It could be due to a viral infection of your mice. Then the virus may have been cleared away and you couldn't detect the isotype switching anymore.

Hi, @Morgane Henry, thank you very much for your answer.

Our lab mice was infected by some kind of E.coli (from their food) before, then the lab technician changed the food and the mice condition is much better now. However, we tried to repeat the isotype switching by ip / po mice with E.coli. (sounds merciless), then i.p. our hybridoma cells, but still failed. I do not know whether our mice have viral infection before, but your answer is reasonable.

Have you ever seen or heard the phenomen by yourself ? While It is hard to explain why the viral or bacterial infected mice have the ability of somatic hypermutation or isotype switch. It is because the AID express in these mice? we tried transfect AID gene into the igG3 hybridoma cells, but still failed to get the functional IgG2a antibody. As the IgG2a mAb is very important for us, we want to find it back. It will be very appreciated of any advice on this.

Thanks again for your nice answer.

Sorry i cannot help further as I only have a little experience in asthma and IgE class switching.

Hi Jingyun. This is an interesting question. You may try (at least in vitro) to get the cells going on LPS in the presence of IFNgamma. IFNg promotes switching to IgG2a in the mouse (the original report was by Finkelman FD et al, J Immunol 140:1022, 1988). A second possibility is to stimulate the cells with LPS with simultaneous addition of TGFb and IL2, which would facilitate switching to IgA, but cutting the stimulation short, trying to catch some cells before they switch to IgA since IgA is the last isotype switch. Check Lebman DA et al (J Immunol 144:952), it's an old paper from 1990 but you can follow the trail and get some fresher report on the subject. In any case, it is a bit of a gambling experiment, just hoping to get some instability in the Ig genes of your hybridoma. Good luck.

Hi, Morgane Henry, that is all right, thank you all the same.

Hi, Maria Villacres, just as you says, replicating the IgG2a functional switching does be a bit of gambling experiment. Besides the instablility of the hybridoma Ig genes that may due to the complicated some kind of inflam environment in vivo, I think that my hybridoma switching is aslo antigen-dependent, while I do not have any solid evidence yet... Anyway, I am very appreciated for your advice, I will try it as soon as possible. Thank you very very much.

I am a bit mystified by your statement that "only the IgG2a mAb is functional". Functional in what way? If you are getting no binding to the target antigen recognised by the IgG2a antibody, it would suggest that class switching is not what is going on. Are you priming your mice with Pristane? Are they Balb/C mice? You may have an entirely new mineral oil-induced IgG3 myloma.

Hi, Dr. Graham Mayrhofer.

Well, It is a little complicated for the functional IgG2a mAb. Here is the detailed story : the molecule we are interested is specificaly expressed in human pDC (RNA level). It is a receptor, and its ligand is unknown at the moment. So we decided to make functional mAb ourselves and use this agonist mAb to examine the role of this molecule on pDCs. The IgG2a mAb stain pDCs very well and does change pDCs function to some extent. (While the IgG3 could not)

At the early start, we did not realize that the mAb comes from switching until we want to produce much more ascites... we tried many times to replicate the switching but not succeed yet.

The IgG2a mAb can bind the target antigen as I have confirmed it on overexpressed 293T cells. The mice we used are Rag2-/-γC-/- mice or NOD/SCID mice (No T,B and NK cells) and both of the chemical substance (pristane or mineral oil) were used to prime the mice before. I tried the switching on the wide type Balb/c mice too, but still failed.

We are re-screening some new functional mAb but still not succeed yet. The difficulty may due to that the angen expressed on pDCs is different from that over-expressed on cell lines(maybe the modification of folding). As the IgG2a can stain pDCs very well and the IgG3 could not stain pDCs at all, while the IgG3 bind the overexpressed 293T much better than IgG2a.

Well, on the whole, we really want to find the IgG2a mAb back. It will be very appreciated of any advice on this.

Thank you very much.

