- Email: [email protected]
[125I]PIP HOE 140, a high affinity radioligand for bradykinin B2 receptors
Life Sciences, Vol. 53, pp. 1879-1886 Printed in the USA
Pergamon Press
[125I]PIP HOE 140, A HIGH AFFINITY RADIOLIGAND FOR BRADYKININ B 2 RECEPTORS N.J. Brenner, G.Y. Stonesifer, K.A. Schneck, H.D. Burns and R.W. Ransom Merck Research Laboratories Departments of New Lead Pharmacology and Radiopharmacology West Point, PA 19486 (Received in final form October 12, 1993) Somma~ A high affinity radioli~and for bradykinin B 2 receptors was prepared by coupling an activated ester of [lz5I]4-iodobenzoic acid to the amino terminus nitrogen of the potent B 2 antagonist HOE 140. The ligand, [125I]para-iodophenyl HOE 140 ([125I]PIP HOE 140), bound to a homogeneous set of sites in guinea pig ileal membranes with an equilibrium dissociation constant of 15 pM and a maximal binding density of 193 fmole/mg protein. Competition studies with a number of BK-related peptides indicated that the lit~and specifically labeled B 2 receptors in the preparation. The results suggest that [I25I]PIP HOE 140 will be a useful tool for future studies of B 2 receptors. Bradykinin (BK) and related kinins display a variety of physiological activities (1). The predominant kinin receptor that mediates the pro-inflammatory, nociceptive and smooth muscle contractile/relaxant effects of BK has been designated as being of the B 2 type (1). BK B 1 receptors, which are preferentially activated by carboxyl-terminal des Arg-kinins (e.g. des Arg 9BK), have been less thoroughly studied, but there is increasing evidence that they may also play a role in pathophysiological processes mediated by kinins (2,3). Radioligand binding studies of B 2 receptors have, for the most part, utilized [3H]BK as a ligand, although iodinated agonist analogues such as [125I]Tyr8-BK have been employed (4,5). There has been one recent report of a radioiodinated B 2 antagonist ligand but it apparently labeled sites other than the B 2 receptor (6). One reason for the lack of radiolabeled antagonists to explore B 2 receptors has been their relatively low affinity. Recently, several reports have appeared regarding HOE 140, a very potent and selective B 2 antagonist (7,8). The present report describes the synthesis and binding activity of a radioiodinated analogue of HOE 140. The results indicate that it should be a useful tool in future investigations of BK B 2 receptors. Materials and Methods Materials. [3H]BK (102 Ci/mmole) was obtained from DuPont NEN (Boston, MA). HOE 140 was synthesized in the Medicinal Chemistry Department of Merck Research Laboratories (West Point, PA). The other peptides used in these studies were either from Bachem California (Torrence, CA) or Peninsula Laboratories Inc. (Belmont, CA). Na125I was obtained from Nordion (5 mCi in 40 I.tl NaOH, pH=10). N-succinimidyl 4-(tri-n-butylstannyl)benzoate was prepared by the method of Zalutsky and Narula (9). All other materials were of the highest grade commercially available. 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
1880
Bradyldnin Receptor Radioligand
Vol. 53, No. 25, 1993
Methods Tissue Preparation and Bindin~ Assavs. Male Hartley guinea pigs (300-600 g) were anesthetized with ether and sacrificed by decapitation. At a level beginning 2 cm above the ileocecal junction, a 10 cm section of ileum was removed, dissected free of fat and flushed with isotonic saline. The tissue was minced and homogenized with a Polytron in 20 vol. of ice-cold 25 mM Ntris(hydroxymethyl)methyl-2-aminoethane sulfonic acid (TES, pH 6.8 at room temperature) containing 1 mM 1,10-phenanthroline. The homogenate was centrifuged at 48000 x g for 15 minutes and the resulting pellet was rehomogenized in fresh buffer and centrifuged once again. The final pellet was resuspended in assay buffer [above buffer with 0.1% protease-free bovine serum albumin, 1 gM MK-422 (enalaprilat) and 140 gg/ml bacitracin] by use of a motor-driven teflon-glass tissue grinder. Protein determinations were performed by a method of Bradford (10), using bovine IgG as the standard. Membrane-binding assays using [3H]BK (20 pM for competition experiments) followed previously described methods (11,12). Briefly, [3H]BK was incubated for 60 minutes with membranes (25 gg protein) at 25°C in a final volume of 1 ml. The assay was terminated by filtration over 0.1% polyethyleneimine soaked (3 hours) Whatman GF/B filters using a Brandel M-24 cell harvester. The tubes were rinsed two times with ice-cold 10 mM TES and filter trapped radioactivity was quantitated by liquid scintillation spectrometry. Nonspecific binding was determined by performing incubations in the presence of 1 I.tM BK and represented less than 10% of the total binding at 20 pM [3H]BK. Binding assays using the [ 125I]4-iodobenzoate conjugate of HOE 140 (referred to as [125i]para_ iodophenyl or [125I]PIP HOE 140) were performed under identical conditions except that incubations were 90 minutes in length and the protein concentration was 10 ~g/tube. Dissociation experiments were performed by incubating 10 pM [125I]PIP HOE 140 