CHAPTER 1: Introduction to Food Biosensors

Food quality and safety surveillance is an important global issue. Monitoring the safety and quality level of foods is of critical importance to ensure that food reaching consumers is safe to eat. Food is often preserved at a desired level of characteristics and qualities to ensure its beneficial properties are maintained.¹  Since overall food quality and safety are also determined by proper handling, preparation and storage it is therefore imperative to fully understand the effect of food preservation methods and each step involved in food processing.²  Information about food contents can also be found on food packaging to provide assurance to consumers that food has been tested and is free of harmful and undesirable substances. If they fail to follow international food guidelines, manufacturers can face serious legal actions with economic consequences.³  Currently, traditional analytical techniques such as high-performance liquid chromatography and gas chromatography are well accepted and taken as gold standards for food quality and safety monitoring, but these conventional procedures are cumbersome and time consuming, requiring expensive instrumentation and skilled operators. Alternatively, biosensors can provide an invaluable method for agro-food diagnostics since they are convenient, portable and do not need particular skills to operate.^1,4–9  Biosensors in the food industry may be used to analyze nutrients, to detect natural toxins and antinutrients, for monitoring of food processing, and for detection of genetically modified organisms. Through enzymatic and immunogenic reactions, biosensors can be used to determine the level of pesticides, antibiotics, proteins, vitamins B complex and fatty acids found in foods.¹⁰  Figure 1.1 depicts the different applications of biosensors used in food industries. As shown in Figures 1.2 and 1.3, research and development in food biosensors have been remarkably rapid in the past decade and are set to expand further with the advances in material science and biotechnology.^11–14  The various types of instruments required for the agro-food diagnostics market can be mainly categorized as large multi-analyzers, bench-top portable instruments and single-use disposable sensors. Table 1.1 lists the desirable characteristics for biosensor commercialization.

A biosensor is a measurement device constituting of a biological element that functions as a target recognition entity, in conjunction with a transducer that converts a biological recognition episode to a measurable signal (Figure 1.4). A biosensor setup is basically comprised of three parts:

• a biological receptor for biomolecular recognition of sample analyte

• a transducer to translate recognition event into an appropriate signal

• a detection technique for signal analysis and processes.¹⁶ 

Typical biological recognition elements may be tissue, living cells, enzymes, antibody or antigen. Signal transduction can be in the format of electric current, electric potential, intensity and phase of electromagnetic radiation, mass, conductance, impedance, magnetic, temperature and viscosity, or a combination of these techniques.¹⁷  A biosensor’s experimental performance is based on its sensitivity, limit of detection, linear range, reproducibility, selectivity, interference response, response time, easy operation, portability, storage, and operational stability.¹⁸  This introductory chapter aims to provide readers with an overview of different types of biosensor with sections dedicated to discussions of the important components of biosensors.

Biosensors can be categorized either on the basis of their biological receptors (e.g., aptasensor or immunosensor), their signal transduction mechanism (e.g., amperometric or optical sensors), or their final application (e.g., clinical sensors).

Biomolecular recognition is essential in biosensor application. In early biosensor development, recognition receptors were primarily obtained from living organisms. Now, with the emergence of modern recombinant techniques, the engineering and manipulation of synthetic receptors in the laboratory have opened up endless possibilities for biosensing design beyond what is possible in nature.

Receptor-based biosensors are categorized into two types: biocatalytic and bioaffinity-based biosensors. A biocatalytic biosensor predominantly uses enzymes as the biological mediator that catalyzes a signalling biochemical reaction, whereas a bioaffinity-based biosensor monitors the binding episode itself. In the latter type, biomolecular recognition uses specific binding proteins, lectins, receptors, nucleic acids, membranes, whole cells, antibodies, or antibody-related substances.¹⁹ 

