The first total synthesis of (+)-mucosin

Article abstract of DOI:10.1039/c2cc17915f

The first total synthesis of (+)-mucosin has been completed allowing assignment of the absolute stereochemistry of the natural product. A zirconium induced co-cyclisation was utilised to install the correct stereochemistry of the four contiguous stereocentres around the unusual bicyclo[4.3.0]nonene core.

Full text of DOI:10.1039/c2cc17915f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 3409–3411
www.rsc.org/chemcomm
COMMUNICATION
The first total synthesis of (+)-mucosinwz
Alan R. Henderson, Jozef Stec, David R. Owen and Richard J. Whitby*
Received 19th December 2011, Accepted 10th February 2012
DOI: 10.1039/c2cc17915f
The first total synthesis of (+)-mucosin has been completed
allowing assignment of the absolute stereochemistry of the
natural product. A zirconium induced co-cyclisation was utilised
to install the correct stereochemistry of the four contiguous
stereocentres around the unusual bicyclo[4.3.0]nonene core.
Our synthetic strategy (Scheme 1) was to use the zirconocene-
(1-butene) induced co-cyclisation of triene 6 to zirconacycle 7 to
control the relative stereochemistry of the four contiguous chiral
centres in mucosin.4 Further elaboration of the zirconacycle
through allyl carbenoid insertion to afford 8 followed by
1,4-addition to an acrylate equivalent should provide a concise
completion of the synthesis with control of the alkene
stereochemistry.5
In 1997 (ꢀ)-mucosin 1, a novel bicyclic eicosanoid was isolated
from the marine sponge Reniera mucosa in the Mediterranean.1
The structure of its methyl ester 2 was determined using mass,
infrared and high field NMR spectroscopy. The relative
stereochemistry was determined using a series of 2D NMR
experiments. Mucosin is an interesting synthetic target given
its unusual bicyclo [4.3.0]nonene core and close structural
relation to the biologically important leukotrienes and prostanoids,
including PGE2 (3), thromboxane and prostacyclin, with
which it presumably shares a common precursor in arachidonic
acid (4).2 The only similar structure of which we are aware is the
C22 compound dictyosphaerin (5).3 Herein we report the first
total synthesis of (+)-mucosin and hence establish the absolute
stereochemistry of the natural product.
The synthesis of triene 6 was achieved in eight steps from the
commercially available tetrahydrophthalic acid anhydride 9
(Scheme 2). The first step was the asymmetric ring opening of
anhydride 9 for which there are wide variety of methods.6 We
found that catalysis of methanolysis with a chiral amine
(alkaloid) as established by Bolm to be the simplest and
cheapest.7 Use of quinidine gave the (1R,6S) enantiomer of
acid ester 10 in 100% yield and 90% ee. To obtain the opposite
enantiomer quinine could have been used as the catalyst.7 The
selective reduction of acid ester 10 using Super-Hydrides
furnished lactone 11.8 Partial reduction of 11 using DIBAL-H
gave a lactol which underwent Wittig methylenation to yield
alcohol 12.9 The second alkene branch was installed by conversion
of 12 to its mesylate followed by cyanide displacement to give
nitrile 13. DIBAL-H reduction and treatment of the resultant
aldehyde with butylidenetriphenylphosphorane gave triene 6
as a 4 : 1 mixture of Z : E isomers.
Control of the four contiguous stereocentres of mucosin is
challenging, but precedent is that 3-zirconabicyclo[3.3.0]octanes
form with trans-ring junction stereochemistry (thus controlling
the relative stereocentres c and d—Scheme 3), and there is
generally good trans-1,2-control between adjacent substituents
and the zirconacyclopentane ring, which would give the correct
Chemistry, University of Southampton, Southampton,
Hampshire SO17 1BJ, UK. E-mail: R.J.Whitby@soton.ac.uk;
Fax: +44 238059 3781; Tel: +44 238059 2777
w Electronic supplementary information (ESI) available: Experimental
procedures and characterisation data for all new compounds; compar-
ison of data with that reported for the natural product; and additional
crystallographic evidence for the relative stereochemistry of the major
isomer formed. See DOI: 10.1039/c2cc17915f
z Dedicated to Prof. Phil Parsons on the occasion of his 60th birthday.
Scheme 1 Synthetic strategy for the synthesis of mucosin.
Chem. Commun., 2012, 48, 3409–3411 3409
c
This journal is The Royal Society of Chemistry 2012

