De novo asymmetric synthesis of cladospolide B-D: Structural reassignment of cladospolide D via the synthesis of its enantiomer

Article abstract of DOI:10.1021/ol9000119

The enantioselective synthesis of cladospolide B, C, and (ent)-cladospolide D has been achieved in 11-15 steps from 1-nonyne. The route relies upon an alkyne zipper reaction to relay an ynone and dienoate functional groups across a nine carbon fragment, which enables a highly enantioselective Noyori ynone reduction and a diastereo-and regioselective Sharpless dihydroxylation of a dienoate. In addition to being a flexible approach to three members of the cladospolide natural products, this route for the first time correctly established the structure for cladospolide D.

Full text of DOI:10.1021/ol9000119

ORGANIC
LETTERS
2009
Vol. 11, No. 5
1107-1110
De Novo Asymmetric Synthesis of
Cladospolide B-D: Structural
Reassignment of Cladospolide D via the
Synthesis of its Enantiomer
Yalan Xing and George A. O’Doherty*
Department of Chemistry, West Virginia UniVersity, Morgantown, West Virginia 26506
george.odoherty@mail.wVu.edu
Received January 4, 2009
ABSTRACT
The enantioselective synthesis of cladospolide B, C, and (ent)-cladospolide D has been achieved in 11-15 steps from 1-nonyne. The route relies
upon an alkyne zipper reaction to relay an ynone and dienoate functional groups across a nine carbon fragment, which enables a highly enantioselective
Noyori ynone reduction and a diastereo- and regioselective Sharpless dihydroxylation of a dienoate. In addition to being a flexible approach to three
members of the cladospolide natural products, this route for the first time correctly established the structure for cladospolide D.
The cladospolides (A-D) are a family of 12-membered macro-
lactones, isolated from various Cladosporium species, and
possess a range of biological activities (Figure 1).1 Although
fungal metabolites, the cladospolides appear to act as plant
pheromones.1b,c For instance, cladospolides A-C inhibit the
shoot elongation in rice seedlings.1c Interestingly, clado-
spolides A and B have the opposite effect on the root growth
of lettuce seedlings.1b More detailed mechanisms of action
studies have suggested that cladospolide C acts via the
inhibition of gibberellin biosynthesis.1c Of the cladospolides,
cladospolide D possesses the most interesting activity with
IC50 values of 0.1 and 29 µg/mL against Mucor racemosus
and Pyricularia oryzae, respectively.1d
cladospolides A-C have been prepared by Banwell from
1-hepten-2-ol and cis-2,3-dihydroxychlorobenzene which can
be prepared in optically pure form via enzymatic resolution and
biotransformation,2j,3 whereas others have chosen carbohy-
drate starting materials.2 In contrast, at the start of this inves-
tigation, the absolute and relative stereochemistry of clado-
spolide D has not been determined.1d,4 In addition, the stereo-
chemical assignment of the C2/C3 double bond was in doubt
being solely based on a 13.5 Hz vicinal coupling constant
(2) For the synthesis of cladospolide A, see: (a) Banwell, M. G.; Loong,
D. T. J. Org. Biomol. Chem. 2004, 2, 2050. (b) Banwell, M. G.; Jolliffe,
K. A.; Loong, D. T. J.; McRae, K. J.; Vounatsos, F. J. Chem. Soc., Perkin
Trans. 1 2002, 22. (c) Solladie, G.; Almario, A. Tetrahedron: Asymmetry
1995, 6, 559. (d) Solladie, G.; Almario, A. Pure Appl. Chem. 1994, 66,
2159. (e) Ichimoto, I.; Sato, M.; Kirihata, M.; Ueda, H. Chem. Express 1987,
2, 495. (f) Maemoto, S.; Mori, K. Chem. Lett. 1987, 109. (g) Mori, K.;
Maemoto, S. Liebigs Ann. Chem. 1987, 863. For the synthesis of clado-
spolide B, see: (h) Pandey, S. K.; Kumar, P. Tetrahedron Lett. 2005, 46,
6625. (i) Austin, K. A. B.; Banwell, M. G.; Loong, D. T. J.; Rae, A. D.;
Willis, A. C. Org. Biomol. Chem. 2005, 3, 1081. For the synthesis of
cladospolide C: (j) Banwell, M. G.; Loong, D. T. J.; Willis, A. C. Aust.
J. Chem. 2005, 58, 511. (k) Chou, C.-Y.; Hou, D.-R. J. Org. Chem. 2006,
71, 9887. (l) Sharma, G. V. M.; Reddy, K. L.; Reddy, J. J. Tetrahedron
Lett. 2006, 47, 6537.
The absolute and relative stereochemistry for the clado-
spolides A, B, and C have been confirmed by several total
syntheses from known chiral starting materials.2 For instance,
(1) (a) Hirota, A.; Isogai, A.; Sakai, H. Agric. Biol. Chem. 1981, 45,
799. (b) Hirota, A.; Sakai, H.; Isogai, A. Agric. Biol. Chem. 1985, 49, 731.
(c) Fujii, Y.; Fukuda, A.; Hamasaki, T.; Ichimoto, I.; Nakajima, H. Phyto-
chemistry 1995, 40, 1443. (d) Zhang, H.; Tomoda, H.; Tabata, N.; Miura,
H.; Namikoshi, M.; Yamaguchi, Y.; Masuma, R.; Omura, S. J. Antibiot.
2001, 54, 635. (e) Hirota, A.; Sakai, H.; Isogai, A.; Kitano, Y.; Ashida, T.;
Hirota, H.; Takahashi, T. Agric. Biol. Chem. 1985, 49, 903. (f) Hirota, H.;
Hirota, A.; Sakai, H.; Isogai, A.; Takahashi, T. Bull. Chem. Soc. Jpn. 1985,
58, 2147. (g) Smith, C. J.; Abbanat, D.; Bernan, V. S.; Maiese, W. M.;
Greenstein, M.; Jompa, J.; Tahir, A.; Ireland, C. M. J. Nat. Prod. 2000, 63,
142.
(3) Hudlicky, T.; Gonzalez, D.; Gibson, D. T. Aldrichimica Acta 1999,
32, 35
.
(4) As this manuscript was being completed, a report of a synthesis and
stereochemical assignment for cladospolide D appeared, see: Lu, K.-J.; Chen,
C.-H.; Hou, D.-R. Tetrahedron 2009, 65, 225. This assignment unfortunately
must be considered erroneous, vide infra
.
10.1021/ol9000119 CCC: $40.75
Published on Web 02/04/2009
2009 American Chemical Society

