I am trying to measure the IC50 of a kinase inhibitor in cells. The approach I am using now is Western blotting. However, I have a lot variations from experiment to experiment and with/without normalization to loading controls (such as actin). Does anyone have experience of doing this? Or should I use a more quantitative approach, such as ELISA?
ELISA is much more quantitative and will be less ambiguous of a result, of course.
If you have access to an IR scanning system for Westerns such as a LiCor Odyssey, then you can do a 2-color Western which allows same-sample normalization and lets your WB become quantitative as well (more than just based on loading controls). In this situation you would use a total-ProteinX antibody and a phospho-ProteinX antibody from different species and divide your phospho signal from your total signal.
I agree with Jordan. Also in ELISA you can run more replicates in parallel when compared to Western where you are kind of limited by the number of lanes you have on the gel.
Also, from my experience with drugs and viability as readout, your absolute values will also strongly depend on standardized conditions wrt initial cell density, keeping time points and things like that. Small changes in the conditions might sum up to considerable changes in the IC50 values you determine. Even when you use a quantitative, reliable readout.
Good luck,
Cindy
Definitely, ELISA. Western blotting is only semi-quantitative at best. With ELISA you can include a calibration curve with each experiment to allow normalization between experiments.
Hello Wei,
If you do not have access to ELISA I can suggest something similar to to what Jordan said above. Use the same lysate und 2 membranes (gels with same µl loading run in parallel) and stain for total prot on M1 and phospho prot on M2.
You results may still vary (in WB and ELISA) due to a lot of variables as Cindy mentioned. It really depends how accurate you need/want to be. Do it 3-fold and take the mean.
Hope that helps
Thomas is correct that you can do that if you do not have access to a 2-color system, but it will still only be semi-quantitative. Be careful about how sure you are about loading, because using 2 membranes means there are new variables (different rates of membrane transfer, different blocking efficiencies, different exposure times, etc). The ability to do a 2-color WB on the same membrane means you are detecting from the exact same loaded sampled, not parallel samples. This is the only way to get a true quantitative WB.
- Badges
- Science method