INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly fi^om the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter 6ce, \riiile others may be from aity type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e g., maps, drawings, charts) are reproduced by sectioning the original, b%inning at the upper left-hand comer and continuing fi-om left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & HowellInformation Conqtany 300 NorthZeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
PHARMACOLOGICAL CHARACTERIZATION OF THE CALCIUM CHANNEL
MEDIATED PROPERTIES OF 2-(4 •-ISOTHIOCYANATOBENZYL)
IMIDAZOLINE ANALOGS IN SMOOTH AND SKELETAL MUSCLES
A DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of the Ohio State University
By
Longping Lei, M.D., M.S. *****
The Ohio State University
1997
Dissertation Committee: Approved by Dr. Dennis R. Feller, Advisor Dr. Dennis B. McKay îdvisor Dr. Norman J. Uretsky College of Pharmacy UMI Number: 9721125
UMI Microform 9721125 Copyright 1997, by UMI Company. Ail rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103 ABSTRACT
Studies investigated the pharmacological mechanism of 2-
(4 '-isothiocyanatobenzyl) imidazoline [IBI] and analogs for interaction with imidazoline preferring receptors (IPRs), a- adrenoceptors (a-ARs) and calcium channels in cardiovascular and skeletal muscle systems. IBI differs from tolazoline by attachment of an electrophilic isothiocyanato (NCS) group.
IBI produced an irreversible, progressive and sustained contraction of rat aorta with a maximal contraction greater than that of tolazoline. The IBI-induced contractions were dependent upon calcium channels (inhibited by verapamil and nifedipine) , and independent of a-ARs (not inhibited by phenoxybenzamine ) or IPRs (not inhibited by efaroxan, moxonidine, idazoxan and cirazoline) . In rabbit diaphragm skeletal muscle, IBI produced contractile responses which were inhibited by dantrolene, an inhibitor of calcium channels present in the sarcoplasmic reticulum. In binding assays, IBI was an inhibitor of dihydropyridine (DHP) binding to calcium
1 1 channels of t-tubule membranes in rabbit skeletal muscle. In contrast to the complete displacement by DHP compounds
(nifedipine), IBI and NCS derivatives partially displaced specifically bound radioligand and gave Hill slope values which differed from unity. These results indicate that these
NCS derivatives interact allosterically with the DHP binding site. In structure activity relationship studies, IBI analogs were tested to examine their mechanisms of contractile response in rat aorta. Only NCS containing derivatives produced phenoxybenz amine-insensit ive, calcium channel dependent contractions similar to that of IBI. The data indicate that the isothiocyanato (NCS) group is responsible for the calcium channel dependent contractile properties of
IBI and its analogs in smooth and skeletal muscles. Thus, IBI and analogs may be useful as affinity probes to characterize sites of interaction on L type calcium channels.
Xll Dedicated to
Ling, Gracia and
my family ACKNOWLEIXaœNTS
I sincerely appreciate my academic advisor Dr. Dennis Feller, for his invaluable guidance, assistance and understanding throughout the course of this work; and Dr. Dennis McKay and Dr. Norman Uretsky for serving on the dissertation committee and providing me assistance. I also wish to thank Dr. Duane Miller, Dr. Meri Slavica and Dr. De Los Angeles (Department of Pharmaceutical Sciences, College of Pharmacy, the University of Tennessee, Memphis, TN) for synthesizing the compounds used in this dissertation ; to Dr. Pal Vaghy and Dr. David Johnson (Department of Medical Biochemistry at the Ohio State University) for allowing me generous access to their laboratories; to Dr. Yeindong Jiang (Dr. Johnson's laboratory) and Ms. Sherri Mazetis (Dr. Vaghy's laboratory) for co-performing some experiments; and to the faculty and the graduate fellows of Division of Pharmacology for making my studying here an enjoyable and unforgettable one. I am deeply indebted to my family with which I should have spent more time, and from which I have been constantly recharged with unending encouragement and resources during this research period and hereto journeys.
V VITA
August 25, 1958...... B o m in Hunan, The People's Republic of China 1977-1982...... M.D., Hunan College of Chinese Medicine, Changsha, China 1982-1984...... Resident Physician, Jiahe Hospital, Hunan, China 1984-1987...... M.S., Hunan College of Chinese Medi c ine, Changsha, China 1987-1990...... Physician, Department of Heart & Kidney, China-Japan Friendship Hospital, Beijing, China 1990-1992...... M.S., Division of Pharmacology, College of Pharmacy, The Ohio State University 1992-present...... Research Assistant, Division of Pharmacology, College of Pharmacy, The Ohio State University
PUBLICATIONS
X. Zhang, X. Yao, J.T. Dalton, G. Shams, L. Lei, P. Patil, D. Feller, F. Hsu, C. George and D.D. Miller: Medetomidine analogs as ttj- adrenergic ligands. 2 . design, synthesis and biological activities of conformationally restricted naphthalene derivatives of Medetomidine, Journal of Medicinal Chemistry 39: 3001-3013, 1996.
K.J. Romstedt, L. Lei, D.R. Feller, D.T. Witiak, F . Loiodice and V. Tortorella: Differential eudismic ratios in the antagonism of human platelet function by phenoxy- and thiophenoxyacetic acids. II Farmaco 51: 107-14, 1996.
VI L. Lei, M. Slavica, L. De Los Angeles, P.N. Patil, D.D. Miller and D.R. Feller: a-Adrenoceptor and imidazoline receptor- independent contractions of rat aorta : role of the 4'- isothiocyanate group on tolazoline (abstract) . The FASEB Journal 9: 3, A398, March 9, 1995.
L. Lei, M. Slavica, P. Emsberger, M.E. Graves, P.N. Patil, D.D. Miller and D.R. Feller : Calcium channel-dependent and I^- imidazoline receptor binding properties of 2 -(4'- isothiocyanatobenzyl) imidazoline analogs in vascular and brain tissues. The Annals of the New York Academy of Sciences 763: 283-286, July 12, 1995.
M. Slavica, L. Lei, P.N. Patil, A. Kerezy, D.R. Feller and D.D. Miller: Synthesis and biological activities of a new set of irreversibly acting 2-(4 '-isothiocyanatobenzyl) imidazoline analogs in rat thoracic aorta. Journal of Medicinal Chemistry 37: 1874-81, 1994.
L. Lei : Some pharmacological activities of 2- (4 ' - isothiocyanatobenzyl) imidazoline in rat aorta, guinea pig ileum and human platelets, thesis, in the pharmacy library, the Ohio State University, 1993 .
L. Lei, M. Slavica, K.J. Romstedt, P.N. Patil, D.D. Miller and D.R. Feller: Non-a-adrenergic actions of 2-(4'-isothiocyanato benzyl) imidazoline [IBI] analogs in rat aorta (abstract). The Pharmacologist 34: 148, 1992.
L. Lei: Case report on treatment of a patient with congestive cardiomyopathy. The Chinese Journal of Chinese Medicine 10: 60, 1988.
L. Lei : Anti-arrhythmic effect of Ping Lu on chloroform- induced arrhythmia in rabbits and epinephrine-induced arrhythmia in guinea pig. thesis, the library of Hunan College of Chinese Medicine, Changsha, PRC, 1985.
FIELDS OF STUDIES
Major fields : Pharmacy Specific field: Pharmacology
V I 1 CONTENT
Page Abstract...... ii Dedication...... iv Acknowledgments...... v V i t a ...... vi List of Tables...... x List of Figures...... xii
Chapter 1. Introduction 1.1 Fundamentals of drug-receptor theory...... 1 1.2 Calcium channels...... 12 1.3 a-Adrenergic receptor...... 30 1.4 Imidazoline preferring receptors...... 37 1.5 2-(4'-Isothiocyanatobenzyl) imidazoline [IBI] Analog ...... 55 1.6 Statement of the problems...... 60
2. Mechanism of 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] action: calcium channel dependent properties of the contractile responses in smooth and skeletal muscles of rat and rabbit 2.1 Introduction...... 65
2.2 Materials and methods...... 6 6 2.3 Result...... 81 2.4 Discussion...... 95 2 .5 Summary and conclusions...... 110
3. Structure activity relationship studies on pharmacological activities of 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] analogs in rat aorta and bovine brain 3.1 Introduction...... 135 3.2 Specific aims...... 136 3.3 Materials and methods...... 137 3.4 Results...... 141 3.5 Discussion...... 145 3.6 Summary and conclusions...... 150
viii 4. In vivo vascular effects and general toxicity of 2- (4 ' - isothiocyanatobenzyl) imidazoline [IBI]
4.1 In vivo vascular effects of IBI and tolazoline in anesthetized rat ...... 164 4.1.1 Specific aims...... 164 4.1.2 Chemicals and methods...... 164 4.2.3 Results...... 167 4.2 Acute toxicity and neurobehavioral activity of IBI and tolazolinein mice...... 170 4.2.1 Specific aims...... 170 4.2.2 Chemicals and methods...... 171 4.2.3 Results...... 172 4.3 Summary and conclusions...... 174
5. References...... 181
6 . Appendix...... 202
IX LIST OF TABLES
Page Table 1.1 Comparison of ryanodine receptors (RYRs) and inositol triphosphate receptors (IP^Rs) ..... 15
Table 1.2 Comparison of the properties of voltage dependent calcium channels...... 18
Table 1.3 Calcium channels antagonistsand agonists.... 22
Table 1.4 Comparison of li and I; binding sites.... 44
Table 2.1 Effects of pretreatment with phenoxybenz amine on the contractile responses to phenylephrine, tolazoline and 2 -(4'-isothiocyanato benzyl) imidazoline [IBI] on rat aorta...... 117
Table 2.2 Effects of pretreatment with imidazoline ligands (efaroxan, moxonidine, cirazoline or idazoxan) on the contractile responses to 2 -(4'- isothiocyanatobenzyl) imidazoline [IBI] on rat aorta. . 119
Table 2.3 Comparative effects of calcium deficient media and calcium channel antagonists on the contractile activities of tolazoline, 2- (4 ' -isothiocyanatobenzyl imidazoline [IBI] and S (-) Bay K 8644 on rat aorta...... 123
Table 2.4 Effect of pretreatment with NCS derivatives (analog VII, VIII) on the contractile responses to 2 -(4'-isothio-cyanatobenzyl) imidazoline [IBI] on rat aorta...... 129
Table 2.5 Effects of pretreatment with d-tubocurare or dantrolene on IBI induced contraction in rat di aphragm...... 131 Table 2.6 Comparative inhibitory constant (K^) , Hill slope (n^) of and percentage inhibition of nifedipine, tolazoline, naphazoline and their NCS derivatives in displacing (+) [^H]PN 200-110 labeled sites from rabbit skeletal t-tubule membrane...... 133
Table 3.1 Chemical structures and abbreviations of 2-(4 '-iso-thiocyanatobenzyl) imidazoline [IBI] and related analogs...... 154
Table 3 .2 Effect of phenoxybenzamine pretreatment on the contractile responses to 2 -(4'-isothiocyanato benzyl) imidazoline [IBI] analogs on rat aorta...... 155
Table 3.3 Effect of imidazoline receptor ligands (cirazoline and idazoxan) and calcium channel blocker (nifedipine) on contractile responses of rat aorta to 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI]analogs ...... 156
Table 4.1 Effect of propranolol, atropine and pyrilamine pretreatment on the cardiovascular activities induced by acetylcholine, histamine, isoproterenol and IBI in anesthetized rat ...... 177
Table 4.2 Effect of IB I -pre treatment on the mean artery pressure induced by norepinephrine, acetylcholine, histamine and isoproterenol in anesthetized rat ...... 178
Table 4.3 Evaluation of the mean lethal dose on mice by IBI and tolazoline...... 179
Table 4.4 Effects of IBI and tolazoline treatment on mobility of mice in the rotor-rod test...... 180
Appendix Table A1 Synthesized compounds...... 202
Appendix Table A2 Comparison of ligand displacing ICgo values of 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] analogs before and after phenoxybenzamine-treatment on bovine ventrolateral medulla membranes bound to [^^®I]p- iodoclonidine...... 203
xi LIST OF FIGURES
Page Figure 1.1 Chemical structures of three major types of calcium inhibitors...... 2 1
Figure 1.2 Chemical structures of dihydropyridines as calcium activators and inhibitors...... 23
Figure 1.3 Schematic representation of L-type calcium channel (A) and the Domain interface model of calcium channels (B)...... 28
Figure 1.4 Chemical structures of some important catecholamines...... 31
Figure 1.5 Chemical structures of selective imidazolines...... 38
Figure 1.6 Imidazolines used for identifying imidazoline preferring sites in radioligand assays..... 45
Figure 1.7 Chemical structures of isothiocyanato derivatives of tolazoline...... 56
Figure 1.8 Schematic representation of the Easson- Stedman hypothesis...... 60
Figure 2.1 Time course and absolute tension changes produced by phenylephrine, tolazoline and 2 -(4'- isothiocyanatobenzyl) imidazoline [IBI] in rat aorta.... 115
Figure 2.2 Effect of phenoxybenzamine pretreatment on contractile responses of phenylephrine, tolazoline and 2-(4'-isothio-cyanatobenzyl) imidazoline [IBI] in rat aorta...... 116
XXI Figure 2.3 Effects of imidazoline ligands pretreatment on potency (A), and the maximal contraction (B) to 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] in rat aorta...... 118
Figure 2.4 Effects of calcium deficient media and pre treatment with nifedipine and verapamil on the contractile activities of 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] and tolazoline on rat aorta...... 120
Figure 2.5 Effects of pre treatment with R (+)-Bay K 8644 (0.3 pM) on the contractile activities of S (-) -Bay K 8644 and 2- (4 ' - isothiocyanatobenzyl) imidazoline [IBI] on rat aorta...... 1 2 2
Figure 2 . 6 Fura-2 fluorescence intensity measurements for pig right coronary artery in response to KCl ...... 124
Figure 2.7 Representative tracing of KCl-induced augment of fura - 2 fluorescence coupled with the tension chagnes in pig coronaryartery ...... 125
Figure 2 . 8 IBI-induced changes in fura-2 fluorescence and tension...... 126
Figure 2.9 Inhibitory effects of 100 pM verapamil on the plateau of calcium signal induced by IBI in pig coronary artery...... 127
Figure 2.10 Effects of pretreatment with NCS derivatives on the contractile responses to 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] on rat aorta...... 128
Figure 2.11 Effect of pretreatment with dantrolene on IBI induced contractile response of rat diaphragms . . . 130
Figure 2.12 Displacement of ( + )[^H]PN 200-110 by tolazoline, naphazoline, nifidipine and their NCS derivatives from rabbit skeletal t-tubule membrane 132
Figure 2.13 Displacement of S(-)[^H]Bay K 8644 by tolazoline, naphazoline and their corresponding NCS derivatives from rabbit skeletal t-tubule membrane 134 xiii Figure 3.1 Effects of pretreatment with phenoxybenzamine on the contractile responses to tolazoline, analog II and analog III on rat aorta...... 157
Figure 3.2 Effects of pretreatment with phenoxybenzamine on the contractile responses to IBI, analog IV, V and VT on rat aorta...... 159
Figure 3.3 Comparative contractile concentration- response curves of analogs IX, X, XI and XII on rat aorta in the presence of nifedipine, phenoxybenz amine followed with cirazoline or idazoxan...... 161
Figure 3.4 Effect of phenoxybenzamine or nifedipine pretreatment on contractile responses of 3'-isothio cyanatobenzyl imidazoline (3 ' -IBI) on rat aorta...... 163
Figure 4.1 Representative tracings of the mean artery pressure of anesthetized rats...... 176
Appendix Figure Al. Displacing of [ I ] p-iodoclonidine by 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] analogs on bovine ventrolateral medulla membranes...... 204
XIV CHAPTER 1
INTRODUCTION
1.1 Fundamentals of drug-receptor theory
1.1.1 Receptor occupancy and intrinsic activity
Originated around the beginning of this century, drug- receptor theory has become the keystone of the contemporary pharmacology. It has provided researchers with useful quantitative methodology for studying drug action. The early conceptualization of receptor stemmed from "locus" (Claude
Bernard, 1813-1878), "receptive substance" (J.N. Langley,
1909) and the hypothesis of drug action specificity and selectivity (Paul Ehrlich, 1854-1915). With the contributions in the discipline of biochemistry, the distinctive era of pharmacology as a quantitative science developed. Most often encountered in the field of enzyme kinetics, the Michaelis-
Menten (1913) equation (v = V„h x [S]/{K„ + [S] }) was established and used to predict a hyperbolic relationship between the steady-state enzyme velocity (v) , maximal velocity (V^) , dissociation constant of the equilibrium enzyme-substrate complex (Km) , and the substrate concentration ( [S] ) . A pioneer of quantitative pharmacology, Clark (1926; 1937) employed the same similar principles of mass action law to agonist-receptor interactions giving an equation: = [Ago,^st ] [Receptor ] /
[RAcompiex] > where is the dissociation constant at equilibrium assuming [A] + [R] = [RA] . He described receptor occupancy theory (RA/Rr, where Rr = [RA] + [R] ) by putting the concept on a qualitative basis for the adsorption isotherm curves proposed by Langmuir (1918) . Thus a fundamental pharmacological equation was developed as [RA /Rr] = [A] / (Ka +
[A] ) which obeyed the assumptions of the Michaelis-Menten equation. This equation dealt with relationship between receptor occupancy, agonist concentration ( [A] ) and dissociation constant, and proposed that the magnitude of the response to an agonist was directly proportional to the fractional receptor occupancy. Thus, the maximal response was only achieved when the entire receptor population was saturated with drug. However, since biological responses induced by drugs are not merely an agonist phenomenon, Ariêns
(1954) complemented Clark's theory by raising two distinct terms for agonist and antagonist interactions with receptors, affinity and intrinsic activity. While the direct proportional response versus receptor occupancy was retained, he defined affinity to reflect the ability of both agonists and antagonists binding the receptors to form a drug-receptor complex. Thus, intrinsic activity was used only to describe responses stimulated by agonists which occur subsequent to binding. Antagonists are able to bind to the receptors but unable to produce a biological response. Thus, the intrinsic
activity of an antagonist is designated as 0 as opposed to
that of an agonist as 1 , and to that of a partial agonist
(later termed by Stephenson, 1956) as less than l and greater than 0. Affinity was measured at median effective dose of the maximum, or EDjo, at which 50% total receptors are occupied.
1.1.2 Efficacy and intrinsic efficacy
The theories advocated by Clark and Ariêns were challenged by Stephenson (1956), Nickerson (1956) and
Furchgott (1966) who proposed that a maximal effect can be produced by an agonist when occupying only a small fraction of the total receptors in the tissue and that the response was not always linearly proportional to receptor occupancy. Thus, drugs may have varying capacities to initiate a response and consequently occupy different proportions of receptors when producing equal responses. In Stephenson's rationalization, response magnitude elicited by an agonist was some unknown function (f) of the biological stimulus (S) regardless the receptor occupancy, and expressed the concept as EA/EL,v = f (S) , where S was proposed to be equal to the eff±ca.cy (e) of the drug. Accordingly, Clark-Langmuir equation was rationalized as EA/E^ax = f(e[A]/Ka + [A]). Efficacy differs from intrinsic activity in that two agonists occupying varied fractions of receptors may produce the same amount of response, e.g., 100% contraction. Thus, intrinsic activity follows receptors occupancy theory, whereas efficacy does not.
Furchgott extended the theory of Stephenson by proposing that efficacy was the product of intrinsic efficacy (e) and total receptor concentration ( [RT] ) such that e = e [Rrl , thus EA/E^
= f (e [Rr] [A]/Ka + [A]) which represents the current status of occupancy theory. The equation symbolizes non-linearly proportional responses versus receptor occupancy considering unknown factor f (which does exist but not prevent the equation from utilizing) and efficacy (e) of drug in a given concentration of total available receptors [R?]. Due to these possibilities, the efficacy of a drug may vary between tissues or be tissue-dependent, and the maximal response can be obtained by a full agonist occupying only a fraction of the receptors, and the maximal response cannot be produced by a partial agonist no matter how large a dose is used.
Additionally, ED 50 and maximal effect may not reflect the affinity and intrinsic activity or intrinsic efficacy of a
drug for the tissue population of receptors. Rather, ED 50 represents the concentration of drug that produces a half maximal functional effect.
1.1.3 Dissociation constant for agonist and antagonist
Several quantitatively applicable methods (for review see
Ruffolo, 1984) have been developed to determine dissociation constant (K^) of full agonists (Furchgott, 1966) and partial agonists (Waud, 1969), and dissociation constant (Kq) of competitive antagonists (Schild 1949). To determine the dissociation constant of a full agonist, Furchgott (1966) utilized an irreversible agent to occupy the fraction of available receptors with the assumption that equieffective responses before and after receptor inactivation will occur only when there is a equieffective biological stimulus before and after the receptor inactivation. Plotting the reciprocals of the equieffective concentrations of agonist before (1/ [A] ) against that of after receptor inactivation (1/[A]') , a straight line is created and agonist constant value is determined using K* = (slope - 1)/ Y intercept. Despite less frequently used than EDgo, the value of an agonist provides with better evaluation of affinity than the EDgo value in the occupancy point of view. Determination of partial agonist was established by Waud (1969) . Since the application of the same concentration of the antagonist would inactivate a certain amount of available receptor sites, the addition of agonist to the system may yield varied responsive intensity for full and partial agonists. Thus, a plot of the reciprocal concentration of a full agonist against the reciprocal equieffective concentration of a partial agonist will yield a straight line with the = slope/Y intercept value.
For determination of the disaocia.tion constant of an antagonist, Schild realized that it was not the same situation as that of a full or partial agonist. He assumed that the dose-response curve of an agonist would be shifted in a parallel manner to the right by an antagonist regardless of whether the effect of the agonist was a linear or non-linear function of agonist receptor occupancy. The factor causing the rightward shift of the concentration-response curve of agonist was due to the presence of an applied concentration of the antagonist {[B] ) with a certain dissociation constant (Kg) in a given tissue, i.e., 1 + ( [B]/Kg) . A compound is considered to be a competitive antagonist if the curve of agonist is shifted to the right in a parallel manner; and if, on the other hand, the maximal response was suppressed in the presence of antagonist, a fashion of non-competitive antagonism is assumed. Kg is determined by acquiring the EDgo of agonist alone and in the presence of at least three concentrations of a receptor antagonist as described by Schild
(1947). The log concentration ratio of initial over apparent
ED 50 values can be plotted at the y-axis against log concentration of agonist using the equation, log (dose ratio -
1 ) = log [B] - log I^, as is derived from rearrangement of the equation of receptor occupancy by Clark (see page 2 of this dissertation). A positive linear relationship with a slope
equal to 1 denotes competitive interaction between antagonist and agonist (Arunlakshana and Schild, 1959) . At the interception of x and y-axes of the straight line of the plot, pAj (or -log Kg) value is determined, and numerically is equal to agonist concentration (log [B] ) when log (dose ratio
- 1) is zero. In other words, it represents the negative logarithm of the molar concentration of antagonist in the presence of agonist whose potency is reduced by two-fold,
namely, 2 times more agonist is required to produce a given response in the presence than in the absence of antagonist.
The importance of the pA; lies in its independency of the agonist used, thus, it is useful quantitative tool for the classification of receptors. Usually, an identical receptor interacting site is proposed if an equal degree of antagonism against two agonists (demonstrated by identical pA; values) is observed by the same antagonist in a given tissue.
1.1.4 Radioligauid binding theory
Radioligand binding assay in the 1970's has been played important role in receptor characterizations and subdivision of receptors (for review see Bylund and Yamamura, 1990). In general, there are two major types of radioligand binding assays employed for pharmacodynamic studies : saturation binding and competitive binding. The purpose of saturation binding is to estimate the dissociation constant (I^) of the radioligand and to determine the maximal binding to receptors (Bmax) • These parameters are determined by adding increasing
concentrations of radioligand to a fixed amount of tissue or
cell membrane alone and in the presence of a saturating
concentration of unlabeled drug that interacts with the
designated receptor population. There are three approaches in
the analysis of the results of radioligand binding assay; they are saturation plot, Scatchard (1949) plot and Hill (1910) plot. The saturation plot is formulated as B = [F] / (K^, +
[F]) which describes the relationship among receptor- radioligand bound complex (B), free radioligand concentration
([F]), Bma% and dissociation constant of radioligand (K^) . The equation is analogous to the equation of receptor occupancy; thus, it depicts a rectangular hyperbolic curve. The total bound radioligand is definedas the specific bound (SB) plus the nonspecific bound (MSB, the observed binding in the presence of an adequate excess of drug that is chemically dissimilar from the radioligand) . UnliJce SB that is saturable because of the limited receptor binding sites, NSB occurs due to binding to glass fiber or membrane lipid, and thus is not saturable and is more rapid to attain steady state. In order to avoid artifacts of high nonspecific bound, the initial segment of the SB curve can be used to determine the proper tissue concentration at which less than 1 0 % of the added
radioligand is bound to the tissue. With the data plotted according to the saturation equation, it is difficult to determine Kp and values owing to the non-linear curve. The equation was transformed into linear relationship by Scatchard
(1949) and Rosenthal (1967) by rearrangement of the saturation equation into: B/F = -B/Kg + Bj^/K d • A straight line with a negative slope is plotted using ratio of bound over free concentration of radioligand versus bound concentration of radioligand. The Kg value is calculated according to -l/slope and the x-intercept represents B„^^. Hill (1910) plot analysis of saturation binding is plotted as positive straight line using the equation log [B/(B^-B)] = n log [F] - n log Kg .
Therefore a plot of log [B/ (B^-B) ] as a function of log [F] yields a line whose slope is n^ or Hill slope. Hill coefficient. This plot was initially used in describing the positive coopérâtivity seen with Og bound to hemoglobin, and is also used to measure the slope of the Hill plot. When n^ =
1 .0 , the ligand binds to a single class of noninteracting
(noncooperative) receptors, abiding the law of mass action;
when ng > 1 .0 , it indicates a positive coopérât ivity, whereby occupancy of one binding site by a ligand enhances the
10 likelihood of preferential binding of other coupled site on the same oligomeric protein by the ligand of related ligand; and when ng < 1, there exists either negative cooperativity or multiple binding sites (Taylor and Insel, 1990).
With the determination of the proper concentration of radioligand and receptor protein, a competitive binding assay can be considered. It involves the addition of an increasing concentration of unlabeled competing ligand to fixed concentrations of radioligand and receptor protein. In the presence of the competing drug, the receptor-bound radioligand dissociates from the complex intermediate into the free or unbound form. With varying concentrations of the competing drug, the binding of the radioligand to the receptor is consequently affected, thus the dissociation constant of radioligand (Kq*) becomes a product of its interaction with a factor designated as 1 + I/K^. The dissociation constant of the competing drug (Kj) is defined as the concentration of the competing or inhibitory drug [I] needed to occupy 50% of the receptors. By incorporating this factor into the saturation equation, the Cheng and Prusoff equation (1973) is simplified as Ki = IC so/(l + [D*]/Kd*), assuming an equilibrium of half displacement of the receptor-bound radioligand by the
11 competing drug (where I = I5 0 ) . By eliminating the presence of possible variance of concentration ([D*] ) and dissociation constant (Ko*) of radioligand, the inhibitory constant (Ki ) value of the competing drug is considered to be more reliable
than IC 5 0 , which is defined as the concentration of competing drug needed to displace 50% of the receptor-bound radioligand.
In other words, ICgo is dependent of the concentration and dissociation constant of the radioligand as used in the binding system, whereas Ki is independent of these parameters.
1.2 Calcium channels
1.2.1 Calcium chauinels and calcium pools of cells
Extracellular pool of calcium. Although there is some involvement of Na* currents, depolarization of vascular smooth muscle cells is primarily dependent on the influx of Ca^*
(Bolton, 1979) . There are a number of ways that calcium can gain entry into the cells (for review see Partridge, 1991;
McDonald et al., 1994). The most extensively studied class of calcium channels are voltage dependent calcium channels
(VDCs) . They open in response to depolarization of the membrane, and extracellular calcium moves down its
12 electrochemical gradient into the cells. As opposed to VDCs, receptor-operated channels (ROCs) can be divided into either opened in direct response to the binding of a ligand to the external side of the channel and allow the entry of calcium, or through an indirect manner such as G-protein coupled mechanism as seen in «i-adrenoceptor activation. The direct type of ROCs was found as N-methyl-D-aspartate (NMDA) receptor in neurons of the central nervous system, or as a non-NMDA receptor sensitive to glutamate in retinal bipolar cells
(Gilbertson et al., 1991) . The indirect, or second messenger- operated channels are channels that open in response to intracellular second messengers produced by the receptor- ligand interaction, and are found in muscles, neutrophils, platelets, lymphocytes and mast cells. Several additional less defined calcium channels have been proposed to exist within the cell membrane of excitable tissues. "Stretch-activated"
(or mechanical-activated) calcium channels, found on endothelial cells, are responsive to hemodynamic stresses on the valvular tone of smooth muscle cells (Hwa and Seven,
1986) . A "leak" channel was described in which calcium enters the cell at rest due to the high concentration gradient which exists at the membrane (Flaim et al., 1984) .
13 Intracellular pool of calcium. The increase of the intracellular calcium concentration is also dependent upon calcium release from the intracellular pool through two major intracellular calcium channels (for review see Karaki, 1989;
Berridge, 1993; McPhemson et al., 1993), ryanodine-sensitive receptors (RYRs) and inositol triphosphate-sensitive receptors
(IP3RS) .
The RYRs are sensitive to openers such as ryanodine, caffeine, ADP and Ca^*. The most important features of these channels are their responsiveness to depolarization (or VDCs mediated) and to non-depolarization (e.g., dihydropyridines,
DHP) . VDCs respond to voltage changes by gating a small amount of calcium, which in turn functions as a trigger activating the RYRs to release stored calcium. Due to such property of
"calcium induced calcium release" (CICR), RYRs are also referred as CICR receptors (Kuriyama et al., 1995). RYRs are typically found in skeletal (RYRl), cardiac (RYR2) and smooth muscle/non-muscle tissues (RYR3) (Kuriyama et al., 1995).
