- Email: [email protected]
Differentiation between two ligands for peripheral benzodiazepine binding sites, [3H]R05-4864 and [3H]PK 11195, by thermodynamic studies
Pergamon Pres!
Life Sciences, Vol. 33, pp. 449-457 Printed in the U.S.A.
DIFFERENTIATION BETWEEN Two LIGAN S FOR PERIPHERAL !I BENZO!jIAZEPINEBINDING SITES, [ H]R05-4864AND [ H]PK 11195, BY THERMODYNAMIC STUDIES G. Le Fur, N. Vaucher, M.L. Perrier, A. Flamier,
J. Benavides,C. Renault, M.C. Dubroeucq, C. Cud&my and A. Uzan PHARMUKA Laboratoires,35, quai du Moulin de Cage, 92231 Gennevilliers,France (Received in final form llay 16, 1983) SUMMARY
The L~H] PK 11195, I-(2-chlorophenyl)-N-methyl-N-(l-methylpropyl)-3-isoquinolinecarboxamide, binding sites in rat cardiac membranes are saturable,with high affinity, specific GABA-independent and correspond to the peripheral type of benzodiazepine. The order of potency of displacing agents was : PK 11195 > R05-4864 > dipyridyole > diazepam b cloobtained with $ H]PK 11195 was equivnazepam. alent of tlEBBmax obtained with [ H]RO5-4864 in the same s However thermodynamic analysis experimental c&%ti H]PK 11195 binding was entropy driven indicates that 5he ["r. whereas the [ H] R05-4864 binding was enthalpy driven. Consequently PK 11195 might be an antagonist of these binding sites and ~05-4864 an agonist or a partial agonist. The simultaneous use of both drugs might help to elucidate the physiological relevance of peripheral benzodiazepine binding sites.
Shortly aftgr the descriptionof brain benzodiazepinereceptors, binding sites for [ Hldiazepam have been described in peripheral tissues such as kidney, lungs or liver (1) and further in platelets (21, mast cells (3), heart (4) and brain (5). These peripheralbinding sites were distinguised from the brain type by the opposite affinities of pharmacological compounds such as clonazepam, specific for the brain type, and R05-4864, specific for the peripheral type. Recently we have shown that a new compound, chemically unrelated to the benzodiazepines,PK 11195 or l-(2chlorophenyl)-N-methyl-N-(l-methylpropyl)-~-isoquinolinecarboxamide (fig. l), was potently active 9 the "peripheraltype" of benzodiazepine binding sites as labelled by [ H]R05-4864 in vitro (6) an4 in vivo (7). Now we have tritiated PK 11195 in order to compare it to [ H]R05-4864 on two parameters : firstly the number of binding sites in rat heart membranes and secondly the temperature effect on their bindin? parameters for [ HIPK 11195 Our results revealed t t the B l!? :a %%&3.6~ were similar and that [ H]PK lll%%inding was entropy driven whereas [ H]R05-4864was enthalpy driven. 0024-3205183$3.00 + .OO Copyright (c) 1983 Pergamon Press Ltd.
450
BenzodiazepineBinding Thermodynamics
O-N’
Vol. 33, No. 5, 1983
w ‘TH-CH, FIG. 1
CH,-CH,
Chemical structure of PK 11195
MATERIAL AND METHODS Chemical and reagents 13H]R05-4864 (73.8 Ci/mmol) was purchased from NEN. L3H]PK 11195 was prepared by catalytic tritiationof 1-(2-chlorophenyl)-N-methyl-N-(l-methyl2-propenylj-3-isoquinolinecarboxamide. RO5-4864 was generously supplied by Hoffmann-La Roche Inc., Nutley, N.J. All other chemicals used were reagent grade. Membrane preparation Membranes were prepared according to Williams et al (8). Briefly male Sprague-Dawley rats (200 g) from Charles River France were killed and the ventricular tissue was homogenized cl:4 w/v) in cold sucrose(0.25M), Tris HCl(5 m& MgCl (1 mM) buffer pH 7.4. The homogenateswere filtered through a double laye& of cheese cloth then centrifuged at 1,000 g for 10 minutes. The supernatant was recentrifugedat 40,000 g for 30 minutes. The resulting pellet was resuspendedin the incubationbuffer. Binding assays The assays were performed in 50 mM Tris HCl pH 7.5 MgCl 10 mM buffer in a final volume of 1 ml containing 0.2 mg of cardiac membr2 ne prot3in and the radioactive ligands. After y minutes at 25-C in the case of [ H]PK 11195 or 45 minutes at O'C with [ ~1~05-4864, the membranes were filtered over GF/C Whatrnanfilters followed by 3 x 5 ml washes with cold buffer. Specific binding (95 $ total binding for both ligands) was defined as the amount of r dioactivity displaced by 1 uM unlabelled ~05-4864. To determine whether 3 [ H]PK 11195 was degraded after binding to heart membranes, a binding assay was carried out and the radioactivity in the filters was extracted