Fluorimetry is the quantitative study of the fluorescence of fluorescent molecules. Many biomolecules are fluorescent or can be labelled with fluorescent molecules, making fluorimetry a widely used tool in analytical and imaging methods. As the available photon-detecting devices are highly sensitive—even a single photon can be detected—and one fluorophore can emit millions of photons in a second, fluorimetry is suitable for and is often used in single-molecule experiments.
The phenomenon of fluorescence was discovered and published by Sir John Fredrick William Herschel in the mid-1800s. He observed that, when illuminated with white light, a solution of quinine emitted a strange blue light perpendicular to the direction of the illumination, even though it remained colourless when observed facing the light source.
Demonstrating the sensitivity of fluorescence measurements, such methods were used to prove that the rivers Danube and Rhine are connected by underground waterways. In 1877, researchers poured fluorescein (a fluorophore) into the Danube and could detect its green fluorescence 60 hours later in a small river flowing into the Rhine. Fluorescein is still used to aid the detection of space cabins that returned to Earth and fell into an ocean.
Photons of a given wavelength are absorbed by the fluorophore and excite some of its electrons. The system remains in this excited state for only a few nanoseconds and then relaxes into its ground state. (Note that light travels 30 centimetres in a single nanosecond.) When returning from the excited state to the ground state, the electron may emit a photon. This is known as fluorescent emission. The wavelength of the absorbed photon is always shorter than that of the emitted photon (i.e. the energy of the emitted light is lower than that of the absorbed one). This phenomenon, the so-called Stokes shift, is an important attribute of fluorescence both in theory and practice.
The relations between the wavelength (λ, nm), frequency (ν, 1/s) and energy (E, J) of light are the following:
λ = c/ν, where c is the speed of light (approximately 300 000 km/s)
ν = c/λ
E = hν, where h is the Planck constant (approximately 6.63 *10-34 Js)
The Stokes shift facilitates the creation of highly sensitive methods of detection of fluorescence. As the wavelengths of the exciting and detected (emitted) light differ, the background created by the exciting light can be minimised by using a proper setup. There are two ways to avoid that the exciting light get into the detector:
(1) Measurements are often carried out in a geometric arrangement in which the detection of emission is perpendicular to the exciting beam of light.
(2) Light filters are placed between the light source and the sample and also between the sample and the detector. Light of only a certain wavelength range can pass through these filters. Photons of the exciting light leaving the sample will not reach the detector as they are absorbed by the emission filter (Figure 4.12). In many cases, monochromators are used instead of filters. Their advantage is that the selected wavelength can be set rather freely and more precisely compared to filters that are set to a given interval and adjustments can only be made by replacing them (Figure 4.13).
This double protection of the detector from the exciting light is necessary due to the fact that the intensity of fluorescent light is usually two or three orders of magnitude smaller than that of the exciting light. This means that even if only 1 or 0.1 % of the exciting light reaches the detector, half of the detected signal intensity would arise from the exciting light and only the other half from the emission of the sample. This would result in a 50 % background signal level, as the detector is unable to distinguish photons based on their wavelength.
Fluorophores are characterised by specific fluorescence spectra, namely their excitation (absorption) spectrum and emission spectrum. The excitation spectrum is recorded by measuring the intensity of emission at a given wavelength while the wavelength of excitation is continuously changed. The emission spectrum is recorded by measuring the intensity of the emitted light as a function of its wavelength while the wavelength of the exciting light is kept constant.
The shape of the excitation spectrum is usually the same as the shape of the emission spectrum. However, due to the Stokes shift, the emission spectrum is shifted towards red compared to the excitation spectrum, and usually the shape of the two spectra are mirror images of each other (Figure 4.14).
The intensity of fluorescence of a molecule is sensitive to its environment. Emission intensity is significantly affected by the pH and the polarity of the solvent as well as the temperature. Usually, an apolar solvent and a decrease in temperature will increase the intensity. The immediate environment of the fluorophore is an important factor, too. Another molecule or group moving close to the fluorophore can change the intensity of fluorescence. Due to these attributes, fluorimetry is well suited to the study of different chemical reactions and/or conformational changes, aggregation and dissociation. In proteins, two amino acids have side chains with significant fluorescence: tryptophan and tyrosine (Figure 4.15). The fluorescence of these groups in a protein is called the intrinsic fluorescence of the protein. Tryptophan is a relatively rare amino acid; most proteins contain only one or a few tryptophans. Tyrosine is much more frequent; there are usually five to ten times more tyrosines in a protein than tryptophans. On the other hand, the fluorescence intensity of tryptophan is much higher than that of tyrosine.
