On the origin of siphonariid polypropionates: Total synthesis of caloundrin B and its isomerization to siphonarin B

Article abstract of DOI:10.1021/ol300432y

Enantioselective synthesis of the enantiomer of caloundrin B was achieved by strategic aldol coupling of an enantiopure trioxaadamantane-containing ketone with a racemic pyrone-containing aldehyde via kinetic resolution. In the presence of imidazole, ent-

Full text of DOI:10.1021/ol300432y

ORGANIC
LETTERS
2012
Vol. 14, No. 6
1648–1651
On the Origin of Siphonariid
Polypropionates: Total Synthesis of
Caloundrin B and Its Isomerization to
Siphonarin B
ꢀ
Fabiola Becerril-Jimenez and Dale E. Ward*
Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon
SK S7N 5C9, Canada
dale.ward@usask.ca
Received February 21, 2012
ABSTRACT
Enantioselective synthesis of the enantiomer of caloundrin B was achieved by strategic aldol coupling of an enantiopure trioxaadamantane-
containing ketone with a racemic pyrone-containing aldehyde via kinetic resolution. In the presence of imidazole, ent-caloundrin B is cleanly
isomerized to ent-siphonarin B confirming the proposed structure and absolute configuration for caloundrin B and establishing that it is a
plausible biosynthetic product from which siphonarin B and baconipyrones A and C can originate.
Caloundrin B (2),1 siphonarin B (3),2 baconipyrone A
(4),3 and baconipyrone C (5)3 are γ-pyrone containing
decapropionates isolated from extracts of the pulmonate
mollusk(falselimpet) Siphonariazelandica.4 The structural
diversity represented in 2ꢀ5 has been postulated to origi-
nate from a common precursor (e.g., 1) via alternative
cyclization/rearrangement cascades possibly orchestrated
by the labile configuration at C-8 and perhaps occurring
during isolation (Scheme 1).5 To test that hypothesis,
we prepared 1 by total synthesis and demonstrated its
isomerization to 3 (70%) under thermodynamic control.6
A mixture of 4 and 5 was obtained from 3 by retro-Claisen
fragmentation of the C8ꢀC9 bond in the presence of
alumina.6 Similar treatment of 1 also gave 4 and 5; how-
ever, the major product was a new structural isomer arising
from a different retro-Claisen pathway (i.e., cleavage at
C7ꢀC8). Those observations suggested that 4 and 5 were
plausible isolation artifacts derived from 3 (but not 1).
However, the origin of 2 was a missing piece of the puzzle
as we were unable to detect its presence in attempts to
induce isomerization of 1 or 3 under a variety of condi-
tions. Consequently, we speculated that 2 was either
thermodynamically unstable relative to 1 and 3,7 the
kinetic barrier for its formation was high, or the proposed
(1) Blanchfield, J. T.; Brecknell, D. J.; Brereton, I. M.; Garson, M. J.;
Jones, D. D. Aust. J. Chem. 1994, 47, 2255–2269.
(2) (a) Hochlowski, J.; Coll, J.; Faulkner, D. J.; Clardy, J. J. Am.
Chem. Soc. 1984, 106, 6748–6750. (b) Garson, M. J.; Jones, D. D.; Small,
C. J.; Liang, J.; Clardy, J. Tetrahedron Lett. 1994, 35, 6921–6924. (c)
Paterson, I.; Franklin, A. S. Tetrahedron Lett. 1994, 35, 6925–6928.
(3) (a) Manker, D. C.; Faulkner, D. J.; Stout, T. J.; Clardy, J. J. Org.
Chem. 1989, 54, 5371–5374. (b) Brecknell, D. J.; Collett, L. A.; Davies-
Coleman, M. T.; Garson, M. J.; Jones, D. D. Tetrahedron 2000, 56,
2497–2502.
(4) For a review on synthesis of pyrone containing natural products,
see: Sharma, P.; Powell, K. J.; Burnley, J.; Awaad, A. S.; Moses, J. E.
Synthesis 2011, 2865–2892.
(5) (a) Garson, M. J.; Goodman, J. M.; Paterson, I. Tetrahedron Lett.
1994, 35, 6929–6932. (b) Davies-Coleman, M. T.; Garson, M. J. Nat.
