Proof-of-principle direct double cyclisation of a linear C15-precursor to a dibrominated bicyclic medium-ring ether relevant to Laurencia species

Since the original isolation of Laurencin (1a) in 1965, marine red algae of Laurencia species have provided a wide variety of C15-acetogenic halogenated diastereoand constitutional isomeric monocyclic (C15H21BrO2) and bicyclic (C15H20Br2O2) medium-ring ethers that are oxygenated at both C-6 and C-7 (Fig. 1). Both the monocyclic and bicyclic metabolites have received considerable synthetic attention, with numerous necessarily different strategies used to forge the 7-, 8-, or 9-membered medium-ring, control the cis or trans a,a0-ether stereochemistry, install the requisite halogen(s), and – in the case of the bicyclic ethers – to fashion the second ring. Various recent studies have also been directed at the further understanding of their biogenesis, where the early pioneering work of Murai demonstrated enzymatic bromoetherifications of straight-chain co-isolated unsaturated C15-diols – laurediols (3E,6R,7R)-7a and (3Z,6S,7S)-7b 8 – to monocyclic medium-ring ethers deacetyl laurencin 1b and prelaureatin 2 respectively, albeit in very low yields (Scheme 1, top). We have recently advanced an alternative biogenesis for the monocyclic (C15H21BrO2) medium-ring ethers from Laurencia species from (6S,7R)-epoxide 8 via an intramolecular bromonium ion assisted epoxide ring-opening (IBIAERO) reaction with water functioning as the external nucleophile (Scheme 1, bottom, 8-B-O/O0-1b/2), and experimentally corroborated this with a model epoxide for the concurrent formation of 7-, 8and 9-ring ethers corresponding to the halogenated medium-ring ethers of known metabolites from Laurencia species. The bicyclic metabolites are generally considered to originate by further bromoetherification of the residual unsaturation of the monocyclic compounds – the Z-configured medium-ring alkene or the pendant enyne – using the free alcohol of the original monocyclic compound located either at C-6 or C-7 as the nucleophile (Scheme 1, top). Several laboratory demonstrations of these later transformations have been successful, either as enzymatic-mediated bromoetherifications of naturally occurring monocycles, or as part of the synthetic strategy in a total synthesis of the bicyclic natural products. Interestingly, although bromocyclisation events had been postulated for both monocycle and bicycle formation, prior to our 2012 report and Snyder’s recent elegant work, a non-enzymatic bromonium-ion induced cyclisation process to directly form medium-ring ether cores relevant to Laurencia species had not been reported. Moreover, to the best of our knowledge, there has been no report of a C15-dibrominated bicyclic medium-ring ether relevant to Laurencia species being formed directly from a linear unsaturated C15-precursor by two successive bromination events in the same pot. Herein we report on a successful strategy to effect such a transformation. To investigate the proof-of-principle demonstration of a direct double cyclisation of a C15 unsaturated linear precursor to a bicyclic medium-ring ether relevant to Laurencia species we targeted hexahydroepoxide (6S*,7R*)-[H6]-8, with the aim that this Fig. 1 Representative monocyclic and bicyclic halogenated medium-ring ethers of formulae C15H21BrO2 (1b, 2) and C15H20Br2O2 (3–6) from Laurencia species that are oxygenated at C-6 and C-7. Laurencin 1a is related as the acetate of 1b.

Since the original isolation of Laurencin (1a) in 1965, 1 marine red algae of Laurencia species have provided a wide variety of C 15 -acetogenic halogenated diastereo-and constitutional isomeric monocyclic (C 15 H 21 BrO 2 ) and bicyclic (C 15 H 20 Br 2 O 2 ) medium-ring ethers that are oxygenated at both C-6 and C-7 (Fig. 1). 2 Both the monocyclic and bicyclic metabolites have received considerable synthetic attention, with numerous necessarily different strategies used to forge the 7-, 8-, or 9-membered medium-ring, control the cis or trans a,a 0 -ether stereochemistry, install the requisite halogen(s), and -in the case of the bicyclic ethers -to fashion the second ring. [3][4][5] Various recent studies have also been directed at the further understanding of their biogenesis, 6 where the early pioneering work of Murai 7 demonstrated enzymatic bromoetherifications of straight-chain co-isolated unsaturated C 15 -diols -laurediols (3E,6R,7R)-7a and (3Z,6S,7S)-7b 8 -to monocyclic medium-ring ethers deacetyl laurencin 1b and prelaureatin 2 respectively, albeit in very low yields (Scheme 1, top). 9 We have recently advanced an alternative biogenesis for the monocyclic (C 15 H 21 BrO 2 ) medium-ring ethers from Laurencia species from (6S,7R)-epoxide 8 via an intramolecular bromonium ion assisted epoxide ring-opening (IBIAERO) reaction with water functioning as the external nucleophile (Scheme 1, bottom, 8-B-O/O 0 -1b/2), and experimentally corroborated this with a model epoxide for the concurrent formation of 7-, 8-and 9-ring ethers corresponding to the halogenated medium-ring ethers of known metabolites from Laurencia species. 10,11 The bicyclic metabolites are generally considered to originate by further bromoetherification of the residual unsaturation of the monocyclic compounds -the Z-configured medium-ring alkene or the pendant enyne -using the free alcohol of the original monocyclic compound located either at C-6 or C-7 as the nucleophile (Scheme 1, top). 7 Several laboratory demonstrations of these later transformations have been successful, either as enzymatic-mediated bromoetherifications of naturally occurring monocycles, 12 or as part of the synthetic strategy in a total synthesis of the bicyclic natural products. 13 Interestingly, although bromocyclisation events had been postulated for both monocycle and bicycle formation, prior to our 2012 report 10 and Snyder's recent elegant work, 6b,c a non-enzymatic bromonium-ion induced cyclisation process to directly form medium-ring ether cores relevant to Laurencia species had not been reported. Moreover, to the best of our knowledge, there has been no report of a C 15 -dibrominated bicyclic medium-ring ether relevant to Laurencia species being formed directly from a linear unsaturated C 15 -precursor by two successive bromination events in the same pot. Herein we report on a successful strategy to effect such a transformation.
