Immunohistochemistry is like cooking. There are many recipes out there, but some of them do not work out well. However, when they do, they are great! You need the right ingredients, a dose of experience, a few tricks from old cooks, and a grain of common sense. Read more to learn about traps to avoid, and important information to report when you publish your findings.
The validity of scientific results depends to a large extent on the use of appropriate methods providing the required precision and sensitivity, as well as controlled specificity of reagents and procedures. This statement is particularly valid for immunohistochemistry, where sensitivity and specificity of the antibodies, as well as technical procedure are crucial to avoid false-positive and false-negative results.
When using immunohistochemistry, several factors can cause false-negative or false-positive results and all should be verified as much as possible in each experimental set-up used. In particular, primary antibodies can fail to detect their target antigen even if the antigen is present in the tissue for many reasons, including conformation changes induced by fixation/embedding, steric hindrance by interacting proteins/post-translational modifications, low affinity of the antibody for the target, or failure of the antibody to penetrate into the tissue. Conversely, antibodies can bind non-specifically to other targets or tissue components. This observation holds true for both primary and secondary antibodies.
The problem is further confounded by the vast literature describing various procedures – such as the use of distinct “blocking” reagents and antigen-retrieval – to minimize these pitfalls and ensure high-fidelity binding of antibodies. This makes it difficult to choose the adequate method, and testing alternatives can be laborious. However, blindly following an established protocol might prove insufficient. It requires considerable experience, rational thinking and evidence from other methods to determine whether a given staining pattern obtained by immunohistochemistry is likely specific or non-specific. For instance, antibodies directed against a synaptic protein should produce no staining of the cell nucleus (unless this protein plays an additional, previously not described role in this compartment). In such case of wrong labeling, either the antibody or the staining protocol may be unsuitable.
Here, we provide a practical guide of what should be tested and what should be reported in order to provide convincing evidence for the validity of immunohistochemical experiments. This includes dealing with the three main steps that can lead to false negative and false positive results:
- Detection of the antigen of interest by the primary antibody
- Detection of the primary antibody by secondary antibodies
- Tissue preparation
1) Primary antibody specificity
The most stringent specificity test is performed in tissue devoid of the antigen of interest (knockout mouse). When not feasible, the best alternative is to show that two antibodies raised against different epitopes of the antigen of interest yield the same staining pattern. A third control includes inactivation of the antibody by incubation with its antigen prior to use for immunohistochemistry. This control does not exclude, however, that several targets sharing a common epitope are detected by this antibody. If doubts remain, additional information about the expression and localization of the antigen of interest should be provided by alternative methods.
As a consequence of the requirement by the Journal of Comparative Neurology to provide detailed characterization of primary antibodies, there is now a large collection of commercially-available antibodies that fulfill these criteria. To avoid unnecessary duplication, the Journal of Comparative Neurology has published an Antibody database listing all these products, which can be considered as being well characterized. Keep in mind, however, that polyclonal antibodies can vary considerably in their affinity and specificity from batch to batch, even when sold under the same catalog number. Every new batch ought to be tested to avoid bad surprises.
2) Secondary antibody specificity
Secondary antibodies are raised against immunoglobulins (typically IgGs) of the species in which primary antibodies were raised. As they are used in fairly high concentration, the likelihood that they bind non-specifically to tissue components (extracellular matrix proteins, blood vessels, etc) is higher than for primary antibodies. Further, they might cross-react with IgGs from other species, which is particularly relevant in multiple-labeling experiments. To minimize cross-reactivity, whenever available, it is best to use highly cross-adsorbed secondary antibodies.
To test for specificity, secondary antibodies should be applied in the absence of primary antibodies: all residual staining can be considered as being non-specific. In immunofluorescence experiments, beware of possible autofluorescent molecules that may be contained in tissues, their presence can be detected best in the absence of secondary antibodies. Note, however, that non-specific binding to tissue components depends both on the antibody and the tissue type. Embryonic/neonatal brain tissue has much more non-specific binding than adult tissue, and lesions, which cause damage to the blood-brain-barrier and leakage of serum proteins into brain parenchyma, also increase non-specific binding of secondary antibodies. The same can be said for weak tissue fixation over extended time (e.g., poor perfusion-fixation, fixation by immersion, etc.).
3) Influence of tissue preparation
Relative to antibody specificity, this issue is more versatile and complex and it can result in both, false-positive and false-negative results, even when using highly specific and well-characterized antibodies. As noted above, these effects arise mainly from epitope masking due to fixation-induced conformational changes and failure of the antibody to penetrate the tissue, as well as from non-specific binding of secondary antibodies, which can mask the specific signals. Furthermore, it needs to be emphasized that when antibodies do not recognize or do not have access to their epitope, they will typically bind non-specifically to other tissue components, and therefore produce false-positive results. Finally, it is important to realize that control tests performed with knockout tissue only provide information about false-positive results, but not false-negatives.
There are a number of published examples of pitfalls resulting from the influence of tissue preparation, notably fixation and the need in some cases, to use stringent methods for unmasking epitopes, in particular for postsynaptic proteins (Fritschy et al., 1998; Watanabe et al., 1998; Lorincz & Nusser, 2008).
The following aspects should be considered, when aiming at improving immunohistochemistry procedures because of apparent false-negative or false-positive results produced by an antibody. Typically, a compromise between preservation of the (ultra)-structure and staining sensitivity should be sought.
