. Vo-Dinh and Cullum (2000) have reviewed some of the latest developments in this technology [5]. It was shown that a multifunctional biochip (MFB), which can simultaneously detect several diseases, might play an important role for point of care measurements [5]. Recently, Vo-Dinh and coworkers (2003) used the MFB to detect Bacillus anthracis and Escherichia coli, noting that a DNA probe specific to gene fragments of Bacillus anthracis and an antibody probe for Escherichia coli were employed [124]. It is important to note that MFB uses an optical detection platform [124]. By contrast, Wang et al. (2001) developed an electrochemical based microchip and used it detect the mouse 26

29 IgG antigen [65]. The antibody-antigen reaction, electrophoretic separation and electrochemical detection are all preformed in the microchip [65] In Vivo & Implantable Biosensors A survey of the patent and scientific literature revealed that most of the work on implantable biosensor technologies has been directed towards developing a long-term glucose sensor [24, 60, 63, 141]. The diagnosis and management of the worldwide health problem of diabetes has been the impetus behind the development of an implantable glucose biosensor [24, 142, 143]. Most of the implantable glucose biosensors presently available are mainly short-term and have an effective lifetime in blood of less than several weeks [63, 141]. Biofouling of the sensor membrane is still a major obstacle to the widespread application of implantable biosensors, noting that the sensor progressively loses function with time [144]. It has been shown that the accurate long-term usage of implanted sensors is limited by fibrosis formation that develops around the sensor and subsequently inhibits the influx of analyte to the detector [144]. Various methods have been developed to overcome biofouling problems [68]. Although none of these approaches appear to completely eliminate membrane biofouling, strategies based on biomimetics and surface perfusion technologies seem to be making the most progress in prolonging sensor functionality [68]. Recent efforts have been directed towards developing new ways of improving the biocompatibility of implantable biosensors [141]. Wang et al. (2000) demonstrated that depositing heparin onto the surface of an amperometric glucose biosensor can significantly improve sensor biocompatibility [62]. Work by Ward et al. (2003) has shown that vascularization of the foreign body capsule which surrounds a subcutaneous biosensor improves sensor life [145]. Recently, Collyer and coworkers have shown that biosensors coated with calix[4]resorcinarenetetrathiol may play an important role in suppressing electrode biofouling / passivation problems [146, 147]. Polymers have received a lot of attention in relation to chemical modification of electrode and sensor surfaces [4]. Brown and Lowry (2003) examined various Nafion coating procedures for in vivo measurements in the brain [148]. This study demonstrated that ascorbic and uric acid interferences can be removed by coating a microelectrochemical platinum sensor with Nafion [148]. By contrast, Meyerhoff and 27

30 coworkers have used a novel approach and tackled the biocompatibility problem by incorporating NO release polymers [i.e., polyurethane, poly(vinyl chloride) and polydimethylsiloxane doped with diazeniumdiolate functional groups] within the sensor membrane [ ]. It is important to note that nitric oxide (NO) plays an important role in suppressing platelet adhesion and thrombus formation [151, 152]. Upon contact with the sample, NO is released from the polymer and this is responsible for retarding platelet adhesion on the membrane surface. In vivo results obtained on an amperometric-based oxygen sensor revealed a significant improvement in the analytical data after coating with a diazeniumdiolate polymer [151]. Despite extensive research efforts, the use of implantable biosensors has not gained widespread clinical acceptance Some Commercial Biosensors There are well over 200 companies worldwide presently working in the area of biosensors and bioelectronics [155]. Due to the comparatively significant number of commercial biosensors, this report will not be able to give due credit to all the products that are commercially available. Some of these companies are directly involved in biosensor fabrication / marketing (will be discussed afterwards), whereas others play an important role in providing the necessary raw materials / reagents / instruments for biosensor production (e.g., Applied Enzyme Technologies, Biozyme Laboratories, Dupont Ltd, Eco Chemie, Ercon Incorporated, Gwent Electronic Materials Ltd, Palm Instruments, Uniscan Instruments Ltd, etc). Most of these companies are working on existing biosensor technologies that were developed over a decade ago [155]. Few of them are developing new technologies, although they appear to be improving existing technologies in order to move them into the commercial arena. Table 1 summarises some of the biosensor instruments made by various companies for the detection of bacteria. It is obvious that the bioluminescence method appears promising; however, this method works on the fact that all microorganisms, except for viruses, contain ATP, suggesting that there may be some limitations in this technique for disease detection. Commercial biosensors can be divided into two categories on the basis of whether they are laboratory or portable / field devices. The most successful handheld biosensor to date is the blood glucose monitor for people with diabetes, which is based on electrochemical transduction technology [24]. Commercial blood-glucose meters are produced by many companies [61]. However, in terms of laboratory-based 28

