- Email: [email protected]
Further studies on the effects of epimers of thromboxane A2 antagonists on platelets and veins
European Journal of Pharmacology, 111 (1985) 125-128
125
Elsevier
Short communication
F U R T H E R S T U D I E S O N T H E EFFECTS O F E P I M E R S O F T H R O M B O X A N E A 2 A N T A G O N I S T S ON P L A T E L E T S AND VEINS D A L E E. MAIS *.t, COEL D U N L A P *, N. H A M A N A K A t t and P E R R Y V. H A L U S H K A *'**,~:
Departments of * Pharmacology and **Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, U.S.A., t The Drug Science Foundation, Charleston, South Carolina 29425, U.S.A., and tt Ono Pharmaceutical Company, Osaka, Japan Received 29 January 1985, accepted 26 February 1985
D.E. MAIS, C. DUNLAP, N. HAMANAKA and P.V. HALUSHKA, Further studies on the effects of epimers of thromboxane A 2 antagonists on platelets and veins, European J. Pharmacol. 111 (1985) 125-128. The C-15 hydroxy epimers of three T X A J P G H 2 antagonists were resolved by HPLC and their pharmacology evaluated in canine saphenous veins and human platelets. For all three compounds, the epimers were equipotent in their ability to antagonize U46619 (1/~M) induced platelet aggregation. In contrast, one of the epimers for all three of the antagonists was significantly more potent than the other in antagonizing U46619 (10 nM) induced saphenous vein contractions. The data support the notion that the TXA2/PGH 2 receptors in veins may be different from those in platelets. Thromboxane A 2/prostaglandin Thromboxane A 2 antagonists
H2
receptors
Saphenous veins
1. Introduction
Platelets
13-Azapinane
of 13-azapinane analogs (Mais et al., 1984) and evaluated their pharmacology in canine and human platelets and saphenous veins (Mais et al., 1985). Based on differences in the rank order potencies of these agents to inhibit the effects of U46619 on platelets and veins, we hypothesized that the T X A 2 / P G H 2 receptor in platelets was different from that in the saphenous vein. The synthesis of these analogs resulted in C-15 epimeric pairs. We have now resolved these pairs for three of the compounds used in the previous study, ONO-11120, PTA-OH and I-PTA-OH. Evaluation of the pharmacology of these epimers in canine saphenous vein and human platelets was carried out to determine the importance of the C-15 hydroxy orientation and whether this orientation is recognized by one or both of these tissues.
In spite of the fact that thromboxane A 2 (TXA2) has not been isolated or synthesized directly, various stable analogs have been synthesized and used to study putative TXA 2 or prostaglandin H 2 (PGH2) receptors. These have included agonists (Bundy, 1975) and antagonists (Nicolaou et al., 1979; Katsura et al., 1983) of platelet aggregation and vascular smooth muscle contraction. Two such classes of antagonists are the 13azaprostanoic acid (LeBreton et al., 1979) and the pinane series (Nicolaou et al., 1979). Katsura et al. (1983) combined these two classes to produce the 13-azapinane derivative ONO-11120 which was an antagonist of the actions of the stable TXA 2 mimetic U46619 (Bundy, 1975) to induce platelet aggregation and to cause vascular smooth muscle contraction. Recently, we described the synthesis of a series
2. Materials and methods
* To whom all correspondence should be addressed.
Fig. 1 shows the structure of the compounds used in these studies. The synthesis and characteri-
0014-2999/85/$03.30 © 1985 Elsevier Science Publishers B.V.
126
zation of PTA-OH (9,11-dimethylmethano-ll,12methano-16-(4-hydroxyphenyl)-13,14-dihydro-13aza-15afl-~0-tetranor-TXA2) and I-PTA-OH (9,
~,,~
"" ~ - - - ~ c o o .