The unique attributes of enzymes, with their ability to specifically recognize their substrates and catalyze their transformation, makes enzymes a powerful tool for use in analytical devices. This distinctive characteristic is due to the complementary structures of the substrate and its binding site on the enzyme, which is the particular region of the enzyme where enzyme–substrate interactions occur. The concentration of target analyte can then be determined by measuring the catalytic transformation of the analyte by the enzyme. Conversely, an enzyme can also be inhibited by its target analyte and this can be used to determine the concentration of target analyte in correlation with a decrease in enzymatic product formed.²⁰  Not only are enzymatic reactions substrate-specific but they are also product-specific, in contrast to uncatalyzed reactions or reactions catalyzed by chemical catalysts which often generate by-products.²¹  Experimental variables such as substrate concentration, temperature, pH, ionic strength, and the availability of a competitive or non-competitive inhibitor can influence the catalytic performance of an enzyme²²  and consequently its stability.²³  Although some enzymes are stable in their native state when implemented in different environments, most enzymes are less stable and this hampers their effective application as biomolecular recognition units in biosensors. This instability is caused by changes in an enzyme’s three-dimensional configuration, which in effect disfigures the enzyme’s active site due to the disruption of non-covalent bonds holding the native protein in place. The main contributing factors determining the conformation stability of the enzyme are hydrophobic interactions, intrapeptide hydrogen bonds, and the ability to recover the original conformation during dehydration–hydration activities. These three interactions need to be preserved for an enzyme to play a roles as biological recognition tool under harsh operating conditions.²³ 

Another important issue related to enzyme-based immunosensors is immobilization of the enzyme on the transducer surface. The selection of immobilization strategy must be based on a number of factors:

• The enzyme has to be stable during the reaction.

• The reagents forming crosslinking bonds should not reach the enzyme’s active center, which must be protected.

• The washing of unbound enzyme must not be detrimental to the immobilized enzyme.

• The mechanical properties of the carrier should be considered.²¹ 

However, enzymes are now often used for labeling purposes than as actual bioreceptors, especially with the massive improvement in enzyme-labeling techniques during the past decade.²⁴ 

Antibodies, which can be polyclonal or monoclonal, are regarded as the prime choice for use in the biomolecular recognition component of biosensors, due to their target specificity and affinity.²⁵  Polyclonal antibodies derived from the serum of an immunized animal form an array of molecular populations (each arising from a separate cell line) that recognize various regions (haptens) on the immunogen.²⁶  Monoclonal antibodies of predetermined specificity, derived from cell hybrids made by fusing normal spleen cells with malignantly transformed antibody-secreting cells,²⁷  typically recognize a more specific region of the immunogen than their polyclonal counterparts. The use of monoclonal antibodies is more common in immunosensor studies because they are quite homogeneous in their molecular structure, have similar binding characteristics, and can be produced in large quantities.²⁸  Monoclonal antibodies also eliminate the problem associated with the density of binding sites that can be immobilized on the surface of the signal transducer due to the absence of serum proteins and other non-analyte-specific antibodies.²⁶ 

Using antibodies as bioreceptors has its own limitations including high cost, limited lifespan, and susceptibility to high temperature. To improve the robustness of antibodies for biosensor applications, antibodies composed of only heavy chains with very small antigen binding sites, obtained from sharks and camelids, have been explored. These single domains are stable at high temperature (up to 90 °C), stable to detergents, and very soluble.²⁹  Another advantage of using antibody fragments is that they can be tailored and engineered to further improve their affinity and stability to environment stress.

Nucleic acids are employed as bioreceptors through the immobilization of a single-stranded oligonucleotide onto a transducer surface to detect its complementary target sequence. The DNA hybridization event is then translated into a signal.³⁰  The overall performance of a DNA-based biosensor predominantly depends on the immobilization of nucleic acid with the receptors oriented so that they can be readily accessed by the target. A DNA probe can be attachment to a transducer surface using various strategies to ensure optimal probe orientation for the target recognition event. These schemes include straightforward adsorption on carbon surfaces, thiolated DNA to form a self-assembly monolayer on a gold surface, the use of functional alkanethiol-based monolayers for covalent attachment to a gold surface, biotylated DNA coupling with avidin or strepvadin, and carbodiimide attachment to functional groups on carbon electrodes.