with ethanol. The extracts were chromatographedon thin layer plates. More than 95 % of the bound activity had R values of 0.5 [ cyclohexane-ethylacetate (l:l)] and 0.27 [petryleum &her (60'-80'), diethylether (2:8)] identical to that of authentic [ H]PK 11195 Determinationof thermodynamicparameters Equilibrium thermodynamic parameters of binding were determined utilizing classical thermodynamic equations. The standard Gibbs free energy change (A G') of associationwas calculated from the equation AG' q-RTLnKa where R is the gas constant (1.99 cal/mol- deg), T is the temperature in degrees Kelvin and Ka is the equilibriumassociation constant (l/K ). The standard enthalpy change ( A H') was calculated from Van't Hoff pldts of the depen-
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
451
dence of Ka on temperaturebetween 4-C and 37'C. The slope of a Van't Hoff plot (LnKa vs l/T) equals - AH-/R. The standard entropy change ( AS‘) was calculated from the equation AG' F A H' - TAS' after first determiningAG' and AH' as described above. RESULTS
Specific binding of [3H]PK 11195 was linearly dependejlt on protein concentration up to 0.2 mg per assay. The binding of [ H]PK 11195 was time dependent and equilibriumwas achieved after 10 minutes (fig. 2). Binding was completely reversibleby 1 uM RO5-4864. Kinetiz ~~;~~","M_c,om;;;~f,rz . association-disgociation experiments were K K = 0.492 nM (means of 3 experiments at !?5'C) and the equilibrium di&ociation constant ( 1 obtained from kinetic constants WY 0.81 nM. When tissue was incubat with increasing concentrationsof [ H]PK 11195 the specifically bound radioactivity was saturable while non specific binding increased linearly (fig. 3). Scatchard analysis of these data is consistentwith the presence of a single populalionof binding sites with an apparent dissociation constant (K ) of 1.27 - 0.11 nM (n 3) which is fairly close to the value obta&d from +the kinetic constant quotient. ) was 2250 - 325 flnoles/mgprotein. Hill Maximum binding capacity (B plot of the data resulted q$xa single straight line with a Hill coefficient of 1 - 0.1 at all the temperatures studied which is indicative of apparent binding site homogeneity and non cooperativity. Analysis of [ H]R05-4864 bindiJg to the sale membrane @reparation yielded the following : K = 5.4 - 1.3 nM an<; - 271 fmoles/mg protein with #ill coefficient of 1 @Y(,a,'%! the temperaturesstudied. (fig. 'lj, Tps [ H]PK 11195 appeared to label the same number of binding sites as [ H]RO5-4864 in the rat heart membranes. Scatchard analysis of the saturation studies over a temperature range ('I-37-C)showed no change in B for R05-4864 and PK 11195. for PK 11195 was also unaffectedby tempe%z -4864 increases pith temperature.Van'tHoff 11195 and [ H]RO5-4864 at several temperatures are s wn in fig. 5. The plot of r3H]R05-4864was linear over the temperature range studied indicatin that the enthalpy change was independent of temperature. The plot of H]PK 11195 was parallel to the abscissa showing that the binding reaction was not nthalpy driven. In other words the affinity of heart membranes for f H]PK 11195 was not affected by changes in temperature. q
Even if the changy in Gibbs free energy between both ligands were similar, t$e binding of [ H]PK 11195 was entropy driven whereas the binding of [ HlR05-4864was enthalpy driven (table I). The pharmacological specificity of the [3H]PK 11195 binding sites was studied using various drugs. As expected, among the benzodiazepinesthe following order of potency was found : RO5-4864 > diazepam > clonazepam (table II). Dipyridamolewhich has been described as a ligand of peripheral binding sites (9b interacted at rather low concentrations (IC at 25'C) with [ HIPK 11195 binding sites. The selective’ 3 if 3f9H;M_ PK 11195 binding to cardiac membranes was demonstratedby the inability of 5-HT, NA, DA, histamine, glycine, GABA, glutamate at 100 uM, propranolol, prazosin, yohimbine, clonidine, haloperidol, methergoline, p?methazine, cimetidine, atropine, morphine, imipramine at 10 u M to affect [ H]PK 11195 binding (data got shown). R05-4864 and diazepam were 2.5 to 4 times less effective on [ HIPK 11195 binding at 25'C than at 4'C whereas PK 11195 and dipyridamolehad the same affinity of both temperatures (table II) which is in agreementwith the thermodynamicstudy.