Scheme 2. De Novo Synthesis of Dienoate 5
Figure 1. Cladospolides and Grahamimycin A.
for the purported trans-enoate carbon/carbon double
bond.1d
Drawn by the structural and biological similarities between
cladospolides and grahamimycin A(Figure 1),5 we became
interested in the de novo synthesis of the cladospolides.6 In
particular, we were interested in the synthesis and stereo-
chemical assignment of cladospolide D. Herein, we report
the de novo synthesis of cladospolides B-D. The approach
features the use of the alkyne zipper reaction to remotely
link an alkyne functional group across nine carbon atoms,
which enables a Noyori asymmetric ynone reduction and a
Sharpless regio- and stereoselective dienoate dihydroxylation.
As part of our efforts for the use of asymmetric catalysis for
the enantioselective synthesis of antimicrobial lactones, we were
interested in an enantioselective synthesis of cladospolide D.
Because of the ambiguity associated with the cladospolide D
structure, we initially targeted the reduced natural products
cladospolides B and C along with their C4/C5 bisepimers 1
and 2 (Scheme 1). Thus, a selective oxidation of the allylic
alcohol from one of these four diastereomers was likely to give
cladospolide D along with its double bond isomer. Retrosyn-
thetically, we envisioned that cladospolides B and C as well as
the bisepimers 1 and 2 could come from the diastereo- and
regioselective dihydroxylation7 and subsequent lactonization of
dienoate 5. In turn, dienoate 5 could come from the carboxy-
lation and diene conjugation of hydroxyalkyne 6, which could
be prepared from the Noyori asymmetric reduction/alkyne
zipper isomerization of ynone 7. Finally, ynone 7 could be
prepared from commercially available 1-nonyne 8.
To access synthetically useful quantities of dienoate 5, an
efficient seven-step approach was developed (Scheme 2). The
route featured an alkyne zipper reaction8 and the Ph3P
promoted ynoate to dienoate isomerization, developed by
Trost.9 Treatment of the lithium acetylide of 8 with acety-
laldehyde gave good yield (80%) of a racemic propargylic
alcohol (rac)-9, which when oxidized with MnO2 gave the
ynone 7. Exposure of the ynone 7 to our modified Noyori
conditions10 provided an excellent yield (85%) of propargyl
alcohol 9 with high enantiomeric purity (>96% ee). Exposure
of 9 to the KAPA reagent readily isomerized it to the terminal
undecynol 6 in good yield (85%) and with no loss of
enantiomeric purity (>96% ee).11 The secondary alcohol in
6 was cleanly protected as a TBS-ether (TBSCl/imid. in
DMF, 97%), and the terminal alkyne was carboxylated (n-
BuLi/ClCO2Et, 83%) to give ynoate 10. Exposure of
Scheme 1. Retrosynthetic Analysis of Cladospolides B-D
(5) Gurusiddaiah, S.; Ronald, R. C. Antimicrob. Agents Chemother. 1981,
19, 153–165.
(6) Hunter, T. J.; O’Doherty, G. A. Org. Lett. 2002, 4, 4447–4450.
(7) The regioselectivity of the asymmetric dihydroxylation of dienoates
has been studied by Sharpless and our group. See: (a) Berker, H.; Soler,
M. A.; Sharpless, K. B. Tetrahedron 1995, 51, 1345. (b) Zhang, Y.;
O’Doherty, G. A. Tetrahedron 2005, 61, 6337–6351.
(8) (a) Brown, C. A.; Yamashita, A. J. Am. Chem. Soc. 1975, 97, 891–
892. (b) Kimmel, T.; Becker, D. J. Org. Chem. 1984, 49, 2494–2496. (c)
Hoye, R. C.; Baigorria, A. S.; Danielson, M. E.; Pragman, A. A.; Rajapakse,
H. A. J. Org. Chem. 1999, 64, 2450–2453.
(9) (a) Rychnovsky, S. D.; Kim, J. J. Org. Chem. 1994, 59, 2659–2660.
(b) Trost, B.; Kazmaier, U. J. Am. Chem. Soc. 1992, 114, 7933–7935.
(10) Li, M.; Scott, J. G.; O’Doherty, G. A. Tetrahedron Lett. 2004, 45,
1005–1009.
(11) Trost, B. M.; Horne, D. B.; Woltering, M. J. Chem.-Eur. J. 2006,
12, 6607.
1108
Org. Lett., Vol. 11, No. 5, 2009

