Stereoselective synthesis of the western hemisphere of salinomycin

Article abstract of DOI:10.1021/ol052900w

A convergent and module-based strategy for the asymmetric synthesis of the western hemisphere (C1-C17 fragment) of salinomycin has been devised. This new synthetic approach relies on highly stereoselective C-glycosidation and aldol processes.

Full text of DOI:10.1021/ol052900w

ORGANIC
LETTERS
2006
Vol. 8, No. 3
527-530
Stereoselective Synthesis of the
Western Hemisphere of Salinomycin
Igor Larrosa, Pedro Romea,* and Fe`lix Urp´ı*
Departament de Qu´ımica Orga`nica, UniVersitat de Barcelona, Mart´ı i Franque´s 1-11,
08028 Barcelona, Catalonia, Spain
pedro.romea@ub.edu; felix.urpi@ub.edu
Received November 30, 2005
ABSTRACT
A convergent and module-based strategy for the asymmetric synthesis of the western hemisphere (C1−C17 fragment) of salinomycin has
been devised. This new synthetic approach relies on highly stereoselective C-glycosidation and aldol processes.
Salinomycin (1, Figure 1) is a polyether ionophore antibiotic
first isolated from a culture broth of Streptomyces albus in
In our case, we were particularly attracted by the tetra-
hydropyran ring flanked by an R-alkyl acetic acid and a
polyketide chain at the C3 and C7 positions, respectively
(Scheme 1), a structural arrangement that salinomycin shares
with other prominent polyether ionophore antibiotics such
as narasin, zincophorin, and X-206.2 Recognition of this
structural motif suggested to us the opportunity of devising
a common approach to the total synthesis of these molecules.
Supporting this new approach, we document herein a highly
stereoselective construction of the C1-C17 fragment of
salinomycin based on C-glycosidation7 and substrate-
controlled aldol8 processes developed in our group.
Figure 1. Salynomicin (1).
1973 and is widely recognized by its remarkable antibacterial
and anticoccidial properties.1,2 Structurally, its molecular
architecture is dominated by the presence of a plethora of
chiral centers embedded in a polyoxygenated backbone,
which encompass two substituted tetrahydropyrans and a
complex tricyclic bis-spiroacetal system. Then, it is not
surprising that the appealing biological properties combined
with the synthetic challenge raised by such intricate and
densely functionalized structure have stimulated much
synthetic efforts, which have culminated in three total
syntheses3-5 and new approaches to the spiroacetal core.6
(3) First total synthesis of salinomycin disclosed by Y. Kishi in 1982
has been solely published as a lecture. For further comments, see ref 2b.
(4) (a) Horita, K.; Oikawa, Y.; Yonemitsu, O. Chem. Pharm. Bull. 1989,
37, 1698-1704. (b) Horita, K.; Nagato, S.; Oikawa, Y.; Yonemitsu, O.
Chem. Pharm. Bull. 1989, 37, 1705-1716. (c) Horita, K.; Oikawa, Y.;
Nagato, S.; Yonemitsu, O. Chem. Pharm. Bull. 1989, 37, 1717-1725. (d)
Horita, K.; Nagato, S.; Oikawa, Y.; Yonemitsu, O. Chem. Pharm. Bull.
1989, 37, 1726-1730.
(5) Kocienski, P. J.; Brown, R. C. D.; Pommier, A.; Procter, M.; Schmidt,
B. J. Chem. Soc., Perkin Trans. 1 1998, 9-39.
(6) For a review, see: Brimble, M. A.; Fare`s, F. A. Tetrahedron 1999,
55, 7661-7706.
(7) Larrosa, I.; Romea, P.; Urp´ı, F.; Balsells, D.; Vilarrasa, J.; Font-
Bardia, M.; Solans, X. Org. Lett. 2002, 4, 4651-4654.
(8) (a) Figueras, S.; Mart´ın, R.; Romea, P.; Urp´ı, F.; Vilarrasa, J.
Tetrahedron Lett. 1997, 38, 1637-1640. (b) Galobardes, M.; Gasco´n, M.;
Mena, M.; Romea, P.; Urp´ı, F.; Vilarrasa, J. Org. Lett. 2000, 2, 2599-
2602. (c) Solsona, J. G.; Romea, P.; Urp´ı, F.; Vilarrasa, J. Org. Lett. 2003,
5, 519-522. (d) Solsona, J. G.; Romea, P.; Urp´ı, F. Tetrahedron Lett. 2004,
45, 5379-5382.
(1) Kinashi, H.; Otake, N.; Yonehara, H.; Sato, S.; Saito, Y. Tetrahedron
Lett. 1973, 49, 4955-4958.
(2) For recent reviews on polyether ionophore antibiotics, see: (a) Dutton,
C. J.; Banks, B. J.; Cooper, C. B. Nat. Prod. Rep. 1995, 12, 165-181. (b)
Faul, M. M.; Huff, B. E. Chem. ReV. 2000, 100, 2407-2473.
10.1021/ol052900w CCC: $33.50
© 2006 American Chemical Society
Published on Web 01/06/2006

