Background haze in B cell FluoroSpot

[Guest Melissa K]

Hello! I am assessing B cell FluoroSpot data (FluoroSpot Flex mouse IgM/IgG) from my very first development assay and have encountered several issues which i'm not quite sure how to solve. I will start by saying that following restimulation, my cell viability was below the 85% threshold which may likely be a significant factor.

I am seeing quite a bit of background haze in the wells which is more pronounced as the plated cell # increases from 10K to 250K cells/well. It is fairly specific to the FITC channel, where there are also many more spots. However, my no cell control wells seem to have a fair bit of background in both FITC & Cy3. This seems to be in contrast to my mAb MT24/JC5-1 coated wells, where the spot resolution seems clearer (despite the high #).

At the same time, I am observing that my spots do not appear to be antigen specific, as the plus & minus antigen coated wells look extremely similar. Because of the other issues, i'm unsure if this is due to an fundamental assay issue?

I have followed your guidelines for buffer usage (no Tween), tapping plates on a paper towel to fully remove the Enhancer solution & ensuring the membranes were never subject to re-wetting & fully dry when reading. Other than a cell death issue, I would appreciate any recommendations you might have.

FluoroSpot test assay.pptx

[Jens@mabtech.com]

Hi Melissa!

You seem to have been doing everything right (no tween, completely dry plate, etc), and as you see clear IgG and IgM spots in the MT24/JC5-1 coated well, your cell handling and experimental setup looks good. So I'm sorry to hear you're still seeing issues. 

Our experience is that IgM can be quite sticky, and that may be the reason for the "haze" as well as the non-specific background spots in the PBS 'coated' wells. My co-worker Gun Kesa said the following: "IgM from hyperimmunized mice may bind to coated irrelevant antigens and/or uncoated ELISpot/FluoroSpot wells. This phenomenon is probably due to the high avidity of the IgM pentamer that allow it to bind detectably even to weakly cross-reacting Ag. This is quite cumbersome when using in vitro activated B-cells (polyclonally stimulated with R848 + rmIL-2)."

So the background seems to be in the nature of IgM from some samples, unfortunately. However, that doesn't have to be a problem: If you would have antigen specific IgM secreting B cells in your sample, you should see an increase in the number of spots compared to the PBS 'coated' wells. Judging from your ppt-file, you don't really see that increase though, and here lies perhaps a bigger issue: Either you don't have any antigen specific cells in your sample, or, if you indeed expect to see antigen specific responses in these samples, could it be something wrong with your antigen? Is it pure? If you for example have a lot of carrier protein with your antigen, then there is a risk that most of the membrane is coated with the carrier. In addition, is it possible that you have coated a too low of a concentration of the antigen? Do you have the possibility to try out different concentrations?

 

[Guest Melissa K]

Thanks for helping me understand what's happening a bit better here. I've purchased a fully characterized, pure antigen from a commercial source. There was PBS+trehalose in the lyophilized product and I have not added any protein when reconstituting. It's possible that I coated too little at 1ug/mL, 100uL/well. I was planning to increase the coating solution to 5 or 10ug/mL to be sure. In parallel, i've also fluorescently labeled the same proteins (w/an Avi tag) and see a nice positive population in unstimulated memory B cells from long-term vaccinated mice which can be further characterized by staining with IgG or IgM (with higher staining levels for likely the same reason you've provided). Given that I know I have antigen specific antibody producing cells, i'm wondering if it makes sense to run this assay both with & w/o restimulation? Based on your feedback, that would likely help clean up the assay. I've erred on the side of a shorter restimulation period here (2 days), but obviously that's more than enough to run into this issue.

[Jens@mabtech.com]

Hi again, Melissa,

Ok, great, you seem to have a pure and good antigen, and you know you have antigen-specific memory B cells present,  so indeed, it might just be a question of increasing the concentration of it. That will probably do the trick. 

Regarding the question of running the assay without re-stimulation (to possibly get a cleaner readout): I'm not sure. If I understand you correctly, you detected the antigen-specific memory B cells with flow cytometry? The question then remain whether you also have antibody secreting plasma cells in that B cell population - and that depends on how long the period is between vaccination and sample taking. I'm not sure of the kinetics in mice, but for humans you will normally find lots of plasma cells around 7 days post-vaccination. Do you have similar data for mice, and how does that compare with the timing of your samples (you mentioned "long-term vaccinated")?

What I'm trying to say is this: If you have plasma cells already in the non-stimulated sample, then you don't need to re-stimulate, you could just run the sample directly in the FluoroSpot plate. But if there has been a too long time since immunization, you will only have memory B cells present, and they won't secrete antibodies without re-stimulation. 

[Jens@mabtech.com]

Oh, and two other possible things to minimize the haze issue: 

1. I'm sure you wash the cells before adding them to the plate, but if you don't: wash them.

2. Evaluate incubating the cells for a shorter time period in the FluoroSpot plate. As short as 6 hours can work fine.