Aiming for branimycin: Synthesis of the cis-decalin core

Article abstract of DOI:10.1055/s-2005-872222

Starting from quinic acid, a highly substituted, cis-α,β unsaturated nitrile oxide was synthesiszed and involved in a 1,3-dipolar cycloaddition to afford a precursor of the cis-decalin system of branimycin. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-872222

LETTER
2227
Aiming for Branimycin: Synthesis of the cis-Decalin Core
A
iming for Branim
a
ycin lentin S. Enev,* Martina Drescher, Hanspeter Kählig, Johann Mulzer*
Institute of Organic Chemistry, University of Vienna, Währingerstrasse 38, 1090 Vienna, Austria
Fax +43(1)427752189; E-mail: valentin.enev@univie.ac.at; E-mail: johann.mulzer@univie.ac.at
Received 9 May 2005
Abstract: Starting from quinic acid, a highly substituted, cis-a,b
unsaturated nitrile oxide was synthesiszed and involved in a 1,3-di-
polar cycloaddition to afford a precursor of the cis-decalin system
of branimycin.
Key words: branimycin, 1,3-dipolar cycloaddition, cis-a,b-unsat-
urated oximes, quinic acid
The increasing resistance of bacterial pathogens against
standard antibiotics combined with an emerging threat of
bioterrorism has led to the urgent need for developing in-
novative anti-infective drugs.1 A novel family of naturally
occurring compounds that show activity against patho-
genic microbes has been isolated in the early 1980s by
workers at Pfizer and Upjohn: the nargenicins and nodus-
mycin.2 Some years later the related compound colorado-
cin was described.3 Recently, branimycin, an unusual
member of this family has been isolated and structurally
characterized by the Laatsch group in Göttingen.4 First bi-
ological tests have shown that branimycin is highly active
against Streptomyces viridochromogenes. The structure
of branimycin, including the unusual configuration at
C17, has been reliably elucidated by multidimensional 1H
NMR and 13C NMR experiments.4 These antibiotics are of
potential interest as they combine low toxicity with a
broad spectrum of high antibacterial activities combined
with considerable oral availability. Additionally, the com-
plex highly oxygenated cis-fused decalin core, the 1,4-ox-
ygen bridge and the macrolide ring present considerable
challenge for total synthesis.
Our retrosynthetic strategy for the synthesis of branimy-
cin, illustrated in Scheme 1, suggests the disconnection
into the highly substituted cis-decalin moiety A and vinyl-
lithium subunit B5 – which will be coupled at a late stage
via 1,2-addition under concomitant epoxide opening and
formation of the C2–C7-oxygen bridge. The C3-append-
age could then be installed via a Claisen rearrangement of
the C3-allylic alcohol. In this letter a model study is re-
ported which aims for the construction of a highly oxy-
Scheme 1
genated enantiomerically and diastereomerically pure cis-
decalin system.
Quinic acid seemed a reasonable starting material for the
synthesis of the core fragment A, expecting that the exist-
ing chiral centers could control the construction of the re-
maining stereogenic centers. The key step in our approach
is the intramolecular nitrile oxide olefin cycloaddition
(INOC) of intermediate A3 to afford oxazolidine A2,
which in turn should be cleaved to b-hydroxy ketone A1.
The C7–C8 double bond in A1 was to be formed via an
E2-elimination from an 8-OMs derivative.
SYNLETT 2005, No. 14, pp 2227–2229
0
2
.0
9
.2
0
0
5
Advanced online publication: 20.07.2005
DOI: 10.1055/s-2005-872222; Art ID: D12505ST
© Georg Thieme Verlag Stuttgart · New York

2228
V. S. Enev et al.
LETTER
As the most challenging step in this sequence we recog- protection of the secondary alcohol as a TBDPS ether and
nized the synthesis of cis-a,b-unsaturated nitrile oxide A3 deprotection of the primary OH group afforded compound
from aldehyde A4.