The introduction of engineered peptide nucleic acid has provided a new direction in nucleic acid recognition, contributing to impressive sequence specificity for DNA biosensors. Tree-like DNA dendrimers consist of many single-stranded branches that can hybridize to their complementary DNA sequence and this can also be immobilized on a transducer to achieve higher sensitivity. A greatly increased hybridization capacity and hence a substantially amplified response is achieved by immobilizing these dendritic nucleic acids onto the physical transducer.³¹ 

Whole cells, developed from bacterial strains,²⁰  can be used as the molecular biorecognition constituent in a biosensor due to the ability of microorganisms to identify and respond to a number of stimuli.³²  Microorganisms recognize and absorb the analyte of interest and this either increases or inhibits their respiratory activities. Products such as protons and ammonia liberated during metabolic processes can then be detected and converted to an electronic signal by a transducer.³³  Whole-cell based biosensors can also be used to observe changes in the vicinity of the cells by examining their electrical properties, thus making this type of biosensor a reliable tool for the detection of pathogens in food samples.³⁴ 

Immobilization is a crucial in whole-cell biosensor fabrication. Different methods of immobilization protocols such as adsorption, encapsulation, entrapment, covalent interaction, and crosslinking can be employed.³²  There are, however, a number of problems associated with the current immobilization techniques, for example:

• Cell viability and function are affected significantly by covalent binding and crosslinking.

• Physical adsorption can cause desorption of microbial cells.

• Immobilization via entrapment may suffer from poor sensitivity as a result of the extra diffusion resistance caused by the entrapment material.³⁴ 

Microbes have the advantages of long lifetime, cost effectiveness, and being able to work in a wide range of pH and temperature. Nevertheless whole cell-based sensors are not a popular choice in this market since microbial cells are complex, have long response times due to diffusional problems, are less sensitive, show poor detection limits and poor selectivity in multiplex detection, non-homogenous intrinsic cellular genotype and phenotype, and stochastic protein expression.^35,36  This situation has slowly started to change due to the rapid development of recombinant DNA technology, to improve the activity of whole cells, allowing customization and tailoring of microorganisms for detection of particular targets. This makes microbes an excellent source to consume or degrade new substrates under certain cultivation conditions.^34,37 

Despite the excellent affinity and selectivity displayed by natural bioreceptors, e.g., enzymes and antibodies, their use in biosensor applications remains a challenge. For instance, in antibody production variations can occur in the quality and concentration of the antibodies in each batch.^38–40  Other obstacles to the use of antibodies include high cost, stability, renewability, and immunity against low molecular weight chemicals and toxins.⁴¹  A study has shown that only 49% (2726 out of 5436) of commercial antibodies obtained from animals could be validated to recognize only their targets. Additionally, a considerable amount of money has been spent on protein-binding reagents for use on antibodies, where these antibodies themselves may be the laboratory tool most typically causing irreproducible research.⁴²  Thus, new types of replacement for antibody–antigen binding molecules that mimic natural bioreceptors^43,44  have been developed in recent years to overcome the drawback of these bioreceptors in biosensor applications. The most common synthetic ligands are aptamers, i.e. single-stranded DNA or RNA oligonucleotides that fold into a three-dimensional secondary structure with binding affinity for a target molecule.^45,46  Other synthetic recognition ligands include molecularly imprinted polymers (MIPs), scaffolded peptides, combined binding agents derived from low-affinity ligands, and combinatorial chemistry ligands. With the exception of aptamers, many of these synthetic ligands have a major drawback, which is their low K_D (∼10⁻⁶ to 10⁻⁷ mol L⁻¹), 2–3 orders of magnitude lower than that of antibodies.

