Synthesis of cladobotryal, CJ16,169, and CJ16,170

Article abstract of DOI:10.1021/ol049130t

(Equation Presented) Condensation of hydroxypyridone 7 with ethyl pyruvate and p-chlorothiophenol, reduction, and cyclization afforded 77% of lactone 6. Protection of the pyridone, acylation of the enolate with tigloyl chloride, and deprotection provided keto lactone 22. Reduction and dehydrative cyclization completed short and efficient syntheses of 2 and 3.

Full text of DOI:10.1021/ol049130t

ORGANIC
LETTERS
2004
Vol. 6, No. 17
2877-2880
Synthesis of Cladobotryal, CJ16,169,
and CJ16,170
Barry B. Snider* and Qinglin Che
Department of Chemistry MS 015, Brandeis UniVersity,
Waltham, Massachusetts 02454-9110
snider@brandeis.edu
Received May 12, 2004
ABSTRACT
Condensation of hydroxypyridone 7 with ethyl pyruvate and p-chlorothiophenol, reduction, and cyclization afforded 77% of lactone 6. Protection
of the pyridone, acylation of the enolate with tigloyl chloride, and deprotection provided keto lactone 22. Reduction and dehydrative cyclization
completed short and efficient syntheses of 2 and 3.
The antifungal furopyridone cladobotryal (1) was first
isolated from the mycoparasitic fungus Cladobotrium Varium
Nees:Fries (CBS 331.95).1 More recently, cladobotryal (1)
was isolated with seven related furopyridones from Clado-
botrium Varium CL12284 and was shown to possess moder-
ate activity against various Gram-positive bacteria, including
some drug resistant strains.2 The seven congeners include
the analogous primary alcohol CJ16,169 (2) and CJ16,170
(3), which differs from 2 in both the relative stereochemistry
of the two side chains and the formation of the dihydrofuran
with the 4- rather than 2-pyridone oxygen.
reaction of 6 with tiglaldehyde. Finally, we thought that
lactone 6 should be accessible from 5-phenyl-4-hydroxy-2-
pyridone (7). Although it was not clear how this last
transformation would be accomplished, this approach was
attractive because 7 can be prepared in two steps in 74%
overall yield by condensation of phenylacetonitrile and
malonyl chloride.5
Clive and Hang recently reported stereospecific, but
lengthy, syntheses of epi-CJ16,170 (26)3 and cladobotryal
(1)4 by routes in which the pyridone ring was constructed at
the end of the synthesis. Our retrosynthetic analysis suggested
that alcohols 2 and 3 should be available by dehydration of
diol 4, which should be available by reduction of lactone 5
(see Scheme 1). Lactone 5 should be obtainable from an aldol
We have previously used pyridone 7 to construct leporins
A and B.5,6 Knoevenagel condensation of 7 with dienal 8
afforded quinone methide 9, which underwent an intramo-
lecular inverse electron demand Diels-Alder reaction to give
35% of tricycle 10 (see Scheme 2). Oxidation and methy-
lation afforded leporins B (11) and A (12). We have used
(1) Breinholt, J.; Jensen, H. C.; Kjær, A.; Olsen, C. E.; Rassing, B. R.;
Rosendahl, C. N.; Søtofte, I. Acta Chem. Scand. 1998, 52, 631-634.
(2) Sakemi, S.; Bordner, J.; DeCosta, D. L.; Dekker, K. A.; Hirai, H.;
Inagaki, T.; Kim, Y.-J.; Kojima, N.; Sims, J. C.; Sugie, Y.; Sugiura, A.;
Sutcliffe, J. A.; Tachikawa, K.; Truesdell, S. J.; Wong, J. W.; Yoshikawa,
N.; Kojima, Y. J. Antibiot. 2002, 55, 6-18.
(3) Clive, D. L. J.; Huang, X. Tetrahedron 2002, 58, 10243-10250.
(4) (a) Clive, D. L. J.; Huang, X. Chem. Commun. 2003, 2062-2063.
(b) Clive, D. L. J.; Huang, X. J. Org. Chem. 2004, 69, 1872-1880.
(5) Snider, B. B.; Lu, Q. J. Org. Chem. 1996, 61, 2839-2844.
(6) Synthesis of leporin B was completed in 1996,5 although the
compound was not isolated as a natural product and named until 2003:
Zhang, C.; Jin, L.; Mondie, B.; Mitchell, S. S.; Castelhano, A. L.; Cai, W.;
Bergenhem, N. Bioorg. Med. Chem. Lett. 2003, 13, 1433-1435.
10.1021/ol049130t CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/20/2004

