Can we used spot size or morphology to distinguish memory B cell from Plasma cells?

[mazor]

Hi,

We are trying to use B cell ELIspot to evaluate the effect of a drug on the number of activated memoty B cells or plasma cells in mice that have pre-existing antibodies to an antigen.

 

Is there a way to distinguish if the spots we get are from a memory B cell (that were activated in vivo) or from a plasma cell?

 

Thank you,

Ronit 

[Christian@mabtech.com]

Dear Ronit,

 

Thank you for the question. As I understand the situation, you have a drug that you think can effect B-cells in a model system were you evaluate pre-existing antibodies. Your hoping that the drug can reduce the number of antigen specific B-cells, turning them off essentially. Would be nice in a setting of immunogenicity for example. 

 

In your setup, are you taking the blood or the spleen from these mice?

 

Either which, I think my understanding of B-cell immunology essentially says that plasma cells only reside in bone marrow. Not 100%, but almost. As a result, any spots you observe in your ELISpot should in theory originate from memory B-cell that have turned into a form of circulating plasmablast upon antigen exposure. They can be found in the blood and in the spleen.

 

Give me some more info and we can digg deeper. In addition, I will ask people at Mabtech with more experience about B-cells to comment.

[mazor]

Hi Christian,

Thank you for the rapid response.

We tried both bone marrow and spleens.

both work, depending on the time elapsed from the last challenge. 

 

Do you have any references where it was determined that antibody secreting cells in the bone marrow are plasma and antibody secreting cells in the spleen are memory? 

 

In addition, I am going back to the initial question- is there a way to use spot size or shape to determine the type of cell.

If you look at the attached well images you will see that some of the spots are very large and blurry and some spots are small and dens. These spots came from a spleen of an immunized mice (stimulation was 3 days before harvest)

I am wondering if those two types of spots represent different cell populations. and if so which populations. 

[mazor]

x

[Christian@mabtech.com]

Dear Mazor,

 

While being no B-cell expert I think the general consensus is that long lived plasma cells reside in the bone marrow where they secrete antibody for year (even decades). Now like in all things immunology, things are not 100%. That is why I said "almost" in my first response. Here is a link to PDF after doing some googleing: 

 

www.roitt.com/elspdf/Plasma.pdf

 

Regarding your direct question about being able to seperate different B-cells based on the morphology of spots my response would be: No, I dont think so.

 

Your spots look very good but I see no basis for being able to seperate them into different subsets based on morphology. However, would it be possible to do it using FluoroSpot? I am just speculating here but would it be possible to seperate different B-cell populations based on some cytokine that only the plasma cells secrete? 

 

In your setup you are coating with the antigen, right? 
 

At Mabtech we strong advocates of the reversed B-cell ELISpot/FluoroSpot protocol whereby you coat with mouse anti-IgG in the bottom instead of antigen. You then detect spots using a biotinylated antigen (elispot) or fluorephore labeled antigen using FluoroSPot. By this approach you are able to look at B-cells secreting antigen specific IgG but also cytokines, such as IL-6 or IL-10.

 

Perphaps there is basis for saying that only plasma cells secrete IgGs and a particular cytokine, whereas memory B-cells do not?

 

 

I will ask more people to contribute. Hold on.

[Jahnmatz]

Dear Mazor

 

I have looked into the your very interesting question and found a possible way to identify both memory B cells and plasma cells from your samples.

 

Like Christian said, Plasma cells are usually found in the bone marrow after the infection has been cleared, but they can also be found in the spleen of mice during an infection. Recently, antibody-producing plasma cells have also linked as regulators of the immune response. In this review, they have focused on the cytokines produced by plasma cells in mice: http://www.sciencedirect.com.proxy.kib.ki.se/science/article/pii/S0952791514000375. Pretty cool stuff, but it is hard to distinguish between long-lived plasma cells and memory-B-cell plasma cells in this context.

 

I have not yet identified a cytokine that is only secreted by activated memory B cells that has differentiated into plasma cells but not long-lived plasma cells, or the other way around. However, since your problem is an interesting one, I will continue looking.

[mazor]

Thank you both for your help.

It look like it would probably be more simple to use flow cytometry to distinguish the two populations.    

 

I am still curious about the two spot populations that we get from spleens. I think we do not get the small ones when we use bone marrow from the same mouse...

[Kajsa]

Thank you for your question!

 

My short answer would be: No, I would not distinguish between the cells based on spot size.

 

I think that by looking at the size you would get a hint on whether it is plasma cells or memory cells but I would not state it in the result section of your paper. Perhaps you could discuss it in the discussion section.

 

 

WHY?

 

To answer this I would go back to what we all know about plasma cells and memory cells: In the spleen you have the early reaction against antigens which forms plasmablasts (low affinity but antigen specific antibodies, possibly switched) and around 7 days post immunization, germinal centres are seen, generating plasma cells and memory cells (switched and with high affinity). Some plasma cells will subsequently migrate to the bone marrow where they become long lived plasma cells. So, in the bone marrow you will likely have plasma cells and not so many memory cells… but also many B cells under development, meaning they have more polyreactive and autoreactive BCRs. Will they respond to your antigen with antibody production? – Possibly?

Furthermore, memory cells may also form plasma cells upon restimulation. Would you then be certain that your “plasma cells” found in the spleen three days post immunisation really are plasma cells and not restimulated memory cells? I don’t know.

 

BUT THEN AGAIN…

 

I thought I would also just mention the work of the Bryan Charleston lab. It might be of interest since they study plasma cells and memory cells after immunization using ELISpot BUT in cows so I don’t know how applicable their results are to you. Anyway, they argue that if you take PBMC from the immunized animal, the spontaneously formed spots are from plasma cells since they produce antibodies without further stimulation. If the cells are cultured for 6 days without stimulus they don’t see any spots but if the PBMC are stimulated with a mix of cytokines (IL2, IL10), anti-CD40 Ab and pokeweed mitogen they claim that they trigger the memory B cell pool, thereby all spots seen after 6 days in culture are memory B cell spots. The equivalent that we offer for mouse is the polyclonal activator R848 + IL2 in case you want to try  The paper is:  A quantitative assessment of primary and secondary immune responses in cattle using a B cell ELISPOT assay Lefevre et al. Vet. Res. (2009) 40:03.

 

 

You were however thinking that you would just FACS them and see instead. I think that is a good idea, however maybe not instead. Is this your big finding that you will base your paper on? Do both.
 

For inspiration: Dogan et al. Nat Immunol. 2009 Dec;10(12):1292-9. Multiple layers of B cell memory with different effector functions. Interesting for you is that at 3 days post antigen boost AID+ B cells, here memory cells, were transformed into plasmablasts.

Gating for FACS: Kathryn A. Pape et al. Science. 2011 Mar 4; 331(6021): 1203–1207. Different B cell populations mediate early and late memory during an endogenous immune response. I normally do a simple Fas and GL7 staining for GC, pregated on B cells and use with a not B cells dump channel.

 

 

With FACS you may see antigen specific plasma cells and memory B cell percentages (I recommend you enrich the B cells first (using e.g. MACS) to get the chance to even see them – make sure to recalculate the percentage to real numbers) but you can’t see their antibody producing capacity. You can always sort the cells and then plate them for ELISpot or FluoroSpot and check for a difference in the spot size to further straighten you discussion. If you think your treatment might affect the antibody production, do ELISpot and FluoroSpot in addition to FACS. If you think that the treatment only affects cell number – FACS might be enough.

 

Good luck

 

Kajsa