- Social Sciences
- African Studies
- American Studies
- Asian Studies
- Communication Sciences
- Ethnic Studies
- European Studies
- Gender Studies
- Physical Sciences
- Life Sciences
- Animal Communications
- Cell Biology
- Evolutionary Biology
- Food Science and Technology
- Human Anatomy
by Jean-Marc Fritschy, Wolfgang Härtig
Immunofluorescence is one of many techniques in biomedical research and diagnostics that makes use of the sensitivity and selectivity of fluorescence for analysing biological tissues. Its main application is the detection of proteins and other biomolecules in cells and tissues. Closely related techniques include the use of fluorescent reporter gene products, such as green fluorescent protein (GFP), for probing gene expression, tracing cell lineage or as fusion tags to monitor protein localization within living cells, as well as fluorescence in situ hybridization (FISH), to detect nucleic acids (DNA and RNA) with fluorochromated nucleotide probes. Another important application of fluorescence in tissues is ion (ratio) imaging, which takes advantage of fluorescent probes sensitive to the concentration of specific ions (e.g. calcium) to monitor their concentration in the cytoplasm and organelles of single living cells. See also Green Fluorescent Protein (GFP), Fluorescence in situ Hybridization, and Fluorescent Probes Used for Measuring Intracellular Calcium
Immunofluorescence is based on the high selectivity and affinity of antibodies for their antigens as specific cellular constituents, notably proteins. Antibodies are typically raised against purified antigen preparations, recombinant proteins, or synthetic peptides coupled to a carrier protein. Optimal detection depends on the quality of the antibodies (high titre and affinity), the preservation of the tissue or cells to be examined, the sensitivity of the assay, the quantum yield and photostability of the fluorochrome, and the lack of autofluorescence. Equally important is the quality of the detection devices (fluorescence microscope, flow cytometer), which will also be reviewed in this article.
The sensitivity of immunofluorescence can be greatly improved by increasing the number of fluorescent molecules per antigen to be detected. This is usually achieved by indirect immunofluorescence (Figure 1). With this technique, the first (primary) antibodies directed against the target structures or molecules are unlabelled. They are then bound by secondary antibodies raised against immunoglobulins of the host species used for the primary antibodies. The secondary antibodies are labelled either with fluorochromes or with other haptens (e.g. biotin, digoxin), which serve as anchoring sites for enzymes or fluorescent molecules, or as targets for a third antibody. Since the primary antibodies can bind more than one secondary antibody, the signal is amplified. Furthermore, this approach provides considerable versatility, since a given primary antibody can be used in combination with innumerable secondary antibodies and detection procedures.
Figure 1. Signal amplification by indirect immunofluorescence. (a) Direct immunofluorescence: the number of fluorochrome molecules bound to the primary antibody is restricted. (b) Indirect immunofluorescence: the primary antibody is unlabelled, and binds several fluorochromated secondary antibodies, resulting in signal amplification.
The first useful fluorochromated secondary antibodies were prepared by reacting fluorescein isothiocyanate (FITC) with IgGs and could be analysed by excitation with blue light, causing the emission of green light. The availability of secondary antibodies conjugated to red fluorescent rhodamines enabled later the double immunofluorescence labelling of relevant antigens. Subsequently, the introduction of Texas Red-tagged antibodies allowed an improved separation of green and red fluorescence. Carbocyanine (Cy)-conjugated immunoreagents, which display a bright and stable fluorescence, are an excellent tool for indirect immunofluorescence labelling (Wessendorf and Brelje, 1992). Cy5, which emits in the far red range, provides a third colour for triple labelling. The strongly fluorescent Alexa dyes were subsequently developed for coupling to immunoreagents (Panchuk-Voloshina et al., 1999). An entire panel of Alexa Fluor dyes emitting in the visible spectrum, ranging from blue to red, is available. They comprise the brightest fluorochromes available and offer remarkable photostability, pH insensitivity, and water solubility. In addition, their absorption spectra are matched to the principal output wavelengths of common excitation sources. See also Fluorescent Analogues in Biological Research, and Immunofluorescence: Dyes and Other Haptens Conjugated with Antibodies
With some oversimplification, the fluorophores that are preferentially conjugated to antibodies might be subdivided in four main classes according to their spectral properties:
- Blue fluorescent dyes excited at wavelengths in the ultraviolet (UV) part of the spectrum (e.g. AMCA (7-amino-4-methylcoumarin-3-acetic acid); Alexa Fluor 350).
