Fluorescence detection in gel electrophoresis
Fluorescence detection in gel electrophoresis

New! - Transilluminator 101 - the midrange transilluminator
New! - Emission spectra of translluminators
SYBR gold, SYPRO rubyIntroductionSYBR gold, SYPRO ruby

Note: The bulk of what's on this page is 20 years old. There are notable updates, but I don't guarantee currency. I try to update links but, sadly, they're volatile

Chemists and biologists are familiar with absorbance but less comfortable with fluorescence. Facility with fluorescence requires understanding fundamental wave and particle properties of light as described in elementary physics and physical chemistry texts.

Relatively few dyes fluoresce. Fluorescent dyes absorb at shorter wavelengths and emit at longer wavelengths. Fluoresced light intensities are always minuscule compared to those of standard absorbance techniques.

Advantages of fluorescence

1- High sensitivity
a- Absorbance detects small differences between incident and transmitted light. Many photons must be absorbed to give detectable absorbance.

b- Fluorescence is an absolute measurement. Theoretically, a single fluorescing photon can be detected with, for example, a scintillation counter. In practice, stray light severely limits sensitivity. Nonetheless good fluorescence methods can be 10 to 1000 fold more sensitive than good absorbance methods. Single DNA molecules can easily be seen by fluorescence microscopy.

c- Observing fluorescence is similar to observing stars. One sees faint light against a dark background. As in astronomy stray light can limit detection sensitivity. This is why fluorescent objects are usually photographed in hoods.
2- Simplicity
a- Where binding enhances fluorescence 10 to 100 fold, gels can be stained with dilute dye and photographed without destaining. Molecules can be prestained with strongly binding dyes.
NOTE: Prior to the advent of ethidium bromide, appropriate dyes and high quality UV light sources were not available.

Useful dyes

1- Ethidium bromide (EtBr)- EtBr intercalates into and stains double stranded regions of nucleic acids. Intercalation enhances fluorescence about 50-fold. At moderate ionic strengths EtBr will not bind to single stranded regions. LePecq and Paoletti [J. Mol. Biol. 27, 87 (1967)] describe the staining properties of EtBr - Spectrum of EtBr - DNA complex.
a- Staining nucleic acids in agarose gels
Stain by soaking in dilute EtBr (about 1 ug/ml in distilled water or dilute Na2EDTA) for 30 min. Photograph through a red filter which absorbs both incident UV light and fluorescence from agarose. Agarose fluorescence severely limits detection sensitivity. All commercial agaroses fluoresce. The fluorescence is insensitive to pH and dielectric constant of the medium and resists chemical oxidation and reduction.
b- Staining nucleic acids in polyacrylamide gels
Stain by soaking in very dilute EtBr (10 to 50 ng/ml in distilled water or dilute Na2EDTA) for 1-2 hrs. Photograph through an orange filter. Bands with as little as 100 pg of nucleic acid can be detected. At this low level of stain, pay careful attention to stoichiometry; major bands can eclipse minor bands by tying up all the stain.
c- Staining single stranded nucleic acids
EtBr is a poor stain for formaldehyde or glyoxal modified nucleic acids. Silver stains are preferable. Brief alkaline hydrolysis improves ethidium staining of formaldehyde modified nucleic acids [Long and Dawid, Cell 18, 1185 (1979)]. SYBR gold (see below) is now the preferred stain.
2- Ethidium dimer - DNA complex
Ethidium dimer [Gaugain et al, Biochemistry 17, 5071 (1978)] stains much like ethidium bromide. It binds at least 1000 times as strongly to nucleic acids as EtBr and can therefore stain at higher salt concentrations. Unfortunately its cost relative to EtBr parallels its affinity for DNA.
3- SYBR Gold - Spectra DNA complex
SYBR Gold, developed by Molecular Probes (now part of Invitrogen), has strong affinity for nucleic acids combined with high absorbance and superior fluorescence enhancement on binding. The dye is expensive, but can be used in very low concentration. It is the dye of choice for staining nucleic acids in gels. Since it binds strongly it is much superior to EtBr for staining denatured nucleic acids.
4- Hoechst 33258 and related bisbenzimidazoles - spectrum of DNA complex
Properties have been described by Cesarone et al, [Anal. Biochem. 100, 188 (1979)] and Latt and Stetten (J. Histochem. and Cytochem. 24, 24 (1976)]. They have been widely used in cytology especially to stain chromosomes. Their binding mechanism differs from that of EtBr. Staining procedures are similar to those for EtBr. Unfortunately emission maxima of these dyes are at shorter wavelengths where agarose fluoresces strongly. At moderate salt concentrations (above 0.1M), these dyes selectively stain DNA. A mixture of EtBr and Hoechst 33258 stains DNA yellow and RNA orange. DAPI and DIPI are in the same family of A-T, DNA specific dyes. The latter is especially resistant to photobleaching [Schnedl et al, Human Genetics 36, 167 (1977)]. A useful review is Schweizer [Human Genetics 57, 1 (1981)].
5- SYPRO Ruby - spectra
SYPRO ruby, a bathophenanthroline complex of ruthenium, is a fluorescent analog of Coomassie Blue and a dye of choice for proteomics. See the Molecular Probes bibliography.