with a stirred solution of membranes for 90 minutes followed by the addition of HOE 140 or BK (1 ~tM final concentrations). Total binding was determined just prior to the addition of unlabeled peptide and at subsequent timepoints by filtration of triplicate 1 ml aliquots. Nonspecific binding did not change over the course of the experiments. Saturation and dissociation experiments were analyzed using the programs EBDA and KINETIC, respectively, of McPherson (13). Competition studies were analyzed using GraphPAD InPlot (GraphPAD Software Inc., San Diego, CA). General Radi0~hemical Procedures. For specific activity determinations, UV measurements were made with a Hewlett Packard Diode Array UV-VIS detector and radioactivity measurements performed with a Packard Minaxi Auto-Gamma Counter 5000. Radiochemical yields were based on readings taken from a Capintec CRC- 12 radioisotope calibrator. TLC analysis of the radioactive product was performed using a System 200 Radioisotope Imaging Scanner, Bioscan, Inc. A Mineralight lamp (model UV GL- 25, multiband UV-254/366 nm), set to the 254 nm wavelength, was used to visualize the non-radioactive standards. The TLC experiments were performed using normal phase Merck silica gel 60CG254 plates, 5x20 cm in size, with concentrating zones and enscribed lanes. The following three elution methods were used: 15% methanol/85% methylene chloride; 80% toluene/20% ethyl acetate; and nbutanol/acetic acid/water (4:1:1). HPLC analysis was performed using a Waters 600E pump, Rheodyne injectors (model no. 7125), a Waters 990 photodiode array detector, an in-line Beckman 170 radioactivity detector and
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1881
the Waters Powerline System Controller. The HPLC analyses were monitored at 214 and 254 nm wavelengths. Svstem 1. The following system was used for HPLC purification of [125I]N- succinimidyl 4iodobenzoate. An Analtech Econosil C-18 (10 ~tm) column, measuring 10x250 mm, was connected to a C-18 (5 ~tM) Brownlee modular guard column. The Rheodyne injector was equipped with a 2 ml injection loop. Two solvents were used at a flow rate of 2 ml/minute: Solvent A: 10% methanol/90% of 1% acetic acid in water. Solvent B: 98% methanol/2% of 1% acetic acid in water. The gradient was programmed as follows: 50% Solvent A/50% Solvent B for the first 8 minutes, changing linearly to 98% of Solvent B over the next 8 minutes and continuing at 2% Solvent A/98% Solvent B for the duration of the HPLC gradient program. System 2. For purification of the final radiolabeled peptide product, a Vydac Peptide-Protein C18 (5 grn) column, measuring 4.6x250 mm, was connected to a C-18 (5 gM) Brownlee modular guard column. The injector was equipped with a 200 gl injection loop. The following linear HPLC gradient programs were used, each operating at a 1 ml/minute flow rate: Program A: Program B: Program C: Program D:
100% of 0.1% TFA in water, changing to 100% of 0.1% TFA in acetonitrile over 60 minutes; 95% of 0.025 M sodium phosphate buffer, pH 7.4/5% acetonitrile, changing to 5% buffer/95% acetonitrile over 30 minutes; isocratic system of 50% water/50% acetonitrile; 100% of 0.1% TFA in water, changing to 100% of 0.1% TFA in acetonitrile over 30 minutes. Results
Synthesis of N-Succinimidvl [125114-Iodobenzoate. The synthesis was performed similarly to the method of Wilbur, et al (14). To a 0.3 ml Wheaton conical vial was added 12 ~tl (0.50[xmol) of N-succinimidyl 4-(tri-n-butylstannyl)benzoate in methylene chloride. The solvent was removed via rotary evaporation. To the residue was added 50 ~tl of 2% acetic acid in methanol, 10 ~tl Dulbecco's PBS and 10 ~tl of a solution of N-chlorosuccinimide (NCS) (1 mg NCS/ml 2% acetic acid in methanol). The solution was thoroughly mixed with vortexing, then withdrawn from the vial using a 0.5 ml disposable syringe. A magnetic stir bar was added to the Na125I shipping vial and the shipping vial was sealed with the septum into the shipping vial of Na125I. The reaction was stirred at room temperature for 60 minutes and then quenched with 20 ~tl of a solution of sodium thiosulfate (2 mg/ml of water). The entire reaction solution was loaded onto the injector and the product was purified using HPLC System 1. The product, N-succinimidyl [125I]4-iodobenzoate, which eluted from 22.6 to 23.9 minutes, was collected in a 13x100 mm polypropylene test tube, dried in a centrifugal evaporator and used within 6-12 hours. Radiochemical yields were 50-75%. Svnthesis of [125I]PIP HOE 140 (Fig. 1). To the tube containing the dried N- succinimidyl [125I]4-iodobenzoate was added 100 ~tg (0.13 ~tmol) of HOE 140 (L- 369,554) in 50 ~tl of Sigma PBS (150 mM, pH=7.2) and 50 ~tl of borate buffer (200 mM, pH=8). The reaction was allowed to run for 4 hours at room temperature with periodic, gentle agitation. The mixture was purified using HPLC System 2 and gradient Program A. The largest radioactive peak was collected after 31 minutes. This fraction was analyzed b~¢ HPLC gradient Program B and found to contain the product, [125I]PIP HOE 140, as well as [125I]4-iodobenzoic acid.