- Green fluorescent dyes excited by blue light (e.g. FITC, Cy2, Alexa Fluor 488).
- Red fluorescent dyes excited by green light (e.g. rhodamines, Texas Red, Cy3, Alexa Fluor dyes 546, 564 and 594)
- Dyes excited with far-red light (e.g. Cy5), to be visualized with electronic detectors (CCD cameras, photomultipliers)
Appropriate combination of these fluorochromes enables up to four markers to be visualized simultaneously in a single preparation.
When selecting fluorochromes for a practical application, several factors have to be considered in addition to their spectral properties. These include fluorescence intensity and stability, as well as sensitivity to environmental influences. Fluorescence intensity depends on the molar extinction coefficient (epsi) for absorption and the quantum yield for fluorescence. Both are constants under specific environmental conditions. The value of epsi is specified at a single wavelength (usually the absorption maximum), whereas quantum yield is a measure of the total photon emission over the entire fluorescence spectral profile. These parameters range from about 5000 to 200 000 cm–1 L mol–1 for epsi and 0.05 to 1.0 for quantum yield for the most widely used fluorochromes.
Fluorescence stability is limited by photobleaching, which refers to the irreversible destruction of the excited fluorophore. Photobleaching is one of the main factors limiting fluorescence detectability. The best way to minimize photobleaching is to maximize detection sensitivity, which allows the excitation intensity to be reduced. As described below under ‘Fluorescence microscopy’, detection sensitivity can be enhanced using high numerical aperture objectives, wide emission bandpass filters, and sensitive detectors, such as CCD cameras. ‘Antifade’ reagents offer an additional way to reduce photobleaching (Longin et al., 1993); however, their effectiveness in a given preparation may be difficult to predict and they are not compatible with living cells. To limit photobleaching, the region to be visualized in fluorescence should be selected first using phase contrast or Nomarski illumination of the specimen.
Among environmental factors influencing fluorescence properties, the solvent polarity and the pH of the aqueous medium appear to the most important (see, e.g. Brismar et al., 1995).
For optimizing fluorescence signals, antigen retrieval and signal amplification are considered crucial procedures (Fritschy et al., 1998; McNicol and Richmond, 1998). In practice, antigen retrieval is based on microwave irradiation (Werner et al., 1996), whereas signal amplification is achieved using avidin-biotin or antibody-hapten secondary detection techniques, or using secondary antibodies conjugated to enzymes in conjunction with fluorogenic substrates. Increased sensitivity may also be achieved by applying a tertiary fluorochrome-labelled antibody directed against the host species of the secondary antibodies tagged with the same fluorochrome (Svensson, 1991) or with the tyramide amplification procedure (Van Gijlswijk et al., 1997). It is important to note that simply increasing the probe concentration can be counterproductive, since this often affects the chemical and optical characteristics of the probe. In addition, increased labelling of antibodies may reduce their specificity and binding affinity.
Major application domains for immunofluorescence labelling are histology and cytology, diagnostics, and cell sorting.
Immunofluorescence of tissue sections is described in more detail in another article. Histology and cytology can be applied to a large variety of tissues and cell types, derived either from organisms or cell cultures. The range of applications is enormous and encompasses the entire range of multicellular organisms. Visualization with fluorescence microscopy provides a detailed view of the localization of one or several antigens of interest as exemplified by Figure 2a. Plants represent special challenges, notably due to the cell wall, which is a barrier to both fixatives and antibodies and has to be partially digested for optimal results, and to the presence of chlorophyll, which has spectral properties comparable to rhodamine-type dyes. Animals and embryos too small to be cut can be stained as whole-mounts, using procedures similar to those used for tissue sections. For invertebrates, such as worms and insects, penetration of antibodies through the cuticle or exoskeleton has to be ensured.
