Binding orientation of amphetamine and norfenfluramine analogues in the benzonorbornene and benzobicyclo[3.2.1]octane ring systems at the active site of phenylethanolamine N-methyltransferase (PNMT)

In a continuation of studies directed at characterizing the conformational basis of binding beta-phenylethylamines at the active site of phenylethanolamine N-methyltransferase (PNMT), anti-10-amino- (12) and syn-10-amino-5,6,7,8-tetrahydro-5,8-methano-9H-benzocycloheptene (13) were prepared and evaluated as substrates and inhibitors for PNMT. These conformationally defined amphetamine analogues mimic a low energy half-chair form of 2-aminotetralin (2AT). Further, in order to determine the active site binding orientation of beta-phenylethylamines bearing aryl lipophilic substituents, the aryl trifluoromethyl-substituted derivatives of 12 and 13 (20-27), as well as anti-9-amino-5-(trifluoromethyl)-(18) and anti-9-amino-6-(trifluoromethyl) benzonorbornene (19), were prepared and evaluated. The competitive inhibition displayed by the fully extended analogue 12 coupled with the uncompetitive kinetics exhibited by the folded isomer 13 supports previous findings that a fully extended side chain conformation is optimal for binding to the active site of PNMT. In addition, the fact that 12 displayed enhanced affinity as an inhibitor over its beta-phenylethylamine counterparts in the benzonorbornene and 1,4-ethanonaphthalene ring systems suggests that a half-chair conformation is preferred when 2AT analogues interact at the active site of the enzyme. This would be consistent with previous results that PNMT preferentially binds molecules with a more coplanar relationship between the aromatic ring and the amino nitrogen. The lack of activity as a substrate in 12 indicates that the negative steric interactions of the ethano bridging unit prohibits it from binding in a manner consistent with the known PNMT substrates exo-2-amino- (6) and anti-9-aminobenzonorbornene (8). Given the emergence of activity as a substrate in 20 and 21 (the 1-trifluoromethyl- and the 2-trifluoromethyl-substituted derivatives of 12), it appears that the positive interaction of the trifluoromethyl group orients these analogues in a manner in which the ethano bridge lies in regions of steric bulk tolerance. This would suggest that the region of steric intolerance has a degree of directionality. Finally, although the aromatic ring binding region of the active site of PNMT contains a large degree of lipophilic character, only specific spatial orientations between the trifluoromethyl group and the amino nitrogen of aryl trifluoromethyl-substituted beta-phenylethylamines allow both to interact simultaneously in a manner that allows the amine to bind in a region of the active site in which methylation can occur.

Synthesis of halogenated trimetoquinol derivatives and evaluation of their beta-agonist and thromboxane A2 (TXA2) antagonist activities.

The 5,8-difluoro (4), 5-iodo (5), 8-iodo (6), and 5-trifluoromethyl (7) derivatives of trimetoquinol (TMQ, 1) have been synthesized and evaluated for their ability to stimulate beta 1 (guinea pig atria) and beta 2 (guinea pig trachea) adrenoceptors as well as for their inhibitory activity against U46619 [a thromboxane A2 (TXA2) mimetic]-mediated contraction of rat thoracic aorta and human platelet aggregation. Both 5 and 6 were considerably less active than TMQ on both beta-adrenergic systems and gave a rank order of stimulatory potency of 1 much greater than 6 greater than or equal to 5. Similarly, iodine substitution at either position also caused a reduction in TXA2 antagonist activity with a rank order potency of 1 greater than 6 much greater than 5. Compared to 1, however, 5-iodo-TMQ (5) showed a marked selectivity for blockade of U46619 responses in rat aorta over human platelets. On beta-systems, 4 had reduced potency compared to TMQ and was similarly nonselective. Introduction of a trifluoromethyl group at the 5-position of TMQ completely abolished both beta 1- and beta 2-adrenergic agonist activities while imparting weak antagonist activity on beta 1 receptors. On TXA2 systems, both 4 and 7 possessed significantly decreased inhibitory activity compared to TMQ. The synthetic approaches to the synthesis of 8-(trifluoromethyl)-TMQ (8) are also described. The enantiomers of the 8-fluoro derivative (3) of TMQ were separated on a preparative Chiralcel OD column and evaluated on beta-adrenergic systems and TXA2 systems. On beta-adrenergic systems, (S)-(+)-8-fluoro-TMQ was at least 10-fold more potent than (R)-(-)-8-fluoro-TMQ. Conversely, (R)-(-)-8-fluoro-TMQ was approximately 14-fold more potent as an antagonist of TXA2-mediated aggregation in human platelets than (S)-(+)-8-fluoro-TMQ. In contrast to platelets, (S)-(+)-8-fluoro-TMQ was an agonist in rat aorta whereas (R)-(-)-8-fluoro-TMQ was an antagonist.