Protocol Online logo
Top : New Forum Archives (2009-): : Immunology

Monoclonal Antibodies - Hybridoma Screen (Jul/22/2009 )

Pages: 1 2 Next

I am continually having problems with monoclonal antibody production. I think our fusions are working since we do get hybridoma lines that are secreting antibody, but it seems that the majority of what we end up with is IgM antibodies or we lose them completely. We have tested various things, but nothing we have tried seems to improve our end results. More recently, after one round of single cell cloning, we ended up with a line that was predominantly IgG3/IgM. We then single cell cloned again and now it seems to be IgA with the possibility of some IgM as well (After narrowing our lines down by ELISA, we sometimes use Isotyping Strips from Roche Biochemicals and/or run Westerns on the supernatants in duplicate by loading them on SDS-PAGE gels. We then transfer, block, and probe with either IgG or IgM secondary antibodies. We can tell the isotype(s) by the signal from the Heavy Chain(s)).

After reading various publications as well as posts on this forum, I have a feeling that most of our problem lies in the post-fusion culturing/screening phase. I know some people may point to the immunization schedule, but our pre-fusion titers are good and our injection schedule/amounts are within range of most standard protocols. Our fusion protocol also seems to be similar to most posted in the literature. Our antigens range from GST or MBP fusion proteins to peptides conjugated to KLH or BSA. Although I would love for someone to be able to tell me exactly what to change, I realize that many things could be occurring to cause this problem. Any pointers would be greatly appreciated, but I have a few specific questions below that might be a step in the right direction.

1. When screening, the most thorough way to test is by using secondary antibodies to various immunoglobulins. If you had to pick just one, what class of secondary would you use? (At 96 well, we only have enough supernatant to test with one or maybe two secondary antbodies).

NOTE: We have been using an anti-mouse IgG (H+L). In recent months, we have added an anti-mouse IgM, u chain to the screens after the 96 well stage.


2. In most protocols that I have read, the suggested secondary is Alkaline Phosphatase, but nothing is mentioned about the class or whether or not it should be whole molecule. Should we be using secondary antibodies that are to the heavy chain only or does it matter?

In the past couple of days, my IgG (H+L) has given decent ELISA readings, but upon Western blotting of the supernatants, it seems that I have IgAs. Why would the IgG (H+L) react with the IgA? Perhaps through the light chain???


3. I have read about class switching in hybridoma cultures. It seems that it isn't the easiest thing to do when somone wants to try to switch Ig classes. How common is this in general culturing of hybridomas?
(I know I'm reaching here, but we don't understand why a line that didn't seem to have any IgA seems to have quite a bit of it after thawing and cloning).


4. Does the type of ELISA plate make a difference? We usually use high binding ELISA plates. Assuming our 0.8 - 1 ug of antigen saturates each well and we block thorougly, do these plates have higher background over regular 96 well plates or lower binding plates? (Our control wells seem to discount this).


5. For future reference...All of our previous fusions have used NS1 (ATCC TIB-18), but we just bought some Sp2/0-Ag14 (ATCC CRL-1581). It seems that both NS1 and Sp2/0 are used quite frequently for hybridoma fusions. Has anyone had a better experience with one line over the other?


I apologize for asking so many questions, but I wanted to be as thorough as possible.

Thanks! --Roo

-Roo-

Roo on Jul 23 2009, 06:17 AM said:

I am continually having problems with monoclonal antibody production. I think our fusions are working since we do get hybridoma lines that are secreting antibody, but it seems that the majority of what we end up with is IgM antibodies or we lose them completely. We have tested various things, but nothing we have tried seems to improve our end results. More recently, after one round of single cell cloning, we ended up with a line that was predominantly IgG3/IgM. We then single cell cloned again and now it seems to be IgA with the possibility of some IgM as well (After narrowing our lines down by ELISA, we sometimes use Isotyping Strips from Roche Biochemicals and/or run Westerns on the supernatants in duplicate by loading them on SDS-PAGE gels. We then transfer, block, and probe with either IgG or IgM secondary antibodies. We can tell the isotype(s) by the signal from the Heavy Chain(s)).

After reading various publications as well as posts on this forum, I have a feeling that most of our problem lies in the post-fusion culturing/screening phase. I know some people may point to the immunization schedule, but our pre-fusion titers are good and our injection schedule/amounts are within range of most standard protocols. Our fusion protocol also seems to be similar to most posted in the literature. Our antigens range from GST or MBP fusion proteins to peptides conjugated to KLH or BSA. Although I would love for someone to be able to tell me exactly what to change, I realize that many things could be occurring to cause this problem. Any pointers would be greatly appreciated, but I have a few specific questions below that might be a step in the right direction.

