spa
|
|
Competitive binding data with one class of receptors
Fitting data to a one-site competitive binding curve
Follow these steps to fit data to a one-site competitive binding equation:
2. From the curves section, choose nonlinear regression.
3. Choose the one-site competitive binding equation
4. If you choose to minimize the sum of the relative distances (as percent of Y), click on the Methods option button and choose "Minimize relative distances".
5. If you want to fix the top and bottom plateaus to constant values, click the Constants button and enter the values.
6. From the nonlinear regression dialog, choose the option Ki from IC50 and enter values for the Kd of the radioligand and its concentration. Enter both in nM (or any concentration units; only the ratio matters). Enter concentrations, not the logarithm of concentration. The Kd must be known from previous saturation binding experiments
Checklist for competitive binding results
When evaluating results of competitive binding, ask yourself these questions:
| Question |
Comment |
| Is the logIC50 reasonable? |
The IC50 should be near the middle of the curve, with at least several concentrations of unlabeled competitor on either side of it. |
| Are the standard errors too large? Are the confidence intervals too wide. |
The SE of the logIC50 should be less than 0.5 log unit (ideally a lot less). |
| Are the values of TOP and BOTTOM reasonable? |
TOP should be near the binding you observed in the absence of competitor. BOTTOM should be near the binding you observed in the presence of a maximal concentration of competitor. If the best-fit value of BOTTOM is negative, consider fixing it to a constant value equal to nonspecific binding determined in a control tube. |
| Has binding reached equilibrium? |
Competitive binding incubations take longer to equilibrate than saturation binding incubations. You should incubate for 4-5 times the half-life for radioligand dissociation. |
| Does only a small fraction of the radioligand bind? |
The equations are based on the assumption that the free concentration of labeled ligand is essentially identical to the concentration you added. Compare the total binding in the absence of competitor in cpm, to the amount of ligand added in cpm. If the ratio is greater than 10% at any concentration, then you've violated this assumption. Try to revise your experimental protocol, perhaps using a large incubation volume. |
| Does the curve have the expected steepness? |
The competitive binding curve has a Hill slope (or slope factor) of -1. If your data form a curve shallower than this, see Shallow competitive binding curves. |
Ki from EC50
Prism first fits the curve to find the EC50, the concentration of competitor that competes for half the specific binding. This is the same as the IC50.
The value of the EC50 is determined by three factors:
|
|
|
 |
• |
The affinity of the receptor for the competing drug. If the affinity is high, the EC50 will be low. The affinity is usually quantified as the equilibrium dissociation constant, Ki. The subscript i is used to indicate that the competitor inhibited radioligand binding. You can interpret it the same as you interpret a Kd. The Ki is the concentration of the competing ligand that will bind to half the binding sites at equilibrium, in the absence of radioligand or other competitors. If the Ki is low, the affinity of the receptor for the inhibitor is high. |
|
|
|
|
• |
The concentration of the radioligand. If you choose to use a higher concentration of radioligand, it will take a larger concentration of unlabeled drug to compete for half the radioligand binding sites. |
|
|
|
|
• |
The affinity of the radioligand for the receptor (Kd). It takes more unlabeled drug to compete for a tightly bound radioligand (low Kd) than for a loosely bound radioligand (high Kd). |
Prism calculates the Ki, using the equation of Cheng and Prusoff (Cheng Y., Prusoff W. H., Biochem. Pharmacol. 22: 3099-3108, 1973).
EC50 or log(EC50)?
The equation built-in to Prism is defined in terms of the log(EC50), so Prism finds the best-fit value of the log(EC50) along with its SE and 95% CI. Prism also reports the EC50 and its 95% CI. It does this by taking the antilog of the log(EC50) and of both ends of the 95% CI. Since the confidence interval is symmetrical on the log scale, it is not symmetrical when converted to EC50.
If the concentrations of unlabeled compound are equally spaced on a log scale, the uncertainty of the log(EC50) will be symmetrical, but the uncertainty of the EC50 will not be. That is why the equation is written in terms of log(EC50).
If you average together results from several experiments, it is better to average the log(Ki) values, rather than the Ki values. If you average Ki values, one value that is far from the rest will have too much influence on the mean. See Why Prism fits the logEC50 rather than the EC50.
|