Scheme 1. Retrosynthetic Analysis of Cladobotryal
Scheme 3. Synthesis of Lactone 6
similar strategies for the synthesis of pyridone alkaloids
pyridoxatin and fusaricide.7 Although 7 is a suitable starting
material for both the leporins and 1-3, the presence of the
quaternary center in 1-3 required the use of the very
different strategy shown in Scheme 1.
Our experience with Knoevenagel condensations of 7
suggested that the three carbons needed for lactone 6 could
be introduced by reaction of 7 with ethyl pyruvate, which
should give quinone methide 13, which should undergo a
1,5-hydride shift to give acrylate 14. Neither 13 nor 14 will
be stable to the Knoevenagel condensation conditions and
must be trapped by the addition of a nucleophile, such as a
thiol, which is known to trap quinone methides formed from
condensations with aldehydes.8
from the reaction. Reduction of 16 with Raney Ni8 in
acetone/EtOH for 12 h at 25 °C provided 95% of 17, which
was thus obtained in 83% overall yield from pyridone 7.9,10
Lactonization to give 6 was achieved in 93% by heating 17
in 50:1 toluene/TFA at reflux for 12 h.11
Protection of the pyridone nitrogen of 6 was required
before the formation of the lactone enolate and addition of
the five-carbon side chain. Reaction of 6 with excess
TBSOTf and Et3N in CH2Cl2 afforded the unstable bis-
(silyloxy) furopyridine 18 (see Scheme 4). Chromatography
Scheme 2. Synthesis of Leporin A and B
Scheme 4. Protection of Lactone 6
Condensation of 7 with ethyl pyruvate (4 equiv) and
p-ClPhSH in EtOH containing Et3N and HOAc in a sealed
tube in a 110 °C oil bath for 3 days afforded 72% of propio-
nate 17 and 12% of 3-arylthiopropionate 16 (see Scheme
3). The latter is formed by conjugate addition of p-ClPhSH
to acrylate 14. Propionate 17 is probably formed by addition
of p-ClPhSH to quinone methide 13 to give 2-arylthiopro-
pionate 15, which was not observed as a reaction product.
Presumably, reduction of 15 with excess p-chlorothiophenol
led to major product 17, and (p-ClPhS)2, which was isolated
on silica gel cleaved the pyridine silyloxy group to give 81%
of furan 19. Stirring crude 18 with aqueous Et3N cleaved
the furan silyloxy group to provide 91% of the desired
protected lactone 20.
(9) Condensation of 7 and ethyl pyruvate can also be carried out using
formic acid as a reductive trapping agent.10 Heating pyridone 7 and ethyl
pyruvate (2 equiv) in (Et3N)2(HCO2H)5 in a sealed tube at 140-150 °C for
5 h gave 40% of the propionic acid, which was esterified to give 17 in
93% yield with H2SO4 in EtOH at 60 °C.
(10) (a) Sekiya, M.; Yanihara, C. Chem. Pharm. Bull. 1967, 17, 747-
751. (b) To´th, G.; Molna´r, S.; Tama´s, T.; Borbe´ly, I. Org. Prep. Proced.
Int. 1999, 31, 222-225.
(7) (a) Snider, B. B.; Lu, Q. J. Org. Chem. 1994, 59, 8065-8070. (b)
Snider, B. B.; Smith, R. B. Synth. Commun. 2001, 31, 111-123.
(8) (a) Moreno-Manas, M.; Pleixats, R. Synthesis 1984, 430-431. (b)
Fuchs, K.; Paquette, L. J. Org. Chem. 1994, 59, 528-532. (c) Tabakovic,
I.; Tabakovic, K.; Gaon, I. Org. Prep. Proced. Int. 1997, 29, 223-226.
2878
Org. Lett., Vol. 6, No. 17, 2004

