Why Is Antibody Validation Important?
For an antibody to be a useful research tool it must bind as specifically as possible, meaning it binds the correct epitope and is clearly visible above any background. Non-specific binding or sub-optimal assays can mislead researchers with false positive results or create sufficient background noise as to mask the protein of interest. These experiences are not trivial and can be particularly painful when the trust in your research tools is dashed alongside the loss of what can be several months of hard work and valuable samples
Unfortunately it is very common when purchasing an antibody that validation data is not available and protocols for use in particular assays are limited. It then falls to the researcher to test any new antibody – first by assay optimization to minimize background, and then antibody validation to confirm on-target binding.
The ability of scientists to validate the specificity of antibodies has improved significantly in the last decade or so, as more robust controls have become available – first through RNAi based knockdown, and now through knockout cell lines generated using gene editing.
Optimization of experimental parameters for every new antibody is still required to ensure confidence in results. Scientists can use a variety of experimental approaches to assess their antibody, including:
- Western blotting
- Immunocytochemistry (ICC) or immunohistochemistry (IHC)
- Flow cytometry
Key factors to keep in mind as you optimize and validate any new antibody
The way in which a sample is prepared should be determined by what is to be studied. For example, different cell lysis buffers can be used to isolate different subcellular fractions. Similarly different cell fixatives (for ICC) can expose or mask different epitopes. Many antibodies work only in certain conditions (for example, recognising only proteins in their native, non-denatured form), and so if a certain sample preparation approach is desired this should inform the choice of antibody.
Most suppliers will give an indication of the conditions in which their antibodies will work. They may not have been exhaustively tested however, so it can be worth testing non-specified conditions and applications if time allows.
Type and concentration of block
With blocks, you are aiming to strike a balance between masking non-specific epitopes and leaving the specific epitope visible.
- For Western blotting, a good starting point is to compare 5% milk powder in TBS-T (strong block) versus 3% Bovine Serum Albumin (BSA) in TBS-T (weak block).
*Tip – for a cleaner Western blot, after preparing a milk block, spin it briefly in a centrifuge at 4000rpm to remove any not dissolved particulate.
- For ICC experimenting with different blocks such as Foetal Calf Serum (FCS), BSA or Fish skin gelatin can significantly reduce noise and improve image clarity.
Concentration of antibody and length of incubation
Trialling a variety of concentrations and incubation times during initial validation not only ensures experiments are optimal moving forward, but can potentially save money if lower antibody concentrations are found to work effectively.
Length of incubation required can also vary significantly, with some highly efficient antibodies binding in a little as 15 minutes, while others requiring overnight incubation steps with agitation. Having this information to hand at the beginning of a project can have a big impact on how you structure your experimental workflows – if you can perform a complete Western in a day versus requiring an overnight incubation, this will influence when you begin your work!
The number and length of washes required can also vary between antibodies, although it is rare that too much washing (within reason) is a bad thing. Most protocols recommend three 5-15 minute washes in TBS-T buffer – some particularly clean antibodies can be washed less, but will not suffer from longer wash steps.
Test multiple different samples
Good practice when using any antibody is to look at its binding across a panel of cell line or tissue backgrounds. This not only builds confidence in the true positive nature of any results, it can often provide an interesting data set in terms of the protein expression profile across different tissues.
Multiple cell line lysates are unlikely to have the same protein composition, and comparison of banding patterns by Western can provide insights into binding specificity. Further to this, the RNA and protein expression profiles of many cell lines are now publicly available, and this data can be used to select a panel of lines with varying expression levels to confirm differences in observed expression using the antibody.
Validate antibody binding using an RNAi knockdown
Another excellent option for controlling antibody validation is to knockdown the protein of interest with RNAi, and look for a concomitant reduction in binding using Western blot or ICC. As knockdowns can often be performed in 72-96h, antibodies can be tested in this manner in less than a fortnight.
One limitation of this approach is that if the knockdown is ineffective, the scientist might discount a good antibody. Another potential (although less likely) limitation is that off-target effects of RNAi result in a reduction of the non-specific protein band – further compounding the false positive result.
Overcoming these limitations requires standard best practices when using RNAi, namely use of a non-targeting control, use of multiple RNAs targeting the same gene (in pool and then de-convoluted) and then potentially showing rescue of a knockdown using overexpression of siRNA resistant forms of the coding sequence.
By far the most robust method for antibody validation now available is to use genome editing to generate a knockout cell line as a negative control. This has several advantages over cell line panels and RNAi:
- A knockout can be validated at the genetic level , leaving no doubt about whether the protein is present in the cell. Functional protein expression can be ablated by introduction of frameshift mutations into the coding sequence, or the epitope can be excised completely using CRISPR-Cas9. In either case, a true negative control is ensured.
- Cell lines in which expression of the gene is also confirmed by mRNA can be selected for knockout, resulting in a genetically identical (isogenic) positive and negative control.
- Consequently, the observed loss of a protein from a cell sample can be attributed directly to the alterations made in the genome at the target site, ensuring extremely high confidence in antibody specificity.
Generate a knockout cell line control
You are of course not limited to cell lines as knockout controls. In their 2013 paper, Davies et al use LRRK2 knockout rats to characterize and optimize monoclonal LRRK2 antibodies. The value of doing this is highlighted especially well by Figure 5, where two antibodies show much clearer specificity and a lack of non-specific bands.
Knockout cell line controls won’t always be appropriate, as essential genes cannot be knocked out. A work around for this is to create inducible knockouts or mutate epitope coding regions in situ,
Furthermore, by contrast to RNAi genome editing is labour intensive – a knockout cell line can take more than 8 weeks to generate. The benefits of having validated antibodies therefore have to be stacked against the significant time investment of generating such cell lines.
That is of course unless the knockout cell line has already been made.
Ready made knockout cell line controls
At Horizon we’ve built a high throughput knockout cell line generation pipeline that allows us to generate tens of knockout cell lines each month rapidly and cost-effectively. We use our proprietary line, the HAP1, to generate these knockouts – this line has been analysed by RNAseq to determine which genes are expressed, and by genome wide knockout screening to determine which genes are essential and can't be knocked out.
If you want better controls for validating your antibody, browse our catalog to see if your knockout cell line is already available.
Can't find the right cell model for your research?
Our Express and Custom engineering services could help you.