Scheme 1. Retrosynthetic Analysis of Salinomycin
As outlined in Scheme 1, our retrosynthetic analysis relies
Ti(i-PrO)Cl3/i-Pr2NEt followed by the addition of a solution
of 10 at -78 °C provided the corresponding syn/anti Felkin
aldol adduct 8 as a single diastereomer (85%, dr > 97:3).
Finally, protection of the C13 alcohol and regioselective
removal of the C10 OTBS group using SmI2 12 afforded the
required ketone 3 in 85% yield over the two-step procedure
from 8.
on the strategic disconnection of the C17-C18 bond. Thus,
salinomycin is split in two halves of similar size and
structural complexity that we have named hemispheres.
Further elaboration of the western hemisphere reveals the
protected C1-C17 fragment 2 shown in Scheme 1 as our
actual target. Following this analysis, we anticipated that the
successful construction of 2 would mainly depend on three
crucial steps: (i) a double asymmetric anti aldol reaction of
ketone 3 and aldehyde 4; (ii) a Lewis acid mediated
C-glycosidation involving titanium enolate from (S)-N-
butanoyl-4-isopropyl-1,3-thiazolidine-2-thione (6) and glycal
7; and (iii) a double asymmetric syn aldol reaction arising
from R-hydroxy ketone 9 and aldehyde 10. In summary, our
synthetic approach is rooted in a highly convergent and
module-based strategy that depends on the stereoselective
coupling of subunits represented in Scheme 1.
Scheme 2. Synthesis of the C10-C17 Subunit 3
Keeping in mind these ideas, we first proceeded to the
synthesis of the C10-C17 subunit 3. Taking advantage of
the highly stereoselective titanium-mediated aldol reactions
based on lactate-derived ketones developed in our group,8
we envisaged that it could arise from a matched double
asymmetric aldol reaction.9 Particularly, the synthesis of
ketone 3 began with a titanium-mediated aldol reaction of
the readily available R-tert-butyldimethylsilyloxy ketone 910
and chiral aldehyde 1011 (Scheme 2). Enolization of 9 with
Having prepared the C10-C17 subunit, our attention was
focused on the preparation of the tetrahydropyran moiety.
Studies on C-glycosidation processes involving titanium
enolates derived from N-acyl-1,3-thiazolidine-2-thiones have
established that the stereochemical outcome of such cross-
coupling reactions mainly depends on the C6 group of the
glycal and the configuration of the chiral auxiliary.7 In our
case, ester 5 might arise from N-butanoyl derivative 6 and
glycal 7 shown in Scheme 1. Then, simultaneous hydrogena-
tion of the alkene and removal of the benzyl protecting group
followed by oxidation of the resulting alcohol should lead
to the C1-C9 subunit 4.
(9) For double asymmetric aldol reactions involving lactate-derived
ketones, see: Solsona, J. G.; Nebot, J.; Romea, P.; Urp´ı, F. Synlett 2004,
2127-2130.
(10) Ferrero´, M.; Galobardes, M.; Mart´ın, R.; Montes, T.; Romea, P.;
Rovira, R.; Urp´ı, F.; Vilarrasa, J. Synthesis 2000, 1608-1614.
(11) Chiral aldehyde 10 was prepared by Swern oxidation of the parental
alcohol, which in turn was obtained following reported procedures. See:
Vong, B. G.; Abraham, S.; Xiang, A. X.; Theodorakis, E. A. Org. Lett.
2003, 5, 1617-1620.
(12) Paterson, I.; Wallace, D. J.; Cowden, C. J. Synthesis 1998, 639-
652.
528
Org. Lett., Vol. 8, No. 3, 2006

