Why is the secondary antibody conjugated to the enzyme in ELISA, instead of the primary antibody? Wouldn't it be easier to conjugate the enzyme to the primary antibody?


Making an antibody-enzyme conjugate isn't trivial. By using a primary/secondary set-up you can use the same well-characterised conjugate in combination with many different primary antibodies (as long as these primaries are all raised in the same species). There is also the possibility of some amplification: for example, if the secondary is an anti-Fab then two secondary Igs will bind to each primary.

Response to OP comment

Most primary antibodies in common use are derived from rabbit or mouse, and most are IgG. So, for example, the secondary antibodies goat anti-rabbit IgG and goat anti-mouse IgG will cover most experiments.

  • $\begingroup$ In other words; otherwise you would have to make a conjugate for every primary antibody? $\endgroup$ – JohnPhteven Dec 15 '12 at 21:04
  • $\begingroup$ Yes, that's right. $\endgroup$ – Alan Boyd Dec 15 '12 at 21:54
  • $\begingroup$ @ZafarS: in fact, that is what you do for most immunological techniques, like Western blotting, immunohistochemistry etc.: use an unconjugated primary and a conjugated secondary. $\endgroup$ – nico Dec 16 '12 at 10:20
  • $\begingroup$ @AlanBoyd Sorry, I still have some doubt whether I grasp secundary antibodies. Every species has (or better: can have) a secondary antibody per animal right? So for example; we can have secondary antibodies for primary antibodies of rabbits, chimps, mice, etc. So that would still be a lot of possible secondary antibodies, but anyway less than primary antibodies. $\endgroup$ – JohnPhteven Dec 17 '12 at 21:13
  • $\begingroup$ This answer could be improved with an explanation that many times primary antibodies are monoclonal, providing a high degree of specificity to the interrogation of the protein in questions and reducing off-target binding, while the secondary ABs are polyclonal and will bind to multiple sites on the primary AB, thus increasing the signal, say of a fluorescent marker, HRP, or radioactive probe. Though alluded to, cost and time of preparation are important considerations. You can have a stock of goat-antirabbit secondary ABs that will work with every assay you use a rabbit primary AB. $\endgroup$ – AMR Sep 15 '15 at 13:36

Another reason that secondary antibodies are used is to look for the production of antibodies by a test animal. Follow me for a second with an example that I think will make it clear.

Our Purpose: We have identified a protein (call it 'V_Entry') that is vital for Ebola to infect human cells. We wonder if antibodies against V_Entry will block Ebola infection. We then use this V_Entry protein as a vaccine in a clinical trial.

Our Question: Do patients have generate their own anti-V_Entry antibodies as a response to the vaccine. (note: at this point we are not addressing whether these antibodies will be protective or not- that's a separate question)

Our basic protocol:

  1. Day 0: Take blood sample Vaccinate patients
  2. Day 14 Take Blood Sample Booster Patients
  3. Day 30 Take Blood Sample

To test the ability of the vaccine to generate an antibodies specific to V_Entry, we will assay serum from our cohort via ELISA.

Our Test: (I'm ignoring controls and other details here to be clear) Coat ELISA plates with V_Entry protein Add aliquots of serum from each of out patients at each timepoint. Wash Add a secondary antibody conjugated with a detection enzyme Wash Add substrate for Enzyme Allow Enzyme to develop the substrate (usually colorimetric) Read on a plate reader

Some expected results:

  1. Nothing happens - a control well works, but none of our patients show antibody generation

    1. Patients have pre-existing antibody that binds V_Entry (maybe these people were already exposed to Ebola?)

    2. Patients have no pre-existing antibody, but they generate antibody in response to the vaccine (control patients not given vaccine show no new antibody)

Why Secondary Antibody was Necessary: We can't practically label every antibody made by the patient, but we can use a secondary antibody (perhaps Goat Anti-Human IgG) to detect any human IgG that has stuck to the V-Entry on our plate.

Was this a long run for a short slide? I considered making a slide for each step of the protocol, but I started wondering if I'm avoiding something...


Not the answer you're looking for? Browse other questions tagged or ask your own question.