The same antibody can be excellent on a Western blot and useless on a tissue section. The three workhorse immunoassays ask different things of an antibody, and most "bad antibody" stories turn out to be a mismatch between the antibody and the assay it was put into. Below is what each method needs, and a few habits that get more out of the antibodies you already own.
Western blotting
On a blot the target has been boiled in SDS, run through a gel and stuck to a membrane. Whatever three-dimensional shape it had is gone, so the antibody has to recognize a linear stretch of sequence. Polyclonal antibodies raised against peptides or denatured protein tend to do well here. Antibodies raised against native, folded protein may not. Blocking matters as much as the antibody does: 5% non-fat milk or 3 to 5% BSA in TBS-T, with the same protein carried into the antibody diluent, keeps background down. If the band is weak, try an overnight primary incubation at 4 °C before you raise the concentration.
Immunohistochemistry
Formalin fixation cross-links proteins and can mask the very epitope the antibody was raised against. That's why antigen retrieval (heating sections in citrate or Tris-EDTA buffer) is routine, and why an antibody that worked on frozen sections may need re-optimizing on paraffin. Titrate the primary on a tissue known to express the target, include a section with the primary left out, and where you can, a tissue that lacks the target. Our neuroscience antibodies, raised in rabbit, sheep and mouse against synaptic and catecholaminergic markers, are affinity purified, and the product pages show the staining you should expect to see.
ELISA
ELISA presents the antigen folded, either coated on the plate or captured by another antibody. A sandwich format needs two antibodies that bind non-overlapping epitopes at the same time, so a pair raised in different species, or a monoclonal capture with a polyclonal detector, avoids competition. Polyclonals make good detection reagents because they bind several epitopes and amplify the signal; monoclonals give the cleaner, more reproducible capture. Blocking and washing decide the background here too, and a matched pre-immune or isotype control shows you what non-specific signal looks like in your system.
Habits that help in all three
- Validate in the assay you will actually use, on a sample where the target is known to be present and one where it is absent.
- Titrate every new lot instead of assuming the last dilution still holds.
- Aliquot antibodies when they arrive and avoid repeated freeze-thaw.
- Keep the blocking protein the same between the block and the diluent.
- Write the lot number down with the result, so a change in staining can be traced later.
Every antibody we sell has its validation data, a Western blot or immunostaining, on the product page, so you can see the application it was tested in before you order. If you're working on a target we don't list, or need a second host species for double labelling, contact us about custom production.