Pharmacological evaluation of the β-adrenoceptor agonist properties of N-benzyl substituted trimetoquinol analogues

European Journal of Pharmacology, 184 (1990) 251-256

251

Elsevier EJP 51427

Pharmacological evaluation of the fl-adrenoceptor agonist properties of N-benzyl substituted trimetoquinol analogues Gamal Shams 1, Marc W. Harrold

2, Barbara

Grazyl, Duane D. Miller and Dennis R. Feller

Divisions of Pharmacology, and Medicinal Chemistry and Pharmacognosy, College of Pharmacy, The Ohio State University, Columbus, OH 43210 U.S.A.

Received 2 May 1990, accepted 8 May 1990

The ill- and fl2-adrenoceptor agonist properties of trimetoquinol (TMQ, I) and N-benzyl ring substituted TMQ analogues (II, 4'-methylbenzylTMQ; III, 4'-chloro-benzylTMQ; IV, 4'-methoxybenzylTMQ; V, 4'-nitrobenzylTMQ; VI, 3',4'-dichlorobenzylTMQ; and VII, 4"-aminobenzylTMQ) were studied in guinea pig atria and trachea. All compounds gave concentration-dependent responses in atria and trachea, and the rank order of fl-adrenoceptor agonist potency was I > VII > II > V > IV > VI > III and I > VII > IV = VI > V > III > II, respectively. Whereas the N-benzyl substitution reduced potency for fl-agonist activity, the fl2/fll-selectivity ratio was enhanced by addition of groups to the N-benzyl ring, and the rank order of/~2-selectivity was VI (10-fold) > III (8-fold) = IV (8-fold) > VII (3-fold) > V -- I > II. The results show that varying the nature of substituents on the N-benzyl ring of TMQ produces compounds which retain greater fl2-selectivity. fl-Adrenoceptor agonists; Trimetoquinol; N-Benzyl-substituted analogues; fl2/fll-Selectivity ratio; Atria (guinea-pig); Trachea (guinea-pig)

1. Introduction Trimetoquinol (TMQ, 1-(3',4',5'-trimethoxybenzyl)-6,7-dihydroxy-l,2,3,4-tetrahydroisoquinoline) is a cyclized phenethylamine and is used for the treatment of bronchial asthma ( Y a m a m u r a and Kishmoto, 1968). T M Q was the most active fl2-adrenoceptor agonist of a series of tetrahydroisoquinoline analogues prepared by Iwasawa and Kiyomoto (1967), and was more potent for relaxa-

1 Current address: Department of Pharmacology, College of Veterinary Medicine, Zagazig University, Zagazig, Egypt. 2 Current address: Department of Medicinal Chemistry, College of Pharmacy, Duquesne University, Pittsburgh, PA 15282, U.S.A. Correspondence to: D.R. Feller, Ph.D. Division of Pharmacology, College of Pharmacy, The Ohio State University, Columbus, OH 43210, U.S.A.

tion of bronchial smooth muscle and less active on the heart (chronotropy) than isoproterenol (Farmer et al., 1970; M u k h o p a d h y a y et al., 1982). We (Mukhopadhyay et al., 1985; Ahn et al., 1988) also showed that T M Q possesses e n d o p e r o x i d e / thromboxane A 2 (TXA2) antagonist properties, and that the tetrahydroisoquinoline class is a unique chemical entity among T X A 2 antagonists (Lefer and Darius, 1987; Kattelman et al., 1986). Since the discovery of T M Q (Iwasawa and Kiyomoto, 1967), our laboratories have prepared and evaluated m a n y T M Q analogues in ill- and fl2-adrenoceptor and T X A 2 receptor systems (Mukhopadhyay et al., 1982; 1985; Clark et al., 1988; Adejare et al., 1986; Fedyna et al., 1987). Adejare et al, (1986) and Fedyna et al. (1987) reported on the synthesis and the/3-adrenoceptor agonist and platelet antithromboxane activities of a series of N-substituted T M Q analogues. These