To investigate the proof-of-principle demonstration of a direct double cyclisation of a C 15 unsaturated linear precursor to a bicyclic medium-ring ether relevant to Laurencia species we targeted hexahydroepoxide (6S*,7R*)-[H 6 ]-8, with the aim that this would undergo an initial IBIAERO reaction via [H 6 ]-B where water functions as both the solvent and the nucleophile (Scheme 2). The use of water in this manner thus guarantees a free hydroxyl group for any subsequent bromoetherification reaction (e.g., [H 6 ]-1b-[H 6 ]-3, Scheme 2) with a second equivalent of an electrophilic bromine source. While we had previously demonstrated successful IBIAERO reactions in water with NBS as the electrophilic bromine source, 11 the attempted IBIAERO reaction of a model epoxide as a truncated C 12 alcohol (inset, Scheme 2) under the same conditions had failed. 10 ‡ We considered that hexahydroepoxide [H 6 ]-8 offered distinct benefits compared to this earlier model and also to epoxide 8 for the proposed experiment: (i) the hydrophilic hexahydro chain may encourage folding of the substrate in water thus inherently facilitating the IBIAREO reaction; (ii) post-IBIAERO reaction, the only region of unsaturation will be located in the medium ring and -compared with the hypothetical use of the putative biosynthetic precursor itself, epoxide 8 -there can be no complicating bromoetherifications to form bromoallene adducts by cyclisation onto any C 1 -C 4 enyne moiety; (iii) hexahydrobicyclic compounds of formulae C 15 H 26 O 2 Br 2 are known in the literature as a consequence of the structural elucidation of the naturally occurring compounds via hydrogenation, 14 providing data for identification of bicyclic products.
Accordingly, epoxide (6S*,7R*)-[H 6 ]-8 was synthesised from bromide 12, itself prepared from (E)-2-penten-1-ol (9) via a known sequence 10,15 with minor modifications. Subsequent coppermediated coupling 16 with hept-1-yne gave novel enediyne 13 (Scheme 3). † Chemoselective and stereoselective hydrogenation 17 afforded (E,Z,Z)-doubly skipped triene 14. Epoxidation of triene 14 with DMDO 18 was found to be entirely selective for the Z-olefins, 19 giving a mixture of mono epoxides (6S*,7R*)-[H 6 ]-8 and 15 which could be separated by chromatography. § ¶8 With epoxide (6S*,7R*)-[H 6 ]-8 in hand, it was treated with two equivalents of NBS -a water stable reagent -under high dilution conditions in water (Scheme 4).** Here, various dibromination adducts, bromohydrin regioisomers, and dibromotetrahydrofurans are expected to be formed by competing processes. 10 In the event, as expected, a complex mixture was obtained that was subjected to extensive chromatography, where 'non-polar' components could be separated away from 'polar' components. † † Much to our delight, by further chromatography of the non-polar components, hexahydrolaureoxanyne [(AE)-[H 6 ]-3] 12a was isolated as a bicyclic medium-ring ether with 1 H NMR data identical to that previously reported, † ‡ ‡ along with dibromoepoxides 16. Thus the desired proof-of-principle has been achieved. This also constitutes the first synthetic route to the laureoxanyne bicyclic medium-ring ether scaffold, and the isolated yield of (AE)- In conclusion, we have demonstrated the proof-of-principle direct cyclisation of a linear unsaturated C 15 -precursor into a C 15 -dibrominated bicyclic medium-ring ether relevant to Laurencia species -where hexahydrolaureoxanyne (AE)-[H 6 ]-3 has an identical bicyclic medium ring ether framework to laureoxanyne 3 -by two successive bromination events in the same pot. These studies are also consistent with epoxide (6S,7R)-8 acting as the biogenetic precursor 10 for bromocyclisation to bicyclic medium-ring ethers of Laurencia species via IBIAERO reactions followed by subsequent bromoetherification events.
We thank the Dinu Patriciu Foundation for funding (to D.-T. S.).
Notes and references ‡ We speculate that the truncated C 12 epoxide suffers from an intramolecular hydrogen bond from the alcohol functional group reducing its nucleophilicity. § 25% of a bis-epoxide was also observed. ¶ Attempted epoxidation of 14 with mCPBA was unselective for the Z-olefins. 8 1 H-13 C and 1 H-1 H NMR correlation spectroscopy were used to distinguish between epoxides (6S*,7R*)-[H 6 ]-8 and 15. † ** In an experiment with 1 equivalent of NBS in water, (AE)-[H 6 ]-3 was isolated in 1.8% yield after extensive chromatography. † † The 'polar' components were expected to contain regioisomeric bromohydrins and dibromohydrins by reference to our earlier work (ref. 10) and were not further characterised. ‡ ‡ The medium-ring bicyclic structure of [H 6 ]-3 is also supported by a characteristic NOESY cross-peak between H 7 and H 9 as previously reported (as an nOe) for 3 (ref. 12a). †