7 replies to this topic

[lical208]

Hellow,
I am trying to use ELISA to measure the murine serum Ig concentrations. I used the kits “SBA clonotyping system” (SouthernBiotech) and “mouse Ig isotype panel” (SouthernBiotech) for the measurement. But the concentrations I got are much lower than the published data. I have no idea what is the problem. Attached is the measured data and the analysis. The concentration range for IgM standar is 0-100 ng/ml.
Could anybody give me some ideas? Many thanks!

Attached Files

•  IgM.xls   21.5K   122 downloads

[PAO_ahac]

You don't indicate if the kit is designed to measure IgM (which I am assuming you are measuring) or IgG.

Is the kit designed for all isotypes or specific ones?  Is the calibrator representative of a specific isotype?

The kit could measure isotypes differently.  One type measures low another high.  Both the samples in the system and the calibrator isotypes should match.  If not...then purchase a purified one and substitute that into the system using your sample matrix.

Another issue could be the matrix of your samples and that of the kit do not match and behave differently. Serial dilutions of your high sample in the kit 0 calibrator should give you a straight line.

[lical208]

PAO_ahac, on 14 February 2012 - 03:26 AM, said:

You don't indicate if the kit is designed to measure IgM (which I am assuming you are measuring) or IgG.

Is the kit designed for all isotypes or specific ones?  Is the calibrator representative of a specific isotype?

The kit could measure isotypes differently.  One type measures low another high.  Both the samples in the system and the calibrator isotypes should match.  If not...then purchase a purified one and substitute that into the system using your sample matrix.

Another issue could be the matrix of your samples and that of the kit do not match and behave differently. Serial dilutions of your high sample in the kit 0 calibrator should give you a straight line.

Thanks PAO_ahac.

The kit can be used for measuring all the isotypes or a specific Ig, such as IgM. For IgM measurement, the purified IgM (100ug/ml) was use to build up the calibrator.

I will check the matrix of my sample...

[PAO_ahac]

I found the package insert and it indicates you basically have to optimize the kit and determine the performance characteristics on your own. there are no guarantees!


I would obtain some purified IgM (powder if possible) from several sources and see if they produce similar signals at the same concentration (midpoint of the curve).  Sometimes I just don't trust the do it yourself test kits!  

You also stated that your results are lower than published data...are you duplicating someone elses experiment...is everything exactly the same...including this kit and the test conditions used in this kit to determine concentrations?

[lical208]

Hi PAO_ahac, thanks for you reply.

Actually, I just followed the manufacturers' introduction. Briefly, Corning Incorporated plate was coated with 5 μg/ml of unlabeled goat anti-mouse capture antibody (SouthernBiotech), 4C overnight. Then, the plate was blocked with 1% BSA in PBS at RT for 1 h. After 3xPBST (0.05% Tween 20) washes, serum sample was incubated on the plate in a dilution of 1:1000 at RT for 1h (the online protocol refers to 37C, is it necessary?), with shaking. Plate was washed with PBST and bound autoantibodies were detected with HRP-conjugated goat anti-mouse IgM antibody (SouthernBiotech) (RT, 1h). Colorimetric reaction was produced by incubation with ABTS substrate. OD was measured at 405 nm at 10 min or 20 min. The IgM standard curve was run as described above but using a purified IgM instead of serum sample.

Because there is no control, I have no idea if my performance is correct or not. Normally, people used different kits gave different values. I search the IgM standard cure online and found that usually people detected the absorption of HRP at 450 nm and their OD values are much lower than mine with the same IgM concentration (http://www.allerresp...gmImage002.gif; http://www.mediomics.com/image/IgM.gif).

I did not duplicate someone else's experiment. I just had a look at their published data. But I also found different people had different values of the WT control mice (not only for IgM but also for other isotypes). Could this be due to different mouse strains or different conditions? However, the IgM concentration of WT mice in several papers displayed a value of ~200-500 ug/ml, while I got only 50-70 ug/ml. That's why I argued myself if I did everything right?

A online protocol suggested the IgM standard curve should start at 100 ng/ml. But from my data, after 100 ng/ml, the curve should not be linear any more. This really confuses me. Is there something wrong with my standard curve??

[lical208]

The purified IgM is from southerBiotech too.

[PAO_ahac]

Now I have a better understanding.  Your dose response curve is fine.  I would extend the concentration range maybe up to 500 to 1000 ng/ml; until you reach a plateau to see the actual range of the curve. Also throw in some purified IgM from companies other than southernbiotech.

In the research you did: different articles, different conditions, different kits, etc etc.  I bet the way the values were assigned to the standards/calibrators were all different and that is the reason your concentrations are off by a factor of 3-4.   It is the value assignment of the calibrators that I would focus on.

And remember, this is a research kit it is not a clinical diagnostic where values are much similar from instrument to instrument lab to lab day to day month to month and where commercial controls are readily available.

[lical208]

Thanks so much PAO_ahac.

I got confidence after your statement. I see what the problem is and I will repeat the experiment as you suggested.

Many many thanks .