Feature

Biosensors

Screen-printed enzymes and mashed mushrooms are providing a fast, cheap method of chemical analysis.

By Alan Hart

It is probably impossible to over-estimate the demand that modern societies have for information about the chemical composition of substances.

• Are there herbicide residues in this water?
• What are the glucose levels in this blood?
• How much lactic acid is in this yoghurt?
• What is the concentration of nitrogen oxides in this exhaust?

Demand for information of this sort is met by analytical chemistry in its various guises. Chemical analyses usually entail collecting a sample, taking it to a laboratory, possibly purifying it and then determining its composition by some combination of "wet" chemistry and instrumental analysis, such as gas chromatography or mass spectrometry.

Laboratory methods can be very sensitive and accurate, but for some situations they have disadvantages. They can be quite costly. It can be impossible or dangerous to collect samples for analysis. The delay in getting results caused by the necessity to transport and treat samples may mean that information is generated too slowly to make the necessary decisions; often "real-time" information is required. If an effluent treatment process goes wrong we need to know now, not several hours later when water supplies may have become contaminated.

Chemical "sensors" have been developed to allow analyses away from the analytical laboratory. A chemical sensor consists of a recognition element attached to a transducer of some kind which converts the reaction between the substance to be detected -- the analyte --  and the recognition element into a physical signal such as a current or voltage.

A well known example of a chemical sensor is the pH electrode, where diffusion of hydrogen ions across a glass membrane generates a voltage which is measured and amplified by the pH meter. The ideal chemical sensor is selective enough not to respond to anything apart from the analyte, doesn't require any sample pretreatment or addition of reagents, and gives results effectively in real-time.

Living organisms -- including chemists in the laboratory -- are very good at doing analytical chemistry. Substances are selected from the environment, often with extreme accuracy and sensitivity, and analyses are done in real-time. Some moths can detect pheromones from another moth 16 kilometres away; enzymes in our bodies react specifically with substrates to synthesize proteins, provide energy in respiration and so on. Why not make use of the sensitivity and selectivity of biological systems to make sensors?

Biology Meets
Electronics

"Biosensors" are sensors in which the recognition element, in contact with a physico-chemical transducer such as an electrode or an optical fibre, is of biological origin. When the biological part interacts with the analyte, there is a change in the biological component, and that change is converted into a measurable signal such as a current or change in light intensity.

A very wide range of biological systems from enzymes and antibodies to cells and pieces of tissue have been coupled to an equally wide range of physical transducers, from optical to electrochemical to piezo-electric to calorimetric to mechanical.

Enzymes -- a large group of biological proteins -- are catalysts with a high degree of selectivity. While enzymes are usually thought of as functioning inside living cells, it is possible to immobilise them in a working state onto the surface of electrodes which are part of an electrochemical circuit.

In one example, glucose oxidase, an enzyme which reacts with glucose, is immobilised with a compound called a mediator, onto the surface of a graphite electrode. This biosensor can then be used to detect glucose in blood, as electrons from the glucose flow to the electrode, forming a current proportional to the amount of glucose in the blood. An answer is obtained in 30 seconds; no other reagents or chemical procedures are required. The demand for clinical glucose analyses has helped make glucose biosensors the most well-known and successful enzyme-based biosensors.

Some biosensors are "immunosensors" which depend on antibody-antigen interactions. One such immunosensor can detect human chorionic gonadotrophin (HCG) in urine, and is used as the basis of a home pregnancy test. The components of the sensor are housed in a small plastic case. An absorbent sampler protruding from one end is held for a few seconds in the urine stream. If HCG is present in the urine, it is carried towards HCG antibodies labelled with a blue dye. An HCG-HCG antibody/dye complex is formed and moves towards a window in the case where there is another set of HCG antibodies. At the window, these combine to form an HCG antibody/dye-HCG-HCG antibody "sandwich". The accumulation of complexes shows up as a blue line at the window, indicating a positive pregnancy test within just three minutes.

Simple, robust photosynthetic organisms called cyanobacteria have been used in a biosensor to detect herbicides in water. The cyanobacteria were immobilised in alginate beads. These were placed over a graphite electrode, positioned under a light source built into the sensor casing. The light induced photosynthesis in the cyanobacteria. When water containing potassium ferricyanide flowed past them, the ferricyanide was reduced by electrons originating from the photosynthetic electron transport chain in the cyanobacteria. The ferricyanide was then re-oxidised at the graphite electrode, thus generating a current proportional to the activity of the photosynthetic system. If the water also contained herbicide residues, photosynthesis was inhibited and the current decreased, indicating the presence of the residues.

In another group of biosensors, plant tissue is used as the recognition element. A biosensor capable of rapid measurements of the neurotransmitter dopamine was constructed from mushroom tissue and an electrode made from reticulated vitreous carbon. Mushrooms are rich in an enzyme capable of reacting with dopamine, and reticulated vitreous carbon has a porous, honeycomb structure. Mushrooms were crushed with a mortar and pestle and packed into the pores of the electrode. Such a simple procedure using crushed tissue is intriguing, as one usually associates a reasonably high degree of order with scientific systems. It has to be assumed that the fine structure required for enzyme function within the crushed tissue is not destroyed during sensor construction.

One of the outstanding features of biosensor research is its multidisciplinary nature. The examples given above use knowledge from electrochemistry, enzymology, organic chemistry, immunology, biochemistry, polymer science, photochemistry, electronics and optics. Information and skills from these and other disciplines are necessary to meld biological and inanimate components into a stable state. Biosensor research is, at first, reductionist in approach, as biological reactions are thought of in physical terms involving flows of electrons, changes in electrochemical gradients and kinetic constants, but is then synthetic, as various components are brought together to build a functional sensor.

Biosensors can provide analyses in real-time and have inherently high selectivity. The use of biological components avoids having to create complex chemical substitutes for biological reactions. In moving from the laboratory to the outside world, biosensors have to compete with alternative analytical techniques on the basis of ease of use, cost, sensitivity, robustness, operational stability and shelf life. These last two factors represent particular problems for biosensor researchers, as biological systems tend to be unstable to varying degrees once they are removed from intact organisms.

One way in which instability has been circumvented for enzyme-based sensors is to make the sensing part of biosensors as disposable units. For example, screen printing techniques enable large numbers of sensing elements to be made very cheaply. These can be used once and discarded before instability becomes a problem.

One role of biosensors will probably be to provide economical screening of samples in homes, hospitals, factories, environmental monitoring and process control, thus reducing the number of more costly and time-consuming laboratory analyses.

Last year saw New Zealand's first symposium on biosensors. A wide range of topics was discussed, from surface chemistry of metal electrodes to the measurement of antibody-antigen interactions through changes in frequency of oscillation of quartz crystals.

Nothing is uninteresting to biosensor researchers, who always have a beady eye out for phenomena which might enable a biological event to be transduced into a measurable physical signal. The small but active sensor research community will ensure that New Zealand will benefit from advances in these fascinating devices.

Alan Hart is a scientist at AgResearch, Grasslands Research Centre, Palmerston North.