By Skye Longworth

November 7, 2023

Antibody validation is an initiative close to our heart here at CiteAb. We began life as an antibody search engine, built to help researchers choose antibodies that work for their experiments. Considering antibody validation is an essential step in this process.

As part of our commitment to increasing awareness and driving improvements in the field, we have been partners in running the International Antibody Validation meeting for several years, learning much on this topic.

What is antibody validation testing for specificity?

Testing an antibody to show that you get signal for a given application is standard practice in life science and essential ahead of performing any experiments. This is of particular importance given differences in antibody performance between common applications, such as flow cytometry compared to IHC.

However, validating your antibody for specificity, by showing that your antibody binds to the intended target with minimal cross-reactivity, takes this a step further. 

Often based upon the ‘five pillars of antibody validation’, this testing gives you supporting evidence that your antibody will work as you expect, against the target you want. 

Why is antibody validation important?

When antibody validation is not done properly the result can be severe. Paper retractions, irreproducible research, and even whole research fields being led astray.

Below we provide some examples to illustrate the scale of the issue.

Researchers have found many commercially available antibodies are not specific to their target in a given application. This issue can impact entire research fields.

For example, in a study focusing on antibodies used for epigenetic research, it was found that around a quarter of the 246 commercially available antibodies tested were not specific [1].

In a more recent paper on eLife, 65 neuroscience-related proteins were tested and over 50% failed in at least one validation test [2].

This issue can derail emerging, exciting fields as well as those already well established. A new field based around the discovery of oestrogen receptor β (ERβ/ESR2) did not evolve in the exciting direction that was expected. Why? A group tested 13 anti-ERβ antibodies and found that only one was specific in IHC, contradicting the findings of many early studies that shaped the field [3]. 

Undoubtedly, the common thread running through all these examples is the importance of antibody validation.

What are common antibody validation techniques?

There are many ways to carry out antibody validation, and no ‘one size fits all’ approach.

When evaluating data or selecting a strategy, it is important to weigh up the pros and cons and consider the application and experimental context.

Often a combination of strategies may give the best results. 

The International Working Group on Antibody Validation brought forward five pillars of antibody validation to help with this process and provide a useful framework for discussing the topic [4].

These pillars are:

• Genetic strategies
• Orthogonal strategies
• Independent antibodies
• IP-MS
• Expression of tagged proteins

Below, we give an overview of these pillars. Plus, we outline two further important strategies: biological models and cell treatment, given their common use by some of the suppliers we partner with.

Other methods of validation could also be considered, such as: neutralisation, peptide array and protein array. These methods are beyond the scope of this current review.

An overview of the Five Pillars of Antibody Validation (and more)

Genetic Strategies

An overview

Many researchers consider Genetic Strategies to be one of the most reliable ways to validate your antibody.

To do so, a knockdown or knockout of the target protein is generated (or bought from a supplier) to act as a true negative control. In a KO sample, you would expect a complete loss of antibody-target binding as the target protein is completely removed.

In KD validation the target protein is down-regulated, meaning the signal should be reduced. 

At CiteAb we have collected >21,000 KO/KD supplier validations for commercially available antibodies. All CiteAb statistics are accurate as of 2022.

How are antibody validation genetic strategies done in practice?

Genetic strategies often make use of CRISPR/Cas9 technology. CRISPR/Cas9 involves a guide RNA coupled with a Cas9 endonuclease to mutate your gene of interest. Knocking-out the gene will produce your true negative control.

In Knockdown, siRNA/shRNA/micro RNA acts on the RNA and disrupts protein translation so less of the target gene is produced. It is worth noting that this down-regulation is temporary.

It is worth noting that this method can not be applied universally. For example, with important house-keeping genes a knockout would result in cell death – putting an end to your testing. On top of this, performing validation in a certain cell line does not guarantee specificity across all cell lines or tissues. 

Orthogonal Strategies

An overview

Orthogonal strategies for antibody validation involve the use of an antibody-dependent and an antibody-independent method in parallel. Similar results should be obtained for what the antibody is trying to measure in a variety of different samples.

At CiteAb, we have collected over 14,000 supplier orthogonal validations for commercially available antibodies. 

How are antibody validation orthogonal strategies done in practice?

An orthogonal approach could involve an antibody-dependent strategy, such as IHC or western blot staining, and an antibody-independent approach, such as the use of omics data, RNA-seq or in situ hybridization.