ll-dimethylmethano-ll,12-methano-16-(3-iodo4-hydroxyphenyl-13,14-dihydro-13-aza-15a/3-c0-tetranor-TXA 2 were carried out as described by Mais et al. (1984). ONO-11120 (9,11-dimethylmethano-ll,12-methano-16-phenyl-13,14-dihydro13-aza-15a/3-~0-tetranor-TXA 2) was a gift of ONO Pharmaceutical Company (Osaka, Japan). U46619 [(15 S)-hydroxy- 1 la, 9a-(epoxymethano)prosta5Z,13E-dienoic acid] was a gift from the Upjohn Company (Kalamazoo, MI). All other reagents were of the highest purity available from Sigma Chemical Company (St. Louis, MO). Epimerswere resolved by HPLC using a Whatman Partisil PXS 10/25 PAC semi-preparative column and a flow rate of 3 ml/min. The mobile phase consisted of 90:10 chloroform/methanol. Detection of the solutes was monitored at 254 nm. Probe mass spectrometry of the methyl esters of the epimers or fast atom bombardment mass spectrometry of the free acids gave spectral data consistent with the resolved epimeric structures. The epimers were pure by TLC and analytical HPLC. Stock solutions of agonist and antagonists were prepared in absolute ethanol stored at - 20°C and diluted in the appropriate buffer before each experiment. Preparation of human platelet rich plasma (PRP) and canine saphenous veins and the conditions utilized for these studies were as described previously (Mais et al., 1985). All values are expressed as means + S.E.M. The Student's t-test was used to determine if a significant difference existed between ICs0 values for the epimers.
(sl
Fig. 1. Chemical structures of ONO-11120 (R = H), PTA-OH (R = 4-OH) and I-PTA-OH (R = 3-iodo-4-OH) and the two possible configurations for the hydroxyl group which can exist about the C-15 carbon. While the (S) and (R) configurations as drawn are correct in the figure, it is not possible based solely on elution characteristics in H P L C or TLC systems to assign an absolute configuration for the three antagonists.
the C-15 position of the separated epimers, therefore, they were assigned (1) for the fastest eluting peak and (2) for the slowest. The ability of these epimers to antagonize U46619 induced canine saphenous vein contraction and human platelet aggregation is shown in
TABLE 1 ICso values for epimers in canine saphenous vein and h u m a n PRP.
PTA-OH
3. Results I-PTA-OH
Two configurations are possible at the C-15 center of prostaglandins and their synthetic analogs; the (S) configuration, found in all natural prostanoids, and the unnatural (R) configuration (fig. 1). The HPLC system used in this study gave good separation of the two components. However, based on elution characteristics alone it is not possible to assign an absolute configuration about
(R)
ONO-11120
Canine saphenous vein ( ~ M ) a
H u m a n platelets (/~M) a
(1) b 0.73+0.06 c (6) a (2) 2.00_+0.17 (9) (1) 0.04+0.01 (6) a (2) 0.11 +0.01 (6) (1) 0.09=t:0.03 (7) d (2) 0.03 + 0.0003 (6)
0.85+0.15 0.76+0.07 0.63+0.13 0.52_+0.09 0.55+0.12 0.46 + 0.09
(5) (5) (4) (4) (4) (4)
a Saphenous veins were contracted with 10 nM U46619. Platelet aggregation was induced with 1 # M U46619 (see Methods). b (1) and (2) refer to order of elution in HPLC system used to achieve resolution (see Methods). c Data are expressed as ~+S.E.M. (n). a Signifies a significant difference (P <
0.05).
127 table 1. For all three compounds in saphenous veins, one of the epimers was nearly three times more potent that the other epimer (P < 0.05) in antagonizing U46619 induced contractions. In contrast, the epimers exhibited equal potency in their ability to antagonize U46619 induced platelet aggregation (table 1).