Aptamers, also known as “chemical antibodies,” are acquired from screening large combinatorial libraries through an in vitro selection and amplification process known as systematic evolution of ligands through exponential enrichment (SELEX), which has allowed the isolation of oligonucleotide sequences capable of recognizing targeted antigens with high affinity and specificity.^46–48  The basic principle of SELEX is to simulate (and stimulate) evolution (systematic evolution of ligands). Evolution in nature is a slow and complex process, but it can be artificially speeded up by the incorporation of a number of selection rounds (usually 10–15 rounds) and exponential amplification of ligands by polymerase chain reaction (PCR) at each round of selection (exponential enrichment).⁴⁹  Unlike antibodies, nucleic acid-based aptamers are able to withstand harsh operational conditions due to the comparatively rigid backbones and limited flexibility of nucleic acids in comparison to proteins, which have more torsional freedom and multiple conformational states with respect to their backbones and side chains.^47,50 

The outstanding molecular recognition of aptamers is due to their ability to adopt specific and complex three-dimensional shapes characterized as stems, loops, bulges, hairpins, pseudoknots, triplexes, or quadruplexes. These three-dimensional configurations allow the binding of targets ranging from small molecules to large ones such as peptides.⁵¹  Another advantage of nucleic acid aptamers is their excellent affinities for their targets, typically with dissociation constants ranging from picomolar to millimolar. Figure 1.5 shows the different recognition schemes for aptasensing according to their level of integration.⁵² 

The structural folding and stability of peptides has been extensively studied,^53–55  thus making peptides a particularly attractive alternative in biosensors. In the early days, the study of peptide receptors was a formidable prospect⁵⁶  and it still remains a challenge. This is due to our ever-changing understanding of the underlying molecular recognitions and the interactions involved in these systems. The use of peptides as bioreceptors for the detection of antigens is a relatively new area, so little information is available in the literature.

The chemical synthesis of peptides can be implemented more rapidly than that of antibodies, with higher stability (without denaturation).⁵⁷  Combinatorial chemistry, phage display technology, rational design approach using computational methods, and molecular imprinting are employed to design a peptide targeting an analyte with high affinity.⁵⁸ 

Combinatorial chemistry is an innovative way of rapidly generating a large number of related compounds, which was first reported from the Eötvös Loránd University in Budapest at international scientific meetings in Prague and Budapest in 1988.⁵⁹  The basic concept of combinatorial chemistry involves simultaneous building of many blocks in one synthesis step, forming all possible combinations of building blocks, with each peptide receptor being evaluated against the targeted antigen. The most active compound is then identified for further development by high-throughput screening.⁶⁰ 

Phage display technology, introduced in 1985,⁶¹  is another effective technique to determine target-specific molecular interactions by exploring in vitro discriminative affinity selection.⁶²  Phage display allows the production of large numbers of various peptides and proteins with the function-specific molecules being selected and isolated.⁶¹  The use of phage display provides a promising approach in biosensor applications to recognize a broad range of target analyte that cannot be achieved by natural receptors. Using this technique, antibodies with the desired properties can be designed and produced where a library of single-chain antibody fragments is displayed on the surface of filamentous phages and subsequently the required recombinant phages are selected through specific binding to target antigen. A number of selection strategies offered by phage display can improve antibody affinity and selectivity by means of a given set of clones with predetermined affinity and selectivity profiles. Clones with optimal attributes can then be isolated from the existing library of clones, thus providing biorecognition elements with high affinity towards their target and other desired properties.⁶³ 

One limitation of the phage selection process is that immobilization of targets is required, and for small compounds specific functionalization is necessary.⁶⁴  Nevertheless, phage display technology opens up a promising prospect in the selection of recognition elements for reliable and robust biosensors. Due to the high stability of phages in fluctuating environmental conditions, phages are suitable for use in sensors for food monitoring purposes.