6- New! Coomassie Blue
Recently found to fluoresce in the IR. Commonly detected with LiCor Odyssey imager. In theory it's more easily and cheaply detected with a blue light box and a near IR sensitive CCD camera. Should be more resistant to photobleaching than standard fluorescent dyes.
7- Deep Purple (Epicocconone)
- recently discovered protein dye - See Proteomics 3, 2273 (more sensitive to photobleaching than Sypro Ruby)
7- Fluorescent amino group reagents
Dansyl chloride is the most generally useful; pyrenesulfonyl chloride) is also useful. They react with protein amino groups to give fluorescent derivatives. Detection with midrange UV light sources is about ten-fold more sensitive than Coomassie Blue staining, but about ten-fold less sensitive than silver. Fluorescamine, o-phthalaldehyde, and 2-methoxy-2,4-diphenyl-3(2H)furanone (MDPF) fluoresce only after reaction with amino groups. Of these only MDPF is stable and resistant to photobleaching. Proteins to be run on SDS gels should be reduced and alkylated prior to amino group modification.
Dansylation of proteins for SDS PAGE - spectra of dansyl group
1- Dissolve protein in 10% Ficoll, 1% SDS, and 10mM NaPO4 buffer, pH 7.2.
2- Add 0.02 vol of 0.25M dithiothreitol and boil 1.5 min.
3- Add 0.09 vol of 0.25M iodoacetamide and heat 15 min at 50oC.
4- Add 0.1 vol of 0.1% thymolphthalein (which turns blue at pH 10), then bring to pH 10 with 0.02 vol of 2N NaOH and add 0.2 vol of dansyl chloride (2 mg/ml in acetone). Incubate 10 min at 50oC.
5- Bring to pH 7.0 with 0.02 vol of 2N HCl.
6- Electrophorese aliquot.
7- Following electrophoresis, examine gel with midrange UV light.
5- 8-Anilino-1-Naphthalenesulfonic acid (ANS)
ANS fluoresces weakly in aqueous solution, but brightly in hydrophobic environments [Brand and Gohlke, Ann. Rev. Biochem. 41, 843(1972)]. ANS is structurally similar to anionic detergents and has been widely used as a "hydrophobic probe" of protein structure. The dimer of ANS, bis-ANS (Molecular Probes) binds strongly to denatured proteins. We have used it as a substitute for Coomassie Blue (it has similar sensitivity). SDS must be soaked from the gel prior to staining with 250 ng/ml bis-ANS in 12% acetic acid. Unfortunately, both ANS and bis-ANS photobleach severely.
8- Other potentially useful reagents
a- Fluorescent Schiff reagents
Procedures have been described in Analytical Biochemistry and in the histochemistry literature. The dyes include acriflavine, dansyl hydrazine (actually a hydrazide) and hydrazinoacridine. In our hands, the latter is a promising reagent.
b- Rare earth cations
Several of these, notably terbium, fluoresce on binding to light absorbing polyanions. Terbium selectively stains ssDNA [Topal and Fresco, Biochemistry 19, 5531 (1980)]. Use terbium below pH 7 since its hydroxide is insoluble. Binding can strikingly enhance fluorescence (>104!).
c- Sulfhydryl reagents
Among several fluorescent SH reagents, the bromobimanes [Newton et al, Anal. Biochem. 114, 383 (1981)] are probably best known.
d- Others
Histochemical literature contains many staining protocols. Unfortunately most give little idea of sensitivity. Molecular Probes does a good job of describing properties and providing bibliographies. Molecular Probes also creates new dyes and staining techniques by combining principles with synthetic chemistry. H.J. Conn's Biological Stains by R.D. Lillie, 9th edition, Williams and Wilkins, 1977 is a useful (but older) general reference. Also check out Anaspec.
Properties desirable for a fluorescent dye
1- Specific binding.
2- High extinction coefficient: to fluoresce intensely a dye must absorb intensely.
3- High quantum yield (i.e. emits as many photons as it absorbs).
4- Stable to heat, pH extremes, irradiation (resistant to photobleaching), etc.
5- Excitation spectrum well matched to the light source.
6- Substrate binding enhances fluorescence.
7- Excitation and emission maxima well separated. Fluorescence microscopy illuminates only small areas using intense light sources and high quality filters. This combination permits use of fluoresceins which have exceptional molar absorptivity and high quantum yield, but overlapping absorption and emission spectra.
8- Reacts rapidly with and has high affinity for substrate.
9- Favorable binding stoichiometry; note: many fluorescent protein stains react only with amino groups.