View Article Online
Only the more stable epimers are shown as the energy differences
were calculated to be >16 kJ molꢀ1
.
It is known that the formation of zirconacyclopentanes is
reversible and heating is sometimes needed to form the
thermodynamically more stable product.13 Zirconium induced
co-cyclisation of triene 6 was monitored by quenching aliquots
to give the protonated compounds 14, which were assayed by
GC. After 1.25 h at room temperature we observed a
26 : 68 : 4 : 2 ratio, in order of GC retention times, of isomers
formed with around 34% of 6 uncyclised, whereas after
heating for 0.5 h at 65 1C the cyclisation was complete and
the corresponding isomer ratio was 63 : 24 : 11 : 2. Further
heating did not change the ratio significantly and additional
products started to form. In order to identify if the kinetic or
thermodynamic isomer was needed for mucosin we further
elaborated the zirconacycles formed under each condition
by insertion of 1-lithio-1-chloro-2-propene. The insertion
occurred exclusively into the unsubstituted C–Zr bond to
afford 15, followed by the BF3ꢁEt2O promoted addition of
benzaldehyde to give the alcohols 16.14 Chromatography gave
samples of 16 comprising 3.1 : 1 and 1 : 1.5 of major diastereo-
isomers (ignoring the hydroxyl stereochemistry) from the
thermodynamic and kinetic conditions, respectively. By
comparing the spectral data of the obtained alcohols 16 with
those of mucosin 1, we were able to establish that the major
product obtained under thermodynamic conditions was 16a
derived from zirconacycle 7a with the desired stereochemistry.
We speculate, but have not proven, that the major isomer
under kinetic conditions is the zirconacycle 7b.
Scheme 2 Synthesis of triene 6. Reagents and Conditions: (a) MeOH,
quinidine, PhMe, ꢀ55 1C, 8 h, then ꢀ18 1C, 3 days then 2.0 M HCl;
(b) Li(Et)3BH, THF, 0 1C, 1 h then rt, 15 h then 2.0 M HCl;
(c) DIBAL-H (fast addition), PhMe, ꢀ78 1C, 1 h, then 3.0 M HCl;
(d) MePh3P+Brꢀ, n-BuLi, THF, 0 1C - rt, 2 h, then 1.0 M HCl;
(e) MsCl, Et3N, DMAP, THF, 0 1C, 2 h; (f) KCN, NaI, 18-crown-6,
90 1C, 66 h; (g) DIBAL-H (dropwise addition), THF, ꢀ78 1C - rt,
2 h, then MeOH, aq NaHCO3; (h) BuPh3P+Brꢀ, n-BuLi, THF,
0 1C - rt, 2 h, then 1.0 M HCl.
With reasonable conditions for predominant formation of
the correct stereoisomer we examined introduction of the
carboxy containing side chain. We have previously shown that
allyl-zirconacycles similar to 15 react with the diethylacetal of
acrolein to afford around 20% of the 1,4-addition product.15
However, despite extensive investigation of different acrylate
equivalents and conditions on model systems we were unable
to achieve good yields of selective 1,4-addition.
An alternative stepwise approach was developed to complete the
total synthesis. Insertion of the silyl carbenoid 1716 into zircona-
cycle mixture 7 formed under thermodynamic conditions gave the
zirconacyclohexane 18. Subsequent protonation and the Woerpel
modification of the Fleming–Tamao oxidation17 yielded alcohol 19
as a 2.7 : 1 mixture of diastereoisomers in 75% overall yield
from triene 6. The pure desired diasteroisomer 19a could be
seperated by HPLC. Removal of the minor diastereoisomer was
also achieved by refluxing the mixture with iodine in benzene
then purification by normal chromatography. We assume that
the unwanted diastereoisomer was removed via iodoetherifica-
tion although such a product was not isolated. Unfortunately
the pure alcohol 19a was only recovered in 39% yield. Despite
the low recovery this method provided further evidence for the
relative stereochemistry of the major diastereoisomer given
that alcohol 19a cannot undergo iodoetherification, and some
evidence that the minor isomer is 19b which can.
Scheme 3 Zirconocene induced co-cyclisation of 6 and synthesis of
alcohols 16.
relative stereochemistry at centres b and c.4 The effect of the
stereocentre at a, and indeed of the fused 6-member ring was
harder to predict.10 Energies from Density Function Theory
calculations have been shown to be useful in predicting the
relative stability of zirconacycles.11 The four most stable
zirconacycle structures 7a–d, which should form from 6 are
shown in Scheme 3 with their calculated relative energies and
indicate that the desired isomer 7a is favoured. There are four
additional stereoisomers differing in the configuration of the propyl-
substituent, but we have shown in other systems that this centre is
independent of the stereochemistry of the starting alkene due to
rapid epimerisation via exocyclic b-hydride elimination/readdition.12
Oxidation of the alcohol 19a to aldehyde 20 followed by
Takai olefination18 with diiodide 21 and removal of the silicon
protecting group with TBAF gave the alcohol 23 in 81% overall
yield. Finally alcohol 23 was oxidised to mucosin 1 in 84% yield
using PDC in DMF (Scheme 4).19 Conversion of acid 1 to its
c
3410 Chem. Commun., 2012, 48, 3409–3411
This journal is The Royal Society of Chemistry 2012

