Catalytic asymmetric synthesis of a potent thiomarinol antibiotic

Article abstract of DOI:10.1021/ja042827p

Thiomarinols, isolated from the bacterium Alteromonas rava sp. nov. SANK 73390, are potent marine antibiotics that possess both Gram-positive and Gram-negative activity. The first total synthesis of a member of the thiomarinol class of marine antibiotics was achieved in a remarkable global yield of 22% (from 3-boronoacrolein pinacolate). The highlight of this synthesis is the efficient catalytic enantio-, regio-, E/Z-, and diastereoselective three-component inverse electron demand Diels-Alder/allylboration sequence. This key operation provides a rare example of an enantioselective HDA reaction involving acyclic 2-substituted enol ethers, and it featured an unusual but fortuitous kinetic selection that favored the requisite Z-dienophile. Copyright

Full text of DOI:10.1021/ja042827p

Published on Web 01/25/2005
Catalytic Asymmetric Synthesis of a Potent Thiomarinol Antibiotic
Xuri Gao and Dennis G. Hall*
Department of Chemistry, UniVersity of Alberta, Edmonton, AB, T6G 2G2, Canada
Received November 29, 2004; E-mail: dennis.hall@ualberta.ca
Scheme 1
The isolation and characterization of the pyran-containing
pseudomonic acids have led to the commercial success of Bac-
troban, a topical antimicrobial medication for skin infections.1 The
active ingredient, pseudomonic acid A (mupirocin, 1, Figure 1),
displays poor oral absorptivity and low metabolic stability. Con-
sequently, there has been significant interest in the development
of improved, less toxic analogues that could also be suitable as
bloodstream antibiotics. Thiomarinol A (2)2 and derivative 33 are
rare marine natural products recently isolated from the bacterium
Alteromonas raVa sp. nov. SANK 73390.
ethyl vinyl ether.6 It is known, however, that 2-substituted enol
ethers are difficult substrates in inverse electron demand Diels-
Alder (IEDDA) cycloadditions because of the added steric effect
of the 2-substituent.11 Our work with 4 highlighted its exceptional
reactivity as a heterodiene,6 which we hoped could overcome the
low reactivity of a 2-substituted enol ether as a dienophilic partner.
Thus, model studies examined the behavior of model Z-enol ether
5a12 in the IEDDA/allylboration process (eq 1). The IEDDA
Figure 1. Pseudomonic acid A (1) and thiomarinol derivatives 2 and 3.
The structures of 2 and 3 differ from 1 by the presence of a
C4-hydroxyl, a shorter C1-alkoxy chain, and the replacement of
the C10-C11 epoxide with an E alkene unit. Compounds 2 and 3
are distinguishable, respectively, by their holotin and anhydro-
ornithine C1 amide end groups. These natural substances were found
to be much more potent than pseudomonic acid A and possess a
wider spectrum of activity (Gram-positive and Gram-negative
bacteria). For example, the activity of derivative 3 against S. aureus
was shown to be comparable to that of tetracycline and strepto-
mycin.3 These impressive properties justify the development of an
efficient synthetic route to 2 and 3, which is key to any future efforts
at generating simplified analogues with higher potency and oral
bioavailability. The five contiguous stereocenters of 2 and 3 (C4-
C8), in particular the C4 hydroxyl, represent a significant challenge.
Herein, we disclose the first total synthesis of a thiomarinol
antibiotic.4 Our concise route to 3 uses a stereoconvergent three-
component strategy amenable to the design of analogues with
attractive skeletal variations.
We envisioned that the pyran core and the C4-hydroxyl of 3
could be constructed stereoselectively with an endo-selective three-
component hetero[4 + 2] cycloaddition/allylboration approach5-7
from 3-boronoacrolein pinacolate 4, a suitably functionalized (Z)-
2-substituted enol ether 5, and aldehyde 8 (Scheme 1). From
intermediate 9, alkene dihydroxylation and acetal reduction would
provide the correct functionalization of the pyran ring, and
completion of the C8 chain would involve a trans olefination. The
requisite C12-C13 propionate fragment could be accessed via either
Brown crotylboration8 or our own catalytic system.9
reaction was indeed more difficult with a 2-substituted enol ether.
Whereas ethyl vinyl ether reacted with 4 in less than 2 h with 1
mol % of catalyst 6,6 5a required more than 5 h with 3 mol % of
6 (neat, 20 °C). Nonetheless, the all-cis endo adduct 7a was cleanly
formed as a single stereoisomer. To our surprise, it was the
allylboration step that proved problematic. Whereas the adduct of
ethyl vinyl ether (7 with R ) H) reacted with various aldehydes at
a temperature of 45 °C,6 an astonishing 110 °C was needed to
produce 9a from 7a and 8. As a result, the tandem process could
not be performed in “one-pot”, and we found it necessary to remove
catalyst 6 with a quick silica filtration to afford a clean allylboration
product. In the putative allylboration transition state leading to 9a
(eq 1), the ring assumes an unfavorable chairlike conformation with
both pseudoaxial boronate and ethoxy substituents.13 It is possible
that gauche interactions from the C8 chain (R) further increase the
barrier for conformational change. The C4-C5 stereochemistry can
be explained as before.6
We have previously demonstrated that Jacobsen’s chiral Cr(III)
complex 610 efficiently catalyzes the cycloaddition between 4 and
9
1628
J. AM. CHEM. SOC. 2005, 127, 1628-1629