1. When screening, the most thorough way to test is by using secondary antibodies to various immunoglobulins. If you had to pick just one, what class of secondary would you use? (At 96 well, we only have enough supernatant to test with one or maybe two secondary antbodies).

NOTE: We have been using an anti-mouse IgG (H+L). In recent months, we have added an anti-mouse IgM, u chain to the screens after the 96 well stage.


2. In most protocols that I have read, the suggested secondary is Alkaline Phosphatase, but nothing is mentioned about the class or whether or not it should be whole molecule. Should we be using secondary antibodies that are to the heavy chain only or does it matter?

In the past couple of days, my IgG (H+L) has given decent ELISA readings, but upon Western blotting of the supernatants, it seems that I have IgAs. Why would the IgG (H+L) react with the IgA? Perhaps through the light chain???


3. I have read about class switching in hybridoma cultures. It seems that it isn't the easiest thing to do when somone wants to try to switch Ig classes. How common is this in general culturing of hybridomas?
(I know I'm reaching here, but we don't understand why a line that didn't seem to have any IgA seems to have quite a bit of it after thawing and cloning).


4. Does the type of ELISA plate make a difference? We usually use high binding ELISA plates. Assuming our 0.8 - 1 ug of antigen saturates each well and we block thorougly, do these plates have higher background over regular 96 well plates or lower binding plates? (Our control wells seem to discount this).


5. For future reference...All of our previous fusions have used NS1 (ATCC TIB-18), but we just bought some Sp2/0-Ag14 (ATCC CRL-1581). It seems that both NS1 and Sp2/0 are used quite frequently for hybridoma fusions. Has anyone had a better experience with one line over the other?


I apologize for asking so many questions, but I wanted to be as thorough as possible.

Thanks! --Roo


All fusions will give you IgM secreting clones. Often they greatly outnumber IgG producing clones. The proportion of IgG to IgM hybridomas depends on the antigen and the length of the immunization scheme. A longer scheme and more boosts give a better opportunity for IgM to IgG class shifts (potential for more IgG producing hybrids). Also longer immunizations gives you antibodies with better affinities due to "affinity maturation" by somatic mutation.
We produce monoclonals for immunometric immunoassays. Since we need very high affinities and only IgG antibodies we give mice 3 doses of antigen over 5 - 7 months before fusion. We also screen using only anti-IgG secondary antibodies (thus we exclude IgMs clones during the 1st screening). If you want both IgMs and IgG producing clones why not use a single rabbit anti-mouse immunoglobulin secondary reagent. This will pick up all IgM, IgG, IgA etc. By the way, if you are fusing spleen cells, IgA antibodies are very, very, very.... rare!

You should expect to lose <10% of the positives found in the primary screen. If you are losing more you should 1) check the myeloma for mycoplasma, 2) check that your assay works every time (use good controls eg some of the immunized mouse serum as the positive) 3) make sure you do not grow the primary clones too long before subcloning. The best positives from the primary screen should be subcloned directly from the 96 well fusion plate!

After subcloning you say you have a clone that is IgG3/IgM. You either have two clones (failed subcloning) or there is something wrong with your isotyping. (a "monoclonal" is monoclonal!!!)

I have not seen people use westerns for isotyping. Are you SURE that the anti-immunoglobulin reagents you use are specific on denatured immunoglobulins (check supplier info that they are fine for westerns).

Reply if I can help further

-klinmed-

Thanks for the reply. I know that all fusions will give you both IgM and IgG (and possibly other isotypes), but after doing numerous fusions using various different antigens (peptides and recombinant protein), we always seem to get a high number of IgM antibodies, but perhaps our assay methods should be changed. Below, I've listed more info about what we do, but I will look over your suggestions more closely with the people that I am working with at the moment.

Immunizations
Our immunization schedule usually includes a priming injection (50 -100 ug protein /mouse; sometimes with Freund's other times injected without FCA but still bound to resin (GST, His, or MBP)). Our boosting injections are done in approximately 3 week intervals (usually 3; ~50 ug/mouse +FIA or resin). We do retro-orbital bleeds 10-12 days after each boost to check titers (by Western - HRP and ELISA-AP). Assuming the titer is decent after the 3rd boost, we rest the mouse for 1 -2 months before fusing. (We've compared this to published protocols as well as company protocols, and it seems ours is in line with what other labs are doing).

