The Importance of Antibody Validation

With poorly characterized antibodies being blamed for the reproducibility crisis, researchers have become more cautious with antibody selection. This editorial suggests what to look for when choosing an antibody, and the types of tests to perform when you receive it, as well as includes tips for addressing common immunoassay problems.

Scrutinize the antibody datasheet

When selecting a primary antibody, most researchers instinctively check to see that it is validated for their chosen application and species. However, such claims should never be taken at face value. “It is important to ensure that the vendor provides example data in the application of interest, as performance in one application does not guarantee performance in another,” advises Katie Crosby, Sr. Director for Antibody Applications and Validation at Cell Signaling Technology. “Researchers should also confirm that a protocol is supplied since antibody performance can differ based on a number of experimental variables. A highly validated antibody can perform poorly if the wrong protocol is utilized.”

The stated species reactivity should likewise be carefully scrutinized. “If you’re looking for an antibody with reactivity in IHC on mouse tissue, you will want to see mouse IHC data,” says Brian McWilliams, Ph.D., Senior Product Manager at Bethyl Laboratories (a Fortis Life Sciences brand). “The term ‘mouse reactivity’ can encompass a vast definition, including flow validation on mouse cells, or even a recombinantly over-expressed mouse homolog in a human cell line, so make sure that the relevant data is presented. Also, I recommend ensuring that the antibody is able to detect endogenous expression, not just a purified or over-expressed protein. Those conditions are not usually biologically relevant and require a customer to have access to these sorts of reagents as well.”

What to look for on an antibody datasheet

• Antibody target
• Immunogen used for antibody production
• Epitope location—intracellular or extracellular
• Host species
• Validated species and applications—with example data
• Polyclonal or monoclonal—including the clone number if monoclonal
• Isotype information
• Concentration and formulation—azide can present issues for some applications
• Purification method and antibody purity
• Known cross-reactivities
• Compatibility with desired workflow—fixation and permeabilization may not always be possible
• Links to references or customer reviews

Information provided by Sartorius*

Consider the pillars of antibody validation

In 2016, Uhlen et al. proposed five pillars for antibody validation as a means of addressing the reproducibility crisis. These now form the basis of many commercial validation processes. During a panel discussion* with Sartorius, it was noted that, in general, antibody manufacturers tend to use genetic knockdown, recombinant expression, and testing with methods such as ELISA, flow cytometry, and SDS-PAGE for validation. In addition, Bio-Layer Interferometry (BLI) is increasingly being applied for real-time, label-free analysis of antibody binding kinetics and affinities. This has been made possible with the introduction of user-friendly platforms like the Octet^® BLI system, which makes characterizing antibody-antigen interactions more efficient and provides a better representation of the native state.

Both Bethyl Laboratories and Cell Signaling Technology have adapted the Uhlen model to include six pillars, not five. “The principles proposed by Uhlen et al. have underpinned our validation process for years,” reports McWilliams. “However, the publication did a wonderful job in helping us to expand our thinking and include even more rigor in the validation of our products.” Crosby agrees, noting that while it may not always be possible to utilize each of the six strategies for any given target, Cell Signaling Technology always incorporates as many strategies as is feasible and necessary to confirm antibody specificity.

The pillars of antibody validation

• Genetic strategies—antibody specificity is evaluated by comparing samples expressing the target protein with samples in which the target protein has been knocked out or depleted
• Orthogonal strategies—antibody-based target detection is compared with an antibody independent method such as mass spectrometry, transcriptomics, or in situ hybridization 
• Independent antibody strategies—antibody binding activity is compared with that of another antibody validated for the same target 
• Expression of tagged proteins—the target protein is expressed with a fusion tag to allow for a comparison of the antibody signal to that of a tag-specific antibody 
• Immunocapture followed by mass spectrometry—the antibody is immobilized on a bead and used for target capture, then any bead-bound proteins are subjected to MS analysis 

Adapted from Uhlen et al.

Two of the antibody validation pillars critical to the validation process for Bethyl Laboratories are Complimentary Assays and Biological Conditions. In this example phospho-KAP-1 expression was induced by etoposide treatment in HEK293T and HeLa cells. This biological condition was detected across multiple assays including western blot (A), ICC (B), and flow cytometry (C).

Cell Signaling Technology has adapted the Pillars of Antibody Validation to include a multiple antibody strategy, whereby the immunostaining data from two or more antibodies against distinct, non-overlapping epitopes on the same target is compared. This example shows immunohistochemical analysis of paraffin-embedded human squamous cell lung carcinoma using MAGE-A4 (E7O1U) XP^® Rabbit mAb (left) and MAGE-A4 Rabbit mAb (right). The similar staining patterns observed help to confirm the specificity of both antibodies.

Don’t rush in

Once you have received your new antibody reagent, you will want to confirm that it works as expected before using it with precious samples. This should include testing against known positive and known negative sample types and titrating to determine the optimum working range, always using appropriate controls (e.g., stained and unstained samples) to rule out any potential non-specific interactions. For immunoassays that will produce a fluorescent readout, checking the fluorophore brightness versus the antigen concentration is important, particularly if the antibody will form part of a panel.

“Ultimately, the best way of evaluating an antibody reagent is to test it in whatever assay you plan to use it for,” comments McWilliams. “During this process, make sure you understand how to utilize the target protein’s biology to scientifically support your findings. For example, are there certain cell lines or tissues that you know the target is or is not expressed in? Can you co-IP with a second antibody, or blot an IP with another antibody against a different epitope of the same target protein? The manufacturer’s recommended protocol is generally a good place to start when running these kinds of in-house validation studies.”

Addressing common problems

Many common immunoassay problems are often relatively easy to resolve. “If you are not seeing a signal in your immunoassay, or your detected signal is only very weak, running a positive control is the best first step to investigate the issue, ideally in combination with a careful replication of the vendor’s protocol,” says Crosby. “A lack of a signal in a known positive control when using the recommended protocol would certainly warrant a call to the vendor.”

If an unexpected signal is observed, modifying the concentration of the primary antibody, secondary antibody, or sample can usually overcome the problem. Performing a side-by-side comparison of a positive control and the sample that yielded the unexpected signal can help to rule out sample-specific artefacts. Other ways of addressing unexpected signal include optimizing any blocking or washing steps, always using fresh buffers to eliminate any risk of contamination.

Lastly, poor reproducibility between experiments is best tackled by considering what has changed between runs. “If poor reproducibility is seen within a single lot of antibody, it is typically associated with user error or inconsistent application of the antibody,” says McWilliams. “Variation between lots, especially with polyclonal antibodies, can be harder to control. If you find yourself in this situation, performing a direct lot-to-lot comparison can help. You may also wish to reach out to the manufacturer to understand any lot-to-lot changes.”

* The Sartorius panel comprised David Apiyo, Ph.D., Manager of BioAnalytics Applications Development; Nicola Bevan, Manager, Application Development; Mark Carter, Manager of Cell Assays, Sartorius Bioanalytical Instruments; and Carrie Weldon, Training Professional.