- Reduce (or enhance) tissue fixation: type of fixative, pH, concentration, duration of the fixation (and postfixation), use of additives (e.g. picric acid). These effects are age-dependent (neonatal, but not adult, brain tissue benefits from a longer postfixation) and related to the nature of the antigen (soluble proteins tend to diffuse away in weakly fixed tissue).
- When fixation has deleterious effects, consider using weak fixation by brief immersion of tissue slices or tissue sections, as described in detail in (Schneider Gasser et al., 2006).
- Antigen-retrieval methods (e.g., heating/boiling tissue in acidic buffer; enzymatic digestion).
- Use of detergent or repeated freeze/thaw to enhance penetration of the antibodies into the tissue.
- Use of blocking solutions before and during primary antibody incubation. Note that in perfusion-fixed tissue, good results can be obtained without any pre-blocking step (see Fritschy & Mohler, 1995).
- Temperature and duration of incubation in primary antibody and secondary antibody solution, as well as duration of the rinsing steps between antibody solutions. Here, an increase in signal-to-noise ratio can be achieved, sometimes with striking reduction of non-specific binding to tissue components.
Submitting to EJN
In the Author guidelines, EJN emphasizes the need to provide enough information in the Materials and Methods section to enable proper evaluation of the results and reproduction of the data. With regards to immunohistochemistry, detailed information about the antibodies used and about control experiments validating their specificity and the sensitivity of the method is mandatory. Authors should indicated which exact antibody was used (catalog and batch number, species of origin, immunogen information) and should provide information on the characterization of the antibody (description -or reference to experiments previously reported- performed to test its specificity, see EJN author guidelines). These requirements are in line with the policy of the Journal of Comparative Neurology, which refuses publication of articles that do not provide detailed characterization of antibodies used for immunohistochemistry. EJN encourages its contributors to consult the antibody database of the Journal of Comparative Neurology when selecting antibodies for experiments and when reporting their use in their manuscripts.
Despite these pitfalls of immunohistochemistry, only a minority of articles submitted to EJN fulfills these explicit requirements of the author guidelines. In 2008, in order to improve this situation, EJN published a technical spotlight emphasizing the necessity to check whether negative findings (failure to detect the antigen of interest in the sample analyzed) might be due to the method – in particular, tissue fixation – used for analysis (Fritschy, 2008).
We hope this article will help you to prepare and analyze your immunohistochemistry experiments. Please feel free to ask questions!
The antibody database of the Journal of Comparative Neurology
Is my antibody staining specific? (Fritschy and Sarter, 2008)
Fritschy, J.M. (2008) Is my antibody-staining specific? How to deal with pitfalls of immunohistochemistry. Eur. J. Neurosci., 28, 2365-2370.
Fritschy, J.M. & Mohler, H. (1995) GABAA-receptor heterogeneity in the adult rat brain: differential regional and cellular distribution of seven major subunits. J. Comp. Neurol., 359, 154-194.
Fritschy, J.M., Panzanelli, P., Kralic, J.E., Vogt, K.E. & Sassoè-Pognetto, M. (2006) Differential dependence of axo-dendritic and axo-somatic GABAergic synapses on GABAA receptors containing the a1 subunit in Purkinje cells. J. Neurosci., 26, 3245-3255.
Fritschy, J.M., Weinmann, O., Wenzel, A. & Benke, D. (1998) Synapse-specific localization of NMDA- and GABAA-receptor subunits revealed by antigen-retrieval immunohistochemistry. J. Comp. Neurol., 390, 194-210.
Lorincz, A. & Nusser, Z. (2008) Specificity of immunoreactions: the importance of testing specificity in each method. J. Neurosci., 28, 9083-9086.
Panzanelli, P., Bardy, C., Nissant, A., Pallotto, M., Sassoè-Pognetto, M., Lledo, P.M. & Fritschy, J.M. (2009) Early synapse formation in developing interneurons of the adult olfactory bulb. J. Neurosci., 29, 15039-15052.
Schneider Gasser, E.M., Straub, C.J., Panzanelli, P., Weinmann, O., Sassoè-Pognetto, M. & Fritschy, J.M. (2006) Immunofluorescence in brain sections: simultaneous detection of presynaptic and postsynaptic proteins in identified neurons. Nature Protocols, 1, 1887-1897.
Tyagarajan, S.K., Ghosh, H., Yevenes, G.E., Nikonenko, I., Ebeling, C., Schwerdel, C., Sidler, C., Zeilhofer, H.U., Gerrits, B., Muller, D. & Fritschy, J.M. (2011) Regulation of GABAergic synapse formation and plasticity by GSK3beta-dependent phosphorylation of gephyrin. Proc. Natl. Acad. Sci. USA, 108, 379-384.
Watanabe, M., Fukaya, M., Sakimura, K., Manabe, T., Mishina, M. & Inoue, Y. (1998) Selective scarcity of NMDA receptor channel subunits in the stratum lucidum (mossy fibre-recipient layer) of the mouse hippocampal CA3 subfield. Eur. J. Neurosci., 10, 478-487.
Zattoni, M., Mura, M.L., Deprez, F., Schwendener, R., Engelhardt, B., Frei, K. & Fritschy, J.M. (2011) Brain infiltration of leukocytes contributes to the pathophysiology of temporal lobe epilepsy. J. Neurosci., 31, 4037-4050.