Screening
Typically 10 days to 2 weeks after the fusion, we perform our first ELISA screen. We use ELISA plates coated with 0.8 -1 ug of antigen (one time we used 0.1 ug and it worked just fine). We block in PBS + 1% BSA, although I tried adding FBS as well based on a something I read in a book, but it didn't seem to change anything. We incubate in 100 ul of supernatant from the 96 well for 2 hours and wash 4x with PBS or PBS+0.02% Tween 20 (tried both ways, it doesn't seem to make a huge difference). We then incubate 1 hour in goat a-mouse IgG (H+L). In our last fusion, we screened 642 wells (96 well) and only 43 were positive. However, we then transferred them to 24-well, waited a day or so and screened again. Only 39 lived through the 24-well stage and of those, 28 were positive. Once again, we transferred these to 6-well, waited 1 - 2 days depending on their growth rate and screened again. In the end, only 25 were positive. These got frozen down and screened by Western before picking one or two for single cell cloning.

Single Cell Cloning
After single cell cloning the last time, it was apparent that our "single cell clone" was probably not single cell cloned. I was able to see signal by Western blot against two forms of the antigen using either the IgG (H+L) or IgM, u chain only secondary antibody. I then isotyped using a commercial chromatography style isotyping strip (Roche Biochemicals). It confirmed that I in fact had IgG3 and and IgM, both Kappa light chain. We performed another round of single cell cloning. I then took one of the clones that had a high positive at 96 well and did an ELISA with either IgG (H+L) or IgM, u chain only secondary (0.638 IgG/0.599 IgM ELISA readings). I also did an isotyping Western (HRP secondaries that we always use for Western). This is when the IgG (H+L) secondary only showed the light chain and the IgM, u chain only came up a little in a couple of the lanes (multiple "clones"). I took one that seemed to have a decent titer based on the Western and isotyped it with one of the Roche isotyping strips. It comes up as IgA/IgM. Yes, I know they are probably not single cell cloned again. Right now this is the least of my worries. We just don't understand why we now have an IgA when there wasn't one before, and where did the IgG3 go? Did these cells die? Even if they did, we still can't believe there is an IgA in the mix, especially since they are so rare.

I just ordered heavy chain only secondary antibodies for IgA and IgG. I will do an ELISA on several with each of these to see if we can fish out what we are looking for. We really would like an IgG antibody, but if all we get is an IgM or an IgA, then we will go with that. We thought that using an IgG secondary would only pick up IgG clones, but after doing this for a long time, we are finding that the majority of what we are ending up with is IgM. I'm thinking that maybe some of the antibodies are reacting through the light chain so perhaps using a heavy chain only antibody will alleviate this problem. I also read in a couple of publications and texts that IgMs are notorious for high background, especially in plasticware. I do believe this is true since the ones that have the highest IgM titer seem to have the highest background by Western blot. I'm having to block longer and wash longer than normal to see my ECL signal in these blots. Still, we do have some wells that are low to IgM so I'll assume for now that background isn't a huge problem. (We are also running positive and negative controls to address these issues).

I hope I haven't confused you, but I'm trying to give you as much info as possible so hopefully I can figure out the best way to approach this.
Thanks for the pointers so far!

By the way, you said...
I have not seen people use westerns for isotyping. Are you SURE that the anti-immunoglobulin reagents you use are specific on denatured immunoglobulins (check supplier info that they are fine for westerns).

I know this isn't the norm, but I have done this for years with great success. We got this idea after seeing our light chain and IgG heavy chain in IP reactions. We use Protein A/G resin for our IPs, but as you know, this binds the antibody used for IPs. In addition to protein bands that we are looking for, we often see the heavy and light chain bands come up in the Western because the secondary ab recognizes it. We decided to use this as another way to isotype. It isn't the only way, but it is cheaper than using the isotyping strips and I can do a lot a once. Once we have narrowed it down, we then use more widely used methods. For this assay, I use the same secondary antibodies I always use for my Western blots (ECL, but AP should work too). Running 2 - 5 ul of antibody supernatant (purified ab works too) on an SDS-PAGE gel denatures it and separates the Heavy and Light Chains. If you block, then hybridize with the secondary, it should pick up the heavy chain (and light chain if a whole molecule antibody) of the particular Immunoglobulin the antibody is made to. Remember, the light chain is the same so you will see a 25 kD band in all if using whole molecule secondary. I have only done this with IgG and IgM. I was thinking about trying it with the IgA, but I haven't done it yet.

I'll post again if my heavy chain only ELISA work later this week.

-Roo-

Update...

