I used software to fit Sigmoid curve to get IC50 but the curve generated was not a typical Sigmoid shape. Why does it so? I have series of 10 different concentration of a compound and tested for in vitro cell inhibitions to get IC50 of the compound. Is that due to 2-fold serial dilution only? OR it does not matter either 2-fold or 10-fold serial dilutions? How should we decide whcih serial dilution is best to test?
I would say that 10-fold serial dilutions are too far apart to define a typical IC50 curve. People usually do 2-fold, half-log (i.e. 1, 3, 10,...) or 4-fold dilutions. Cell-based assays sometimes have steep IC50 curves. If the concentrations are too far apart, you might miss the transition from low to high inhibition, making the curve look like a step-function. I prefer to use 2-fold dilutions for enzyme assays, in which the Hill slope of the IC50 curve is usually expected to be 1, with concentrations covering a 1000-fold range.
Another issue can be insolubility of the inhibitor. This might make the curve flatten out before reaching 100% inhibition, or even rise then fall.
According to my experience, it is depending on the type of inhibitors. Synthetic compounds can be used for two fold, 3 fold and four fold serial dilutions and screen it. However, if a inhibitor is antibodies, it is impossible to use 2, 3, 4 and even 10 fold serial dilutions. If we use 2, 3, 4 or 10 fold dilutions, we will never get a nice Sigmoid curve but will get flatten or strange curve because absorbent are not much different each other concentration. In this case, we have to use 50 fold, 100 fold or 1000 fold serial dilutions.
Pls share your experience on this. Thanks.
I have never done an IC50 curve using an antibody as the inhibitor. Can you say more about the experimental system?
A respective antibody against a growth factor for specific cells of interest was used to get IC50 of the antibody. I found depending on how effective the antibody, it is necessary to make antibody dilution either 10 folds or 50 folds or 100 or 1000 folds or maybe 2 to 3 folds etc. It is a bit tricky than using synthetic compound. Sometimes, highest dose of antibody was similar effect to a lowest dose of antibody too. I am not sure why it is so? I have never seen such thing in using synthetic compound. For synthetic compound, if I used more, it showed more inhibition and never see similar effect as low dose of the same synthetic compound. Does anyone knows about this? This is why hard to get nice sigmoid curve of IC50 of some antibodies but not all. How should I solve this?
- Badges
- Science method