Molecular Imprinted Polymers (MIPs) are able to mimic natural molecular receptors (e.g., antibodies) and selectively recognize targets using imprinting technology. MIPs are obtained by crosslinking large amounts of monomers using a “print” molecule (template).⁶⁵  Once the polymerization process is completed, the template is then removed, leaving specific cavities that can discriminately re-bind template molecules depending on their structure and functionality. This scheme is depicted in Figure 1.5. In comparison to natural receptors, MIPs have the advantages of stability at extreme pH and temperature, easy preparation, low cost, and reusability for biosensor applications.⁶⁶  Polymerization between the template and monomers can be realized via self-assembly or preorganized strategies. Self-assembly uses non-covalent interactions, for instance hydrogen bond and Van der Waals forces, similar to those found in natural recognition processes. In preorganized strategies, covalent reversible bonds are used resulting in homogenous binding sites and less non-specific sites. In the latter approach, removal of the template from the matrix requires cleavage of covalent bonds.⁶⁷ 

The transducer is an integral part of a biosensor. The past decades have seen the development of a number of detection methods based on a variety of transduction modes.

The detection principles of electrochemical techniques can be divided into amperometry, potentiometry, voltammetry, and impedimetry. Electrochemical detection presents the advantages of low cost, ease of miniaturization, portability, ease of construction, and ability to work with turbid samples. It can be coupled with other detection techniques for enhanced performance.⁶⁸ 

Amperometry involves the generation of current at a given constant potential between the working electrode and reference electrode. Signal current output is produced as a result of the reduction or oxidation of an electroactive product at the working electrode. Current detected and displayed correlates with concentration of electroactive species in the sample.³⁴  An amperometric biosensor uses the oxidation or reduction potential that is characteristic to the analyte, which makes this type of transducer highly selective.¹⁸ 

In voltammetry, current and potential are measured and recorded⁶⁹  where the potential is being scanned over a set potential range. The current output is translated into a peak or plateau where the position of peak current corresponds to a specific analyte and the peak height is proportional to concentration of the analyte. Voltammetric detection has the ability for simultaneous detection of multiple analytes, with different characteristic peak positions, in a single experiment. Also, the low background signal and effective preconcentration of sample (electrochemical stripping analysis) make this technique a sensitive electroanalytical method.³² 

Potentiometry measures the ion activity for a given electrochemical reaction. The potentiometric measurement determines the accumulated charge potential at the working electrode versus the reference electrode in an electrochemical cell at zero current. The relationship between concentration and potential is dictated by the Nernst equation, where E_cell represents the observed cell potential when current is zero.²⁴  pH electrodes are the most typical potentiometric instruments, although other ions (F⁻, I⁻, CN⁻, Na⁺, K⁺, Ca²⁺, NH₄⁺) or gas (CO₂, NH₃) selective electrodes are also available.¹⁵  Although potentiometry has the advantage of concentration range measureable over several orders of magnitude, potentiometric biosensors for food diagnostics have not been extensively reported.⁷⁰ 

Electrochemical impedance technique (EIS) is a powerful electrochemical procedure for measuring the electrical resistance and capacitance of materials in response to changes of surface properties by applying a small-amplitude sinusoidal voltage signal as a function of frequency.³⁴  In general, impedance detection techniques can be divided into two types according to the presence or absence of specific biorecognition entities. The first detection principle involves the measurement of impedance change resulting from the binding of targets to bioreceptors (antibodies and nucleic acids) attached onto the electrode surface. The second category is based on the detection of metabolites produced by bacterial cells as a result of growth. EIS is a highly suitable method for monitoring the recognition interactions between receptors and target at the interface of a transducer. After immobilization of bioreceptors and recognition of the target, the surfaces or layers are characterized using an equivalent circuit that is employed to curve-fit the experimental data (Figure 1.5). From this, information is obtained about the electrical parameters responsible for the impedance change.⁷¹ 

