Search

A synthesis of dimethyl gloiosiphone A (2) has been realized. The key step involves the boron trifluoride-catalyzed eliminative ring expansion of a 2′-(dimethoxymethyl)-1′-(cyclopenten-1′-yl)-4-hydroxy-2-cyclobuten-1-one, which was directly assembled by condensation of the appropriate cycloalkenyllithium with dimethyl squarate. Following arrival at spirocyclic diketone 17 in only two steps, the cyclopentenedione A-ring was subjected to controlled reductive removal of one carbonyl group. Subsequent oxidation of the enol ether to unsaturated aldehyde 21 set the stage for proper elaboration of the α-hydroxy ketone part structure in ring B. The spectral properties of the synthetic material were identical to those of natural 2, which was originally isolated as a racemate.

The bicyclic ketone (+)-17b reacts with vinylmagnesium bromide from its exo face, thus enabling exploitation of the anionic oxy-Cope rearrangement and direct alkylation of the enolate so formed with ethyl iodoacetate to provide 24 in an efficient, stereocontrolled manner. Tricyclic lactone 26b, readily produced from 24, proved to be a common precursor to both synthetic targets. The construction of vulgarolide (9) proceeded by transforming the SEM group in 26b into a mesylate for the purpose of conversion to epoxy acid 39 in advance of oxidative cleavage of the double bond. This transformation resulted in direct conversion to 9 and its epimer 42. Central to the companion synthesis of deoxocrispolide (4) was the ability of 37 to undergo the Michael addition of methoxide ion and of 54 to experience regiocontrolled dehydration to bridgehead olefin 55 in a protocol that involves the smooth translocation of bridgehead unsaturation with concomitant functionalization as an allylic alcohol.

The formation of the same cyclic ethers from the sulfenylation of unsaturated alcohols and from acid catalysed phenylthio migration on diols via the same intermediate can be explained by kinetic and thermodynamic control. Both reactions are synthetically useful. A result from a previous Tetrahedron Letter is corrected

The bicyclic ketone (+)-17b reacts with vinylmagnesium bromide from its exo face, thus enabling exploitation of the anionic oxy-Cope rearrangement and direct alkylation of the enolate so formed with ethyl iodoacetate to provide 24 in an efficient, stereocontrolled manner. Tricyclic lactone 26b, readily produced from 24, proved to be a common precursor to both synthetic targets. The construction of vulgarolide (9) proceeded by transforming the SEM group in 26b into a mesylate for the purpose of conversion to epoxy acid 39 in advance of oxidative cleavage of the double bond. This transformation resulted in direct conversion to 9 and its epimer 42. Central to the companion synthesis of deoxocrispolide (4) was the ability of 37 to undergo the Michael addition of methoxide ion and of 54 to experience regiocontrolled dehydration to bridgehead olefin 55 in a protocol that involves the smooth translocation of bridgehead unsaturation with concomitant functionalization as an allylic alcohol.

A synthesis of dimethyl gloiosiphone A (2) has been realized. The key step involves the boron trifluoride-catalyzed eliminative ring expansion of a 2′-(dimethoxymethyl)-1′-(cyclopenten-1′-yl)-4-hydroxy-2-cyclobuten-1-one, which was directly assembled by condensation of the appropriate cycloalkenyllithium with dimethyl squarate. Following arrival at spirocyclic diketone 17 in only two steps, the cyclopentenedione A-ring was subjected to controlled reductive removal of one carbonyl group. Subsequent oxidation of the enol ether to unsaturated aldehyde 21 set the stage for proper elaboration of the α-hydroxy ketone part structure in ring B. The spectral properties of the synthetic material were identical to those of natural 2, which was originally isolated as a racemate.

The discovery of the potent and selective prostaglandin D2 (PGD2) receptor (DP) antagonist [(3R)-4-(4-chlorobenzyl)-7-fluoro-5-(methylsulfonyl)-1,2,3,4-tetrahydrocyclopenta[b]indol-3-yl]-acetic acid (13) is presented. Initial lead antagonists 6 and 7 were found to be potent and selective DP antagonists (DP Ki = 2.0 nM for each); however, they both suffered from poor pharmacokinetic profiles, short half-lives and high clearance rates in rats. Rat bile duct cannulation studies revealed that high concentrations of parent drug were present in the biliary fluid (Cmax = 1100 microM for 6 and 3900 microM for 7). This pharmacokinetic liability was circumvented by replacing the 7-methylsulfone substituent present in 6 and 7 with a fluorine atom resulting in antagonists with diminished propensity for biliary excretion and with superior pharmacokinetic profiles. Further optimization led to the discovery of the potent and selective DP antagonist 13.