4. Discussion
Previous work using agonists and antagonists of TXA 2 and P G H 2 has suggested that the T X A 2 / P G H 2 receptor in platelets may be different from that in blood vessels (Gorman et al., 1981; Lefer et al., 1980). More recently, we have demonstrated a significant difference in the rank order potencies of a series of 13-azapinane analogs as T X A 2 / P G H 2 receptor antagonists in platelets when compared to saphenous veins in both dog and man (Mais et al., 1985). The results of the current study are also consistent with the hypothesis that platelet and vascular receptors for T X A 2 / P G H 2 are different. For this series of compounds, the orientation of the C-15 hydroxyl group significantly influenced the potency of the epimers in the saphenous vein whereas that was not the case in platelets. This is demonstrated by the three fold difference observed between the potencies of the epimers in the saphenous vein and no difference in the potencies of each of the epimers in platelets. We have previously shown that the rank order potencies of these antagonists are the same in both canine and human saphenous veins (Mais et al., 1985), thus this difference in the potencies of the epimers between platelets and veins cannot be attributed to species differences between human and dog. The first epimeric peak eluted from the HPLC column for both PTA-OH and I-PTA-OH gave the more potent activity in the saphenous vein whereas it was the second peak for ONO-11120 which was most potent in this tissue. For many synthetic prostaglandins, the order of elution in a straightphase partition system has been utilized for assigning the (R) or (S) configuration (Wilson et al., 1982). Utilizing a TLC system which previous workers (Nicolaou et al., 1979) have used to assign
the configurations at C-15, we also found a reversal of migration characteristics of the epimers (data not shown). We cannot exclude the possibility that the configuration at C-15 yielding the highest activity in vessels was reversed for PTA-OH and I-PTA-OH compared to ONO-11120. This would seem unlikely and perhaps the change in Rfs is due to the presence of the phenolic hydroxyl in I-PTA-OH and PTA-OH. Therefore, rather than trying to assign (R) or (S) configurations at the present time, we have used (1) and (2) in table 1 to represent the order of elution from the HPLC column. No matter which configuration exists at C-15 clearly further studies are needed to assign absolute configurations. Further experiments are also indicated to assess the potential selective effects of epimers of other antagonists on platelets and blood vessels. The data presented are consistent with the hypothesis that TXA 2 / P G H 2 receptors are different in platelets compared to saphenous veins. The exact nature of and the differences between these receptors, remains unclear. However, the differences observed here support the nomenclature that we previously introduced for these receptors. The platelet receptor has been designated (TXA 2 / P G H 2 ) , , a for aggregation and the vascular receptor ( T X A 2 / P G H 2 ) ~ , ~ for tone (Mais et al., 1985).
Acknowledgements
This work was supported in part by N.I.H. HL 29566 and HL07260. P.V. Halushka is a Burroughs Weilcome Scholar in Clinical Pharmacology. References
Bundy, G., 1975, The synthesis of prostagiandin endoperoxide analogs, Tetrahedron Lett. 24, 1957. Gorman, R., R. Shubuski, J. Aiken and G. Bundy, 1981, Analysis of the biological activity of azoprostanoids in human platelets, Fed. Proc. 40, 1997. Katsura, M., T. Miyamato, N. Hamanaka, K. Kondo, T. Terada, Y. Ohgaki, A. Kawasaki and M. Tsuboshima, 1983, In vitro and in vivo effects of new powerful thromboxane A 2 antagonists (3-alkyl-aminopinane derivatives),in: Advances in Prostaglandin, Thromboxaneand LeukotrieneResearch, Vol. 11,351.
128 LeBreton, G.C., D.L. Venton, S.E. Enke and P.V. Halushka, 1979, 13-Aza-prostanoic Acid: A specific antagonist of the human blood, platelet thromboxane/endoperoxide receptor, Proc. Natl. Acad. Sci. U.S.A. 76, 4097. Lefer, A., E. Smith, H. Araki, J. Smith, D. Aharony, D. Clareman, R. Magolda and K. Nicolaou, 1980, Dissociation of vasoconstrictor and platelet aggregatory activities of thromboxane by carbacyclic thromboxane A 2, Proc. Natl. Acad. Sci. U.S.A. 77, 1706. Mais, D., D. Knapp, K. Ballard, P. Halushka and N. Hamanaka, 1984, Synthesis of thromboxane receptor antagonists with the potential to radiolabel with 1251, Tetrahedron Lett. 25, 4207. Mais, D., D. Saussy, A. Chaikhouni, P. Kochel, D. Knapp, N. Hamanaka and P. Halushka, 1985, Pharmacologic char-
acterization of human and canine thromboxane A 2 / prostaglandin H 2 receptors in platelets and blood vessels. Evidence for different receptors, J. Pharmacol. Exp. Ther. (in press). Nicolaou, K., R. Magolda, J. Smith, D. Aharony, E. Smith and A. Lefer, 1979, Synthesis and biological properties of pinane-thromboxane A2, a selective inhibitor of coronary artery constriction, platelet aggregation and thromboxane formation, Proc. Natl. Acad. Sci. U.S.A., 76, 2566. Wilson, N., V. Peesapati, R. Jones and K. Hamilton, 1982, Synthesis of prostanoids with bicyclo [2.2.1] heptane, bicyclo [3.1.1] heptane, and bicyclo [2.2.2] octane ring systems. Activities of 15-hydroxy epimers on human platelets, J. Med. Chem. 25, 495.