31 instrumentation an optical detection system appears to be more commercially viable. Companies such as Affymetrix [156] and Agilent [157] have developed various commercial microarray optical detectors and scanners for genomic and proteomic analysis. Optical sensors that employ surface plasmon resonance (SPR) detection have also been successfully used in many laboratories and universities [158]. In terms of implantable biosensors, several companies are investigating such systems, but are only looking at monitoring the level of glucose in the blood. The device developed by Medtronic Minimed, the diabetes management business of Medtronic [159], is a tiny enzyme-based sensor that is implanted under the skin for up to three days. By contrast, VeriChip are presently developing an implantable microprocessor [160]. Although a step in the right direction, these devices still suffer from the limitations, which have already been highlighted (see Section 4.12). Table 1. Some manufactures of commercial biosensor instruments for bacteria detection [39]. Detection Detection limit Commercial instrument Analysis time method (cells/ml) Midas Pro (Biosensori SpA, Milan, Italy) Amperometry min PZ 106 Immuno-biosensor System (Universal Sensors, New Orleans, USA) Piezoelectric min Bactometer (Bactomatic Inc., Princeton, USA) Impedimetry h Malthus 2000 (Malthus Inc., Stoke-on-Trent, UK) Conductance h Unilite (Biotrace, Bridgend, UK) Bioluminescence min Lumac Biocounter (Lumac B.V., Schesberg, Netherlands) Bioluminescence min Coulter counter (Coulter Electronics, Canada) Coulter counter 5x min Thermal activity monitor (Thermometric, Northwich, Cheshire, UK) Microcalorimetry h BIA-core (Pharmacia, Uppsala, Sweden) Surface Plasmon Resonance h Vitek AutoMicrobic System (BioMerieux Vitek, Hazelwood, MO) Optical h It is important to note that several multinational companies dominate the biosensor industry. Reports suggest that MediSense, Bayer and Roche Diagnostics (Roche- 29

32 Boehringer Mannheim) are the major players in terms of market hand-held meter style devices with disposable one-shot electrodes [143]. However, the recent merger of Therasense and i-stat with Abbott has significantly reinforced its position in the top three biosensor companies worldwide [143]. By contrast, commercially available optical bench-size immunosensor systems such as BIAcore (Biacore AB, Uppsala, Sweden) and IAsys (Affinity Sensors, Cambridge, UK) have found their market in research laboratories for the detection and evaluation of biomolecular interactions, noting that these technologies are based on the principles of surface plasmon resonance. Pharmacia Biosensor AB [161], now BIAcore AB, was the pioneer of the commercialbased SPR and currently holds approximately 90% of the market in this technology [158]. It is important to note that BIAcore offer a range of biosensors with various specifications, and a recent review suggests that BIAcore instruments are the most sensitive [131]. Unfortunately, most of the SPR instruments are relatively expensive and are not designed for studies in the field. However, Texas Instruments [162] have recently developed a low cost, rapid, and portable SPR-based biosensor (Spreeta TM ), which can be deployed in the field. Some reports suggest that this technology is not as sensitive compared to the standard enzyme-linked immunosorbent assay (ELISA) [163]. Nevertheless, there appears to be a general push for developing hand-held devices. The development of disposable sensors in conjunction with handheld devices for point of care measurements has featured prominently. Microfabrication technology has played an important part in achieving miniaturised biosensors. Such technology has provided cheap mass-producible and easy-to-use / disposable sensor strips. Similarly, electrochemical methods have played a pivotal role in detecting the changes that occur during a biorecognition event, and the merging of microfabrication with electrochemical detection has enabled various handheld biosensor devices to be developed. In fact, i-stat have developed the world s first hand-held device for point-of-care clinical assay of blood (see Figure 6), noting that this biosensor array employs several electrochemical-based transduction methods (i.e., potentiometric, amperometric, conductometric). i-stat Corporation is an international company that manufactures and markets diagnostic products for blood analysis [164]. The i-stat Portable Clinical Analyser is a hand-held silicon-based multiple-analyte sensor array [164], which is used to monitor various blood electrolytes (i.e., sodium, potassium, chloride, calcium, ph), gases (i.e., carbon dioxide, oxygen) and molecules (i.e., urea, glucose, hematocrit). 30

