Certain metabolites derived from polyunsaturated fatty acids (PUFAs) play a key role in mammalian physiology, where they orchestrate both inflammatory response as well as the return to homeostasis [1,2,3,4]. By combining total synthesis with chemical biology and molecular pharmacology, a number of distinct eicosanoids and docosanoids have been identified, which are active in the cascade elicited by noxious stimuli [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23]. As a result, natural products with an underlying PUFA motif are of great interest as potential immunomodulators.
Since antiquity, sea dwelling organisms have proven to be a particularly abundant source of new chemical entities, set apart from those found in the terrestrial environment [24,25]. Thus, the ancient Phoenicians were renowned for their trading with Tyrian purple from the Murex sea snail [26,27]. Rising above mere prospecting, modern-day discovery, enabled by the advent of powerful analytical instruments and methods, has found a wealth of bioactive compounds in the marine environment [28,29,30,31,32,33,34].
Ostensibly, mucosin (1) is a natural product that was isolated from the Mediterranean sponge Reniera mucosa as methyl ester 2 . Formally classified as an eicosanoid, it has been conjectured to originate from arachidonic acid (3), based on the C20-architechture (Figure 1). While sharing some noticeable features with the prostane scaffold, the compound differs by having an unusual bicyclic core. Clearly, in the structure proposed for mucosin (1), the characteristic cyclopentane ring is integrated in a cis-fused bicyclo[4.3.0]non-3-ene system. However, turning to the elucidation, the assignment of topology poses a challenge. Though only a small molecule, the structure is compact in terms of the four contiguous stereocentres. In NMR experiments on the isolated methyl ester, pertaining to both 1D- and 2D-techniques, the distinguishing resonances/correlations are ensconced in a crowded aliphatic region. Consequently, with the absence of any coupling pattern to corroborate the configuration of the carbocycle, the assignment published by Casapullo et al. does not convince on its own .
Fascinated by the structure and the prostane-like motif, we devised a practical, divergent and synthetically unambiguous strategy to establish the proposed stereochemistry. At the end of the campaign, capitalizing on X-ray crystallography to pinpoint the relative arrangement, it was concluded that mucosin (1) does not represent the portrayed compound . In a pursuit to identify the natural product isolated from Reniera mucosa, our intent is to achieve the goal by manipulation of the bicyclo[4.3.0]non-3-ene system. We herein detail synthesis of the mucosin diastereomer 1*, demonstrating aspects of the chosen route with regard to stereochemical control (Figure 2). From the point of potential biological activity, the putative structure of mucosin shares some apparent structural similarities with bicyclic prostaglandins. Thus, providing that sufficient amounts could be made available, material could be screened in similar assays. Currently, it is the bearer of unknown properties.
2. Results and Discussion
In 2012, Whitby and co-workers reported that they had completed the first total synthesis of antipodal mucosin (ent-1) . Using zirconium induced co-cyclisation as the pivotal feature, the preliminary experimental work led them to conclude that thermodynamic control would favour the relative stereochemistry assigned by Casapullo et al. . Applied to the actual sequence, elaboration of the key zirconacycle afforded a ~3:1 mixture of diastereomers . While it was conjectured that the major component could be processed to ent-1, the minor component would in turn yield ent-1*. However, although Whitby and co-workers provided data that aligned with the natural product, the authors of the present paper demonstrated irrefutably, that mucosin is not represented by the relative topological connectivity featured in structure 1 . This therefore raised the question as to which diastereomer had been taken on by Whitby and co-workers, and whether mucosin in fact corresponds to structure 1*. In order to resolve this pressing issue, we designed a strategy to access structure 1*.