0014-2999/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

252 findings show that the larger the size of the N-substituent, the lower was the /3-adrenergic activity and the higher was the f12//31-tissue selectivity and platelet antithromboxane activity. N-BenzylTMQ was the most potent in this series against U46619 (a TXA 2 agonist)-induced platelet aggregation, and the most selective fl2-adrenoceptor agonist. In this regard, N-benzyl TMQ was 41- and 15-fold more fl2-selective than isoproterenol and TMQ in isolated tracheal and atrial tissues (Fedyna et al., 1987). In an attempt to improve the profile of biological activities for TMQ, a new chemical series of N-benzyl ring substituted analogues of TMQ was synthesized in our laboratories (Harrold et al., 1988) (fig. 1). Among these compounds there was a direct correlation between platelet antfthromboxane activity and the presence of electron-donating substituents on the N-benzyl ring. In contrast, the antithromboxane activities of these compounds in rat thoracic aorta were not dependent on substituent induced changes in the electron density of the N-benzyl ring. Whereas substitution on the N-benzyl aromatic ring is important for differentiating the biological effects of these compounds as antagonists on TXA 2 receptor systems, no studies have examined the profile of/3-adren-

oceptor stimulatory activities for this chemical series of TMQ analogues. Our present study was undertaken to evaluate the 131- and fl2-adrenoceptor activities of the Nbenzyl substituted TMQ analogues and to determine whether they possess an improved/32//31selectivity as compared to TMQ, as was previously reported for N-benzylTMQ (Adejare et al., 1986; Fedyna et al., 1987). The /3-adrenoceptor potencies of these new analogues were assessed in guinea pig atria and trachea as representative /31- and /32-adrenoceptor systems.

2. Materials and methods

2.1. Drugs Racemic-TMQ was synthesized by Miller et al. (1975). The N-benzyl TMQ analogues were synthesized in our laboratories for use in these biological experiments (Harrold et al., 1988). Structures and abbreviations used for TMQ and these analogues are shown in fig. 1. The compounds used and their sources are as follows: carbachol chloride (Aldrich Chemical Co., Milwaukee, WI), and (-)-isoproterenol HC1

R1

II NH ~

HN O~ ~R OCH 3

TMQ

2

H

O

~

OCH 3 -OCH3

V -OCH3 OCH 3

No. Compound Trimetoquinol (TMQ) 4'-Methylbenzyl TMQ II 4'-Chlorobenzyl TMQ III 4'-Methoxybenzyl TMQ IV 4'-Nitrobenzyl TMQ V 3',4'-Dichlorobenzyl TMQ Vl 4'-Aminobenzyl TMQ vii

OCH 3 RI

CH3 C1

R2

H H

OCH3 H

NO 2 H Cl Cl NH 2

H

Fig. 1. Chemicalstructuresof trimetoquinol(TMQ) and N-benzylsubstitutedanalogues.

253

(Sigma Chemical Co., St. Louis, MO), (-)-propranolol HCI (Ayerst Laboratories, Inc., New York, NY), reserpine (Ciba-Geigy, Summit, N J).