452
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
TIME
50
(m’n)
FIG. 2 Time course of [3H]PK 11195 binding to rat heart membranes. Association of 1 nM [ H] PK 11195 was measured under the s ndard assay conditions, 11195 (A.). At the as a function of the time from the addition of time indicated by the arrow (15 minuys) 1 uM of R05-4864 was add? to prevent the binding of dissociated [ H]PK 11195 ( A 1. Specific [ HIPK 11195 binding was assayed as described in Methods. Each point is the mean of three determinationswhose standard errors were less than 5 %.
I 2ooc
>_
n
I
n Kd = 1.41 1500
Bm,2250
nM fmol/mg
prot*in
\ bound 5
10
T_~ --15
J
zoo’
(3HPK
11195
( nM
FIG. 3 Specific binding of r3H]P5 11195 to rat cardiac membranes as a function The inset shows a Scatchard plot of the concentrationof [ H]PK 11195. of the binding data. Results are the mean of 4 experiments.
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
1000
453
bound
FIG. 4 Specific binding of [3~]Rg5-4864to rat cardiac membranes as a function of the concentrationof [ H]R05-4864. The inset shows a Scatchard plot of the binding data. Results are the mean of 4 experiments.
FIG. 5 Van't Hoff plot of K of [3H]PK 11195 (A) and [3~]RG5-4864(+) binding on temperature (T) u&ing rat cardiac membranes. KA values were calculated from the equilibrium dissociation constants (K = l/K ) determined by Scatchard analysis. Each point represents the kean 2 8.e.m. of 3 to 5 determinations. The s.e.m. of points with no error bans indicated were less than the size of the symbols. The density of binding sites did not vary with temperature.
- 10.63
- 10.45
- 10.45
- 10.74
4-c
37'C
0
0
- 12.41
- 11.16
H'
(Kcal ml-')
A
37'C
4-c
(Kcal mol-')
a G’
40.3
+
+
0.60
1
+ 40
+
(Entropy units)
A S'
Thermodynamicparameterswere calculated from VanIt Hoff plots (fig. 5) as described in (10, 11). ffi'=- RTLnKa ; AS' = AH' - AGO/T and AH'/RT is the slope of the Van't Hoff plot.
R05-4864
PK 11195
LIGANDS
THERMODYNAMICPARAMETERSOF LIGAND BINDING TO THE "PERIPHERAL"BENZODIAZEPINEBINDING SITES OF RAT CARDIAC MEMBRANES
Table I
50
12.8 [11.3 - 14.71
48.3 [37.4 - 62.31
1384 - 4691
> 1000
> 1000
425
1145 - 2991
[231 - 4591
326
1286 - 9321
208
i4.6 - 20.31
r9.2 - 17.31
517
9.6
4-c
12.6
25-C
Results are expressed as IC in nM and 95 confidence limits. PK 11195 concentration was 1 nM. Each val% IC is the mean of three membrane preparations.