A central feature in our divergent strategy (Scheme 1) was to take advantage of the efficient desymmetrization of meso-ketone 4 . (Supplementary Materials pp. S3, S4 provides the synthetic sequence to obtain meso-ketone 4) After a chiral foothold had been established, it would then be a matter of introducing a functional pattern amenable for subsequent stereoiteration. Ideally, in order to uncover the topologically deviant point(s), the prerequisite cis-fused keto ester 5 should also be interconvertible with the trans-fused system if need be. However, we first chose to examine the configuration at the appended positions. Thus, along these lines and having previously established a diastereochemical bias, addition of some suitable nucleophile to conjugate ester 7 was judged to follow the precognised trend. By completing the sequence, a new compound 1* with the topology inverted at C8 and C16 would result. Aptly, this could then be named exo-mucosin 1*, since the bulky group added during the stereodifferentiating step was projected to occupy the exo face of the bicycle (Figure 3). Once exo-mucosin 1* had been made, the physical data recorded could be compared against those published for the natural product.
By the developed protocol, our synthesis commenced with desymmetrization of meso-ketone 4 , using Mander’s reagent in combination with the lithium amide of (+)-bis[(R)-phenyethyl]amine (Scheme 2). This chiral amide is sometimes also referred to as Simpkins’ base [38,39,40]. Then, with asymmetric keto ester 9 in hand, conjugated ester 10 was prepared by a three-step procedure, involving sequential manipulation of the keto moiety. Accordingly, the ketone in 9 was reduced, whereupon the corresponding alcohol was turned into a mesylate. Finally, the intermediate mesylate was subjected to base-induced elimination, whereby the Michael acceptor motif was produced.
Having carried out the delineated transformation, the key stereoiterative concept could be tested in the elaboration of conjugated ester 10. While addition to the less hindered exo-face seemed inevitable, the resulting stereochemistry at the ester-appended chiral centre was somewhat uncertain a priori. Simplistically, depending on whether the supervening ester enolate is intercepted by H+ at the equatorial or axial position of C8, the protonated species will correspond to the kinetic and the thermodynamic product, respectively. Reflecting the ambivalent stereochemical nature of the C8-carbanion, and based on our previous experience , conjugate addition to 10 could consequently lead to a mixture of epimers.
In reality, with Cu(I)-catalysed conjugate addition, using BuMgCl as nucleophile in the presence of TMSCl, the reaction gave ester 11 as the sole compound (Scheme 3). Presumably, the Lewis acid takes on dual roles: Not only does TMSCl lower the LUMO of the Michael acceptor, but also stabilizes the ester enolate [41,42,43,44,45,46]. Hence, ester 11 ought to be the conjectured thermodynamic product. Subsequent reduction provided the corresponding carbinol 12, which could also be readily derivatized for the purpose of X-ray analysis. By obtaining suitable crystals of the dinitrobenzoate 12-DNB, the relative configuration of the four contiguous stereocentres could be established (Figure 4). This also confirmed the exo-facial and thermodynamic preference in the reaction of 10, using the specified conditions. (Supplementary Figure S-74 provides a side perspective of the single crystal X-ray structure 12-DNB).
With the intended topological pattern confirmed, carbinol 12 was taken through a course of four steps to install an alkyne handle by the Ohira-Bestmann protocol [47,48,49,50]. For the last step, it may be noted that Taber et al. have provided an interesting alternative to the rather pricy reagent . Although 1H-NMR of the natural product clearly indicates the presence of an E-alkene , the en route aldehyde 13 could also serve as a relay point for Z-selective olefination. However, with the cited observation in mind, alkyne 14 was transformed accordingly to provide the featured E-configured alkenyl ester motif. This was achieved by performing three consecutive reactions in one-pot. Thus, by means of stereospecific hydrometallation [51,52,53,54,55] and halodemetallation , alkyne 14 rendered the corresponding E-vinyl halide as substrate for Pd-catalysed cross-coupling with a commercial zinc reagent [57,58]. The target molecule, exo-mucosin 1*, was then obtained after hydrolysis of ester 15. Finally, re-esterification gave methyl ester 2*, to be compared with the data published by Casapullo et al. .