Using an immunohistochemical procedure, the ryanodine receptor is characterized mainly distributed on the superficial sarcoplasmic reticulum (SR) vesicle in the guinea pig taenia
14 coli, but not in the aorta (Lesh et al., 1993; Kuriyama et al., 1995). On the other hand, RYAS located in the sarcoplasmic reticulum of skeletal muscle contribute to the transverse (t)-tubule foot structure responsible for excitation-contraction coupling (McPherson and Campbell,
1993) . The cytosol RYAS are thought to interact with sarcolemma (or t-tubule) dihydropyridine receptor (or VDC) in a direct manner, leading to calcium release from SR, In such a VDC-RYR-SR triad, the dihydropyridine - receptor (see Table
1.3 for receptor agonists) in the surface membrane functions as voltage sensor which senses a change in voltage and undergoes conformational change which is transmitted through the bulbous head of the RYAS to open the channels in the SR
(Schneider et al., 1973; McPherson and Campbell, 1993;
Berridge, 1989; 1993).
Unlike the RYAS, IPjRs release calcium from internal calcium stores by an indirect process involving in membrane bound proteins coupling with effector target and second messenger elicited by activation of ROCs such as «i-AR. The process describes "pharmacomechanical coupling" because of a dissociation between the membrane potential change and contraction (Somlyo and Somlyo, 1968), and the mechanism of
15 Parameter ______m s ______lEafiS______
Mol. weight 4x565 kDa 4x313 kDa
Types RYRl (skeletal) IP3 R family: RYR2 (cardiac) IP3R1,2,3,4,5 RYR3 (smooth muscle, endothelial cells)
Structure tetramer/monomer tetramer
Distribution cardiac, Xenopus, spinesskeletal, neurons both: atrium, vascular cells, chromaffin cells, cerebellar Purkinje cells
Receptor voltage-dependent Go-linked PLC-S coupling channels (via « 1 -adrenoceptors) Tyrosine kinase- linked PLC-y
Physiologic Ca^ IP, activators cADP ribo_.
Ca^*-modulât ion activity activity increases: <0.1mM Ca^* increases: <0.3/zM Ca** decreases : >lmM Ca^* decreases: >0.3fM Ca^*
ATP-modulation activity increases activity increases : < 2mM decreases : > 4mM
Pharmacologic ryanodine (< 10 /iM) activators doxorubicin caffeine imperatoxin activator (skeletal isoform)
Pharmacologic ryanodine (>10 /zM) heparin inhibitors heparin caffeine ruthenium red ethanol imperatoxin inhibitor
Modified from Berridge, 1993, Ehrlich et ai., 1994 and McPherson et ai., 1993; IP3 : inositol (1,4,5)-triphosphate.
1.1 Comparison of rycuiodine receptors (RYAS) and inositol triphosphate receptors (IP3 RS)
16 IPjRs activated by IP3 was first elucidated in smooth muscles
by Suematsu et al. (1984) . In fact, the second messenger IP3 has a key role in controlling the intracellular calcium concentration by mobilization of internal stores as well as an increased entry of external calcium (Berridge, 1993; Kuriyama
at al., 1995). Cells generate IP3 through two major signaling pathways involving phospholipase C (PLC). In one pathway, these receptors are linked to PLC-S mediated via G protein, and the other pathway is coupled to PLC-y via autophosphorylation of specific tyrosine kinase (Berridge,
1993) . Some electrophysiological and pharmacological properties of these two intracellular calcium channels are listed in Table 1.1,
1.2.2 Calcium channel classification emd characterization
It was first realized that voltage dependent calcium influx occurred through work with crab leg fibers in which
"calcium spike" was described as the result of calcium influx that was enhanced by quaternary ammonium ion or procaine (Fatt and Katz, 1953). These studies led to the eventual observation that every excitable cell has calcium channels (Hille, 1984).
Based on the electrophysiological and pharmacological natures,
17 Type L-type T-type N-type P-type
Unitary 20-27 8-10 13 1-12 conductance (pS)
Activation -30 to -10 -70 -30 to -10 -60 to -40 range (mV)
Inactivation No Yes Yes Yes
Blockers Organic Tetramethin, u-conotoxin Funnel Web calcium, amiloride, Spider toxin DHP, octanol diltiazem phenylalkyl- amine
Tissue cardiac, specialized neuronal cerebellar distribution smooth and conducting eps. Purkinje cell skeletal cardiac dendrites muscles, cells nerve and secretory cells
Modified from Spedding and Paoletti (1992).
Table 1.2 Comparison of the properties of voltage dependent calcium channels
four major types of VDC have been characterized (Spedding and
Paoletti, 1992); see Table 1.2 for a summary. The L type
(long-lasting, larger molecular weight channels), T type
18 (transient tiny molecular weight channels), N type (neuronal tissue and resembling neither of the other two in kinetics and inhibitor sensitivity) and P type (Purkinje cells) of calcium channels are most widely characterized.
1.2.3 Calcium channel modulators
The primary modulator of VDCs is the membrane potential.
In skeletal muscle cells, which are the best studied tissue type, the potential is spread through sarcolemma, a plasma membrane flanking the cells. Then it activates t-tubules which are the continuation and invagination of the sarcolemma at the
Z-line of a sarcomere unit. It is believed that the potential- dependent membrane depolarization of t-tubules is coupled to the calcium release form the sarcoplasmic reticulum (SR) which consists of a network of interconnected tubules forming a closed space inside the cytoplasm. Under resting condition, these excitable cells maintain a membrane potential at about
-90 mV with a low free intracellular calcium concentration
(100 nM) which is 4 orders of magnitude lower than that (1 mM) in extracellular space. This large concentration gradient represents an enormous driving force for calcium to enter the cell and can be maintained only by a membrane that is largely
19 impermeable to calcium and contains calcium pump to remove calcium out of the cell. Upon the excitation of the cell membrane, depolarization results in a action potential which is caused by sodium influx through sodium channels. This raises the potential from -90 mV to -30 mV, a threshold voltage at which calcium channel opens due to increased conductivity to calcium. The mechanism for this change in calcium conductance involves a depolarization of the sarcolemma which, in turn, causes charge movement in the voltage sensitive calcium channel of the t-tubules (Schneider et al., 1973). Thus it is the charge movement that is responsible for the excitation-contraction coupling. Opening of the calcium channels results in calcium release from the SR
into the cytoplasm, rapidly raising the free calcium from 1 0 0 nM to 10 /iM. Calcium occupies binding sites of intracellular binding proteins such as troponin C (in skeletal and cardiac muscles) and calmodulin (in smooth muscles). These calcium- bound proteins then interact with other regulatory proteins and enzymes, e.g., troponin I in cardiac and skeletal muscles and myosin light-chain kinase in smooth muscles. This results in conformational changes to these proteins in such a way that actins of the thin filament are exposed to myosins of the
20 thick filaments (in skeletal and cardiac muscles), ultimately- facilitating cross-bridge formation between actin and myosin leading to muscle contraction. During repolarization induced by potassium efflux, relaxation occurs because calcium is removed by calcium ATPase from the calcium specific regulatory sites of troponin C back into the sarcoplasmic reticulum and to extracellular space (Murphy et al., 1993).
OCH: CN CH OCH:
Verapamil (phenylalkylamlne)
OCH
NO 2 OCOCH; HgCOOC COOCH3
(CH3)2NCH2CH2
Diltiazem (benzothiazepine) Nifedipine (dihydropyridine)
Figure 1.1 Chemical structures of three major types of calcium inhibitors.
21 In addition to membrane potential, other endogenous and exogenous modulators also affect VDCs. Hormones and neurotransmitters (catecholamines) regulate the VDCs by activation of protein kinases (A or C) that phosphorylate the channels, or through guanine nucleotide-binding (G) proteins that link S-adrenoceptors to calcium channels. Inorganic ions such as Co^*, Ni^* and Cd^* are inhibitors of the VDCs. The drugs that act directly on calcium channels include both calcium antagonists and agonists. These chemicals are heterogeneous in their structures and can be classified as the following groups (see Figures 1.1 and 1.2 for structures).
Antagonistg Agonists
1,4-Dihydropyridines R(+)Bay K 8644 S(-)Bay K 8644 (-)202-791 (+)202-791 Nitrendipine CGP 283 92 Nifedipine YC-170 PN 200-110 Phenylalkylamines Verapamil
Desmethoxyverapamil (D-8 8 8 ) D-600 Benzothiazepines Diltiazem TA 3090
Modified from Vaghy et al., 1987b.
Table 1.3 Calcium channels antagonists and agonists
22 a OoN COOCH3 02N\>k^COOCH3
HaC CHa'N..LX, CHc
S(-) Bay K 8644 R(+) Bay K 8644 (Calcium channel activator) (Calcium channel inhibitor)
N
N OOC. NO.
CH; JL X H S 202-791 R 202-791 (Calcium channel inhibitor) (Calcium channel activator)
OOC COOCHa
HaC^ "N" "CHa rl
PN 200-110 (isradipine, calcium channel inhibitor)
Figure 1. 2 Chemical structures of dihydropyridines as calcium activators and inhibitors.
23 Calcium antagonists were initially described to the effects of certain phenylalkylamines such as prenylamine and verapamil on isolated cardiac papillary muscles preparations, which were indistinguishable from the effects of calcium removal (Fleckenstein, 1977). The calcium channel blocking
effects could be reversed by calcium addition, 6 -adrenergic agonists and cardiac glycosides. Verapamil and prenylamine, together with other drugs that inhibit excitation-contraction coupling, were termed as calcium antagonists (for review, see
Fleckenstein, 1977; 1983) . The calcium antagonists consist of three chemically distinct groups of compounds. Figure 1.1 shows the chemical structures of representative compounds from
each group, and includes nifedipine (a 1 ,4-dihydropyridine
(DHP) , verapamil (a phenylalkylamlne) , and diltiazem (a benzothiazepine). Thus, calcium antagonists act by inhibiting the influx of calcium into cells through specific voltage- dependent calcium channels located in cell membranes. Since the chemical structures of these three representative drugs are completely different, it has been proposed that each of the drugs has a distinct receptor site on L type calcium channels. Also, one drug may allosterically bind to the receptor sites of the other two drugs. For instance, verapamil
24 inhibits the binding of diltiazem and vice versa. These receptors are located on the same Ofi- subunit on L type VDCs
(Vaghy et al., 1987b; Dumont et al., 1988),
Within the dihydropyridine class of calcium channel modulators are confounds which have opposite acting effects on the L type calcium channel. These drugs, known as calcium agonists, such as Bay K 8644 (Figure 1,2) exert their action by increasing calcium influx through L type calcium channels and producing contraction of vascular smooth muscle and positive inotropy in the heart (Schramm et al,, 1983; Bean et al,, 1986; Kuriyama et al,, 1995), What also makes this group of dihydropyridines interesting is the fact that properties of calcium channel agonism and antagonism can reside within the enantiomers of a single dihydropyridine molecule (Dube et al,,
1985a, 1985b; Franckowiak et al,, 1985; Williams et al,,
1985), The DHP calcium agonists are structurally similar to their antagonists which are featured with ester functionality in both of the C-3 and C-5 positions on the 1,4-DHP ring.
Conversely, DHP calcium agonists possess one ester, rending a
C-4 asymmetric center to the DHP, Although few data are available with which to compare structural requirements for activation and blockade, such a chiral center was reported to
25 impose effects of stereoselectivity. It is possible that discrete binding sites exist for activator and blocker species
(Kokubun et al., 1986; Janis and Triggle, 1991).
Calcium channel blockers, particularly those of L-type, have therapeutic applications in the treatment of angina pectoris, supraventricular tachycardia, hypertension, post hemorrhagic cerebral vasospasm, and Raynaud's phenomenon.
Other potential uses of these drugs include pulmonary hypertension, asthma, premature labor, epilepsy, glaucoma,
Alzheimer's disease, and dysmenorrhea (Janis and Triggle,
1991).
1.2.4 Molecular mechanisms of L type calcium channels.
The L type channel has been extensively studied and is characterized by high-threshold activation and non inactivating currents in most excitable tissues. These channels are responsible for the entry of "slow calcium " during the cardiac potential (Hille, 1984), and exist in a variety of excitable tissues including neurons, cardiac, skeletal and smooth muscles and endocrine cells (for review, see Tsien, 1983; Bean, 1989; McDonald et al., 1994).
The voltage dependent L type calcium channel was
26 initially isolated from skeletal muscles, and purified as five protein subunits (429 kD) called Oi, (Xg, 6, y and 5 (Tanabe et al., 1987; see Figure 1.3). The primary structure of the subunit (170 kD) in rabbit skeletal muscles is the ion selective pore of the channel. With the hydrophobic ity analysis, the subunit was predicted to possess a transmembrane topology with four homologous repeats each containing six helical membrane spanning segments S1-S6. It operates as voltage sensor as well as a calcium channel (Beam et al., 1992). The fourth segment in each repeat contains positively changed residues which are responsible for the voltage sensor activity. Using [^H]ludopamil as a photoaffinity probe, the phenylalkylamine binding site on purified skeletal muscle calcium channels was determined.
After digestion with proteolytic enzymes, the radioligand remained bound to the subunit of the skeletal calcium channel. The binding site on the subunit was mapped with labeled peptide fragments with antipeptide antibodies, and the
phenylalkylamine binding site was located in segment S 6 of domain IV on the C terminal cytoplasmic loop (Nakayama et al.,
1991; Catterall and Striessnig, 1992; see Figure 1.3). Using a similar antibody mapping procedure, the dihydropyridine
27 B
Figure 1.3 (A) Schematic Representation of L-type Calcium Channel (Catterall, 1991); (B) The "Domain interface" model suggests that the dihydropyridine binding site is formed at the extracellular end of the interface between repeat III and IV (Nakayama et al., 1991; Catterall aind Striessnig, 1992).
23 binding site on subunit of calcium channel was labeled with
either azidopine or diazipine, and was located in segment S 6 of domain III and IV from the extracellular surface calcium channel (Striessnig et al., 1991; Catterall and Striessnig,
1992; see Figure 1.3).
The channel forms a water-filled pore and allows calcium to move in the direction of its electrochemical potential gradient. It has been proposed that there are three conformational states of calcium channels, resting (closed) , open, and inactivated (closed) states. The calcium channels are assumed to contain activation (A) and inactivation (I) gates that are moved or altered by the potential difference across the membrane and by drugs so as to open or close the transmembrane flux of calcium. The channels open only when both the proposed A and I sites are open, and close when either the A or I sites are not open. At rest, two Ca^* are bound to binding sites in the pore of the calcium channel.
Depolarization causes a calcium channel to open and Ca^* to dissociate from its binding site and enter the cytoplasm. As the intracellular calcium concentration increases, the calcium channel is inactivated presumably due to a conformational change of the calcium channel caused by calcium binding to the
29 sites at the intracellular mouth of the channel (Godfraind,
1983; Tsien, 1983; Tsien and Tsien, 1990; Kuriyama et al.,
1995).
1.3 • a-Adrenergic receptors
1.3.1 Classifications with functional studies
Adrenergic receptors (ARs) interact with the endogenous catecholamines, epinephrine and norepinephrine, which exert their physiological effects by direct binding with cell surface receptors of target tissues. Norepinephrine acts as a primary neurotransmitter released from postganglionic sympathetic fibers, whereas epinephrine and norepinephrine are secreted from adrenal medullary cells. The latter structure is embryologically and functionally analogous to sympathetic ganglia. In addition to their crucial roles in central and peripheral nervous regulation, these catecholamines are important in controlling cardiovascular function, airway smooth muscle reactivity and nutrient metabolism.
Pharmacological studies of the receptors activated by these molecules have resulted in the development of drugs to treat cardiovascular disease, asthma, nasal congestion, and various
30 CNS disorders (see Figure 1.4 for the structure of
catecholamines).
The early explanation of the existence and
subclassification of the ARs was initially conducted in an
indirect fashion. Specific receptor moieties, termed as adrenoceptors or adrenergic receptors, were first postulated based on in vivo and in vitro pharmacological studies of the
intrinsic physiologic effects provoked by a-adrenergic agents.
Ahlquist (1948) initially proposed a division of adrenergic responses into two general classes called a- and 13- adrenoceptors. Based on different effects of structurally related agonists in different tissues, Of-AR were characterized
OH OH
HO— ^y \y CHrw CHg NHgNIM„ HO— HO—/ ^ \ y CH CH CH, CHa NH NH CH 3
Norepinephrine Epinephrine
-CH— CHa NH CH ^ C H a Isoproterenol
Figure 1.4 Chemical structures of some important catecholamines. The asterisk (*) denotes the presence of an asymmetric center.
31 by the rank order of potency of sympathetic agents which was epinephrine > norepinephrine > isoproterenol. In contrast, the
P-AR receptor had a potency rank order of isoproterenol > epinephrine i norepinephrine.
The concept of two discrete types of AR responses led to the development of highly selective and potent antagonists which supported the differentiation of ARs, and promoted the division of each receptor family into additional subtypes.
Subsequently, P-AR responses were divided into two types, 3i- and Pg-ARs (Lands et al., 1967). This subclassification was based on the rank order of agonist potencies of isoproterenol, epinephrine and norepinephrine. p^-ARs displayed approximately equal affinity for epinephrine and norepinephrine whereas P;-
ARs had a higher affinity for epinephrine than that of norepinephrine. Lands and co-workers found that p^-ARs were distributed in adipose tissue and cardiac muscle while Pj-ARs were found mostly in blood vessel and trachea. Activation of
Pi-ARs accelerated lipolysis and heart rates whereas P 2 -ARs stimulation caused relaxation of aorta and trachea smooth
muscle. More recently, a P3 -AR or atypical P-AR has been described (Kaumann et al., 1989; Zaagsma et al., 1991) in adipose tissue, GI tissue, skeletal muscle and heart. This
32 receptor is activated by unique classes of agonist molecules
and the response is resistant to competitive antagonists like
propranolol. All subtypes (3i-,32~ and Ps-AR) have been cloned
cuid recombinant cDNAs expressed in host cells (Kenakin, 1996) .
.. The existence of distinct a^- and aj -AR subtypes was
proposed according to differences in the relative potency of
selective agonists and antagonists (Langer, 1974; Berthelsen
et al., 1977). a^-ARs were stimulated by phenylephrine and
inhibited by prazosin whereas of;-ARs were stimulated by
clonidine and blocked by yohimbine or rauwolscine.
1.3.2 Subclassifications of ax- and % -ARs with radioligand
binding assays
As of the 1990's, there appeared to be at least six different a-AR subtypes designated as ax^, Oxa, Oic, Ofa*/ ofaa and
« 2 C (Bylund, 1992) . Classification of the subtypes of a^- and
Oj-AR is attributed essentially to the application of radioli gand binding assays. Using radioactive agents bound to
isolated membrane bound receptors, this approach makes it possible to quantify the interactions between the drugs and receptors and to minimize the effect of tissue diffusion barriers complicating studies in isolated tissues (Minneman et
33 al., 1980).
The existence of «i*- and «ig- sub type s was first proposed by Battaglia et al. (1983) and subsequently extended by Morrow et al. (1986) based upon radioligand binding studies in rat cerebral cortex. While prazosin had similar affinity for the
two « 1 -AR binding sites, the «-antagonists, WB4101 and phentolamine, were approximately 40- and 20-fold more potent
at the « 1A site as opposed to the a ig site. These two «j. subtypes were corroborated by Johnson and Minneman (1987) who demonstrated that only «ig-AR binding sites in different tissues were inactivated by chloroethylclonidine (CEO , a site-directed alkylating agent. Further, the subtypes distin guished by CEC were the same as those defined with the use of
WB4101 (Minneman et al., 1988; Wilson and Minneman, 1989) . An excellent correlation was observed between the percent of «ig receptors determined by these two methods, CEC sensitivity and
WB4101 affinity (Bylund, 1992). Consequently, was recognized as a subtype of a-AR that has high affinity for antagonists WB4101, niguldipine and methylurapidil but is insensitive to CEC inactivation. In contrast, «ig-ARs subreceptor exhibits properties opposite to those described above for the «i*-subtype (Bylund, 1992; Bylund et al., 1996) .
34 Subtypes of (Xz-AR are usually labeled by [^H] -yohimbine as well as a more selective isomer [%] -rauwolscine (Perry et al., 1981). Both ligands exhibit relatively high nonspecific binding and have low specific receptor binding. Norepinephrine and epinephrine have similar affinities at both Og-AR subtypes. Ogs-AR subtypes were recognized by several otz adrenergic antagonists including prazosin and ARC239, which were observed to have significantly different affinities in inhibiting [^H]yohimbine binding in rat lung, rat cerebral cortex, rat kidney and NGIOB cell line. By contrast, the partial agonist, oxymetazoline, has a higher affinity for the
(XzA- subtype which is distributed abundantly in human platelets and the HT29 cell line (Bylund, 1988). More recently a third pharmacological subtype, ofjc-AR, has been identified by radioligand binding studies in opossum kidney (OK) cell line
(Blaxall et al., 1991; Murphy et al., 1988). [^H] Idazoxan, in addition to binding to Oj-AR, labels other sites termed imidazoline preferring binding sites (Boyajian et al., 1987a) .
This complicates data interpretation when idazoxan is used as an antagonist or radioligand for a-AR investigations. The fourth subtype, designated as ago, has been found in bovine pineal, rat pancreatic islet cell tumor (RINmSF) (Simonneaux
35 et al., 1991; Michel et al., 1989a; Retnaury and Paris, 1992;) .
This subtype has a lower affinity for [%] rauwolscine than the other subtypes and, like the adrenoceptors, a low affinity for prazosin, spiroxatrine and ARC 239 (for review, see
Bylund, 1995; Bylund et al., 1996; Kenakin, 1996).
1.3.3 Second messenger cascade
It has been suggested that adrenoceptors be divided into three families, the a^-ARs, a % -ARs, and the S-ARs (Bylund,
1988) . Each of the three receptor groups is associated with a specific second-messenger system. Cfi-ARs increase intra
cellular calcium concentrations, whereas 0 (2 -ARs and 6 -ARs inhibit and stimulate adenylyl cyclase, respectively. There do not appear to be differences in signal transduction mechanisms between individual subtypes in each of the three major families (Bylund et al., 1996) . Activation of all known Oi-AR subtypes is accompanied by an elevation of intracellular calcium, which is mediated by a G-protein coupling leading to activation of phospholipase C (Minneman et al., 1991). In
turn, this causes the release of inositol 1 ,4,5-triphosphate
(IP3) and diacylglycerol (DAG) from phosphatidylinositol
(PIP;). IP3 stimulates the release of sequestered stores of
36 calcium, leading to an increased concentration of cytoplasmic
calcium. The at^-AR subtype activates calcium influx from an
extracellular site while the a^a-AR subtype activates calcium
release from sarcoplasmic reticulum (Minneman, 1988) . It has
become clear that in at least some systems, Oi^-ARs can also
increase inositol phosphate formation and release stored
intracellular calcium (Wilson et al., 1990). «g-AR ligands
inhibit adenylate cyclase (AC) by causing dissociation of the
inhibitory G-protein, Gi, into its subunits, hence lowering
adenylate cyclase activation and cytoplasmic cAMP levels
(Limbird, 1974) . Activation of ofjA- and « 2 3 -ARs subtypes
inhibit adenylate cyclase activity.
1.4 Imidazoline preferring receptors (ZPR)
1.4 .1 Evidence for IPR existence
Evidence gathered over the past ten years (for review,
see Lehmann efc al., 1989; Michel et al., 1989; Atlas et al.,
1991; Regunathan et al., 1996) supports the existence of a unique class of imidazoline compounds, which appear to
interact with unique receptors. These receptors have been
termed the imidazoline preferring receptors (IPRs) . Because of
37 OH
N— CH. N‘ H
Phentolamine Tolazoline
CH.
Tetrahydrozoline Naphazollne
ÇH 3 /
CH 3 PI H
Xylometazollne
N— CHa
Antazollne
Figure 1.5 Chemical structures of selective imidazolines. The imidazoline(s) in the upper panel possess a-adrenoceptor antagonist activity; the middle panel are decongestants; and the lower panel is a histamine inhibitor.
38 their wide spread use in basic research experiments and clinic, imidazoline compounds such as clonidine, tolazoline, phentolamine and moxonidine, are receiving increasing attention for interaction with IPRs in central and peripheral nervous systems (Papp et al., 1994; Hieble et al., 1995).
Research leading to the differentiation of imidazoline preferring receptors and related ligands from adrenoceptors has been based on the hypothesis that catecholamines
(norepinephrine, epinephrine) and imidazolines (oxymetazoline, tolazoline, clonidine) activate the adrenoceptors in a different manner (Ruffolo et al., 1977; Boyajian et al.,
1987a; 1987b) . This was thought to be due largely to heterogeneous interactions of catecholamine ligands and
imidazoline ligands with the of-adrenoceptors, especially « 2 "
ARs (Atlas at al., 1991).
The pharmacological properties of imidazolines have been well documented. Most imidazolines are active at a-ARs and inactive at 3-ARs. Also, they do not engage as a substrate for the uptake or affect the transport of other types of sympathomimetics (Bowman, 1980a) . Although some may serve as a-ARs antagonists (e.g., tolazoline, phentolamine; Figure
1.5), their effects are directly related to an interaction
39 with a-adrenoceptors in peripheral tissues. The long duration of the vascular responses to imidazolines makes them applicable as topical nasal decongesteints or conjunctival decongestants (e.g., naphazollne, tetrahydrozoline, xylometazollne; see Figure 1.5) . In addition, some of the imidazoline ring-containing compounds can even act as histamine antagonists (e.g., antazollne; Figure 1.5).
On the central nervous system, most of the imidazolines exert hypotensive effects presumably through presynaptic Oj-
ARs, producing sedation and sometimes other symptoms such as dry mouth (Bowman, 1980b). A typical imidazoline compound with agonist activity on a-ARs within the central nervous system, clonidine, is clinically used to lower blood pressure. This was thought to be due largely to a reduction in the heart rate and cardiac output, which in turn decreases mean arterial pressure (Kobinger, 1978). A prejunctional receptor activation on a noradrenergic neuron has been suggested to produce this blood pressure lowering effect (Bowman, 1980b). However, this theory has been questioned, since neither neurotransmitter depletion with reserpine (Kobinger, 1978) nor destruction of
terminals with 6 -hydroxydopamine (Dollery et al., 1973) substantially attenuates the ability of clonidine to decrease
40 blood pressure. The centrally mediated blood pressure lowering effect of clonidine is blocked by selective cfg-AR antagonists, yohimbine and rauwolscine. These antagonists caused a dose- related blockade of the hypotensive effect produced by intravertebral administration of clonidine to anesthetized cats (Timmermans et al., 1981). Activation of (Xg-ARs produces multiple physiological responses in the central nervous system, and to a lesser degree, in some peripheral tissues.
Stimulation of (Xg-AR in the peripheral tissues causes human platelet aggregation, inhibition of lipolysis, inhibition of acetylcholine release in guinea pig ileum, inhibition of insulin secretion of pancreatic cells and inhibition of the field-stimulated contraction of rat vas deferens (Starke et al., 1989).
Extensive evidence has demonstrated that some imidazolines interact with non-adrenergic receptors, while other imidazolines function as classical agonists and antagonists at a^- and (Xg-adrenoceptors. The peripheral smooth muscle contraction or central neuroinhibition of imidazoline a-agonists such as clonidine, cirazoline, oxymetazoline and UK
14,304 can be blocked by imidazoline a-antagonists such as prazosin or rauwolscine (Ruffolo et al., 1982; Ruffolo et al.,
41 1985; Oriowo et al., 1990). Imidazoline a-agonists can be relatively nonselective for a^- and a % -adrenoceptors (e.g., oxymetazoline) , selective for aj.-adrenoceptors (St 587, cirazoline) or Og - adrenoceptors (clonidine, UK 14,304) . Hence, these imidazolines may interact at heterogeneous binding sites where they initiate responses via binding to the domain related to both a-ARs and imidazoline receptors. From the receptor point of view, multiple effects of clonidine analogs on the central nervous system could not be mimicked by the
"classical" aj-AR agonists, epinephrine or norepinephrine.
These findings suggest that ag-ARs are a heterogenous population of receptors which can be further subdivided into novel subfamilies by the use of imidazoline and catecholamines.
One of the early studies performed by Ruffolo and co workers (1977) demonstrated that desensitization of the rat vas deferens to the contractile action of oxymetazoline, an imidazoline, produced a nearly complete blockade of contraction induced by tetrahydrozoline, another imidazoline, yet had no effect on the contractile response to the phenethylamine agonists such as norepinephrine, methoxamine or phenylephrine. This finding suggests that these two chemical
42 classes of a -adrenoceptor agonists may interact at different
receptor sites. In the pithed rat, different stereo-structure
activity relationships were observed between phenethylamines
and imidazolines on blood pressure lowering effects (Ruffolo
et. al., 1983). Bousquet et al. (1984) noted that various
imidazolines, irrespective of their affinity and efficacy at known a-AR subtypes, lowered blood pressure upon direct injection into nucleus reticularis later is of brainstem cardiovascular area, whereas catecholamines were inactive.
1.4.2 Radioligand binding assays of ZPR
Numerous articles regarding imidazoline receptor identification have been published using the radioligands,
[^H] clonidine, [^H] p-aminoclonidine ([PAC]) aind PH] idazoxan
(see Figure 1.6).
A. PH]-Para-aminoclonidine and PH] -Clonidine. The earliest direct evidence for the existence of IPRs was reported by
Emsberger and colleagues (1986; 1987) using [^H] PAC in bovine brain. Upon injection into bovine rostral ventrolateral medulla, the antihypertensive area and also the ligand binding sites labeled by clonidine (Tibirica et al., 1991) and
43 rilmenidine (Gomez et al., 1991), separated 70% of the total
[^H]PAG binding sites were displaced by norepinephrine and
a-2 A ,B ,C .D II
RadioligcUids [miRauwolscine [*H] Clonidine [*H] Idazoxan C'H] Idazoxan [“*11 p-Iodoclonidine [’H] Clonidine [’H] Moxonidine [“*11p-lodoclonidine
Ligand affinity Imidazolidine z Clonidine = Cirazoline > Imidazoline > Phentolamine = Idazoxan = BFI > Phenylathylamine • Idazoxan » Naphazoline = Imidazoles Rilmenidine = Tolazoline Moxonidine > (Aicuiabenz • Efaroxan ■ Clonidine ■ Epinephrine = Epinephrine = Rauwolscine Rauwolscine
Endogenous ligetnd Epinephrine Agmatine? Agmatine? Agmatine? cCDS cCDS
G-protein coupling Yes Yes No
Subcellular fraction Plasma membrane Plasma membrane Mitochondria
Signaling mechanism Adenyl cyclase Prostaglandin MAO inhibition? inhibition release, K channel
Distribution Widespread Brainstem reticularis Forebrain; circum- Neurons z glia kidney, chromaffin ventricular organs, cells, PC12 cells, glia > neurons, platelets kidney, liver, adrenal medulla
Modified form Emsberger et al. (1995) and Regunathan and Reis (1990 .
