Overview

Protocols for immunohistochemistry vary widely, due to the differences between antigens and their recognition by antibody. Some epitopes are destroyed by the high temperatures and organic solvents used in paraffin fixation, while others cannot survive freeze and thaw. Each fixative works by a different mechanism, which may either help to uncover the epitope, or may destroy it or make it less accessible. Some antibodies may even adhere to the blocking agent used, so that it causes rather than eliminates background. For success it is crucial to try a number of different techniques and compare the results. If at all possible, obtain all details of the procedure used by other researchers for that antibody with successful results; this can save much effort.

Once a "signal" is obtained, proving that signal truly reflects the distribution of the molecule of interest is still a matter of some difficulty. The simplest negative control is the absence of expression in tissues in which the RNA for the molecule of interest is known not to be expressed by RNase protection, and restriction of expression to regions of a target tissue that have expressed RNA as seen by in situ hybridization. A better negative control is the elimination of the signal by pre-incubating the antibody with an excess of the peptide or protein with which it was raised. A Western blot can suggest specificity of the interaction if only one band is seen; nonetheless, the difficulty of optimizing conditions for a Western blot should demonstrate the possibility that conditions in the immunohistochemistry reaction are less than optimal. On the other hand, there can also be doubts as to whether the antigen is visualized everywhere; in particular, one must determine whether the fixation protocol used actually permeabilizes the nuclear membrane to a sufficient degree for nuclear antigens to be fully visible.

This protocol represents the basic procedure recommended for proteins present in relatively high abundance. We have also made use of a more sensitive protocol, using Triton X-100 permeabilization, and gold chloride/silver developer for signal intensification.

Paraffin sections

1. After dissecting tissues, fix in 2% paraformaldehyde, Bouin's solution, or other fixative from 30 minutes to overnight. Larger tissue sections require longer periods of time; the time of fixation will affect the characteristics of sectioning the tissue (crumbling or cracking of the tissue) and possibly antigen recognition. In any case, the tissue will be exposed to ethanol for a prolonged period during dehydration, which itself causes some fixation. A basic paraffin embedding protocol is given in the Lee lab protocol for in situ hybridization. Section the tissues 5-10 micrometers thick.
2. Incubate 2-3 times in xylene for 10 minutes each.
3. Incubate twice in 100% Ethanol for 2 minutes each.
4. Hydrate by placing in 95%, 70%, 50%, 30% ethanol for 2 minutes each.
5. Optional: If endogenous peroxidases are a problem in this tissue type, and you are visualizing antigen with horseradish peroxidase, then inactivate the peroxidase by incubation for 5-15 minutes (longer for paraffin tissues) in 0.1% hydrogen peroxide in PBS (Santa Cruz protocol) , or 30 minutes in 0.3% hydrogen peroxide in methanol (Vectastain kit). Afterwards, wash in buffer (3 changes) or running water for 20 minutes. For my studies in intestinal sections, this did not appear to be necessary.
6. Place slides in buffer for 5 minutes. Buffer: 0.25 M Tris-HCl at pH 7.5; other protocols often use PBS.
7. Optional: Post-fix slides for 2 minutes in the original fixative used for the paraffin slides. This decreases the tendency of the sections to detach from the slide --- slightly. In general, treat paraffin sections in this procedure GINGERLY. Even if they do not completely detach, the tendency of edges of the tissue to lift up allows antibody to become trapped beneath the edges of the tissue, causing high background.
8. Optional: Block slides with 10% serum from the species from which the secondary antibody was taken. Incubate for 20 minutes at room temperature in a humidified chamber. Wash in buffer for 5 minutes.
9. Place slides into buffer containing 0.5% BSA and 2% Fetal Calf Serum for five minutes [note: the biotin present in this small quantity of FCS did not appear to interfere with recognition of biotinylated antibody].
10. Incubate slides in a humidified chamber overnight with primary antibody. To create the "chamber", use cut disposable pipettes to form a support for the slides above H2O saturated paper towels in a large petri dish; cover for incubations). Dilute antibody in the Tris with protein solution above. Always dilute the antibodies in a solution containing protein! Many other protocols use 3% BSA or the like. Antibody is usually diluted from 1:20 to 1:200; as with so many parts of this procedure, this must be determined empirically for a new antibody. Use Kimwipes to remove excess buffer from the slides, then add 50-60 microliters of antibody to the slide. The antibody should be restricted to the wet region of the slide by surface tension alone; if you have trouble, "PAP Pens" and other hydrophobic markers are sold by many companies.
11. Rinse slides first with a gentle stream of buffer from a squirt bottle, that flows across the sections from above.
12. Wash 5 minutes in buffer.
13. Block 5 minutes in Tris with protein.
14. Blot slides as before, and incubate with secondary antibody in a humidified chamber for 30 minutes or longer.
15. Rinse from squirt bottle and wash 5 minutes in buffer as before.

Fluorescence visualization

The secondary antibody will usually be labeled with FITC or TRITC, so the slides can be mounted after this rinse. Mounting should be done in an aqueous medium. Gel/Mount, available from Fisher, preserved my slides for over a year before they began to noticeably degrade; however, the sharpness of the image was not quite as high as with DAB and Permount. Temporary glycerol mounts, in my hand s, produced much light scattering and very poor image quality.

Peroxidase visualization

1. Mix 2 drops Vectastain "A" and 2 drops Vectastain "B" in 10 ml PBS as per kit instructions at the beginning of the secondary antibody incubation (30-60 minutes before use).
2. Incubate 1 hour with peroxidase-anti-peroxidase mix in a humidified chamber. Rinse with buffer as before.
3. Incubate in fresh DAB solution. (10 mg DAB + 20 ul 38% H2O2 in 20 ml 0.1 M Tris pH 7.2; filter; optionally, add 200 ul 1 M imidazole) Stop the reaction by washing in running water when a uniform brown color first becomes visible on the section.

Frozen sections

1. Freeze and section as in the Tyner lab in situ protocol. Fix in one (or more) of the following ways:
2. 5 minutes in Bouin's at room temperature.
3. 10 minutes in acetone at 4 degrees C.
4. 15 minutes in methanol at -20 degrees C.
5. 2 minutes in 2% paraformaldehyde at 4 degrees C.
6. 2 minutes in 4% paraformaldehyde at 4 degrees C, then 10 minutes in 70% EtOH at 4 degrees C, then dehydrate (as per in situ hybridization; a starting point for combined protocols?).
7. Incubate in three changes of fresh buffer (2 minutes each).
8. Proceed with primary antibody as for paraffin-embedded sections.