Hi Dr. Jingyun Li,

sincerely, I never observed this phenomenon in my laboratory. In this context, your question is important for all of us to be aware concerning this issue.

I have a question for you: how have you done the limit dilution to isolate the clone you were interested in?

Hi, Dr.Jose Mineo.

First of all, I am not a Dr. yet, I am a phD student at present. :-)

Well, When we did the hybridoma limit dilution, we just follow the "Current protocol of Immunology". Four round sub-clonings were conducted until we get the single clone, and the isotype of the hybridoma we get is IgG3.

While the clone we are interested in is the switched IgG2a, as I have mentioned above, it is coming from isotype switching in the peritoneal cavity. At the early start we did not know the isotype switching happened, and when we harvested the ascites we did not harvest the hybridoma cells...As we could not repeat the switching at the moment, we lost the IgG2a hybridoma cells we interested in.

Weid phenomenon, isn't it? My experience tell me that small probability event could happen. If your hybridoma is very important, I recommend that you'd better harvest the hybridoma cells when you harvest the ascites.

Hi Jingyun,

thank for the details of your procedure to carry out the hybridoma clonage. We actually use the same protocol. So, we have to be aware about this (low?) probability of the isotype switching when growing hydromas in ascites.

I still find this puzzling. Class switching should not change the specificity of the antibody, although it can change its biological properties. While I accept that the differences in binding between the targets that you use (pDC and transfected 293T cells) could be due to differences in folding or glycosylation of the antigen, this cannot explain differences between the IgG2a and IgG3 binding to the antigen when you use the same target. I am still suspicious that the IgG3 clone is not derived from the parent IgG2a parent clone (for whatever reason). Have you got frozen stocks of the parent from which you can produce culture supernatants and avoid ascites?

Hi, Dr. Graham Mayrhofer,

According to my results, the differences between my IgG2a and IgG3 mAbs is not only their isotype. Their binding properties does change, only the IgG2a can bind pDCs. In my opinion, besides the class switch recombination, the somatic hypermutation may happened at the same time and it maybe a progress of affinity-based selection. (although It sounds a bit unbelievable)

I believe that the IgG2a hybridoma is derived from the IgG3 parent clone, as the mice themselves do not have B cell. What's more, I am sure that the IgG3 hybridoma is a single clone without any other cell pollution before the i.p.

We frozed stocks of the IgG3 hybridoma cells, and their supernatants are IgG3. Unfortunately, we do not harvest the switched IgG2a hybridoma cells from ascites as we did not realize the swiching had happened.

Since you have the hybridoma, you could clone the heavy and light chain VDJ regions and express a recombinant IgG2a. Although this a a lot of work, it would definitively answer your question. I suspect that the IgG2a clone is binding very strongly to Fc Receptors (the high affinity Fc Receptor binds monomeric IgG2a the best). That might explain higher binding on pDC and lower binding on 293T (which lack FcR).

Hi, Amanda Gavin, first of all, thank you for your advice.

Well, We do have IgG3 hybridoma cells, and we have cloned its VDJ region, and combined it to some IgG2a Fc fragament. But, the expressed recombinant IgG2a could not stain pDCs either. So,I think besides the class switch recombination, the somatic hypermutation may happened at the same time and its VDJ maybe changed (maybe it is affinity-based selection). Besides, the different binding ability on pDCs and 293T is not because of FcR, as before I start the mAbs staining, I use the FcR block buffer.

Jingyun, I didn't realize that the specificity of your mAb (IgG2a) was also changed in addition to the isotype switch. It seems to me that a reasonable explanation is the accidental infection of the mice used to expand the hybridoma. Your choice here is to try to replicate this process, but with a virus, preferable one infecting DC. If I understand correctly, your ligand for the IgG2a is a DC surface protein. Some strains of LCMV infect preferentially DC, but would be safer to use inactivated virus. LCMV can quickly spread in a mouse colony if not properly handled. Have you considered an alternative like to isolate DC membranes to start a fresh immunization protocol?