|Home | About | Journals | Submit | Contact Us | Français|
Immunofluorescence microscopy of cultured cells often gives poor preservation of delicate structures. We have obtained dramatically improved results with a simple modification of a standard protocol. Cells growing on a coverslip are rapidly dehydrated in a cold organic solvent, and then are rehydrated in a solution containing a homobifunctional crosslinker. The crosslinking reaction stabilizes cellular structures during subsequent incubation and wash steps, usually without compromising antigenicity. This method reproducibly yields high-quality images of endomembrane compartments and cytoskeletal elements.
Studies of the secretory pathway rely on methods to determine the intracellular locations of secretory cargo proteins and of trafficking machinery components. A common approach is to overexpress fluorescently tagged proteins, but both overexpression and tagging can cause aberrant localization. Immunofluorescence microscopy of cultured cells avoids this limitation by detecting endogenous proteins at their normal expression levels .
Unfortunately, the quality of immunofluorescence images is often poor. This problem can be ascribed to the gentle fixation procedures that are used to retain antigenicity. The most commonly used fixative is formaldehyde . However, when we fixed cultured mammalian cells with formaldehyde to visualize transitional ER sites (tER sites; also known as ER exit sites), much of the information was lost. As judged by GFP tagging, discrete tER sites were present in the cell periphery and concentrated in the juxtanuclear Golgi region [2, 3]; yet when tER sites were viewed by immunofluorescence microscopy, the peripheral sites were often invisible while the juxtanuclear sites appeared as a diffuse blob. If the cells were fixed with organic solvents rather than formaldehyde , the results were better but still not satisfactory. Our troubleshooting suggested that cellular architecture was being disrupted by liquid flow during the wash steps. We therefore developed an improved immunofluorescence method that incorporates a chemical crosslinking reaction. This method reliably preserves tER sites as well as other cellular structures, including Golgi compartments and microtubules .
Our approach is to dehydrate cultured cells and then rehydrate them in a fixative solution. Cells grown on a coverslip are transferred to acetone or methanol at −20°C. This treatment appears to extract the lipids almost instantaneously while precipitating the proteins in place . The dehydrated cells are then rehydrated in the presence of a homobifunctional amine-reactive crosslinker. At this point, the sample can be processed using standard procedures for antibody labeling and DNA staining. The crosslinking reaction strongly stabilizes cellular architecture, typically without compromising antibody binding [2, 3]. A representative image is shown in Fig. 1, which displays tER sites, microtubules, and nuclear DNA in a dividing cell.
This work was supported by NIH grant GM-61156. The anti-Sec13 antibody was a kind gift of Bor Luen Tang and Wanjin Hong (National University of Singapore).
1Most of our experiments have employed single-well SecureSlip™ coverslips (cat. no. MSR12–0.5), which can be placed in individual wells of a 6-well culture dish. The multi-well versions (such as cat. no. MSR-12) are suitable for placing in a Petri dish.
2We have used traditional glycerol-based mounting media, but newer formulations that eliminate refractive index mismatch may significantly improve image quality .
3Octyl glucoside is optional but reduces background labeling with some antibodies. This detergent will not solubilize transmembrane proteins at the concentration used. Octyl glucoside readily absorbs water from the air and should be stored with desiccation. If the octyl glucoside powder was refrigerated, the bottle should be warmed completely to room temperature before opening.
4Other additives can be also used to make a blocking buffer. The formulation given here has worked well for us with a variety of antibodies.
5The confluency and culture medium can be varied as needed for the purposes of the experiment.
6Many antigens are visualized well with acetone, but some are visualized better with methanol, which tends to denature proteins more extensively. The choice of organic solvent should be made empirically.
7Removing the coverslip from the culture dish can be tricky. The easiest method is to lift one edge of the coverslip with a syringe needle before grabbing the coverslip with forceps.
8If a significant amount of culture medium is transferred with the coverslip, the dehydration will initially be incomplete, and the mixture of water and organic solvent will disrupt cellular architecture. It is therefore essential to remove as much as possible of the culture medium before dropping the coverslip into the organic solvent. However, the cells cannot be allowed to dry completely.
9If desired, coverslips can be left for days or even weeks in organic solvent at −20°C.
10After prolonged storage of a SecureSlip™ coverslip in organic solvent, the silicone backing may detach. In this case, dry the coverslip and silicone backing separately, and then reattach the silicone backing, taking care not to perturb the cells.
11Some epitopes might be blocked by treatment with BS3, which reacts mainly with lysine side chains. If the staining is weak after BS3 treatment, rehydration can be done with PBS+ lacking BS3. In this case, the washes should be exceptionally gentle to avoid disrupting cellular architecture.
12Ethylenediamine is a very potent quencher of amine-reactive crosslinkers. More common quenchers such as glycine would presumably also be effective, but the incubation period might need to be prolonged.
13If any droplets of liquid remain, they may not mix completely with the mounting solution, and the resulting refractive index mismatch will distort the images.
Dibyendu Bhattacharyya, Dibyendu Bhattacharyya, Ph.D., Department of Molecular Genetics and Cell Biology, The University of Chicago, 920 East 58th Street, Chicago, IL 60615.
Adam T. Hammond, Institute for Biophysical Dynamics, The University of Chicago, 929 East 57th St., Chicago, IL 60637.
Benjamin S. Glick, Department of Molecular Genetics and Cell Biology, The University of Chicago, 920 East 58th Street, Chicago, IL 60615.