Comparison of binding and cellular profiles
of multi-target G protein-coupled receptors
B. Fouchaq, S.R Wang, D. Piwnica, L. Coussy, T. Jolas, G. Néliat - Cerep - Le Bois l'Evêque, 86600 Celle l'Evescault, France - www.cerep.com - [email protected]
Box 1
A+R
a
Partial agonist effect
EA
Antagonist effect
0
KA
Inverse agonist effect
=
Em
KA
a[A]
[A]+KA
AR
=
a[A.R]
[Rt]
EA ➞ fraction of maximal response
Em ➞ maximal response
A ➞ concentration of ligand
a ➞ efficacy
AR ➞ concentration of drug receptor complex
KA ➞ equilibrium dissociation constant
-1
For agonist or inverse agonist compounds, the efficacy
constant is not equal to 0 and direct comparison of binding
(Ki) and functional results(EC50/IC50) is not possible.
n
To control for the quality of our results in terms of homogeneity and reproducibility, we produce large batches of cells which express GPCR receptors.
From these batches prepared membranes and whole cells are frozen and stored for binding assays and functional assays respectively.
Two technologies were selected for use in this platform: HTRF® to determine cAMP concentration and real time fluorescence to monitor calcium
concentration both in living cells. Measuring these second messagers allows us to developed functional assays for receptors coupled to Gi, Gs, and Gq.
Cell culture in batches
s
Cells in
frozen vials
(1 vial/1 plate)
Compounds
fresh 10X cc
solution in
water
-8
-6
@10-4M
-5
-4
-3
REACTION
REACTION
Neca: 28.5 nM
s
[cAMP]
-7
s
0
s
20
s
(C) Fresh culture
% of control activity
80
60
40
20
80
60
40
30 min stimulation
& cAMP HTRF® kit
detection
Filtration & radioactivity
measurement
20
0
-7
-6
-5
-4
-3
Log [*Neca (p2)] (M)
[cAMP] @10-4M Neca: 49.5 nM
-8
-7
-6
-5
-4
-3
Log [*Neca (p1)] (M)
s
-8
s
% of control activity
s
Real-time fluorescence
detection using dual
run protocol
100
100
s
(B) Frozen cells @ H-24h
s
We acknowledge Mr Jacques Migeon for his
assistance, Mr Olivier Godet for IT part of this
work. We thank Mrs Catherine Moreau and Mr
Loïc Dorgeret for poster preparation.
Frozen
reference
compounds,
controls &
radioligands
40
0
Acknowledgements
Membrane
preparation
in frozen
plates
s
Hill , S. J. (2006), G-protein-coupled receptors:
past, present and future. British journal of
pharmacology,147 (S1): S27-37
60
Compounds
fresh 10X cc
solution in
HBSS
s
Kenakin, T. et al. (2003), Predicting therapeutic
value in the lead optimization phase of drug
discovery. Nature reviews, 2: 429-38.
80
Frozen
reference
compounds,
controls &
agonist
s
Ariën, E. J. et al. (1979), The receptor, a
comprehensive treatise. Edited by R. O’Brien, p
33-91. Plenum Press, NY
% of control activity
100
Bibliography
s
Neca was tested using cAMP
generic
protocol and compared in three conditions:
n A - Thawed cells at
(A) Frozen cells @ H-1h
37°C and directly
seeded in reaction
plate
n B - Thawed cells at
37°C and incubated
24h in a flask
Log [*Neca (p1)] (M)
n C - Fresh culture.
5
150
5-HT2A (h) agonist effect
5-HT2A (h) antagonist effect
5-HT2A (h) binding effect
120
90
60
30
0
-30
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
Compound number
Fig 4. Comparison of the binding and functional effect of 50 BioPrint® compounds in the D1 receptor (using cAMP HTRF® detection)
Comparison of protocols using fresh and
frozen cells for the functional study of the
human A2B receptor Second
Results analysis EC50 and PKBapp
Results analysis IC50 and Ki
[cAMP] @10-4M Neca: 35 nM
2
D – Comparison of binding and functional profiles
Fig 3. Comparison of the binding and functional effect of 50 BioPrint® compounds in the 5HT2A receptor (using calcium fluorometric detection)
B – Material and methods
HTRF®
2
In the figure 2, we observe a poor correlation between KI and EC50 of agonist references compounds tested on the same panel of assays. These results
15
confirm that the cellular response to a pharmacological agonists depends on the efficacy of the agonist (which is a property of each individual agonist) and
the sensitivity of the system (Kenakin, 2003).
