 Methods & Techniques For Histologists and Immunohistochemists |
Cell Biology Applications of Fluorescence Microscopy
Stephen Rogers
Manager, Microscopy Suite, Light and Confocal Microscopy
Imaging Technology Group
Beckman Institute for Advanced Science and Technology
University of Illinois at Urbana-Champaign
405 N. Mathews, Urbana, IL 61801
Techniques for Staining Specific Cellular Structures
Microtubules
Microtubules are highly dynamic structures within the cell, and can be very difficult to preserve satisfactorily. Fixation in ice-cold methanol for 10 minutes works well, but distorts their three-dimensional structure. Crosslinking fixatives are, therefore, better suited for confocal work. Formaldehyde, however, does not preserve microtubules well and so glutaraldehyde is the best choice. It is important to warm the fixative to the physiological temperature of the cells (i.e. 37 C for mammalian cells) as microtubules spontaneously disassemble at lower temperatures. Microtubules are also picky about the buffer used for the fixative - they are best preserved in PIPES, pH 6.9 or in imidazole. Following fixation, I have found that permeabilization with 0.5% SDS for 10 minutes consistently yields the best results in terms of uniformity of staining as well as decreased background fluorescence. Numerous tubulin-specific antibodies are available commercially, the best of them being DM1a, a mouse monoclonal antibody. Microtubule preservation may be evaluated by examining their ends located in the cell periphery. In a good fixation, they should appear as long continuous filaments, not as beaded, broken segments. In addition, the astral microtubules of mitotic spindles should be present and distinct.
Actin
Filamentous actin may be efficiently labeled using fluorescently-conjugated phallotoxins - bicyclic peptides isolated from the poisonous mushroom Amanita. The most commonly used member of this family, phalloidin, may be purchased conjugated to a wide variety of fluorescent dyes. Phalloidin selectively binds and stabilizes polymerized, filamentous actin without binding monomeric actin and its non-specific staining is negligible. These properties make phalloidin more attractive than actin-specific antibodies for fluorescence microscopy. Furthermore, phalloidin binds to actin from different species (including plant, animal, and fungal cells) and does not discriminate between different actin isoforms.
Phalloidin is usually purchased as a lyophilized powder and is reconstituted in methanol or DMSO. I find it convenient to prepare it as a 1000X stock of 3 uM in small aliquots and store it below 0 C. For staining, it should be diluted immediately prior to staining into PBS or whatever buffer is being used, and may be included with fluorescent secondaries for double-label experiments. Phalloidin is cell-impermeable, so specimens to be stained must be fixed and detergent-permeabilized. Cells to be stained should be fixed either with aldehydes or cold acetone, but NOT methanol. Methanol fixation destroys the phalloidin-binding site on actin, thereby eliminating staining. For this reason, many people advise that phalloidin be dried down in a vacuum microfuge, then resuspended in buffer - but I have not found this additional step to be necessary. Permeabilization with SDS also eliminates phalloidin binding, so Triton X-100 should be used for permeabilization. Phalloidin-stained specimens should be imaged within a few days of staining as it will dissociate into the mounting medium over time and produce a background autofluorescence that may obscure fine detail.
DNA
Fluorescent staining of the nuclei of cells is often useful in order to determine the status of the cell cycle of individual cells, or to provide a cytological 'landmark' for double staining, or as a counter-stain to delineate different populations of cells within tissues. There are literally dozens of DNA-specific fluorescent dyes to chose from, but I have used five of them predominantly, depending on the desired wavelength and microscope to be utilized. They are all intercalation agents, and only fluoresce when bound to nucleic acids.
The first two - DAPI and bisbenzimide (or Hoescht 33245) - are dyes that bind to DNA with very high specificity and photostability. They are cell permeable and readily cross the cell membrane. Both dyes are dissolved in de-ionized water at a concentration of 10 mg/mL to make a 1000X stock and are kept stored as aliquots below 0 C. They should be diluted into buffer and used to stain specimens for 10 to 30 minutes. Both dyes exhibit excitation wavelengths around 370 nm and emission peaks around 450 nm. These spectral characteristics make them particularly useful for triple labeling for fluorescence microscopy. [Note: the confocal microscope housed in the Beckman Institute is not currently equipped to excite dyes in the UV wavelengths.] Samples stained with these dyes should be documented with in one to two weeks after mounting as they will dissociate slowly and produce background autofluorescence in the hydrophobic conditions of some mounting media. The third is propidium iodide, which is a red dye (ex. 530 nm/em. 615 nm). PI is extremely photostable and is very robust even under the confocal microscope. It is prepared by dissolving in de-ionized water at a concentration of 10 mg/mL to make a 1000X stock and is stored as aliquots below 0 C. It should be diluted into buffer immediately prior to use and used to stain specimens for 10 to 30 minutes. PI will bind to RNA as well as DNA, so there is an additional step that must be performed when using this dye to stain nuclei. Following fixation and permeabilization, samples must be treated with RNAse (10 U/mL diluted into buffer for 30 minutes at 37 C in a humidified chamber) in order to avoid background staining of the cytoplasm. The fourth and fifth are green DNA-specific dyes - sytox green and chromomycin A3 (ex. 450 nm/em. 570 nm). Both are cell-impermeable, and so require that the specimens be fixed and permeabilized. Sytox green is available from Molecular Probes and comes as a solution in DMSO at 1000X concentration. It may be diluted and used from this stock. This dye is very bright, but is not terribly photostable. Chromomycin A3 is available from numerous sources as a lyophilized powder. It should be reconstituted in buffer (i.e. PBS) to 10 mg/mL (for a 1000X stock) supplemented with 5 mM magnesium chloride. Chromomycin A3 must be complexed to magnesium in order to bind to DNA, so all staining solutions should be supplemented accordingly. This dye is fairly photostable, and is a good choice for a green nuclear stain for confocal microscopy.
