If you're going to use an agonist, it's better to make curves dose-responde to obtain the EC50; if you're going to use an antagonist, you should construct curves dose-response to obtain the IC50. That allow you to see how much of a particular substance you need to inhibit or increase a response asociated to an enzymatic event or binding to a receptor.
The Kd gives you information about the affinity of the drug with your receptor (or enzyme).
If i where you, i'll use IC50 to observate the effect of the response.
What if I don't know whether my drug is agonist or antagonist? I only know from computational study that this drug may target this protein ... I want to emphasize this information and also know the certain action of the drug whether agonist or antagonist. what do you recommend?
I recommend you serach in databases structurals comparison of your drug in first place, to know if will act like an agonist or antagonist. Other thing it's test this drug and the possible effect that will cause in some celullar line or an standard animal. There a lot of things you can do.
Kd is the dissociation constant for the enzyme-drug complex, and it is equivalent to Ki, the inhibition constant. Usually Kd or Ki are determined through biophysical techniques (ITC, SPR, spectroscopy,...) or enzymatic assays. Kd and Ki are equilibrium constant that depend on the experimental conditions, but should not be dependent on the concentration of enzyme.
IC50 is the 50% inhibitory concentration of a given drug. It is operationally defined as the concentration of drug required to inhibit 50% of the enzyme in a given in vitro assay. Therefore, IC50 depends on the concentration of enzyme, because, even if the affinity of the drug is extremely high, IC50 will be at least equal to the concentration of enzyme employed in the assay.
EC50, 50% effective concentration, is similar to IC50, but employed in cell-based assays. Similarly to IC50, EC50 depends on the amount (concentration) of cells employed in the assay.
Kd or Ki can be compared among different experiments (provided that the experimental conditions are similar), because they are intrinsic equilibrium constants. However, IC50 and EC50 can be difficult to compare if different experiments are performed with different enzyme or cell concentrations.
You can search Inhibition mode of your substrate (Competitive inhibitor, Noncompetitive inhibitor and Uncompetitive inhibitor) using Lineweaver–Burk plots, 1/[V] (M/min) and 1/[S] (mM).
To try with few points to start (concentration of inhibitors like 100uM; 10 uM; 1 uM) to have a first idea of the inhibition capacity of your product. Then, try an IC50 to have a more accurate value and then if you want informations of the mode of action of your inhibitor, you could try to obtain Ki.
It seems that you expect that your compound may increase the activity of the enzyme, which would make it an "activator" not an an "agonist," a term used for receptors. It would be useful to measure the maximal -fold increase in enzyme activity, and the concentration that produces half of the maximal effect (EC50) when comparing the effects of multiple compounds. The Kd (dissociation constant), a measure of the affinity of the compound for the enzyme, would also be of interest but might be more difficult to obtain.
I agree with aLeidy Gome, to have an idea about the relationship between the drug and the enzyme, you should do the enzyme assay in the presence & the absence of the drug( using different concentration of your substrate & then doing the experiment in using the same substrate concentration but this time in presence of your drug) then from line -weaver plot you can have an idea what type of effect you have and calculate the required constants ( Ki , Vmax,& Km