Response to: Should we ignore western blots when selecting antibodies for other applications?

Uhlen replies:In the report1 from the International Working Group for Antibody Validation (IWGAV), we concluded that “approaches for antibody validation must be carried out in an application- and context-specific manner.” Our argument is that samples are treated differently in different applications and that this influences the epitopes exposed on the target protein, which might have profound consequences for the ability of a given antibody to bind specifically to its target. As an example, proteins that are analyzed by immunohistochemistry (IHC) are normally first cross-linked with formalin and then heated to very high temperatures (normally >100 °C) in a procedure that is sometimes termed 'epitope retrieval'. Obviously, this procedure might influence the target protein differently than the procedure used to prepare proteins for a western blot, in which the sample is instead treated with a detergent (SDS) before the electrophoresis step. Thus, as concluded by the members of the IWGAV1, the results obtained for a given antibody in western blot applications cannot be used to predict the specificity of the antibody in another assay based on an entirely different epitope-retrieval method, such as IHC.In the Human Protein Atlas (HPA) program, we have validated more than 24,000 in-house-generated antibodies directed to 17,000 human target proteins2. Although there is often a correlation between performance in different applications, we have observed many examples of antibodies that show strong support for specificity in IHC or immunofluorescence microscopy (IF) but which do not stain the correctly sized band in a western blot, and vice versa3, 4, 5. However, as pointed out by Lund-Johansen and Browning, western blot analysis has indeed been found to be useful as a general validation tool for antibody specificity. Many antibody providers use western blot analysis to show whether a band of the right size is stained or whether additional bands are present, the latter indicating off-target binding. This is indeed a practical procedure for a 'first-line' screening of antibodies for specificity. In the HPA program, we always screen our in-house-generated antibodies by western blot analysis.Similarly, protein arrays, in which the target protein is arrayed together with hundreds or thousands of unrelated proteins, are also useful tools for probing antibody specificity6. In this application, however, target proteins are often presented as purified (noncomplex) spots, and the proteins usually have not undergone any prior treatment with denaturing agents. This means that caution should also be applied when using protein arrays for validation of an antibody for its use in other applications, such as IHC and IF.In conclusion, western blot and protein array analyses can indeed be useful tools when selecting specific antibodies for other applications. The use of these methods is encouraged both for antibody providers and users, and antibodies with signs of cross-reactivity in these applications should be treated with caution. However, the formal validation of an antibody for a specific application must be performed in an application- and context-specific manner as suggested by the working group1.