Prod. Rep. 1998, 15, 477–493.
(6) Beye, G. E.; Ward, D. E. J. Am. Chem. Soc. 2010, 132, 7210–7215.
(7) MM2 calculations have suggested that 2 is ca. 13 kJ/mol less
stable than 3 (ref 5a).
(8) The relative configuration for the trioxaadamantane motif in 2
was established by extensive analysis of NMR spectra; however, the
proposed relative configuration for C10ꢀC14 and the absolute config-
uration for 2 were assigned based on the presumed relationship with its
cometabolite 3 (see ref 1). The sample decomposed to unknown products
during the NMR studies, and additional 2 has apparently not been
isolated.
r
10.1021/ol300432y
Published on Web 02/27/2012
2012 American Chemical Society

structure was incorrect.6,8 In this paper, we report the total
synthesis of ent-2 and its facile isomerization to ent-3.
These results fully corroborate the proposed structure for
2 and imply that it is unlikely to be an isolation artifact.
ring system is a ringꢀchain tautomer of a 3-hydroxy-1,5,7-
trione and all reported syntheses of this acid- and base-
sensitive ring system have relied on a favorable equilib-
rium with ring-opened tautomers established under mild
conditions.15 In the context of synthetic studies on muamvatin
(15), both 1415a and ent-10-epi-1415b were prepared in mod-
erate to excellent yields under thermodynamic control.
Although formation of the caloundrin B ring system is
more challenging,16 we selected the (4S,10S) diastereomer
12 as the core trioxaadamantane fragment with hopes that
it would be a stable tautomer of 13. The thiopyran route to
polypropionates17 was expected to provide stereoselective
access to 13 from the known precursors 9,18 10,19 and 11.20
Ketone 7 would be available from 12 by straightforward
functional group manipulation.
Scheme 1. Siphonariid Decapropionates
The synthesis commenced with desulfurization of 16
(>98% ee), prepared in two steps from 9 and 10,14,18
followed by aldol reaction of the resulting ethyl ketone via
its Li enolate with 1120 to afford 17 as a mixture of
diastereomers in good yield (Scheme 3). Hydrolysis of
the ethylene acetal and silyl ether in 17 followed by
chemoselective oxidation of the C-9 alcohol with the
Dess-Martin periodinane (DMP) and equilibration of
the 3-hydroxy-1,5,7-trione product in the presence of
imidazole provided the hemiacetal 19 in moderate yield.
Despite extensive experimentation, we were unable to
obtain the desired 12 by tautomerization of 19.21 We
speculated that 19 might be favored in the equilibrium
because of a stabilizing intramolecular hydrogen bond
between the anomeric OH group and the benzyl (Bn) ether.
The juxtaposition of these groups is enforced by avoidance
of a syn-pentane interaction between the C-8 Me group
and the substituents at C-10. The C-10 epimer of 19 should
not be similarly stabilized, and to test that hypothesis, a
mixture of adducts 18 was prepared by desulfurization of
ent-16 followed by aldol reaction with 11 as described
above. Gratifyingly, subjecting 18 to the same three-step
sequence that gave 19 from 17 produced a separable 3:1
equilibrium mixture of trioxaadamantane 20 and the
corresponding hemiacetal (i.e., ent-10-epi-19).22
The anticipated7,8 instability of 2, corroborated by our
failed attempts at the synthesis of 2 by isomerization of 1
or 3,6 predicated a synthetic plan based on elaboration
of a preformed 2,4,6-trioxatricyclo[3.3.1.13,7]decan-1-ol
(hereafter “trioxaadamantane”) fragment (Scheme 2).