33 Oxford Biosensors have also developed a portable hand-held device (Multisense ) for cholesterol detection [165]. The biosensor consists of disposable test strips (microelectrodes) and uses the electrochemical detection strategy. Likewise, SenDx Medical Inc. (acquired by Radiometer) has fabricated a compact and portable potentiometric sensor array for determining various ions in the blood [166]. NH 4+ selective membrane immobilised urease ground AgCl Ag Cl Electrical contact K + reference Na + Urea Figure 6 The i-stat sensor array for monitoring various blood electrolytes, gases, and metabolites [164]. Various organisations have directed their efforts towards developing DNA chips and lab-on-chip devices. DiagnoSwiss is a company specialising in protein analysis on miniaturised platforms [167]. They have recently fabricated a lab-on-a-chip (or biochip) device for high-performance and high-throughput immunological testing. The unique feature of this device is that it incorporates both a micro-analytical system for separating various components along with a detection platform (electrochemical-based). GeneOhm Sciences have developed a DNA chip that employs the electrochemical detection platform [168]. Established in San Diego, California in 2001, GeneOhm Sciences is a company that focuses on molecular diagnostics for a wide range of diseases. Likewise, Motorola Life Sciences Inc. is another organisation that has produced an electrochemical-based DNA chip [169]. 31

34 A number of small companies appear to be making some progress in the development of various handheld devices. Chemel (Lund, Sweden) have developed a portable biosensor (SIRE biosensor) that is based on enzymatic / amperometric measuring principles [170]. This technology has only been used to measure various sugars and alcohols. Sensor Tech Ltd (Cambridge, UK) has recently developed an immunosensor (Universal Transducer System) for in vitro diagnostic and biosensor applications [171]. The Universal Transducer System (UTS ) employs a potentiometric detection platform and it is reported that this biosensor is rapid (<15 mins), stable (i.e., 4 months), reproducible (CV <5% at 0.1 ng/ml), and sensitive (~50 fm). Sensor Tech Ltd has taken a patent, which covers this platform technology [172]. The biosensor is fabricated using screenprinting technology and this allows a multiarray sensor to be developed. 5 Biosensor Activities Around the Globe Since there are well over 500 companies and research organisations worldwide involved in biosensor development, the authors have decided to provide a selected review of some of them, noting that several of the chosen organisations have similar objectives and interests to the AB-CRC (i.e., Lawrence Livermore National Laboratory, The US Naval Research Laboratory, Oak Ridge National Laboratory). It is important to note that the major biosensor players were identified in terms of the following criteria: publication output / quality, scientific impact, reputation, and size. 5.1 Universities & Laboratories Cranfield University (Silsoe, England) The Institute of Bioscience and Technology (IBST) at Cranfield University ( is a recognised authority on the research and development of biosensors [173]. Some of the scientists working at IBST were responsible for the development of the world s most successful biosensor to date, the Medisense TM home blood glucose monitor. Microelectrodes for the fabrication of electrochemical sensors and biosensors are another major area of expertise at Cranfield. Microelectrode arrays are normally produced by photolithography and laser ablation techniques; however, at Cranfield a new and patented approach for the manufacture of microelectrode arrays has been developed. Consequently, ultrasonic sensor electropolymerisation and fabrication facilities allows the processing of >250,000 sensor strips per day. Figure 7 shows the ultrasonic ablation set-up used at IBST. Some 32

35 biosensor research activities currently being undertaken at IBST are as follows: a) design and fabrication of diagnostic devices; b) the development of enzyme electrodes; c) biomimetic sensors; d) development of immunosensors; e) DNA chips; and f) detection of infection and infectious agents. Some recent work emerging out of IBST is the development of a hand-held electrochemical-based biosensor for the trace level detection of pesticides. Other developments taking place at IBST involve the collaboration with Pelikan Technologies in Palo Alto (USA) to develop a painless and convenient system for blood sampling and glucose measurement device. Together they have created the first, fully integrated, blood sampling and glucose measurement device. Figure 7 Ultrasonic ablation process used at IBST Lawrence Livermore National Laboratory (Livermore, USA) Lawrence Livermore National Laboratory (LLNL) is managed by the University of California for the Department of Energy s National Nuclear Security Administration, and is committed to addressing the technological challenges facing the United States in relation to biosecurity threats. The Chemistry and Materials Science (CMS) directorate comprises the Biosecurity and Nanoscience Laboratory (BSNL), which conduct scientific research into biological threats and natural-disease outbreaks [174]. BSNL scientists are presently developing various biosensors and sensing materials for emerging infectious diseases ( The following is a list of some projects currently being undertaken at BSNL a) the development of carbon nanotube arrays for the detection of anthrax / botulinum toxin; b) creating and designing synthetic high-affinity ligands (SHALs) for the detection of toxins and 33