Scheme 5. Acylation to Form 22
Scheme 6. Reduction to Give Diol 4
Although, the overall yield of the four products from 22 is
75%, the selectivity for 2 is not good. We thought that the
biosynthesis of 2 and 3 might involve a similar cyclization
in water. We were pleased to observe that cyclization of
crude 4 in 4:1 H2O/TFA for 12 h at 25 °C provided 24% of
2, 13% of 3, 30% of 25, and 8% of 26. The yield of 2
increased from 15% in CH2Cl2/TFA to 24% in H2O/TFA.
Oxidation of 2 with Dess-Martin periodinane as previously
described4 gave cladobotryal (1) in 95% yield.12,13
Treatment of lactone 20 with n-BuLi in THF at -78 °C
followed by addition of tiglaldehyde failed to give 23 (see
Scheme 5). The formation of enolate 21 under these
conditions was established by trapping in high yield with
MeI or D2O. We hypothesized that the aldol reaction to give
23 might be thermodynamically unfavorable. Furan enolate
21 is aromatic, the conjugation of tiglaldehyde is lost on the
aldol reaction, and adduct 23 is sterically congested. We
therefore investigated the irreversible Claisen condensation
of tigloyl chloride with enolate 21, which proceeded in high
yield. Deprotection of the crude product by stirring in 10:1
CH2Cl2/HOAc cleaved the silyl ether to give keto lactone
22 in 78% yield from 20.
Scheme 7. Cyclization of 4
We now turned to the reduction of keto lactone 22 to form
diol 4. Reduction of 22 with NaBH4 and CeCl3 in MeOH
gave the methyl ester corresponding to 17. Presumably,
reduction of the ketone provided secondary alcohol 5, which
underwent a retro aldol reaction to give lactone 6, which
opened to give the methyl ester. This supports our analysis
that aldol adduct 23 is thermodynamically disfavored.
Concomitant reduction of both the lactone and ketone should
prevent the retro aldol reaction. Reduction of 22 with LAH
in THF afforded 80% of a 1:1 mixture of the desired diol 4
and alcohol 24, which was formed by reduction of the retro
aldol product, lactone 6 (see Scheme 6). We then investigated
the use of DIBAL-H, which reduces lactones rapidly to
lactols, which might further reduce the extent of retro aldol
reaction. We were pleased to find that treatment of 22 with
DIBAL-H for 3 h at -78 °C, followed by addition of LAH
and stirring for 18 h at 25 °C to complete the reduction,
yielded 75% of 4 as an unstable 1:1 mixture of diasteromers
and only 15% of the undesired alcohol 24.
To understand the details of the cyclization, we needed to
separate the diastereomers of 4, establish their stereochem-
istry, and cyclize the individual isomers. Flash chromatog-
raphy of crude 4 afforded 20% of 70% pure 4a and 22% of
80% pure 4b. Recrystallization (CHCl3) gave pure 4a in 12%
yield from 22, while reverse-phase chromatography gave
pure 4b in 16% yield from 22. Reaction of 4a with TsOH
and 2,2-dimethoxypropane for 24 h afforded 35% of 28,
whose structure was established by the NOEs shown in
Scheme 8.14 A similar reaction of 4b provided 42% of 30,
whose structure was established by the NOEs shown in
Scheme 8.14 The low yields of 4a, 4b, 28, and 30 result from
partial cyclization during chromatography and acetal forma-
tion.
Diol 4 cyclized partially on flash chromatography to give
a complex mixture of 2, 3, 25, 26, and other products.
Therefore, we cyclized the crude mixture by stirring in 4:1
CH2Cl2/TFA for 3 h at 25 °C to provide 15% of CJ16,169
(2), 19% of CJ16,170 (3), 34% of epi-CJ16,169 (25), and
7% of epi-CJ16,170 (26) (see Scheme 7). The spectral data
of 2,2,4 3,2,4 and 263 are identical to those previously reported.
(12) As previously observed,2,4 the NMR spectra of 1, 2, and 25 are
broadened, presumably due to slow prototropic tautomerism on the NMR
time scale. There is less broadening in CD3OD.13
(13) Elguero, J.; Katritzky, A. R.; Denisko, O. V. AdV. Heterocycl. Chem.
2000, 76, 1-84.
(11) (a) Faber, K.; Stu¨ckler, H.; Kappe, T. J. Heterocycl. Chem. 1984,
21, 1177-1181. (b) Schley, D.; Radspieler, A.; Christoph, G.; Liebscher,
J. Eur. J. Org. Chem. 2002, 369-374.
Org. Lett., Vol. 6, No. 17, 2004
2879

