Previous observations implicating PgH2 as a direct activator of platelets suggested that derivatives of U46619, a well-characterized TxA2 receptor agonist having structural homology with PgH2, might possess antiplatelet activity. The present work describes the synthesis of [1S-(1 alpha,2 beta,3 alpha,4 alpha)]-3-[(tetrahydropyranyloxy)methyl]- 2-[2-[(triphenylmethyl)oxy]ethyl]-5-oxabicyclo[2.2.1]heptane (14) a potentially useful intermediate for the synthesis of various epoxymethano derivatives. The latter was converted to [1S-(1 alpha,2 beta (Z),3 alpha,4 alpha)]-7-[3-[[2- [(phenylamino)carbonyl]-hydrazino]methyl]-5-oxabicylo[2.2.1]hept-2 - yl]-5-heptenoic acid (23), an epoxymethano derivative of PgH2 containing a hydrazide lower side chain as previously used in the TxA2 antagonist, SQ 29,548. The intermediate 14 was also converted to [1S-(1 alpha,2 beta (Z),3 alpha,4 alpha)]-7- [3-[(hexylamino)methyl]-5-oxabicyclo[2.2.1]hept-2-yl]-5-heptenoic acid (25) which contained a simple aza side chain as used in earlier antagonists. Derivatives 23 and 25 appeared to be specific antagonists of the human platelet TxA2 receptor as evidenced by their inhibition of U46619 (1.5 microM) induced aggregation of human platelet rich plasma (IC50 = 22 and 7 microM, respectively), while having little effect on ADP (2 microM) induced aggregation at much higher concentrations. In addition, one of these derivatives, the bicycloamine 25, was shown to compete for [3H]U46619 binding to washed human platelets with an IC50 value of 25 microM, supporting the notion that these derivatives were acting at the thromboxane receptor. However, the potency of these derivatives was less than for previously reported TxA2 antagonists, suggesting that simple linear combinations of functionality from molecules active at the human platelet thromboxane receptor will be of limited predictive value.