Scheme 3
.
De Novo Synthesis of Cladospolide C and
4,5-bis-epi-Cladospolide B 1
Scheme 4. Synthesis of Cladospolide B and
4,5-bis-epi-Cladospolide C 2
alkynoate 10 to the Rychnovsky variant of the Trost
isomerization (Ph3P/PhOH) cleanly gave E,E-dienoate 5 in
excellent yield (90%) and stereoselectivity.
Exposure of 14 to the Yamaguchi conditions yielded a 6:1
mixture of macrolactones 15 and 16. The lactones 15 and 16
were separated and deprotected with TFA to give cladospolide
B (81%)14 and 4,5-bis-epi-cladospolide C 2 (82%), respectively.
With the required four stereoisomeric macrolactone diols
with defined stereochemistry, we turned our attention to the
synthesis and structural proof of cladospolide D. This began
with the investigation of the selective oxidation of the allylic
alcohol in cladospolide C (Scheme 5). In our grahamimycin
A synthesis, we had previously found that the enone
functionality could be installed by a stoichiometric TEMPO
We next turned to the oxidative assembly of the macrolactone
natural product cladospolide C and its double bond isomer 4,5-
bis-epi-cladospolide B 1 (Scheme 3). The sequence began with
the use of the Sharpless protocol for the regio- and stereose-
lective dihydroxylation of dienoate 5 to provide diol 3.7 Acid-
catalyzed acetonide diol protection/TBS deprotection of 3
followed by base-promoted ester hydrolysis gave good yields
of the hydroxy acid 11 (74%). Unfortunately, the seco-acid 11
only gave polymeric products with exposure to the Keck-DCC
protocol for macrocyclization.12 In contrast, when the Yamagu-
chi conditions13 were used, good yields of a separable mixture
of macrolactones 12 and 13 were formed in a 6:1 ratio. Acid-
catalyzed deprotection of the two acetonides 12 and 13 with
TFA gave good yields of cladospolide C (80%)14 and 4,5-bis-
epi-cladospolide 1 (78%), respectively.
Scheme 5. Synthesis of Purported Cladospolide D
Following a nearly identical protocol, dienoate 5 was
converted into the macrolactone natural product cladospolide
B and its double bond isomer 4,5-bis-epi-cladospolide C 2
(Scheme 4). Simply switching the (DHQD)2PHAL ligand to
(DHQ)2PHAL in the Sharpless dihydroxylation gave the dia-
stereomeric diol 4 in the same yield and regioselectivity but
opposite diastereoselectivity. Similarly, acid-catalyzed acetonide
diol protection/TBS deprotection of 4 followed by base-
promoted ester hydrolysis gave the hydroxy acid 14 (70%).
(12) Keck, G. E.; Murry, J. A. J. Org. Chem. 1991, 56, 6606–6611.
(13) Inanaga, J.; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M. Bull.
Chem. Soc. Jpn. 1979, 52, 1989–1993.
(14) The 1H and 13C NMR spectral data, optical rotation, and mp match
the reported data for cladospolide B and C, although it should be noted
that Banwell reports the opposite value for cladospolide B. See Supporting
Information section E.
(15) The enantiomer of structure 22 was incorrectly assigned as
cladospolide D by Hou et al. See ref 4.
Org. Lett., Vol. 11, No. 5, 2009
1109

