Our perception of reality often differs due to individual biases and viewpoints. Scientists aim to eliminate such biases with the help of methods and tools that make scientific results reproducible and replicable for other researchers in the field, a cornerstone of good scientific practice. Researchers therefore need to rely on that their tools do what they are expected to do. When looking at the same cells in the same setup through a microscope for example, the picture should always be the same, no matter how many different people look through the ocular lenses.

Antibodies in life sciences

One of the main tools employed by biologists and other life scientists are antibodies. Discovered in the 19^th century they are protective proteins found in many organisms. They recognize invading pathogens or toxins (antigens) and bind to their epitopes, which are specific structures on the surface of antigens. By that they neutralize or render antigens immobile, being an essential part of the immune system. In the 1970’s scientists found out that they can clone antibodies from a single immune cell by the hybridoma technology. Hybridoma cells secrete so called monoclonal antibodies that recognize a specific epitope. Furthermore, to make even those proteins visible, for which no specific antibodies exist, researchers have developed “tagging” techniques. Hereby a short peptide “epitope-tag” is attached to the protein of interest, thereby making the protein detectable by available tag-specific antibodies.

Do antibodies show what they are supposed to?

In their first paper, Stefan Schüchner and his colleagues from Egon Ogris’ group took a closer look at some widely used Myc tag antibodies. The team found out that depending on where and in which amino acid sequence context the tag is attached to the protein of choice, several of the examined antibodies showed fluctuations in their ability to detect the tagged proteins. One antibody called 9E10 stood out even more, showing drastic differences in its ability to recognize the tagged protein.

In their second study, first author Ingrid Frohner examined antibodies that are directed at the phosphatase PP2A, a major regulatory enzyme in cells. They found out that the antibodies in use to detect total PP2A in cells only recognize a minor form of the enzyme that is missing an important chemical modification necessary for its activation. Also, some of the antibodies showed cross-reactions and interacted with other related enzymes. The researchers concluded that these antibodies are not suitable to assess the activity of PP2A in the cell.

Antibodies need to be validated

Nowadays thousands of antibodies against different cellular targets exist. They are used in basic research as well as in medical applications like cancer immune therapies, where their ability to recognize and attach to antigens is exploited to direct the immune system against cancer cells, making antibodies indispensable tools for researchers and clinicians alike.

At the same time, however, inconsistent, non-specific, and cross-reacting antibodies are suspected to be a common cause for misinterpretation of data and lack of reproducibility, a serious problem in science. The results published in these two latest publications from the Ogris group prove that like all precision instruments, scientific tools need to be calibrated and fine-tuned before using them, meaning that proper validation of antibodies should be employed when conducting research with these reagents.

Original Publications:

Ingrid E. Frohner, Ingrid Mudrak, Stephanie Kronlachner, Stefan Schüchner, and Egon Ogris: Antibodies recognizing the C terminus of PP2A catalytic subunit are unsuitable for evaluating PP2A activity and holoenzyme composition.

Science Signaling  28 Jan 2020, Vol. 13, Issue 616
DOI: 10.1126/scisignal.aax6490

Stefan Schüchner, Christian Behm, Ingrid Mudrak, and Egon Ogris: The Myc tag monoclonal antibody 9E10 displays highly variable epitope recognition dependent on neighboring sequence context.

Science Signaling  28 Jan 2020, Vol. 13, Issue 616, eaax9730
DOI: 10.1126/scisignal.aax9730