Table 1.4 Comparison of « 2A-Df Ii and 1% binding sites
44 Cl
Cl H \ , H
Clonidine p-Aminoclonidine
idazoxan
Figure 1.6 Imidazolines used for identifying imidazoline preferring sites in radioligand assays. [^H]Clonidine and [^H] />aminoclonidine labele Ir site, and [4f]idazoxan labeles I2* sites.
other phenethylamines. The remaining 30% binding sites were displaced by imidazoline a-adrenoceptor agonists as well as certain molecules bearing an imidazole moiety, e.g., histamine and cimetidine. Similar sites have also been demonstrated in rat kidney (Emsberger et al., 1988); however, these [^H] PAC binding sites are not recognized by guanidino (e.g. guanabenz)
(Emsberger et al., 1988) or by benzazepine compounds (SK&F-
86466) (Emsberger et al., 1988) or by the diuretic drug,
45 amiloride (Reis et al., 1992). Using [^H] clonidine, it was observed that the binding to human nucleus reticularis lateralis (NRL) membranes was clonidine-specific and catecholamine-insensitive. Thus, the human NRL region provides the first model of a homogenous population of IPRs (Bricca et ai., 1989).
Recent reports have demonstrated that [^H]moxonidine, an imidazoline agonist radioligand, can label the sites identified by [%] FAC and [%] clonidine in brain as well as in renal medulla and adrenal chromaffin tissue (Emsberger et al., 1993; 1994). The drug has been received intensive research with an agonist mechanism, and used in clinical trials as an antihypertensive agents (Emsberger et ai., 1993;
Ollivier et ai., 1994).
B. H] -Idazoxan. The discovery of high-affinity binding sites for [^H] idazoxan, a radiolabeled imidazoline and cXg- adrenergic antagonist, to non-adrenergic receptors in cerebral cortex, rather than in medulla, suggested the existence of an
"idazoxan receptor" which was distinct from a-ARs (Boyajian et ai., 1987a; 1987b). [^H] Idazoxan labels three-fold more sites than [^H] rauwolscine (an isomer of yohimbine) in several areas
46 of the rat brain. Hence, these results suggest that
[^H] idazoxan recognizes a heterogenous population of Oj-AR
sites, one of which is selectively identified by
[^H] rauwolscine. Subsequent studies in diverse tissues from different species generally support the finding that idazoxan binds to more sites than rauwolscine in these tissue preparations (Lehmann et al., 1989) . The idazoxan sites share
some unique features for these tissues: 1 ) a low affinity for catecholamines (norepinephrine, epinephrine) , or for non
imidazoline « 2 -adrenoceptor antagonists (yohimbine,
rauwolscine) ; 2 ) clonidine is weak or inactive in inhibiting idazoxan binding to these sites; and 3) cirazoline auid guanabenz are potent displacing inhibitors of these sites.
1.4.3 IPR nomenclature and receptor characterization
Based on the analysis of the findings exhibiting apparent heterogeneity of imidazoline recognition sites, subclassification of imidazolines receptors has been postulated (Michel et al., 1989b). In the light of recent studies, it is now evident that IPRs, like most other receptors, exist as multiple subtypes. A uniform nomenclature was proposed for the two major subtypes of imidazoline
47 receptors, I^- sites and I % - sites, based on the ligand affinities (Michel et al., 1992). sites are labeled with nM affinity by clonidine analogs whereas Ig- sites have tnM affinity for clonidine and are usually labeled by
[^H]idazoxan. By Michel's standard, the conception for "I" was designated as all those potential ligands at imidazoline sites including not only imidazolines, imidazoles and imidazolidines but also such related structures as guanidines and oxazolines.
Therefore, extensive structure activity relationships for imidazoline ligands may be of assistance in understanding ligand-receptor interactions at imidazoline receptors and in exploiting novel imidazoline ligands.
The rank order of potency for ligands on imidazoline subtypes has also been determined recently (Reis et al.,
1992) . Clonidine-preferring subtypes (or I^-) exhibit the rank order of potency for displacing agents (K^) of phentolamine > para-aminoclonidine > idazoxan > cimetidine > imidazole 4- acetic acid >> cirazoline >> amiloride. The other, an idazoxan-pref erring subtype (Ig-) has a rank order for displacement of [^H] idazoxan of cirazoline > idazoxan > naphazoline > amiloride > clonidine >> imidazole 4-acetic acid. Furthermore, the subtype is subdivided on the basis
48 of whether it has high da*-) and low (Igg -) affinity for
amiloride (also see Table 1.4).
The clonidine-preferring site has been well characterized
in a discrete area within the medulla of several species,
including humans. A sympatho-inhibitory action producing a blood pressure lowering effect is seen when clonidine and other imidazolines are locally injected into this portion of
the brain. The relative contribution of imidazoline and « 3 - adrenoceptors to the cardiovascular actions of centrally acting antihypertensive drugs when administered systemically to experimental animals or humans has yet to be established.
The idazoxan-pref erring site has also been found in a variety of other tissue sites, including the cerebral cortex, platelet, adipocyte and renal tubular epithelial cell (Michel et ai., 1992) . No functional effect has yet been attributed to activation of this site.
Multiple research tools such as purification, selective antisera, photoaffinity probes and cloning have been employed to characterize the nature of imidazoline receptors. Parini et al. (1989) reported that non-catecholamine imidazoline
receptor sites can be physically separated from 0 (2 -ARs in rabbit kidney by heparin-agarose or lectin affinity
49 chromatography, and appear to be monomeric, non-glycosylated
proteins with apparent molecular masses of 50-66 kD. Reis et
al. (1992) reported that sites can be separated and
biologically expressed in the astrocytes of bovine chromaffin
cells. By using immunocytochemical analysis, antisera against
the partially purified protein selectively stained astrocytes
and some neuronal terminals and cell bodies which contain
imidazoline binding sites. Electromicrographs have determined
that some binding sites are associated with mitochondria,
supporting previous data demonstrating co-purification of 1 %
sites with mitochondrial marker enzymes; sites, however, are not found extensively in mitochondrial fractions and are pharmacologically distinct (Michel et al., 1989b).
1.4.4 Endogenous ligands for IPRs
Isolation auid partial purification of an endogenous clonidine-like substance from rat brain (Atlas et al. 1984b) and bovine brain (Atlas et al. 1984a) were reported. Tissue extracts were able to inhibit [^H] -clonidine binding in a reversible and competitive manner and were named clonidine- displacing substance (CDS) . CDS has long been considered an endogenous ligand for I sites and has a reported selectivity
50 of II > 0 2 -adrenoceptors ^ I 2 - It was reported that CDS is a low molecular weight substance of 588 daltons (Atlas et al.,
1984a; 1984b), ninhydrin and fluorescamine negative, and heat and acid resistant (Atlas et al., 1987) . Except for its action in brain where it increased blood pressure upon microinj ected into cat rostral ventrolateral medulla (Bousquet et al.,
1984) , the extract mimicked the action of clonidine in peripheral tissues. Unfortunately, CDS is merely an extract and not a chemically defined pure substance. Some of the earlier CDS preparations may have contained GABA or glutamate, which may be responsible for their opposing effects on blood pressure when administered centrally (Michel et al., 1992).
Questions have arisen about the early functional results produced by the undefined active ingredient in CDS.
Nevertheless, numerous reports have shown that the endogenous
CDS but not catecholamines can competitively bind to non catecholamine receptive sites labeled by [%] -idazoxan or [%] -
PAC (Atlas, 1991) . This adds proof that the catecholamine and imidazoline binding sites are distinct entities. More recently, Li et al. (1994) demonstrated that a substance
displacing [^H]PAC from 0 f2 -adrenergic receptors of membranes of rat cerebral cortex with CDS-like activity was present in
51 bovine brain. With a procedure using HPLC, elution and mass spectroscopy, the CDS was determined as agmatine (or decarboxylated arginine) with less than 300 Da.
1.4.5 Mechamistic aspects of IPR
Zonnenschein and co-workers (1990) observed that Gpp(NH)p and sodium did not modify the affinity of clonidine, UK 14,304 or guanabenz, strongly suggesting a mode of action of signal transduction for these ligauids at the imidazoline receptor that is unrelated to a G-protein coupling mechanism. Reis et al. (1992) also reported that Ij sites are mediated not through a G-protein but in some manner are linked to Ca^* fluxes, which may result from an interaction of IPRs with K* channels and or an Na*/H* exchanger. In addition to a significant inhibition of specific binding by 4-aminopyridine
(IC5 0 = 0.38 mM) , binding of [^H]-idazoxan to imidazoline receptors was observed to be sensitive to monovalent ions
which interfere with potassium channels, e.g., NH 4 * and Cs*
(Zonnenschein et al., 1990). In kidney membranes, K* was shown to increase the dissociation rate of idazoxan from 0.13 to 0.33 min*"-, suggesting that K binds to an allosteric site of the imidazoline receptors (Coupry et al., 1989).
52 In summary (also see Table 1.4) , during the last two
decades, substantial progress has been made in the
characterization of the existence, properties and
pharmacological role of the imidazoline receptors (for review,
see Bylund, 1995; Hieble and Ruffolo, 1992; 1995; Emsberger
at al., 1995; Regunathan and Reis, 1996). The general
hypothesis for imidazoline receptors has been on the basis of
heterogeneity of a-adrenoceptors ligeinds, especially otz-
adrenoceptor ligands interacting with the receptors in a
different manner or sites. Nonadrenergic binding sites
recocfnizing molecules that contain an imidazoline ring or
structurally related moiety have been referred to by a variety
of names including imidazoline guanidinium receptor sites
[IGRS] (Tesson et al., 1991), imidazoline preferring receptors
(IPR), nonadrenergic idazoxan binding sites (NAIBS), idazoxan
receptors, and imidazoline receptors. To date,
subclassification of imidazoline receptors have been
standardized, with sites identified by [%]p-aminoclonidine or
[%] clonidine being designated as Ij. receptors and those
identified by [^H] idazoxan as 1% receptors. Early demonstrations for the existence of imidazoline receptors
include a functional (Bousquet et al,, 1984) and radioligand
53 binding assay (Emsberger et al., 1986; 1987; 1988) in which imidazolines, but not catecholamines, were shown to interact with the recognized brain region responsible for hypotensive activity. Endogenous substances were isolated and identified to displace clonidine labeled sites (for review, see
Regunathan and Reis, 1996) including peptide (Atlas and
Burstein, 1984a; 1984b) and non-peptide (Li et al., 1994) substances. Perhaps the most important recent development was the identification of agmatine, a non-peptide, as a potential endogenous ligand for the imidazoline receptors. Agmatine
interacted with subreceptors of imidazolines as well as « 2 - adrenoceptor sites. Along with other evidence, it is still difficult to dissociate imidazoline receptors from cXg- adrenoceptors both pharmacologically and functionally. No antagonist that selectively blocks subtypes of imidazoline receptors is available. Recent reports have challenged the existence of imidazoline receptors (MacMillan et al., 1996;
Link et al., 1996). In these studies, however, catecholamines were shown to be critical in producing hypotensive effect by an interaction with brain regions previously shown to be ineffective to catecholamines. Point mutation (Asp''® to Asn''®) of UjA-adrenoceptors in mice also supported the principal role
54 of an antihypertensive effect mediated by 0 3 *-adrenoceptors.
Although there are subtle differences in the distribution of imidazoline receptors and Oj-adrenoceptors within the central nervous system, most central nuclei contain both receptor classes. In many cases, stimulation of receptors and oj- adrenoceptors produced identical physiological responses.
Despite this controversy, drugs having agonist activity at Ii receptors are currently in clinical use as antihypertensive
(Papp and Ollivier, 1994; Haxhiu et al., 1994; Ollivier and
Christen, 1994) and anti-arrhythmic (Lepran et al., 1994) agents, with central Ij. receptor activation postulated as their primary mechanism of action. One of these drugs being extensively studied is moxonidine. This drug is identified as an Ii agonist in the central nervous system, and exhibits less adverse effects than clonidine (Emsberger et al., 1993;
Lepran et al., 1994).
1.5 2-(4 '-Zsothiocyanatobenzyl) Imidazoline [IBI] analogs
Tolazoline analogs substituted on the phenyl ring are direct acting a-adrenergic agonists in rabbit ileum (Struyker et al., 1974; Sanders et al., 1975; Timmermans and Van
Zwieten, 1977) . Hydroxylation of tolazoline at the meta
55 Tolazoline 2-(4'-lsothiocyanatobenzyl) imidazoline
Figure 1.7 Chemical structures of isothiocyanato (NOS) derivative of tolazoline.
position (Ruffolo et al., 1979a) or at the 3- and 4-positions of the aromatic ring (Ruffolo et al., 1979b and 1980; Banning et al., 1984; Rice et al., 1987) increase agonist potency.
Substitution of sin isothiocyanato moiety on the phenyl ring produces compounds with a sustained action, and related phenyl isothiocyanato derivatives which are used for analyzing amino acids in plasma, urine and food samples (Cohen and Strydom,
1988) . p-Isothiocyanato- and p-methylisothiocyanato-clonidine, alkylating derivatives which are structurally similar to tolazoline, produced non-parallel rightward shifts in the concentration-response curves of norepinephrine and a significant reduction of the maximum response to norepinephrine in rat aortic strips; these effects are
56 characteristic of an irreversible antagonism (Decker et al.,
1983) . With these data in mind, it was thought that preparation of tolazoline analogs with an isothiocyanato group might be useful tools for investigating adrenergic receptor- related mechanisms.
Sengupta et al. (1987) and Venkataraman et al. (1989) found that the 4-isothiocyanato derivative of tolazoline, namely, IBI [2-(4 '-isothiocyanatobenzyl) imidazoline] (Figure
1.7) produces contractions of rat aortic strips with an ECgo value of 16.3 /xM and with a maximum contraction equal to tolazoline and phenylephrine. Pretreatment with IBI (1 /xM) shifted the concentration-response curve of phenylephrine to the right with a reduction in the maximum. Shams et al. (1991) also reported that the IBI agonist activity, unlike tolazoline, was unrelated to stimulation of a-AR. The effects of a-adrenoceptor antagonists (phentolamine, prazosin, SK&F
104078) on the agonist activities of synthesized tolazoline analogs, 2-(4'-aminobenzyl) imidazoline (ABI) and IBI (Figure
1.7) were examined on rat thoracic aorta. The contractile responses to ABI was reduced by an approximate 100-fold in the presence of the a-AR antagonists, whereas the same pretreatments did not affect the responses to IBI. Thus, the
57 addition of the 4'-isothiocyanato (4'-NCS) group to tolazoline produces a compound which is insensitive to a-AR antagonist pretreatments in rat aorta.
IBI inhibited epinephrine-induced aggregation with a Kg of 20 in aspirin-treated human platelets (Venkataraman at al., 1989). Tolazoline and ABI were selective competitive antagonists of epinephrine at az^-platelet adrenoceptor sites.
IBI differed from the these two imidazoline compounds by exhibiting nonselective inhibitory actions at various sites in the pros t agiandin-dependent pathway of human platelet activation (Shams et al., 1991). In comparing the ability of these three compounds to antagonize primary wave aggregation induced by epinephrine, only the slope value in the Schild plot of IBI differed from unity. This finding suggests that the IBI inhibitory action may involve other sites in the platelet aggregation pathway. Only IBI blocked the aggregation responses to ADP (second wave only) arachidonic acid and
U46619 (a TXAj receptor agonist). Malondialdehyde formation (a by-product of arachidonic acid) induced by arachidonic acid was blocked by IBI but not by ABI. This evidence supports the proposal that IBI blocks arachidonic acid release, prostaglandin metabolism and the action of thromboxane A; in
58 human platelets. However, prolonged exposure of platelets to
IBI abolished the inhibition of epinephrine-mediated
aggregation. Thus, IBI is not useful as a specific ligand for
platelet cKg-AR since the blockade of epinephrine-induced
responses was not long-lasting, and because IBI inhibited
numerous sites in the prostaglandin-dependent pathway of
platelet activation (Shams et al., 1991).
Binding studies performed by Hussain et al. (1992)
provided evidence that IBI competes for [%] idazoxan binding
sites in brain. Non-adrenoceptor, imidazoline binding sites
comprised approximately 30% and 100% of the [%] idazoxan
binding with an approximate pK^ value of 6 in guinea pig
cerebral cortex and porcine renal cortex membranes,
respectively. In addition, a similar value of pDj was obtained
that IBI but not tolazoline increases the [^H] -cAMP level in
a concent rat ion - dependent manner in rat cortex. These results
suggest that IBI may interact with imidazoline preferring binding sites in guinea pig cerebral cortex and porcine renal
cortex. Further, the pathway involved in cAMP changes elicited by tolazoline and IBI may be different (Wilson, 1991; Hussain
et al., 1992), suggesting the importance of -NCS group in the
4 '-position of the phenyl ring.
59 1.6 Statement of the problems
The vast majority of a-adrenoceptor agonists can be divided into two main classes, the S-phenylethylamines and imidazolines. Although these two classes of drugs activate a- adrenergic receptor responses, both classes are blocked by antagonists such as phentolamine. Thus, it has been suggested that they may not interact with the receptor in the same manner. In general, the biological activities induced by phenylethylamines strictly adhere to the Easson-Stedman (1933) hypothesis whereas that mediated by imidazoline does not. The
OH OH OH
HO. HO. HO.
OH
(R)-{-)-Norepinephrine (S)-(+)-Norepinephrine
Figure 1.8 Schematic representation of the Easson-Stedman hypothesis.
60 potency of catecholamine adrenergic receptors was determine as
R(-) > S(-) = deoxy. Thus, it was hypothesized that there are three points of interaction for the optically active phenylethylamines with the receptor. These include a nitrogen atom common to all sympathomimetics, an aromatic ring whose potency is enhanced in the presence of meta- and para- hydroxyl groups, and a hydroxyl group in the R(-) absolute
configuration on the 3 -carbon atom of the ethyl amino side chain. In contrast, the optically active imidazolines do not follow the Easson-Stedman theory, and are proposed to interact with the receptor by a two point attachment. This proposal was supported by testing optically active 3,4-dihydroxytolazoline
(or catechoimidazoline) which exhibited an agonist potency sequence of desoxy > R(+) > S(-) (Ruffolo et al., 1983;
Nichols and Ruffolo, 1991). If there is a difference in the receptor site interaction of imidazolines and phenethylamines, then the steric requirements for these two classes of agonists should be different. Patil et al. (1991) postulated that certain active groups, steric specificities and electronic properties may contribute to the differences between these two chemical classes. More recently, the search for imidazoline ligands has intensified, and results suggest that I PR sites in
61 related tissues are recognized by a group of chemically diverse agents including imidazolines (e.g., clonidine, idazoxan, cimetidine, cirazoline), imidazoles (imidazole 4- acetic acid), guanidines (guanabenz, amiloride) and oxazoline
(rilmenidine) analogs. Studies of IPRs involve the use of radioligands such as [^H] idazoxan and [%] p-aminoclonidine.
This heterogeneity of I PR ligands implies a variation and complexity of the ligand-receptor interactions.
A recent report indicates that 2-(4'-isothiocyanato- benzyl) imidazoline (IBI), a tolazoline derivative with a 4'- isothiocyanato group (4'-NCS), displaces [^H]idazoxan from IPR sites in brain tissues; and that binding to these sites may be related to the activation of adenylate cyclase (Wilson, 1991).
Pharmacological studies of tolazoline derivatives have added to our understanding of the non-a-AR mechanisms of imidazoline compounds (Venka tar aman et al., 1989; Shams et al., 1991). IBI was documented to produce concentration- related contractions of the rat isolated thoracic aorta which were resistant to a-AR antagonists (phentolamine, prazosin,
SK&F 104087). In contrast, the contractile responses to tolazoline and a 4'-amino substituted derivative of tolazoline
(ABI) were blocked by a-adrenergic antagonists. It is
62 interesting to note that substitution of a 4 ' -NCS on
tolazoline altered the partial a-AR activity (Sanders et ai.,
1975, Venkataraman et ai., 1989) of tolazoline on rat thoracic aorta. IBI, like tolazoline, also possesses az^-AR antagonist activity in human blood platelets. Thus, modification of the benzene ring with an electrophilic group (NCS) enhances non-a- adrenergic responses in smooth muscle.
Although the IBI action is attributed to activation of non-a-adrenoceptors, as was shown in the thesis research (Lei,
1992), its relationship with calcium homeostasis and imidazoline receptors in smooth and skeletal muscle preparations remains unknown. Thus, more pharmacological studies are needed to characterize the specificity and underlying mechanism of action of the non-a-adrenoceptor actions exhibited by the prototype analog, IBI.
It is hypothesized that the unique fashion of IBI properties on muscles is due to the NCS group attached to tolazoline, and that its effects on tissues are dependent upon activation of calcium channels and elevation in intracellular calcium. The general objectives of these studies are to evaluate the receptor populations which mediate the stimulant actions of IBI agonist in smooth and skeletal muscle
63 preparations; and to investigate the structure activity relationships for interaction with non-a-adrenoceptor sites using a series of recently synthesized IBI analogs.
The specific aims of the dissertation are as follows:
1. To determine the specificity of IBI agonist activity in smooth muscle preparations using calcium channel antagonists and imidazoline receptor antagonists.
2. To determine the mechanism and site(s) of the IBI agonist activity in the involvement of calcium channels, using a calcium channel-specific radiobinding assay and measurement of intracellular calcium changes with a fluorescent indicator.
3. To further investigate the structure-activity relationship of IBI using analogs in smooth muscle of rat aorta preparations.
64 CHAPTER 2
MECHANISM OF 2-(4 •-ISOTHIOCYANATOBENZYL) IMIDAZOLINE [IBI]
ACTION: CALCIUM CHANNEL ACTIVATION PROPERTIES OF THE
CONTRACTILE RESPONSES IN SMOOTH AMD SKELETAL MUSCLES
2.1 Introduction
In order to investigate the structural requirements for imidazoline containing drugs to interact with imidazoline receptors or a-adrenoceptors, IBI vfas synthesized as an analog of tolazoline, and contains an electrophilic isothiocyanato
(NCS) group. With this affinity modification on a drug with such imidazoline-containing feature, IBI was expected to interact with sites of imidazoline preferring receptors (IPRs) and a-adrenoceptors (a-ARs) . In previous research on IBI (Lei et al. 1992; 1992), however, IBI-mediated contractile responses of rat thoracic smooth muscles were not dependent on a-ARs or IPRs. In other words, the presence of the NCS group on tolazoline altered its Oi-adrenoceptor-interacting action.
65 Additionally, such an IBI-like action in smooth muscles was neither linked to other receptor systems like serotonergic and histaminergic receptors or channels (Lei, 1992) . With the recent discovery of new ligands for IPRs for the evaluation of
IBI on smooth muscles, it is of importance to further characterize the mechanism of IBI action.
The present studies were designed to acquire evidence to support the hypothesis that IBI contracts smooth muscle via a calcium dependent mechanism. The experiments were undertaken to examine the importance of NCS group in IBI-like activities using tolazoline as a prototypical a-AR agonist on various pharmacological systems.
2.2 Materials and methods
2.2.1 Smooth muscle contraction studies
Source of chemicals. 2-(4'-Isothiocyanatobenzyl) imidazoline
(IBI) was synthesized by Drs. Meri Slavica, J. De Los Angeles and Duane D. Miller (Department of Pharmaceutical Sciences,
College of Pharmacy, The University of Tennessee, Memphis, TN
38136) and provided for these studies. Other chemicals used in the experiments and their sources are as follows :
66 phenoxybenzamine hydrochloride from Smith, Kline & French
(King of Prussia, PA) ; /-phenylephrine hydrochloride, neostigmine methyl sulfate, nifedipine from Sigma Chemical
Company (St. Louis, MO); cirazoline from Synthelabo Recherche
(Paris, France); (±)-verapamil hydrochloride, idazoxan hydrochloride, efaroxan hydrochloride, R(+)-Bay K 8644, S(-)-
Bay K 8644 from Research Biochemicals Inc. (Natick, MA) ; tolazoline hydrochloride from CIBA Pharmaceutical Co. (Summit,
NJ); moxonidine (free base) as a gift from Dr. D. Ziegler of
Solvay Pharma Deutschland GMBH (Hannover, Germany); tubocurarine chloride and physostigmine sulfate [eserine] from
Mallinckrodt Chemical Works (St. Louis, MO) . Chemicals used for preparing physiological salt solution were purchased from commercial suppliers and were of reagent quality.
Phenoxybenzamine was dissolved in 100% methanol as 0.01
M stock solutions and diluted in 0.9% saline; nifedipine and
Bay K 8644 isomers in 100% ethanol and further diluted with
0.9% NaCl. IBI and all other compounds used in this study were dissolved in 0.9% NaCl.
Animals. Male Sprague-Dawley albino rats ranging from 300-400 grams were purchased from Harlan Sprague-Dawley Inc.
67 (Indianapolis, IN). Animals were housed in a vivarium maintained at 25-26 °C, a relative humidity of 45-55%, an
alternating 1 2 hour light and dark cycle, and with free access to Purina rat chow water. The animal vivarium in the College of Pharmacy has been accredited by the American Association for the Accreditation of Laboratory Animal Care since 1968.
Preparation of Tissues. Animals were anesthetized by exposure to carbon dioxide and tissues from albino rat thoracic aorta were dissected and cut into either spiral strips. The segment of aorta removed was placed in a petri dish containing a physiological salt solution (PSS) containing (all in mM) :
NaCl, 118; KCl, 4.7; CaClj. HjO, 2.5; MgSO^. HjO, 0.5; NaHgPO^,
1.0; NaHCO], 25; glucose, 11; and EDTA, 0.03; pH 7.4 according to Venkataraman et al. (1989) . Fat and excessive tissues were trimmed away using fine curved scissors without stretching the tissues. Helical strips were prepared by cutting the segment at an angle of about 15-30 degrees, depending on the diameter of the aorta. Usually 2 to 3 strips of 2 mm width x 20 mm length were prepared from each rat thoracic aorta. One end of
each tissue was placed in a 1 0 ml water-jacketed tissue bath containing a physiological salt solution maintained at 37 ±
68 0.5 °C and aerated with a mixture of 95% oxygen and 5% carbon dioxide. The other end was connected via a silk suture to an
FT-03 isometric force displacement transducer which was coupled with the preamplifier of a Grass polygraph (Model 7) to record the changes in tension. Resting tension maintained
for helical aortic strips was 1 gm.
Experiments with Rat Thoracic Aorta. About one hour of equilibration was allowed for each tissue during which the solution was changed every 20 min. Each strip was primed with
1 phenylephrine with washout (3-4 times, 10 min intervals) .
Subsequently, a cumulative concentration response curve (CRC) to phenylephrine was established. The tissues were washed at least three times so that they were thoroughly relaxed and baselines were readjusted to the constant tension before the addition of test drug. Prior to the construction of the CRC to a given compound, some tissues were incubated with receptor- selective antagonists. In most instances, a matched control tissue was used and was allowed to rest for a similar period of time before addition of drug. Construction of CRCs to each compound followed the method by Van Rossum (1963) .
Upon completion of the initial control CRC of phenylephrine
69 and a washout period (volume replaced four times within 30 min) , strips were treated with vehicle or one of the following receptor antagonists: 30 nM phenoxybenzamine for 20 min followed with a washout (two times within 15 min) ; 1 /xM idazoxan for 60 min; 30 nM phenoxybenz amine for 20 min with washout (two times within 15 min) followed by addition of 1 fiM idazoxan for 90 min; or 30 nM phenoxybenzamine for 20 min with washout (two times within 15 min) followed by addition of 3 /xM cirazoline for 60 min.
Parameters measured in rat aorta studies were changes in the maximal tension in mg, the percent maximal response (%
Response, using the maximal contraction by 30 /xM phenylephrine
as 1 0 0 percent) and the molar drug concentration producing half of the maximal response (ECgo) .
Pharmacological Parameters and Statistical Analysis. Data are given as means and standard errors of the mean (SEM) . The percent maximal contraction was calculated using the maximal contraction by 30 /xM phenylephrine as reference. The median maximal concentration or ECgg value was determined graphically using a standard curve generated by GraphPad Inplot (GPIP,
GraphPad Software Inc., San Diego, CA) to interpolate sample
70 concentration curve from each individual experiment.
Differences in sample means between groups of data were tested
using the Student's t-test. A p value less than 0.05 was
considered as a statistically significant difference in sets
of observations between means of control and analog-treated
data (Mansfield, 1986; Tallarida and Murray, 1981).
2.2.2 Intracellular calcium detection with fura-2 AM
Chemicals. The acetoxymethyl (AM) ester of fura-2 (Molecular
Probes, Inc.) was dissolved to 2 mM stock in dimethyl-
sulfoxide (DMSO) used as a penetration enhancer that contains a solubilizing agent, pluronic acid (20%, w/v). The stock
solutions were kept at -15 °C until use, thawed and diluted with a physiological salt solution (PSS in mM: NaCl, 118; KCl,
4 .7 ; CaClj. H 2 O, 2.5; MgSO^. HgO, 0.5; NaH 2 PO4 ,1 .0 ; NaHCOj, 25; glucose, 11; EDTA, 0.3 mM; pH 7.4).
Tissue collection and preparation. Fresh warm pig hearts were obtained from slaughterhouse of Ohio Packing Company,
Columbus. The hearts were removed from regular adult pig, then immediately transferred into a chilled PSS, and transported to the lab. The right coronary artery was dissected free from
71 connective tissues, and cut open longitudinally along the
lumen. Then, the lumen intimai layer was scraped 2-3 times
with a knife to eliminate intimai cells. Tissue strips (0.5 x
1 x 6 mM) were prepared by vertical cutting and tearing away
form outer layer of the artery. Strips were then incubated in
PSS solution containing 5 (M fura-2 at 37 °C and aerated with
5% CO; in oxygen for 4-5 hours. The fura-2 loaded strips were
stored in PSS solution at 4 °C until use within 48 hours.
A strip was mounted horizontally on two metal hooks in a plexiglass chamber filled with 1 ml PSS buffer at 37 °C maintained by circulating fluid through channels surrounding
the chamber. One hook was fixed to a micrometer to adjust muscle length, and the other attached to a stationary
isometric force transducer (BG 10 Semiconductor Products,
Inc., Leonie, NJ) . By adjusting the micrometer and allowing equilibration under minimal passive tension, an initial tissue length of 3 mm was increased to optimal length of 5 mm (about
1.5-1.7 fold of the initial length). The optimal length was
equivalent to a resting tension about 1 gm at which a strip was maintained for the duration of the experiments.