1882
Bradykinin Receptor Radioligand
Vol. 53, No. 25, 1993
TABLE I Results of TLC Analysis of [125I]PIP HOE 140 in the following elution systems: Eluant 1: 15% methanol/85% methylene chloride; Eluant 2: 80% toluene/20% ethyl acetate; and Eluant 3: n-butanol/acetic acid/water (4:1:1). RFVALUESFOR: COMPOUND HOE 140 [125I]PIP HOE 140 4-Iodobenzoic acid [ 125I]4-iodobenzoic acid
ELUANT 1 0.00 0.00 1.00 1.00
ELUANT 2 0.00 0.00 0.35 0.35
ELUANT 3 1.00 1.00 1.00 1.00
An additional purification of [125I]PIP HOE 140 was performed, using a series of consecutive gradient programs. The HPLC Program B was used to elute the contaminant, [125I]4-iodobenzoic acid, at 12 minutes. The column, which retained the product, was rinsed under Program C to remove further contaminants. The final Program D was used to elute the product at 23 minutes. A radiochemical purity greater than 99% was confirmed using the three TLC methods described above. The TLC results are summarized in Table I. Final radiochemical yield was 20%, based on the Na125I starting material. The specific activity of the final product was 600-700 Ci/mmole. Binding Activity. Equilibrium binding of [125I]PIP HOE 140 (10000 DPM/tube) to guinea pig ileal membranes was achieved within 75 minutes at room temperature. Subsequently, a 90-minute incubation period was used to obtain the data reported here. At 10000 DPM/tube, specific binding was 400-500 DPM, which represented approximate 80% of the total binding. Saturation studies were performed with [125I]PIP HOE 140 over a concentration range of 2-200 pM. Transformation of the data by the method of Scatchard (15) showed that the ligand labeled a uniform population of sites with an equilibrium dissociation constant (KD) of 15 + 3 pM (mean -+ SEM, n=4) and a maximal binding density ( B mxa) of 193 + 12 fmole/mg protein (Fig. 2). Similar studies with [3H]BK in the same membrane preparations demonstrated that it labeled a similar number of sites (12). When a fixed concentration of BK was present in the [125I]PIP HOE 140 saturation experiments, there was a reduction in the affinity of the radioligand but no change in the Bma x value (Fig. 2). This result is consistent with a competitive interaction between the two peptides. The effects of HOE 140 on [3H]BK saturation binding parameters were also examined. Fifty pM HOE 140 reduced the apparent [3H]BK K D from 19 + 3 p M (n=3) to 57 + 5 pM without affecting the maximum density of binding sites (not shown). [125I]PIP HOE 140 was found to dissociate monophasically from ileal membranes when either HOE 140 or BK was used to prevent ligand rebinding (Fig. 3). The data are consistent with the interaction of [125I]PIp HOE 140 with a uniform population of sites. The dissociation rate constant from four determinations was 0.025 -+ 0.002 min -1 (mean _+SEM). The pharmacology of the binding of [3H]BK and [125I]PIP HOE 140 were compared. The results in Table II show that the rank order of potencies for both agonist and antagonist inhibition of the binding of each radioligand were in agreement. However, differences in the absolute potencies were observed. These differences consistently followed the trend that agonists more potently displaced [3H]BK binding than [125I]PIP HOE 140 binding while the opposite was true for peptides that are B 2 receptor antagonists (Table II). The Hill coefficients obtained for the inhibition of each radioligand by all competitors were not significantly different from one (not shown). Also, all peptides were found to completely reduce the binding of [125I]PIP HOE 140 down to the level of nonspecific binding as defined using 1 pM BK.
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1883
0
oArg-Arg-Pro-Hyp-Gly-Thi.Ser-oTic-Oic-Arg HOE 140
O
~2h--~
Icl--oArg-Arg-Pro-Hyp.Gly-Thi-Ser-DTic-Oic-Arg [I~I]PIP HOE 140
Fig. 1 Synthesis of [125I]PIP HOE 140 Discussion HOE 140 has been shown to be a potent B 2 receptor antagonist in several species (7,8). Evaluation of the peptide in competition studies indicated that, under the conditions of our binding assay, it was about 100-fold more potent than previous B 2 antagonists. It was, therefore, thought that a radiolabeled analogue of HOE 140 would provide a useful ligand for B 2 receptors. The activated ester, N-succinimidyl [125I]4-iodobenzoate, was chosen for radiolabeling the peptidebecause this radioiodinated conjugating reagent provided a means for covalently attaching a radioiodine that is not susceptible to enzymatic deiodination (as are iodinated tyrosine and other phenolic moieties) (14). The derivatization of the peptide through acylation of the N-terminus o¢amine was not expected to significantly change its binding affinity since other BK antagonists
15
09
10
LL cO r'n
5
.1
.2
Bound (pmole/mg protein) Fig. 2 Scatchard transformations of saturation binding studies with [125I]PIP HOE 140 (O) [3H]BK ( I ) and [125I]PIP HOE 140 in the presence of 100 pM BK (Q). The K D values generated from the slopes were 13 pM, 18 pM and 38 pM.
1884
Bradykinin Receptor Radioligand
~x
50
E
I
89 o
Vol. 53, No. 25, 1993
100
oo,-
".
,
LU
0
7t]..
100
-
I
I
I
I
I
I
10
20
30
40
50
60
Time (min)
Fig. 3 [125I]PIP HOE 140 dissociation from ileal membranes using HOE 140 (11, solid line) or BK to prevent rebinding of the radioligand. The data points represent specific binding (mean + SEM) from a single experiment. The inset is a semilogarithmic plot of the same data. TABLE II Inhibition of [125I]PIP HOE 140 and [3H]BK binding to membranes from guinea pig ileum. Ki (pM) P~ptide [3HIBK [ 125i]Pi P H O E BK Lys-BK Met,Lys-BK Tyr8-BK Des Arg9-BK Antazonists HOE 140 D-Arg0[Hyp2,3,Thi5,8,D-Phe7 ] BK D-Arg0[Hyp 3,D-Phe7]BK (Thi5,8,D-Phe7]BK Des Arg 9 ,Leu8-BK
21±4 34±8 66±8 211±26 >106
73±13 97±17 245±44 865±126 >106
32_+3 4650±1200 8100+_2100 17600±4140 > 106
17_+4 930±135 2510±578 6050±1120 > 106
The results are the mean + SEM from at least three determinations for each competitor. Dissociation constants (Ki) were derived from IC50 values using the equation of Cheng and Prusoff (20).