References; Southern, Methods in Enzymology 68,152; Ultraviolet and Fluorescence Photography, Eastman Kodak Publication No. M27.
1- Light sources; direct (transilluminator) or indirect (epi-illuminator)
Suppliers of UV transillumators include Ultra-Violet Products Inc. Fotodyne Inc., Spectronics. More recently the "Dark Reader " (visible transilluminator) has come on the market. Summary of imaging platforms. LEDs may be the light sources of the future, but they have a way to go.
a- Short wavelength - potential safety hazard, low stray light, on older models filters rapidly solarize (become opaque to UV light).

b- Medium wavelength - less hazard, low stray light, less solarization, less photobleaching, emission spectrum well-suited to excite common fluorescent dyes such as EtBr.

c- Long wavelength - little hazard, some stray visible light, not well-suited to excite common dyes such as EtBr.

d- Visible - no hazard, doesn't damage nucleic acids

Added 3/26/04 Transmittance of UV light boxes - a must see if you use these regularly
2- Exciter filter
The light box itself normally contains the exciter filter. It passes shorter wavelength exciting light and absorbs longer wavelength light that might interfere with fluorescence detection. Commercial exciter filters, however, pass far red light which can expose panchromatic (far red sensitive) films. The purple glow of UV light box is a mixture of blue light, below 450nm, and red light, above 650nm. The filters are opaque to a surprisingly narrow band of light.
3- Barrier filter
The barrier filter is above the gel and usually attached to the camera lens. It absorbs exciting and stray light, passing only fluoresced light from the sample. Many different filters, (commonly yellow, orange or red) can be used (Kodak Filters, Eastman Kodak Technical Publication B-3H). Filters and filter combinations should be visually inspected for fluorescence under UV illumination. If they fluoresce, the barrier filter should be protected from exciting light by a lower wavelength cutoff filter (usually UV). Remember that the order of filters can be critical. Commercial filter producers pay little attention to fluorescence so the user must inspect each filter.
4- Film - Fluorescence photography info and resources
Light intensities are minuscule relative to "normal" photography. Films are exposed beyond their intended range in the region of "reciprocity failure" where grain activation is no longer proportional to exposure. This is because activation of a grain requires several photons (roughly 5). The first few photons are absorbed reversibly. When the photon flux falls below a certain level the loss of photons from a grain becomes significant. We use Polaroid type 665 film (ASA 75). The f-stop is wide open with exposures up to 10 min. Many films are satisfactory, but high ASA films do not generally improve detection sensitivity. Sensitivity can be improved by pre-flashing film, which enables a smaller number of irreversibly absorbed photons to activate a grain, and increasing development time beyond manufacturer's specifications. Kodak Ektachrome and Technical Pan Film 2415 are sensitive, low cost alternatives. The latter can be purchased in 100 ft rolls and developed in the darkroom. Far red light which passes through midrange exciter filters exposes both films, particularly the 2415. A short pass interference filter (about 660nm cutoff, Ditric Optics Inc., 312 Main St., Hudson, MA 01749) solves the problem. Wide band interference filters are an alternative. We have not checked the sensitivity of the Kodak films but suspect they are slightly better than Polaroid 665. If your results are best with short exposures and high f-stops, your sensitivity is low. The very highest film sensitivities can be obtained by "hypersensitization" which involves heating film in a hydrogen atmosphere [Babcock, et al, The Astronomical Journal 79, 1479 (1974)]. A small group of astronomers and photographic sensitivity specialists use this technique.
5- Dedicated systems
Integrated systems including transilluminators, hoods, barrier filters and CCD cameras are now commonly available. For those with larger budgets laser scanners are available. Most fluorescent dyes absorb more strongly in the visible than in the UV (e.g. SYBR Gold). The combination of a laser light source and an interference filter will give optimum sensitivity. The major drawback of lasers is fixed wavelength which limits versatility.
6- Ideal systems
Ideally one would have a light box with a high power tunable monochromatic source. One could watch the output (CCD device) in real time on a computer screen and pick the optimal excitation wavelength. One would also have a continuously variable barrier filter where one could optimize by varying wavelength and bandwidth of the emission filter in real time. Since many dyes fluoresce in the IR, false color adjustments would be useful. Fancier still one could have a tunable dichromatic source and continuously monitor resolution and contrast of a doubly stained source.
Sensitivity considerations

1- Stray light
Stray light is any light reaching the detector other than fluoresced light from the stained subject. It limits sensitivity. Stray light sources can be detected by exposing film to the light source in the absence of sample or simply observing the gel and filters during photography. Sources of stray light are:
a- Light leaks in the darkroom or photographic hood
b- Fluorescence of paper, cloth, etc. in the darkroom
c- Poor exciter filter
d- Fluorescence in barrier filter
e- Reflected light
f- Unbound dye
g- Contaminants in the gel (intrinsic fluorescence of agarose) or staining reagents
NOTE: The human eye is an excellent visible light detector. Ignoring visually obvious stray light is a common darkroom oversight.

2- We can detect as little as 100 pg of nucleic acid on polyacrylamide gels. Fluorescence free agarose would improve sensitivity and utility of DNA analysis. More sensitive film would stimulate development of more sensitive techniques.

3- Useful hint - If bands diffuse during staining, use a more concentrated (or gradient) gel to selectively restrict macromolecular diffusion.

Relationship between colors and wavelengths

fluorescence detection, gel electrophoresis