Experimental setup for tension and intracellular calcium
72 measurement. Isometric force changes were recorded through a chart oscillograph (Gold 240OS Recorder), and also displayed on the video screen of a computer work station integrated in parallel.
The calcium sensitive fluorescent dye, fura-2 AM, was used as intracellular calcium indicator (Himpens et al., 1988;
Gilbert et al., 1991). Incubation of tissue strips for five hours loaded the fluorescent dye in to myoplasm. The ester form of the dye is able to penetrate through the sarcolemma into the myoplasm where it is hydrolyzed by non-specific esterases into the free acid. The free acid of the dye is impermeable to the sarcolemma membrane and sensitive in binding myoplasmic calcium. After a loading period of 4-5 hrs, the fura-2 solution was washed out at least 45 min before data collection. A glass window at the bottom of chamber was secured to the moving arm of an inverted microscope stage. A
40X fluorescence objective lens (Nikon) illuminating at the strip muscles from 0.2-0.3 mm underneath was connected to a fiber optic system (Photon Technology International [PTI] ,
South Brunswick, NJ) . With a spectrofluorometer, the system emits alternately (60 Hz) at the excitation wavelength of 340 nm and 380 nm, and measures the amount of fluorescence at 510
73 nm. The ratio of the fluorescence at 340 to that at 380
(F3 4 0 /F 3 8 0 ) represents the calcium signal. The fluorescence outputs and the raw force signals were converted to
digitalized data, recorded ( 2 0 points per sec) aind monitored simultaneously via a computerized PTI software.
Protocol. PSS solution was replaced with oxygenated buffer
every 1 0 min and allowed for about 2 0 min to reach equilibrium. In order to facilitate diffusion and to synchronous ly activate the muscle preparation, 500 fil of bath
solution was removed, mixed with 1 0 - 2 0 /xl drug stock solution, and then added back to the chamber to restore the final volume as 1 ml. Histamine (10 fM) and potassium chloride (KCl, 35 mM) were used to produce maximal contraction as references. IBI
( 1 0 0 (iM) induced contractions were tested by either post
administration with verapamil ( 1 0 0 ^M) at the contraction plateau, or pre-administration with verapamil (100 /xM, 20 min) used to suppress the potassium-induced contraction.
Data analysis and statistics. Data were analyzed with the computerized PTI program and expressed in values describing half time and peak wave for changes both in tension and
74 intracellular calcium concentration. A Student's t test was performed for statistical significance. All statistics are presented as means ± SEM.
2.2.3 Skeletal muscles contraction studies
Chemicals. The chemicals used in these experiments were obtained as following: tubocurarine chloride and neostigmine methyl sulfate form Sigma Chemical Company (St. Louis, MO); physostigmine sulfate [eserine] form Mallinckrodt; dantrolene sodium from Procter & Gamble Pharmaceuticals (Norwick, NY
13815) . All drugs used for preparing physiological salt solution were purchased from commercial suppliers and were of reagent quality. IBI and all other chemicals used in this study were dissolved in 0.9% NaCl.
Animals and Tissue Preparation. Male Sprague-Dawley rats were used as in these studies. Animals were sacrificed by decapitation and rat diaphragm from both sides were removed along with the phrenic nerve as described by Bûlbring (1946) .
The ribs were cut along the sternum to the upper thorax, then from the base of the sternum laterally to the flank, then to the upper thorax. The phrenic nerve was gently dissected free
75 from surrounding tissues and ligated with thread at a point near thymus where the nerve was cut. The semidiaphragm was then removed in a triangle shape, one edge with a rib and the opposite tip with the tendon portion of the diaphragm.
After nicely trimming the diaphragm with the rib equivalent to 1.3 cm, the tissue was attached to a Perspex Electrode
(C.F. Palmer, Bucks, England) . The two ends of the tissue were tied to the stable base of the electrode, and the nerve was centered in the electrode used for nerve stimulation. This phrenic nerve electrode was placed into a 30 ml bath containing a physiological salt solution (PSS, in mM: NaCl,
118; KCl, 4.7; CaClj. HjO, 2.5; MgSO,. HgO, 0.5; NaH gPO 4 , 1.0;
NaHCOj, 25; glucose, 11; and EDTA, 0.03; pH 7.4). The solution was maintained at 37 ± 0.5 °C and aerated with a mixture of
95% oxygen and 5% carbon dioxide. The tendon of the semidiaphragm was connected via a silk suture to an FT-03 isometric force displacement transducer (Grass Instruments,
Quincy, MA) coupled with the preamplifier of a Grass Model 7
Polygraph. A 500 mg resting tension was used and maintained for the duration of the experiment.
Rat semidiaphragms were allowed to stabilized for at least
1 hr before drug administration. During the equilibration
76 period, preparations were stimulated (Grass S9 Stimulator) indirectly through the phrenic nerve using about 5 V rectangular wave pulses of 0.4 msec duration at a frequency of
0.2 Hz, giving a basal twitch tension of 200-300 mg.
2.2.4 Binding assay in t-tubules of rabbit skeletal muscles labeled with [^H] PN 200-110/(+) [ % Bay K 8644
Materials. Chemicals and drugs were obtained as following:
(+) [%] PN 200-110 (SA = 85.1 Ci/mmol) and (+) [ % Bay K 8644 (SA
= 87 Ci/mmol) from Amersham Corporation or DuPont New England
Nuclear; PN 200-110 as a gift from Sandoz Ltd. (Basel,
Switzerland); nifedipine from Sigma Chemical Company (St.
Louis, MO); R(+)Bay K 8644, S(-)Bay K 8644 from Research
Biochemicals Inc. (Natick, M A ) ; naphazoline, tolazoline hydrochloride from CIBA Pharmaceutical Co. (Summit, NJ) .
Compounds synthesized and provided by Professor Duane Miller ' s group (Department of Pharmaceutical Sciences, University of
Tennessee at Memphis) were: 2-(4'-isothiocyanatobenzyl) imidazoline (IBI) , naphazoline-NCS (analog VI) and nifedipine-
NCS (analog VIII; see Table 3.1 for chemical structures).
All drugs were dissolved in double distilled water except for the S(-)- and R(+)- isomers of Bay K 8644, nifedipine and
77 nifedipine-NCS. These latter compounds were dissolved in 100% ethanol at 10"^ M and diluted in water.
Membrane isolation of t-tubules. T-tubules were prepared from skeletal muscle of rabbit as follows. Rabbits (about 1 kg) were killed by cervical dislocation. Fast white muscle was removed from the back and hindlimbs. T-tubule membranes of rabbit skeletal muscle were isolated at 4 °C as described by
Glossmann and Ferry (1985) , and by Nakayama et al. (1991) with modifications. The tissue was minced with scissors and homogenized using an Omni-Mixer at maximal speed in 4 times for 35 sec in chilled 20 mM NaHCOa containing 0.5 mM phenylmethane sulfonylfluoride (PMSF) . The homogenates were further homogenized with a Brinkman Polytron at 60% of the maximal speed for 7 times for 25 sec. An interval of 35 sec on cooled ice was allowed between homogenization bursts. After resuspension, the extract was filtered through cheesecloth and centrifuged at 15,000 x g for 20 min. The supernatant was then filtered through cheesecloth and centrifuged at 45,000 x g for
20 min. Thepellet was resuspended in ice-cold 50 mM Tris buffer (pH 7.4 at 37 °C) containing 0.5 mM PMSF and centrifuged at 45,000 x g for 20 min. After repeating this
78 step, the final pellet was resuspended in the same buffer to obtain a membrane suspension. Protein content was determined using the Pierce Protein Assay Reagent (Pierce Chemical
Company, Rockford, IL). The protein suspensions (10 mg/ml) were stored at -70 °C until use.
Radioligand binding. All binding assays were conducted in 50 mM Tris HCl buffer, pH 7.4, containing 5% glycerol at 25 °C.
Kinetic experiments were carried out to determine the specific activities (concentration versus dpm/mg bound) of the calcium channel preparation. Two types of radioligands were used to label calcium channels sites and included: 1) 0.3 nM of
(+) PH]PN 200-110 incubated with 0.1 mg/ml protein in a final volume of 4 ml per tube for 1 hr, and non-specific binding was defined by 1 [xM (±)PN 200-110; and 2) 20 nM of (+)PH]Bay K
8644 was incubated with 0.05 mg/ml protein in a final volume of 0.5 ml per tube at for 1 hr, and non-specific binding was defined by 10 /xM (±)PN 200-110. The experiments were conducted in yellow light to keep the dihydropyridines (nifedipine and nifedipine-NCS) from light sensitive decomposition. Incubation buffer (50 mM Tris-HCl pH 7.4 at 25 °C, 5% glycerol) and washing buffer (50 mM Tris-HCl pH 7.4 at 4 °C) were used for
79 each experiment. Each drug concentration was tested in duplicate. Proteins were recovered on filter discs under vacuum with four washes of cold washing buffer (4-5 ml/tube) on a Brandel Cell Harvester (Model: M-24). The harvester possesses 24 filter disc channels with 12 in one row that was convenient for duplicate samples. Filter discs (Brandel FP-
200, Whatman GF/C) were submerged and mixed in 5 ml scintillation cocktail (Scintiverse BD, Fisher Scientific).
The radioactivity ( [^H] ) was measured by a Beckman (Model LS-
7000) counter.
Data collection and analysis. Specifically bound ( + ) [^H]PN
200-110 and ( + ) [^H] Bay K 8644 were defined in each assay system as the amount of radioligand bound in the absence of competing ligand minus the amount in the presence of 1 /xM
(±)PN 200-110 or 10 /xM (±)PN 200-110, respectively. The initial curve estimates were determined using equilibrium binding data analysis (EBDA) software. The estimates were then analyzed by non-linear curve fitting to determine the Ki value for each lab using a PC-version of the radioligand binding program LIGAND (McPhenson, 1985). The values represent the dissociation constant of the competing ligand for the receptor
80 and was calculated by the equation of Cheng and Prusoff
(1973) . In this equation Ki = IC 5 0 /(1 + [D]/K j) , where IC 50 is the molar concentration of competing drug displacing 50% of receptor bound radioligand. The radioligand was incubated at certain molar concentrations ([D]), and the dissociation constant (K^) of (+) [^H] PN 200-110 was determined to be 0.48 nM from saturation binding assay (n = 3) . The values was utilized for the calculation of inhibitory constant (Ki) of competitors. However, the K^ of [^H]Bay K 8644 was not used since most competitors were so weak that Ki value determination was unnecessary.
2.3 Results
2.3.1 Effects of phenoxybenzamine pretreatment on the contractile responses of phenylephrine, tolazoline and IBI analogs in rat aorta
The studies in this tissue were designed to evaluate pharmacological properties of IBI on a-adrenergic receptors using a-adrenergic receptor antagonist pretreatment (Figure
2.2) on rat thoracic aorta. The time course of response, maximal contractions and ECgq values for phenylephrine, IBI and
81 tolazoline were compared (Table 2.1; Figure 2.1).
Phenylephrine, an «i-AR selective agonist, produced
concentration-dependent contractile responses on rat aorta.
The EC 50 of phenylephrine was 18 ± 5 nM and its maximal
contraction was 679 ± 32 mg (n = 27) . The maximal effect of
phenylephrine (30 /xM) was set as 100% with each strip. After
pretreatment with 30 nM PBZ, the contractile responses to
phenylephrine were abolished (1 nM - 30 /xM) . In this tissue,
tolazoline produced concentration dependent contraction with
an EC50 of 0.329 ± 0.120 /xM and a 59 ± 6 % maximal contraction
(n=4) in relation to that of phenylephrine, confirming that
this drug is a partial «i-adrenoceptor agonist. The
tolazoline-induced contraction was also abolished by pretreating the tissue with PBZ (30 nM) . In contrast, IBI
produced contractile responses with a lower potency (EC5 0 = 5
fiM) and a signif icsintly (p < 0 .0 0 1 ) increased maximal contraction (116 %) as compared to that of 30 phenylephrine. The IBI-induced contractile activities were not
significantly affected by PBZ pretreatment (EC50 = 5.16 /xM and maximal contractions 112 %) , suggesting that contractions of rat aorta by IBI were mediated via an a-AR independent mechanism.
82 The time course for each of the three treatments was also summarized in Table 2.1. IBI-induced contractile responses in rat aorta were characterized by a slow onset and a sustained contraction. The time duration for these drugs to reach maximal contraction (in min) was IBI (44) > phenylephrine
(8.2) > tolazoline (5.8). Thus, IBI's effect requires a 5.9-
and 8 .3-fold longer time to reach the maximal contraction plateau than phenylephrine and tolazoline, respectively.
Furthermore, the IB I-induced slow onset of the maximal contraction in rat aorta was resistant to reversibility by repeated washings every 10 min over a 2-3 hrs period; whereas the responses induced by phenylephrine or tolazoline were readily returned to baseline by washing every 10 min over a 40 min period. In contrast to either phenylephrine or tolazoline,
IBI possesses a unique irreversible non-a-AR agonist activity in smooth muscle.
2.3.2 Effects of pretreatment with imidazoline ligands on the contractile responses of IBI analogs in rat aorta.
The purpose of these experiments was to further explore the potential interactions of IBI with imidazoline receptors in rat aorta using pretreatment with recognized selective
83 imidazoline receptor ligands (efaroxan, idazoxan, moxonidine,
or cirazoline) . Moxonidine alone is an agonist contracting rat
aorta tissues with an ECgo value of 10.0 ± 2.4 nM and a maximal
response equal to that of phenylephrine. This contraction by
moxonidine was abolished by 30 nM phenoxybenzamine (PBZ)
pretreatment. Thus, the a-AR-dependent activities induced by
moxonidine were inhibited, and allowed this compound to
interact with the Ii-imidazoline receptor sites that may be
present in this system. In these PBZ treated preparations,
moxonidine (5 piM, 60 min) did not alter the concentrâtion-
response curve of IBI (Figure 2.3; Table 2.2). Similarly,
pretreatment of the tissue with efaroxan (10 /xM, 60 min) ,
another subtype antagonist, did not affect IBI induced
contractile activities in rat aorta. These results support previous observations in which IBI-induced contractions of rat
aorta were not blocked by I PR ligands such as idazoxan or
cirazoline (Lei, 1992) . Taken collectively, these results
imply that contractile responses of IBI in rat aorta are
independent of imidazoline preferring receptors of the 1 % or
Il subtypes. In this regard, IBI-induced activities were
unaffected by currently recognized imidazoline ligands of 1 %
(cirazoline, idazoxan) and Ii(moxonidine, efaroxan) subtypes
84 of IPRs (Emsberger et al., 1995; Regxinathan et al., 1996).
2.3.3 Effects of calcium deficient buffer and calcium restoration on the contractile responses of IBI and tolazoline in rat aorta
In order to examine the possible role of the extracellular calcium in the IBI-induced contractions, calcium was omitted from PSS buffer solution (calcium deficient buffer). At the end of these experiments, the calcium deficient buffer was replaced with normal buffer, and the maximal contraction was measured again. The results are summarized in Table 2.3 and
Figure 2.4. In the calcium deficient buffer, the IBI-induced concentration-response curve was shifted to the right by 2.5 fold (ECso=12.6 fiU) , and the maximal contraction was reduced to be 34 ± 9% (phenylephrine-induced maximum as 100%) . This reduction of IBI-induced maximal contraction was reversed by replacement of calcium deficient buffer with a normal modified
Kreb's solution. In contrast to partial contractile responses observed by IBI, tolazoline did not produce a contraction in calcium deficient buffer (n=3); and the addition of normal calcium containing buffer to the tissue bath did not reverse such an effect.
85 2.3.4 Effects of pretreatment with nifedipine or verapamil on the contractile responses of IBI and tolazoline in rat aorta
The purpose of these studies was to examine the possible calcium channel dependent activities of IBI in rat smooth muscle using the calcium channel blockers nifedipine, R(+)Bay
K 8644 (dihydropyridines [DHP] ) , and verapamil (a
phenylalkylamine [PAA] ) . Data of EC 5 0 , pECgo and maximal responses of IBI or tolazoline alone and in the presence of blockers are summarized in Table 2.3.
IBI produced an EC 50 value of 5.02 (pECgo = 5.30). In the presence of either nifedipine (1 /xM) or verapamil (1 fiM) incubated for 60 min, the concentration-response curve of IBI was significantly shifted to the right with a value of 3.9 fold (19.8/5.02 fiM) or 4.3 fold (21.4/5.02 ^M) in nifedipine or verapamil treated tissue, respectively (Table 2.3). The maximal contraction of IBI was also reduced in these preparations. Nifedipine or verapamil pretreatment suppressed about 2/3 or 1/3 of the maximal contraction by IBI in comparison to the control (Figure 2.4 upper panel). Thus, these calcium channel blockers appear to antagonize IBI induced contractions of rat aorta in a non-competitive manner.
IBI action is sensitive to inhibition by calcium channel
86 blockers of two different chemical classes (DHP and PAA).
Tolazoline was observed to produced concentrât ion-dependent
response that was also suppressed in the presence of 1 nifedipine or 1 verapamil (Figure 2.4 lower panel. Table
2.3) but with only a slight rightward shift in the concentration-response curve. Pretreatment with nifedipine produced a greater inhibition (46%) than that of verapamil
(23%). Thus, it appears that verapamil and nifedipine inhibit the IBI-induced contractions on rat aorta by interfering with the entrance of extracellular calcium into the smooth muscle cells.
2.3.5 Effects of R(+) Bay K 8644 pre treatment on the contractile responses of IBI in rat aorta
To further study the contraction of the IBI-induced contraction, the isomers of Bay K 8644 were employed in these studies. The S(-) and R(+) isomers of Bay K 8644 act as a calcium channel activator and inhibitor, respectively (Bechem et al., 1989; Janis and Triggle, 1991; seeFigure 1.2 for chemical structures). In this study, S(-)-Bay K 8644 caused a concentration-dependent contraction with an ECgo of 18 nM. The average time required to reach the maximal contraction (4.8
87 min) was shorter than either phenylephrine (8.2 min) or IBI
(44 min) . The maximal contraction produced by S(-) Bay K 8644 was 67% of that produced by phenylephrine (Figure 2.5, Table
2.3), and its concentration-response curve was shifted toward the right (11.3 fold shift) with a reduction of 50% in the maximal response by R(+)Bay K 8644 (0.3 /xM) . Similarly, the concentration-response curve of IBI was shifted to the right
(4.5 fold) in a parallel fashion in the presence of 0.3 pM
R(+)Bay K 8644 without a change in the maximal contractile response (Figure 2.5). Thus, the IBI-induced contraction on rat aorta was blocked competitively by pretreatment with R(+) -
Bay K 8644, and the data provide further evidence for a calcium channel-dependent mechanism of IBI action in vascular smooth muscle.
2.3.6 Intracellular calcium transient and tension chamges induced by IBI in pig coronary artery
To obtain additional information on the role of
intracellular calcium, studies were carried out using fura - 2
AM as an indicator of intracellular calcium changes in pig coronary artery strips. In these experiments, transient calcium changes were measured in the presence of KCl and IBI.
88 KCl (35 mM) produced calcium transient and tension changes, and was used as a control compound in each preparation. The authenticity of the fluorescence ratio changes was confirmed
by observing F340 and F 330 moving in opposite directions (Figure
2.6). Measurement of fluorescence ratio values was not useful in these experiments due to the long-lasting time course of
IBI-induced calcium signal which may undergo photobleaching with time. However, the half time (t^/;) needed to reach the peak values induced by KCl (Figure 2.7) and IBI (Figure 2.8) were measured and compared. Both KCl (Figure 2.7) and histamine (not shown) induced changes in calcium signal and tension were rapid (within min range) ; and the changes in calcium signal preceded the tension in a comparable manner.
In KCl treated tissues, the lag time between the fluorescence
elevation (ti/2 =ll ± 2 sec) and the tension increase (t^2 = 2 0 ± 3 sec) was about 9 sec. In contrast, at a concentration of 100
(which was 20 fold greater than the ECgo of 5 fiM in rat thoracic smooth muscle), IBI produced a slight initial relaxation followed by a slow contractile response. These responses were similar to that seen in rat aorta when a large concentration of IBI was applied. The half time (t^/j) required to achieve the peak activation for calcium signal and
89 contraction is 170±31 sec or 25±4 min. These data indicate a non-immediate contractile response despite the relatively rapid increase of IBI-induced intracellular calcium signal, suggesting that another mechcinism may be also involved in the
IBI-mediated contraction of vascular smooth muscle.
On the contraction plateau of IBI, a high concentration of verapamil (100 fiM) was added and observed (n=3) to partially reverse (15%) the intracellular calcium signal evoked by 100
(iM IBI (Figure 2.9) . This small reduction in calcium transient by verapamil was also coupled initially with a small, then with a complete relaxation (within 30 min) of the contraction produced by IBI. These data indicate the intracellular calcium increase plays a role in the contractions produced by IBI.
Therefore, these results provide evidence for calcium channel- involvement of IBI in contracting the smooth muscles.
Nifedipine (100 ^M) was also used for this purpose, but produced very weak effect in calcium transient and tissue tension changes due to its possible photosensitivity and rapid degradation. Therefore, only verapamil was used as calcium channel blocker in these experiments.
90 2.3.7 Effect of pre treatment with NCS containing derivatives on the contractile responses of IBX in rat aorta.
The purpose of these experiments was to examine whether 1)
IBI induced activities in smooth muscles are altered by pretreatment with 2 '-IBI isomer (analog VII in Table 3.1); and
2 ) the calcium channel blocking properties of nifedipine are enhanced by the presence of an NCS group (nifedipine-NCS, or analog VIII; see Table 3.1 for structure). The experiments were carried out by testing the contractile activities of the
-NCS analogs of IBI and nifedipine alone, or as antagonists of
IBI action in rat aorta.
The results of the experiments are summarized in Table 2.4.
Neither analog produced a significant contraction of the
smooth muscles in the concentration range up to 1 0 - 1 0 0 /xM.
2 '-IBI isomer shifted the concentration-response curve of IBI to the right by about 4-fold without suppression of the maximal contractile response (Figure 2.10, upper panel) . These results suggest that the 2'-IBI isomer blocked the IBI activities in a competitive fashion. In contrast, nifedipine-
NCS (10 fiM, 45 min) suppressed the maximal contraction by IBI and increased the ECgo value to 27.8 (Figure 2.10 lower panel). This observed reduction of the IBI maximal contraction
91 of IBI by the nifedipine-NCS was reversible. Therefore, nifedipine-NCS blocked the action of IBI on rat aorta in a non-competitive manner, suggesting that this compound interacts at a different site from that of IBI.
2.3.8 Functional studies of IBI on rat skeletal muscle
IBI alone contracted the skeletal muscles of rat diaphragm with a similar time course and concentration-dependency to that in rat aorta tissue. The ECgo of IBI in skeletal muscle
(37 ± 0.4 [iM, n=3; Table 2.5) is 7-fold greater than that in thoracic aorta (5.0 /xM, n=13 ; Table 2.1). The IBI-induced contractions of skeletal muscle were increasingly tonic and raised the basal tension to such a level that electrically driven twitches did not produce further contraction. The contractile responses to IBI were not blocked in the presence of 10 /xM tubocurarine, a concentration which abolished the electrically driven twitches in control preparation (data not shown). However, the concentration-response curve of IBI was shifted to the right in the presence of 30 /xM dantrolene
(Figure 2.11, Table 2.5). These results indicate that IBI actions in skeletal muscle are independent of nicotinic receptors, and dependent on calcium channels which are
92 sensitive to inhibition by dantrolene.
2.3.9 Inhibitory effects of IBI analogs on (+) [^H]PN 200-110
binding to rabbit skeletal t-tubule membranes
The results of the binding assay with the labeled
dihydropyridine (DHP), (+) [^H] PN 200-110, and competition by
IBI and selected analogs of naphazoline and nifedipine are
summarized in Table 2.6. The values shows a binding potency
(in iiM) sequence of nifedipine (0.012) >> nifedipine-NCS
(0.83) > naphazoline-NCS (2.6 ^M) > IBI (49) . Tolazoline and
naphazoline at concentrations ranging from 3 to 3 mM did
not displace the radioligand from specific binding sites.
While nifedipine completely displaced ( + ) [%] PN 200-110 from
DHP binding sites on calcium channels, the NCS containing cuialogs (tolazoline-NCS [IBI] , naphazoline-NCS and nifedipine-
NCS) displaced the specific radioligand binding by 84-88% at the highest concentrations. These results indicate that -NCS group of IBI interacts with DHP binding sites on calcium channels, and that this interaction may be of an allosteric type. Interestingly, the presence of the NCS group on nifedipine (nifedipine-NCS) reduced the affinity for DHP sites, and nifedipine-NCS incompletely displaced ( + ) [^H] PN
93 200-110 from the DHP sites (Figure 2.12) , In addition, naphazoline-NCS, but not naphazoline, was a weak competitor of the DHP binding sites. Thus, the NCS group acts as a mediator that enables compounds to actively bind to calcium channels.
In other words, the presence of this functionality confers calcium channel binding properties to both naphazoline and tolazoline and reduces but does not abolish calcium channel binding properties of nifedipine (Figure 2.12, Table 2.6).
2.3.10 Inhibitory effects of IBI analogs on S (-) [^S]Bay K
8644 binding to rabbit skeletal T-tubule membreuies
S(-)-Bay K 8644, the compound with the same chemical structure as the radioligand, displaced the radioligand binding to the same extent as 10 /xM PN 200-110. IBI and naphazoline-NCS displaced S(-) [^H] Bay K 8644 from DHP sites, whereas tolazoline and naphazoline did not reduce specific binding of this radioligand at concentrations up to 1 mM
(Figure 2.13). Therefore, the presence of NCS group on tolazoline (IBI) and naphazoline (naphazoline-NCS) was responsible for the interaction with C^H] Bay K 8644 binding sites. These results, like those for (+)[^H]PN 200-110, indicate that the NCS group plays an important role in
94 facilitating interaction of tolazoline-like compounds with L- type calcium channels.
These data mainly agree with the previous results obtained by (+) [^H]PN 200-110 labeling in the same tissue preparation.
This may be due to the similar manner of structure between these two radioligands, thus a similar receptor-ligand interaction. However, the extent of displacement of S(-) [^H]
Bay K 8644 by IBI and naphazoline-NCS is about 50% the maximal displacement by addition of cold Bay K 8644 ligand.
2.4 Discussion
2.4.1 Irreversible agonist activities of IBI independent of a-Adrenergic amd imidazoline receptors
Isothiocyanate group (NCS) . In the hope of differentiating imidazoline receptor sites from those of a-adrenoceptors, IBI was synthesized. IBI differs from tolazoline, an imidazoline containing a-adrenergic ligand by the addition of an isothiocyanato (NCS) group on the 4 '-position of the phenyl ring. IBI was evaluated as a selective site directed affinity probe on the basis of the electrophilic properties of the NCS group. Affinity label probes of imidazolines have been
95 previously reported and include chlorethylclonidine (CEC) , p- azidocloinidine, p-isothiocyanato clonidine (Atlas et al.,
1982) and p-methylisothiocyanato clonidine (Decker et al.,
1983) . These affinity probes exhibited diverse pharmacological activities including either a-, non-a-adrenergic agonist or a- antagonist actions. In the present studies, IBI produced a contractile response in smooth muscle mediated via non-a- adrenergic and non-imidazoline receptor dependent mechanisms, and exhibited antiaggregatory activities against human platelets via both ttj-adrenoceptor dependent and independent pathways (Lei, 1992).
It is hypothesized that IBI, by the NCS group, is capable of interacting with a nucleophilic site of the receptor.
Presumably, IBI interacts with those receptors to form a covalent bond. This proposal is supported by several
observations : (1 ) the tissue responses were of slow onset amd
long-lasting duration; (2 ) the responses were not easily reversed by repeated washings; and (3) the contractile responses were not significantly reversed by addition of calcium channel blockers.
a.-Adrenoceptor independent contractions of IBI. The
96 contractile responses induced by IBI contrast to that of phenylephrine (a full agonist) and tolazoline (a partial agonists) in several aspects. As opposed to the reference agonist phenylephrine, IBI-induced contractions are characterized by a slow onset (44 min) , lower potency (ECgo =
5 iM) , higher intrinsic activity (116%) , and a resistance to an irreversible a-AR blocker, phenoxybenz amine (PBZ; 30 nM) , and to a reversal of the contractile response by an extensive washing procedure (3 hrs). These parameters are in close agreement with those previously reported for IBI. In the present experiments, the ECgo value of IBI (5 /xM) is lower than those by Venkataraman et al., 1989 (16.3 /xM) and Lei,
1992 (45.3 /xM) . This discrepancy may be due to the use of a longer time course for development of contractile response to an individual dose were adopted in the present studies, or to older rats (350-450 gm) . Due to the longer times needed between doses, the variability in ECgo values in tissues was also reduced in these experiments. More importantly, these changes in experimental design do not interfere with the measurement of parameters of agonist activity on smooth muscle for IBI and related analogs.
PBZ is a haloalkylamine which irreversibly interacts with
97 both the « 1 - and -adrenergic receptors (ARs) and blocks responses to catecholamines (Hoffman and Lefkowitz, 1990) . It covalently binds to a-ARs producing a non-equilibrium type of blockade for a prolonged duration of time (14-48 hr) . PBZ is generally classified as non-selective a-AR blocker even though it is somewhat selective for a^-AR; and does not interfere with p-AR mediated effects. PBZ also inhibits reuptake of released catecholamines into both presynaptic adrenergic nerve terminals and extraneuronal tissues (Cubeddu et al., 1988).
When used for studying adrenoceptors, this property may provide the experimentalist with an advantage if a reuptake blocker (e.g., cocaine) is not used in the experiment. In addition to a-ARs, higher concentrations of PBZ also block histamine (H^-subtype) , acetylcholine and serotonin receptors
(Hoffman and Lefkowitz, 1990) . Hence, the concentration and time of incubation are important in controlling the degree of blockade that is achieved by PBZ.
In the present studies, 30 nM phenoxybenzamine was
incubated for 2 0 min prior to addition of the test compounds.
The phenoxybenzamine-pretreated preparations proved to be a useful tool since this treatment abolished the contractile responses of the tissue to known a^-AR agonists, phenylephrine
98 and tolazoline, and previously was found to shift the a- adrenoceptor dependent concentration response curve of norepinephrine on rat aorta by over 100-fold (Lei, 1992).
However, PBZ-pretreatment did not significantly affect the stimulatory effects of IBI in terms of either ECgo or maximal contraction response. The results indicate that the irreversible stimulatory activity of IBI is unrelated to activation of a-adrenergic receptors in rat aorta, and that the substitution of NCS group at the 4'- position of tolazoline is responsible for the contractile responses.