2.2. Guinea pig right atria and trachea Male guinea pigs (Hartley strain, Glenn Carr, Columbus, OH) weighing between 300-450 g were reserpinized (5 mg/kg i.p.) 12-16 h before experiments. Animals were killed using CO 2 and the atria and trachea were quickly removed by standard procedures (The Staff, University of Edinburgh, 1968). Tracheal strips and spontaneously beating atria were prepared and mounted in 10 ml tissue baths filled with Krebs-Henseleit solution warmed to 37°C and bubbled with 95% 02-5% CO 2 as described previously (Mukhopadhyay et al., 1982). Tensions of 1 and 3 g were applied to the atria and trachea, respectively. Following a 1 h equilibration period, the ,Baadrenoceptor properties of the TMQ analogues in spontaneously beating guinea pig fight atria were analyzed by constructing cumulative concentration-response curves of the compounds according to the method of Van Rossum (1963). All drug responses were expressed in terms of a maximal response to 10 -5 M isoproterenol added after the completion of each concentration response curve. Following a 1 h equilibration period, 3 × 10 -7 M carbachol was added to induce a 60-70% maximal contraction of tracheal strips. Cumulative concentration response curves of the fl2-adrenoceptor (relaxant) actions of TMQ analogues were constructed as described by Van Rossum (1963). All drug-induced responses were expressed in terms of the response to 10 -5 M isoproterenol added after the completion of each concentration response curve. Maximum effects (mean + S.E.) of isoproterenol on guinea pig atria (beats/min) and trachea (tension changes, gm) were 120 +_4.8 (n = 11) and 2.07 4-0.15 ( n = 11), respectively. Only one drug was tested on each tissue preparation. In experiments where propranolol was used to demonstrate the fl-adrenoceptor activities of TMQ analogues, tissues were incubated for 30 min with 3 × 10 -8 M propranolol prior to the construction of the concentration response curve to each TMQ analogue, and data were analyzed for the determination of the pK Bvalue for propranolol.

3. Results

3.1. Chronotropic effects The chronotropic effects of trimetoquinol (TMQ) and the N-benzyl substituted TMQ analogues in guinea pig atria are shown in fig. 2. All N-benzyl TMQ analogues are agonists which are less potent than TMQ giving a rank order of agonist potency of TMQ (I)> 4'-aminobenzyl-

PERCENT MAXIMAL CHRONOTROPIC RESPONSE (ISO= I00%)' I00

ATRIA

. ~ e - - .

{/*

I/,i-J

80 60 40 20 0 I

L

L

-9

-8

-7

-6

I

I

-5

-4

•

,.....- { 7 ~

-5

PERCENT MAXIMAL RELAXATION (ISO=lO0%) TRACHEA

I00

40 20

o,

, -9

-8

,

.

-7 -6 -5 -4 LOG MOLAR CONCENTRATION

-5

Fig. 2. Concentration-response curves of trimetoquinol (TMQ) and N-benzyl substituted analogues in guinea pig atria (upper panel) and guinea pig trachea (lower panel). Key: TMQ, I, O; 4'-aminobenzylTMQ, VII, A; 4'-nitrobenzylTMQ, V, I ; 4'chlorobenzylTMQ, III, ©; 4'-methylbenzylTMQ, II, zx, 4'methoxybenzylTMQ, IV, n; and 3',4'-dichlorobenzylTMQ, VI. 0 - The responses are expressed relative to the maximal response elicited by 10 -5 M isoproterenol and the data represent the means + S.E. of 4-10 tissues.

254

TMQ (VII) > 4'-methylbenzylTMQ (II) > 4'nitrobenzylTMQ (V) > 4'-methoxybenzylTMQ (IV) > 3',4'-dichlorobenzylTMQ (VI) > 4'-chlorobenzylTMQ (II) (table 1). Several N-benzyl TMQ analogues were only able to produce a response equal to 60-83% of the maximum response produced by 10 -5 M isoproterenol. Only 4'-methylbenzylTMQ (II) and TMQ gave a maximal response similar to isoproterenol (table 1).