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1885
have been modified with large substituents at this position without reducing activity (16). The results presented here show that [125I]PIP HOE 140 retained the high affinity of the parent peptide. Saturation, competition and dissociation studies indicated that the radioligand binds to a homogeneous class of sites in ileal membranes. The identity of these sites as BK B 2 receptors was demonstrated by the pharmacological properties of the binding and the finding that [125I]PIP HOE 140 and [3H]BK maximum binding densities are virtually identical. Interestingly, agonists were found to more potently compete against agonist ([3H]BK) radioligand binding while antagonists exhibited the reverse preference. This result may indicate that there are distinct agonist and antagonist preferring conformations of the B 2 receptor in guinea pig ileum. It has been reported that HOE 140 blocks BK contractions of guinea pig ileum through both competitive and noncompetitive mechanisms (17). The present results indicate that the peptide behaves competitively in the [3H]BK binding assay and BK is observed to act as a competitive inhibitor of [125I]PIP HOE 140 binding. Thus, HOE 140 inhibited [3H]BK binding with a Hill coefficient near one and, in saturation experiments, it reduced the apparent affinity of [3H]BK without changing the maximum number of binding sites. Likewise, BK and its analogues inhibited [125I]PIP HOE 140 binding with Hill slopes close to one and BK's effects on [125I]PIP HOE 140 saturation binding parameters (K D, Bmax) were consistent with a competitive type of interaction. Insurmountable receptor blockade in organ bath studies by high affinity, competitive antagonists has been recently demonstrated for the angiotensin II AT 1 receptor (18,19). The observed noncompetitive component in these studies was ascribed to the slow dissociability of the antagonists which, under the conditions of the assay, produced an apparent insurmountable receptor blockade. A similar mechanism may be the cause of the noncompetitive inhibition by HOE 140 in the report of Griesbacher and Lembeck (16). Tousignant et al. (6) recently described the binding of a radioiodinated B 2 receptor antagonist ([125I]Tyr-D-Arg[Hyp3,D-Phe7,Leu8]BK) to guinea pig ileum epithelial membranes. The ligand was observed to label two sites, of which the high affinity site (KD = 1.6 nM), exhibited the pharmacological properties of a B 2 receptor. Binding to the low affinity site (K D = 16.8 nM) was displaceable by B 2 receptor antagonists but not agonists. Results using HOE 140 to inhibit binding were not reported. In the present studies, [125I]PIP HOE 140 bound to a single site and all agonists and antagonists reduced binding to the same extent. Neither the nature of the additional site reported by Tousignant et al. (6) nor the reason it would not be labeled by [125I]PIP HOE 140 are clear. It is unlikely that their results differ from those reported here because of dissimilar membrane preparations. Our preparation contained both mucosal and smooth muscle membranes, but this should not have precluded detection of their low affinity site as its density was said to be over 10-fold greater than the B 2 receptor density (2.08 pmole/mg protein vs. 0.156 pmole/mg protein). A reasonable explanation would be that [ 125i]Pi P HOE 140 is sufficiently structurally distinct from [125I]Tyr-D-Arg[Hyp3,D-Phe7,Leu8]BK that it does not recognize the latter's low affinity site. In conclusion, the results presented here indicate that [125I]PIp HOE 140 binds specifically and with high affinity to BK B 2 receptors. As an antagonist radioligand, it should be a useful probe in both membrane and autoradiographic studies of these receptors. References 1. 2.
D. REGOLI and J. BARABI~. Pharmacol. Rev. 32 1-46 (1980). F. MARCEAU, A. LUSSIER, D. REGOLI and J.P. GIROUD, Gen Pharmacol. 14 209229 (1983).
1886
3. 4. 5. 6. 7. 8.
9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19.
20.
Bradykinin Receptor Radioligand
Vol. 53, No. 25, 1993
S.G. FARMER, B.A. MCMILLAN, S.N. MEEKER and R.M. BURCH, Agents Actions 34191-193 (1991). C. TOUSIGNANT, G. GUILLEMETTE, J. BARABI~, N.-E. RHALEB and D. REGOLI, Can. J. Physiol. 69 818-825 (1991). H.M. COX, K.A. MUNDAY and J.A. POAT, Br. J. Pharmacol. 87 201-209 (1988). C. TOUSIGNANT, D. REGOLI, N.-E. RHALEB, D. JUKIC and G. GUILLEMETI'E, Eur. J. Phannacol. 225 235-244 (1992). F. LEMBECK, T. GRIESBACHER, M. ECKHARDT, S. HENKE, G. BREIPOHL and J. KNOLLE, Br. J. Pharmacol. 102 297-304 (1991). K. W I R T H , F.J. HOCK, V. ALBUS, W. LINZ, H.G. A L P E R M A N N , H. ANAGNOSTOPOULOS, ST. HENKE, G. BREIPOHL, W. KONIG, J. KNOLLE and B.A. SCHOLKENS, Br. J. Pharmacol. 774-777 (1991). M.R. ZALUTSKY and A.S. NARULA, Appl. Radiat. Isot. 38 1051-1055 (1987). M.M. BRADFORD, Anal. Biochem. 72 248-254 (1976). D.C. MANNING, R. VAVREK, J.M. STEWART and S.H. SNYDER, J. Pharmacol. Exp. Ther. 237 504-512 (1986). R.W. RANSOM, G.S. YOUNG, K. SCHNECK and C.B. GOODMAN, Biochem. Pharmacol. 43 1823-1827 (1992). G.A. MCPHERSON, J. Pharmacol. Methods 14 213-218 (1985). D.S. WILBUR, S.W. HADLEY, M.D. HYLARIDES, P.G. ABRAMS, P.A. BEAUMIER, A.C. MORGAN, J.M. RENO and A.R. FRITZBERG, J. Nucl. Med. 30 216-226 (1989). G. SCATCHARD, Ann. NY Acad. Sci 51 660-672 (1949). B. LAMMEK, Y.-X. WANG, I. GAVRAS and H. GAVRAS, J. Pharm. Pharmacol. 43 887-888 (1991). T. GRIESBACHER and F. LEMBECK, Eur. J. Pharmacol. 211 393-398 (1992). R.G. PENDLETON, G. GESSNER and E. HORNER, J. Pharmacol. Exp. Ther. 248 637-643 (1989). R.S.L. CHANG, P.K.S. SIEGL, B.V. CLINESCHMIDT, N.B. MANTLO, P.K. CHAKRAVARTY, W.J. GREENLEE, A.A. PATCHETT and V.J. LOTTI, J. Pharmacol. Exp. Ther. 262 133-138 (1992). Y.-C. CHENG and W.H. PRUSOFF, Biochem. Pharmacol. 22 3099-3108 (1973).