In previous studies (Lei, 1992), another alkylating blocker, chloroethylclonidine (CEC) was also examined for its ability to block IBI actions in rat aorta. CEC contains an imidazoline ring and is able to alkylate a^g-adrenoceptors. It is reported to unmask a non-a-adrenoceptor responsive site to norepinephrine in rat aorta (Oriowo et al., 1990), Thus, pretreatment with CEC was expected to reveal the importance of a^g-adrenoceptor and non-a - adrenoceptor imidazoline sites. Like the results with PBZ, pretreatment with CEC did not significantly affect the IBI induced contractile responses in rat thoracic aorta.
99 Imidazoline receptor independent contractions of IBI.
Pharmacological studies on imidazoline ligands have resulted in the isolation and identification of imidazoline receptors in a variety of tissues. Subdivision of imidazoline receptors has been proposed as and Ij- sites, which were studied by use of the radiolabeled ligands, [^H]clonidine and
[^H] idazoxan, respectively (Michel et al., 1992).
Previous studies suggest that IBI contracted smooth muscle tissues independent of imidazoline receptor activation (Lei,
1992). These results are inconsistent with data obtained from binding studies which showed that IBI displaced [^H]idazoxan binding in guinea pig cerebral and porcine renal cortex membranes (Hussain et al., 1992). In functional experiments, the mechanism of IBI-induced contractions was examined using cirazoline («i-AR agonist and I;- receptor ligand), idazoxan
(«2 - and Ig- receptor antagonist) and phenoxybenzamine (PBZ) pretreated rat aortic strips. Since IBI contains an imidazoline ring that may be a potential bioactive group, it is important to examine whether IBI induced activities can be blocked by recently identified I^- imidazoline subtype ligands such as moxonidine and efaroxan (Emsberger et al., 1993) . The results of the present study indicate that pretreatment with
100 either efaroxan or moxonidine (Emsberger et al., 1993) in
PBZ-pretreated tissues did not affect IBI activity in smooth muscles. Since cirazoline and moxonidine were agonists in rat aorta, PBZ-treated tissues were used in these experiments.
With the abolishment of a-AR activities by PBZ-pretreatment, these imidazoline antagonists did not affect the ECgo or the maximal contraction response of IBI relative to the control responses. In agreement with the previous observations by Lei
(1992) , and as recently published (Lei at al., 1995), the data indicate that imidazoline preferring receptors (IPR) are not involved in the mechanism of IBI action.
A negative result was observed with the various imidazoline ligand pretreatments in rat aorta. In view of the importance of the imidazoline ring in these a-AR and IPR antagonists in characterizing imidazoline receptor activity, we assumed that the affinity or efficacy of tolazoline should be influenced by addition of the NCS group at the 4 ' -position of the phenyl ring (as in IBI) . Ideally, these imidazoline ligands may exert their antagonism by an interaction with domains on both a-AR and IPR. If the imidazoline ring of IBI plays a role in initiating the IBI action, pre-incubâtion with imidazoline containing a-AR antagonists might modify IBI action by pre
101 occupying the imidazoline ring binding region of IBI receptor.
Thus, the negative findings suggest that the imidazoline ring
of IBI is not an important factor in eliciting the non-a-
adrenoceptor mediated stimulatory activity of IBI in rat
aorta. However, whether IBI acts at another subtype of
imidazoline receptor remains to be established.
It is important to mention the results of previous studies
(Lei, 1992) with tolazoline, a partial agonist and antagonist
in a-AR system. Tolazoline was inactive as an agonist on guinea pig ileum, indicating that a-AR is unresponsive to imidazolines, or that there may be little or no a-AR in this tissue. Since tolazoline is a close structural analog of IBI, the effect of pretreatment of ileum with tolazoline on the concentration-response curve to IBI was examined. No blockade was observed with tolazoline on IBI stimulatory activity.
These results indicate that tolazoline and IBI interact with different receptor sites. Since these two compounds differ only by presence of a 4-isothiocyanato group, it appears that the presence of this functional group in IBI is essential for producing the non-a-adrenergic contractile activities in these tissues.
102 2.4.2 Calcium channel dependent activities of IBI in smooth and skeletal muscles
2.4.2.1 Functional studies
The present studies demonstrated that IBI-produced contractile responses on rat aorta are dependent upon the presence of L type calcium channels. Contractile responses of
IBI were blocked by nifedipine and verapamil, and by use of calcium deficient media (Slavica et al., 1994; Lei et al.,
1995). In most cases, the concentration-response curves of IBI were similarly shifted to the right with reduction in the maximal response. The reduction in IBI responses by calcium deficient buffer was reversed upon replacement with normal calcium-containing buffer (Table 2.3). Structure-function studies with IBI analogs containing -NCS group (see Chapter 3) also indicate that their contractile effects on rat aorta are antagonized by calcium channel blockers, but not by PBZ- treatment.
To further characterize the calcium channel dependent property of the IBI-induced contraction, the isomers of Bay K
8644 were employed in this studies. The contractile responses of IBI and S (-)-Bay K 8644 (a calcium channel activator) were
103 tested in the absence and presence of R( + )-Bay K 8644 (a calcium channel blocker; see Figure 1.2 for structures).
Responses to both IBI and S (-) Bay K 8644 were attenuated by the calcium antagonist R(+)Bay K 8644, resulting in rightward shifts in the concentration-response curves of both agonists.
Since these effects on IBI were similar to S (-)Bay K 8644, these findings suggest that IBI is an activator of calcium channels.
The contractions induced by tolazoline, phenylephrine or
IBI were blocked by verapamil, suggesting that there is common pathway regulating calcium channels. However, the contractile responses to tolazoline and phenylephrine were blocked by PBZ, whereas PBZ-treatment did not block IBI-induced contraction.
These results strongly suggest that the mechanism of action for a-adrenoceptor agonist (phenylephrine, tolazoline) and IBI on calcium channels are different. This may be explained due to the existence of either different receptor/channel binding sites, or to a different pathway involved in the activation of calcium channels by these two groups of drugs. For example,
IBI may directly activate calcium channels whereas the other two drugs mediate activation of calcium channels via the a- adrenoceptor dependent pathway.
104 2.4.2.2 Intracellular calcium transient
Fura-2 AM as an intracellular calcium indicator. To date,
fura - 2 is probably one of the best cytosolic calcium indicators available at present (Grynkiewicz et al., 1985). It functions as fluorescent calcium chelator and provides a simple method for simultaneous measurements of [Ca^*] ^ and contractile tension in vascular and other types of smooth muscles (Himpens et al., 1988). Normal loading conditions do not change cell functions since the contractile and relaxing
activities of smooth muscle are not affected by the fura - 2 loading (Karaki, 1989). Despite being taken up by the nucleus
(which contains a similar concentration of calcium to that of
cytoplasm) , there is no distribution of fura - 2 into mitochondria or sarcoplasmic reticulum which contain high calcium concentrations (Moore et al., 1991).
The use of this indicator, however, presents some practical problems including formation of intermediate metabolites
(Scanlon et al., 1987), partial photobleaching (Becker et al.,
1987) and leaking out of the cell from the cytosolic compartments. To minimize these possible artifacts while conducting the experiments with fura-2 AM, special attention
105 was given to monitoring F 340 and F^go movement in opposite directions (see Figure 2.6), and limiting the intensity and duration of excitation light to reduce photobleaching. Owing to the sustained elevation of intracellular calcium and tension, absolute value of intracellular calcium concentration becomes less practical. The latter is calculated from the
fura - 2 fluorescence intensity and the dissociation constant of fura-2 for calcium (Grynkiewicz et al., 1985). Since this parameter measurement requires relatively stable fluorescence intensity, this fluorescent dye is suited for experiments of short duration. In our studies, relative fluorescence intensity was used instead of absolute calcium concentration.
This type of measurement provides comparable data for sets of activators using the maximal calcium transient and tension evoked by high concentration (35 mM) of KCl as the reference standard. Such contractions, termed as calcium-induced contractions triggered by the potassium-depolarization, may be mediated via voltage dependent calcium channels and are susceptible to calcium inhibitors, but do not involve direct mobilization of intracellular calcium stores (Golenhofen et al., 1977; Brading and Sneddon, 1980; Spedding, 1982).
Histamine was also utilized in the system as a reference
106 representing an increased intracellular calcium concentration from both external influx and internal calcium release.
Calcium signal and tension. Preliminary experiments using rat thoracic aorta were unsuccessful. This was due to the very high fluorescence background counting of photons in the tissue preparation. Experiments were then continued with pig coronary artery as an alternative tissue that proved to be appropriate for these studies with IBI.
In these preparations, IBI produced increases of the intracellular calcium signal and a maximal contraction which was much higher than that by KCl. It may be possible that IBI contracts the smooth muscle through more than one mechanism.
Addition of verapamil at the plateau partially reversed the
IBI-induced calcium signal and tension changes. One of the mechanisms for IBI apparently involves the activation of cell membrane calcium channels leading to influx of extracellular calcium. Additionally, intracellular calcium release from the myoplasmic apparatus may also play important role in this IBI induced calcium and tension changes, for instance, by intracellular calcium release from sarcoplasmic reticulum by
the IP3 pathway (Berridge, 1989; 1993).
107 Taken together, the data indicate that IBI produced an elevation of both intracellular calcium concentration and tension. The time course of both calcium signal and contraction induced by IBI was longer than that of either histamine or potassium, suggesting that a much slower rate of intracellular calcium elevation was stimulated by IBI.
Moreover, both calcium signal and maximal tension induced by
IBI were greater than that by potassium and smaller than that by histamine, suggesting a mechanism involving extracellular calcium influx as well as possible association with intracellular calcium release through such calcium channels as ryanodine receptors or its effector pathways.
2.4.2.3 Binding of IBI to raüobit skeletal muscles
Role of the isothiocyanato (NCS) group. The electrophilic nature of the NCS group makes it a valuable tool for reaction with nucleophilic nitrogen of the polypeptide bound on a receptor protein. This may imply an interaction of the NCS group on IBI with a nucleophilic domain such as nitrogen atom of the protein in the membrane. Further, the formation of a covalent bond to the receptor may contribute to the long lasting, irreversible action of IBI induce-contraction in
108 smooth and skeletal muscles.
Potential protein site(8) of action. Functional studies indicated that calcium channel blockers antagonized IBI contraction. IBI may regulate ligand interactions with dihydropyridine [DHP] sites or phenylalkylamine [PAA] sites on calcium channels. To test this possibility, binding assays
using (-) [^H]D 8 8 8 for PAA sites or [^H] PN 200-110 (or [^H]
Bay K 8644) for DHP sites were used to determine sites of interaction for IBI related compounds. It is also possible that the -NCS binding site may be different from the DHP or PA sites. The marked diversity in the chemical structures of
DHP, PAA and diltiazem calcium antagonists and their potencies on L-type channels suggests that different interaction sites exist for these channels classes (Janis and Triggle, 1991) . On the channel homogeneity point of view, a proposal of a multiplicity of binding sites existing on the L channels is not surprising. The sequence of the «i-subunit of L type channel is highly homologous to that of the sodium channel which possesses at least five distinct categories of binding sites (Tanabe et al., 1987; Strichartz et al., 1987). With binding assay, allosteric interactions of calcium channel
109 agents with the channels have frequently been reported (Vaghy et al., 1987b; Janis et al., 1991). These data can be divided into positive or negative heterotropic interactions which result in conformational changes of channels in such a way that either facilitates or impedes the gating of calcium through the membranes. In the present studies, while inhibition by control competitors (nifedipine or S (-)Bay K
8644) completely displaced the DHP sites, only a fraction (84-
8 8 %) of the DHP sites were inhibited by the -NCS derivatives.
These results of a partial displacement of radioligands (PN
200-110/Bay K 8644) from the calcium channels indicate that the NCS group may interact allosterically with DHP sites.
2.5 Summary and conclusions
Prototype drug. Tolazoline is a known a-adrenergic ligand in smooth muscles. In comparison to a full agonist phenylephrine
(ECso=12 nM) , it produced a partial contraction (59% phenylephrine maximal response) and was less potent (ECgo =
0.32 [iM) . Like phenylephrine, the responses to tolazoline were completely abolished by PBZ-treatment.
Functional studies. Attachment of an isothiocyanato (NCS)
110 group at the 4 ' -position of the phenyl ring of tolazoline yielded 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI]. Such a chemical modification has altered the pharmacological properties of tolazoline. IBI has been shown (Lei, 1992;
Slavics et al., 1994) to produce an irreversible, progressive and sustained contraction of rat aorta with an ECgo of 5 ptM and
a maximal contraction which is nearly 2 -fold greater than that of tolazoline. The IBI-induced contractions are mediated through a mechanism dependent upon calcium channels (inhibited by verapamil and nifedipine) , and independent of cx- adrenoceptors (not inhibited by phenoxybenzamine) or imidazoline preferring receptors (not inhibited by efaroxan, moxonidine, idazoxan and cirazoline). Similarly, studies on structure-activity relationships (see Chapter 3) of IBI analogs revealed that the NCS group is important for the IBI- like activities in smooth muscles. In the rabbit diaphragm skeletal muscle preparation, IBI produced a similar time course and concentration dependency of contractile responses to that observed in smooth muscles. In addition, the IBI- induced contractions in rat diaphragm were inhibited by dantrolene (10 /xM) , an inhibitor of calcium channels present in the sarcoplasmic reticulum (Strazis et al., 1993).
Ill Intracellular calcium and tension changes. Addition of KCl (35 mM) to pig coronary artery strips produced immediate increases in intracellular calcium and tension in a parallel manner, and were used as control. The lag time between the calcium and the tension was 9 sec. In contrast to KCl, IBI (100 /xM) produced a much slower onset of changes in both calcium transient and tension, and markedly increased lag time (22 min) . A greater dissociation between these two parameters was evident with IBI as opposed to KCl. However, IBI-induced increases in calcium transient and tension were blocked or partially reversed by pre incubât ion (data not shown) or addition (Figure 2.9) of verapamil (0.1 mM) . Thus, the data suggest that calcium is an important mediator for IBI-induced contraction. However, other mechanism unrelated to immediate intracellular calcium increases by IBI may be involved in the contractile response.
Calcium channel binding assays. The present study has demonstrated that IBI is an inhibitor of DHP binding to calcium channels of t-tubule membranes in rabbit skeletal muscle. Moreover, whereas tolazoline and naphazoline were not competitive inhibitors of DHP binding sites, their corresponding isothiocyanato (NCS) derivatives (IBI and
112 naphazoline-NCS, respectively) were able to displace the radioligands.
Differences were observed in the maximal displacement of the specific binding of DHP ligands by IBI and related compounds. In contrast to the complete displacement observed for DHP compounds [nifedipine and S (-)Bay K 8664], all NCS
derivatives displaced up to 8 8 % of the specific radioligand bound. Taken collectively with the % not equal 1, these differences in maximal displacement may indicate an allosteric interaction of these NCS derivatives with DHP sites on calcium channels.
Nifedipine competed for the DHP site in a manner of single receptor species (ng = 0.97) via a reversible bi-molecular reaction. Nifedipine-NCS exhibited an positive cooperativity
(ng = 1.37) in displacing the radioligand from the DHP site, although the binding affinity was reduced. An increased Hill slope value were also observed for the other NCS derivatives with the exception of naphazoline-NCS. Due to the presence of the bulky naphthalene ring, attachment of an NCS group to naphazoline (naphazoline-NCS) resulted in a calcium channel interaction, which exhibited a negative cooperativity (ng =
0.81) for the ligand-receptor interaction. This observation
113 for naphazoline-NCS in the binding assay correlates well to its a-adrenoceptor dependent contractile responses in rat aorta. Hence, the result implies that both allosteric and facilitated interactions of NCS derivatives are required to achieve the IBI-like contractile responses with non-a-AR and non-imidazoline mechanisms.
Summary. It appears that: (1) the NCS group attached to imidazoline a-adrenoceptor ligands has resulted in the
discovery of a novel group of calcium channel agonists; (2 ) the NCS containing derivatives interact with DHP binding sites in an allosteric manner; (3) NCS attachment to these compounds either confers calcium agonist properties to non-DHP compounds or facilitates binding to DHP sites; (4) both allosteric and positive cooperative binding to DHP sites are required in order to achieve the IBI-like activities elicited by NCS derivatives; and (5) the results from both functional and binding studies on muscles are highly correlated, and strongly indicate that IBI interacts directly with L type calcium channels, which increases in intracellular calcium signals and leads to the unique contractile responses for this novel class of compounds.
114 l.Ch Phenylephrine
0.8- IBI E (0 l_ U) c o tfl Tolazoline c
0. 2-
0.0-
20 3 0 4 0 5 0 M inutes
Figure 2.1 Time course and absolute tension changes produced by 30 /xM phenylephrine (PE) , 100 /xM tolazoline (TOL) and 100 fiM 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] in rat aortic strips. Data represent the mean ± SEM of 4-27 experiments. Notice that the sec[uence of time course (min) is IBI (44) > PE (8.2) > TOL (5.8), and that of the maximal contraction (30 PE as 100%) is IBI (116) > PE (100) > TOL (59) . The differences in time course and the maximal contraction between agonists are statistically significant (p < 0.05). Data are summarized in Table 2.1.
115 ■ IBI (n=13) □ PBZ + IBI (n=13) A P E (n = 9 ) 100- A PBZ + PE (n=9) c • TOL (n=4) o 4-> o PBZ + TOL (n=4) % C o ü 50- X
- l O _• 10 Ï0 “ ® 10 “ ® 10 ï6“®' 10 “ ® ■ 1 0 “ ^ [D ru g ] (M)
Figure 2.2 Effect of phenoxybenzamine (PBZ) pretreatment on contractile responses of phenylephrine (PE), tolazoline (TOL) and 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] in rat aorta. The time course of contraction for these agonists are listed in Table 2.1. Contractions produced by 30 ptM phenylephrine were used as 100%. PBZ (30 nM) was incubated for 20 min followed with wash twice. Data represent the mean ± SEM of 4-13 experiments.
116 Cmpd n Time Tension %Emaxb EC 50 (pM)° course (mg) Alone With PBZd Alone With PBZd (min)
PE 27 8.2±0.3 679±32 1 0 0 NA“ 0.018±0.005 NA“
TOL 4 5.8±0.2 407±41 59±6f NA“ 0 .329±0.120 NA“
IBI 13 44±3 787±43 116±5f 112±5 5.02±0.69 5.16±0.60
“ Data represent the mean ± SEM of 6-27 experiments. *’ = percent maximal contractile responses using 30 fxM phenylephrine as 100%
° EC 5 0 = median effective concentration. ^ PBZ = phenoxybenzamine, 30 nM was incubated for 20 rain with washout twice. “ No activity was observed at concentrations up to 3 0 (iM. Significantly different from control (p < 0 .0 0 1 ). * Statistically different from the control (PE), p < 0.05.
Table 2.1 Effects of pretreatment with phenoxybenzamine on the contractile responses to phenylephrine [PE], tolazoline [TOL] and 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] on rat thoracic aorta.“
117 □ iBI (n=13) gPBZ&MOX + IBI (n=4) ^E FA + IBI (n=3) [0PBZ&CIR + IBI (n=4) IDA + IBI (n=4)
100
O
o in 50
Figure 2.3 Effects of imidazoline ligands pretreatment on potency (A), and the maximal contraction (B) to 2-(4'- isothiocyanatobenzyl) imidazoline [IBI] in rat thoracic aorta. Efaroxan (EFA) or idazoxan (IDA) was incubated at 10 ptM for 60 min without wash; Phenoxybenzamine (PBZ, 30 nM) was incubated for 20 min with washout twice and then incubated with 5 moxonidine (MOX) or 3 cirazoline (CIR) for 60 min without wash. Percent maximal contractile responses were calculated using 30 /xM phenylephrine as 100%. Data represent the mean ± SEM of 3-13 experiments.
118 Compound*’ n EC 5 0 (pM)=
IBI 13 5.02 ± 0.7 116 ± 5
EFA + IBI 3 5.64 ± 0.46 119 ± 9
IDA + IBI 4 4.95 ± 0.9 113 ± 5
PBZ & MOX + IBI 4 5.99 ± 0.25 114 ± 2
PBZ Sc CIR + IBI 4 4.92 ± 1.2 102 ± 4
“ Data represent the mean ± SEM. ** Efaroxan (EFA) or idazoxan (IDA) was incubated at 10 fiM for 60 min without wash; Phenoxybenzamine (PBZ, 30 nM) was incubated for 2 0 min with washout twice and then incubated with 5 ^M moxonidine (MOX) or 3 ^M cirazoline (CIR) for 60 min without wash.
= EC 50 = median effective concentration. ^ = percent maximal contractile responses using 30 [iM phenylephrine as 1 0 0 %.
Table 2.2 Effects of pretreatment with imidazoline ligands (efaroxan, moxonidine, cirazoline or idazoxan) on the contractile responses to 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] on rat thoracic aorta.
119 Figure 2.4 Effects of calcium deficient media and pretreatment with nifedipine (1 /xM) or verapamil (1/xM) on the contractile activities of 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] (the upper panel) and tolazoline (the lower panel) on rat aorta. Percent maximal contractile responses were calculated using 30 /xM phenylephrine as 100%. Data represent the mean ± SEM of 4-13 experiments.
120 ■ Control A Verapamil • deficient buffer o Nifedipine
100- c o
50-
-8 -7 -4 [IBI] (M)
100-
50-
-8 -5 -4 [Tolazoline] (M)
121 ♦ (-)Bay K (n=4) « (+)Bay K + (-)Bay K (n=4) ■ IBI (n=13) □ (+)Bay K + IBI (n=4) 100- c o
O 5 0 X
- l O ^ 8 ^7 ^5 10- 10 ' [D rug] (M)
Figure 2.5 Effects of pretreatment with R( + )-Bay K 8644 ( [ ( + ) Bay K] ,0.3 [iM) on the contractile activities of S (-) -Bay K 8644 [ (-) Bay K] and 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] on rat aorta. Percent maximal contractile responses were calculated using 30 /xM phenylephrine as 100%. Data represent the mean ± SEM of 4-13 experiments.
122 Treatment^ n EC 5 0 (/iM) pECso'
TOL 4 0.329±0.120 6.48±0.20 59±6 Ca** deficient 4 NA^ NA^ NAf buffer + TOL Add Ca** back 4 — — - NAf NIF + TOL 4 0.48±0.27 6.32±0.21 13+4=
VP + TOL 4 0.221±0.048 6 .6 6 ± 0 . 1 1 26±1=
IBI 13 5.02±0.7 5.30±0.2 116±5 Ca*+ deficient 4 4.90±0.01 5.31±0.1 34±9= buffer + IBI Add Ca** back 4 - - - 115±3 NIF + IBI 4 19.8±6.7= 4.70±0.17® 33±6= VP + IBI 4 21.4±7.9= 4.67±0.19® 64±12=
(-)BK 4 0.018±0.005 7.75±0.12 78±7 ( + )BK + (-)BK 4 0.11±0.017= 6.96±0.04® 40±11= (+)BK + IBI 4 22.5±3.4« 4.65±0.07® 117±4
® Data represent the 1mean ± SEM.
^ Before construction of tolazoline (TOL) or IBI curves, 1
/iM nifedipine (NIF), 1 /iM verapamil (VP) or 0.3 /iM R( + ) Bay K 8644 was incubated for 60 min, respectively.
pECso = -log EC 5 0 , the median maximal concentration. = the maximal percent contractile responses using the maximal contraction by 30 piM phenylephrine as 100%. ® Statistically different (p < 0.01) from control. ^ NA = no activity up to 30 /iM. Statistically different (p <
0 .0 0 1 ) from control.
Table 2.3 Comparative effects of calcium deficient media and calcium channel antagonists (verapamil, nifedipine, R(+)Bay K 8644) on the contractile activities of tolazoline, 2-(4'- isothiocyanato-benzyl imidazoline [IBI] and S (-) Bay K 8644 on rat aorta.®
123 2.5
0 ) U c (D U (/);% 2.0 en c (U 0) O c
_ 2 ü _
1.5 380
1 min
Figure 2.6 Fura-2 fluorescence intensity (F340/F 3 8 0 ) measurement for pig right coronary artery in response to 35 mM
KCl. The opposite direction of fluorescence F 340 and F 3S0 represents the authenticity of fluorescence change.
124 KCI-induced calcium and tension change l.O i
Calcium 0.8- Wash
0: 'o 0.6- Tension u Ü o 8 n 0.4- 5 0 mg
0.2-
KCI ad d ed 0.0 50 100 150 S e c o n d s •
Figure 2.7 Representative tracing of 35 mM KCl-induced augment of fura - 2 fluorescence coupled with the tension changes in pig coronary artery. The half time (ti/j) of KCl- induced peak activation for fluorescence and tension changes is 11+2 sec or 20±3 sec, respectively.
125 A B Changes in calcium signal induced by IBI Changes in tension induced by IBI 1.0-,
-600 0 8- —4
s % 0 ^ r L- -400 o o n 0 4- u_ -200 0. 2-
IBI (100 mM) added 0.0 100 200 300 Seconds Minutes
Figure 2.8 100 fiM IBI-induced changes in fura-2 fluorescence and tension. The half time (ti/z) of IBI-induced peak activation for fluorescence (panel A) or tension change (panel B) is 170±31 sec and 25±4 min, respectively. No tension change was observed within 300 s e c .
126 Reversal of IBI-induced calcium signal
•0.2 On by verapamil (O.lmM. n=3)
QL 0 CO <ü m 015“ tn ^ 0) m b D L.
010 - 50 1 0 0 150 200 Seconds
Figure 2.9 Inhibitory effects of verapamil (O.lmM) on the the plateau of calcium signal induced by IBI (100 for 40 min) fluorescence in pig coronary artery. This reversal effect by verapamil represents 15% of the total calcium elevation induced by IBI. Nearly complete reduction in tension (not shown) occurred 3 0 min after the calcium signal attenuation. Data are means + SEM of n=3.
127 ■ IBI o 2'-IBI isomer +- IBI
100-
50-
C o 4u -J (9
C o A Nifedipine-NCS (wash) + IBI o D Nifedipine-NCS (no wash) + IBI 5^ ■ IBI
100-
50-
[Drug] (M)
Figure 2.10 Effects of pretreatment with NCS derivatives on the contractile responses to 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] on rat aortic strips. Upper panel: incubated with 2'-IBI isomer (analog VII, 100 /xM for 45 min); lower panel: incubated with naphazoline-NCS (analog VIII, 10 /xM for 45 min) . Data represent the mean ± SEM of n = 3. Analogs VII, and VIII alone did not produce contraction, and their structures are listed in Table 3.1.
128 Compound** n EC 50 (/xM) = %E^<*
IBI 4 3.95 ± 0.57 1 2 1 ± 13
VII + IBI 3 16.3± 2.25 109 ± 7
VIII + IBI (no wash) 3 27.8 ± 1 2 . 2 45 ± 2
VIII + IBI (wash) 3 16.6 ± 0.30 1 0 1 ± 7
“ Data represent the mean ± SEM. Chemical structures of analogs are given in Table 3.1. ^ VII (2'-IBI isomer, 100 fjM) or VIII (nifedipine-NCS, 10 /xM) was incubated for 45 min without or with wash. ^ ECgo = median effective concentration. = percent maximal contractile responses using 3 0 /xM phenylephrine as 1 0 0 %.
Table 2.4 Effect of pretreatment with NCS derivatives (2'-IBI isomer [analog VII] and nifedipine-NCS [analog VIII]) on the contractile responses to 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] on rat thoracic aorta.®
129 ♦ Dantrolene + IBI (n = 3) 100 □ IBI (n = 3) □
X c o in ^ 50
-5.0 -4.0 -3.0 Log [IBI] (M)
Figure 2.11 Effect of pretreatment with dantrolene (30 {iM for
20 min) on 2 - ( 4 '-isothiocyanatobenzyl) imidazoline [IBI] induced contractile responses of rat diaphragms. The contractions, expressed as percent tension, were calculated using the maximal IBI-induced contraction as 100%. Data are the mean ± SEM of 3-4 experiments.
130 IBI induced contraction
Treatment n ECso (/iM) % E ^
IBI 3 3 7 . 4 ± 0 . 4 100
Dantrolene +IBI 4 1 2 5 .0 ± 2 . 1 100 d-Tubocurare + IBI 3 37.3 ± 0.9 100
Tissue were pretreated with d-tubocurare 10 [iM for 20 min, or dantrolene (30 fiM for 20 min) . The percentage contractions were calculated and expressed using the maximal IBI-induced contraction as 100%. Data represent the mean ± SEM of n = 3-4.
Table 2.5 Effects of pretreatment with d-tubocurare and dantrolene on 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] induced contractions in rat diaphragm.
131 □ Tolazoline IBI A Naphazoline Naphazoline-NCS o Nifedipine Nifedipine-NCS
100
? 2 Si ÛÛ o _ CVJ II 5 0
-3 -8-9 I' 10 ” [D rug] (M)
Figure 2.12 Displacement of (+) [^H]PN 200-110 by tolazoline naphazoline, nifedipine and corresponding isothiocyanato (NCS) derivatives from rabbit skeletal t-tubule membrane. The binding assay was carried out using 0.3 nM of ( + ) [^H] PN 200-
1 1 0 incubated with 0 . 1 mg/ml protein of rabbit t-tubule membrane in a final volume 4 ml at 25 °C for i hr. Non specific bound was defined by 1 /xM (±) PN 200-110. Values are the mean + SEM of n = 4-6. Notice that all NCS containing derivatives (solid symbols) displaced 84-88% of the total specific binding, and that the slope of curves of -NCS derivatives are steeper than that of nifedipine (with the exception of naphazoline-NCS; also see text).
132 Compound (n) Ki(pM)b ng* %Inhibition=
Tolazoline(4) ND* ND* 18±4
Tolazoline-NCS 49.0±2.0 1.46±0.11 84±2 (4) [IBI] (43.0-55.0) (1.10-1.82)
Naphazoline(4) NDd ND* 47±0
Naphazoline-NCS 2.6±3.3 0 .8 ± 0 . 0 2 8 8 ± 1
(6 ) (1.8-3.4) (0.76-0.86)
Nifedipine 0.012±0.0028 0.97±0.06 99±2 (5) (0.0072-0.015) (0.79-1.14)
Nifedipine-NCS 0.83±0.053 1.37±0.05 86±3
(4) (0 .6 -1 .1 ) (1.21-1.52)
® The binding assay was carried out using 0.3 nM of ( + ) [^H]PN
2 0 0 - 1 1 0 incubated with 0 . 1 mg/ml protein of rabbit t-tubule membrane in a final volume 4 ml at 25 °C for one hour. Non specific bound was defined by 1 fiM (±)PN 200-110 (see method for details). Numbers in parenthesis: 95% confidence interval. Percentage maximal inhibition of specific radioligand binding. ND = not determined. Displacement did not exceed 47% at the highest concentrations used (1 mM).
