Transcript Lecture 05 - binding quant - Cal State LA

Quantifying the drug-target binding affinity
Receptors as targets
(Receptors are 45% of current drug targets)
•Receptors are areas of proteins found in cellular membranes or
nuclei
•They are classified by their function (superfamilies)
•Any natural (or unnatural) chemical that binds is a ligand
(Natural: neurotransmitters, hormones, growth factors, insulin,
some vitamins)
Observed responses: effect of addition of an agonist or antagonist
Dose-response or concentration-response curve: plot of
response (y axis) vs. -log[ligand](semilogarithmic plots):
Agonist
Antagonist
Observed responses: Types of antagonists
How do different antagonists affect the activity of a ligand (agonist)?
Ex.: ACh = Acetylcholine = ligand (agonist)
ACh+Competitive antagonist Y:
•Effect of Y is reversed by
addition of more ligand
ACh+Noncompetitive antagonist Y’:
•Effect of Y’is unaffected by [ligand]
Observed responses: Some definitions
•EC50 is the molar concentration that produces half the maximum
biological response observed. This is often used to estimate the
value of KD, but the correlation is not always reliable.
•ED50 is the dose administered that produces half the maximum
biological response observed.
•IC50 is the concentration at which an antagonist exerts its
half-maximal effect.
Theories of Drug-receptor interactions
Can they explain the observations?
(Receptors are 45% of current drug targets)
Occupancy theory: intensity of the pharmacological effect is directly
proportional to the number of receptors occupied by the drug
(fraction bound, or fMbound).
A = drug; M = receptor
AM
A+ M
fMbound =
Keq =
[AM]
[M]0
[A][M]
[AM]
= Kd
Occupancy Theory: Types of antagonists - one receptor
Unoccupied
(no activity)
Agonist
(full activity)
Receptor
Ligand (agonist)
binding site
Competitive
antagonist
(no activity)
Ligand (agonist)
binding site
May bind in the ligand binding site
OR nearby, partially blocking it
Ligand (agonist)
binding site
Noncompetitive
antagonist
(no activity)
Ligand (agonist)
binding site
May bind at allosteric site,
altering ligand binding site
Theories of Drug-receptor interactions
Additional known expressions (besides fMbound) can be used to
generate a different equation for fraction bound:
[AM]= Ka[A][M]
Ka
[M]0 = [AM] + [M]
mass balance
Use mass balance and K to generate the following:
[M]0= Ka[A][M] + [M] = [M](1+Ka[A])
Substitute this into the expression for fraction bound:
Ka[A]
[A]
Ka[A][M]
[AM] Ka[A][M]
=
=
=
=
fMbound =
[M]0
[M]0
[M](1+Ka[A]) (1+Ka[A])
(KD+[A])
Theories of Drug-receptor interactions
Plot fraction bound versus [A]: direct or saturation plot
Ka[A]
[A]
=
fMbound =
(1+Ka[A])
(KD+[A])
Theories of Drug-receptor interactions
Data may be displayed in a semi-logarithmic plot as well.
fMbound =
Ka[A]
(1+Ka[A])
=
[A]
(KD+[A])
Theories of Drug-receptor interactions
While the occupancy theory simulates actual dose-response curves,
(theoretically, KD = EC50), it does not account for agonists that do
not produce the maximum effect.
Modified occupancy theory: modified to separate the binding
affinity from the intrinsic activity () of the compound. That is, a
compound can bind tightly, but cause a little or no effect.
Theories of Drug-receptor interactions
The modified occupancy theory is also incomplete:
How can two molecules binding to one receptor cause different
effects? (one as an agonist; on as an antagonist)
How can desensitization be explained?
Rate theory: stimulus occurs only when the ligand first occupies
the receptor.
Induced fit theory: a specific conformational change in a receptor
is responsible for initiation of the biological response.
•Agonist binding induces this conformational change.
•Partial agonist induces a partial conformational change.
•Antagonist binding would either not cause a conformational
change OR it would cause a conformational change without
eliciting a biological response.
Determination of affinity constants K.
Keq =
A+ M
AM
[A]eq[M]eq
[AM]eq
= Kd
•Simply measure the equilibrium concentrations of each of the
species involved, (A, M, AM), and plug in the Kd expression.
•It is generally difficult to directly measure these concentrations.
•Many alternative methods have been developed to measure ligand
binding strength.
1. Determination of concentration of unbound [ligand] using phase
separation (equilibrium dialysis; ultrafiltration)
AM
A
M
A
A
A
AM
Left side (L)
A
A
Right side (R)
semipermeable membrane
Determination of affinity constants K.