CLONAZEPAM
DIPYRIDMLE
DIAZEPAM
m5-4864
PK 11195
DRUGS
EFFECT OF PK 11195, DjPYRIDAMOLEAND SEVERAL BENZODIAZEPINESON THE IN VITRO BI@JDIM3 OF I. HIPK 11195 TO CARDIAC MEMBRANES AT 4' AND 25'C
Table II
456
Benzodiazepine Binding Thermodynamics
Vol. 33, No. 5, 1983
DISCUSSION The PK 11195 binding sites in rat heart are specific, saturable, with high
affinity GABA-independent and correspond to the peripheral type of benzodiazepines, based on the relative potencies of RO5-4864, diazepam, clonazepam and dipyridamole. Moreover Bmax is equivalent for both ligands. The binding of L3,1 PK 11195 to rat cardiac membranes is associated to little or no chaye in enthalpy and substantial increases in entropy whereas the binding of [ H]R05-4864 is enthalpy driven. These results comply with a two step model as proposed from p receptors (10,111 1
L+Rz
2 LR
-
LR'
where L is the ligand, A the receptor and R' the agonist induced state of the receptor protein. In this model antagonists would participate only in reaction 1 with little or no change in enthalpy and large increases in entropy. Agonists could participate to both reactions 1 and 2 by inducing a conformation change of the receptor to R'. This reaction 2 would have negative enthalpy. Thus the binding of agonists is enthalpy driven with marked net decreases in entropy whereas the binding of antagonists seems to be almost exclusively entropy driven. Such a model has been successfully applied to the brain type of benzodiazepinereceptors (12,131. At 37-C the clonazepam is associated with a large decrease in enthalpy ~?&",~~pe~!!!&ed for an un3f avourable decrease in entropy whereas the binding of the antagonist [ H]ROl5-1788 was associated with a positive entropy change analogy with these results, in the case of the peripheral benwdiazepine binding sites, PK 11195 might be an antagonist but R05-4864 an agonist or a partial agonist. In the case of the B receptors for instance some antagonists have the same thermodynamicparameters (AG', AH', AS*) as PK 11195 and some partial agonists similar thermodynamic parameters as While it is evident that the peripheral benzodiazepine R05-4864 (lo). binding sites might be labelled by a benzodiazepinederivative ~05-4864 and by a chemically unrelated to benzodiazepine compound (isoquinoline) i.e. PK 11195, their functional role or physiological significance remains unkown. Moreover PK 11195 seems to label a subclass or a specific conformation of these binding sites. If the analogy with the thermodynamic analysis between ligands and receptors applied for instance top receptors (10,ll) or to the brain type of benzodiazepinereceptors (l2,l3) is true in the case of the peripheral benzodiazepinebinding sites, PK 11195 might be an antagonist and R05-4864 an agonist or a partial agonist of a hypothetical receptor coupled to these binding sites. Thus the simultaneoususe of both compounds might help to elucidate if peripheral benzodiazepine binding sites are receptors or acceptors (14). ACKNOWLEDGEMENTS The authors wish to thank Hoffmann-La Roche for their generous supply of R05-4864 and Mrs. E. Berthier for her secretarialassistance. REFERENCES 1. 2.
C. BRAESTRUP and R.F. SQUIRES, Proc. Natl. Acad. Sci. USA 74 3805-3809 (1977) J.K. WANG, T. TANIGUCHI and S. SPECTOR, Life Sci. 27 1881-1888 (1980) T. TANIGUCHI, J.K. WANG and S. SPECTOR, Life Sci. 27 171-178 (1980) T. TANIGUCHI, J.K. WANG and S. SPECTOR, Biochem. Pharmacol. 31 589-590 (1982)
Vol. 33, No. 5, 1983
5. 6.
BenzodiazepineBinding Thermodynamics
457
H. SCHOEMAKER,M. BLISS and H.I. YAMAMURA, Europ. J. Pharmacol. 71 173175 (1981)
G. LE FUR, M.L. PERRIER, N. VAUCHER, F. IMBAULT, A. FLAMIER, A. UZAN, C. RENAULT, M.C. DUBROEUCQ and C. GUEREMY, Life Sci. 32 1839-1847 (1983) 7. G. LE FUR, F. GUILLOUX, P. RUFAT, A. UZAN, C. RENAULT, M.C. DUBROEUCQ and C. GUEREMY, Life Sci. 32 1849-1856 (1983) 8. R.S. WILLIAMS and R.J. LEFKOWITZ, Circ. Res. 43 721-727 (1978) L.P. DAVIES and U. HUSTON, Europ. J. Pharmacol. 73 209-211 (1981) 9. 10. G.A. WEILAND, K.P. MINNEMAN and P.B. MOLINOFF, Nature 281 114-117 (1979) 11. G.A. WEILAND, K.P. MINNEMAN and P.B. MOLINOFF, Mol. Pharmacol. 18 341-347 (1980) 12. R.C...SPETH, G.J. WASTEK and H.I. YAMAMURA, Life Sci. 24 351-358 (1979) 13. H. MOHLER and J.G. RICHARDS, Nature 294 763-765 (1981) 14. J.G. RICHARDS, H. MOHLER and W. HAEFELY, T.I.P.S. 3 233-235 (1982)
Life Sciences, Vol. 33, pp. 449-457 Printed in the U.S.A.