We chose to assemble the carbon framework of 2 by aldol
coupling of 7 with 8.9 Although both enantiomers of 8 are
known,10 the syntheses are lengthy and the product is
unstable and prone to racemization. We prepared (()-8
from the readily available 611 by treatment of its NaN-
(SiMe3)2 generated anion12 with formaldehyde followed
by oxidation with 2-iodoxybenzoic acid (IBX). Model
studies13 indicated that aldol reactions of (()-8 with Li
and B enolates of 3-pentanone were highly Felkin selective
suggesting that kinetic resolution would be possible in
analogous reactions with suitable enantiopure ketones
(e.g., 7).14 The possibility of selectively obtaining different
adduct diastereomers simply by altering the reaction
conditions was seen as an advantage of this synthetic
approach,14 especially if the proposed relative configura-
tion of 2 was incorrect.8 Formally, the trioxaadamantane
(16) Computational comparison (ref 15e) of truncated models (i.e.,
with an Et substituent at C-9) of the caloundrin B trioxaadamantane
(4R,8R relative configuration) and muamvatin trioxaadamantane
(4S,8R relative configuration) suggests the former is 2 kJ/mol less stable
and the equilibrium for its formation 5 kJ/mol less favorable.
(17) Ward, D. E. Chem. Commun. 2011, 47, 11375–11393.
(18) Ward, D. E.; Jheengut, V.; Beye, G. E.; Gillis, H. M.; Karagiannis,
A.; Becerril-Jimenez, F. Synlett 2011, 508–512.
(19) Ward, D. E.; Rasheed, M. A.; Gillis, H. M.; Beye, G. E.;
Jheengut, V.; Achonduh, G. T. Synthesis 2007, 1584–1586.
ꢀ
(20) de Lemos, E.; Poree, F.-H.; Bourin, A.; Barbion, J.; Agouridas, E.;
Lannou, M.-I.; Commerc-on, A.; Betzer, J.-F.; Pancrazi, A.; Ardisson, J.
Chem.;Eur. J. 2008, 14, 11092–11112.
(21) The anomer of 19 is the direct precursor of 12. See ref 15e for
reaction conditions used for isomerization.
(22) The relative configuration for the trioxaadamantane in 20 was
(9) Formation of the C11ꢀC12 bond by aldol coupling was rejected
because of anticipated difficulties in forming an enolate in the presence
of the pyrone and in achieving the required stereoselectivity.
(10) (a) Arimoto, H.; Yokoyama, R.; Okumura, Y. Tetrahedron Lett.
1996, 37, 4749–4750. (b) Paterson, I.; Chen, D. Y.-K.; Acena, J. L.;
Franklin, A. S. Org. Lett. 2000, 2, 1513–1516. Also see ref 2c.
(11) Mullock, E. B.; Suschitzky, H. J. Chem. Soc. C 1967, 828–830.
(12) Sengoku, T.; Takemura, T.; Fukasawa, E.; Hayakawa, I.;
Kigoshi, H. Tetrahedron Lett. 2009, 50, 325–328.
(13) Details of this study will be reported separately.
(14) Ward, D. E.; Becerril-Jimenez, F.; Zahedi, M. M. J. Org. Chem.
2009, 74, 4447–4454.
confirmed by NMR (CDCl3) as described in ref 15e; δC‑6 = 35.8, δC‑8
=
3
3
(15) (a) Paterson, I.; Perkins, M. V. J. Am. Chem. Soc. 1993, 115,
1608–1610. (b) Dahmann, G.; Hoffmann, R. W. Liebigs Ann. Chem.
1994, 837–845. (c) Hoffmann, R. W.; Dahmann, G. Chem. Ber. 1994,
127, 1317–1322. (d) Lister, T.; Perkins, M. V. Org. Lett. 2006, 8, 1827–
1830. (e) Beye, G. E.; Goodman, J. M.; Ward, D. E. Org. Lett. 2009, 11,
1373–1376.
34.9, δH3CC‑6 = 0.89, JHC4ꢀHC5 = 3.5 Hz, and JHC6ꢀHC8 ≈ 0 Hz are
particularly diagnostic. The absolute configuration of the trioxaadamantane is
established by the absolute configuration of the known 16. The (S) config-
uration at C-10 originates from the known (S)-11; isomerization can be ruled
out because ent-20 was not formed from 17 and ent-19 was not formed
from 18.