Table 2.6 Comparative inhibitory constant (Ki) , Hill slope (ng) and percent inhibition of nifedipine, tolazoline, naphazoline and their NCS derivatives in displacing (+) [^H] PN
2 0 0 - 1 1 0 labeled sites from rabbit slceletal t-tubule membrane.®
133 o (-)Bay K 8644 a Naphazollne ■ IBI A Naphazoline-NCS □ Tolazoline
100-
CL
-5 M l l f-4 ^3 10 ’ 10’ 10 ’ 10’ [Drug] (M)
Figure 2.13 Displacement of S(-) [%] Bay K 8644 by tolazoline, naphazoline and corresponding isothiocyanato (NCS) derivatives from rabbit skeletal T-tubule membrane. The binding assay was carried out using 20 nM of S(-) [^H] Bay K 8644 incubated with 0.05 mg/ml protein of rabbit t-tubule membrane in a final volume 0.5 ml at 25 °C for 1 hr. Non-specific bound was defined by 10 fiM (±)PN 200-110. Data represent mean ± SEM of n=4 for each compound.
134 CHAPTER 3
STRUCTURE ACTIVITY RELATIONSHIP STUDIES ON PHARMACOLOGICAL
ACTIVITIES OF 2-(4 •-ISOTHIOCYANATOBENZYL) IMIDAZOLINE [IBI]
ANALOGS IN RAT AORTA AND BOVINE BRAIN
3.1 Introduction
Except for direct antagonism studies for a given drug, as is shown in Chapter 2, other approaches such as structure- activity relationships (SARs) also play an essential role in elucidating the mechanism of action of a drug. The previous
SAR data for IBI action (Lei, 1992) was obtained by testing a series of IBI-related analogs (n=10) in phenoxybenzamine treated rat aorta. The results have shown that the IBI like activities in blood vessels appear to rely on the presence of the NCS group, and unrelated to the imidazoline ring. To further investigate the SAR for the non-imidazoline and non-a- adrenoceptor (AR) activities of IBI in smooth muscles, a number of IBI analogs were synthesized (Table 3.1) . In these
135 studies, it was important to determine whether these IBI analogs possess pharmacological activities through activation of Of-ARs or imidazoline preferring receptors (IPR) , and whether IBI analogs, like their prototype drug IBI, are also mediated by activation of calcium channels in contracting smooth muscles of rat aorta. Selective agonists for a-AR and
IPR and calcium channels may have therapeutic potential for treatment of nasal congestion, hypertension and cardiovascular diseases.
3.2 Specific alms
The purpose of these experiments was to determine the structural requirements of IBI ligands for imidazoline, a- adrenergic receptors or calcium channels by testing the contractile responses of vascular smooth muscles with each IBI analog. Experiments were conducted in rat aortic strips pretreated with 1) an irreversible a-AR antagonist (PBZ); 2)
IPR ligands (idazoxan and cirazoline); 3) or a calcium channel inhibitor (nifedipine). In addition, some of the IBI analogs were selected and evaluated for their ability to bind to a- adrenergic and imidazoline receptor sites in bovine ventrolateral medulla membranes.
136 3.3 Materials and methods
Materials. 2-(4 '-Isothiocyanatobenzyl) imidazoline (IBI, analog I ) and other analogs given in Table 3.1 were synthesized by Drs. Meri Slavics, J. De Los Angeles and Duane
D. Miller (Department of Pharmaceutical Sciences, College of
Pharmacy, The University of Tennessee, Memphis, TN 38136) and provided for these studies. The chemical structures, abbreviations and properties of IBI and related analogs are given in Table 3.1 and Appendix Figure A1. Compounds used were
2-(4 '-aminomethylbenzyl) imidazoline (analog II), 2-(2'- nitrobenzyl) imidazoline (analog III), 2-isothiocyanatophenyl-
2-aminoimidazoline (analog IV), N-(4'-isothiocyanato- phenylethyl) piperidine (analog V), 2-(4'-isothiocyanato- naphthylmethyl) imidazoline (analog VI) , iraidazolino(2,1-c)-
5,10-dihydro-4H-l,3-benzodiazepine-4-thione (analog VII) , 2,6- dimethyl-4- (4-isothiocyanatophenyl-1,4-dihydropyridine-3,5- dicarboxylic acid dimethyl ester (analog VIII), 2-(4'- isothiocyanatophenethyl) imidazoline (analog IX), 2-(4'- isothiocyanatoxymethyl) imidazoline (analog X) , N, N-dimethyl-
2-(4 '-isothiocyanatophenyl) ethylamine (analog XI), 2-(4'- isothiocyanatobenzyl) imidazole (analog XII), 2-(3'- isothiocyanatobenzyl) imidazoline (compound XIII) and 2-(4'-
137 chloroacetamidobenzyl ) imidazoline (XIV) . Except for compound
XII that was in hydrous form, all analogs were available as hydrochloride salts (Table 3.1). Compound VIII was soluble in
1 0 0 % ethanol, and other compounds were readily dissolved in
0.9% NaCl. The molecular weights of these compounds varied from 242-385 (see Appendix Table Al) .
Other chemicals used in the experiments and their sources were as follows: phenoxybenzamine hydrochloride from Smith,
Kline & French (King of Prussia, PA) ; /-phenylephrine hydrochloride, cirazoline from Synthelabo Recherche (Paris,
France) ; idazoxan hydrochloride from Research Biochemicals
Inc. (Natick, MA); tolazoline hydrochloride from CIBA
Pharmaceutical Co. (Summit, NJ) ; and nifedipine (anhydrous) from Sigma Chemical Company, St. Louis, MO) . Chemicals used for preparing physiological salt solution were purchased from commercial suppliers and were of reagent quality.
IBI and most analogs were prepared as 0.01 M solutions in
0.9% NaCl. Analog VIII and nifedipine were dissolved in 100% ethanol, and phenoxybenzamine in 100% methanol, as 0.01 M stock solutions and diluted in 0.9% NaCl for use in experiments.
138 Methods for rat aorta experiments. Animals used in the experiments were male Sprague-Dawley albino rats weighing from
300-350 grams each. Rats were anesthetized by exposure to carbon dioxide and the thoracic artery was dissected and removed. The connective tissue was cleeined off from the aortic segment. Helically-cut thoracic aorta strips (2 x 20 mm, 2-3 strips from each aortic segment) were prepared. Each strip was
placed in a 1 0 ml tissue bath containing a physiological salt solution (PSS) of the following composition (in mM) : NaCl,
118; KCl, 4.7; CaClj. HjO, 2.5; MgSO,. H gO, 0.5; NaHzPO^, 1.0;
NaHCOj, 25; glucose, 11; pH 7.4 according to Venkataraman et al. (1989) , and maintained at 37±0.5 °C with aeration (95% oxygen and 5% carbon dioxide) . Tissues were mounted with a
resting tension of 1 gm and tension changes were monitored with a Grass polygraph (Model 7, Quincy, MA).
About a 90 min period was allowed for each strip to achieve
equilibrium, including a priming with 1 ^M phenylephrine with repeated washes. A concentration-response curve of phenylephrine (1 nM to 30 /xM) was constructed in each strip, and washed several times to return to baseline. Some tissues were then pretreated with 30 nM phenoxybenzamine for 20 min
139 with wash twice, and followed by incubation of selected imidazoline ligands (3 /zM cirazoline for 60 min, or 1 idazoxan for 60 min); other tissues were pretreated with the calcium channel blocker nifedipine (1 /zM) for 60 min. Drugs were added to construct cumulative concentration response curves according to the method of Van Rossum (1963). The time interval between each addition of drug was 30 min for IBI and analogs with similar slow contractions, and 5-10 min for each of the remaining analogs.
The binding experiments were conducted by Marilyn E, Graves and Paul Emsberger (School of Medicine, Case Western Reserve
University, Cleveland, OH 44106). Binding of various IBI analogs to Oz-AR and I^-imidazoline receptor sites was studied in isolated BVLM membranes incubated with ["^I] p- iodoclonidine. Displacement of the radioligand from czg- adrenergic and I^- receptors (control) and from Ii-receptors after alkylation of Ofg-ARs in BVLM with phenoxybenzamine was performed (Felsen et al., 1994). A Student's t test was used for data analysis. Data were expressed as the mean ± SEM.
140 3.4 Results
3.4.1. Effect of phenoxybenzamine pretreatment on the contractile responses of 2-(4*-isothiocyanatobenzyl) imidazoline [IBI] and related analogs II, III, IV, V and VI on rat aorta
Values of ECso and maximal contractions by each analog were calculated and summarized in Table 3.2. All compounds produced agonist activities on rat aortic strips. The rank order of
agonist potency (EC5 0 , fiM) was analogs VI (0.06) > III (0.16)
> I [IBI] (4.0) > IV (7.2) > V (11.2) > II (45.2). The CRC of tolazoline (ECgo = 0.32 fM) is plotted in comparison to analogs
II and III which lack an NCS group (Figure 3.1) . Analog II (4- aminomethylbenzyl analog) produced a less potent contraction
(ECso = 45.2 /xM) than that of tolazoline; whereas analog III
(2 -nitrobenzyl analog) is more potent (ECgo = 0.16 ^M) than tolazoline on smooth muscles. Similarly, the maximal contraction induced by tolazoline (59%) is greater than that produced by analog II (38%) , but smaller than that by analog
III (80%) . The contractile activities of both analogs were, like tolazoline, abolished by PBZ pretreatment (Figure 3.1).
141 In contrast to IBI, which contracted rat aorta with a slow onset of the plateau response (44 min), both of these analogs exhibited a rapid time course (about 2-4 min) for each drug concentration to achieve plateau contraction. These results along with the findings with tolazoline (Chapter 2) suggest that substitution with a 2-nitro or 4 -aminomethyl group on the phenyl ring of tolazoline yield compounds that mimic tolazoline activity on ot-ARs in smooth muscles.
Analogs IV and V contain the NCS group, but differ from IBI in other parts of the chemical structure. Analog IV has a carbon atom on the bridge between the phenyl and imidazoline rings replaced with nitrogen atom; and analog V has the imidazoline ring replaced by an N -ethylpiperidine group. The contractile activities of these analogs in rat aorta were not significantly altered by PBZ pretreatment (30 nM, 20 min with washout twice; Figure 3.2). Thus, these results provide more evidence for the hypothesis that the presence of the NCS group on the phenyl ring of IBI is required for the non-a- adrenoceptor mediated contractile activities in the smooth muscles of rat aorta.
In these experiments, naphazoline-NCS (analog VI) induced contractions of rat aorta that were abolished by PBZ
142 pretreatment (Figure 3.2, left lower panel). It appears that the presence of additional benzene ring attached to 4 ' -NCS group of IBI abolished non-a-AR activity, while retaining significant a-AR activity like that of naphazoline.
3.4.2 Effect of imidazoline receptor ligands (cirazoline and idazoxan) and the calcium channel blocker (nifedipine) on contractile responses of rat aorta to 2- (4 ' -isothiocysmato- benzyl) imidazoline [IBI] and related analogs
The results of these experiments are summarized in table
3.3. All IBI analogs (IX, X, XI and XII) possess an NCS group and were observed to produce contractile responses on rat aorta with similar manner to that of IBI activity, indicating that they produced a slow onset (20-30 min) and a sustained duration of action. Similar to the findings in the thesis research (Lei, 1 9 9 2 ), all compounds showed a similar potency
(ECgo around 1 6 -1 8 jxM) in smooth muscles, except for analog XII that was less potent (ECso=28.6 pM). No significant changes in the concent rat ion - response curves were observed in the presence of either cirazoline or idazoxan in PBZ-retreated tissues (Figure 3,3). However, pretreatment with nifedipine shifted the CRC of each IBI analog to the right and
143 significantly decreased contractions to less than 34% of the
control maximum. These results indicate that these analogs,
like IBI, contract the tissue through a calcium-channel
dependent mechanism.
Figure 3.4 shows a biphasic contractile response to 3 '-IBI
(analog XIII). The first contractile phase (from 0.1-3 /xM)
exhibited a rapid onset of action for each concentration and
an ECso value of 1 ^M. This activity was blocked by
pretreatment with phenoxybenzamine (30 nM, 20 min with wash) .
The second phase displays a more pronounced progressive
contractile response with a longer time course and an ECgo
value of about 2 /xM. This phase was similar to the fashion of
IBI-induced contraction and was blocked by nifedipine (1 /xM,
60 min) , but not by phenoxybenzamine pretreatment.
3.4.3 Binding of selected IBI analogs to bovine ventrolateral medulla (BVLM) membranes
Each compound displaced [^^®I] p-iodoclonidine from I^- and
Oj-AR binding sites, giving the following rank order of
potency (in /xM, see Appendix Table A2 and Figure Al) : X (1.55
> IX (3.16) > IBI (7.41) > XI (7.94) > XIV (16.2). Alkylation
of ttj-ARs with PBZ treatment did not change the rank order
144 potency of the IBI analogs in BVLM membranes. However, the displacement curves of IBI, analogs X and XI were significantly (p < 0.05) shifted to the left by about 2 to 4- fold in phenoxybenz amine - pr e t reat ed membranes, implying that these compounds exhibit a modest binding selectivity for Ii~ receptors over «z-AR sites. Hill slope values (nj were also measured (Appendix Table A2) and show a value of about 1 for each competitor. The changes in Hill slope in the absence and presence of PBZ for most of the sinalogs were not significantly altered (p > 0.05). The exception was IBI (ng = 1.39) which agrees with the results of studies with IBI in displacing
[^H]PN 200-110 from labeled DHP sites in rabbit skeletal t- tubules (Chapter 2).
3.5 Discussion
3.5.1 Non-imidazoline, non-a-adrenoceptor but calcium channel dependent activities of IBI analogs in rat aorta
Previous structure-activity relationship studies on IBI and analogs have shown that IBI contracts smooth muscles of rat aorta via a non-imidazoline, non-a-adrenergic receptor pathway
(Lei, 1992). Several types of IBI analogs were used in these
145 experiments including analogs with 1) a replacement of the NCS
group on phenyl ring with an aminomethyl (sinalog II), 0£-
chloroacetamido (analog XIV) or a nitro group (analog III) ; 2)
an elongated bridge connecting phenyl and imidazoline rings
with a carbon or an oxygen atom (analogs IX and X) ; 3)
replacement of the imidazoline ring with an imidazole,
piperidine ring, or opened ring (N, N-dimethylethylamine)
structure (analogs XII, V and XI) ; and 4) a transfer of NCS
group from 4 ' - to 3 ' - position of the phenyl ring (analog
XIII, 3'-IBI), and the addition of a benzene ring to the
phenyl group of IBI (analog VI, naphazoline-NCS; see Table 3.1
for chemical structrues). These studies were carried out in
tissues pretreated with phenoxybenzamine (PBZ) , with
recognized imidazoline preferring receptor ligands such as
cirazoline («i-AR agonist and I;- receptor ligand) and idazoxan
(«2 - and I2 - receptor ligand) in PBZ treated tissues, or with nifedipine. Similarly to the results obtained from IBI, the
results here indicate that the contractions of IBI analogs
containing the NCS group (analogs IV, V, IX-XIII) were unaffected by imidazoline ligands or PBZ treatments. In
contrast, the contractile responses by analogs which lack the
NCS group (analogs II, III) were abolished in PBZ-treated
146 tissues, suggesting that the effects of the compounds were
mediated via a-AR activation in this tissue. Moreover, the
presence of imidazoline ring in these analogs does not
necessarily ensüDle them to interact with imidazoline
receptors, and further that the presence of NCS group in these
analogs is required for the non-a-adrenoceptor activities
which are susceptible to inhibition of calcium channel
blockers.
Addition of an NCS group at the 4' position of tolazoline,
as in IBI, contracted rat aorta via a non-imidazoline and non-
a-adrenoceptor mechanism. In order to determine the importance
of the position of the NCS group on the phenyl ring of
tolazoline, an analog was synthesized with substitution of -
NCS group at the 3 '-position of 2-benzyl ring of tolazoline,
as in 3'-IBI. This analog was observed to produce a biphasic
contractile response in rat aorta, differing from the monophasic contractions of IBI. Therefore, 3 '-IBI contractions were composed of an a-AR related response at lower
concentrations, and a non-a-AR dependent component of action
at higher concentrations. The results indicated the
requirement of 4 ' position for maintaining the IBI like
activities. These findings are similar to our work with the
147 3'-iodo-IBI analog that also exhibited on a-AR component of action, suggesting the 3', but not 4' substituents, in this
IBI series possess dual contractile actions in the rat thoracic aorta (Figure 3.4).
Another electrophilic functional group in IBI is the positively charge imidazoline ring. Together with the size and shape, the degree of ionization of the imidazoline compound molecule may be crucial in determining the drug activity
(Miller, 1986) . The imidazoline ring has been postulated to interact with the anionic domain of the receptor due to its positive charge (Patil et al., 1991). Thus, questions arose about 1) whether the non-a-adrenoceptor mediated IBI contractile responses in smooth muscle are solely due to the
presence of imidazoline or due to its positive charge; and 2 ) whether other forms of positively charged amine groups induce a similar or different response as compared to IBI. Since imidazoline ring and opened imidazoline ring analogs are secondary and tertiary amines which possess pKa values of about 9, the compound exists predominantly in a charged or protonated species at a pH of 7.4. To evaluate the influence of electronic changes in imidazoline ring on the IBI action, replacement of imidazoline ring with imidazole and opened
148 imidazoline ring were previously made and examined (Lei,
1992). These three compounds produced IBI-like responses in rat aorta which were not affected by the pretreatment with phenoxybenzamine, suggesting that either form of positively charged functional group produces the same non-a-adrenoceptor mediated IBI-like response in rat aorta. Besides, the data indicate that the maximal response but not potency induced by imidazole analog is less than that of the imidazoline analog
(IBI). This may be owing to changes of conformational and electronic factors, since the introduction of double bonds into imidazoline results in a flat ring structure, namely imidazole, as well as less protonated form of imidazole due to lowered pKg value of imidazole (Patil et al., 1991).
Previously, Miller et al. (1986) and Venkataraman et al.
(1991) showed that imidazole analogs are less potent than imidazoline compounds as a-adrenoceptor agonists. Our studies provide further evidence for the differentiation of IBI- sensitive receptor sites from a-adrenoceptor populations.
3.5.2 Binding of IBI analogs to proteins from bovine ventrolateral medulla (BVLM) membrsm.es
IBI was previously reported to compete for [^H]idazoxan
149 binding in guinea pig cerebral cortex and porcine renal cortex
membranes (Hussain et al., 1992). In the present studies,
several IBI analogs were selected to examine the structural
requirements for interaction with a-adrenergic and imidazoline
receptors in bovine brain tissues. Use of BLVM membranes was
chosen because it is a well characterized source of Ii-
imidazoline receptors (Emsberger et al., 1987). The results from these binding studies suggest that the IBI analogs possess Ii-subtype binding properties in bovine brain tissue.
Although the functional effects of IBI and analogs on rat aorta are independent of a-AR and are likely coupled to activation of voltage-gated L-type calcium channels, these compounds appear to interact with both a-AR and Ii-imidazoline receptor sites in bovine brain stem tissue. Interestingly, analog X, which bears a structural similarity to cirazoline and idazoxan, was the most potent agent for displacement of
^^I]p-iodoclonidine from I^-receptor and on a-AR sites.
3.6 Summary smd conclusions
The SAR approach was employed to further investigate the mechanisms of IBI-induced contractile response in rat aorta.
The synthesized IBI analogs (I-XIV) tested in these
150 experiments fall into several groups on the basis of chemical structures: tolazoline derivatives without NCS substitution
(analogs II, III and XIV) ; NCS containing derivatives with modifications on the phenyl ring (analog VI) or the bridge connecting the phenyl and imidazoline rings (analogs IV, V,
IX, X) or the imidazoline ring (analogs XI and XII) ; and IBI isomer by transferring the NCS group from the 4 ' to the 3 '
(analog XIII) or to the 2' (analog VII; see Chapter 2 for data) position.
The results from these experiments showed that tolazoline derivatives (analogs II and III) exhibited similar pharmacological activities to tolazoline with more or less potency and maximal contraction in smooth muscles, and that their contractile responses were abolished by phenoxybenzamine
(PBZ; 30 nM) pretreatment. Similarly, PBZ-treatment abolished the contractions by the prototype compound, tolazoline.
Conversely, NCS containing derivatives (analogs IV, V, IX, X,
XI, XII and XIII) produced PBZ-insensitive contractions with similar manner to that of IBI, and were characterized by a slow onset and sustained contraction in rat aorta. The contractions by selected analogs (IX, X, XI and XII) were resistant to cirazoline/idazoxan pretreatment, and blocked by
151 the calcium channel inhibitor (nifedipine, 1 ^M) . Presence of an additional benzene ring adjacent to the phenyl ring of IBI
(analog VI, naphazoline-NCS) produced a-AR dependent contractile responses in rat aorta, and this a-AR dependency correlates well with the binding data showing a reduced Hill slope value (0.8±0.02; Table 2.6). 3 '-IBI differs from IBI by exhibiting an additional a-AR dependent contraction at lower
concentrations (< 1 /xM) and retained a calcium channel dependent contraction at higher concentrations (> 1 /xM) , the latter effect corresponding closely to that of IBI. Some NCS containing derivatives (IBI analogs: IX, X and XI) were also shown to bind to both imidazoline I^-and a-AR sites in bovine ventrolateral medulla membranes labeled by [^^^I] p- iodoclonidine. Interestingly, the a-chloroacetamido analog
(XIV) , which is an affinity label that lacks the NCS group, also binds to both sites in this preparation (Appendix Table
A2) .
These results of SAR with IBI and analogs indicate that 1) modifications on the phenyl ring of tolazoline that lacks the
NCS group do not alter a-adrenergic receptor (AR) dependency; but, those compounds with an NCS group yield novel vasoconstricting compounds whose effects are mediated via non
152 a-AR mechanism; 2) imidazoline-containing compounds do not necessarily interact with a-AR or imidazoline ring dependent receptors in smooth muscles ; 3) NCS derivatives are sensitive to inhibition by the calcium channel blocker nifedipine; 4) the position of the NCS group is important for maintaining
IBI-like, non-a-AR contractile activities; 5) the size of the phenyl ring attached to the NCS group, as in a naphthalene ring (analog VI), interferes with the interaction of the NCS
group to non-a-AR sites; and 6 ) NCS derivatives may interact with imidazoline and a-AR sites in the CNS.
153 Compound Structure Nomenclature H 2-(4'-lsothiocyanatobenzyl) I II :i SCN N. imidazoline [IBI] 2-(4'-Aminomethylbenzyl) 2HCI II +I/4H2O imidazoline H N 2-(2'-Nitrobenzyl) .HCI ^ N imidazoline NOg 2-lsothiocyanatophenyl-2-amino IV imidazoline
N-(4'-lsothiocyanatophenylethyl) V piperidine SCN 2-(4'-lsothiocyanatonaphthylmethyl) VI imidazoline [Naphazoline-NCS] SCN lmidazolino(2,1 -c)-5,10-dihydro-4H-1,3- VII benzodiazepine-4-thione [Isomer of IBI]
HsCOgC^ ^CHa 2,6-Dimethyl-4-(4-isothiocyanatophenyl) VIII -1,4-dihydropyridine-3,5-dicarboxylic acid dimethyl ester [Nifedipine-NCS] H3 CO0 C
2-(4'-lsothiocyanatophenethyl) IX imidazoline
HCI 2-(4'-lsothiocyanatophenoxymethyl) X imidazoline S C N - O ^ CH, .HCI N,N-Dimethyl-2-(4'-isothiocyanato XI r r " CH, phenyl) ethylamine S C N ^ ^ t N ^'^.IMHgO 2-(4'-lsothiocyanatobenzyl) XII SCN H imidazole .N 2-(3'-lsothiocyanatobenzyl) XIII N^.HCI imidazoline NCS H 2-(4'-Chloroacetamidobenzyl) XIV Cl J l^ l^ ^ N L /.H C I imidazoline H Table 3.1 Chemical structures and abbreviations of 2-(4'-isothiocyanatobenzyl) imidazoline [IBI] analogs. The assigned compound numbers (l-XIV) for each structure along with the name are used in this dissertation. 154 Compound ECc« [uM]b %Emax? (n) Alone With PBZ** Alone With PBZ*
IBI (13) 5.0±0.7 5.2±0.6 116±5 112±5
II (3) 45.2±5.7 NA® 38±5 NA®
III (3) 0.16±0.01 NA® 80±5 NA®
IV (3) 7.2±1.4 6 . 8 ± 0 . 6 100±3 106±3
V (3) 1 1 .2 ± 0 . 1 13.6±2.5 102±4 110±7
VI (3) 0.06±0.01 NA® 80±5 NA®
^ Data represent the mean ± SEM of 3-4 experiments. Chemical structures of these analogs are given in Table 3.1. ^ ECso = median effective concentration. = %Emax = percent maximal contractile responses using 3 0 /xM phenylephrine as 100% which is 679 ± 32 mg of n = 27. ^ PBZ = phenoxybenzamine, 30 nM was incubated for 20 min with washout twice. ® NA = not active. No contractile activity was observed at concentrations up to 0.3 mM.
Table 3.2 Effect of phenoxybenzamine pretreatment on the contractile responses to 2 -(4'-isothiocyanatobenzyl) imidazoline [IBI] analogs on rat aorta.®
155 Treatment*’ n ECso (pM) =
IX alone 4 18.3 ± 0.7 1 0 0 With PBZ & CIR 3 14.9 ± 1.3 103 ± 9 With PBZ &IDA 3 18.1 ± 0 . 6 114 ± 8 With NIF 3 67.5 ± 6.5* 2 1 ± 6 *
X alone 3 17.2 ± 0.9 1 0 0 With PBZ & CIR 3 15.4 ± 0.3 117 ± 11 With PBZ & IDA 3 16.1 ± 1 . 1 115 ± 8 With NIF 3 52.1 ± 0.5* 15 ± 6 *
XI alone 3 16.8 ± 0 . 2 1 0 0 With PBZ & CIR 3 16.9 ± 0.3 91 ± 6 With PBZ & IDA 3 16.2 ± 1 . 1 93 ± 8 With NIF 3 48.5 ± 6 .1 * 26 ± 7*
XII alone 3 28.6 ± 11.7 1 0 0 With PBZ &CIR 3 25.5 ± 8 . 6 116 ± 13 With PBZ & IDA 3 31.9 ± 9.5 94 ± 16 With NIF 3 76.1 ± 13.2* 34 ± 7*
XIII alone 5 12.9 ± 2 . 0 1 0 0 With PBZ 5 22.4 ± 3.7 103 ± 8 With NIF 6 36.7 ± 5.7* 53 ± 6 *
® Data represent the mean ± SEM of 3-6 experiments. Chemical structures are given in Table 3.1. ^ PBZ = phenoxybenzamine, 30 nM was incubated for 20 min with washout twice; CIR = cirazoline, 3 /fM was incubated for 60 min; IDA = idazoxan, 1 ^M was incubated for 60 min; NIF = nifedipine, 1 0 fiM was incubated for 60 min. = ECso = median effective concentration. %Emax = percent maximal contractile responses to the analogs as 1 0 0 %. ® Statistically different from the analog control value (p < 0.05) .
Table 3.3 Effect of imidazoline receptor ligands (cirazoline and idazoxan) and calcium channel blocker (nifedipine) on contractile responses of rat aortic strips to 2-(4'- isothiocyanatobenzyl) imidazoline [IBI] analogs.®
156 Figure 3.1 Effects of pretreatment with phenoxybenzamine (30 nM for 20 min with washout twice) on the contractile responses to tolazoline [upper panel], analog II [middle panel], analog III [lower panel] on rat aortic strips. Data represent the mean ± SEM of n = 3.
157 lOCh • Tolazoline c o o PBZ + tolazoline
1-8 1-71-6 1-5 1-3 lOOi A Analog II A PBZ + analog II
50-
1-8 |-6 -5 ,-31-7 lOOi ♦ Analog III c A PBZ + an alo g I I I o -4-» o
2 "c o 50 o
,-7 1-6 1-5 1-4 1-3 [Drug] (M)
158 Figure 3.2 Effects of pretreatment with phenoxybenzamine (30 nM for 20 min with washout twice) on the contractile responses to IBI [upper left], analog IV [upper right], V [lower right] and VI [lower left] on rat aortic strips. Data represent the mean ± SEM of n = 3-13.
159 ■ IBI ♦ Analog IV □ P B Z + IBI o PBZ + analog IV
100- 100-
c o
* 1-6 1-5 i-4 1-8 1-7 1-5 1-4 10" ® 10 "? -4-» C o A Analog VI CD • Analog V A PBZ + analog VI o PBZ + analog V
100- 100-
50-
ttmy -7 -5 1-7 -6 -4
160 Figure 3.3 Comparative contractile concentration-response curves of analogs IX [upper left], X [upper right] , XI [lower left] and XII [lower right] on rat aorta in the presence of nifedipine (NIP, 1 iiM for 60 min) , phenoxybenzamine (PBZ, 30 nM for 20 min with washout twice) followed with cirazoline (CIR, 3 /xM for 60 min) , or phenoxybenzamine followed with idazoxan (IDA, 10 fiM for 60 min) . The data represent the mean ± SEM of n = 3-4.
161 □ Control ^ PBZ + CIR o NIP o PBZ + IDA
100- 100-
50-
c g 4-) o -6 [Analog IX] (M) [Analog X] (M) o o
100- 100 -
50-
-6 1-4 1-6 [Analog XI] (M) [Analog X II] (M)
162 • Control (n=5) o Phenoxybenzamine (n=5) A Nifedipine (n=6)
100-
o 50-
T I i I I' I I rj T I I I I n r y r 7 -5 -4
Figure 3.4 Effect of phenoxybenzamine (PBZ, 30 nM for 20 min with washes twice) or nifedipine (NIF, 10 [iM for 60 min) pretreatment on contractile responses of 3'-isothiocyanato- benzyl imidazoline (3'-IBI) on rat aorta strips. The data are expressed as the mean ± SEM of n = 5-6.