2. Add one component to the other in successively larger
amounts, with the concurrent measurement of some observable
(titration). The observable (often spectral) change is from
either the ligand OR the macromolecule.
Observed data is fit to a binding model that incorporates the
spectroscopic observables. Best fit of data to model reveals K.
Determination of affinity constants K.
3. Competetive binding assays
a. Quantitative affinity chromatography.
b. Biosensor technology (BIAcore or IASys). Quantitation of
bound ligand by mass.
c. Radioligand binding assay.
Step1. Determine binding constant (Kd) of ligand
Step2. Determine binding constant of drugs by competition
Determination of K - Radioligand competition binding assays
Step 1: Determine binding constant Kd of ligand (radiolabeled) with
receptor
•Incubate different ratios of [radiolabeled ligand] and [receptor];
equilibrate; separate receptor-bound ligand from unbound ligand;
determine [ ]bound using scintillation counting.
•Correct for nonspecific binding of radioligand to other substances
Determination of K - Radioligand competition binding assays
Step 1(cont): Correction for nonspecific binding (example):
Determination of K - Radioligand competition binding assays
Step 1: Determination of radioligand KD: Rearrange our equations…
[AM] =
fMbound =
[M]0
[AM] =
Ka[A]
=
(1+Ka[A])
[M]0 Ka[A]
(1+Ka[A])
=
[A]
(KD+[A])
[M]0 [A]
(KD+[A])
•Plot corrected [AM] versus [A] (direct plot). Evidence of saturation
of receptor must be shown.
•Analyze data using nonlinear regression. (Minimizes the sum of
squares of deviations from fit and iteratively tries may solutions to
get the best fit of [M]0 and KD to the experimental data). Usually
linear transformations (Scatchard) of data are shown in publications,
even if nonlinear regression analysis was used.
Determination of K - Radioligand competition binding assays
Step 1: Determination of radioligand KD: plots (example)
Scatchard:
Determination of K - Radioligand competition binding assays
Step 2: Competition Studies: Determine KI (drug). Fix the
concentration of receptor and radioligand, while varying the
concentration of the test compound (drug).
•This results in a decrease of radioligand (L1) binding with
increasing [drug] (L2).
• the observed IC50 will depend on the amounts of radioligand and
drug as well as KD (ligand) and KI (drug).
Determination of K - Radioligand competition binding assays
•KI can be determined mathematically (Cheng and Prusoff, 1973).
KI (drug) = KL2
KD (ligand) = KL1
[Radioligand]
References for quantification of binding:
Ninfa A. J.; Ballou, D. P. Fundamental Laboratory Approaches for Biochemistry and Biotechnology;
Fitzgerald Science Press: Bethesda, MD, 1998.
Patrick, G. L. An Introduction to Medicinal Chemistry; Oxford University Press: New York, NY, 2001
Silverman, R. B. The Organic Chemistry of Drug Design and Drug Action ; Academic Press: San
Diego, CA, 1992.
Thomas, G. Medicinal Chemistry An Introduction; John Wiley and Sons, Ltd.: New York, NY, 2000.
Example problems
Biochemistry, 41 (32), 2002, 10262-10269.
Trp
F3Trp
Dose-response curves for the application of 5-hydroxytryptamine (5HT, a ligand) to nAchR
variants (receptor mutants). Each receptor variant has a binding site residue with 0, 1, 2, or
3 fluorines on a key tryptophan residue.
What is the EC50 of 5HT to each variant 1-4 (labeled curves from left to right)?
NH3
Which variant binds to 5HT most strongly?
5HT has the following structure. Can you explain the observed EC50’s?
HO
N
H
J Med Chem 45 (15) 2002 p. 3234
Nicotinic acetylcholine receptor agonists: 3b4 receptor
Using KI (table 1), rank agonists 1, 2, and 3 from best to worst for binding the 3b4
receptor.
Does the functional assay (right) reflect this ranking?
J. Med. Chem., 46 (7), 1153 -1164, 2003
All these compounds have been tested as influenza
endonuclease (enzyme) inhibitors.
•Compare the IC50’s of compounds 9-12. Which is the best
inhibitor?
•This data (for 9-12) reveals that a structural feature is
important for inhibition. What is it?
•The IC50 vs. pH graph for compound 8 is shown to the right.
What is the optimal pH for inhibition?
•All of these inhibitors are thought to bind to a metal ion in the
enzyme active site. How might the IC50 vs. pH graph support
this?