3.2. Tracheal relaxant effects The concentration response curves of TMQ and the N-benzyl TMQ analogues are shown in fig. 2. The data show that all the N-benzyl substituted analogues are fl2-agonists which are less potent than the parent compound (TMQ). The rank order of fl2-adrenoceptor agonist potency is TMQ (I) > 4'-aminobenzylTMQ (VII) > 4'-methoxybenzylTMQ (IV) --- 3',4'-dichlorobenzylTMQ (VI) > 4'nitrobenzylTMQ (V) > 4'-chlorobenzylTMQ (III) > 4'-methylbenzylTMQ (II) (table 1). In contrast to results of the compounds in the atria, all the N-benzyl TMQ analogues are capable of eliciting

86-98% of the maximal isoproterenol response in guinea pig tracheal strips. Concentration response curves of all N-benzyl TMQ analogues were shifted to the right in the presence of 3 × 10 -8 M propranolol giving pK B values (mean + S.E., n = 4-6) of 8.92 + 0.09, 8.48 + 0.05, 8.30 + 0.11, 8.22 + 0.23, 8.60 _ 0.08 and 8.59_ 0.08 for analogues VII, IV, VI, V, III and II, respectively. Table 1 also summarizes the comparative agonist potencies and fl2/fla-selectivities of TMQ and the N-benzyl substituted analogues in guinea pig atria and trachea. Of these compounds TMQ, 4'-methylbenzyl TMQ (II) and 4'-nitrobenzyl TMQ (V) were equally potent as agonists on both fl-adrenoceptor systems. By contrast, the remaining N-benzyl TMQ analogues were more potent as agonists on guinea pig trachea. Using TMQ as a reference standard the fl2-selectivity indices for 4'-aminobenzylTMQ (VII), 4'-methoxybenzylTMQ (IV), 4'-chlorobenzylTMQ (III) and 3',4'dichlorobenzylTMQ (VI) were 2.5-, 7.8-, 8.3- and 9.6-fold, respectively (table 1).

TABLE 1 ill" and fl2-adrenoceptor activities of trimetoquinol (TMQ) and N-benzyl substituted T M Q analogues in guinea pig atria a and

trachea a. Drug (No.)

TMQ, I 4'-AminobenzylTMQ, VII 4'-MethoxybenzylTMQ, IV 3 ',4 '-DichlorobenzylTMQ, VI 4'-NitrobenzylTMQ, V 4 '-ChlorobenzylTMQ, III 4'-MethylbenzylTMQ, II

Guinea pig atria

Guinea pig trachea

n

p D 2 + S.E. b

I.A. c

Potency ratio d

4

6.76+0.15

0.97+0.02

1.0

4

5.54 + 0.12

0.83 + 0.07

4

4.19 + 0.10

4

n

Selec-

p D 2 4- S.E. b

I.A. c

Potency ratio d

tivity ratio

10

7.01+0.19

0.93+0.01

1.00

1.00

0.060

6

6.18 + 0.04

0.93 + 0.05

0.15

2.50

0.82 4- 0.08

0.0027

6

5.32 4- 0.13

0.98 4- 0.01

0.021

7.78

4.10 4- 0.30

0.69 4- 0.05

0.0022

6

5.32 +_0.06

0.95 + 0.03

0.021

9.55

4

4.71 4-0.13

0.604-0.05

0.0089

6

5.024-0.17

0.864-0.01

0.010

1.13

4

3.73 4- 0.11

0.64 4- 0.07

0.00093

6

4.89 4- 0.16

0.98 4- 0.01

0.0077

8.28

4

4.73 + 0.13

1.00 4- 0.0

0.0093

7

4.87 4- 0.06

0.95 4- 0.01

0.0073

0.78

a Values are the m e a n s + S . E , of n = 4 - 1 0 , b p D 2 = - - l o g ECs0; ECs0=effective concentration-50, c I.A. =intrinsic activity= m a x i m u m response of drug relative to the response of 10 - 5 M isoproterenol, a Potency r a t i o = E C s 0 ( T M Q ) / E C s 0 (drug). c Selectivity ratio = potency ratio (trachea)/potency ratio (atria).