16 in 54% yield over 3 steps. Finally, Dess–Martin oxida-
tion of alcohol 16 gave a,b-unsaturated aldehyde 17. After
intensive experimentation we found that in presence of
molecular sieves (4 Å) and three equivalents of polymer-
supported di-tert-butyl pyridine as a base, aldehyde 17
reacts with hydroxylamine hydrochloride to isomerically
pure cis-a,b-unsaturated oxime 18 in 80% yield
(Scheme 3).
We started from the known ketone 1,6 which after reduc-
tion and MOM protection gave ester 2 in 89% yield. Re-
duction of 2 and methylation of the resulting alcohol
afforded 4 in 87% yield. Hydroboration of 4 followed by
oxidative work up,7 furnished alcohol 5 as a single isomer
in excellent yield, which was subsequently protected as
TBS ether 6. The cyclohexylidene ketal 6 was trans-
formed to the olefin 9, according to the Corey–Hopkins To the best of our knowledge this is the first example for
protocol.8 Compound 9 was then deprotected to give the a stereoselective synthesis of a cis a,b-unsaturated oxime.
allylic alcohol 10 (Scheme 2). Wittig–Still rearrangement
of 10, followed by Dess–Martin oxidation produced alde-
hyde 13 as a single isomer in good overall yield. Treat-
O
SnBu3
ii
OH
ment of aldehyde 13 with the lithium salt of TBS
protected propargyl alcohol afforded 14 as a 1:3 mixture
of a- and b-alcohols which were easily separated to give
the desired 14b epimer. The absolute stereochemistry of
14b was assigned by NMR spectra of the corresponding
Mosher esters. Reduction of 14b with Lindlar catalyst,
MeOH2C
MeOH2C
MeOH2C
MeOH2C
i
OMOM
OMOM
10
11
MeOOC
MeOOC
O
O
O
O
MeOH2C
ii
i
iv
iii
O
OH
OMOM
O
OMOM
OMOM
13
1
2
12
OTBS
v
OR
OTBS
ROH2C
MeOH2C
O
O
O
MeOH2C
iv
vi
vi
H
H
O
OMOM
OMOM
OH
MOMO
14
OH
MOMO
5 R = H
6 R = TBS
α:β - 1:3
15
3 R = H
v
iii
4 R = Me
O
OH
H
MeOH2C
MeOH2C
viii
vii
OTBS
OTBS
H
MeOH2C
MeOH2C
O
O
viii
OH
vii
S
OTBDPS
17
OTBDPS
MOMO
MOMO
OH
OMOM
OMOM
16
OH
N
7
8
O
N
H
MeOH2C
MeOH2C
OTBS
OH
ix
H
MeOH2C
MeOH2C
ix
H
OTBDPS
OTBDPS
MOMO
MOMO
OMOM
OMOM
18
19
9
10
Scheme 3 Reagents and conditions: i) Bu3SnCH2I, KH, THF, r.t.,
90%; ii) BuLi, –78 °C to –40 °C, 62%; iii) Dess–Martin oxidation,
90%, iv) Li-CCCH2OTBS, THF, –78 °C, 60%; v) H2/Lindlar cat.
62%; vi) TBDPSCl, imidazole, THF, then TsOH, 90%; vii) Dess–
Martin oxidation; viii) NH2OH·HCl, di-tert-butyl pyridine polymer
bound, 4 Å MS, THF, 40 °C, 12 h, 80%; ix) t-BuOCl, CH2Cl2,
–40 °C, 2 h then pyridine, 12 h, 64%.
Scheme 2 Reagents and conditions: i) NaBH4–MeOH, then MOM-
Cl, Hünig’s base 89%; ii) DIBAL, –78 °C, 92%; iii) MeI, NaH, THF–
DMF, 87%; iv) BH3·SMe2, –20 °C then NaBO3·4H2O, THF, 92%; v)
TBSCl, imidazole, DMF, 90%; vi) TsOH, MeOH, r.t., 90%; vii)
Im2C=S, Tol, reflux, 5 h, 82%; viii) 1,3-dimethyl-2-phenyl-1,3,2-di-
azaphospholidene, 90%; ix) TBAF, THF, r.t., 92%.