Scheme 8. Characterization and Cyclization of 4a and 4b
Table 1. Cyclization Product Ratios from 4, 4a, and 4b
diol
temp
solventa
% 2
% 3
% 25
% 26
4b
25 °C
25 °C
25 °C
25 °C
0 °C
0 °C
25 °C
25 °C
CH2Cl2
H2O
20
32
10
32
5
29
7
53
25
17
22
17
21
18
13
12
46
40
63
35
71
39
76
16
9
11
5
16
3
14
4
19
4b
4a c
4bc
4a c
4bc
4a c
4bc
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
H2O
H2O
a Mixture (4:1) of indicated solvent and TFA. b Mixture (1:1) of crude
4 was used. 75% Overall isolated yield of 2, 3, 25, and 26 from 22. c Ratio
1
of products determined by H NMR in CD3OD containing residual TFA.
between the axial alkenyl side chain and the pyridone.
Cyclization therefore proceeded by an SN1 reaction with an
allylic cation intermediate.
Treatment of primary alcohol 24 in 4:1 CH2Cl2/TFA for
24 h at 25 °C provided only trifluoroacetate ester 31,
indicating that the protonated primary alcohol does not
cyclize by an SN2 displacement under these conditions.
Heating 31 with KI in DMF for 3 h at 70 °C afforded 32 in
88% yield from 24 (eq 1). Presumably, the primary iodide
formed from the trifluoroacetate and underwent an SN2
cyclization.
The ratio of products formed in the cyclizations of 4, 4a,
and 4b in CH2Cl2/TFA at 0 and 25 °C and H2O/TFA at
25 °C are shown in Table 1. All four products are stable
when resubjected to the cyclization conditions, indicating
that the product ratios are kinetically controlled. Cyclization
of 4a preferentially formed 25 (63-76%) with inversion,
possibly by an SN2-like displacement through protonated
intermediate 27. CJ16,170 (3) was formed in 13-22% yield
by displacement with inversion from the other pyridone
oxygen. Cyclization of 4b was less selective but did give
53% of CJ16,169 (2) in water by displacement with
inversion. The SN2-like reaction through protonated inter-
mediate 29 is less favorable because of steric repulsion
In conclusion, we have completed short, efficient syntheses
of cladobotryal (1) and CJ16,169 (2) and the first synthesis
of CJ16,170 (3). Condensation of pyridone 7 with ethyl
pyruvate and p-chlorothiophenol, reduction, and cyclization
afforded 77% of lactone 6. Protection of the pyridone,
acylation of the enolate with tigloyl chloride, and deprotec-
tion provided keto lactone 22. Reduction and dehydrative
cyclization completed the syntheses.
Acknowledgment. We thank the National Institutes of
Health (GM-50151) for financial support.
Supporting Information Available: Full experimental
1
details and copies of H and 13C NMR spectral data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(14) For similar analyses, see: (a) Crump, R. A. N. C.; Fleming, L.;
Hill, J. H. M.; Parker, D.; Reddy, N. L.; Waterson, D. J. Chem. Soc., Perkin
Trans. 1 1992, 24, 3277-3294. (b) Ziegler, F. E.; Zheng, Z.-L. J. Org.
Chem. 1990, 55, 1416-1418.
OL049130T
2880
Org. Lett., Vol. 6, No. 17, 2004