255

4. Discussion

methoxy group on the N-benzyl ring increased the

fl2/fll-selectivity as compared to the parent drug, TMQ and N-benzyl substituted analogues were agonists and gave different rank order potencies in /31 (atria)- and/32 (trachea)-adrenoceptor systems. In analyzing the/3-adrenoceptor tissue data (table 1), no apparent relationship exists between illand/32-agonist activities and substituent effects on the electron density of the N-benzyl ring. In /3adrenoceptor systems, TMQ analogues possessing either electron withdrawing groups (mono- or dichloro substituents) or an electron-donating group (methoxy substituent) were the most /32-selective. Harrold et al. (1988) studied the inhibitory effects of these N-benzyl substituted TMQ analogues against U46619 (a TXA 2 agonist)-mediated contractions of rat thoracic aorta and also found no apparent correlation between antithromboxane antagonist activity and substituent-induced changes in the electron density of the N-benzyl ring. The rank order of inhibitory potency against U46619-mediated contractions of aorta was TMQ > II > IV > V > VII while the two chloro-substituted analogues III and VI were inactive. Moreover, the structural requirements for antagonism of TXA2-mediated responses in vascular smooth muscle and stimulation of/3-adrenoceptors among these TMQ analogues are different. Indeed, the chloro-substituted TMQ compounds are promising since they are considerably more fl2-selective than TMQ and devoid of TXA 2 antagonist activity in vascular tissue. In our earlier study (Harrold et al., 1988), electron-donating substituents (amino, methoxy or methyl groups) on the N-benzyl ring increased platelet antithromboxane activity whereas electron-withdrawing substituents (chloro or nitro groups) decreased this activity. Our results in /3adrenoceptor systems do not show this dependency of substituent-induced electronic effects on potency for these compounds. For example, 4'aminobenzyl TMQ and 4'-methylbenzyl TMQ were the most and least active as /32 agonists. Taken collectively, our results show that (1) the interaction of the N-benzyl ring of TMQ with /3-adrenoceptor sites is tolerant to the changes in electron density induced by substituents and (2) the presence of one or two chlorine atoms or of a

TMQ. Similar to our previous reports (Adejare et al., 1986; Fedyna et al., 1987) the present studies also provide additional evidence that structural requirements for biological activities of TMQ as antagonists on TXA 2 receptors versus agonists on fl-adrenoceptor systems are different. Salbutamol (albuterol), terbutaline, orciprenaline and fenoterol represent agonists of the catecholamine class which exhibit a greater selectivity for activation of fl2-adrenoceptors of bronchial smooth muscle versus fll-adrenoceptors of heart tissue, in vivo and in vitro (Daly and Levy, 1979; Lunts, 1985). In comparison to isoproterenol, albuterol is 3- to 7-fold and 80-fold whereas TMQ is 2- to 3-fold and 3-fold more selective for the stimulation of ]32- versus flladrenoceptors in vivo and in vitro, respectively (Lunts, 1985; Fedyna et al., 1987). Our results show that several N-benzyl substituted TMQ analogues exhibit a greater fl2-selectivity and a lower potency than TMQ. The reduced potency of these TMQ analogues is expected since tertiary amines of catecholamines and TMQ are less active as fl-agonists (Pratesi and Grana, 1965; Iwasawa and Kiyomoto, 1967). It is notable; however, that these tertiary amine analogues of TMQ exhibit a much improved fl2-selectivity of 8- to 10-fold over that of the parent drug, and the fl2/fll-selectivity ratios of N-benzyl substituted TMQ analogues are similar to those reported for other fl2-selected agonists such as albuterol (Lunts, 1985).

References Adejare, A., D.D. Miller, J.S. Fedyna, C.H. Ahn and D.R. Feller, 1986, Synthesis and fl-adrenergic agonist and antiaggregatory properties of N-substituted trimetoquinol analogues, J. Med. Chem. 29, 1603. Alan, C.H., K.J. Romstedt, L.J. Wallace, D.D. Miller and D.R. Feller, 1988, Characterization of the inhibition of U46619mediated human platelet activation by the trimetoquinol isomers, Biochem. Pharmacol. 37, 3023. Clark, M.T., A. Adejare, G. Shams, D.R. Feller and D.D. Miller, 1988, 5-Fluoro- and 8-fluorotrimetoquinol: selective fl2-adrenoceptor agonists, J. Meal. Chem. 30, 86. Daly, M.J. and G.P. Levy, 1979, The subclassification of fl-adrenoceptors: evidence in support of the dual fl-adren-