All molecules have atomic nuclei and electrons in different orbital states and thus have the capability to interact with electromagnetic fields flowing through them. When molecules are placed in oscillating electromagnetic fields corresponding to propagation of light, electrons within the molecules will experience a force that makes them oscillate. This oscillation causes free electrons to be polarized in the presence of light’s electromagnetic field, producing a polarization current which travels slower through the molecule than it would through space. The fundamental principle behind an optical biosensor’s ability to detect analyte relies on the fact that all proteins, cells, and nucleic acids have dielectric permittivity larger than air and water. This causes these biological molecules to slow the propagation speed of electromagnetic fields passing through them.⁷²  Optical biosensors measure the change in phase, speed of polarization, or frequency of input light upon recognition of bioreceptors by their targets passing through them. Depending on the sensor’s assembly, the electromagnetic wave may be moving or standing and confined in such a way that the wave will pass through the samples. Various types of transducers that can be employed to produce optical change are grating couplers, resonant mirror, surface plasmon resonance (SPR) interferometry, reflectrometric interference spectroscopy, ellipsometry, and total internal reflection fluorescence (TIRF).⁷³  For detection using colorimetric and fluorescence, labeling either the target or bioreceptor with a chromogenic/fluorescent tag (e.g., a dye) will generate a change in color intensity/fluorescence signal that signifies the presence of the target molecule. Label-free detection is comparatively straightforward and cost-effective since the target entities are not labeled and are detected in their native configurations.⁷⁴ 

SPR is an optical phenomenon that is sensitive to the optical changes of a medium close to a metal surface.⁷²  Surface plasma wave excitation leading to anomalous diffraction on diffraction gratings was first reported by Wood at the beginning of the twentieth century. In the late 1960s, optical excitation of surface plasmons was successfully demonstrated by attenuated total reflection and since then surface plasmons and their major properties have been intensively researched.⁷⁵  Surface plasmons are distinctive electromagnetic modes associated with charge-density oscillation that may exist at the interface of two media with dielectric constants of opposite signs, such as a metal and a dielectric.⁷⁶  At a resonant frequency characteristic of a particular bulk metal, surface conductive electrons collectively oscillate and reach their maxima along the metal–dielectric interface and decay exponentially into both media.⁷⁵  An SPR biosensor is based on this principle. Bioreceptor molecules are first immobilized on a metallic thin film that is layered on a glass plate and irradiated from the back surface through a prism by a light. The target analyte is then introduced.⁷⁷  Binding of antigen to the surface of the metal film alters the resonance to longer wavelengths and the degree of change correspondingly depends on the concentration of bound target.⁷⁸  SPR allows the detection of target molecules in real-time mode as the optical wave varies with the progression of the binding reaction.⁷⁹ 

SPR-based detection has advantages over other conventional analytical procedures (e.g., microbiological assays) in terms of convenience of use, greater simplicity, rapid preparation of samples, and shorter analysis time (in some cases, only minutes for analysis). Furthermore, SPR biosensors provide apparent advantages in rapidity compared to other methods depending on biological readouts, for instance inhibition of microbial growth for detecting antibiotics.⁸⁰  Labeling is not required and rich information is readily obtainable, making SPR a particularly attractive technique in small-molecule screening.⁸¹ 

The main drawback of optical-based biosensors is the high cost of the instrumentation. Other disadvantages include

• requirement for immobilization of biomolecules, as material losses are observed during the process of immobilization of biomolecules on solid substrate

• contamination due to biomolecules and chemicals leaking out of the biosensor

• requirement for sterilization, since biomolecules may be denatured if non-sterile probes are used.⁸² 

Currently, SPR is a leading sensor technology for observation of biomolecular interactions in real time, and has been commercialized by several companies.⁸³ 

Mass-sensitive biosensing is based on mass change upon recognition of analyte, which is determined by a corresponding change in acoustic parameters of the biosensor.⁸⁴  This mass alteration can be studied by using a piezoelectric crystal that can produce vibration at a certain frequency coupled with the use of an electrical signal of a specific frequency.²⁴  Hence, the oscillation frequency is closely related to the applied electrical frequency and mass of the crystal. The increase in mass resulting from a biorecognition event causes changes in the oscillation frequency of the crystals and the difference in mass can be detected electrically.