10
Effect
(% of inhibition / % of stimulation)
Agonist effect
Log EC50 (functionnal assays)
For antagonist compounds, the efficacy constant (a in
Box 1) is very low and we can suppose that functional
effects only depend on the affinity component of the
equation (according the law of mass action at the origin of
occupancy theory). Using this approach, the comparison
between binding (affinity component) and functional
(affinity and efficacy component) can be performed by
a comparison of the Ki and the KB app of one compound
(only for competitive antagonists).
n
s
s
1
Effect
(% of inhibition / % of stimulation)
Description of the pharmacophoric properties of compounds
Log KB app (functionnal assays)
To analyze the correlation beFig 1.
Fig 2.
-10
-4
-3
-11
-9
-8
-7
-6
-5
-5
-4
-10
-9
-8
-7
-6
-11
-3
tween functional and binding
-5
Correlation
Correlation
-4
assays, Log KI & LogKB deter-6
between
between
-5
binding and
binding and
-7
mined for reference antagonist
-6
-7
functional
functional
-8
compounds were compared on
-8
assays for
assays for
-9
56 different GPCR targets. As
-9
antagonist
agonist
y = 0.982x + 0.4364
y = 0.5948x - 3.5127
-10
R = 0.7788
R = 0.2377
-10
reference
reference
shown in figure 1, we obtain a
-11
-11
compounds
compounds
very strong correlation between
log Ki (agonist binding assay)
log Ki (agonist binding assay)
KI and KB with an R2 near 0.78
for the linear fit. We confirm that application of occupancy theory (see part B) is a good model with 25
which to analyze antagonist binding and functional effects. Moreover, the functional assay protocols, using only the frozen cells, produces the same quality20of results as the corresponding binding assay.
s
Using generalized and uniform
conditions, Cerep is currently
developing a complete highthroughput functional profiling
GPCR platform adapted to monitor
Gs, Gi and Gq-coupled receptors.
To understand the correlation
between receptor binding and
intracellular signalling (Calcium
flux and cAMP production) we
have compared results obtained
in GPCR receptor radioligand
binding and functional assays.
Using Ki calculations, Radioligand Binding Assays (RBA) are powerful tools to determined the affinity of one compound for specific targets. While
binding affinity is important, the functional downstream effect is of greatest importance. A compound has the several potential functional behaviors,
such as agonist, antagonist, partial agonist, inverse agonist, etc. As is shown below, the functional effects of a compound can be described using
affinity and efficacy property components.
s
In designing leads, it is of
interest to consider that GPCRs
can couple to a large variety of
cellular effects. A new approach
based on functional profiling of
compounds can provide high
density information necessary to
optimize compounds during the
development of successful drugs.
C – Correlation between binding and functional assays
s
nIntroduction
A – Difference between agonist and antagonist
D1 (h) agonist effect
D1(h) antagonist effect
D1(h) binding effect
120
100
80
60
40
20
0
-20
We performed binding and functional
(agonist and antagonist mode) profiles
of compounds selected for their high hit
rates on GPCR targets. In figures 3 & 4,
we show the profiles of 50 compounds
on the 5HT2A and D1 receptors. The
typical effect of agonist or antagonist
compounds are shown in blue and
yellow circles respectively.
Remark: Profiles of agonist compounds
differ between detection technologies. In
fact, using calcium fluorimetry, desensitization of the calcium system (Gq coupling
receptors) induces a false antagonist effect.
Conversely, for Gs coupled receptors,
overproduction of cAMP induced by agonist
induces negative inhibition.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
-40
-60
Compound number
nConclusion
To understand the correlation between receptor binding data and intracellular signalling data (Calcium flux and cAMP production)
we have compared results obtained in GPCR receptor binding and functional assays using reference compounds. For the antagonist
compounds, we obtained a good correlation between binding and functional effects due to the single affinity component of functional
effect. For the agonist compounds, both efficacy and affinity drive the functional effects of the compound. In fact, the affinity, an intrinsic
property of the molecule, can easily be determined by binding assays, but the efficacy depends on the environment of the receptor,
expression level of receptors & coupling proteins and the method of detection. New advances in pharmacology have introduced new
descriptions of molecules. Inverse agonists, allosteric modulators and proteans can be described in many systems (Hill, 2006). In the
case of the work presented here, we use a simple approach to functionality. Inverse agonists appear as antagonists and we are unable
to detect allosteric potentiators. Nevertheless profiling molecules using this new approach provides information about their abilities to
simulate or to block a signal via a receptor. During the lead optimization phase, functional information combined with binding results
now opens a new dimension in pharmalogical profiling which allows us to more efficiently approach on and off target biological effects.