Fluorescent Biosensors

Otto S. Wolfbeis

University of Regensburg, Institute of Analytical Chemistry,
Chemo- and Biosensors, D-93040 Regensburg, Germany

Abstract. Fluorescence spectroscopy in is numerous variations has become a most powerful bioanalytical and diagnostic tool in the past 20 years and - in terms of versatility - seems to be second only to nmr spectroscopy, but with an entirely different field of application. This development reflects the transition of fluorescence spectroscopy from a merely academic area of research into a highly practical tool and was paralleled by a gradual move of fluorescence spectroscopy from physics to chemistry and biology. We also note an increasingly better understanding of the basic work (that had been performed by physicists) by chemists and biologists. As a result, fluorescence has become highly interdisciplinary, and its main applications are clearly in the biosciences. One of the exciting fields is in biosensor technology, and specific examples are given on how the numerous methods known in fluorescence spectroscopy have been implemented to design biosensors.


1. Introduction

Fluorescence spectroscopy is highly interdisciplinary: Numerous biophysical studies have revealed that most bio-organic matter is fluorescent. For cases in which the intrinsic fluorescence of biomaterials is of limited use, organic chemists have designed smart synthetic fluorophors which may be used as probes and labels in areas such as membrane biophysics, cell sorting, ion transport and immunoassay, to mention only a few. In addition, new spectroscopic methods have been developed that make use of new components like (diode) lasers and LEDs, fiber optics, fast imaging devices, data loggers, and - of course - intelligent software. If one had to name one or two single factors for the tremendous increase in the popularity of fluorescence spectroscopy in the 1980s, it probably is the availability of numerous specific fluorescent probes, along with the success of fluorescence lifetime measurements. The 90s, in turn, saw a tremendous progress in the areas of imaging and single molecule detection.

The commercially most successful application of fluorometry is in luminescence immunoassay, now followed by the diverse applications of fluorescence activated cell sorting (FACS) and other studies on the function of cells. Other exciting novel areas include fluorescence correlation spectroscopy which enables the detection of single molecules, multi-photon excitation with its inherent advantages over conventional excitation, fluorescence imaging and sensing. This contribution will focus on bioanalytical and sensing aspects of fluorescence and - in a wider sense - luminescence.

2. What is a biosensor, and what is it good for?

Biosensors are small devices capable of detecting a chemical or biochemical species and incorporating a biological element such as an enzyme, an antibody, a DNA, or even whole cells. Biosensors are expected find numerous applications which include (a) medical research, (b) clinical diagnosis, (c) environmental testing; (e) bioprocess monitoring and biotechnology in general; (e) food quality control, (f) pharmacology and brain research, and (g) development of new pharmaceuticals. In the ideal case, a biosensor is contacted with the sample and the analytical result is displayed in short time. This has been accomplished in a limited number of cases only so far, but tremendous research activities are undertaken to design new and better sensors.

Biosensors can be based on a variety of detection schemes. Among those, optical sensors form a major group and display features that can make them advantageous over other systems such as electrochemical, mass-sensitive, thermal, acoustic, or other devices. Typical optical schemes are based on absorption spectrocopy (from the UV to the deep infrared), Raman and conventional fluorescence spectroscopy and imaging, but also on more sophisticated methods such as surface plasmon resonance, evanescent wave and near-field spectroscopy, fiber optic spectroscopy, correllation spectroscopy, and - last not least - luminescence lifetime, polarization and energy transfer.

Luminescence spectroscopy knows numerous parameters to be determined. In analytical applications these include measurement of light intensity, its decay time, polarization, quantum yield and quenching efficiency, radiative and non-radiative energy transfer, and numerous combinations thereof. Luminescence can be excited by light (including laser light), electrochemically, by (bio)chemical energy, pressure or sound. Luminescence can be measured of gaseous, liquid and solid samples, of tissue and cells, by directly illuminating the object or by using waveguide optics including optical fibres. Its versatility is obvious. Luminescence can be presented in Cartesian format, but also in multidimensional form to result in so-called excitation-emission matrices or contour plots. Fluorescence lends itself to imaging which has become the basis for exciting studies in cell biology, dermatology, and aviation research, but also for remote sensing, e.g. from satellites.

3. Many biomolecules are fluorescent

Most biological matter is fluorescent, albeit in the uv. Repective species include the three aromatic amino acids and proteins containing them, the green fluorescent protein (GFP), nucleic acids, flavine nucleotides and NADH, whilst NAD+ and most saccharides and lipids are non-fluorescent. In addition, most colored matter in nature including coumarins, flavones, anthocyans and chlorophylls (but with the notable exception of hem) is fluorescent.

4. What if not fluorescent?

Fluorescent labels render a biomolecule (or a biological system) fluorescent so to make it amenable to fluorescence spectroscopy. Labelling of proteins is most common in biosciences, and millions of fluorescent immunoassays are performed world-wide every year. Numerous tests ("kits") are commercially available and based on measurement of either intensity, polarization, energy transfer, or delayed fluorescence. It is also worth to recall that the human genom project (the largest single project in science at present) is based on fluorescent labelling of nucleic acids which enables a faster sequencing than any other method at present.