Synlett 2005, No. 14, 2227–2229 © Thieme Stuttgart · New York

LETTER
Aiming for Branimycin
2229
(7) Kabalka, G.; Shoup, T.; Goudgaon, M. J. Org. Chem. 1989,
54, 5930.
(8) Corey, E. J.; Hopkins, P. B. Tetrahedron Lett. 1982, 23,
1979.
(9) Stevens, R. V.; Beaulieu, N.; Chan, W.; Daniewski, A. R.;
Takeda, T.; Waldner, A.; Williard, P. G.; Zutter, U. J. Am.
Chem. Soc. 1986, 108, 1039.
Finally, treatment of compound 18 with one equivalent of
t-BuOCl at –40 °C9 followed by pyridine at room temper-
ature afforded the desired cis-isoxazoline 19 in a 64%
yield.10 The stereochemistry of 19 was confirmed by
NOESY experiments.11
In conclusion, we have demonstrated that quinic acid
could be a suitable starting material toward the cis-decalin
core of branimycin. Further investigations on the synthe-
sis of branimycin are currently under way in our laborato-
ries.
(10) Selected Data.
Compound 18: IR (neat): 3344, 2947, 1613, 1514, 1463,
1376, 1249 cm–1. 1H NMR (600 MHz): d = 7.62 (2 H, d,
J = 7.7 Hz), 7.59 (2 H, d, J = 8Hz), 7.53 (1 H, dd, J = 10.5,
0.9 Hz) 7.46–7.32 (6 H, m), 7.27 (1 H, br s), 6.04 (1 H, ddd,
J = 11.3, 9.1, 0.9 Hz), 5.79 (1 H, ddd, J = 11.4, 10.6, 1,2),
5.77 (2 H, m), 4.88 (1 H, ddd, J = 9.2, 2.1, 1.2 Hz), 4.41 (1
H, d, J = 7.0 Hz), 4.30 (1 H, d, J = 7Hz), 3.69 (1 H, ddd,
J = 11.6, 8.7, 3.7 Hz), 3.36 (3 H, s), 3.31 (1 H, dd, J = 8.8,
6.4 Hz), 3.28 (1 H, dd, J = 8.9, 6.6 Hz), 3.24 (3 H, s), 2.50 (1
H, m), 2.21 (1 H, m), 2.15 (1 H, m), 1.32 (1 H, ddd, J = 12.3,
11.7, 11.7 Hz), 1.01 (9 H, s). 13C NMR (150 MHz): d = 147.3
(CH), 139.8 (CH), 136.0 (CH), 135.9 (CH), 133.9 (C), 133.3
(C), 129.2 (CH), 129.7 (CH), 129.6 (CH), 127.6 (CH), 127.5
(CH), 125.6 (CH), 121.0 (CH), 96.4 (CH2), 76.8 (CH2), 75.8
(CH), 68.6 (CH), 58.9 (CH3), 55.4 (CH3), 51.3 (CH), 36.4
(CH), 32.8 (CH2), 26.9 (CH3), 19.5 (C). HRMS: m/z calcd
for C30H41NO5Si: 523.2754; found: 523.2778.
Acknowledgment
We thank S. Schneider for performing the HPLC separation and the
Austrian Science Fund (FWF Project P15929) for financial support.
References
(1) (a) Antibiotic Resistance: Origin, Evolution, Selection and
Spread; Chadwick, D. J.; Goode, J., Eds.; Wiley: Chichester,
1997. (b) Levy, S. B. Sci. Ann. 1998, 278, 32; and cited
literature.
(2) Review: (a) Kallmerten, J. Studies in Natural Products
Chemistry, Vol. 17; Atta-ur-Rahman, Ed.; Elsevier:
Amsterdam, 1995, 283–310; and cited literature.
(b) Nargenicin: Celmer, W. D.; Chmurny, C. N.; Moppett, C.
E.; Ware, R. S.; Watts, P. C.; Whipple, E. B. J. Am. Chem.
Soc. 1980, 102, 4203. (c) See also: Tone, J.; Shibakawa, R.;
Maeda, H.; Yamauchi, Y.; Niki, K.; Saito, M.; Tsukuda, K.;
Whipple, E. B.; Watts, P. C.; Moppett, C. E.; Jefferson, M.