36 pathogen. SHALs act as reagents to detect toxins / pathogens, noting that these molecules possess some of the traits of a natural antibody and can bind to a protein with high affinity and high specificity; c) development of a Bio-Aerosol Mass Spectrometer (BAMS) for the characterisation of airborne / bioaerosol toxic particles. Figure 8 shows the BAMS instrument developed at LLNL and it is suggested that individual airborne particles can be identified at the single-cell level in ~100 milliseconds; d) developing optical microscopy and micro-raman spectroscopy methods to detect individual toxin molecules; e) synthesising smart membranes of various pore sizes (diameters range from 300nm to 2.5μm) to identify pathogens; and f) developing methods (i.e., atomic force microscopy) for finding the biological signatures of spores, viruses, and bacteria. Other potentially ground-breaking work at BSNL is the development of a new immobilisation technique that allows covalent anchoring of enzymes on an active silicon platform while retaining enzymatic activity. This research paves the way for a new class of extremely compact and inexpensive biosensors based on silicon technology. Figure 8 The Bio-Aerosol Mass Spectrometer (BAMS) developed at Lawrence Livermore National Laboratory [174] Lund University (Sweden) Prof Lo Gorton heads the biosensors and bioelectrochemistry group in the Division of Analytical Chemistry at Lund University ( Research in the 34

37 Division of Analytical Chemistry is directed mainly at combining traditional and modern aspects of analytical methodology with bioanalytical tools [175]. More specifically, Prof Gorton and his research group are primarily studying electron-transfer (ET) reactions of redox proteins and enzymes for biosensor applications. Current research activities are focused on investigating heterogeneous and intraprotein ET of ligninolytic enzymes for the development of amperometric biosensors for the detection of hydroperoxides, carbohydrates, phenols, and catecholamines. Studies of the heterogeneous redox reactions of NAD+/NADH at chemically modified electrodes constitute one of the main research activities of the group. Development of detection techniques based on spectroelectrochemical methods is another emerging area of bioelectrochemical research in the group. The group have also undertaken extensive work on the development of biosensors for on-line analysis of fermentation processes / systems. The research is focussed on developing autoclavable microdialysis probes. The group is also involved in the development of analytical methods for the determination / characterisation of chemically modified starch and cellulose Naval Research Laboratory (Washington, USA) The US Naval Research Laboratory (NRL) conducts a broad based multidisciplinary program of scientific research and technological development directed toward maritime applications of new and improved materials, techniques, equipment, and systems [176]. A majority of the biosensor developmental work is undertaken at the Centre for Bio-Molecular Science & Engineering (CBMSE) ( CBMSE comprises a wide variety of professionals (e.g., biochemists, biophysicists, cell / molecular biologists, chemists, engineers, physicists, etc) working on various projects that deal with the development of biomimetic materials, bioengineered systems, biosensors, responsive materials, tissue engineering, proteogenomics, microfluidics, etc. Since the early 1990s, NRL has been working to develop field-portable biosensors for environmental monitoring. In particular, Dr Frances Ligler and her research group have developed an array biosensor that uses a fluorescence-laser transduction system to detect infectious diseases and toxins in clinical fluids (see Figure 9). It consists of a patterned array of biological recognition elements (i.e., antibodies) immobilised on the surface of a planar waveguide, and a fluorescence array is performed on the patterned surface, which yields an array of fluorescent spots. Signal transduction is achieved by 35

38 using a diode laser for fluorescence excitation and a CCD camera to capture the image. The biosensor is a portable device and has been used for the detection of proteins, toxins, bacteria, and viruses in a variety of physiological and environmental matrices. Another project currently being undertaken at CBMSE involves the development of a portable fiber optic biosensor that is capable of simultaneously testing a number of different agents and biological threats (i.e., Staphylococcal enterotoxin type-b, Francisella tularensis, Bacillus anthracis, and Bacillus globigii spores) in ten minutes or less. This work led by Dr George Anderson has resulted in a commercial device known as the RAPTOR bio-agent detector. Similarly, work by Dr Anne Kusterbeck and her group have developed a portable flow immunosensor for the quantitative analysis of explosives in contaminated soil samples. The method is based on a displacement immunoassay performed in a commercial instrument, the FAST Figure 9 Optical (A) and fluidic (B) components of an array sensor developed at NRL [121] New Mexico State University (Las Cruces, USA) In the department of chemistry and biochemistry at New Mexico State University ( Professor Joseph Wang (has just recently moved to Arizona State University, Tempe, USA) and his research group (comprises ~12 postdoctoral staff and PhD students) are responsible for performing a majority of the biosensor research and development [177]. Most of the research effort is directed towards nanobioelectronics in which nanomaterials are applied to the analysis of biomolecules. A number of research projects currently being performed in the group include the following: a) development of nucleic acid and protein biosensors; b) implantable in vivo glucose biosensor c) the development of miniaturized analytical systems, micro separation chips and microfluidic devices (see 36

