To the Editor:

We read the commentary “A proposal for validation of antibodies” by Uhlen et al. with interest1. The authors explain that “Because of differences in protein conformation and target accessibility, antibodies that perform well in one context may perform inadequately in others”, and they conclude that “approaches for antibody validation must be carried out in an application- and context-specific manner”.

Assessing antibody specificity in immunohistochemistry (IHC) and immunofluorescence microscopy (IF) is difficult since there is no comprehensive and definitive source of information about the distribution of proteins in tissues and subcellular compartments. It is therefore common to use results from western blotting (WB) analyses as supportive evidence. In fact, most of the antibodies used in IF and IHC were initially selected on the basis of their performance in WB analyses, as manufacturers often use WB as a first-pass test in product development. The expectation is that antibodies that stain a single band corresponding to the mass of the intended target (i.e. confirmatory) are more likely to be specific in IF and IHC analyses than those that are evidently cross-reactive in WB analyses (i.e. non-confirmatory).

Strict adherence to the principle of application-specific validation would imply that WB results have no relevance for IF and IHC analyses and that researchers should disregard non-confirmatory WBs when selecting antibodies for use in these applications. This seems counterintuitive, and there is evidence that WB can uncover cross-reactivity and trigger a more cautious interpretation of subsequent IF and IHC data2,3,4,5,6.

We would like to ask the authors to clarify their opinion about the predictive value of confirmatory versus non-confirmatory WB analyses for antibody specificity in IF and IHC. Is there compelling evidence that cross-reactivity in WB has no relevance?

Author contributions

F.L.-J. and M.D.B. selected the references and wrote the text.