Mitochondria
I have used several different dyes to stain mitochondria, including rhodamine 123 and DiOC(6), but have found the MitoTracker dyes from Molecular Probes to be the most satisfactory for several reason. This family of dyes is available in different forms that may be selected for different wavelengths. They are very specific for active mitochondria, producing very little background staining. The MitoTrackers are cell permeable - cells are labeled by adding the stain to the culture medium for a period of time during which it accumulates in the organelles. This property makes it suitable for live cell microscopy, but it is also fixable using aldehydes, facilitating double-label immunofluorescence. Lastly, they are fairly photostable, unlike the other dyes I have tried. I cannot provide specific protocols for using MitoTracker dyes, as different cells accumulate the dye at variable rates and concentrations. Molecular probes provides protocols for determining optimal staining procedures in the technical information packaged with the product.
Golgi Apparatus
Probably the best probes to visualize the Golgi are antibodies specific for proteins that reside inside the lumen of this organelle. Several of these, such as COPII, are available commercially from many vendors.
I have had some success labeling the Golgi using fluorescently-conjugated lectins. As one of the primary functions of the Golgi is glycosylation of proteins, the Golgi apparatus and Golgi-derived vesicles may be stained using lectins specific for sugar residues present in this compartment. I have found wheat germ agglutinin (WGA) to be particularly useful as it does not require any special buffer conditions. Cells may be fixed by either precipitation or cross-linking, although I have found that the GA's morphology is best preserved by aldehyde fixation. Samples should be extracted with 1% Triton X-100 for 15 minutes in order to allow access to the lectin, as well. I have used rhodamine-conjugated WGA from Vector Laboratories, which is provided as an aqueous solution at 1 mg/mL, diluted to 10 ug/mL to stain cells for one hour. If performing double labeling with antibodies, it may be wise to perform the lectin staining after staining with secondaries as IgGs are, themselves, glycosylated and simultaneous incubation my induce formation of antibody aggregates. Using rhodamine-WGA, I have been able to visualize the juxtanuclear Golgi apparatus in many cell types. In addition, small vesicles are labeled along with the plasma membrane, although not as intensely. Unfortunately, WGA also labels the serum proteins present in the growth medium that adhere to the coverslip, producing a uniform, low level of background fluorescence. This background is usually low enough that it may be eliminated from final images by manipulating their gray scale histograms or look-up tables.
Endoplasmic Reticulum
As with the Golgi apparatus (above), the best choice of probe for labeling the ER is probably an antibody against a resident protein, such as BiP, commercially available from many sources.
I have used 3,3'-dihexyloxacarbocyanine iodide, DiOC6(3) to label the ER in cultured cell lines. This dye possesses a positively charged polycyclic head domain and two hydrocarbon tails. It is cell-permeable, and readily accumulates in intracellular membranes. At low concentrations it accumulates in mitochondria, while at higher concentrations it labels the ER, which may be identified by its tubular morphology. This dye is not fixable and is extracted by detergents, and so is not particularly useful for double-labeling with antibodies. Several laboratories have used it to study membrane dynamics in living cells, however. DiOC6(3) is a very bright dye, but is not very photostable, and I have found it useful to include N-propyl gallate in specimen preparations to retard photobleaching. The dye may be made as a 0.5 mg/mL stock in ethanol and is stored in the dark below 0 C. It may be diluted to 2.5 ug/mL into growth medium and used to stain living cells for 5 minutes. Alternatively, cells may be fixed with low (0.25%) concentrations of glutaraldehyde and then stained with 2.5 ug/mL diluted into PBS for 10 seconds. At higher concentrations, the dye begins to label cytosolic structures indiscriminately. Following labeling, samples should be mounted in PBS supplemented with 3% N-propyl gallate and imaged immediately. Cellular membranes will begin to bleb shortly thereafter, and the dye will redistribute.
References.
Virtanen, I., Ekblom, P., and Laurila, P. (1980) J. Cell Biol. 85, 429-434
Tartakoff, A., and Vassalli, P. (1983) J. Cell Biol. 97, 1243-1248
Terasaki, M. (1989) Meth. Cell Biol. 29, 125-135