DIFFERENTIATION BETWEEN Two LIGAN S FOR PERIPHERAL !I BENZO!jIAZEPINEBINDING SITES, [ H]R05-4864AND [ H]PK 11195, BY THERMODYNAMIC STUDIES G. Le Fur, N. Vaucher, M.L. Perrier, A. Flamier,
J. Benavides,C. Renault, M.C. Dubroeucq, C. Cud&my and A. Uzan PHARMUKA Laboratoires,35, quai du Moulin de Cage, 92231 Gennevilliers,France (Received in final form llay 16, 1983) SUMMARY
The L~H] PK 11195, I-(2-chlorophenyl)-N-methyl-N-(l-methylpropyl)-3-isoquinolinecarboxamide, binding sites in rat cardiac membranes are saturable,with high affinity, specific GABA-independent and correspond to the peripheral type of benzodiazepine. The order of potency of displacing agents was : PK 11195 > R05-4864 > dipyridyole > diazepam b cloobtained with $ H]PK 11195 was equivnazepam. alent of tlEBBmax obtained with [ H]RO5-4864 in the same s However thermodynamic analysis experimental c&%ti H]PK 11195 binding was entropy driven indicates that 5he ["r. whereas the [ H] R05-4864 binding was enthalpy driven. Consequently PK 11195 might be an antagonist of these binding sites and ~05-4864 an agonist or a partial agonist. The simultaneous use of both drugs might help to elucidate the physiological relevance of peripheral benzodiazepine binding sites.
Shortly aftgr the descriptionof brain benzodiazepinereceptors, binding sites for [ Hldiazepam have been described in peripheral tissues such as kidney, lungs or liver (1) and further in platelets (21, mast cells (3), heart (4) and brain (5). These peripheralbinding sites were distinguised from the brain type by the opposite affinities of pharmacological compounds such as clonazepam, specific for the brain type, and R05-4864, specific for the peripheral type. Recently we have shown that a new compound, chemically unrelated to the benzodiazepines,PK 11195 or l-(2chlorophenyl)-N-methyl-N-(l-methylpropyl)-~-isoquinolinecarboxamide (fig. l), was potently active 9 the "peripheraltype" of benzodiazepine binding sites as labelled by [ H]R05-4864 in vitro (6) an4 in vivo (7). Now we have tritiated PK 11195 in order to compare it to [ H]R05-4864 on two parameters : firstly the number of binding sites in rat heart membranes and secondly the temperature effect on their bindin? parameters for [ HIPK 11195 Our results revealed t t the B l!? :a %%&3.6~ were similar and that [ H]PK lll%%inding was entropy driven whereas [ H]R05-4864was enthalpy driven. 0024-3205183$3.00 + .OO Copyright (c) 1983 Pergamon Press Ltd.
450
BenzodiazepineBinding Thermodynamics
O-N’
Vol. 33, No. 5, 1983
w ‘TH-CH, FIG. 1
CH,-CH,
Chemical structure of PK 11195
MATERIAL AND METHODS Chemical and reagents 13H]R05-4864 (73.8 Ci/mmol) was purchased from NEN. L3H]PK 11195 was prepared by catalytic tritiationof 1-(2-chlorophenyl)-N-methyl-N-(l-methyl2-propenylj-3-isoquinolinecarboxamide. RO5-4864 was generously supplied by Hoffmann-La Roche Inc., Nutley, N.J. All other chemicals used were reagent grade. Membrane preparation Membranes were prepared according to Williams et al (8). Briefly male Sprague-Dawley rats (200 g) from Charles River France were killed and the ventricular tissue was homogenized cl:4 w/v) in cold sucrose(0.25M), Tris HCl(5 m& MgCl (1 mM) buffer pH 7.4. The homogenateswere filtered through a double laye& of cheese cloth then centrifuged at 1,000 g for 10 minutes. The supernatant was recentrifugedat 40,000 g for 30 minutes. The resulting pellet was resuspendedin the incubationbuffer. Binding assays The assays were performed in 50 mM Tris HCl pH 7.5 MgCl 10 mM buffer in a final volume of 1 ml containing 0.2 mg of cardiac membr2 ne prot3in and the radioactive ligands. After y minutes at 25-C in the case of [ H]PK 11195 or 45 minutes at O'C with [ ~1~05-4864, the membranes were filtered over GF/C Whatrnanfilters followed by 3 x 5 ml washes with cold buffer. Specific binding (95 $ total binding for both ligands) was defined as the amount of r dioactivity displaced by 1 uM unlabelled ~05-4864. To determine whether 3 [ H]PK 11195 was degraded after binding to heart membranes, a binding assay was carried out and the radioactivity in the filters was extracted with ethanol. The extracts were chromatographedon thin layer plates. More than 95 % of the bound activity had R values of 0.5 [ cyclohexane-ethylacetate (l:l)] and 0.27 [petryleum &her (60'-80'), diethylether (2:8)] identical to that of authentic [ H]PK 11195 Determinationof thermodynamicparameters Equilibrium thermodynamic parameters of binding were determined utilizing classical thermodynamic equations. The standard Gibbs free energy change (A G') of associationwas calculated from the equation AG' q-RTLnKa where R is the gas constant (1.99 cal/mol- deg), T is the temperature in degrees Kelvin and Ka is the equilibriumassociation constant (l/K ). The standard enthalpy change ( A H') was calculated from Van't Hoff pldts of the depen-
