Abstract
Antibodies are commonly used to detect or isolate proteins from biological samples. Much attention has been paid to the potential for poorly-characterized antibodies to lead to misleading results, but antibody-independent artefacts may also occur. Here, we recount two examples of antibody-independent artefacts that have confounded the interpretation of results in our search for molecular entities associated with memory loss in Alzheimer's disease (AD). First, when using biotin-avidin systems for antibody detection, endogenous biotinylated proteins created spurious bands in Western blots of brain lysates from AD patients and transgenic mouse models of AD. These artefactual bands occurred in a transgene- and strain-dependent manner. A second, unexpected artefact occurred when Protein A-conjugated Sepharose beads were used to deplete lysates of endogenous immunoglobulins prior to immunopurification of target proteins. In these assays, Protein A shed from the beads, then bound to (and was eluted from) an immunoaffinity matrix designed to capture AD-related proteins. The Protein A then bound detection antibodies when the immunoaffinity eluates were analyzed by Western blot. Both of these artefacts-the endogenous biotinylated proteins and the Protein A artefact-can be monitored by including an "irrelevant" antibody as an experimental control (e.g., running a parallel protocol in which the antibody directed against the target of interest is replaced by a non-specific antibody).