In crystals, the piezoelectric effect takes place without a center of symmetry. Application of pressure to the crystal will cause a distortion of the crystal lattice resulting in a dipole moment in the molecules of the crystal. Although numerous kinds of crystal demonstrate the piezoelectric effect, quartz is most commonly used in biosensing applications due to its electrical, mechanical, and chemical properties.⁸⁴  Bulk wave or quartz crystal microbalance (QCM) and surface acoustic wave (SAW) are the main two types widely reported in mass-based biosensing application.

In QCM a thin quartz disc, which mechanically oscillates in the presence of an alternating electrical field, is placed between two electrodes.⁸⁵  Biorecognition elements are immobilized on the surface of the quartz resonator, followed by the administration of analyte solution. During the biorecognition event and binding of target by bioreceptors, the effective mass of the oscillator increases causing a decrease in the resonance frequency of the oscillator. This difference in frequency is evaluated to determine the change in mass.²⁴ 

Using the piezoelectric properties of a quartz crystal, SAW sensors transmit a surface wave generated by electrodes along a crystal face from one location to another.⁸⁵  The transducer behaves both as a transmitter and a receiver. Due to the use of high frequencies (ranging from several hundred MHz to GHz), SAW sensors have higher sensitivities than QCM.^64,86  The generated wave travels along the surface of the crystal and changes as mass is loaded on its path, which alters the phase-wave frequency. SAW instruments are layered with sensing receptors and the capture of analyte is detected by assessing the changes in surface wave velocity resulting from mass loading on the transducer.⁸⁷  The change in the resonant frequency can then be quantified and correlates with the mass load on the sensing transducer.⁸⁸ 

In a biochemical reaction, heat is either evolved or absorbed and this becomes the key component in thermal-based biosensors. Thermometric biosensors use variations in reaction enthalpies to monitor and detect changes in the reactions and structural dynamics of biological molecules.⁸⁹  Since temperature variations can be directly converted to electrical signals, an electrical method for temperature determination is the most efficient technique in biosensor fabrication.⁹⁰  The total amount of heat evolved or absorbed corresponds to the molar enthalpy and to the sum of product molecules produced in biochemical reactions.⁸⁹  A thermistor is often used as a temperature transducer with a very high negative temperature coefficient of resistance.⁹¹ 

The advantages of thermal biosensors include long-term stability since there is no chemical contact between transducer and analyte of interest; low cost; freedom from varying optical or ionic influence of sample attributes; no complications or interference in some multienzyme systems. They can be used for numerous applications.⁹²  However, the inherent disadvantage of this biosensor is its non-specific nature, because all enthalpy changes in a reaction system are responsible for the final measurement. Consideration therefore needs to be given to avoid non-specific enthalpy changes resulting from dilution and solvation effects.

In the last decade, biosensing has been extensively developed in every field but there is still a long journey ahead before conventional methods can be substituted totally by biosensors. The use of biosensors has been highly successful in the biomedical diagnostic market, but their potential in the food sector still needs to be firmly evaluated and established. Commercially available biosensors for food industries (Table 1.2) are still rather limited, regardless of the large amount of research and development work being carried out around the globe. These biosensors are used mainly to detect glucose and lactic acid concentrations.^93–95 

Problems such as limited active life of biomolecules, mass production, reliability, and easy handling must be addressed. Nevertheless, advances in biotechnology, bioelectronics, electronics and material sciences will permit these obstacles to be overcome in the near future, since biosensors are cost-effective and time saving in the food and agricultural diagnosis sectors. In the food industry the use of biosensors is predicted to grow steadily as robust analytical tools for quick onsite and online detection for food quality and safety applications and to close the gap not covered by conventional analytical methods. The commercialization of a handful of biosensors in the food industry is a preliminary indication of the massive potential of biosensors in the near future.

Minhaz Uddin Ahmed would like to acknowledge the financial support for the project given by Universiti Brunei Darussalam (UBD), (Grant# UBD/PNC2/2/RG/1 (255)) and Brunei Research Council (Grant# BRC-10) of Negara Brunei Darussalam. Syazana Abdullah Lim wishes to thank the UBD’s Graduate Research Scholarship (GRS) for her PhD fellowship.