39 Figure 10); d) designing new detection schemes for microchip electrophoresis; e) nanoparticle-based bioassays; f) micro fabricated enzyme electrodes; g) remote sensors for environmental monitoring and security surveillance; h) biorecognition-induced formation of nanomaterials; i) designing nanomaterials for electrical assays; j) developing techniques for ultra-trace measurements. Some recent cutting-edge science emerging from Wang s group is the use of carbon nanotubes (CNTs) for dramatically amplifying the recognition and transduction events of enzyme-based biosensors. Other ground-breaking work has shown that it is possible to develop a microfabricated device, which integrates the multiple steps of electrochemical enzyme immunoassay on a chip platform. Figure 10 Lab-on-a-chip system developed at New Mexico State University [177] Oak Ridge National Laboratory (USA) Oak Ridge National Laboratory (ORNL) ( is another organisation devoted to biosensor / sensor research and development [178]. In the Life Sciences Division (LSD) at ORNL their objective is to advance science and technology in order to understand complex biological systems and its relationship with human health and the environment. LSD has a range of expertise and facilities in mammalian genetics / genomics, computational biology, biophysics / biomedical technologies, and toxicology / risk analysis. The programs are supported by a number of federal and nonfederal agencies and institutions, such as the Department of Energy, the National Institutes of Health, the Environmental Protection Agency, and several industrial 37

40 partners. The Advanced Biomedical Science and Technology Group (ABSTG), which is led by Dr. Tuan Vo-Dinh is responsible for a large portion of biosensor research at ORNL. The major areas of research at ABSTG are: a) the development of biochips for the diagnosis of genetic susceptibility, diseases and pathogens; b) cancer diagnostics using laser-induced fluorescence (LIF). A fiberoptic probe can be inserted through an endoscope and used to directly measure autofluorescence of normal and malignant tissues; c) nano-biosensors for exploring the sanctuary of individual living cells. There are several reports on optical nano-biosensors for in situ intracellular measurements of single cells using antibody-based nano-probes; d) the development of nanoparticles for labelling of biomolecules; e) the development of a near-field scanning optical microscope (NSOM) for non-destructive imaging of biomolecules in nanoscale domains is of considerable interest; and f) investigating novel alternative techniques such as surface-enhanced Raman scattering (SERS) for interrogating biological / biochemical materials. Another area of interest at ORNL is the development of medical telesensors, which are being used by military troops in combat zones to provide information on vital functions to remote recorders. An array of chips has been developed to collectively monitor bodily functions. These chips are attached at various points on a person using a nonirritating adhesive and can send physiological data by wireless transmission to an intelligent monitor on another person s helmet. The monitor can also be used to receive and transmit global satellite positioning data to help medical staff locate a wounded soldier. Figure 11 shows a 2 2-millimeter silicon chip attached to the skin. The chip contains a temperature sensor in an integrated circuit, a lithium thin-film battery that supplies the very low level of power required by the circuit and signal processing and transmission electronics, and an antenna that sends the data by radio signals (radiofrequency transmission) to a monitor when the chip is queried. The development of miniaturised devices is an area of considerable interest at ORNL's Chemical Technology and Engineering Technology Divisions. ORNL researchers have made dramatic progress in miniaturising laboratory-based scientific instruments. One useful miniature device developed at ORNL's Engineering Technology Division is the infrared micro-spectrometer (see Figure 12). The device can be used for a wide array of applications ranging from blood chemistry analysis, gasoline octane analysis, 38

41 environmental monitoring, industrial process control, aircraft corrosion monitoring, and detection of chemical warfare agents. The plastic device uses a light source to excite certain types of compounds and these excited compounds give off infrared light of various wavelengths. Figure 11 An image of a typical medical telesensor developed at ORNL [179]. Figure 12 The infrared microspectrometer developed at ORNL [179]. A new class of biosensors is also being developed at ORNL's Chemical Technology Division, which transforms a biological system into a tiny electronic device. These biomolecular optoelectronic sensors are being used to analyze biological / physiological processes and to detect and identify bacteria. 39