Org. Lett., Vol. 14, No. 6, 2012
1649

Scheme 2. Retrosynthetic Analysis of Caloundrin B (2)
Scheme 3. Synthesis of ent-7
Our inability to prepare a trioxaadamantane with the
desired relative configuration (e.g., 12) under thermody-
namic control prompted consideration of routes involving
modification of stable analogues. In that regard, previous
studies showed that protection of the trioxaadamantane
hemiacetal OH group as its trimethylsilyl (TMS) ether
resulted in a much more stable ring system that could
be deprotected under mild conditions.15e Eventually we
settled on a rather lengthy but efficient isomerization
sequence to correct the relative configuration at C-10 in
20 (Scheme 3). After ’locking’ the trioxaadamantane in 20
as its TMS ether, the benzyl ether was hydrogenolyzed and
the resulting alcohol was oxidized with 2-iodoxybenzoic
acid (IBX) to afford aldehyde 21. Treatment of 21 with
imidazole in CHCl3 gave a separable 4:1 equilibrium
mixture of 21 and 22, respectively. After six cycles of
equilibration, 22 (dr 17) was obtained in 49% yield along
with recovered 21 (28%; dr 17). The desired ent-7 was
easily obtained from 22 by treatment with EtMgBr fol-
lowed by IBX oxidation of the resulting alcohol.
that reaction of the (E)-boron enolate of ent-7 with (()-8
should selectively give 23 via preferential reaction with
(S)-8.14 In the event, attempted formation of the boron eno-
late by reaction of ent-7 with (c-C6H11)2BCl and Et3N led to
aldol adducts with very low conversion. However, reaction
of ent-7 with LiNtBu(SiMe3)25 followed by treatment of the
resulting putative (E)-Li enolate with (c-C6H11)2BCl and
then (()-8 (3.8 equiv) produced the desired 23 (60%) along
with a 2:1 mixture of two (of the seven possible) other
diastereomers (18%).26 Comparison of the 13C NMR spec-
trum of 23 with those of model compounds indicated a
12,13-anti-13,14-syn relative configuration as shown;13 how-
ever, the 10,12-syn relative configuration was presumed
based on literature precedent.23
Scheme 4. Synthesis of ent-2 and Its Isomerization to ent-3
To obtain the relative configuration present in 2 from
ent-7 requires stereoselective aldol reaction with (S)-8
(Scheme 4). Considering the high Felkin selectivity
observed13 for aldol additions of 3-pentanone to 8 and
the propensity for (E)-boron enolates of chiral ethyl
ketones bearing an R-methyl substituent to give aldol
adducts with anti relative topicity and 1,3-syn methyl
groups,23 application of the multiplicativity rule24 suggests
(23) (a) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Rieger, D. L. J. Am.
Chem. Soc. 1995, 117, 9073–9074. (b) Jung, M. E.; Salehi-Rad, R.
Angew. Chem., Int. Ed. 2009, 48, 8766–8769. (c) Jung,M. E.;Chaumontet,
M.; Salehi-Rad, R. Org. Lett. 2010, 12, 2872–2875.
(24) (a) Heathcock, C. H.; Pirrung, M. C.; Lampe, J.; Buse, C. T.;
Young, S. D. J. Org. Chem. 1981, 46, 2290–2300. (b) Masamune, S.;
Choy, W.; Petersen, J. S.; Sita, L. R. Angew. Chem., Int. Ed. Engl. 1985,
24, 1–30.
1650
Org. Lett., Vol. 14, No. 6, 2012

Diastereoselective reduction27 of 23 directed by the OH
group at C-13 gave diol 24 (dr >20) (Scheme 4). Reaction
of 24 with Et3SiOSO2CF3 led to selective protection of the
of the C-11 OH group; however, because any 11,13-bis-
(silyloxy) byproduct could not be recycled, it was most
efficient to run the reaction to <50% conversion. IBX
oxidation of the resulting 25 followed by treatment
for 2. In addition, these results firmly establish that calo-
undrin B (2) is thermodynamically much less stable than
siphonarin B (3) (and 1) confirming previous calculations7
and explaining our failure6 to produce 2 by isomerization
of 3 (or 1). Consequently, 2 cannot be an isolation artifact
of 1 or 3 but presumably is formed in an enzyme-mediated
process. Accordingly, 2 must now be considered as a
plausible biosynthetic product from which the formation
of 3ꢀ5 can be readily explained.6
with HF pyridine gave ent-2 in near-quantitative yield.