I don't know if this will help anyone else, but I just finished the test ELISA using heavy chain only secondary antibodies to IgA, IgG, and IgM. I tested several of our pools that are always positive to both IgG and IgM when using IgG (H+L) 2˚Ab and IgM, µ chain only 2˚Ab. This time, they were only positive to IgM. My guess is that the majority of lines that we have pulled out of our hybridoma screens are IgM, but we were following them as if they were IgG since we pulled them out with the IgG whole molecule secondary. Why this continues to happen is a mystery to me. All I know is that from now on, we will use the heavy chain only antibodies for the ELISAs or at the very least confirm any positives (early on) pulled out with the whole molecule secondary antibodies by retesting with the heavy chain only secondaries.

--Roo

-Roo-

Hi, You'll be pleased to know that we too are experiencing the same thing with regard to our fusions. We also seem to be getting predominantly IgM. We originally isotyped with the Roche strips and based on this all our clones were IgM, with a kappa light chain. I couldn't believe this as with our immunization schedule, class switching should have taken place. I then purchased isotyping reagents from another supplier and performed an ELISA. The ELISA yielded similar results, with one of our clones now appearing to be IgM/IgG1! I just don't know what to think anymore!

-MouseMab-

Hi,

We have a similar problem, and I can not understand why. ;)

After the immunization schedule we have good antibody titers (using an anti-IgG secondary antibody in ELISA), we make the fusion using myeloma NS1 cell line (mycoplasma free) and after some days of culture (using HAT and HT media) we found positive clones using an ELISA to detect them, using two different ELISA, colorimetric (secondary antibody anti-IgG-peroxidase), and fluorometric (secondary antibody anti-IgG-Europium) ... but when we do the isotype (CBA, Becton Dickinson) almost all the clones are IgM. The mouse serum as control gives all isotypes.

I think the problem may be in the selection of positive clones according to the ELISA results, we selected and cloned according to the results of ELISA. Do you know if IgM hybridomas may divide faster than IgG hybridomas and grow before?

For established IgM-secreting hybridomas, is possible make switching in vitro inducing mutagenesis by ICR-191, but it's a long process, and does not guarantee to have a stable hybridoma clone secreting specific IgG.

I will follow and participate in this link to see if between all find an answer to this mystery.Lucky for all


buscar

-tonix37-

Roo on Jul 28 2009, 11:23 AM said:

Update...

I don't know if this will help anyone else, but I just finished the test ELISA using heavy chain only secondary antibodies to IgA, IgG, and IgM. I tested several of our pools that are always positive to both IgG and IgM when using IgG (H+L) 2˚Ab and IgM, µ chain only 2˚Ab. This time, they were only positive to IgM. My guess is that the majority of lines that we have pulled out of our hybridoma screens are IgM, but we were following them as if they were IgG since we pulled them out with the IgG whole molecule secondary. Why this continues to happen is a mystery to me. All I know is that from now on, we will use the heavy chain only antibodies for the ELISAs or at the very least confirm any positives (early on) pulled out with the whole molecule secondary antibodies by retesting with the heavy chain only secondaries.

--Roo


I assume you have this worked out, but I'll throw in my two cents in case it helps others...

As you reasoned out here, you are seeing double positives for IgG and IgM when screening by ELISA using anti-IgG(H+L) and anti-IgM becasue the antiIgG(HL) has anti-kappa and anti-lamba antibodies which will make an IgM,k monoclonal hybridoma look identical to a pool of IgG and IgM antibodies when it is not. We use a very specific anti-Fc or light chain antibody for our hybridoma screening depending on what we are looking for. (You can do this at the titering stage as well to look just for IgG vs. IgM antigen-specific titers) You can do lots of screens all at one if you want (anti IgM, anti-IgG, anti-kappa, anti lambda, anti-IgA) just by aliquoting small amounts the supernatant onto all of these plates in replicates - we have dons as many as 5 at once (a drag, needless to say).

As for your immunizations, the place where isotype is usually determined, it can depend on your specific antigen, adjuvant and schedule. Using a protein antigen like you are should elicit mostly IgG, but that still depends on your specific antigen. If you are using immunizing with a carbohydrate antigen, you may elicit mostly/only IgM antibodies. Your schedule looks right on, but one question, after resting your mice 1-2 months, do you i.v. boost with antigen w/o adjuvant 3-4 days prior to fusion? This is important, so I assume you are doing it, but I didn't see it in your methods.

As for the number of hybridomas, it can vary from 0-75% of the wells positive, and to be honest, I find that the titer of an animal still does not directly correspond to the success rate of good hybridoma production. It's still a numbers game.
We also see a fair amount of attrition after expansion. In the old days, people were very very careful with the growing hybridomas because they are so fragile - some will limiting-dilute directly out of the 96 well plate.

tonix37-
It looks like you are using an IgG whole molecule antibody which has anti-light chain antibodies. This will have you chasing IgM antibodies instead of looking specifically for IgG from the initial titer check.