Pergamon Press
[125I]PIP HOE 140, A HIGH AFFINITY RADIOLIGAND FOR BRADYKININ B 2 RECEPTORS N.J. Brenner, G.Y. Stonesifer, K.A. Schneck, H.D. Burns and R.W. Ransom Merck Research Laboratories Departments of New Lead Pharmacology and Radiopharmacology West Point, PA 19486 (Received in final form October 12, 1993) Somma~ A high affinity radioli~and for bradykinin B 2 receptors was prepared by coupling an activated ester of [lz5I]4-iodobenzoic acid to the amino terminus nitrogen of the potent B 2 antagonist HOE 140. The ligand, [125I]para-iodophenyl HOE 140 ([125I]PIP HOE 140), bound to a homogeneous set of sites in guinea pig ileal membranes with an equilibrium dissociation constant of 15 pM and a maximal binding density of 193 fmole/mg protein. Competition studies with a number of BK-related peptides indicated that the lit~and specifically labeled B 2 receptors in the preparation. The results suggest that [I25I]PIP HOE 140 will be a useful tool for future studies of B 2 receptors. Bradykinin (BK) and related kinins display a variety of physiological activities (1). The predominant kinin receptor that mediates the pro-inflammatory, nociceptive and smooth muscle contractile/relaxant effects of BK has been designated as being of the B 2 type (1). BK B 1 receptors, which are preferentially activated by carboxyl-terminal des Arg-kinins (e.g. des Arg 9BK), have been less thoroughly studied, but there is increasing evidence that they may also play a role in pathophysiological processes mediated by kinins (2,3). Radioligand binding studies of B 2 receptors have, for the most part, utilized [3H]BK as a ligand, although iodinated agonist analogues such as [125I]Tyr8-BK have been employed (4,5). There has been one recent report of a radioiodinated B 2 antagonist ligand but it apparently labeled sites other than the B 2 receptor (6). One reason for the lack of radiolabeled antagonists to explore B 2 receptors has been their relatively low affinity. Recently, several reports have appeared regarding HOE 140, a very potent and selective B 2 antagonist (7,8). The present report describes the synthesis and binding activity of a radioiodinated analogue of HOE 140. The results indicate that it should be a useful tool in future investigations of BK B 2 receptors. Materials and Methods Materials. [3H]BK (102 Ci/mmole) was obtained from DuPont NEN (Boston, MA). HOE 140 was synthesized in the Medicinal Chemistry Department of Merck Research Laboratories (West Point, PA). The other peptides used in these studies were either from Bachem California (Torrence, CA) or Peninsula Laboratories Inc. (Belmont, CA). Na125I was obtained from Nordion (5 mCi in 40 I.tl NaOH, pH=10). N-succinimidyl 4-(tri-n-butylstannyl)benzoate was prepared by the method of Zalutsky and Narula (9). All other materials were of the highest grade commercially available. 0024-3205/93 $6.00 + .00 Copyright © 1993 Pergamon Press Ltd All rights reserved.
1880
Bradyldnin Receptor Radioligand
Vol. 53, No. 25, 1993
Methods Tissue Preparation and Bindin~ Assavs. Male Hartley guinea pigs (300-600 g) were anesthetized with ether and sacrificed by decapitation. At a level beginning 2 cm above the ileocecal junction, a 10 cm section of ileum was removed, dissected free of fat and flushed with isotonic saline. The tissue was minced and homogenized with a Polytron in 20 vol. of ice-cold 25 mM Ntris(hydroxymethyl)methyl-2-aminoethane sulfonic acid (TES, pH 6.8 at room temperature) containing 1 mM 1,10-phenanthroline. The homogenate was centrifuged at 48000 x g for 15 minutes and the resulting pellet was rehomogenized in fresh buffer and centrifuged once again. The final pellet was resuspended in assay buffer [above buffer with 0.1% protease-free bovine serum albumin, 1 gM MK-422 (enalaprilat) and 140 gg/ml bacitracin] by use of a motor-driven teflon-glass tissue grinder. Protein determinations were performed by a method of Bradford (10), using bovine IgG as the standard. Membrane-binding assays using [3H]BK (20 pM for competition experiments) followed previously described methods (11,12). Briefly, [3H]BK was incubated for 60 minutes with membranes (25 gg protein) at 25°C in a final volume of 1 ml. The assay was terminated by filtration over 0.1% polyethyleneimine soaked (3 hours) Whatman GF/B filters using a Brandel M-24 cell harvester. The tubes were rinsed two times with ice-cold 10 mM TES and filter trapped radioactivity was quantitated by liquid scintillation spectrometry. Nonspecific binding was determined by performing incubations in the presence of 1 I.tM BK and represented less than 10% of the total binding at 20 pM [3H]BK. Binding assays using the [ 125I]4-iodobenzoate conjugate of HOE 140 (referred to as [125i]para_ iodophenyl or [125I]PIP HOE 140) were performed under identical conditions except that incubations were 90 minutes in length and the protein concentration was 10 ~g/tube. Dissociation experiments were performed by incubating 10 pM [125I]PIP HOE 140 with a stirred solution of membranes for 90 minutes followed by the