Let’s use an omics approach as an example: levels of the target using proteomic or transcriptomic data can be compared to levels of the target detected using the antibody based approach in the same tissue. If you observe a correlation, this would support antibody specificity. Some databases that can be of use for this method include: ProteomicsDB and NCBI Gene. 

Using orthogonal strategies provides opportunities for streamlined validation. However, there can be challenges with interpretation of the data. For example if levels of RNA don’t correlate with levels of protein, and if expression levels of the target do not differ between samples.

Biological Models

An overview

Conceptually, the biological model strategy for antibody validation has some similarities to an orthogonal approach. The expected localisation or expression profile of the protein in a given cell line can be compared to the antibody-target binding signal.

Although this isn’t one of the ‘five pillars’, we have seen valuable data come from the suppliers who use this strategy. 

How are antibody validation biological strategies done in practice? 

There are many ways to develop a suitable biological model. Examples include the use of a disease model that results in increased or decreased protein expression. Alternatively, an enzymatic assay which reveals expression or localisation of the protein can be used.

Databases are available to help with developing a biological model. These include: the Cancer Cell Line Encyclopedia, or Gene Cards, the human gene database.

Compare results from the relevant biological model to what the antibody shows. A correlation can help give you confidence in antibody specificity.

One drawback of this strategy is that information about protein expression or localisation needs to be accessible. This makes it difficult to perform for proteins where little information is publicly available.

Independent Antibody Strategies

An overview

This strategy compares the staining of your target of interest to the staining produced from another antibody, which binds to a different epitope on the same target. You would expect to see a correlation between the results produced by the two independent antibodies.

At CiteAb, we have collected over 4,000 independent antibody validations for commercially available antibodies. 

How are antibody validation independent strategies done in practice?

To carry out this strategy, you would use independent antibodies in the same application binding different epitopes. The results of the independent antibodies are compared to a control antibody. The control antibody does not bind an epitope of the target. If the results from multiple antibodies and your test antibody are consistent, this would support antibody specificity to the target. 

With this strategy, results can be misleading if the independent antibodies used give the same non-specific staining. It can also be challenging to find multiple independent antibodies, and to ensure that the antibodies are all truly distinct.

Expression of Tagged Proteins Strategy

An overview 

The concept of this strategy is to compare the results of antibody-target binding using a tagged protein and a control. You would expect to see matching staining between your test antibody against the untagged protein, and an antibody against the tagged protein target. Similarly, scientists use a protein that’s been recombinantly overexpressed for a variation on this method.

At CiteAb, we have collected nearly 5,000 supplier validations using the expression of tagged proteins for commercially available antibodies. 

How are antibody validation tagged protein strategies done in practice?

To carry out validation with a tagged protein, you would express your target with an affinity or fluorescent tag and measure the signal produced by an anti-tag antibody. This would be measured alongside your test antibody against the same untagged protein. If overlapping signals are observed, this can give some evidence that your test antibody is specific. 

While this method is selective for your target, it is not the most specific of the five pillars. When using recombinant techniques, protein expression can often be higher than endogenous levels. This can give you confidence that if your antibody is binding your target over closely related family members. On the flipside, you don’t have as much evidence that your antibody will be specific to your target at lower levels of endogenous expression.

Validation using Mass Spectrometry

An overview

In an IP-MS method, an original pillar of antibody validation, the test antibody is used to immunoprecipitate the target of interest. This is followed by mass spectrometry to identify what has been immunoprecipitated, thereby showing you what the antibody binds to.

In a newer version of validation using mass spectrometry, the capture-MS method, a western blot is compared to a mass spec of the target of interest. For the mass spec, the sample is first run through an SDS-PAGE to separate the protein extract. Antibody specificity is shown by a correlation in signal at the same size between this analysis and a western blot.

At CiteAb, we have collected over 300 supplier capture-MS antibody validations for commercially available antibodies. 

How is validation using mass spectrometry done in practice?

After immunocapture via the antibody, mass spectrometry can be carried out on the sample. Mass spectrometry allows you to calculate the molecular weight of the target. You can then identify if the antibody is binding to what you think it should be. For comparison, a negative control of an antibody against an unrelated protein target should also be used.