256 oceptor hypothesis, in: Trends in Autonomic Pharmacology, ed. S. Kalsner (Urban and Schwarzenberg, Inc., Baltimore-Munich) Vol. 1, p. 347. Farmer, J.B., I. Kennedy, G.P. Levy and R.J. Marshall, 1970, A comparison of the beta-adrenoceptor stimulant properties of isoprenaline with those of orciprenaline, salbutamol, soterenol and trimetoquinol on isolated atria and trachea of guinea pig, J. Pharm. Pharmacol. 22, 61. Fedyna, J., A. Adejare, D.D. Miller and D.R. Feller, 1987, Pharmacological evaluation of the fl-adrenoceptor agonist and thromboxane antagonist properties of N-substituted trimetoquinol analogs, European J. Pharmacol. 135, 161. Harrold, M.W., B. Grazyl, Y. Shin, K.J. Romstedt, D.R. Feller and D.D. Miller, 1988, Synthesis and thromboxane A 2 antagonist activity of N-benzyl-trimetoquinol analogues, J. Med. Chem. 31, 1506. Iwasawa, Y. and A. Kiyomoto, 1967, Studies on tetrahydroisoquinolines (THI) 1-bronchodilator activity and structure activity relationship, Jap. J. Pharmacol. 17, 143. Kattelman, E.J., D.L. Venton and G.C. LeBreton, 1986, Characterization of U46619 binding in inactivated, intact human platelets and determination of binding site affinities of four TXA2/PGH2 receptor antagonists (13-APA, BM 13.177, ONO 3708 and SQ 29.548), Thromb. Res. 41, 471. Lefer, A.M. and H. Darius, 1987, A pharmacological approach to thromboxane receptor antagonism, Fed. Proc. 46, 144. Lunts, L.H.C., 1985, Salbutamol, a selective /~2-stimulant bronchodilator, in: Medicinal Chemistry, The Role of Organic Chemistry in Drug Research, eds. S.M. Roberts and B.J. Price (Academic Press, Ltd. London) Chap. 4, p. 49.

Miller, D.D., P. Osei-Gyimah, J. Bardin and D.R. Feller, 1975, Synthesis and biological evaluation of fragmented derivatives of tetrahydroisoquinolines-2-trimetoquinol studies, J. Med. Chem. 18, 454. Mukopadhyay, A., S.S. Navran, H. Amin, S.A. Abdel-Aziz, J. Chang, D.J. Sober, D.D. Miller and D.R. Feller, 1985, Effect of trimetoquinol analogs for antagonism of endoperoxide/thromboxane A2-mediated responses in human platelets and rat aorta, J. Pharmacol. Exp. Ther. 232, 1. Mukopadhyay, A., D.J. Sober, J. Chang R.T. Slenn, H. Amin, D.D. Miller and D.R. Feller, 1982, Broncho-sdective action of a new series of trimetoquinol analogues, European J. Pharmacol. 77, 209. Pratesi, P. and E. Grana, 1965, Structure and activity relationships in adrenergic receptors of catecholamines and certain related compounds, in: Advances in Drug Research, eds. N.J. Harper and A.B. Simmonds (Academic Press, New York) Vol. 2, p. 127. Staff of the Department of Pharmacology, 1968, University of Edinburgh, in: Pharmacological Experiments on Isolated Preparations (Livingstone, London) p. 104. Van Rossum, J.M., 1963, 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. 143, 299. Yamamura, Y. and S. Kishmoto, 1968, Clinical effectiveness of a new bronchodilator, Inolin, on bronchial asthma, Ann. Allergy 26, 504.