The spectra in Figure 4.15 clearly show that the fluorescence of tryptophan can be studied specifically even in the presence of tyrosines, since if the excitation is set to 295 nm and the detection of emission is set to 350 nm, the fluorescence of tyrosine can be neglected. Both the intensity of the fluorescence and the shape of the emission spectrum are sensitive to the surroundings of the side chain, which often changes upon conformational changes of the protein. Tryptophan fluorimetry is therefore suitable to detect conformational changes of enzymes and other proteins. It can also be applied to detect the binding of ligands to proteins as well as the di- or multimerisation of proteins, provided that the reaction results in a change in the surroundings of a tryptophan side chain. The environment of tryptophans obviously changes on unfolding of proteins. Consequently, fluorescence is well suited also for following denaturation of proteins.
Tryptophan and tyrosine fluorescence is not the only way to detect and investigate proteins using fluorescence. There are proteins that undergo post-translational modifications including the covalent isomerisation of three amino acids that makes them fluorescent. The first such protein discovered was the green fluorescent protein (GFP), which is expressed naturally in the jellyfish Aequorea victoria (phylum Cnidaria) (Figure 4.16, left panel). Since then, fluorescent proteins were isolated from many other species. A large number of recombinantly modified forms of GFP were created in the last 20 years, all different in their fluorescence and colour (Figure 4.16, right panel). The intrinsic fluorescence of GFP can be used to label proteins. If we create a chimera from the genes of GFP and another protein of interest—in other words, we attach the gene of GFP to the 5’ or 3’ end of the gene encoding the other protein—this construct will be transcribed and translated into a protein that will have GFP fused to it at its N- or C-terminus. Thus, if using an appropriate vector we transform an organism and introduce this new gene into it, its product will show a green fluorescence when excited. Through this phenomenon, we can easily locate proteins on the tissue, cellular or subcellular levels. As a variety of differently coloured fluorescent proteins are at our disposal, we can even measure colocalisation of labelled proteins in vivo. The application of fluorescent proteins in biology was such a significant technological breakthrough that its pioneers were awarded a Nobel prize in 2008.
Proteins and other biological molecules can also be made fluorescent by using extrinsic modifications. We can attach extrinsic fluorophores to biomolecules by either covalent or non-covalent bonds.
Covalent attachment of fluorophores is most often achieved by using the reactive side chains of cysteines. To this end, researchers use fluorophores that bear iodoacetamido or maleimido groups that alkylate the sulfhydryl group of cysteine side chains under appropriate conditions.
Proteins can form complexes with fluorescent substrates or inhibitors also via non-covalent bonds. There exist also fluorophores that can bind to certain regions of proteins with a high affinity. For example, 8-anilononaphtalene-1-sulfonic acid (ANS) binds to hydrophobic regions of proteins specifically and becomes strongly fluorescent when bound. We can take advantage of this phenomenon in experiments. A change in the amount of hydrophobic surfaces can occur in conjunction with structural changes induced by the binding of a ligand. Thus, addition of the ligand may cause the decrease of the amount of protein-bound ANS and thus the binding of the ligand can be studied by measuring the changes in the fluorescence of ANS. This way the binding constant of the protein and the ligand, as well as the kinetics of the binding can be examined in a simple yet quantitative manner (cf. Chapter 8).
Labelling of double-stranded DNA can also be achieved, for example, with ethidium bromide in vitro. When intercalated between the bases of DNA, the fluorescence of ethidium bromide will rise markedly. Formerly, visualisation of DNA in agarose gel electrophoresis was generally achieved using ethidium bromide. The dye was mixed into the agarose gel to form a complex with the DNA passing through it. When illuminated by ultraviolet light, the ethidium bromide accumulated in the DNA becomes visible due to its fluorescence. As ethidium bromide is carcinogenic, nowadays rather non-carcinogenic alternatives (e.g. SYBR Safe) are used.
Nucleic acids can also be labelled fluorescently through covalent modifications. Fluorophores can be bound to the 5’ or 3’ hydroxyl group. The most common way to create DNA labelled on its 5’ end is to synthesise it in a PCR reaction with primers labelled at their 5’ end.
There is a large number of different fluorophores available commercially that exhibit different fluorescent properties. We can make our choice based on the excitation and emission wavelengths. For example, fluorescein, one of the first fluorophores used, exhibits its absorption maximum at 494 nm and its emission maximum at 521 nm. The intensity of the fluorescence of a given fluorophore is determined by its absorption coefficient and emission efficiency (i.e. the probability that the electron emits a photon when it relaxes). This can provide another way to optimise our experiment. The extent of the Stokes shift is also an important aspect. Technically, fluorophores with a greater shift are more advantageous. The greater the difference between the excitation and detection wavelengths, the easier it is to prevent (by using filters or monochromators) the exciting light from getting into the detector. This significantly decreases the actual background.