Hi Bo-Cheng

May I possibly know about the concentration of your capture antibody and the concentration of your Gold-conjugated antibody which mix with the MMP9 analytical?

Dear Lu,

Capture antibody I used is 1 mg/ml 0.5 ul on the membrane.

Detection antibody which mixed with gold is 1 mg/ml also, but only apply 0.25 ul in 20 ul gold solution.

Do you know or can you calculate the final concentration of your Au-labeled antibody?

1mg/mL is a very high concentration of immune assay.

It better to test a serials dilution of captured and labeled antibody combination. you may find a optimized concentration.

We often use 20~200ug/mL (mostly 50,100ug/mL) for capture antibody immobilization on PEG-modified PDMS substrate. 

"3. Subsequently mixed the antibody-gold conjugate with MMP9 analyte (1000ng/ml) for 30 mins,centrifuge 13000rpm 10mins. Replace supernant with PBS."

At this step, your aim is to let Au-labeled antibody combine with the antigen MMP9, is it right? You  need to know the relative amount of your antigen and antibody, it better be 1:1 of the molecular number for them.

Dear Lu,

I also consider that issue before, in my opinion, I think high density of capture antibody on the membrane should result higher intensity of color revealment as long as more antigen be captured. So that's why I choose 1 mg/ml as my prelim-capture antibody concentration.

Centrifuge is for removing the analyte which wasn't bind to the antibody might cause the hook effect and false negative.

PDMS is a 2 dimemsion platform and transparent material, but NC membrance is a 3D paper structure makes the color revealment more difficult. That's the another reason I put the concentration so high.

I will put the antigen and antibody amount be 1:1 and evaluate the result =)

I am a newhand in LFIA, so if there is anything wrong please comment.

 Have you tried diluting your analyte solution?  Since you say that you are using a high concentration of analyte, it may be that you are seeing a severe "hook effect".

Dear Richard,

I choose 1000ng/ml since the threshlod appximately 100ng/ml.

So it should be okay.

It might come from the capture and detection antibody will interfere each other. Because I didn't choose the antibodies already tested in sandwich assay. But the MMP9 surely can be used in sandwich. I think I didn't clearly elaborate this in detail at first.

If hook effect take place, it should also have some relatively low signal rather than nothing right?? because I silver enhance the reaction place and nothing happend.

BTW, I use O.D.=1 conjugate 20ul simply apply on membrane, is it too low??

So briefly speaking, for now

I am doing a plate based ELISA, and set capture antibody 1 ug/ml to test the performance. But I haven't done it before. Hope it will have some signal beyond 500nm ( gold label used as detection method.)

 I agree that even in the worst hook-effect case, you should see some signal in the capture zone.  You're other suggested approaches sound very reasonable and show that you have a good appreciation of the scientific method.  Good luck in your continued research.

Dear prof. Richard,

I think I should optimize the conjugate method first, becuase little aggregation will interfere the signal intensity.

But if the antibody couldn't be used in sandwich, I have to optimzie another antibody again to produce high signal.

Plate based might be appropriate now.Long way to go...

Thanks for your recommendation and encouragement