addition of HOE 140 or BK (1 ~tM final concentrations). Total binding was determined just prior to the addition of unlabeled peptide and at subsequent timepoints by filtration of triplicate 1 ml aliquots. Nonspecific binding did not change over the course of the experiments. Saturation and dissociation experiments were analyzed using the programs EBDA and KINETIC, respectively, of McPherson (13). Competition studies were analyzed using GraphPAD InPlot (GraphPAD Software Inc., San Diego, CA). General Radi0~hemical Procedures. For specific activity determinations, UV measurements were made with a Hewlett Packard Diode Array UV-VIS detector and radioactivity measurements performed with a Packard Minaxi Auto-Gamma Counter 5000. Radiochemical yields were based on readings taken from a Capintec CRC- 12 radioisotope calibrator. TLC analysis of the radioactive product was performed using a System 200 Radioisotope Imaging Scanner, Bioscan, Inc. A Mineralight lamp (model UV GL- 25, multiband UV-254/366 nm), set to the 254 nm wavelength, was used to visualize the non-radioactive standards. The TLC experiments were performed using normal phase Merck silica gel 60CG254 plates, 5x20 cm in size, with concentrating zones and enscribed lanes. The following three elution methods were used: 15% methanol/85% methylene chloride; 80% toluene/20% ethyl acetate; and nbutanol/acetic acid/water (4:1:1). HPLC analysis was performed using a Waters 600E pump, Rheodyne injectors (model no. 7125), a Waters 990 photodiode array detector, an in-line Beckman 170 radioactivity detector and
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1881
the Waters Powerline System Controller. The HPLC analyses were monitored at 214 and 254 nm wavelengths. Svstem 1. The following system was used for HPLC purification of [125I]N- succinimidyl 4iodobenzoate. An Analtech Econosil C-18 (10 ~tm) column, measuring 10x250 mm, was connected to a C-18 (5 ~tM) Brownlee modular guard column. The Rheodyne injector was equipped with a 2 ml injection loop. Two solvents were used at a flow rate of 2 ml/minute: Solvent A: 10% methanol/90% of 1% acetic acid in water. Solvent B: 98% methanol/2% of 1% acetic acid in water. The gradient was programmed as follows: 50% Solvent A/50% Solvent B for the first 8 minutes, changing linearly to 98% of Solvent B over the next 8 minutes and continuing at 2% Solvent A/98% Solvent B for the duration of the HPLC gradient program. System 2. For purification of the final radiolabeled peptide product, a Vydac Peptide-Protein C18 (5 grn) column, measuring 4.6x250 mm, was connected to a C-18 (5 gM) Brownlee modular guard column. The injector was equipped with a 200 gl injection loop. The following linear HPLC gradient programs were used, each operating at a 1 ml/minute flow rate: Program A: Program B: Program C: Program D:
100% of 0.1% TFA in water, changing to 100% of 0.1% TFA in acetonitrile over 60 minutes; 95% of 0.025 M sodium phosphate buffer, pH 7.4/5% acetonitrile, changing to 5% buffer/95% acetonitrile over 30 minutes; isocratic system of 50% water/50% acetonitrile; 100% of 0.1% TFA in water, changing to 100% of 0.1% TFA in acetonitrile over 30 minutes. Results
Synthesis of N-Succinimidvl [125114-Iodobenzoate. The synthesis was performed similarly to the method of Wilbur, et al (14). To a 0.3 ml Wheaton conical vial was added 12 ~tl (0.50[xmol) of N-succinimidyl 4-(tri-n-butylstannyl)benzoate in methylene chloride. The solvent was removed via rotary evaporation. To the residue was added 50 ~tl of 2% acetic acid in methanol, 10 ~tl Dulbecco's PBS and 10 ~tl of a solution of N-chlorosuccinimide (NCS) (1 mg NCS/ml 2% acetic acid in methanol). The solution was thoroughly mixed with vortexing, then withdrawn from the vial using a 0.5 ml disposable syringe. A magnetic stir bar was added to the Na125I shipping vial and the shipping vial was sealed with the septum into the shipping vial of Na125I. The reaction was stirred at room temperature for 60 minutes and then quenched with 20 ~tl of a solution of sodium thiosulfate (2 mg/ml of water). The entire reaction solution was loaded onto the injector and the product was purified using HPLC System 1. The product, N-succinimidyl [125I]4-iodobenzoate, which eluted from 22.6 to 23.9 minutes, was collected in a 13x100 mm polypropylene test tube, dried in a centrifugal evaporator and used within 6-12 hours. Radiochemical yields were 50-75%. Svnthesis of [125I]PIP HOE 140 (Fig. 1). To the tube containing the dried N- succinimidyl [125I]4-iodobenzoate was added 100 ~tg (0.13 ~tmol) of HOE 140 (L- 369,554) in 50 ~tl of Sigma PBS (150 mM, pH=7.2) and 50 ~tl of borate buffer (200 mM, pH=8). The reaction was allowed to run for 4 hours at room temperature with periodic, gentle agitation. The mixture was purified using HPLC System 2 and gradient Program A. The largest radioactive peak was collected after 31 minutes. This fraction was analyzed b~¢ HPLC gradient Program B and found to contain the product, [125I]PIP HOE 140, as well as [125I]4-iodobenzoic acid.
