Hi,

how did you exactly confirm that the antibodies do bind to your virus?

If the antibodies are very specific for your virus and normally bind to the virus material, it seems as if somtehing within your assay is inhibiting the binding of the antibodies

Hi Anna,

Thank you very much for your response.

totally we used three antibodies. They are commercial ones. One Ab was tested positive for the virus with Immunoflorescence, western blot and ICC. The other was tested positive with Immunoflorescence and the third one was tested positive with Immunohistochemistry.

But they do not seem to be binding with the virus in our lateral flow assay. I tried indirect assay, where I coat virus on the nitrocellulose membrane and then come up with Gold NP-Ab. The interaction of Ab with the virus should show up red color corresponding to Gold-NP. But it did not show up any color.

The virus is around 40nm diameter and the antibodies are raised against the envelop protein of the virus. I guess it is OK to coat 40nm Virus on the nitrocellulose membrane ?

I also tried sandwich assay. but that also did not show up any color.

was trying to figure out the reason. Thought any experienced people in this forum can share their thoughts.

Thank you

There are many possible issues but you get a control line, which would suggest that at least some conjugation of Ab to the particles has occurred. The results after spotting virus on membrane could be false unless you can prove that the virus has bound.

As the antibodies have been validated in non-LFA applications you cannot make the assumption that they will work in LFA. They may do, but not necessarily so. The on-rate of the antibody is particularly important in LFA , whereas in Westerns, for example, you would incubate for an hour or more, and there is much more time for binding to occur (and the antigen is denatured too, so very different from LFA). I am not suggesting you measure on-rate as that parameter in itself does not mean the LFA will work either!

The best thing to do is simplify the assay to identify the nature of the problem, but you should retain the intended final capture format. So, spot the capture antibodies on a membrane without a conjugate pad and allow to dry. Then dip the strips in ~50ul of conjugate (a 96 well plate is perfect for this type of testing). If you get a red spot you can conclude that the conjugate before was probably inactivated upon drying into the conjugate pad. If you do NOT get a spot then you can conclude one of the following (i) the antibodies are not suitable for LFA (ii) the antibodies have been inactivated during conjugation (iii) the conjugation is not efficient enough, (iv) the capture antibody has been inactivated upon drying (v) there is a pH/buffer or steric problem.

There are other possible explanations too, but depending on what you find in the above testing we should be able to home in on the problem and think of solutions.

Dear Nick,

Thank You Very Much for your elaborate response. I really appreciate for your time and thoughts.

I was reading the simplified assay you suggested. I have one doubt which I am explaining below

.

As I understood, there are two Abs involved in Sandwich format: capture Ab on the membrane, detection Ab on the conjugate pad. Given below are some of my doubts in the procedure you explained

Step 1: I will spot the capture Ab on the membrane without conjugate pad

Step 2: Dip the strips in 50 ul of conjugate. By conjugate, do you mean Gold NP- Detection Ab ?

Step 3: We should see a red spot for a positive test. But, we have not introduced the virus in our previous steps. The red spot indicative of the sandwich should be seen only in the presence of virus right ? Dear Nick, did I miss out anything ? Do I need to incubate the conjugate with virus in the second step ?

Also, is there any way I can find out whether a given Ab is binding to the virus in a LFA format ? How do I know that the virus has bound to the membrane if I spot it ?

Thank you very much for your time

Sorry, you are correct of course, I meant to write conjugate plus virus. (Do the test without the virus too, as that is an important control).

With the format suggested above you actually have more flexibility than would normally be the case. i.e. you could add the virus sample between steps 1 and 2, followed by buffer, and then the conjugate in step 2 (yes, this is the gold conjugate). This avoids the risk of a false negative because the antigen concentration is far too high, but in a real test situation the sample is mixed with the gold conjugate (i.e by solubilisation of the gold conjugate in the conjugate pad), so I would be inclined to mix varying amounts of virus with the gold conjugate and run the mixture directly up the strip. Use a broad range of dilutions of virus (log dilutions initially perhaps) to make sure that you are not working in the wrong range of virus concentration.

To answer your other point, I'm not sure how much virus you would be spotting or whether a protein stain might pick it up. However, this is not the actual test situation and may not give an answer with predictive value. If you want to check that the antibodies are sticking down you can always make/buy a polyclonal anti-species gold conjugate.

Hope this helps .

Dear Nick,

Thank you very much for your response again. Will do what you suggested.

Thank you

First you have to verify that you have pair antibidies that work with the virus by dot blot or elisa. Then apply the best pair to the membrane.

Dear Oded Babai,

Thank you very much for your response. Will test them as you suggested.

Before going to the antibody pair, is there a way I can test the afinnity of every individual antibody on a dot blot or lateral flow ?

Thank you

You can do a serial dilutions of the antigen as dot blot and see which antibody reacts with the low dilution

Thank you very much Oded Babai.

Why did u mention "Low Dilution" ? Does excess of antigen prevent binding of antibody ?

If you want to identify the best pair for lateral flow you MUST test in lateral flow, not dot blot or ELISA.