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
451
dence of Ka on temperaturebetween 4-C and 37'C. The slope of a Van't Hoff plot (LnKa vs l/T) equals - AH-/R. The standard entropy change ( AS‘) was calculated from the equation AG' F A H' - TAS' after first determiningAG' and AH' as described above. RESULTS
Specific binding of [3H]PK 11195 was linearly dependejlt on protein concentration up to 0.2 mg per assay. The binding of [ H]PK 11195 was time dependent and equilibriumwas achieved after 10 minutes (fig. 2). Binding was completely reversibleby 1 uM RO5-4864. Kinetiz ~~;~~","M_c,om;;;~f,rz . association-disgociation experiments were K K = 0.492 nM (means of 3 experiments at !?5'C) and the equilibrium di&ociation constant ( 1 obtained from kinetic constants WY 0.81 nM. When tissue was incubat with increasing concentrationsof [ H]PK 11195 the specifically bound radioactivity was saturable while non specific binding increased linearly (fig. 3). Scatchard analysis of these data is consistentwith the presence of a single populalionof binding sites with an apparent dissociation constant (K ) of 1.27 - 0.11 nM (n 3) which is fairly close to the value obta&d from +the kinetic constant quotient. ) was 2250 - 325 flnoles/mgprotein. Hill Maximum binding capacity (B plot of the data resulted q$xa single straight line with a Hill coefficient of 1 - 0.1 at all the temperatures studied which is indicative of apparent binding site homogeneity and non cooperativity. Analysis of [ H]R05-4864 bindiJg to the sale membrane @reparation yielded the following : K = 5.4 - 1.3 nM an<; - 271 fmoles/mg protein with #ill coefficient of 1 @Y(,a,'%! the temperaturesstudied. (fig. 'lj, Tps [ H]PK 11195 appeared to label the same number of binding sites as [ H]RO5-4864 in the rat heart membranes. Scatchard analysis of the saturation studies over a temperature range ('I-37-C)showed no change in B for R05-4864 and PK 11195. for PK 11195 was also unaffectedby tempe%z -4864 increases pith temperature.Van'tHoff 11195 and [ H]RO5-4864 at several temperatures are s wn in fig. 5. The plot of r3H]R05-4864was linear over the temperature range studied indicatin that the enthalpy change was independent of temperature. The plot of H]PK 11195 was parallel to the abscissa showing that the binding reaction was not nthalpy driven. In other words the affinity of heart membranes for f H]PK 11195 was not affected by changes in temperature. q
Even if the changy in Gibbs free energy between both ligands were similar, t$e binding of [ H]PK 11195 was entropy driven whereas the binding of [ HlR05-4864was enthalpy driven (table I). The pharmacological specificity of the [3H]PK 11195 binding sites was studied using various drugs. As expected, among the benzodiazepinesthe following order of potency was found : RO5-4864 > diazepam > clonazepam (table II). Dipyridamolewhich has been described as a ligand of peripheral binding sites (9b interacted at rather low concentrations (IC at 25'C) with [ HIPK 11195 binding sites. The selective’ 3 if 3f9H;M_ PK 11195 binding to cardiac membranes was demonstratedby the inability of 5-HT, NA, DA, histamine, glycine, GABA, glutamate at 100 uM, propranolol, prazosin, yohimbine, clonidine, haloperidol, methergoline, p?methazine, cimetidine, atropine, morphine, imipramine at 10 u M to affect [ H]PK 11195 binding (data got shown). R05-4864 and diazepam were 2.5 to 4 times less effective on [ HIPK 11195 binding at 25'C than at 4'C whereas PK 11195 and dipyridamolehad the same affinity of both temperatures (table II) which is in agreementwith the thermodynamicstudy.