In accordance with this analysis, we applied our efforts
to preparing glycal 7. Unfortunately, no stereoselective
methodology afforded it in synthetically useful amounts. This
failure jeopardized the whole approach, and we were required
to modify our plans. Close inspection of the crucial C-
glycosidation step suggested that other options could be
considered. Indeed, although the mechanistic details of the
above-mentioned C-glycosidation procedure are still under
scrutiny, it is likely that the reaction proceeds through the
oxocarbenium I shown in Scheme 3 via an SN1-like process.
4). Next, clean removal of the chiral auxiliary with DIBALH
produced aldehyde 16 in a single step and excellent yield.15
Submission of 16 to Still-Gennari olefination conditions16
afforded R,â-unsaturated ester 17 (96% yield, Z/E ratio 92:
8), which was easily converted into lactone 18. Finally, a
two-step sequence based on a DIBALH reduction of 18 and
O-methyl glycosidation of the resulting lactol afforded the
desired pseudoglycal 12 as a mixture of diastereomers (dr
90:10) in 97% yield. The stage was now set for the critical
C-glycosidation step. Gratifyingly, experimental conditions
previously optimized for glycals7 could be successfully
applied to (S)-N-butanoyl-4-isopropyl-1,3-thiazolidine-2-
thione (6) shown in Scheme
1
and pseudo-
Scheme 3
glycal 12 in such a way that adduct 19 was isolated as a
1
single diastereomer (dr > 97:3 by H NMR). At this point,
configuration of new stereocenters C2-C3 was secured
through extensive 1D and 2D NMR analysis.
Furthermore, no chromatographic purification of the
reaction mixture was required and methyl ester 5 was
smoothly obtained in 75% overall yield from 12 by removal
of the chiral auxiliary with methanol in the presence of
catalytic DMAP at room temperature, as shown in Scheme
4.
With methyl ester 5 in hand, it was envisaged that
simultaneous hydrogenation of the C4-C5 double bond and
cleavage of the C9 benzyl ether followed by oxidation of
the resulting alcohol would afford the C1-C9 subunit 4
shown in Scheme 1. However, submission of 5 to the usual
reductive conditions employing 10% Pd/C as catalyst
provided the expected hydroxy ester 20 contaminated by epi-
C3 diastereomer 21.17
After careful optimization, it was established that formation
of the undesired diastereomer 21 could be avoided applying
a two-step sequence: CdC double bond hydrogenation in
the presence of PtO2, followed by hydrogenolysis of the
benzyl protecting group with 10% Pd/C as catalyst (Scheme
5). At this point, the stereochemical array of 20 was nicely
Therefore, glycal 7 might be replaced by any other species
able to provide the putative cationic intermediate I by means
of Lewis acid activation, such as pseudoglycal 12 represented
in Scheme 3.
Then, all of the stereocenters present in the required
oxocarbenium I were envisaged to arise from a stereoselec-
tive aldol reaction. According to this approach, titanium-
mediated aldol13 reaction of (S)-4-isopropyl-N-propanoyl-
1,3-thiazolidine-2-thione (13) and (S)-3-benzyloxy-2-methyl-
propanal (14)14 followed by treatment of the resulting
diastereomeric mixture with TBSOTf/lutidine provided enan-
tiomerically pure protected aldol 15 in 60% yield over two
steps after a simple chromatographic purification (Scheme
Scheme 4. Synthesis of the C1-C9 Ester 5
Scheme 5. Synthesis of the C1-C9 Subunit 4: Final Steps
confirmed through comparison with the spectroscopic data
previously reported.4d Eventually, Swern oxidation of alcohol
20 completed the synthesis of the C1-C9 subunit 4.
(13) Crimmins, M. T.; King, B. W.; Tabet, E. A.; Chaudhary, K. J. Org.
Chem. 2001, 66, 894-902.
(14) Roush, W. R.; Palkowitz, A. D.; Palmer, M. J. J. Org. Chem. 1987,
52, 316-318.
(15) Spectroscopic data of 16 match those of ent-16 previously reported.
See Supporting Information.
Org. Lett., Vol. 8, No. 3, 2006
529

With key subunits 3 and 4 in hand, we faced their crucial
coupling to obtain the required C1-C17 fragment 2. Since
initial studies highlighted how heating of salinomycin
derivatives triggers a clean retroaldol reaction affecting the
C9-C10 bond, the synthetic approaches reported so far rely
on the construction of such strategic bond through a
stereoselective aldol reaction.2-5 In our case, the scenario
was designed for a boron-mediated double asymmetric aldol
reaction.18 However, enolization of 3 with Chx2BCl/Et3N
followed by addition of aldehyde 4 gave low and nonrepro-
ducible yields. Fortunately, the more reactive (E) lithium
enolate counterpart19 provided the double asymmetric anti
aldol reaction stereoselectively (dr 90:10), and a simple
purification by flash chromatography permitted the isolation
of enantiomerically pure C1-C17 fragment 2 in 61% yield,
as outlined in Scheme 6.
Scheme 6. C1-C17 Fragment 2: Final Coupling
processes. Our synthetic approach is rooted on a convergent
and module-based strategy that could be applied to the
synthesis of other prominent polyether ionophore antibiotics.
In summary, we have disclosed a concise and efficient
route to the asymmetric construction of the western hemi-
sphere of salinomycin, which is based on highly stereo-
selective C-glycosidation and substrate-controlled aldol
Acknowledgment. Financial support from the Ministerio
de Ciencia y Tecnolog´ıa and Fondos FEDER (grant
BQU2002-01514) and from the Generalitat de Catalunya
(2001SGR00051 and 2005SGR00584), as well as a doctorate
studentship (Generalitat de Catalunya) to I.L., are acknowl-
edged.
(16) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405-4408.
(17) The mechanistic details of such isomerization are unclear, but
migration of the CdC bond promoted by the catalyst followed by
nonstereoselective hydrogenation is presumably the cause of this undesired
secondary reaction.
Supporting Information Available: Experimental pro-
cedures and physical and spectroscopic data, including copies
of 1H and 13C spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
(18) Evans, D. A.; Dart, M. J.; Duffy, J. L.; Rieger, D. L. J. Am. Chem.
Soc. 1995, 117, 9073-9074.
(19) Evans, D. A.; Yang, M. G.; Dart, M. J.; Duffy, J. L. Tetrahedron
Lett. 1996, 37, 1957-1960.
OL052900W
530
Org. Lett., Vol. 8, No. 3, 2006