Labels are preferably attached to the species of interest by covalent binding via a reactive group that forms a chemical bonds with other groups such as amino, hydroxy, sulfhydryl or carboxy. Labels are expected to be inert to other chemical species present in the environment, for example to pH. In order to reduce background luminescence of biological matter, labels preferably have long-wave excitation and emission, and/or long decay times so that background luminescence decays much faster than the luminescence of the label. As a result, there is a substantial interest in the design of longwave and long-decay luminescent labels.

5. Fluorescent Probes

Probes, in contrast to labels, are expected not to be inert but rather to respond to their micro-environment or to a chenmical species. Probes responding to a chemical species (such as an ion, to pH or oxygen) may also be referred to as indicators. Numerous fluorescent probes have become available in the past years and have substantially contributed to the success of fluorescence spectroscopy in biosciences.

6. Direct Fluorescent Sensors

Numerous bioanalytical assays are based on the fact that NADH is fluorescent, while NAD+ is not. As a result, all enzymatic reactions based on NAD/NADH are amenable to fluorescence analysis, and this is widely exploited in practive, even though NADH has to be excited at around 350 nm which can cause substantial background fluorescence from other biomatter. On the other side, most assays are performed in the kinetic mode so that it is the relative signal change (above the background) that is measured rather than the total intensity (including the background) and this reduces interferences a lot.

The co-enzyme FAD is another strongly luminescent species but has found less wide applications because both the oxidized (FAD) and the reduced form (FADH2) display fluorescence so that they are less useful for monitoring the course of a biochemical reaction. Their excitation is at around 450 nm, and fluorescence peaks at 512 nm. Both NADH and FAD have been shown to be useful for purposes of chemical sensing using immobilized reagents and, in some cases, using fiber optics waveguides. However, their practical applications have been been confined mainly to conventional cuvette tests so far.

7. Indirect Fluorescent Sensors

7.1. Chemosensors

There is a tremendous interest in sensing ("monitoring") clinically imprtant species such as oxygen, the proton ("pH"), carbon dioxide, glucose, urea and the like. These are major metabolites or substrates which cannot be visualized directly via luminescence spectroscopy. Hence, use is made of respective indicators that can be added to the system of interest and which allow indirect visualization of the species of interest.

Optical sensing of oxygen is almost exclusively accomplished on the basis of the quenching effect which it exerts on numerous luminophores. Both luminescence intensity and decay time may be measured, the latter being preferred for reasons of system stability, precision and accuracy. The pH of blood and related biological matter is determined via classical pH indicators such as the fluoresceins which are immobilized on solid supports (or inside polymeric matrices) and undergo a pH-induced spectral change. Numerous schemes exist that are based on measurement of either luminescence intensity, decay time, or energy transfer efficiency. Fiber optic sensors for measurement of oxygen, pH, carbon dioxide and hematocrit are on the market, as are microsensors (with ~10 Ám thin tips) for measurement of oxygen in algae and in the deep sea.

7.2. Enzymatic Biosensors

The above chemosensors may also be used to monitor the course of enzymatic reactions during which oxygen or protons are consumed or produced, a typical example being the oxidation of glucose by the enzyme glucose oxidase according to:

glucose + oxygen ===> gluconic acid + hydrogen peroxide (1)

It can be seen that such a reaction can be monitored kinetically in various ways, namely by measureing the consumption of oxygen (via the above fluorescent oxygen sensor), the formation of the acid (via the above pH sensor), or the formation of hydrogen peroxide (via a respective sensor). Since 2 - 4% of caucasians suffer from Diabetes mellitus, there is a substantial need for such sensors. Numerous other enzyme based biosensors have been described in the past 10 years, and typical examples will be given.

Enzymatic reactions can be inhibited by toxic species including organophosphates and heavy metals. This is the basis for so-called inhibition biosensors which respond to environmentally harmful species like herbicides, warfare agents, and heavy metals including lead and cadmium. Respective examples will be presented as well.

7.3. Cellular Biosensors

Fluorescent biosensors may also be obtained by immobilizing whole cells on the surface of a sensor layer. For example, immobilization of yeast on the surface of a luminescent oxygen sensor led to a fiber optic sensor for measurement of biological oxygen demand. This is a parameter for the loading of water with organic waste. Yeast and certain bacteria digest the organic fraction and thereby consume oxygen, and this is detected by the luminescent oxygen sensor.

8. Trends in Biosensor Technology

Current trends include the design of new luminescent probes and labels for the near infrared spectral range, the search for useful new schemes for bioanalytical applications, the design of imaging sensors, micro-sensors and arrays, the use of fiber optic systems in high-technology fields such as distributed sensing, gene probing, high-throughput screening, and the design of methods for 3-dimensional imaging.