Figure 2. Examples of indirect immunofluorescence staining. (a) Double immunofluorescence of astrocytes (arrowheads) and microglia (double arrowheads) in rat cerebral cortex, using rabbit antibodies directed against glial fibrillary acid protein (green) and monoclonal mouse anti-keratan sulfate (red). These primary antibodies were visualized by Cy2-tagged goat anti-rabbit IgGs and Cy3-conjugated goat anti-mouse IgGs, respectively. The corresponding images were captured separately using a CCD camera and were digitally superimposed. (b) Combination of immunofluorescence staining with lectin-histochemistry, visualized by laser scanning confocal microscopy. The calcium-binding protein parvalbumin (red) was detected in a neuron of rat cerebral cortex with monoclonal antibody followed by goat anti-mouse IgGs conjugated to Cy3. Perineuronal nets surrounding these cells (green) were revealed by applying biotinylated Wisteria floribunda agglutinin and subsequently a Cy2-streptavidin conjugate. Stacks of images corresponding to each marker were acquired simultaneously using a dual-channel recording system (Leica TCS). The panel represents the superposition of eight images spaced by 0.3 mum. (c) Indirect immunofluorescence of the synaptic marker PSD-95 in a neuron in primary culture. The marker was revealed with a monoclonal antibody followed by goat anti-mouse IgGs coupled to Alexa Fluor 488 and visualized by video microscopy. The arrow points to the cell body. Scale bars: (a) and (c) 25 mum; (b) 10 mum.
Compared to isolated cells, tissue sections pose a number of difficulties for immunofluorescence staining: the section represents a physical barrier limiting the penetration of antibodies to a few micrometres from the surface; it has to be fixed well to withstand the staining procedure; it is likely to contain sites of nonspecific binding for both primary and secondary antibodies; autofluorescence might represent a major confounding problem. Some of these limitations can be partially circumvented using the following strategies:
- Thinner sections can be cut, depending on the cell type being examined.
- Fixation should always be carried out with freshly prepared fixatives.
- Whenever possible, sections should be stained ‘free-floating’ to ensure antibody penetration from both sides.
- Antibody penetration can be enhanced - at the cost of morphological preservation - by repeated freeze-thaw cycles or by the use of detergents.
- Cryostat sections prepared from fresh-frozen tissue can be fixed by immersion. This procedure is only suitable for nonsoluble antigens (e.g. structural proteins)
- Since antibodies can be rapidly destroyed by bacteria, incubation times should be kept to a minimum, or should be carried out in media containing sodium azide or other antibacterial agents.
Control experiments to verify staining specificity are of major importance in histological studies using immunofluorescence. (1) Examination of unstained sections in the fluorescence microscope will provide information about the source and level of autofluorescence. (2) In indirect immunofluorescence, omission of the primary antibodies will reveal the extent of nonspecific binding of secondary antibodies to the tissue section. (3) Whenever possible, replacement of the primary polyclonal antibodies with crude preimmune serum from the same animal should be done to verify the absence of staining unrelated to the antigen of interest. (4) Preadsorption of the primary antibodies with their antigen should reveal a dose-dependent reduction in specific staining. However, this will by no means prove that the primary antibodies recognize only the antigen of interest. (5) The best control for specificity is the analysis of a tissue known to be devoid of the antigen of interest; for instance, sections from mutant animals carrying a gene-deletion for this protein provide an optimal control.