452
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
TIME
50
(m’n)
FIG. 2 Time course of [3H]PK 11195 binding to rat heart membranes. Association of 1 nM [ H] PK 11195 was measured under the s ndard assay conditions, 11195 (A.). At the as a function of the time from the addition of time indicated by the arrow (15 minuys) 1 uM of R05-4864 was add? to prevent the binding of dissociated [ H]PK 11195 ( A 1. Specific [ HIPK 11195 binding was assayed as described in Methods. Each point is the mean of three determinationswhose standard errors were less than 5 %.
I 2ooc
>_
n
I
n Kd = 1.41 1500
Bm,2250
nM fmol/mg
prot*in
\ bound 5
10
T_~ --15
J
zoo’
(3HPK
11195
( nM
FIG. 3 Specific binding of r3H]P5 11195 to rat cardiac membranes as a function The inset shows a Scatchard plot of the concentrationof [ H]PK 11195. of the binding data. Results are the mean of 4 experiments.
Vol. 33, No. 5, 1983
Benzodiazepine Binding Thermodynamics
1000
453
bound
FIG. 4 Specific binding of [3~]Rg5-4864to rat cardiac membranes as a function of the concentrationof [ H]R05-4864. The inset shows a Scatchard plot of the binding data. Results are the mean of 4 experiments.
FIG. 5 Van't Hoff plot of K of [3H]PK 11195 (A) and [3~]RG5-4864(+) binding on temperature (T) u&ing rat cardiac membranes. KA values were calculated from the equilibrium dissociation constants (K = l/K ) determined by Scatchard analysis. Each point represents the kean 2 8.e.m. of 3 to 5 determinations. The s.e.m. of points with no error bans indicated were less than the size of the symbols. The density of binding sites did not vary with temperature.
- 10.63
- 10.45
- 10.45
- 10.74
4-c
37'C
0
0
- 12.41
- 11.16
H'
(Kcal ml-')
A
37'C
4-c
(Kcal mol-')
a G’
40.3
+
+
0.60
1
+ 40
+
(Entropy units)
A S'
Thermodynamicparameterswere calculated from VanIt Hoff plots (fig. 5) as described in (10, 11). ffi'=- RTLnKa ; AS' = AH' - AGO/T and AH'/RT is the slope of the Van't Hoff plot.
R05-4864
PK 11195
LIGANDS
THERMODYNAMICPARAMETERSOF LIGAND BINDING TO THE "PERIPHERAL"BENZODIAZEPINEBINDING SITES OF RAT CARDIAC MEMBRANES
Table I
50
12.8 [11.3 - 14.71
48.3 [37.4 - 62.31
1384 - 4691
> 1000
> 1000
425
1145 - 2991
[231 - 4591
326
1286 - 9321
208
i4.6 - 20.31
r9.2 - 17.31
517
9.6
4-c
12.6
25-C
Results are expressed as IC in nM and 95 confidence limits. PK 11195 concentration was 1 nM. Each val% IC is the mean of three membrane preparations.