The cis-fused bicyclo[4.3.0]non-3-ene system is not often encountered in nature. Adhering to the supposition that arachidonic acid (3) is the biogenetic origin of mucosin [59,60,61], the geometry proposed for the core structure invokes a formal disrotatory ring-closure . At a more profound level, the machinery leading to the natural product may traverse any number of pericyclic pathways . Of particular interest, though, is the ongoing discussion regarding whether or not enzyme-catalysed Diels-Alder reactions are implicated in biological systems . The preceding biosynthetic transformation of 3, into a suitable conjugated precursor for cycloaddition, is known to take place in several marine species [59,60,61,64,65,66,67,68,69,70,71,72,73,74,75,76,77]. However, in all the cases where a Diels-Alderase could be claimed to provide the transformative impetus [78,79,80,81,82,83,84,85], the authors of this paper have found no example of cis-fusion.
Cycloaddition via a non-enzymatic pathway is also possible. Thus, Gerwick has proposed allylic carbocations as conceptual intermediates in the biogenesis of marine carbocyclic oxylipins, such as prostaglandin A2 (PGA2) [86,87,88]. In this sense, arachidonic acid (3) provides a link between mucosin and the prostanoid scaffold, pointing towards a possible mechanism. Yet, for the majority of examples found, the annulation produces a trans-1,2-disubstituted cyclopentane ring [89,90,91,92,93,94].
Although being an uncommon structural feature, it would be premature to conclude that the cis-fused bicyclo[4.3.0]non-3-ene system was incongruous. Nevertheless, when recordings were made on methyl ester 2*, the data did not match those reported for the compound isolated from Reniera mucosa. This was most convincingly demonstrated by comparing the 13C-NMR spectra (Table 1): Out of the 20 resonances that are observable for the carbon framework, excluding the methoxy group, 16 display deviating shifts (see also Supplementary Figure S-38). Furthermore, the optical rotation of 2* did not only differ in magnitude, but also in sign: While the naturally occurring material and its purported structure 2 have values of and (, hexane), respectively [35,36] the diastereomer 2* had an (
Copper(I) Catalyzed Reactions of Organolithium and Grignard Reagents,
Erdik, E. Tetrahedron1984, 40, 641.
Copper Assisted Nucleophilic Substitution of Aryl Halogen,
Lindley, J. Tetrahedron1984, 40, 1433.
Organocopper Reagents: Substitution, Conjugate Addition, Carbo/Metallocupration, and Other Reactions,
Lipshutz, B. H.; Sengupta, S. Org. Reactions1992, 41, 135.
Selective Synthesis by Use of Lewis Acids in the Presence of Organocopper and Related Reagents,
Yamamoto, Y. Angew. Chem. I.E.1986, 25, 947.
Applications of Higher-Order Mixed Organocuprates to Organic Synthesis,
B. H. Lipshutz Synthesis1987, 325.
SN2' Additions of Organocopper Reagents to Vinyloxiranes,
Marshall, J. A. Chem. Rev.1989, 89, 1503.
New Tools in Synthetic Organocopper Chemistry,
Nakamura, E. Synlett1991, 539.
New Aspects of Organocopper Reagents: 1,3- and 1,2-Chiral Induction and Reaction Mechanism,
Ibuka, T.; Yamamoto, Y. Synlett1992, 769.
Transmetalation reactions in organocopper chemistry,
Wipf, P. Synthesis1993, 537.
Recent Progress in Higher Order Cyanocuprate Chemistry,
Lipshutz, B.H. Adv. Metal-Organic Chem. Vol. 4, 1995, JAI: Greenwich, Connecticut.
Regioselective and Stereoselective Syntheses with Organocopper Reagents,
Krause, N.; Gerold, A. Angew. Chem. Int. Ed. Engl. 1997, 36, 187-204.