Immunofluorescence staining can be combined with a large variety of histological techniques that rely also on fluorescence signals for detection of specific probes. For instance, the combination with lectin-histochemistry (e.g. Härtig et al., 1994; Figure 2b), in situ hybridization, and the rapidly growing number of fluorescent probes for specific organelles and cell constituents (Johnson, 1998), in particular DNA (Schnell and Wessendorf, 1995; Suzuki et al., 1997), is attracting considerable interest. Likewise, immunofluorescence can be combined with neuroanatomical tracing techniques based on fluorescent tracers (e.g. Fluoro-gold, latex microspheres) to identify the labelled projection of a neuron. Immunofluorescence labelling might also be applied in combination with the detection of enzymatic activity based on fluorogenic substrates (Haugland, 1995). For the detection of the same markers at light- and electron-microscopic level, it is useful to convert fluorescence signals into electron-dense reaction products by photooxidation (Sandell and Masland, 1986; Kacza et al., 1997), or by immunoenzymatic detection of either antibodies directed against fluorophores (van der Loos et al., 1989), or fluorochrome-tagged antibodies (Pilowsky et al., 1991) or fluorescent tracers (Chang et al., 1990). See also Cell Staining: Fluorescence imaging of mitochondrial structure and function, Cell Staining: Fluorescent Labelling of Endocytic Compartments, Cell Staining: Fluorescent Labelling of the Endoplasmic Reticulum, Cell Staining: Fluorescent Labelling of the Golgi Apparatus, and Immuno-electron Microscopy
The main applications of immunofluorescence for clinical diagnosis are in the fields of oncology and immunology, notably for the analysis of cell-surface expression of antigens in haematopoietic cells or for the detection of antibodies in autoimmune diseases, such as myasthenia gravis. Microbiology, virology and mycology also rely heavily on immunofluorescence for the identification of pathogens. In clinical practice, a large number of diagnostic procedures also depend on histological analysis of tissue samples, although detection methods based on enzymatic reactions with stable reaction products are usually preferred. For instance, immunoperoxidase staining is more sensitive, can be visualized by conventional bright-field microscopy, and is more stable for long-term storage.
Immunofluorescence staining finds most applications when combined with flow cytometry, which underwent a tremendous expansion during the 1990s. The majority of clinical applications involve measuring cell-surface epitopes in the analysis of haematopoietic cells or analysing DNA content analysis of malignant cells. While DNA content flow cytometric analysis is usually not performed with antibodies, the detection of DNA and RNA probes labelled with a fluorochrome is based on similar principles as immunofluorescence. Measurement of intracellular antigens by flow cytometry has not been extensively employed so far, but the application is expending. The restriction is mostly due to methodological limitations linked with the restricted access of antibodies to intracellular (cytoplasmic and nuclear) compartments (Bauer and Jacobberger, 1994).
Compared to histological studies, diagnostic studies have other specific requirements. Speed can be of crucial importance, and it can be achieved at the cost of signal-to-noise ratio. In diagnosis, the precise localization of the antigen of interest is often less important, since the question usually is whether a selected marker is present in the tissue or cell population to be examined. The time required for the immunofluorescence staining procedure can be markedly reduced by increasing the concentration of antibodies and the temperature of the incubation (e.g. from 4°C to 37°C).
Whether in histology or flow cytometry, the analysis of intracellular probes requires cell fixation and permeabilization to generate pores in the plasma membrane sufficiently large to provide access of antibodies to the cell interior (see for example, Shankey, 1994). These steps are crucial and often determine the quality and reproducibility of immunofluorescence because the majority of antigens are sensitive to fixation.
Cell sorting based on a specific immunofluorescence pattern is a major application of flow cytometry. If the antigens of interest are cell-surface markers or extracellularly located, live cells can be labelled with antibodies (using either direct or indirect immunofluorescence staining) and sorted by flow cytometry. The yield of purified cells is limited, however, by the considerable mechanical stress imposed on the cells. An alternative method, which is gaining increasing popularity, is sorting based on biomagnetic separation technology (Thiel et al., 1998). This technique makes use of antibodies coated with magnetic beads for separation and isolation of cells (Wright et al., 1997; Wang, 1998) or organelles (Kausch et al., 1999) with magnetic equipment.
Detection Techniques for Immunofluorescence
Four elements are required in fluorescence detection systems: an excitation source, a fluorochrome, wavelength filters to isolate emission photons from excitation photons, and a detector to register emitted photons and produce a recordable output.
The basic functions of a fluorescence microscope are to deliver excitation energy to the fluorochrome(s) in the specimen to be investigated, to separate the much weaker emitted fluorescence light from the excitation light, and to send it to the detector, where a high-contrast image is generated (Figure 3). The specificity and high sensitivity of fluorescence microscopy allows the investigation of cellular and subcellular structures in biological tissues with high resolution. It provides spatial information in three-dimensions and, in living cells, temporal changes in the spatial information (‘four-dimensional’ images). The use of fluorescence microscopy offers a number of advantages.
Figure 3. Epifluorescence microscopy. The (relatively strong) excitation light is directed to the specimen by reflection on the dichromatic beam-splitting mirror and focusing through the objective. The (relatively weak) emission light, with a longer wavelength, is separated as it passes the mirror and reaches the photodetector.