1882
Bradykinin Receptor Radioligand
Vol. 53, No. 25, 1993
TABLE I Results of TLC Analysis of [125I]PIP HOE 140 in the following elution systems: Eluant 1: 15% methanol/85% methylene chloride; Eluant 2: 80% toluene/20% ethyl acetate; and Eluant 3: n-butanol/acetic acid/water (4:1:1). RFVALUESFOR: COMPOUND HOE 140 [125I]PIP HOE 140 4-Iodobenzoic acid [ 125I]4-iodobenzoic acid
ELUANT 1 0.00 0.00 1.00 1.00
ELUANT 2 0.00 0.00 0.35 0.35
ELUANT 3 1.00 1.00 1.00 1.00
An additional purification of [125I]PIP HOE 140 was performed, using a series of consecutive gradient programs. The HPLC Program B was used to elute the contaminant, [125I]4-iodobenzoic acid, at 12 minutes. The column, which retained the product, was rinsed under Program C to remove further contaminants. The final Program D was used to elute the product at 23 minutes. A radiochemical purity greater than 99% was confirmed using the three TLC methods described above. The TLC results are summarized in Table I. Final radiochemical yield was 20%, based on the Na125I starting material. The specific activity of the final product was 600-700 Ci/mmole. Binding Activity. Equilibrium binding of [125I]PIP HOE 140 (10000 DPM/tube) to guinea pig ileal membranes was achieved within 75 minutes at room temperature. Subsequently, a 90-minute incubation period was used to obtain the data reported here. At 10000 DPM/tube, specific binding was 400-500 DPM, which represented approximate 80% of the total binding. Saturation studies were performed with [125I]PIP HOE 140 over a concentration range of 2-200 pM. Transformation of the data by the method of Scatchard (15) showed that the ligand labeled a uniform population of sites with an equilibrium dissociation constant (KD) of 15 + 3 pM (mean -+ SEM, n=4) and a maximal binding density ( B mxa) of 193 + 12 fmole/mg protein (Fig. 2). Similar studies with [3H]BK in the same membrane preparations demonstrated that it labeled a similar number of sites (12). When a fixed concentration of BK was present in the [125I]PIP HOE 140 saturation experiments, there was a reduction in the affinity of the radioligand but no change in the Bma x value (Fig. 2). This result is consistent with a competitive interaction between the two peptides. The effects of HOE 140 on [3H]BK saturation binding parameters were also examined. Fifty pM HOE 140 reduced the apparent [3H]BK K D from 19 + 3 p M (n=3) to 57 + 5 pM without affecting the maximum density of binding sites (not shown). [125I]PIP HOE 140 was found to dissociate monophasically from ileal membranes when either HOE 140 or BK was used to prevent ligand rebinding (Fig. 3). The data are consistent with the interaction of [125I]PIp HOE 140 with a uniform population of sites. The dissociation rate constant from four determinations was 0.025 -+ 0.002 min -1 (mean _+SEM). The pharmacology of the binding of [3H]BK and [125I]PIP HOE 140 were compared. The results in Table II show that the rank order of potencies for both agonist and antagonist inhibition of the binding of each radioligand were in agreement. However, differences in the absolute potencies were observed. These differences consistently followed the trend that agonists more potently displaced [3H]BK binding than [125I]PIP HOE 140 binding while the opposite was true for peptides that are B 2 receptor antagonists (Table II). The Hill coefficients obtained for the inhibition of each radioligand by all competitors were not significantly different from one (not shown). Also, all peptides were found to completely reduce the binding of [125I]PIP HOE 140 down to the level of nonspecific binding as defined using 1 pM BK.
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1883
0
oArg-Arg-Pro-Hyp-Gly-Thi.Ser-oTic-Oic-Arg HOE 140
O
~2h--~
Icl--oArg-Arg-Pro-Hyp.Gly-Thi-Ser-DTic-Oic-Arg [I~I]PIP HOE 140
Fig. 1 Synthesis of [125I]PIP HOE 140 Discussion HOE 140 has been shown to be a potent B 2 receptor antagonist in several species (7,8). Evaluation of the peptide in competition studies indicated that, under the conditions of our binding assay, it was about 100-fold more potent than previous B 2 antagonists. It was, therefore, thought that a radiolabeled analogue of HOE 140 would provide a useful ligand for B 2 receptors. The activated ester, N-succinimidyl [125I]4-iodobenzoate, was chosen for radiolabeling the peptidebecause this radioiodinated conjugating reagent provided a means for covalently attaching a radioiodine that is not susceptible to enzymatic deiodination (as are iodinated tyrosine and other phenolic moieties) (14). The derivatization of the peptide through acylation of the N-terminus o¢amine was not expected to significantly change its binding affinity since other BK antagonists
15
09
10
LL cO r'n
5
.1
.2
Bound (pmole/mg protein) Fig. 2 Scatchard transformations of saturation binding studies with [125I]PIP HOE 140 (O) [3H]BK ( I ) and [125I]PIP HOE 140 in the presence of 100 pM BK (Q). The K D values generated from the slopes were 13 pM, 18 pM and 38 pM.
1884
Bradykinin Receptor Radioligand
~x
50
E
I
89 o
Vol. 53, No. 25, 1993
100
oo,-
".
,
LU
0
7t]..
100
-
I
I
I
I
I
I
10
20
30
40
50
60
Time (min)
Fig. 3 [125I]PIP HOE 140 dissociation from ileal membranes using HOE 140 (11, solid line) or BK to prevent rebinding of the radioligand. The data points represent specific binding (mean + SEM) from a single experiment. The inset is a semilogarithmic plot of the same data. TABLE II Inhibition of [125I]PIP HOE 140 and [3H]BK binding to membranes from guinea pig ileum. Ki (pM) P~ptide [3HIBK [ 125i]Pi P H O E BK Lys-BK Met,Lys-BK Tyr8-BK Des Arg9-BK Antazonists HOE 140 D-Arg0[Hyp2,3,Thi5,8,D-Phe7 ] BK D-Arg0[Hyp 3,D-Phe7]BK (Thi5,8,D-Phe7]BK Des Arg 9 ,Leu8-BK
21±4 34±8 66±8 211±26 >106
73±13 97±17 245±44 865±126 >106
32_+3 4650±1200 8100+_2100 17600±4140 > 106
17_+4 930±135 2510±578 6050±1120 > 106
The results are the mean + SEM from at least three determinations for each competitor. Dissociation constants (Ki) were derived from IC50 values using the equation of Cheng and Prusoff (20).