125
Elsevier
Short communication
F U R T H E R S T U D I E S O N T H E EFFECTS O F E P I M E R S O F T H R O M B O X A N E A 2 A N T A G O N I S T S ON P L A T E L E T S AND VEINS D A L E E. MAIS *.t, COEL D U N L A P *, N. H A M A N A K A t t and P E R R Y V. H A L U S H K A *'**,~:
Departments of * Pharmacology and **Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, U.S.A., t The Drug Science Foundation, Charleston, South Carolina 29425, U.S.A., and tt Ono Pharmaceutical Company, Osaka, Japan Received 29 January 1985, accepted 26 February 1985
D.E. MAIS, C. DUNLAP, N. HAMANAKA and P.V. HALUSHKA, Further studies on the effects of epimers of thromboxane A 2 antagonists on platelets and veins, European J. Pharmacol. 111 (1985) 125-128. The C-15 hydroxy epimers of three T X A J P G H 2 antagonists were resolved by HPLC and their pharmacology evaluated in canine saphenous veins and human platelets. For all three compounds, the epimers were equipotent in their ability to antagonize U46619 (1/~M) induced platelet aggregation. In contrast, one of the epimers for all three of the antagonists was significantly more potent than the other in antagonizing U46619 (10 nM) induced saphenous vein contractions. The data support the notion that the TXA2/PGH 2 receptors in veins may be different from those in platelets. Thromboxane A 2/prostaglandin Thromboxane A 2 antagonists
H2
receptors
Saphenous veins
1. Introduction
Platelets
13-Azapinane
of 13-azapinane analogs (Mais et al., 1984) and evaluated their pharmacology in canine and human platelets and saphenous veins (Mais et al., 1985). Based on differences in the rank order potencies of these agents to inhibit the effects of U46619 on platelets and veins, we hypothesized that the T X A 2 / P G H 2 receptor in platelets was different from that in the saphenous vein. The synthesis of these analogs resulted in C-15 epimeric pairs. We have now resolved these pairs for three of the compounds used in the previous study, ONO-11120, PTA-OH and I-PTA-OH. Evaluation of the pharmacology of these epimers in canine saphenous vein and human platelets was carried out to determine the importance of the C-15 hydroxy orientation and whether this orientation is recognized by one or both of these tissues.
In spite of the fact that thromboxane A 2 (TXA2) has not been isolated or synthesized directly, various stable analogs have been synthesized and used to study putative TXA 2 or prostaglandin H 2 (PGH2) receptors. These have included agonists (Bundy, 1975) and antagonists (Nicolaou et al., 1979; Katsura et al., 1983) of platelet aggregation and vascular smooth muscle contraction. Two such classes of antagonists are the 13azaprostanoic acid (LeBreton et al., 1979) and the pinane series (Nicolaou et al., 1979). Katsura et al. (1983) combined these two classes to produce the 13-azapinane derivative ONO-11120 which was an antagonist of the actions of the stable TXA 2 mimetic U46619 (Bundy, 1975) to induce platelet aggregation and to cause vascular smooth muscle contraction. Recently, we described the synthesis of a series
2. Materials and methods
* To whom all correspondence should be addressed.
Fig. 1 shows the structure of the compounds used in these studies. The synthesis and characteri-
0014-2999/85/$03.30 © 1985 Elsevier Science Publishers B.V.
126
zation of PTA-OH (9,11-dimethylmethano-ll,12methano-16-(4-hydroxyphenyl)-13,14-dihydro-13aza-15afl-~0-tetranor-TXA2) and I-PTA-OH (9,
~,,~
"" ~ - - - ~ c o o .