- Specificity. Fluorescence excitation and emission spectra of a given fluorochrome are characteristic for that molecule. Different fluorochromes with distinctexcitation and emission spectra can be used for the analysis of complex mixtures of molecular species.
- Sensitivity. The detection threshold of fluorescence can be reached with only small numbers of fluorescent molecules.
- Quantification. Quantitative measurements are feasible because the emitted fluorescence is directly related to the quantum yield of the fluorophore and hence to the number of fluorescent molecules excited in the tissue.
- High spatial resolution. Fluorescence images of high resolution can be acquired by confocal laser scanning microscopy (see below), which removes the out-of-focus blur emitted by the tissue below and above the focal plane. Interactions between cellular components with dimensions below the diffraction-limited resolution of the light microscope can be visualized using fluorescence resonance energy transfer (FRET) techniques.
Three interrelated parameters determine the quality of the image generated by the fluorescence microscope: (1) the numerical aperture (NA) of the lens, where NA = eta sin alpha, eta being the refractive index of the medium between the specimen and the objective and alpha the maximal angle of the emitted fluorescence relative to the direction of exciting light that is captured by the lens; (2) the resolution (d), which is defined as the smallest distance between two objects that can still be discerned as two separate objects: d = 0.61(lambda/ NA), where lambda is the wavelength of the light (for example, d = 0.3 mum with lambda = 550 nm and NA = 1.3); (3) the magnification, which is defined as the apparent enlargement of an object by an optical instrument. A useful magnification ranges between 500 and 1000 times the NA. Additionally, it is essential that the (relatively weak) emitted light be fully separated from the (intense) excitation light to generate a high contrast image. This is achieved by epifluorescence microscopy, in which the exciting light is reflected into the back aperture of the objective by a dichromatic beam-splitting mirror. The emitted fluorescence is collected by the objective and, since its wavelength is of lower energy (longer wavelength), it will be transmitted through the dichromatic mirror and reach the detector.
Excitation light sources usually are mercury and xenon lamps. The former emit peaks of energy at discrete wavelengths (e.g., 365, 400, 440, 546 and 580 nm), whereas the latter have a uniform intensity profile from the ultraviolet to the far red. The appropriate excitation and emission wavelengths are selected with the corresponding bandpass filters. More sophisticated devices, such as acousto-optical tuneable filters or liquid-crystal tuneable filters, have been introduced to replace glass filters.
Laser scanning confocal microscopy
The confocal microscope is an established tool for biomedical research, providing improved light-microscopic imaging of cells labelled with fluorescent dyes. Its main application is the high-resolution analysis of subcellular constituents in three dimensions. Although the principle of confocal microscopy was patented as early as 1957, its practical development followed the introduction of powerful microcomputers and image analysis software (Paddock, 1999).
The major difference from conventional epifluorescence microscopes lies in the method of image formation. While the former provide an image of the entire specimen, confocal microscopes focus a light spot that is scanned line by line across the specimen; the term laser scanning microscopy refers to the fact that the laser beam is scanning the immobile specimen. The image is formed from the output of a photodetector collecting the light emitted from the scanned specimen, displayed on a computer screen, and stored on a hard disk for further processing. To remove out-of-focus information originating from above and below the focal plane, the emitted fluorescence passes through a pinhole before reaching the detector. If the pinhole is small enough and the specimen bright enough, the resolution of confocal microscopy can be improved by a factor of up to 1.4 compared to conventional microscopy.
In addition to the optical characteristics of the microscope, the performance of confocal laser scanning microscopes is dependent on stable multiwavelength lasers for bright point sources of light, on efficiently reflecting mirrors, on sensitive detectors and on fast computing capabilities for image analysis and display.
Flow cytometry is a technique for deriving measurements of physical or chemical characteristics of cells and other biological particles while they pass in single file in a fluid stream through a measuring apparatus. The measurement of immunofluorescence is a particular application of flow cytometry. It is especially useful for cell sorting, based on the presence or absence of a fluorescent signal, and allows precise definition and quantification of heterogeneity in cell populations.