Hope something in there helps.

-wirly-

Hi all,

wirly, in the case of anti-IgG-peroxidase, this antibody is specific to gamma chain (Sigma Aldrich A3673) in the case of europium-conjugated secondary antibody is an anti-IgG (whole?) (Perkin Elmer AD0124).

The results are similar in both immunoassays. We developed the second based on time resolved fluorescence to increase sensitivity in order to detect the presence of antibody some days before that if we use the classical ELISA with peroxidase-bound secondary antibody, to see if the time to perform the ELISA influence in development and "selection" of clones secreting IgM .... but the results are similar and when we did the isotype were IgM

any idea?

thanks in advance for your comments and suggestions

-tonix37-

tonix37 on Nov 10 2009, 02:52 AM said:

Hi all,

wirly, in the case of anti-IgG-peroxidase, this antibody is specific to gamma chain (Sigma Aldrich A3673) in the case of europium-conjugated secondary antibody is an anti-IgG (whole?) (Perkin Elmer AD0124).

The results are similar in both immunoassays. We developed the second based on time resolved fluorescence to increase sensitivity in order to detect the presence of antibody some days before that if we use the classical ELISA with peroxidase-bound secondary antibody, to see if the time to perform the ELISA influence in development and "selection" of clones secreting IgM .... but the results are similar and when we did the isotype were IgM

any idea?

thanks in advance for your comments and suggestions


Your results are strange. The anti-Gamma specific antibody should only detect gamma chains so unless you have two or more clones in each well (one secreting IgG the other IgM, you should get IgG antibodies out of any wells positive with the Sigma antibody. The Perkin ELmer antibody is anti-mouse Ig (it reacts with all mouse immunoglobulins) so it's really not helpful data if you are looking for IgGs.

You say the results are similar, but are they exactly the same?
How many colonies do you get per well?
Do you perform any media changes between initial 96-well plating and initial screening with anti-gamma? Ho many?
Is every well positive with both secondaries?
Does every antibody end up being IgG in the end?
If not, what is a rough percentage for your double postive wells/single positive wells/resulting IgG/resulting IgM

-wirly-

hi all,

wirly thanks for your help.

You say the results are similar, but are they exactly the same?
In the cases I have tested with both ELISA the result are double-positive. Not all samples are tested by both immunoassays. We have an established cutoff at which the mean +2 sd of the negative control is positive, and high positive sd +4, and is consistent in both ELISA, the high positive clones are found in both ELISA, and negative clones are negative in both ELISA

How many colonies do you get per well?
The problem occurs in different steps of the cell fusion: in the primary wells fo cell fusion, and in the wells of cell clonnings at 1 cell/well, with 1 clone microscopically visible.

Do you perform any media changed between initial 96-well plating and initial screening with anti-gamma? Ho many?
In the case of primary cell fusion wells (postfusion) we did HAT media change (day 7), HT (day 11), 10% FCS (day 18), hybridoma clones are very apparent and we do the ELISA from primary wella between the day 20-25 . In the case of cloning, once we limit dilution add 200 microliters / well and we ELISA 7-10 days.

Is every positive well with both secondaries?
Yes. Not always we do the double ELISA, the ELISA with FTR (time resolved fluorescence) is a screening test to perform before colorimetric test in order to select and expand positive hybridomas more quickly. With qualification + = mean +2 sd negative control, + + = mean +4 sd the negative control wells are positive with both ELISA. The scale is different because the colorimetric ELISA signal are between 0-3, while the signal we FTR are between 0-90000.

Does every end up being IgG antibody in the end?
Although serum hiperimmune have IgG, the resulting hybridomas are all IgM and IgA and IgG3 (postive primary well cloned- positive clon (1cell/well) recloned)

If not, what is a rough percentage for your postive wells double / single positive wells / Resulting IgG / IgM Resulting
I haven't result of all isotypes, but over 40 positive, 37 IgM: 2 IgA: 1 IgG3


I do not know what happens... What do you think about:
- Batch of fetal calf serum can influence the hybridoma growth rate, and IgM-secreting hybridomas grow faster than IgG-secreting hybridomas ?
- We use a layer of macrophages as feeder layer in the early days postfusion, but we found the same problem while to generate other monoclonal and most clones were IgM hybridoma using the enhanced supplement supplement from Sigma. What kind of supplement could be better to grow the hybridomas on the first days postfusion, or after cell cloning??

thanks in advance for your ideas, comments, suggestions and time

-tonix37-
Pages: 1 2 Next