We noted that Edfors et al. published a paper on antibody validation for western blotting. They drew attention to the risk of this strategy for measuring specificity of the antibody to unrelated, bound proteins [5].

Cell Treatment Strategies

An overview

Cell treatment is a validation strategy which makes use of tools often used in cell biology to modify the protein or its localisation in some way. Cell treatment strategies are applied to enrich, deplete, translocate, or post translationally modify the target. This is then compared to a control sample. 

How is cell treatment done in practice?

To test for antibody specificity, you can use multiple methods of cell treatment. Some examples include the use of cytokines such as PDGF leading to phosphorylation in the sample, or the use of TSA for histone enrichment. The antibody-target binding can be compared between a sample which has undergone cell treatment, and a control sample.

An infographic to summarise the pros and cons of the five pillars of antibody validation, proposed by Uhlen et al.

–

Final thoughts – The Reproducibility Crisis and Validation

To wrap up, let’s take a moment to reflect again on why antibody validation is really important. 

In the wake of a paper released in 2012, the term ‘reproducibility crisis’ was coined. The important of reagents in contributing to this issue made clear. [6].

Further papers have reinforced that many antibodies aren’t validated in research, or validation data isn’t shared, and that validation is crucial for reproducibility. 

It is essential, therefore, that the antibody you use in your experiment is specific, and can therefore generate reproducible results. 

We encourage you to carefully consider which antibodies you use in your experiments, evaluate validation data – resources like YCharOS now exist to help with this, which we link from in our search engine – and validate antibodies yourself in the lab.

You can try out our search engine here:

Doing so will save you time and money further down the line. This ultimately accelerates your research progress.

• Skye and the CiteAb team

References:

1. ‌Egelhofer, T.A., Minoda, A., Klugman, S., Lee, K., Kolasinska-Zwierz, P., Alekseyenko, A.A., Cheung, M.-S., Day, D.S., Gadel, S., Gorchakov, A.A., Gu, T., Kharchenko, P.V., Kuan, S., Latorre, I., Linder-Basso, D., Luu, Y., Ngo, Q., Perry, M., Rechtsteiner, A. and Riddle, N.C. (2011). An assessment of histone-modification antibody quality. Nature Structural & Molecular Biology, [online] 18(1), pp.91–93. doi:10.1038/nsmb.1972.
2. Ayoubi, R., Ryan, J., Biddle, M.S., Alshafie, W., Fotouhi, M., Bolivar, S.G., Moleon, V.R., Eckmann, P., Worrall, D., McDowell, I., Southern, K., Reintsch, W., Durcan, T.M., Brown, C.M., Bandrowski, A., Virk, H.S., Edwards, A.M., McPherson, P.S. and Laflamme, C. (2023). Scaling of an antibody validation procedure enables quantification of antibody performance in major research applications. eLife, [online] 12. doi:https://doi.org/10.7554/eLife.91645.
3. Andersson, S., Sundberg, M., Pristovsek, N., Ibrahim, A., Jonsson, P., Katona, B., Clausson, C.-M., Zieba, A., Ramström, M., Söderberg, O., Williams, C. and Asplund, A. (2017). Insufficient antibody validation challenges oestrogen receptor beta research. Nature Communications, 8(1). doi:https://doi.org/10.1038/ncomms15840.
4. Uhlen, M., Bandrowski, A., Carr, S., Edwards, A., Ellenberg, J., Lundberg, E., Rimm, D.L., Rodriguez, H., Hiltke, T., Snyder, M. and Yamamoto, T. (2016). A proposal for validation of antibodies. Nature Methods, 13(10), pp.823–827. doi:10.1038/nmeth.3995.
5. ‌Edfors, F., Hober, A., Linderbäck, K., Maddalo, G., Azimi, A., Sivertsson, Å., Tegel, H., Hober, S., Szigyarto, C.A.-K., Fagerberg, L., von Feilitzen, K., Oksvold, P., Lindskog, C., Forsström, B. and Uhlen, M. (2018). Enhanced validation of antibodies for research applications. Nature Communications, 9(1). doi:10.1038/s41467-018-06642-y.
6. ‌Begley, C.G. and Ellis, L.M. (2012). Raise standards for preclinical cancer research. Nature, 483(7391), pp.531–533. doi:10.1038/483531a.

About the author

Skye is the Scientific Communications Manager at CiteAb, with a background in Biochemistry.

More posts by Skye