Schematically, a flow cytometer comprises three components arranged in an orthogonal configuration: the sample flow (carrying the cells to be measured), the laser beam (for excitation of the fluorochrome), and the optical axis of fluorescence detection, the latter two being right angles to each other (Figure 4). As in the fluorescence microscope, this ensures that emitted fluorescence is completely separated from the much stronger excitation fluorescence. The laser beam is focused on the sample with one or two lenses. Likewise, emitted fluorescence is collected with an objective having a high numerical aperture. Multiple fluorescence measurements are possible, using dichroic prisms to separate the two emission wavelengths and two detectors.
Figure 4. Flow cytometry. A pressurized jet of cells in suspension crosses a laser beam. Emitted fluorescence is captured by a photodetector and the signal is digitized and visualized on a monitor (as fluorescence intensity as a function of time).
Flow cytometers exhibit remarkable performance characteristics. Their detection limit is about 1000 dye molecules per cell, representing 10–18 g of dye per cell. This amount is measured in 1 mus at an average rate of 104 cells s–1. The sensitivity and reproducibility of flow cytometers are limited by intensity and stability of excitation, effective numerical aperture of the collecting objective, sensitivity of the detector and stability of the sample flow. Therefore, depending upon the specific requirements of investigators, immunoflourescence can provide excellent spatial resolution (microscope) or sensitivity (flow cytometry).
Mirror that reflects light below a certain wavelength (usually excitation light) but transmits light above a certain wavelength (usually emission light).
Refers to a popular method of differential interference contrast microscopy, used to visualize either reflective objects (e.g. metallic surfaces) or transparent objects containing organelles differing in refractive index.
A measure to define how fast the irreversible chemical destruction (photobleaching) of an excited chromophore occurs.
Ratio of the intensity between the specific signal and the background.
Tyramide amplification procedure
Enzymatic amplification procedure based on the use of horseradish peroxidase (indirectly bound to the primary antibody) to catalyse the deposition of biotin- or fluorophore-labelled tyramide onto tissue sections. Since numerous biotin or fluorescent molecules are deposited in the vicinity of the enzyme bound to the primary antibody, the signal is amplified considerably compared to conventional indirect immunofluorescence without significant loss of spatial resolution.
Bauer KD and Jacobberger JW (1994) Analysis of intracellular proteins. Methods in Cell Biology 41: 351–376.
Brismar H, Trepte O and Ulfhake B (1995) Spectra and fluorescence lifetime of lissamine rhodamine, tetramethylrhodamine isothiocyanate, Texas Red, and cyanine 3.18 fluorophores: influences of some environmental factors recorded with a confocal laser scanning microscope. Journal of Histochemistry and Cytochemistry 43: 699–707.
Chang HAT, Kuo H, Whittaker JA and Cooper NGF (1990) Light and electron microscopic analysis of projection neurons retrogradely labelled with Fluoro-Gold: notes on the application of antibodies to Fluoro-Gold. Journal of Neuroscience Methods 35: 31–37.
Fritschy JM, Weinmann O, Wenzel A and Benke D (1998) Synapse-specific localization of NMDA- and GABAA-receptor subunits revealed by antigen-retrieval immunohistochemistry. Journal of Comparative Neurology 390: 194–210.
Härtig W, Brauer K, Bigl V and Brückner G (1994) Chondroitin sulfate proteoglycan-immunoreactivity of lectin-labelled perineuronal nets around parvalbumin-containing neurons. Brain Research 635: 307–311.
Haugland RP (1995) Detecting enzymatic activity in cells using fluorogenic substrates. Biotechnic and Histochemistry 70: 243–251.
Johnson I (1998) Fluorescent probes for living cells. Histochemical Journal 30: 123–140.
Kacza J, Härtig W and Seeger J (1997) Oxygen-enriched photoconversion of fluorescent dyes by means of a closed conversion chamber. Journal of Neuroscience Methods 71: 225–232.
Kausch AP, Owen TP, Narayanswami S and Bruce BD (1999) Organelle isolation by magnetic immunoabsorption. Biotechniques 26: 336–343.
Longin A, Souchier C, Ffrench M and Bryon PA (1993) Comparison of anti-fading agents used in fluorescence microscopy: image analysis and laser confocal microscopy study. Journal of Histochemistry and Cytochemistry 41: 1833–1840.