Vol. 53, No. 25, 1993
Bradykinin Receptor Radioligand
1885
have been modified with large substituents at this position without reducing activity (16). The results presented here show that [125I]PIP HOE 140 retained the high affinity of the parent peptide. Saturation, competition and dissociation studies indicated that the radioligand binds to a homogeneous class of sites in ileal membranes. The identity of these sites as BK B 2 receptors was demonstrated by the pharmacological properties of the binding and the finding that [125I]PIP HOE 140 and [3H]BK maximum binding densities are virtually identical. Interestingly, agonists were found to more potently compete against agonist ([3H]BK) radioligand binding while antagonists exhibited the reverse preference. This result may indicate that there are distinct agonist and antagonist preferring conformations of the B 2 receptor in guinea pig ileum. It has been reported that HOE 140 blocks BK contractions of guinea pig ileum through both competitive and noncompetitive mechanisms (17). The present results indicate that the peptide behaves competitively in the [3H]BK binding assay and BK is observed to act as a competitive inhibitor of [125I]PIP HOE 140 binding. Thus, HOE 140 inhibited [3H]BK binding with a Hill coefficient near one and, in saturation experiments, it reduced the apparent affinity of [3H]BK without changing the maximum number of binding sites. Likewise, BK and its analogues inhibited [125I]PIP HOE 140 binding with Hill slopes close to one and BK's effects on [125I]PIP HOE 140 saturation binding parameters (K D, Bmax) were consistent with a competitive type of interaction. Insurmountable receptor blockade in organ bath studies by high affinity, competitive antagonists has been recently demonstrated for the angiotensin II AT 1 receptor (18,19). The observed noncompetitive component in these studies was ascribed to the slow dissociability of the antagonists which, under the conditions of the assay, produced an apparent insurmountable receptor blockade. A similar mechanism may be the cause of the noncompetitive inhibition by HOE 140 in the report of Griesbacher and Lembeck (16). Tousignant et al. (6) recently described the binding of a radioiodinated B 2 receptor antagonist ([125I]Tyr-D-Arg[Hyp3,D-Phe7,Leu8]BK) to guinea pig ileum epithelial membranes. The ligand was observed to label two sites, of which the high affinity site (KD = 1.6 nM), exhibited the pharmacological properties of a B 2 receptor. Binding to the low affinity site (K D = 16.8 nM) was displaceable by B 2 receptor antagonists but not agonists. Results using HOE 140 to inhibit binding were not reported. In the present studies, [125I]PIP HOE 140 bound to a single site and all agonists and antagonists reduced binding to the same extent. Neither the nature of the additional site reported by Tousignant et al. (6) nor the reason it would not be labeled by [125I]PIP HOE 140 are clear. It is unlikely that their results differ from those reported here because of dissimilar membrane preparations. Our preparation contained both mucosal and smooth muscle membranes, but this should not have precluded detection of their low affinity site as its density was said to be over 10-fold greater than the B 2 receptor density (2.08 pmole/mg protein vs. 0.156 pmole/mg protein). A reasonable explanation would be that [ 125i]Pi P HOE 140 is sufficiently structurally distinct from [125I]Tyr-D-Arg[Hyp3,D-Phe7,Leu8]BK that it does not recognize the latter's low affinity site. In conclusion, the results presented here indicate that [125I]PIp HOE 140 binds specifically and with high affinity to BK B 2 receptors. As an antagonist radioligand, it should be a useful probe in both membrane and autoradiographic studies of these receptors. References 1. 2.
D. REGOLI and J. BARABI~. Pharmacol. Rev. 32 1-46 (1980). F. MARCEAU, A. LUSSIER, D. REGOLI and J.P. GIROUD, Gen Pharmacol. 14 209229 (1983).
1886
3. 4. 5. 6. 7. 8.
9. 10. 11. 12. 13. 14.
15. 16. 17. 18. 19.
20.
Bradykinin Receptor Radioligand
Vol. 53, No. 25, 1993
S.G. FARMER, B.A. MCMILLAN, S.N. MEEKER and R.M. BURCH, Agents Actions 34191-193 (1991). C. TOUSIGNANT, G. GUILLEMETTE, J. BARABI~, N.-E. RHALEB and D. REGOLI, Can. J. Physiol. 69 818-825 (1991). H.M. COX, K.A. MUNDAY and J.A. POAT, Br. J. Pharmacol. 87 201-209 (1988). C. TOUSIGNANT, D. REGOLI, N.-E. RHALEB, D. JUKIC and G. GUILLEMETI'E, Eur. J. Phannacol. 225 235-244 (1992). F. LEMBECK, T. GRIESBACHER, M. ECKHARDT, S. HENKE, G. BREIPOHL and J. KNOLLE, Br. J. Pharmacol. 102 297-304 (1991). K. W I R T H , F.J. HOCK, V. ALBUS, W. LINZ, H.G. A L P E R M A N N , H. ANAGNOSTOPOULOS, ST. HENKE, G. BREIPOHL, W. KONIG, J. KNOLLE and B.A. SCHOLKENS, Br. J. Pharmacol. 774-777 (1991). M.R. ZALUTSKY and A.S. NARULA, Appl. Radiat. Isot. 38 1051-1055 (1987). M.M. BRADFORD, Anal. Biochem. 72 248-254 (1976). D.C. MANNING, R. VAVREK, J.M. STEWART and S.H. SNYDER, J. Pharmacol. Exp. Ther. 237 504-512 (1986). R.W. RANSOM, G.S. YOUNG, K. SCHNECK and C.B. GOODMAN, Biochem. Pharmacol. 43 1823-1827 (1992). G.A. MCPHERSON, J. Pharmacol. Methods 14 213-218 (1985). D.S. WILBUR, S.W. HADLEY, M.D. HYLARIDES, P.G. ABRAMS, P.A. BEAUMIER, A.C. MORGAN, J.M. RENO and A.R. FRITZBERG, J. Nucl. Med. 30 216-226 (1989). G. SCATCHARD, Ann. NY Acad. Sci 51 660-672 (1949). B. LAMMEK, Y.-X. WANG, I. GAVRAS and H. GAVRAS, J. Pharm. Pharmacol. 43 887-888 (1991). T. GRIESBACHER and F. LEMBECK, Eur. J. Pharmacol. 211 393-398 (1992). R.G. PENDLETON, G. GESSNER and E. HORNER, J. Pharmacol. Exp. Ther. 248 637-643 (1989). R.S.L. CHANG, P.K.S. SIEGL, B.V. CLINESCHMIDT, N.B. MANTLO, P.K. CHAKRAVARTY, W.J. GREENLEE, A.A. PATCHETT and V.J. LOTTI, J. Pharmacol. Exp. Ther. 262 133-138 (1992). Y.-C. CHENG and W.H. PRUSOFF, Biochem. Pharmacol. 22 3099-3108 (1973).