Solutions to the antibody problem

In the current era of very accurate DNA analyses by in situ hybridization, DNA chip analyses, and deep sequencing, it is often assumed that antibodies have an analogous ability to identify molecular targets accurately. Nothing could be further from the truth. Many investigators are unaware of the potential problems with specificity of antibodies and the need to document antibody characterization meticulously for each antibody that is used.

We have blogged in the past about the many problems which are increasingly coming to the fore with antibodies; problems of cross-reactivity and variability between lots. It has been said that a significant fraction of commercial antibodies appear to be ‘no better than PBS (link is external)’. However, the body of evidence demonstrating problems with swathes of the commercially available antibodies have been concentrated within particular fields, such as neurobiology, physiology, pathology and cancer diagnostics (link is external), and most of these publications will not have been seen by many cell biologists, even those that are heavy users of commercial antibodies. But very few will have failed to notice the articles that have appeared over the past year in Nature and other major journals that discuss the antibody problem in somewhat apocalyptic terms.

Affimer animation still

Antibody validation is the cause of much confusion. There is no consensus in the field on what constitutes suitable validation and this is further complicated by the number of different applications and methods for antibody use. Like all of science, knowledge regarding affinity reagents increases as more work is carried out using the specific antibody, aptamer or protein scaffold. Journals are now calling for contributing authors to provide antibody validation data with submitted manuscripts. In a recent webinar (link is external)regarding antibody validation Science magazine’s editor spoke of their plans to work with a company such as figshare to enable them to store complete datasets to accompany manuscripts, such as a complete set of western blots or stained tissues. None of this will matter though unless we deal with the issue of antibody variability.
If an antibody is seen to have a high demand many manufacturers will rush to make a version of their own. But each new antibody will differ subtly from each other, and the validation done on the original antibody will not support each new clone. A solution to this was suggested by Andrew Chalmers and his colleagues. They propose that all publications using commercial antibodies should state the name of the supplier and catalogue number of the antibody used. This would mean that from a supplier selling multiple varieties of an antibody a researcher will be able to order the same antibody used in a specific publication. This is increasingly being incorporated into different journals instructions for authors.

But obtaining an antibody with the same catalogue number would still leave the variability between different lots of the same antibody, obtained from different bleeds of the same animal. There has been a drive towards the use of sequenced monoclonal antibodies in research to overcome this issue and many are now calling for the end in polyclonal use and a shift entirely towards monoclonal antibodies and synthetic antibodies. Polyclonal antibodies are associated with numerous pitfalls, with lot-to-lot variability being a major problem. This in turn leads to problems with antibody vendors providing datasheets from old antibody batches that are no longer relevant to the purchased product. The interests of antibody vendors are at odds with researchers here, as vendors want to make money and so continue to sell a new lot of a popular antibody as the same product, while researchers are seeking reproducibility and reliability in their chosen antibody.

Although researchers are aware of the variability in polyclonal antibodies and consequently their use in long-term research or clinical assays is avoided, monoclonal antibodies also show a degree of variability.  Dr Chris Kerfoot, co-founder of Mosaic Laboratories, recently spoke of how lot-to-lot variability has been observed in his laboratory not just for polyclonal antibodies (link is external), but for monoclonals too, with different lots giving significantly different histochemical staining patterns.

While better validation of antibodies, aptamers and protein scaffolds will improve confidence in the resulting data the variability of antibodies, both polyclonal and monoclonal, may mean that it is now time to shift to more controlled and reliable reagents. Protein scaffolds, such as the Affimer, offer one such solution. As Affimer technology is generated entirely in vitro, using non-eukaryotic systems lot-to-lot variability is not an issue. All our Affimer products are validated in-house by our expert validation team, testing multiple experimental controls in both protein-specific and whole cell samples across multiple platforms.

While scientists and journals work together to overcome the problems with validation of affinity reagents, the issues around variability can be overcome by the use of synthetic antibodies like the Affimer.