T.; Huang, L. H.; Cullen, W. P.; Celmer, W. D. 20th
Interscience Conf. Antimicr. Agents and Chemoth. New
Orleans, LA, Abstract 62, Sep. 22-24, 1980.
Compound 19: IR (neat): 1922, 1560, 1456, 1377, 1197,
1122, 1072 cm–1. 1H NMR (600 MHz): d = 7.75 (2 H, d,
J = 6.8 Hz), 7.69 (2 H, d, J = 8.0 Hz), 7.50–7.30 (6 H, m),
6.30 (1 H, dd, J = 10.1, 2.8 Hz), 6.10 (1 H, ddd, J = 10.1, 1.5,
1.5 Hz), 4.83 (1 H, d, J = 6.9 Hz), 4.62 (1 H, d, J = 6.9 Hz),
4.60 (1 H, m), 4.41 (1 H, dd, J = 8.7, 8.7 Hz), 3.64 (1 H, ddd,
J = 10.4, 10.4, 3.6 Hz), 3.47 (1 H, dd, J = 9.4, 4.7 Hz), 3.37
(1 H, dd, J = 9.4, 3.6 Hz), 3.10 (1 H, dd, J = 8.2, 5.8 Hz),
3.40 (3 H, s), 3.34 (3 H, s), 2.51 (1 H, m), 1.91 (1 H, ddd,
J = 13.4, 3.7, 3.7 Hz), 1.66 (1 H, m), 1.35 (1 H, ddd,
J = 13.4, 13.4, 10.9 Hz), 1.09 (9 H, s). 13C NMR (150 MHz):
d = 155.4 (C), 143.4 (CH), 136.0 (CH), 135,9 (CH), 133.1
(C), 127.8 (CH), 127.6 (CH), 129.9 (CH), 120.0 (CH), 117.0
(CH), 97.6 (CH2), 78.2 (CH), 73.5 (CH), 73.4 (CH2), 73.1
(CH), 59.2 (CH3), 55.7 (CH3), 48.6 (CH), 44.2 (CH), 38.1
(CH), 33.0 (CH2), 27.1 (CH3), 19.1 (C). HRMS: m/z calcd
for C30H39NO5Si: 523.2598; found: 521.2584.
(d) Nodusmycin: Whaley, H. A.; Chidester, C. G.; Mizak, S.
A.; Wnuk, R. J. Tetrahedron Lett. 1980, 3659. (e) See
further: Whaley, H. A.; Coates, J. H. 21st Interscience
Conference on Antimicrobial Agents and
Chemotherapeutics, Chicago, IL, Abstract 187, Nov. 4-6,
1981.
(3) (a) Jackson, M.; Karwowski, J. P.; Theriault, R. J.;
Fernandes, P. B.; Semio, R. C.; Kohl, W. H. J. Antibiot.
1987, 40, 1375. (b) Rasmussen, R. R.; Scherr, M. H.;
Whittern, D. N.; Buko, A. M.; McAlpine, J. B. J. Antibiot.
1987, 40, 1383.
(4) Speitling, M.; Grün-Wollny, I.; Hannske, F. G.; Laatsch, H.
12 and 13 IRSEER Naturstofftage der DECHEMA e.V. Irsee
2000, 2001, poster sessions.
(5) Rosenbeiger, D. Diploma Thesis; Universität Wien, 2004;
International ASCMC Symposium, Abstract P160, Moscow,
May, 2004.
(6) Ketone 6 is easily available in a five-step sequence from
quinic acid see: Ulibarri, G.; Nadler, W.; Skrydstrup, T.;
Audrain, H.; Chiaroni, A.; Riche, C.; Grierson, D. J. Org.
Chem. 1995, 60, 2753.
(11) Selected NOE data for 19 is shown in Figure 1.
O
N
H
H
H
MeOH2C
H
H
H
MOMO
OTBDPS
Figure 1
Synlett 2005, No. 14, 2227–2229 © Thieme Stuttgart · New York