The alkylcopper species generally have lower stability and reactivity than the cuprates, and are sometimes insoluble, so they are not used extensively. Their solubility and reactivity can be improved by complexation with Lewis acids (BF3 or MgBr2) or Lewis bases (phosphines). The dialkyl cuprates are the most generally useful, and most synthetic applications make use of these. Because only one of the ligands are transferred (after the first transfer the product is a usually unreactive alkylcopper), a variety of mixed cuprates have been developed in which one of the ligands does not transfer rapidly (typically cyanide or an acetylide is used, but many other groups such as other stabilized organolithium reagents (thienyllithium, lithiosulfones), dialkyl amides, phosphides or thiolates can be used.
Organocopper reagents are most frequently made by transmetalation of lithium reagents, but organomagnesium, organozinc and organoboron compounds can also be transmetalated. The reactions with lithium reagents generally require a stoichiometric amount of copper salt, so no free lithium reagent remains, since the lithium reagent itself is very reactive toward most substrates. In contrast, the reactions with Grignard and zinc reagents can be catalytic in copper, since these organometallic reagents are less reactive than the organocopper species with typical substrates.
A catalytic Zn/Cu transmetalation-conjugate addition-aldol sequence in the synthesis of Prostaglandin PGE1 Methyl Ester: Arnold, L. A.; Naasz, R.; Minnaard, A. J.; Feringa, B. L. J. Am. Chem. Soc2001, 123, 5841
There are three principal reactions where the organocopper species are more effective than their precursor organolithium or organomagnesium reagents: conjugate addition to α,β-unsaturated carbonyl compounds, coupling with alkyl halides, epoxides or tosylates, and carbometalation of acetylenes.
Organocopper reagents do not react rapidly with ketones, ester or amides, but will react with aldehydes, α,β-unsaturated ketones and esters, acyl halides, alkyl, aryl and vinyl bromides and iodides, epoxides, as well as allyl and propargyl acetates.
Copper Mediated Conjugate Additions
1,4-Addition Reactions of Organocuprates with α,β-Unsaturated Ketones,
Smith, R. A. J.; Vellekoop, A. S. Advances in Detailed Reaction Mechanisms. Vol. 3. James M. Coxon, Ed., JAI Press: Greenwich, CT 1994.
Phosphoramidites: Ligands in Catalytic Asymmetric Cu catalyzed Organozinc Additions
Feringa, B. L. Acc. Chem. Res. 2000, 33, 346-53.
Enantioselective Conjugate Additions.
Sibi, M. P.; Manyem, S. Tetrahedron2000, 56, 8033-61.
Decoding the ‘Black Box' Reactivity that is Organocuprate Conjugate Addition Chemistry.
Woodward, S. Chem. Soc. Rev.2000, 29, 393-401.
Conjugate addition-alkylation - synthesis of Methyl Jasmonate. Green, Crabbé, Tetrahedron Lett.1976, 4867.
Transmetalation of Li to Mg and Cu: Piers, E.; Wai, J. M. Chem. Commun.1987, 1342
Conjugate addition-elimination: Synthesis of β-Elemenone: Majetich, G.; Grieco, P. A.; Nishizawa, N. J. Org. Chem.1977, 42, 2327.
Trimethylsilyl Chloride Accelerated Conjugate Additions
Alkylation of vinyl epoxide - Prostaglandin-PGF1α: Marino, J. P.; Pradilla, R. F.; Laborde, E. J. Org. Chem.1987, 52, 4898
Lithium-Tin Exchange: Dehydroarachidonic Acids. Corey, E. J.; Kang, J. Tetrahedron Lett.1982, 23, 2651.
Carbocupration of acetylenes
Codling Moth Compound: Marfat, A.; McGuirk, P. R.; Helquist, P. J. Org. Chem., 1979, 44, 1345, 3888
Synthesis of Epothilone B: White, J. D.; Carter, R. G.; Sundermann, K. F.; Wartmann, M.. J. Am. Chem. Soc.2001, 123, 5407.
Oxidative Coupling of Organocuprates
Falck, Mekonnen, Yu, Lai J. Am. Chem. Soc.1996, 118, 6096.
© 2012 Hans J. Reich, All Rights Reserved
Web page created by WINPLT.