Type in product code or name:
Many scientists are very unhappy about the performance of commercially available antibodies. There are many reasons why that is, some come from the provider and others originate from the user. And usually a combination of reasons play part in each individual case, thus making things complicated. The main factor though is a general lack of adequate knowledge on antibodies and how to work with them from both sides. The levels of complexity of immunoassays like Western blot and IHC/ICC are generally severely underestimated and many starting undergraduates and PhD students have to learn things the hard way without much expert guidance. The result is precious money and time wasted on very important biomedical research, and usually the commercial antibody gets the blame for lack of progress. See also our CEO's remarks.
This article aims to set the records straight. Of course some antibodies fail miserably, but we need to systematically dissect all the factors that make an immunoassay fail, even when the antibody is fine. After that, we can address the integrity of the antibody and the reasons why it may have been impaired if it was any good in the first place.
Storage: Antibodies deteriorate with repeated freeze/thaw cycles. Such cycles should be kept to an absolute minimum. One way is to keep aliquots frozen at all times and have one aliquot in the fridge for daily use until finished. Our primary antibodies come with NaN3 to keep them good at 4C for many months. Please be aware that some freezers are opened extremely often during a working day and then antibodies stored at the front may endure freeze/thaw cycles because of the frequent opening in combination with long browsing and possibly with the summer sun shining straight in. When this applies, antibodies are best kept in a different freezer that is less often opened (eg. -80C).
Western Blot (WB): This most common assay is regarded as extremely simple because it is taught in the lower years of an academic study and it seems all very straight forward. None could be further from the truth as many will notice after several fruitless attempts. The most trivial factors that may raise problems are addressed in our . This guide is tailored around our goat primary antibodies, but the reader will get the picture.
Potential problems may have already occurred during preparations of the biological material to be analyzed, well before the assay takes place. Proteolysis and oxidation are two notorious factors that will introduce low reproducibility of results when comparing different lysates from the exact same cell type with each other. The conditions (including temperature) of protein separation in SDSPAGE will influence the banding patterns. And finally, different tissue types and cell types do not necessarily give mutually identical patterns in WB. It depends on the type of protein of interest and whether any posttranslational modifications (PTM) and/or degradation may take place in a different fashion in each cell type. The PTMs may be prone to (partial) removal during the lysate preparation by endogenous enzymes. Proper inhibitors (making sure they did not get inactivated during preparation) should be added to the lysates to minimize such impairments. Hence, a commercial antibody shown to work properly in one tissue type or species should still be validated in other tissue types or species. And it is best advised to generate several lysates of the same cell type or tissue type at different days before starting analysis of all of them, side by side.
Immunohistochemistry and immunocytochemistry (IHC/ICC): When tissue sections have been stored after dehydration, the structure of the proteins in the sections will have changed thus affecting the binding of antibodies. Antigen/epitope retrieval will restore this, but its success is very dependent on the method of retrieval and the method may have to be adjusted from antibody to antibody (or better from target protein to target protein). For heat induced epitope retrieval (HIER) please consult our , but one could also opt for protease induced epitope retrieval (PIER) when the other approach remains unsuccessful.
Microwell platforms: Matrix effects are best dealt with by diluting the matrix in assay buffer. But polyclonal antibodies may perform better after being preadsorbed to abundant serum proteins as a blocking step.
Immunoprecipitation (IP/ChIP): Some antibodies will not bring its target down under certain conditions. Particularly when working with linear epitope antibodies (such as ours) one is advised to compare native conditions with denatured conditions (reduced and in presence of 0.1% SDS) as a positive control.
How to validate an antibody: Useful details can be found on our dedicated . When the scientist is not confident that a commercial antibody is fit for purpose, a test sample should be obtained at low cost. Most vendors and manufacturers can be pursuaded making a sample available when the scientist is willing to provide new validating data for the product sheet. But if the science is confidential, a 100% guarantee needs to be sought (see for example our guarantee for all our products). Small test samples from manufacturers are available through OWL.
Reproducibility: One should not jump to conclusions after a single experiment. Even positive data may sometime be false positive and lack of signals may be due to trivial factors that can be solved by systematic trouble shooting. Therefore conclusions can only be drawn once identical results have been obtained by the same experiment carried out at different times. For further details, see our
Once the user has ruled out everything that he or she could possibly be blamed for, it is time to start blaming the vendor and/or the manufacturer.
Batch-to-batch variations: The vast majority of vendors have not produced all the antibodies on its catalogue by themselves, and most content of their catalogues come from a wide variety of different manufacturers from all over the world. The vendors are not going to tell and keep the appearances that they themselves are the primary source of all its antibodies. This enables them to keep QC data on the datasheet that were generated many years ago thus keeping the sales going, while the actual antibody that generated these data has sold out and been replaced by other batches (from different animals) multiple times while the current batch may no longer be able to generate such data at all.
Even monoclonal antibodies suffer from batch-to-batch variations, but not to such severe extent as some polyclonal antibodies can do. Nonetheless, certain clone numbers are still being used after decades but just like with HeLa and HEK293, such hybridomas cannot be the same after so many passages anymore. It is therefore misleading to use QC data that were generated so many years ago, unless the current batch has proven to be still representative for such data (in which case one might as well show the fresher version of such data).
Assay developers are advised to buy straight from the manufacturer and ask for a free validation sample from a large batch in stock. Once validated for the required platform, the same batch then can be purchased in bulk so to prevent batch-to-batch variations during the entire project. Identifying the manufacturer can be a challenge, but since goat produce the largest batches compared to rabbits and rodents, please do not look further.
Transportation-related impairments: Anyone working in a subtropical climate (eg. southern state of the USA) knows how hot it becomes on a sun baked parking space during summer and how temperatures can soar inside a delivery van parked out there while taking care of a delivery. Antibodies are inherently made to work at 37-42C, but it remains a protein and it can cook until inactive when temperatures go well over 50C. We as the manufacturer dispatch our antibodies on frozen icepacks, but we know that not all vendors who carry our products on their catalogues do the same. It is the user’s responsibility to ensure the antibody is received in proper packaging (particularly during hot weather conditions) and complain to the vendor when this is not the case. We advise to return the goods unused when they arrived hot.
Endorsement by publication: Today’s advanced internet facilities enable vendors to search for publications describing successful use of their antibodies. However, some publications mistakenly mention a vendor for an antibody that is not from this vendor. And sometimes an antibody is correctly mentioned but no data generated by it are to be seen or presented in the paper, nor in the supporting documentations. In such cases the internet search delivers false-positives, and examples can be found in each vendor’s catalogue. Many such papers are inaccessible to the vendors and therefore they cannot always double check themselves. Although this is a relatively rare phenomenon, it is real all the same and therefore the customer is advised to double check the contents of the referenced paper when such endorsement is a requirement for the scientist to purchase this antibody. Each vendor will appreciate the feedback when a customer identifies a false-positive reference, and so we urge customers to contact the vendor when such reference has been identified so they can remove it from the product data sheet.
The best search engine to find literature-endorsed antibodies is Citeab.
Editors of scientific journals still allow antibodies to be described in a submitted paper without mentioning the catalogue number. For every manufacturer and vendor this is frustrating when more than one antibody against the same protein is on the catalogue as it will be impossible to find out which one was used by the authors. We urge both scientists and editors to take care of this lack of accuracy and make sure the catalogue number is specified when describing the antibody used for the presented data. Until then vendors may use the reference for every antibody on their catalogue that possibly applies........
Jan Voskuil 2013