ll-dimethylmethano-ll,12-methano-16-(3-iodo4-hydroxyphenyl-13,14-dihydro-13-aza-15a/3-c0-tetranor-TXA 2 were carried out as described by Mais et al. (1984). ONO-11120 (9,11-dimethylmethano-ll,12-methano-16-phenyl-13,14-dihydro13-aza-15a/3-~0-tetranor-TXA 2) was a gift of ONO Pharmaceutical Company (Osaka, Japan). U46619 [(15 S)-hydroxy- 1 la, 9a-(epoxymethano)prosta5Z,13E-dienoic acid] was a gift from the Upjohn Company (Kalamazoo, MI). All other reagents were of the highest purity available from Sigma Chemical Company (St. Louis, MO). Epimerswere resolved by HPLC using a Whatman Partisil PXS 10/25 PAC semi-preparative column and a flow rate of 3 ml/min. The mobile phase consisted of 90:10 chloroform/methanol. Detection of the solutes was monitored at 254 nm. Probe mass spectrometry of the methyl esters of the epimers or fast atom bombardment mass spectrometry of the free acids gave spectral data consistent with the resolved epimeric structures. The epimers were pure by TLC and analytical HPLC. Stock solutions of agonist and antagonists were prepared in absolute ethanol stored at - 20°C and diluted in the appropriate buffer before each experiment. Preparation of human platelet rich plasma (PRP) and canine saphenous veins and the conditions utilized for these studies were as described previously (Mais et al., 1985). All values are expressed as means + S.E.M. The Student's t-test was used to determine if a significant difference existed between ICs0 values for the epimers.
(sl
Fig. 1. Chemical structures of ONO-11120 (R = H), PTA-OH (R = 4-OH) and I-PTA-OH (R = 3-iodo-4-OH) and the two possible configurations for the hydroxyl group which can exist about the C-15 carbon. While the (S) and (R) configurations as drawn are correct in the figure, it is not possible based solely on elution characteristics in H P L C or TLC systems to assign an absolute configuration for the three antagonists.
the C-15 position of the separated epimers, therefore, they were assigned (1) for the fastest eluting peak and (2) for the slowest. The ability of these epimers to antagonize U46619 induced canine saphenous vein contraction and human platelet aggregation is shown in
TABLE 1 ICso values for epimers in canine saphenous vein and h u m a n PRP.
PTA-OH
3. Results I-PTA-OH
Two configurations are possible at the C-15 center of prostaglandins and their synthetic analogs; the (S) configuration, found in all natural prostanoids, and the unnatural (R) configuration (fig. 1). The HPLC system used in this study gave good separation of the two components. However, based on elution characteristics alone it is not possible to assign an absolute configuration about
(R)
ONO-11120
Canine saphenous vein ( ~ M ) a
H u m a n platelets (/~M) a
(1) b 0.73+0.06 c (6) a (2) 2.00_+0.17 (9) (1) 0.04+0.01 (6) a (2) 0.11 +0.01 (6) (1) 0.09=t:0.03 (7) d (2) 0.03 + 0.0003 (6)
0.85+0.15 0.76+0.07 0.63+0.13 0.52_+0.09 0.55+0.12 0.46 + 0.09
(5) (5) (4) (4) (4) (4)
a Saphenous veins were contracted with 10 nM U46619. Platelet aggregation was induced with 1 # M U46619 (see Methods). b (1) and (2) refer to order of elution in HPLC system used to achieve resolution (see Methods). c Data are expressed as ~+S.E.M. (n). a Signifies a significant difference (P <
0.05).
127 table 1. For all three compounds in saphenous veins, one of the epimers was nearly three times more potent that the other epimer (P < 0.05) in antagonizing U46619 induced contractions. In contrast, the epimers exhibited equal potency in their ability to antagonize U46619 induced platelet aggregation (table 1).