3
Spectroscopic data (1H and 13C NMR, IR, MS) for ent-2
([R]D þ50; c 0.2, CHCl3) were fully consistent with those
reported1 for 2 ([R]D ꢀ19; c 0.16, CHCl3).
In conclusion, the enantioselective total synthesis of ent-
caloundrin B was achieved in 18 linear steps from (()-9.
The key features of the synthesis involved the following:
(i) synthesis of the challenging trioxaadamantane fragment
ent-7 by isomerization of a 10-epi derivative formed under
thermodynamic control; (ii) assembly of the carbon skele-
ton by a novel aldol reaction of (()-8 and ent-7 via kinetic
resolution with formation of the required (E)-boron en-
olate by borylation of the corresponding Li enolate. Iso-
merization of ent-2 into ent-3 confirms the proposed
relative and absolute configuration of 2 and establishes
that 2 is not an isolation artifact but is a probable bio-
synthetic product.
To explore the thermodynamic stability and isomeriza-
tion of caloundrin B, ent-2 was treated with imidazole in
CDCl3 at room temperature.28 After 24 h, a mixture of ent-2,
ent-3 (50% isolated), and ent-1 (mixture of hemiacetals)6
was obtained. Prolonged reaction (5.5 d) led to complete
consumption of ent-2 and produced ent-5 (20% isolated)
and ent-14-epi-56 along with ent-3 (30% isolated), ent-1,
and other unidentified components. A similar experiment
conducted in acetonitrile was monitored by HPLC. After 8
days, a relatively clean transformation to ent-3 (ca.70%)
and ent-1 (mixture of hemiacetals)6 was observed with no
more than traces of ent-2 remaining. Spectroscopic data
(1H and 13C NMR) for ent-3 ([R]D ꢀ50; c 0.1, CHCl3) were
fully consistent with those reported2a for 3 ([R]D þ13.2; c
0.01361, CHCl3). The isolation of ent-3 from these experi-
ments clearly establishes that 2 and 3 share the same abso-
lute configurations at all stereocenters (except C-8) and
confirms the proposed relative and absolute configuration
Acknowledgment. We thank C. K. (Leon) Lai and M.
Mehdi Zahedi (University of Saskatchewan) for prepara-
tion of (()-8 and ent-16, respectively, and Prof. Mary
Garson (University of Queensland) for providing copies
of NMR spectra for caloundrin B. Financial support from
the Natural Sciences and Engineering Research Council
(Canada) and the University of Saskatchewan is gratefully
acknowledged.
(25) Xie, L.; Isenberger, K. M.; Held, G.; Dahl, L. M. J. Org. Chem.
1997, 62, 7516–7519.
(26) Scattered examples of formation of B enolates from Li enolates
have been reported. For ketones, see: (a) Hoffmann, R. W.; Froech, S.
Tetrahedron Lett. 1985, 26, 1643–1646. (b) Hoffmann, R. W.; Ditrich,
K.; Froech, S. Liebigs Ann. Chem. 1987, 977–985. (c) Suginome, M.;
Uehlin, L.; Yamamoto, A.; Murakami, M. Org. Lett. 2004, 6, 1167–
1169. For carboxylicacid derivatives, see: (d) Lang, F.; Zewge, D.; Song,
Z. J.; Biba, M.; Dormer, P.; Tschaen, D.; Volante, R. P.; Reider, P. J.
Tetrahedron Lett. 2003, 44, 5285–5288. (e) Burke, E. D.; Gleason, J. L.
Org. Lett. 2004, 6, 405–407 and references cited therein.
(27) Chen, K. M.; Hardtmann, G. E.; Prasad, K.; Repic, O.; Shapiro,
M. J. Tetrahedron Lett. 1987, 28, 155–158.
Supporting Information Available. Experimental pro-
cedures, characterization data, and copies of 1H and 13
C
NMR spectra for all reported compounds; comparison of
NMR data for natural and synthetic material; 1H NMR
and HPLC data for isomerization of ent-2. This material
is available free of charge via the Internet at http://pubs.
acs.org.
(28) A sample of ent-2 in CDCl3 was unchanged after 3 months
The authors declare no competing financial interest.
at ꢀ20 °C.
Org. Lett., Vol. 14, No. 6, 2012
1651