163 CHAPTER 4
IN VIVO VASCULAR EFFECTS AND GENERAL TOXICITY OF 2-(4'-
ISOTHIOCYANATOBENZYL) IMIDAZOLINE [IBI]
4.1 In vivo Vascular Effects of IBI and Tolazoline In
Anesthetized Rat
4.1.1 Specific aims
Mechanism studies on IBI indicate that IBI possesses a unique pharmacological contraction on vascular smooth muscle which involves an activation of L-type calcium channels; in contrast, the contractile activity of tolazoline, the structural analog of IBI lacking the NCS group, is related to the activation of a-ARs (see Chapter 2). These studies were designed to compare the in vivo cardiovascular activities of
IBI and tolazoline.
4.1.2 Chemicals and methods
Chemicals. The compounds used in the experiments are as
164 follows: ketamine HCl [Ketalar] from Parke-Davis (Morris
Plains, NJ) , xylazine HCl [Rompun] from Bayer Corporation
(Shawnee Mission, Kansas), /-norepinephrine HCl, /- isoproterenol-d-bitartrate, propranolol, acetylcholine chloride and pyrilamine maleate from Sigma Chemical Company
(St. Louis, MO), histamine dihydrochloride from Fisher
Scientific Company, tolazoline HCl [Priscoline] from CIBA
Pharmaceutical Co., Div. CIBA-GEIGY Corp., (Summit, NJ) , atropine sulfate-dl from Mallinkrodt; and heparin sodium from
Solo Pak Laboratories Inc. (Grove Village, XL 60007) . IBI was provided by Professor Duane D. Miller's group (Department of
Pharmaceutical Sciences, College of Pharmacy, The University of Tennessee, Memphis, TN). Compounds were freshly prepared just prior to the experiment and dissolved in normal saline solution. Exceptions were norepinephrine and isoproterenol which were prepared in EDTA (10 mg/1000 ml) with normal saline.
Animals and methods. Animals used were male Sprague-Dawley albino rats ranging from 350-450 grams. Rats were anesthetized i.p. with ketamine (60 mg/kg) and xylazine (15 mg/kg) .
165 Respiratory cannulation was performed to maintain regular breathing activity. The left carotid artery was isolated and
cannulated with a tube filled with heparinized ( 1 0 0 unit/ml) normal saline. The changes in mean artery pressure (MAP, measured as diastolic pressure plus 1/3 pulse pressure) was recorded through a pressure transducer (Gold, P23 ID) , connected to a polygraph instrument. Recovery time was measured as the period required for the MAP to restore the basal line of blood pressure tension. Changes in heart rate were also obtained before and after injection of each compound. The left femoral vein was catheterized as the i.v. injection route for the compounds.
Ten min was allowed for equilibrium completion of the surgical procedure. Each compound was injected i.v. in a volume range of 0.1 - 0.25 ml. The MAP for each compound was obtained by injecting norepinephrine (5 ng/kg), acetylcholine
(30 ng/kg) , histamine (30 ng/kg) , isoproterenol (5 ng/kg) and
IBI (0.5 mg/kg). A fixed amount (0.1 ml) of normal saline was used for flushing of residual solution in the catheter into
the vein, and an interval of 6 - 1 0 min was used in between dosages. Receptor selective blockers, namely, propranolol (1
mg/kg) , atropine ( 1 mg/kg) and pyrilamine ( 2 mg/kg) were
166 injected sequentially to antagonize respective receptors.
The data were analyzed with Student's t-test and expressed as mean ± SEM with significant difference with p < 0.05.
4.1.3 Results
4.1.3.1 IBI induced vascular activities in anesthetized rat
MAP was measured prior to drug treatment and the values in the anesthetized rat averaged 91 ± 7 mm Hg (n = 7). IBI (0.5 mg/kg) alone produced a biphasic response of the MAP. The first phase exhibited a rapid reduction in MAP of -47±5 mm Hg and a recovery time of 253±37 sec. A slowly developed second phase exhibited a small elevation in MAP of 56±7 mm Hg (n = 4) with a time course of 6.8±0.6 min (Table 4.1). Changes in heart rates induced by IBI was insignificant. A three fold higher dose (1.5 mg/kg) of IBI also produced a similar drop in the MAP and a subsequent recovery over a longer period of time
(quantitative aspects were not analyzed).
4.1.3.2 Effects of propranolol, atropine and pyrilamine pretreatment on the mean artery pressure induced by IBI, acetylcholine, histamine and isoproterenol in anesthetized rat
The results of these studies are summarized in Table 4.1.
167 These three blockers were injected i.v. 3 min apart into the rats. They were observed to produce only slight changes in vascular activities (data not shown). After this treatment, isoproterenol was injected and was observed to produce a vasodilation. This isoproterenol-induced vasodilation was significantly inhibited by pretreatment with the blockers, and accompanied with a significantly shorter recovery period needed to return the MAP to basal level. The MAP reduction by histamine was also inhibited by the preparation; however, its recovery time remained unchanged in comparison to the control value. Surprisingly, pretreatment with 1 mg/kg atropine did not significantly affect the vasodilation activities induced by acetylcholine. These results indicate that the system is viable in testing pharmacological blockade to S-adrenoceptors
(AR) and Hi-histaminergic receptors, but not to cholinergic receptors. In comparison to the response to IBI alone, vasodilation by IBI was blocked significantly, and the recovery time was shortened in the presence of the blockers
(Figure 4.1, Table 4.1). These results suggest that IBI- induced in vivo vasodilation may probably be associated with activation of Hi-histaminergic receptors or Pg-ARs.
168 4.1.3.3 Effects of IBI-pretreatment on the mean artery pressure induced by norepinephrine, acetylcholine, histamine in am.esthetized rat
The purpose of these studies was to examine whether IBI possesses in vivo antagonist activities against major receptor systems. The concentration for each drug was approximated according to the ECgo values (IBI had the higher value than the others). The results obtained from these studies are summarized in Table 4.2. While norepinephrine (5 ng/kg) produced an increase in MAP, the remaining drugs
(acetylcholine, 30 ng/kg; histamine, 30 ng/kg; isoproterenol,
5 ng/kg) induced an attenuation in MAP. The MAP values induced by all these drugs were not significantly affected by IBI (0.5 mg/kg), as was measured by repeating each compound following the injection of 0.5 mg/kg IBI. The changes in the recovery time and heart rate by IBI were also statistically insignificant (data not shown in the table). These results indicate that IBI does not act as a blocker for adrenergic, muscarinic and histaminergic receptors in vivo system.
169 4.1.3.4 Effects of tolazoline pretreatment on norepinephrine induced vascular activities in anesthetized rat
As shown in Figure 4.1, tolazoline (10 mg/kg, i.v.) was observed to produce a rapid decrease in MAP. Tolazoline reduced the NE-mediated increase in MAP (Figure 4.1, lower panel) , suggesting an a-AR blockade activity of tolazoline in vivo. These results contrast with the data obtained from thoracic aorta studies that exhibited an a-AR mediated partial contraction by tolazoline (see Chapter 2) . It appears that, while tolazoline possesses a partial a^-AR agonist activity in rat in vitro; it also acts as a blocker for a-AR mediated vascular responses in rat in vivo.
4.2 Estimation of Acute Toxicity and Behavioral Activity of
IBI and Tolazoline in Mice
4.2.1 Specific aims
The studies were undertaken to 1) evaluate the relative toxicity (acute mean lethal dose), and changes in behavioral activities of IBI in mice; and 2) to compare these profiles between IBI and tolazoline.
170 4.2.2 Chemicals amd methods
Animals and chemicals. Tolazoline HCl was provided CIBA
Pharmaceutical Corporation. IBI was provided by Professor
Duane D. Miller's group (Department of Pharmaceutical
Sciences, College of Pharmacy, University of Tennessee,
Memphis, TN) . CD-I albino male mice weighing 25-30 grams
(Charles Rivers Laboratory, Wilmington, Mass.) were used.
Methods. The test was designed according to the description by Miller and Tainter (1944). Before drug injection, each mouse was tested on the rod (diameter = 4 cm) of a rotor rod
(Dodge Manufacturing Corporation, Mishawaka, IN) rotating at
6 rpm. Those animals which fell twice from the moving bar within 3 min were removed from testing group. Both i.v. and i.p. injection of drugs were used for IBI or tolazoline groups. Drugs were dissolved in normal saline at a concentration that limits the final injection volume within
0.5 ml for either administrative route. While the mouse was comfortably confined in a small cage, the drug was given by injection i.v. into the tail vein using a 27 gauge needle. The tail was wiped with xylene to improve clarity of the tail vein for the injections.
171 After drug injection, each animal was re-tested on the rotor rod at 5, 15, 60 min and 24 hours, and behavioral changes were also observed. Mean lethal dose was obtained by observing the number of animals which died within one hour i.v. postinjection.
4.2.3. Results
A. Mean lethal dose. Both IBI and tolazoline showed a similar i.v. lethal dose range (Table 4.3), which approximated a value of 30-40 mg/kg. The animals injected with either IBI or tolazoline died quickly (within 1-3 min) or survived over the days of these experiments. This value of tolazoline observed from our lab is within the range of the value of 56.7 mg/kg reported by Takeuchi's group (1986). This observation indicates that IBI, a compound which differs from tolazoline by only the presence of an isothiocyanato (NCS) group, does not significantly increase the acute toxicity in mice as compared to tolazoline.
B. Behavior changes. Before drug injection, mice showed active behavioral activities such as climbing, exploring, sniffing.
While animals treated by tolazoline did not show apparent
172 changes in the above mentioned activities, animals treated by
IBI (30 mg/kg or higher, i.v.) were observed to show hypoactivity and frequently doing abdominal stretches. These phenomena continued during the second day of the experiment.
C. Drug’-induced changes in mobility. At a concentration of 20 mg/kg (about half of the estimated mean lethal dose values obtained from i.v. injection) , IBI or tolazoline was injected i.p. in mice (three/group) , and mobility was detected by testing the mice on the rotor rod for three min at postinjection 5, 15, 60 minutes and 24 hours. All mice injected with tolazoline were observed to stay on the rotating bar. Only one of the IBI injected mice fell twice from the rotor-rod on the trial at the 5 min postinjection time (Table
4.3, lower panel) . These results suggests that IBI may have a lower threshold range in altering the mobility of mice than does tolazoline.
In comparison to the results obtained from in vivo cardiovascular test with IBI in rat, IBI displayed a biphasic vascular activity (a rapid vasodilation followed by a slow vasoconstrictive response) as seen in changes of the mean artery pressure, whereas tolazoline produced a rapid
173 vasodilation. Tolazoline also blocked norepinephrine-induced vasoconstriction. Thus, it appears that the 4'-NCS substitution on tolazoline does change its in vivo vascular activity in rat whereas its effects on behavior and acute toxicity were similar to that of IBI in mice.
4.3 Summary and conclusions
In vivo vascular activities in rat and parameters on toxicity, behavior and neurology in mice induced by IBI and tolazoline were compared. Unlike the vasoconstriction in vitro, IBI (0.5 mg/kg) produced a biphasic response on rat mean artery pressure (MAP) , a rapid reduction in MAP (-47±5 mmHg with a recovery time of 253±37 sec) in the first phase and a slowly developed increase in MAP (56±7 mmHg with a time
course of 6 .8±0 . 6 min) in the second phase. These IBI-induced activities were attenuated by i.v. sequential injection of the
rat with propranolol (l mg/kg) , atropine ( 1 mg/kg) and
pyrilamine (2 mg/kg) , which similarly blocked the vasoactivities of histamine and isoproterenol. Pretreatment of rat with IBI (i.v., 0.5 mg/kg, 10 min) did not affect the MAP induced by norepinephrine, acetylcholine, histamine and isoproterenol. In contrast, tolazoline (10 mg/kg) produced
174 apparent vasodilation as seen in reduced MAP, and i.v. pretreatment of rat with tolazoline inhibited norepinephrine- induced vasoconstriction. It thus is proposed that, in vivo,
IBI is capable of modulating vascular tension presumably by activation of Pg-^drsnergic or Hi-histaminergic receptors, but does not possess antagonist activities on pharmacologic receptors of the a- or p-adrenergic, histamine or acetylcholine type; and that tolazoline possesses antagonist activities against vasoconstriction through inhibition of a- adrenoceptors. In studies on mice, IBI and tolazoline produced a mean lethal dose about 3 0-40 mg/kg and 57 mg/kg, respectively. IBI-treated mice exhibited hypoactivity and abdominal stretching as opposed to minor changes in behavioral activities in tolazoline-treated mice. In the studies with rotor rod testing, IBI (20 mg/kg) may have a lower threshold range in altering the mobility in mice.
175 Control 100 mmHg L 131 (0.5 mg/kg) 1 min
After propranolol, atropine and pyrilamine
IBI (0.5 mg/kg) V
Tolazoline (10 mg/kg)
NE (5 ng/kg) NE (5 ng/kg)
Figure 4.1 Representative tracings of the mean artery pressure of anesthetized rats. UPPER: IBI (0.5 mg/kg) induced biphasic vascular response; MIDDLE: Blockade of IBI induced vascular
activities by propranolol ( 1 mg/kg), atropine ( 1 mg/kg) and pyrilamine (2 mg/kg) injected 10 min before IBI; LOWER: Blockade of norepinephrine induced vasoconstriction by
tolazoline ( 1 0 mg/kg) .
17? MAP (mmHg) ^ Recovery Time (sec) TJeartRate ( /min) **
Cmpd® alone +blockers* alone +blockers* alone+blockers'
IBI -47±5 -8 ± 2 *** 253±37 36±14** -1±13 3±18
ACH -32±4 -39±4 32±2 101±34 48±27 -5±25
HIST -48±4 -29±2** 78±42 105±31 24±9 1 ± 1 2
ISO -61±1 -29±2*** 480±35 113±39** 59±19 46±17
^ Data represent mean ± SEM with n = 3-4. MAP (mean artery pressure) = diastolic pressure plus 1/3 pulse pressure in mmHg. The values denote the changes in MAP measured with a normal average pressure of 91 ± 7 mmHg (n=7) . = Recovery time = the period required for the MAP to restore the basal line of the pressure tension. Heart Rate = difference of heart beats per min before and after each drug addition. ® All compounds were i.v. injected with the concentration of acetylcholine (ACH) = 30 ng/kg, histamine (HIST) = 30 ng/kg, isoproterenol (ISO) = 5 ng/kg, IBI = 0.5 mg/kg. ^ Three blockers were injected i.p. with a sequence of propranolol ( 1 mg/kg) , atropine ( 1 mg/kg) and pyrilamine ( 2 mg/kg) , respectively, and incubated for 1 0 min.
** p < 0 . 0 1
*** p < 0 . 0 0 1
Table 4.1 Effect of propranolol, atropine and pyrilamine pretreatment on the cardiovascular activities induced by acetylcholine, histamine, isoproterenol and IBI in anesthetized rat*.
177 Treatment n NE ACH HIST ISO (mmHg)
Control^ 3 72±14 -32±4 -48±4 -61±1
After IBI= 3 52±16 -48±8 -43±3 -73±15
® Data represent mean ± SEM and no significant statistical difference were observed between the related data. MAP (mean artery pressure) = diastolic pressure plus 1/3 pulse pressure in mmHg. The values denote the changes in MAP measured with normal average pressure of 91 ± 7 mmHg (n = 7). ^ All compounds were i.v. injected with the concentration of norepinephrine = 5 ng/kg, acetylcholine =3 0 ng/kg, histamine = 30 ng/kg, isoproterenol = 5 ng/kg, respectively. = IBI (0.5 mg/kg) was injected and injected 10 min before the sequential administration of agonists. Changes in the MAP before and after IBI-treatment were not significant (P > 0.05) .
Table 4.2 Effect of IBI-pretreatment on the mean artery pressure (MAP) induced by norepinephrine (NE), acetylcholine (ACH), histamine (HIST) and isoproterenol (ISO) in anesthetized rat®.
178 JËI Tolazoline. dose (mg/kg) route n observation* observation*
2 0 i.p. 3 alive alive
30 i.v. 2 alive/dead
40 i.v. 1 alive dead
50 i.v. 1 dead
60 i.v. 1 dead dead
1 0 0 i.v. 1 dead
* Acute toxicity was studied by observing the number of animals that died within 24 h after-treatment.
Table 4.3 Evaluation of the mean lethal i.v. dose on mice by IBI and tolazoline
179 . - ISI Tglaz-g-line dose (mg/kg) route n rotor rod test** rotor rod test*’
2 0 i.p. 3 1 fell none fell
30 i.v. 2 stayed
40 i.v. 1 stayed
60 i.v. 1
® Acute toxicity was studied by observing the number of animals that died within 24 h after-treatment. ^ Mice were tested during 3 min trial at 5 min postinjection.
Table 4.4 Effects of IBI and tolazoline treatment on mobility of mice in the rotor-rod test.
180 REFERENCES
Ariên, E.J.: Affinity and intrinsic activity in the theory of competitive inhibition, part I. problems and theory. Arch. Int. Pharmacodyn. 99: 32-49, 1954.
Atlas, D. : Clonidine-displacing substances (CDS) and its putative imidazoline receptors. Biochem. Pharmacol. 41: 1541- 1549, 1991.
Atlas, D. and Burstein, Y. : Isolation and partial purification of clonidine-displacing endogenous brain substance. Eur. J. Biochem 144: 287-293, 1984a.
Atlas, D. and Burstein, Y. : Isolation of an endogenous clonidine-displacing substance from rat brain. FESEBS Lett. 170: 387-391, 1984b.
Atlas, D., Diamant, S., Pales, H.M. and Panne 11, L. : The brain's own clonidine: purification and characterization of endogenous clonidine displacing substance from brain. J. Cardiovasc. Pharmacol. 10(Suppl. 12): sl22-sl27, 1987.
Atlas, D., Plotec, Y. and Miskin, I.: An affinity label for Of; - adrenergic receptors in rat brain. Eur. J. Biochem. 126: 537-541, 1982.
Ahlquist, R.P.: A study of the adrenergic receptors. Am. J. Physiol. 153: 586-600. 1948.
Arunlakshana, 0. and Schild , H.O.: Some quantitative uses of drug antagonists. Br. J. Pharmacol. 14: 48-58, 1959.
Banning, J.W., Rice, P.J., Miller, D.D., Ruffolo, R.R., Hamada, A. and Patil, P.N.: Differences in adrenoceptor activation by stereoisomeric catecholimidazolines and catecholamines. In: Neuronal and extra neuronal events in
181 autonomie pharmacology. Edited by Fleming W., Langer S., Graefe K-H,, Weiner N., Raver Press, New York, p. 167-180, 1984.
Battaglia, G., Shannon, M., Borgundvaag, B . and Titeler, M. : Properties of [^H] prazosin labeled «i-adrenergic receptors in rat brain and porcine neurointermediate lobe tissue. J. Neurochem. 41: 538-542, 1983.
Beam K.G., Adam, B.A., Nidome, T., Numa, S. and Tanabe, T.: Function of truncated dihydropyridine receptor as both voltage sensor and calcium channel. Nature 360: 169-171, 1992.
Bean, B.P.: Classes of calcium channels in vertebrate cells. Ann. Rev. Physiol. 51: 367-384, 1989.
Bean, B.P., Sturek, M., Puga, A. and Hermsmeyer, K. : Calcium channels in muscle cells isolated from rat mesenteric arteries: modulation by dihydropyridine drugs. Circ. Res. 59: 229-235, 1986.
Bechem, M., Gross, R., Hebisch, S. and Schramm, M.: Calcium agonists: a new class of inotropic drugs. Basic Res. Cardiol. 84 (Suppl. 1):105-116, 1989.
Bechem, M. and Schramm, M. : Calcium-agonists. J. Mol. Cell. Cardiol. 19: II63-II75, 1987.
Becker, P.L. and Fay, F.S: Photobleaching of fura-2 and its effect on determination of calcium concentration. Am. J. Physiol. 253: C613-C618, 1987.
Berridge, M.J. and Irvine, R.F.: Inositol phosphates and cell signaling. Nature 341: 197-217, 1989.
Berridge, M.J.: Inositol triphosphate and calcium signaling. Nature 361: 315-324, 1993.
Berthelsen, S. and Pettinger, W.A.: A functional basis for the classification of a-adrenergic receptors. Life Sci. 21: 594- 606, 1977.
Blaxall, H.S., Murphy, T.J., Baker, J.C., Ray C and Bylund,
182 D.B. : Characterization of the ajc-adrenergic receptor subtype in the opossum kidney and in the OK cell line. J. Pharmacol. Exp. Ther. 259: 322-329, 1991.
Bolton, T.B.: Mechanism of action of transmitters and other substances on smooth muscle. Physiol. Rev. 59: 606-718, 1979.
Borkowski, K. and Finch, L. : A comparison of the cardiovascular effects of centrally administered clonidine and adrenaline in the anaesthetized rat. J. Pharm. Pharmacol. 31: 16-19, 1979.
Bousquet, P.J., Feldman, J. and Schwartz, J. : Central cardiovascular effects of a-adrenergic drugs : differences between catecholamines and imidazolines. J. Pharmacol. Exp. Ther. 230: 232-236, 1984.
Bowman, W.C. and Rand, M.J. : Textbook of pharmacology, 2nd edition, Blackwell Scientific Publications. Oxford, p. 11-31, 1980a.
Bowman, W.C. and Rand, M.J. : Textbook of pharmacology, 2nd edition, Blackwell Scientific Publications. Oxford, p. 1-32, 1980b.
Boyajian, C.L. and Leslie, F.M.: Pharmacological evidence for aj-adrenoceptor heterogeneity : differential binding properties of [%] rauwolscine and [^H] idazoxan in rat brain. J. Pharmacol. Exp. Ther. 241: 1092-1098, 1987a.
Boyajian, C.L., Loughlin, S.E. and Leslie, F.M.: Anatomical evidence for a-adrenoceptor heterogeneity: differential autoradiographic distributions of [^H] rauwolscine and [^H] idazoxan in rat brain. J. Pharmacol. Exp. Ther. 241: 1079- 1091, 1987b.
Brading, A.F. and Sneddon, P.: Evidence for multiple sources of calcium for activation of the contractile mechanism of guinea pig taenia coli on stimulation with carbachol. Br. J. Pharmacol. 70: 229-240, 1980.
Bricca, G., Dontenwill, M., Molines, A., Feldman, J., Belcourt, A. and Bousquet, P. : The imidazoline preferring
183 receptor: binding studies in bovine, rat and human brainstem. Eur. J. Pharmacol. 162: 1-9, 1989.
Bùlbring, E. : Observations on the isolated phrenic nerve diaphragm preparation of the rat. Brit. J. Pharmacol. 1: 38- 61, 1946.
Bylund, D.B.: Subtypes of ttj-adrenoceptors : pharmacological and molecular biological evidence converge. Trends Pharmacol. Sci. 9: 356-361, 1988.
Bylund, D.B.: Subtypes of «i- and « 2 -adrenergic receptors. The
FASEB Journal 6 : 832-840, 1992.
Bylund, D.B.: Pharmacological characteristics of (Xg-adrenergic receptor subtypes. In: The Imidazoline Receptor: Pharmacology, Functions, Ligands, and Relevance to Biology and Medicine, edited by Donald, J. Reis, Pascal Bousquet, and Angelo Parini. Ann New York Acad. Sci., vol. 763: 1-7, 1995.
Bylund, D.B., Eikenberg, D.C., Hieble, J.P., Langer, S.Z., Lefkowitz, R.J., Minneman, K.P., Molinoff, P.B., Ruffolo, R.R.J. and Trendelenburg, U. : IV International union of pharmacology nomenclature of adrenoceptors. Pharmacological Reviews 46: 121-136, 1996.
Bylund, D.B., Ray-Prenger, C. and Murphy, T.J.: «ja" and «2 3 - adrenoceptor subtypes : antagonist binding in tissues and cell lines containing only one receptor subtype. J. Pharmacol. 162: 1-9, 1989.
Bylund, D.B. and Yamamura, J.I.: Methods for receptor binding. In: Methods in Neurotransmitter Receptor Analysis, edited by Yamamura, H.I. New York: Reven Press, p. 1-35, 1990.
Catterall, W.A. : Functional subunit structure of voltage-gated calcium channels. Science 253: 1499-1500, 1991.
Catterall, W.A. and Striessnig, J. : Receptor sites for Ca^* channel antagonists. Trends Pharmacol. Sci. 13: 256-262, 1992.
Chamley-Campbell, J.H., Cambell, G.R. and Ross, R.: The smooth muscle cell in culture. Physiol. Rev. 59: 1-61, 1979.
184 Cheng, Y. and Prusoff, W.H.: Relationship between the inhibition constant (Ki) and the concentration of an inhibitor that causes a 50% inhibition (ICgo) of an enzymatic reaction. Biochem. Pharmacol. 22: 3099-3108, 1973.
Clark, A.J.: The antagonism of acetylcholine by atropine. J. Physiol. 61: 547-556, 1926.
Clark, A. J. and Raventos, J. : The antagonism of acetylcholine and of quaternary ammonium salts. Quant. J. Exp. Physiol. 26: 30: 86-90, 1937.
Cohen, S.A. and Strydom, D.J.: Amino acid analysis utilizing phenylisothiocyanate derivative. Anal. Chem. 174: 1-16, 1988.
Coupry, I., Atlas, D., Podevin, R.A., Uzielli, I. and Parini, A.: Imidazoline guanidinium receptive site in renal proximal tubule : asymmetric distribution, regulation by cations, and interaction with an endogenous clonidine displacing substance. J. Pharmacol. Exp. Ther. 252: 293-299, 1989.
Cubeddu,, L.X. : New «i-adrenergic receptor antagonists for the treatment of hypertension : role of vascular a receptors in the control of peripheral resistance (review). Am. Heart J. 116: 133-162, 1988.
Decker, N., Leclerc, G., Quennedey, M.C., Rouot, B. and Schwartz, J. : Study of two alkylating derivatives; the p- isothiocyanato- and the p-methylisothiocyanato-clonidine. Eur. J. Pharmacol. 93: 205-211, 1983.
Docker, N., Leclerc, G., Quennedey, M.C., Rouot, B. and Schwartz, J. : Study of two alkylating derivatives: the p- isothiocyanato and the p-methylisothiocyanato clonidine. Eur. J. Pharmacol. 93: 205-211, 1993.
Dollery, C.T. and Reis, J.L.: Central noradrenergic neurons and the cardiovascular action of clonidine in the rabbit. Br. J. Pharmacol. 47: 206-216, 1973.
Dube, G.P., Baik, Y.H., Vaghy, P.L. and Schwartz, A.: Nitrendipine potentiates Bay K 8644-induced contraction of isolated porcine coronary artery: evidence for functionally
185 distinct dihydropyridine receptor subtypes. Biochem. Biophys. Res. Comm. 128: 1295-1302, 1985a.
Dube, G.P., Baik, Y.H. and Schwartz, A.: Effects of a novel calcium channel agonist dihydropyridine analog. Bay K 8644, on pig coronary artery : biphasic mechanical response and paradoxical potentiation of contraction by diltiazem and nimodipine. J. Cardiovasc. Pharmacol. 7: 377-389, 1985b.
Dumont, L, Williams, J.S., Vaghy, P.L. and Schwartz A.: Characteristics of the phenylalkylamine binding site in canine cardiac sarcolemmal membranes. Basic Research in Cardiology. 83: 369-357, 1988.
Easson, L.H. and Stedman E. : Studies on the relationship between chemical constitution and physiological action. IV. Molecular dissymetry and physiological action. Biochem. J. 27: 1257-1266, 1933.
Ehrlich, B.E., Kaftan, E., Bezprozvannaya, S. and Bezprozvanny, I. : The pharmacology of intracellular Ca^* release channels. Trends Pharmacol. Sci. 15: 145-152, 1994.
Emsberger, P., Graves, M.E., Graff, L.M., Zakieh, N . , Nguyen, P., Collins, L.A., Westbrooks, K.L. and Johnson, G.G. : Ii- Imidazoline receptors: definition, characterization, distribution, and transmembrane signaling. In: The Imidazoline Receptor: Pharmacology, Functions, Ligands, and Relevance to Biology and Medicine, edited by Donald, J. Reis, Pascal Bousquet, and Angelo Parini. Ann New York Acad. Sci., vol. 763: 22-42, 1995.
Emsberger, P., Meeley, M.P. and Reis, D.J. : Clonidine binds to imidazole binding sites as well as o^-adrenoceptor in the ventrolateral medulla. Eur. J. Pharmacol. 134: 1-13, 1987.
Emsberger, P., Meeley, M.P., Mann, J.J. and Reis, D.J.: An endogenous clonidine-like substance bindings preferentially to imidazole binding sties in the ventrolateral medulla labeled by [^H]para-aminoclonidine. J. Hypertension 4 (Suppl. 5) : S109-S111, 1986.
Emsberger, P., Giuliano, R., Willette, R.N., Granata, A.R.
186 and Reis, D.J.: Hypotensive action of clonidine analogs correlates with binding affinity at imidazole and not (Xg- adrenergic receptors in the rostral ventrolateral medulla. J.
Hypertension 6 (suppl. 4): S554-S557, 1988.
Emsberger, P., Timothy, H.D., Graff, L.M., Schafer, S.G. and Cheisten, M.O.: Moxonidine, a centrally acting antihypertensive agent, is a selective ligand for Ii-sites. J. Pharmacol. Exp. Ther. 264: 172-182, 1993.
Fatt, P. and Katz, B. : The electrical properties of crustacean muscle fibers. J. Physiol. 129: 171-204, 1953.
Felsen, D., Emsberger, P., Sutaria, P.M., Nehat, R.J., Nguyen, P., May, M., Breslin, D.S., Marion, D.N. and Vaughan, D.J.: Identification, localization and functional analysis of imidazoline and a-adrenergic receptors in canine prostate. J. Pharmacol. Exp. Ther. 268: 1063-1071, 1994.
Flaim, S.F., Wright, D.E., Gleason, M.M. and Swigart, S.C. : Neuromedin-K (tachykinin neuropeptide) has potent cardiotonic effects and stimulates leak channel calcium influx in rabbit aorta without altering vascular tone. Circulation 70: 11-167, 1984.
Fleckenstein, A. : Specific pharmacology of calcium in myocardium. Cardiac pacemakers and vascular smooth muscle. Ann. Rev. Pharmacol. Toxicol. 17: 149-166, 1977.
Fleckenstein, A. : History of calcium antagonists. Cir. Res. 52: 13-116, 1983.
Franckowiak, G., Bechem, M., Schramm, M. and Thomas, G. : The optical isomers of 1,4-dihydropyridine Bay K 8644 show opposite effects on calcium channels. Eur. J. Pharmacol. 114: 223-226, 1985.