4. Discussion
Previous work using agonists and antagonists of TXA 2 and P G H 2 has suggested that the T X A 2 / P G H 2 receptor in platelets may be different from that in blood vessels (Gorman et al., 1981; Lefer et al., 1980). More recently, we have demonstrated a significant difference in the rank order potencies of a series of 13-azapinane analogs as T X A 2 / P G H 2 receptor antagonists in platelets when compared to saphenous veins in both dog and man (Mais et al., 1985). The results of the current study are also consistent with the hypothesis that platelet and vascular receptors for T X A 2 / P G H 2 are different. For this series of compounds, the orientation of the C-15 hydroxyl group significantly influenced the potency of the epimers in the saphenous vein whereas that was not the case in platelets. This is demonstrated by the three fold difference observed between the potencies of the epimers in the saphenous vein and no difference in the potencies of each of the epimers in platelets. We have previously shown that the rank order potencies of these antagonists are the same in both canine and human saphenous veins (Mais et al., 1985), thus this difference in the potencies of the epimers between platelets and veins cannot be attributed to species differences between human and dog. The first epimeric peak eluted from the HPLC column for both PTA-OH and I-PTA-OH gave the more potent activity in the saphenous vein whereas it was the second peak for ONO-11120 which was most potent in this tissue. For many synthetic prostaglandins, the order of elution in a straightphase partition system has been utilized for assigning the (R) or (S) configuration (Wilson et al., 1982). Utilizing a TLC system which previous workers (Nicolaou et al., 1979) have used to assign
the configurations at C-15, we also found a reversal of migration characteristics of the epimers (data not shown). We cannot exclude the possibility that the configuration at C-15 yielding the highest activity in vessels was reversed for PTA-OH and I-PTA-OH compared to ONO-11120. This would seem unlikely and perhaps the change in Rfs is due to the presence of the phenolic hydroxyl in I-PTA-OH and PTA-OH. Therefore, rather than trying to assign (R) or (S) configurations at the present time, we have used (1) and (2) in table 1 to represent the order of elution from the HPLC column. No matter which configuration exists at C-15 clearly further studies are needed to assign absolute configurations. Further experiments are also indicated to assess the potential selective effects of epimers of other antagonists on platelets and blood vessels. The data presented are consistent with the hypothesis that TXA 2 / P G H 2 receptors are different in platelets compared to saphenous veins. The exact nature of and the differences between these receptors, remains unclear. However, the differences observed here support the nomenclature that we previously introduced for these receptors. The platelet receptor has been designated (TXA 2 / P G H 2 ) , , a for aggregation and the vascular receptor ( T X A 2 / P G H 2 ) ~ , ~ for tone (Mais et al., 1985).
Acknowledgements
This work was supported in part by N.I.H. HL 29566 and HL07260. P.V. Halushka is a Burroughs Weilcome Scholar in Clinical Pharmacology. References
Bundy, G., 1975, The synthesis of prostagiandin endoperoxide analogs, Tetrahedron Lett. 24, 1957. Gorman, R., R. Shubuski, J. Aiken and G. Bundy, 1981, Analysis of the biological activity of azoprostanoids in human platelets, Fed. Proc. 40, 1997. Katsura, M., T. Miyamato, N. Hamanaka, K. Kondo, T. Terada, Y. Ohgaki, A. Kawasaki and M. Tsuboshima, 1983, In vitro and in vivo effects of new powerful thromboxane A 2 antagonists (3-alkyl-aminopinane derivatives),in: Advances in Prostaglandin, Thromboxaneand LeukotrieneResearch, Vol. 11,351.
128 LeBreton, G.C., D.L. Venton, S.E. Enke and P.V. Halushka, 1979, 13-Aza-prostanoic Acid: A specific antagonist of the human blood, platelet thromboxane/endoperoxide receptor, Proc. Natl. Acad. Sci. U.S.A. 76, 4097. Lefer, A., E. Smith, H. Araki, J. Smith, D. Aharony, D. Clareman, R. Magolda and K. Nicolaou, 1980, Dissociation of vasoconstrictor and platelet aggregatory activities of thromboxane by carbacyclic thromboxane A 2, Proc. Natl. Acad. Sci. U.S.A. 77, 1706. Mais, D., D. Knapp, K. Ballard, P. Halushka and N. Hamanaka, 1984, Synthesis of thromboxane receptor antagonists with the potential to radiolabel with 1251, Tetrahedron Lett. 25, 4207. Mais, D., D. Saussy, A. Chaikhouni, P. Kochel, D. Knapp, N. Hamanaka and P. Halushka, 1985, Pharmacologic char-
acterization of human and canine thromboxane A 2 / prostaglandin H 2 receptors in platelets and blood vessels. Evidence for different receptors, J. Pharmacol. Exp. Ther. (in press). Nicolaou, K., R. Magolda, J. Smith, D. Aharony, E. Smith and A. Lefer, 1979, Synthesis and biological properties of pinane-thromboxane A2, a selective inhibitor of coronary artery constriction, platelet aggregation and thromboxane formation, Proc. Natl. Acad. Sci. U.S.A., 76, 2566. Wilson, N., V. Peesapati, R. Jones and K. Hamilton, 1982, Synthesis of prostanoids with bicyclo [2.2.1] heptane, bicyclo [3.1.1] heptane, and bicyclo [2.2.2] octane ring systems. Activities of 15-hydroxy epimers on human platelets, J. Med. Chem. 25, 495.