McNicol AM and Richmond JA (1998) Optimizing immunohistochemistry: antigen retrieval and signal amplification. Histopathology 32: 97–103.
PaddockSW (ed) (1999) Confocal Microscopy: Methods and Protocols. (Methods in Molecular Biology, vol. 122). Totowa, NJ: Humana Press.
Panchuk-Voloshina N, Haugland RP, Bishop-Stewart J et al. (1999) Alexa dyes, a series of new fluorescent dyes that yield exceptionally bright, photostable conjugates. Journal of Histochemistry and Cytochemistry 47: 1179–1188.
Pilowsky PM, Lipski J, Prestidge R and Jiang C (1991) Dual fluorescence combined with a two-color immunoperoxidase technique: a new way of visualizing diverse neuronal elements. Journal of Neuroscience Methods 36: 185–193.
Sandell JH and Masland RH (1986) Photoconversion of some fluorescent markers to a diaminobenzidine product. Journal of Histochemistry and Cytochemistry 36: 555–559.
Schnell SA and Wessendorf MW (1995) Bisbenzimide: a fluorescent counterstain for tissue autoradiography. Histochemistry 103: 111–114.
Shankey TV (1994) Antibodies to intermediate filament proteins as probes for multiparameter flow cytometry of human cancers. Methods in Cell Biology 42 (part B): 209–229.
Suzuki T, Fujikura K, Higashiyama T and Takata K (1997) DNA staining for fluorescence and laser confocal microscopy. Journal of Histochemistry and Cytochemistry 45: 49–53.
Svensson MAJ (1991) Increased sensitivity of the indirect immunofluorescence method by use of a tertiary fluorochrome-labelled antibody. Journal of Histochemistry and Cytochemistry 39: 235–237.
Thiel A, Scheffold A and Radbruch A (1998) Immunomagnetic cell sorting - pushing the limits. Immunotechnology 4: 89–96.
van der Loos CM, Das PK, Van den Oord JJ and Houtloff H-J (1989) Multiple immunoenyzme staining techniques. Use of fluoresceinated, biotinylated and unlabelled monoclonal antibodies. Journal of Immunological Methods 117: 45–52.
Van Gijlswijk RPM, Zijlmans HJMAA, Wiegant J (1997) Fluorochrome-labeled tyramides: use in immunocytochemistry and fluorescence in situ hybridisation. Journal of Histochemistry and Cytochemistry 45: 375–382.
Wang EC (1998) Sorting of human peripheral blood T-cell subsets using immunomagnetic beads. Methods in Molecular Biology 80: 365–376.
Wright AP, Fitzgerald JJ and Colello RJ (1997) Rapid purification of glial cells using immunomagnetic separation. Journal of Neuroscience Methods 74: 37–44.
Werner M, von Wasielewski R and Komminoth P (1996) Antigen retrieval, signal amplification and intensification in immunohistochemistry. Histochemistry and Cell Biology 105: 253–260.
Wessendorf MW and Brelje TC (1992) Which fluorophore is brightest? A comparison of the staining obtained using fluorescein, tetramethylrhodamine, lissamine rhodamine, Texas Red, and cyanine 3.18. Histochemistry 8: 81–85.
- DarzynkiewiczZ, RobinsonJP and CrissmanHA (eds) (1994) Flow Cytometry, Parts A and B (Methods in Cell Biology, vols 41 and 42). Orlando: Academic Press.
- Herman B (1998) Fluorescence Microscopy, 2nd edn. New York: Springer Verlag.
- Hockfield S, Carlson S, Evans C et al. (1993) Molecular Probes of the Nervous System, vol. 1, Selected Methods for Antibody and Nucleic Acid Probes. New York: Cold Spring Harbor Laboratory Press.
- Matsumoto B (1993) Cell Biological Applications of Confocal Microscopy. (Methods in Cell Biology, vol. 38). Orlando: Academic Press.
- Robinson JP (1997) Current Protocols in Cytometry. New York: Wiley.
- Slavik J (1996) Fluorescence Microscopy and Fluorescent Probes. New York: Plenum Press.