View Article Online
[a]2D6 = + 38.21 (c = 0.8, n-hexane) compared with that reported
for the methyl ester of the natural product of [a]2D6 ꢀ35.5 1C
(n-hexane, c = 0.8) proving that the natural product is (E)-7-
((1S,2R,3aS,7aS)-2-butyl-2,3,3a,4,7,7a-hexahydro-1H-inden-1-yl)-
hept-5-enoic acid.
In conclusion the absolute stereochemistry of the natural
product (ꢀ)-mucosin has been established and the first total
synthesis of its enantiomer (+)-mucosin has been completed in
14 steps and 7% overall yield. (ꢀ)-Mucosin could easily be made
by using quinine in the enantioselective ring opening of 9.
We thank GlaxoSmithKline, the EPSRC and ERDF
(IS:CE-Chem & InterReg IVa program 4061) for funding
this work.
Notes and references
1 A. Casapullo, G. Scognamiglio and G. Cimino, Tetrahedron Lett.,
1997, 38, 3643–3646.
2 Prostaglandins, Leukotrienes and other Eicosanoids: From biogenesis
to clinical application, ed. F. Marks and G. Furstenberger,
Wiley-VCH, 1999.
¨
3 S. J. Rochfort, R. Watson and R. J. Capon, J. Nat. Prod., 1996, 59,
1154–1156.
4 C. Rousset, D. Swanson, F. Lamaty and E.-I. Negishi, Tetrahedron
Lett., 1989, 30, 5105; D. F. Taber and J. P. Louey, Tetrahedron,
1995, 51, 4495–4506.
5 T. Luker and R. J. Whitby, Tetrahedron Lett., 1995, 36,
4109–4112.
6 I. Atodiresei, I. Schiffers and C. Bolm, Chem. Rev., 2007, 107,
5683–5712.
7 C. Bolm, I. Schiffers, C. L. Dinter and A. Gerlach, J. Org. Chem.,
2000, 65, 6984–6991.
8 G. S. Hamilton, Z. Huang, X. J. Yang, R. J. Patch,
B. A. Narayanan and J. W. Ferkany, J. Org. Chem., 1993, 58,
7263–7270.
9 A. T. Stevens, J. R. Bull and K. Chibale, Org. Biomol. Chem., 2008,
6, 586–595.
10 D. R. Owen and R. J. Whitby, Synthesis, 2005, 2061–2074.
11 D. F. Taber, J. P. Louey, Y. Wang, W. A. Nugent, D. A. Dixon
and R. L. Harlow, J. Am. Chem. Soc., 1994, 116, 9457–9463.
12 E. Thomas, S. Dixon and R. J. Whitby, Tetrahedron, 2007, 63,
11686–11701.
13 D. F. Taber, J. P. Louey and J. A. Lim, Tetrahedron Lett., 1993,
34, 2243–2246.
14 G. J. Gordon, T. Luker, M. W. Tuckett and R. J. Whitby,
Tetrahedron, 2000, 56, 2113–2129.
15 T. Luker and R. J. Whitby, Tetrahedron Lett., 1995, 36,
4109–4112.
16 S. Dixon, S. M. Fillery, A. Kasatkin, D. Norton, E. Thomas and
R. J. Whitby, Tetrahedron, 2004, 60, 1401–1416.
17 J. H. Smitrovich and K. A. Woerpel, J. Org. Chem., 1996, 61,
6044–6046.
Scheme 4 Completion of the synthesis of mucosin. Reagents and conditions:
(a) THF, ꢀ78 1C, 45 mins; (b) NaHCO3, MeOH, ꢀ78 1C - rt, 15 h;
(c). KH, NMP, t-BuOOH, 0 1C - rt, 10 mins then TBAF, 70 1C, 14 h.
(d) Me2SO, CH2Cl2, oxalyl chloride, ꢀ60 1C, 15 min then Et3N,
ꢀ60 1C - rt, 1 h; (e) CrCl2, DMF, THF, rt, 2.5 h; (f) TBAF, THF, rt,
3 h; (g) PDC, DMF, 0 1C - rt, 15 h. (h) CH2N2, Et2O, ꢀ78 1C - rt, 1 h.
18 T. Okazoe, K. Takai and K. Utimoto, J. Am. Chem. Soc., 1987,
109, 951–953.
19 E. J. Corey and G. Schmidt, Tetrahedron Lett., 1979, 20,
399–402.
methyl ester 2 using diazomethane allowed comparison with the
reported data demonstrating excellent agreement.1 The optical
rotation of the synthetic mucosin methyl ester (90% ee) was
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 3409–3411 3411