CLONAZEPAM
DIPYRIDMLE
DIAZEPAM
m5-4864
PK 11195
DRUGS
EFFECT OF PK 11195, DjPYRIDAMOLEAND SEVERAL BENZODIAZEPINESON THE IN VITRO BI@JDIM3 OF I. HIPK 11195 TO CARDIAC MEMBRANES AT 4' AND 25'C
Table II
456
Benzodiazepine Binding Thermodynamics
Vol. 33, No. 5, 1983
DISCUSSION The PK 11195 binding sites in rat heart are specific, saturable, with high
affinity GABA-independent and correspond to the peripheral type of benzodiazepines, based on the relative potencies of RO5-4864, diazepam, clonazepam and dipyridamole. Moreover Bmax is equivalent for both ligands. The binding of L3,1 PK 11195 to rat cardiac membranes is associated to little or no chaye in enthalpy and substantial increases in entropy whereas the binding of [ H]R05-4864 is enthalpy driven. These results comply with a two step model as proposed from p receptors (10,111 1
L+Rz
2 LR
-
LR'
where L is the ligand, A the receptor and R' the agonist induced state of the receptor protein. In this model antagonists would participate only in reaction 1 with little or no change in enthalpy and large increases in entropy. Agonists could participate to both reactions 1 and 2 by inducing a conformation change of the receptor to R'. This reaction 2 would have negative enthalpy. Thus the binding of agonists is enthalpy driven with marked net decreases in entropy whereas the binding of antagonists seems to be almost exclusively entropy driven. Such a model has been successfully applied to the brain type of benzodiazepinereceptors (12,131. At 37-C the clonazepam is associated with a large decrease in enthalpy ~?&",~~pe~!!!&ed for an un3f avourable decrease in entropy whereas the binding of the antagonist [ H]ROl5-1788 was associated with a positive entropy change analogy with these results, in the case of the peripheral benwdiazepine binding sites, PK 11195 might be an antagonist but R05-4864 an agonist or a partial agonist. In the case of the B receptors for instance some antagonists have the same thermodynamicparameters (AG', AH', AS*) as PK 11195 and some partial agonists similar thermodynamic parameters as While it is evident that the peripheral benzodiazepine R05-4864 (lo). binding sites might be labelled by a benzodiazepinederivative ~05-4864 and by a chemically unrelated to benzodiazepine compound (isoquinoline) i.e. PK 11195, their functional role or physiological significance remains unkown. Moreover PK 11195 seems to label a subclass or a specific conformation of these binding sites. If the analogy with the thermodynamic analysis between ligands and receptors applied for instance top receptors (10,ll) or to the brain type of benzodiazepinereceptors (l2,l3) is true in the case of the peripheral benzodiazepinebinding sites, PK 11195 might be an antagonist and R05-4864 an agonist or a partial agonist of a hypothetical receptor coupled to these binding sites. Thus the simultaneoususe of both compounds might help to elucidate if peripheral benzodiazepine binding sites are receptors or acceptors (14). ACKNOWLEDGEMENTS The authors wish to thank Hoffmann-La Roche for their generous supply of R05-4864 and Mrs. E. Berthier for her secretarialassistance. REFERENCES 1. 2.
C. BRAESTRUP and R.F. SQUIRES, Proc. Natl. Acad. Sci. USA 74 3805-3809 (1977) J.K. WANG, T. TANIGUCHI and S. SPECTOR, Life Sci. 27 1881-1888 (1980) T. TANIGUCHI, J.K. WANG and S. SPECTOR, Life Sci. 27 171-178 (1980) T. TANIGUCHI, J.K. WANG and S. SPECTOR, Biochem. Pharmacol. 31 589-590 (1982)
Vol. 33, No. 5, 1983
5. 6.
BenzodiazepineBinding Thermodynamics
457
H. SCHOEMAKER,M. BLISS and H.I. YAMAMURA, Europ. J. Pharmacol. 71 173175 (1981)
G. LE FUR, M.L. PERRIER, N. VAUCHER, F. IMBAULT, A. FLAMIER, A. UZAN, C. RENAULT, M.C. DUBROEUCQ and C. GUEREMY, Life Sci. 32 1839-1847 (1983) 7. G. LE FUR, F. GUILLOUX, P. RUFAT, A. UZAN, C. RENAULT, M.C. DUBROEUCQ and C. GUEREMY, Life Sci. 32 1849-1856 (1983) 8. R.S. WILLIAMS and R.J. LEFKOWITZ, Circ. Res. 43 721-727 (1978) L.P. DAVIES and U. HUSTON, Europ. J. Pharmacol. 73 209-211 (1981) 9. 10. G.A. WEILAND, K.P. MINNEMAN and P.B. MOLINOFF, Nature 281 114-117 (1979) 11. G.A. WEILAND, K.P. MINNEMAN and P.B. MOLINOFF, Mol. Pharmacol. 18 341-347 (1980) 12. R.C...SPETH, G.J. WASTEK and H.I. YAMAMURA, Life Sci. 24 351-358 (1979) 13. H. MOHLER and J.G. RICHARDS, Nature 294 763-765 (1981) 14. J.G. RICHARDS, H. MOHLER and W. HAEFELY, T.I.P.S. 3 233-235 (1982)