Furchgott, R.F.: The use of S-haloalkylamines in the differentiation of receptors and in the determination of dissociation constants of receptor-agonist complexes. In: Advances in Dmg Research, edited by Harper, N.J. and Simmonds, A.B., New York Academic Press, vol. 3, p. 21-55, 1966.
187 Furchgott, R.F. and Zawadzki, J.V. : The obligatory role of endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature 288: 373-376, 1980.
Gilbert, E.K., Weaver, B.A. and Rembold, C.M. : Depolarization decreases the [Ca^*] ^ sensitivity of myosin light-chain kinase in arterial smooth muscle : comparison of aquorin and fura 2 [Ca^*] i estimates. The FASEB Journal 5: 2593-2599, 1991.
Gilbertson, T.A., Scobey, R. and Wilson, M. : Permeation of calcium ions through non-NMDA glutamate channels in retinal bipolar cells. Science 251: 1113-1115, 1991.
Glossmann, H. and Ferry, D.R.: Assay for calcium channel. Methods Enzymol. 109: 513-550, 1985.
Godfraind, T.R.: Actions of nifedipine on calcium fluxes and contraction in isolated arteries. J. Pharmacol. Exp. Ther. 224: 443-450, 1983.
Golenhofen, K., Wagner, B. and Weston, A.H. : Calcium systems of smooth muscle and their selective inhibition. In: Excitation-coupling in smooth muscle, edited by R. Casteels, T Godfraind, J.C. Ruegg. Elsevier, Amsterdam, p. 131-136, 1977.
Gomez, R.E., Emsberger, P., Feinland, G. and Reis, D.J. : Rilmenidine lowers arterial pressure via imidazole receptors in brainstem Cl area. Eur. J. Pharmacol. 195: 181-191, 1991.
Grynkiewicz, G., Poenie, M. and Tsien, R.Y.: A new generation of calcium indicators with greatly improved fluorescence properties. J. Biol. Chem. 260: 3440-3450, 1985.
Hallam, T.J. and Rink, T.J.: Receptor-mediated Ca^* entry: diversity of function and mechanism. Trends Pharmacol. Sci. 10: 8-10, 1989.
Han, C ., Abel, P.W. and Minneman, K.P. : Heterogeneity of Oi- adrenergic receptors revealed by chloroethylclonidine. Mol. Pharmacol. 32: 505-510, 1987.
Haxhiu, M.A., Dreshaj, I., Schafer, S.G. and Emsberger, P.:
188 Selective antihypertensive action of moxonidine is mediated mainly by Ii-imidazoline receptors in the rostral ventrolateral medulla. J. Cardiovas. Pharmacol. 24 (Suppl): S1-S8, 1994.
Head, G.A. : Importance of imidazoline receptors in the cardiovascular actions of centrally acting anti-hypertensive agents. In: 2nd International Symposium on Imidazoline Receptors, New York City, p. s40, 1994.
Hieble, J.P. and Ruffolo, R.R.: Imidazoline receptors: historical perspective. Fundam. Clin. Pharmacol. 6 (Suppl. 1) S7-S13, 1992.
Hieble, J.P. and Ruffolo, R.R.: Possible structural and functional relationships between imidazoline receptors and (Xg- adrenoceptors. In: The Imidazoline Receptor: Pharmacology, Functions, Ligands, and Relevance to Biology and Medicine, edited by Donald, J. Reis, Pascal Bousquet, and Angelo Parini. Ann New York Acad. Sci., vol. 763: 8-21, 1995.
Hill, A.V. : The possible effects of the aggregation of the molecules of hemoglobin on its dissociation curves. J. Physiol. 40: iv-vii, 1910.
Hille, B.: Calcium channels. Chapter 4, In: Ionic channels in excitable membranes. Sinauer Publishers, Sunderland, Massachusetts, 1984.
Himpens, B. and Somlyo, A. P.: Free-calcium and force transients during depolarization and pharmacomechanical coupling in guinea-pig smooth muscle. J. Physiol. 395: 507- 530, 1988.
Hoffman, B.B. and Lefkowitz, R.J.: Adrenergic receptor antagonists. In: Goodman and Gilman's The Pharmacological
Basis of Therapeutics. 8 th edition, Pergamon Press, New York, p. 221-224, 1990.
Hussain, J.F., Patil, P.N., Miller, D.D., Kendall, D.A. and Wilson, V.G.: Isothiocyanate tolazoline interacts with non adrenoceptor, imidazoline binding sites labeled by [^H] idazoxan (abstract). Fundam. Clin. Pharmacol. 6 : 56, 1992.
189 Hwa, J.J. and Beven, J.A.: Stretch-dependent (myogenic) tone in rabbit ear resistance arteries. Am. J. Physiol. 250: H87- H95, 1986.
Janis, R.A. and Triggle D.J.: Drugs acting on calcium channels. In: Calcium channels : their properties, functions, regulation, and clinical relevance, edited by Leon Hurwitz, L. Donald Partidge, John K. Leach, CRC Press, p. 195-250, 1991.
Johnson, R.D. and Minneman, K.P.: Differentiation of aj.- adrenergic receptors linked to phosphatidylinosital turnover and cyclic AMP accumulation in rat brain. Mol. Pharmacol. 32: 505-510, 1987.
Karaki, H.: Calcium localization and sensitivity in vascular smooth muscle. Trends Pharmacol. Sci. 10: 320-325, 1989.
Kaumann, A. J. : Is there a third heart fi-adrenoceptors? Trends Pharmacol. Sci. 10: 316-320, 1989.
Kenakin, T.: The classification of seven transmembrane receptors in recombinant expression systems. Pharmacol. Rev. 48: 413-463, 1996.
Kobinger, W. : Central a-adrenergic systems as targets for hypotensive drugs. Rev. Physiol. Biochem. Pharmacol. 81: 40- 100, 1978.
Kokubun, S., Prod'hom, B., Becker, C., Porzig, H. and Reuter, H. : Studies on calcium channels in intact cardiac cells: voltage dependent effects and cooperative interactions of dihydropyridine enantiomers. Mol. Pharmacol. 30: 571-584, 1986.
Kuriyama, H., Kitamury, K. and Nabata, H. : Pharmacological and physiological significance of ion channels and factors that modulate them in vascular tissues. Pharmacol. Rev. 47: 387- 573, 1995.
Lands, A.M., Arnold, A., McAuliff, J.P., Luduena, P.P. and Brown, T.G.: Differentiation of receptor systems activated by sympathomimetic amines. Nature 214: 597-598, 1967.
190 Langer, S.Z.: Presynaptic regulation of catecholamine release. Br. J. Pharmacol. 60: 481-497, 1974.
Langmuir, I.J.: The adsorption of gases on plane surfaces of glass, mica and platinum. J. Amer. Chem. Soc. 40: 1361-13 72, 1918.
Lehmann, J., Koenig-Berard, E. and Vitou, P.: The imidazoline preferring receptors. Life Sci. 45: 1609-1615, 1989.
Lei, L. : Some pharmacological activities of 2-(4'~ isothiocyanatobenzyl) imidazoline in rat aorta, guinea pig ileum and human platelets. Thesis, The Ohio State University, Columbus, OH. p. 29-37, 1992.
Lei, L., Slavica, M., Emsberger, P., Graves, M.E., Patil, P.N., Miller, D.D. and Feller, D.R.: Calcium channel-dependent and Ii-imidazoline receptor binding properties of 2-(4'- isothiocyanatobenzyl) imidazoline [IBI] analogs in vascular and brain tissues. In: The Imidazoline Receptor: Pharmacology, Functions, Ligands, and Relevance to Biology and Medicine, edited by Donald, J. Reis, Pascal Bousquet, and Angelo Parini. Ann New York Acad. Sci., vol. 763: 283-287, 1995.
Lei, L., Slavica, M., Romstedt, K., Patil, P.N., Miller, D.D. and Feller, D.R.: Non-a-adrenergic actions of 2-(4'- isothiocyanatobenzyl) imidazoline (IBI) analogs in rat aorta (abstract). The Pharmacologist 34: 148, 1992.
Lepran, I. and Papp, J.G.: Effect of Moxonidine on arrhythmias induced by coronary artery occlusion and reperfusion. J. Cardiovasc Pharmacol. 24(Suppl): S9-S15, 1994.
Lesh, R.E., Marks, A.R., Somlyo, A.V., Fleishcer, S. and Somlyo, A.P.: Anti-ryanodine receptor antibody binding sites in vascular and endocardial endothelium. Circ. Res. 72: 481- 488, 1993.
Li, G., Regunathan, S., Barrow, C.J., Eshraghi, J., Cooper, R. and Reis, D.J.: Agmatine is an endogenous clonidine-displacing substance in brain. Science 263: 966-969, 1994.
Limbird L: Receptors linked to inhibition of adenylate
191 cyclase: additional signaling mechanisms. The FASEB Journal: 2686-2695, 1974.
Link, R.E., Desai, K., Hain, L., Stevens, M.E., Chruscinski, A., Bernstein, D., Barsh, G.S. and Kobilka, B.K. : Cardiovascular regulation in mice lacking CKg - adrenergic receptor subtypes b and c. Science 273: 803-805, 1996.
Lloyd C. Miller and M. L. Tainter: Estimation of the EDgo and its error by means of logarithmic-probit graph paper. Proc. Soc. Exp. Biol. Med. 57: 261-264, 1944.
MacMillan, L.B., Hein, L., Smith, M.S., Piascik, M.T. and Limbird, L.E.: Central Hypotensive effects of the ttja" adrenergic receptor subtype. Science 273: 801-803, 1996.
Mansfield, E.: Basic statistics. W. W. Norton & Company Inc., New York, p. 215-356, 1986.
McDonald, T.F., Pelzer, S., Trautwein, W. and Pelzer., D.J.: Regulation and modulation of calcium channels in cardiac, skeletal, and smooth muscle cells. Physiol. Reviews 74: 365- 507, 1994.
McPherson, P.S. and Campbell, K. P. : The ryanodine receptor/calcium release channel. J. Biol. Chem. 268: 13675- 13768, 1993.
McPherson, G.A.: Kinetic, EBDA, Ligand, Lowry: A collection of radioligand binding analysis programs, Cambridge : Elsevier Science Publishers, 1985.
Michel, M.C. and Emsberger, P. : Keeping an eye on the I site: imidazoline-preferring receptors. Trends Pharmacol. Sci. 10: 369-370, 1992.
Michel, A.D., Loury D.N. and Whiting, R.L.: Differences between the 0(2 - adrenoceptors in rat submaxillary gland and the
Oja- and (%2B-adrenoceptor subtypes. Br. J. Pharmacol. 98: 890- 897, 1989a.
Michel, M.C. and Insel, A.P.: Are there multiple imidazoline binding sites? Trends Pharmacol. Sci. 10: 342-344, 1989b.
192 Miller, D.D.: Structure-activity relationship and drug design Ann. Rep. Med. Chem. 21 (283): 422-432, 1986.
Miller, L.C. and Tainter, M.L. ; Estimation of the ED 50 and its error by means of logarithmic-probit graph paper. Proc. Soc. Exp. Biol. Med. 57: 261-264, 1944.
Minneman, K.P., Han, C. and Abel, P.W.: Comparison of a^- adrenergic receptor subtypes distinguished by chloroethylclonidine and WB4101. Mol. Pharmacol. 33: 509-514, 1988.
Minneman, K.P. and Molinoff, P.B.: Classification and quantitation of S-adrenergic receptor subtypes. Biochem. Pharmacol. 29: 1317-1323, 1980.
Moore, E.D., Becker, P.L., Itch, T. and Fay, F.S.: Calcium homeostasis in single intact smooth muscle cells. Advan. Exp. Med. Biol. 304: 171-183, 1991.
Morrow, A.L. and Creese, I.: Characterization of (Xi-adrenergic receptor subtypes in rat brain: a re-evaluation of [^H]yohimbine. Biochem Biophys. Res. Comm. 97: 321-330, 1986.
Murphy, R.A., Beme, R.M. and Levy, M.N. : Muscle/The cardiovascular system. In: Physiology, edited by Robert M Beme, Matthew N. Levy. 3rd ed. Mosby Press, p. 281-326; 397- 417, 1993.
Murphy, Y.J. and Perry, D.W.: Characterization of Oj- adrenergic receptors in the OK cell, an opossum kidney cell line. J. Pharmacol. Exp. Ther. 244: 571-578, 1988.
Nakayama, H., Taki, M., Striessnig, J., Grossmann, H., Catteral, W.A. and Kanaoka, Y. : Identification of 1,4- dihydropyridine binding regions within the (%i-subunit of skeletal muscle Ca^* channels by photoaffinity labeling with diazepine. Proc. Natl. Acad. Sci. USA 8 8 : 9203-9207, 1991.
Newman, A.H.: Irreversible ligands for drug receptor characterization. Ann. Rep. Med. Chem. 251: 271-279, 1991.
Nichols, A.J., Ruffolo, R.R.: Structure activity relationship
193 for a-adrenoceptor agonist and antagonists. In: a- Adrenoceptors: Molecular Biology, Biochemistry and Pharmacology, edited by Ruffolo R.R., Karger Basel, p. 75-114, 1991.
Nickerson, M.: Receptor occupancy and tissue response. Nature 78: 697-698, 1956.
Ollivier J.-P. and Christen, M.O.: I^-Imidazoline receptor agonists in the treatment of hypertension : an appraisal of clinical experience. J. Cardiovas. Pharmacol. 24(Suppl): S39- S48, 1994.
Oriowo, M.A., Hieble, J.P. and Ruffolo, R.R.: Evidence for heterogeneity of prejunctional c^-adrenoceptors. Pharmacology 43: 1-13, 1990.
Papp, J.G. and Ollivier, J.-P.: Selective stimulation of Ii- imidazoline receptors as a novel mechanism for antihypertensive action: focus on moxonidine. J. Cardiovas. Pharmacol. 24(Suppl): v-vi, 1994.
Parini, A., Coupry, I., Graham, R.M., Uzielli, I., Atlas, D. and Lanier, S.M.: Characterization of an imidazoline/ guanidinium receptive site distinct from the c^-adrenergic receptor. J. Biol. Chem. 264: 11874-11878, 1989.
Partridge, L.D. : The physiology of calcium channels: an overview. In: Calcium channels : their properties, functions, regulation, and clinical relevance, edited by L. Hurwitz, L.D. Partridge and J.K. Leach, p. 63-69, 1991.
Patil, P.N., Feller, D.R. and Miller, D.D.: Molecular aspects of the receptor activation by imidazoline: an overview. J. Neural Transm. 34 (Suppl): 187-194, 1991.
Perry, B.C. and U' Prichard, D.C.: [^H] Rauwolscine (a- yohimbine): a specific antagonist radioligand for brain « 2 " adrenergic receptors. Eur. J. Pharmacol. 76: 461-464, 1981.
Regunathan, S. and Reis, D.J.: Imidazoline receptors and their endogenous ligands. Ann. Rev. Pharmacol. Toxicol. 36: 511-514, 1996.
194 Reis, D.J., Regunathan, S., Wang, H., Feinstein, D.L. and Meeley, M.P. : Imidazoline receptors in the nervous system.
Fundam. Clin. Pharmacol. 6 (Suppl. 1): S23-S29, 1992.
Remaury, A., and Paris, J. : The insulin-secreting cell line RINmSF, expresses an a^radrenoceptor and nonadrenergic idazoxan-binding sites. J. Pharmacol. Exp. Ther. 260: 417-426, 1992.
Rice, P.J., Hamada, A., Miller, D.D. and Patil, P.N.: Asymmetric catecholimidazolines : affinity and efficacy relationships at the a-adrenoceptor in rat aorta. J. Pharmacol. Exp. Ther. 242: 121-130, 1987.
Rosenthal, H.E.: A graphic method for the determination and presentation of binding parameters in complex systems. Anal Biochem. 20: 568-580, 1967.
Rosenthal, W., Hescheler, J., Trautwein, W., Schultz, G. : Control of voltage dependent calcium channels by G-protein coupled receptors. The FASEB Journal. 2: 2784-2790, 1988.
Ruffolo, R,R.: Use of isolated, physiologically responding tissues to investigate neurotransmitter receptors. In: Monographs in Neural Science, edited by Karger, B. New York: Karger, Basel, p. 53-84, 1984.
Ruffolo, R.R., Dillard, R.D., Rosing, E.L. and Waddell, J. E.: Receptor interactions of imidazolines. III. structure- activity relationships governing a-adrenergic receptor occupation and receptor activation derivatives in rat aorta. J. Pharmacol. Exp. Ther. 211: 733-380, 1979a.
Ruffolo, R.R., Dillard, R.D., Waddell, J. E. and Yaden, E.L. : Receptor interactions of imidazolines. II. affinity and efficacies of hydroxy-substituted tolazoline derivatives in rat aorta. J. Pharmacol. Exp. Ther. 211: 74-79, 1979b.
Ruffolo, R.R., Rice, P.J., Patil, P.N., Hamada, A. and Miller, D.D. : Differences in the applicapability of the Easson - Stedman hypothesis to the a^- and aj- adrenergic effects of phenethylamines and imidazolines. Eur. J. Pharmacol. 8 6 : 471- 475, 1983.
195 Ruffolo, R.R., Turowski, B.S. and Patil, P.N.: Lock of cross desensitization between structurally dissimilar a-adrenoceptor agonist, J. Pharm. Pharmacol. 29: 378-380, 1977.
Ruffolo, R.R. and Waddell, J. E. : Receptor interactions of imidazolines. IX. Cirazoline is an - adrenergic agonist and an « 2 -adrenergic antagonist. J. Pharmacol. Exp. Ther. 222: 29- 36, 1982.
Ruffolo, R.R., Yaden, E.L., Waddell, J. E. and Dillard, R.D. : Receptor interactions of imidazolines. IV. significance of carbon bridge separating phenyl and imidazoline rings of tolazoline-like a-adrenergic imidazolines. J. Pharmacol. Exp. Ther. 214: 535-540, 1980.
Ruffolo, R.R. and Zeid, R.L. : Relationship between a- adrenoceptor agonist, cirazoline, and the aj-adrenoceptor agonist, BHT-933, in canine saphenous vein. J. Pharmacol. Exp. Ther. 235: 636-643, 1985.
Sanders, J., Miller, D.D. and Patil, P.N. : a-Adrenergic and histaminergic effects of tolazoline-like imidazolines. J. Pharmacol. Exp. Ther. 195: 362-371, 1975.
Scanlon, M., Williams, D.A. and Fay F. S.: A calcium insensitive form of fura - 2 associated with polymorphomuclear leukocytes. Assessment and accurate calcium measurements. J. Biol. Chem. 262: 6308-6312, 1987.
Scatchard, G.: The attractions of proteins for small molecules and ions. Ann. J. Pharmacol. 17: 153-163, 1949.
Schild, H.O. : pA, a new scale for the measurement of drug antagonism, Br. J. Pharmacol. 2: 189-206, 1947.
Schneider, N.W., Jona, I., Vegh, M., Martonosi, A. : Voltage dependent charge movement of skeletal muscles : a possible step in exc it at ion- contract ion coupling. Nature 242 : 244-246, 1973.
Schramm, M., Thomas, G., Towart, R. and Fanckowiak, G. : Novel dihydropyridines with positive inotropic action through activation of calcium channels. Nature 303: 535-537, 1983.
196 Sengupta, J.N., Hamada, A., Miller, D.D. and Patil, P.N.: Interaction of enantiomers of hydroxy tolazoline with adrenoceptors. Naunyn-Schmiedeberg's Arch Pharmacol. 335: 391- 396, 1987.
Shams, G., Venkat araman, B.V., Hmada, A., Miller, D.D., Patil, P.N. and Feller, D.R.: Diversity of the pharmacological actions of some tolazoline analogs in human platelets and rat aorta. Eur. J. Pharmacol. 199: 315-323, 1991.
Simonneaux V., Ebadi, M. and Bylund, D.B: Identification and characterization of Ofgo-adrenergic receptor in bovine pineal gland. Mol. Pharmacol. 40: 235-241, 1991.
Slavica, M., Lei, L., Patil, P.N., Kerezy, A., Feller, D.R. and Miller, D.D.: The synthesis and biological activities of a new set of irreversibly acting 2 -(4'-isothiocyanatobenzyl) imidazoline (IBI) analogs in rat thoracic aorta. J. Med. Chem. 37: 1874-1881, 1994.
Somlyo, A.V. and Somlyo, A. P.: Electro-mechanical and pharmacomechanical coupling in vascular smooth muscle. J. Pharmacol. Exp. Ther. 159: 129-159, 1968.
Spedding, M. : Assessment of "Ca^*-antagonist" effects of drugs in K*-depolarized smooth muscle : differentiation of antagonist subgroups. Naunyn-Schmiedeberg's Pharmacol. 318: 234-240, 1982.
Spedding, M. and Paoletti, R.: III. Classification of calcium channels and the sites of action of drugs modifying channel function. Pharmacol, Rev. 44: 363-376, 1992.
Starke, K., Gothert, M. and Kilbinger, J. : Modulation of neurotransmitter release by presynaptic autoreceptors. Physiol. Rev. 69: 864-898, 1989.
Stephenson, R.P.: A modification of receptor theory. Br. J. Pharmacol. 11: 379-393, 1956.
Strange, P.G.: The use of radiochemicals for studying receptors. In: Radiochemicals in Biomedical Research, edited by E.A. Evans and K.G. Oldham, John Wiley & Sons, N.Y., N.Y.,
197 1988.
Strazis, K.P. and Fox, A.W. : Malignant hyperthermia: a review of published cases. Anesth. Analy. 77(2): 297-304, 1993.
Strichartz, G., Rando, T. and Wang, G.K. : An integrated view of the molecular toxicology of sodium channel gating in excitable cells. Ann. Rev. Neurosci. 10: 237-267, 1987.
Striessnig, J., Murphy, B.J. and Catterall, W.A. : Dihydropyridine receptor of a-type calcium channels: identification of binding domains for [^H] ( + )PN 200-110 and [%]azidopine within the «i-subunit. Proc. Nat. Acad. Sci. USA
8 8 : 10769-10773, 1991.
Struyker, B.H., De Boer, J., Smeets, J., Lien, E.J. and Van Rossum, J.: Structure activity relationships of central and peripheral a-adrenergic activities of imidazoline derivatives. Life Sci. 15: 887-899, 1974.
Suematsu, E., Hirata, M., Hashimoto, T. and Kuriyama, J. : Inositol 1,4,5-triphosphate releases calcium from intracellular store sites in skinned cells of porcine coronary artery. Biochem. Biophys. Res. Comm. 120: 481-485, 1984.
Takeuchi, K., Akatsuka, K. suid Kasuya, Y. : Pharmacological studies of imidazoline derivatives. Ill actions of cardiovascular system and acute toxicity. J. Pharmacobio-Dyn. 9: 395-4041, 1986.
Tallarida, R.J. and Murray, R.B.: Manual of pharmacologic calculations with computer programs. Springer-Verlag New York Inc., 1981.
Tanabe, T., Takeshima, J., Mikami, A., Flockerzi, V., Takanashi, H., Kangawa, K., Kojima, M., Matsuo, H., Hirose, T. and Numa, S.: Primary structure of the receptor for calcium channel blocker from skeletal muscle. Nature 328: 313-318, 1987.
Taylor P. and Inel, P.A. : Molecular basis of pharmacologic selectivity. In: Principles of Drug Action. The Basis of Pharmacology. 3rd ed. Edited by William B. Pratt and Palmer
198 Taylor. Churchill Livingstone, p. 53-56, 1990.
Tesson, F., Prip-Buus, C, Lemoine, A., Pegorier, J.P. and Parini, A. : Subcellular distribution of imidazoline- guanidinium-receptive sites in human and rabbit liver. Major localization to the mitochondrial outer membrane. J. Biol. Chem. 266: 155-160, 1991.
Tibirica, E., Feldman, J., Mermet, C., Gonon, F. and Bousquet, P. : An imidazoline specific mechanism for the hypotensive effect of clonidine: a study with yohimbine and idazoxan. J. Pharmacol. Exp. Ther. 256: 606-613, 1991.
Timmermans, PBMWM, Schoop, AMC., Kwa, H.Y. and Van Zwieten, P.A. : Characterization of a-adrenoceptors participating in the central hypotensive and sedative effects of clonidine using yohimbine, rauwolscine and corynanthine. Eur. J. Pharmacol. 70: 7-15, 1981.
Tsien, R. W. : Calcium channel in excitable membranes. Ann. Rev. Physiol. 45: 341-58, 1983.
Tsien, R.W. and Tsien, R.Y.: Calcium channels, stores, and oscillations. Ann. Rev. Cell Biol. 6 : 715-670, 1990.
Vaghy, P.L., McKenna, E., Itagaki, K. and Schwartz, A. : Resolution of the identity of the calcium antagonist receptor in skeletal muscle. Trends Pharmacol. Sci. 9: 398-402, 1988.
Vaghy, P.L., Striessnig, J., Miwa, K., Knaus, H.G., Itagaki K., McKenna, E., Glossmann H. and Schwartz, A. : Identification of a novel l,4-dihydropyridine- and phenylalkylamine-binding polypeptide in calcium channel preparations. J. Biol. Chem. 262: 14337-14342, 1987a.
Vaghy, P.L., Williams, J.S. and Schwartz, A.: Receptor pharmacology of calcium entry blocking agents. Am. J. Cardiol. 59: 9A-17A, 1987b.
Van Rossum, J.M.: Cumulative dose-response curves. II. Technique for the making of dose-response curves in isolated organs and the evaluation of drug parameters. Arch. Int. Pharmacodyn. Ther. 143: 299-330, 1963.
199 Venkataraman, B.V., Hamada, A., Shams, G., Miller, D.D., Feller, D.R. and Patil, P.N.: Paradoxical effects of isothiocyanate analog of tolazoline on rat aorta and human platelets. Blood Vessels 26: 335-346, 1989.
Waud, D.R.: On the measurement of the affinity of partial agonists for receptors. J. Pharmacol. Exp. Ther. 170: 117-122, 1969.
Wikberg, J.E.S. and Uhlen, S.: Further characterization of the guinea pig cerebral cortex idazoxan receptor-selective drug. J. Neurochem. 55: 192-203, 1990.
Williams, J.S., Grupp, I.L., Grupp, G., Vaghy, P.L., Dumont, L., Schwartz, A, Yatani, A., Hamilton, S. and Brown, A.M: Profile of the oppositely acting enantiomers of the dihydropyridine 202,791 in cardiac preparations: receptor binding, electrophysiological, and pharmacological studies. Biochem. Biophys. Res. Comm. 131: 13-21, 1985.
Wilson, K.M. and Minneman, K. P. : Regional variations in ttj.- adrenergic receptor subtypes in rat brain. J. Neurochem. 43: 1782-1786, 1989.
Wilson, K.M. and Minneman, K. P. : Pertussis toxin inhibits norepinephrine-stimulated inositol phosphate formation in primary brain cell cultures. Mol. Pharmacol. 38: 274-281, 1990.
Wilson, V . G . : A letter (to Dr. Patil) from Department of Physiology and Pharmacology, Medical School, Queen's Medical Center Nottingham, NG7, 2UH, London, UK. personal communication, 10-10-1991.
Yanagisawa, M., Kurihara, J., Kimura, S., Tomobe, Y., Kobayashi, M., Mittsui, Y., Yazaki, Y., Goto, K. and Masaki, T. : A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 332: 441-445, 1988.
Zaagsma, J. and Hollenga, C. : Distribution and function of atypical (S3 ) adrenoceptors. In: Adrenoceptors: Structure, Mechanisms, Function, edited by Szabadi E., Birkhanser, Germany, p. 47-59, 1991.
200 Zonnenschein, R., Diamant, S. and Atlas, D. : Imidazoline receptors in rat liver cells: a novel receptor or a subtype of
« 2 -adrenoceptors? Eur. J. Pharmacol. 190: 203-215, 1990.
201 APPENDIX
Compound Salt f.w. m.p. solubility (g/mol) (°C)
IBI (I) HCl 253.74 151 - 155 water II 2HC1+ WHzO 266.69 191 - 293 water III HCl 241.68 246 - 247 water IV 2HC1+ WHzO 259.25 134 - 136 water V HCl 282.84 217 - 219 water VI HCl 303.82 239 - 243 water
VII HCl 253.74 2 0 1 - 203 water VIII 358.41 172 - 173 alcohol IX HCl 276.77 171 - 172 water X HCl 269.75 133 - 135 water XI HCl 242.81 174 - 175 water
XII XHzO 210.71 209 - 2 1 1 water XIII HCl 253.74 water XIV HCl 288.17 214 - 216 water
® See Table 3.1 for chemical structure.
Table Al. Synthesized Compounds®
202 Analog ICcn» nt,® Control With PBZ® Control With EBZ®
IBI 7.41±0.34 2.34±0.1lf 1.39±0.8S 0.96±0.87 IX 3 .16±0.29 3.38±0.39 1.05±0.81 0.79±0.B5 X 1.55±0.40 0.42±0.05f 0.98±0.78 0.74±0.S7 XI 7.94±0.91 4.17±0.48f 0.96±0.89 0.92±0.81 XIV 16.2±1.67 8.13±0.58 0.84±0.8S 0.83±0.89
® Data represent mean ± SEM of n = 3-5. Chemical structures are given in table 3.1, and results and discussion are in Chapter 3.
^ IC50 = concentration for 50% maximal binding displacement. ng = Hill number or Hill slope. Hill coefficient describing the manner of receptor-ligand interaction. ® PBZ = phenoxybenzamine (1 /xM) with N-ethylmaleimide (0.25 mM) to irreversibly inactivate Oj-ARs. ^ Statistically different from the control (p < 0.05).
Appendix Table A2. Comparison of ligand displacing IC50 values of 2-(4 '-isothiocyanatobenzyl) imidazoline [IBI] analogs before and after PBZ-treatment on bovine ventrolateral medulla membranes bound to [^^®I] p-iodoclonidine
203 Inhibition of Specific [^^^I]PIC Binding to Bovine Ventrolateral Medulla Membranes
100-
4)
II 50 □ IBI o Analog IX II o Analog X X Analog XI A Analog XIV
J "I- ' '— T g' 10" ^ 10~ ® 10"® 10-5 [Drug] (M)
Appendix Figure Al. Displacing of p-iodoclonidine by 2- (4'-isothiocyanatobenzyl) imidazoline [IBI] analogs before and after PBZ-treatment (1 piM phenoxybenzamine with 0.25 mM ethylmaleimide to irreversibly inactivate aj-ARs) on bovine ventrolateral medulla membranes. Data represent mean ± SEM of n = 3-5. Chemical structures are given in Table 3.1, and results and discussion are in Chanter 3.
204