10.1021/ja042827p CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
The synthesis of 3 was initiated with the sequential three-
Acknowledgment. This work was funded by the Natural
Sciences and Engineering Research Council (NSERC) of Canada,
an AstraZeneca Chemistry Award to D.G.H., and the University
of Alberta. X.G. thanks the University of Alberta for a Province of
Alberta Graduate Fellowship.
component coupling between 4,6 enol ether 5b,12 and commercial
aldehyde 8 (Scheme 2). Interestingly, it was difficult to obtain
isomerically pure enol ether Z-5b but luckily, the Z-isomer was
found to be more reactive than the E-isomer. Reactions with a
mixture of isomers only afforded product consistent with a
kinetically selective cycloaddition of the Z-isomer. It appears that
Z-5b may be more reactive simply as a result of steric control by
the huge catalyst 6.14 In any event, the desired pyran 9b was isolated
in 76% yield as a single diastereomer in >95% enantiomeric excess.
Remarkably, all three key stereocenters of thiomarinols: C4, C5,
and C8 are set in this process. After protection of the secondary
hydroxyl of 9b, a high-yield sequence for selective oxidative
cleavage of the exocyclic alkene furnished alcohol 10. Reduction
of the cyclic acetal with Et3SiH/TiCl4 was best planned at this stage.
The primary hydroxyl was then transformed into the tetrazolyl
sulfone 11 required for the olefination.15 Ring dihydroxylation was
carried out first, occurring selectively from the face opposite to
the C5/C8 substituents. The protected diol 12 was then subjected
to the Julia-Kocienski coupling15 with aldehyde 1312 to give
advanced intermediate 14 in good yield. More than 3 g of this
intermediate was obtained, which is testimony to the efficiency of
the chosen sequence. Hydrolysis of the C1 ester, re-esterification
with side chain alcohol 1512 to afford 16, and final removal of the
protecting groups afforded 3.16
Supporting Information Available: Experimental details and
spectral reproductions for all experiments. This material is available
free of charge via the Internet at http://pubs.acs.org.
References
(1) Class, Y. J.; DeShong, P. Chem. ReV. 1995, 95, 1843-1857.
(2) (a) Kodama, K.; Shiozawa, H.; Ishii, A. Sankyo Kenkyusho Nenpo 1993,
45, 131-136. (b) Shiozawa, H.; Kagasaki, T.; Kinoshita, T.; Haruyama,
H.; Domon, H.; Utsui, Y.; Kodama, K.; Takahashi, S. J. Antibiot. 1993,
46, 1834-1842. (c) Shiozawa, H.; Takahashi, S. J. Antibiot. 1994, 47,
851-853.
(3) Stierle, D. B.; Stierle, A. A. Experientia 1992, 48, 1165-1169.
(4) For synthetic studies on 2 and 3, see: Khan, N.; Xiao, H.; Zhang, B.;
Cheng, X.; Mootoo, D. R. Tetrahedron 1999, 55, 8303-8312.
(5) Deligny, M.; Carreaux, F.; Carboni, B.; Toupet, L.; Dujardin, G. Chem.
Commun. 2003, 276-277.
(6) Gao, X.; Hall, D. G. J. Am. Chem. Soc. 2003, 125, 9308-9309.
(7) Deligny, M.; Carreaux, F.; Toupet, L.; Carboni, B. AdV. Synth. Catal.
2003, 345, 1215-1219.
(8) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293-294.
(9) Lachance, H.; Lu, X.; Gravel, M.; Hall, D. G. J. Am. Chem. Soc. 2003,
125, 10160-10161.
(10) Gademann, K.; Chavez, D. E.; Jacobsen, E. N. Angew. Chem., Int. Ed.
2002, 41, 3059-3061.
(11) These reactions are further complicated by difficulties in making iso-
merically pure 2-substituted acyclic enol ethers. For examples of catalytic
enantioselective HDA reactions of cyclic enol ethers, see: (a) Evans, D.
A.; Johnson, J. S.; Olhava, E. J. J. Am. Chem. Soc. 2000, 122, 1635-
1649. (b) Audrain, H.; Thorauge, J.; Hazell, R. G.; Jørgensen, K. A. J.
Org. Chem. 2000, 65, 4487-4497.
In summary, we have achieved the first total synthesis of a
member of the thiomarinol class of marine antibiotics. Compound
3 was reached in a remarkable global yield of 22% (from 4). The
highlight of this synthesis is the efficient catalytic enantio-, regio-,
E/Z-, and diastereoselective three-component IEDDA/allylboration
sequence. This key operation provides a rare example of an
enantioselective HDA reaction involving acyclic 2-substituted enol
ethers. Moreover, this reaction featured an unusual but fortuitous
kinetic selection that favored the requisite Z-dienophile from a
mixture of isomers. It is also noteworthy that two key allylboration
reactions have been employed to set four of the eight stereogenic
centers of 3. The convergent synthetic strategy should facilitate
our future efforts at generating improved thiomarinol analogues.
(12) See Supporting Information for more details and references on the
preparation of this compound.
(13) Alternatively, a stereoelectronically viable boatlike conformation with a
pseudoequatorial ethoxy substituent may also be proposed.6
(14) Isomeric 2-substituted acyclic enol ethers were found to be equally reactive
with other substrates and catalysts: Wada, E.; Pei, W.; Yasuoka, H.; Chin,
U.; Kanemasa, S. Tetrahedron 1996, 52, 1205-1220. Our preliminary
experiments indicate a similar outcome with 4 and Yb(FOD)3. This issue
will be addressed in a future full account.
(15) Bellingham, R.; Jarowicki, K.; Kocienski, P.; Martin, V. Synthesis 1996,
285-296.
(16) The NMR spectroscopic data and the absolute configuration of synthetic
3 are in full agreement with the reported data for natural 33 (see Supporting
Information for details). Authentic samples of 3 were no longer available.
JA042827P
9
J. AM. CHEM. SOC. VOL. 127, NO. 6, 2005 1629