initially began with 4,5-bis-epi-cladospolide B 1 and clado-
spolide B (Scheme 6). As before, mono-TBS protection of
diol 1 was unselective, yielding a separable mixture of
alcohols 23 and 24 with the desired product isolated as the
minor isomer. Oxidation of the allylic alcohol 23 gave a good
yield of enone 25 (95%), which was deprotected with 5 mol
% HF/CH3CN (86%) to give one of the two possible cis-
diastereomers of the purported cladospolide D structure.
Similarly, the monoprotection of cladospolide B was unse-
lective to give an inseparable mixture of regioisomers 27/
28, which when oxidized gave a poor yield of the cis-enone
29. To our delight, upon exposure of 29 to 5 mol % HF/
CH3CN, TBS deprotection occurred to give γ-keto-enoate
30, with which 1H NMR and 13C NMR matched the reported
data for cladospolide D (vide infra).1d
Scheme 6. Synthesis of Cladospolide D
With the knowledge of the structure of the desired target
molecule, we next sought an improved synthesis (Scheme
7). This brought us back to the trans-enone 31, which was
made from the Dess-Martin oxidation of 20. Interestingly,
when 31 was treated with HF/Py (instead of 5 mol % HF/
CH3CN), a clean double bond isomerization occurred to give
the previous prepared cis-isomer 29. As before, TBS depro-
tection with 5 mol % HF/CH3CN occurred to give γ-keto-
enoate 30. Synthetic 30 material was spectroscopically (1H
NMR, 13C NMR, IR and MS; see page S107 in the
Supporting Information) identical with natural cladospolide
D, although the optical rotation was opposite in sign ([R]D
-58 vs + 56 in MeOH). Thus, structure 30 must be the
enantiomer of cladospolide D, and the structure should be
revised to the cis-isomer as shown in Figure 1.
In conclusion, a short and enantioselective synthesis of clado-
spolide B, cladospolide C, and (ent)-cladospolide D has been
developed along with the synthesis of all the possible stereo-
isomers of cladospolide D. Because the Noyori reduction for 7
can produce either enantiomer of 9, this synthesis is also a
formal synthesis of the natural stereoisomer of cladospolide D.
This de novo asymmetric approach installs all the stereocenters
in the cladospolides using two catalytic asymmetric reactions
(Noyori reduction and Sharpless asymmetric dihydroxylation).
The overall route to this class of natural products was achieved
in less steps and greater overall efficiency than the previous
routes from chiral starting materials and is amenable for the
preparation of either enantiomer. In addition, this study for the
first time correctly establishes the absolute and relative stere-
ochemistry of cladospolide D and serves as a correction of the
originally assigned natural product structure. Further studies on
the use of this de novo asymmetric sequence for the preparation
of other macrolactone antibiotics as well as cladospolide
analogues will be reported in due course.
oxidation of the allylic alcohol in colletodiol, without the
need for selective protection.6 Unfortunately, when we
applied these same conditions to cladospolide C, no simple
oxidation products were detected. Thus, we turned to a
selective protection/oxidation/deprotection sequence.
Mono-TBS protection of cladospolide C occurred when
exposed to TBSCl/imid. in CH2Cl2 (90%) to give a separable
mixture of regioisomers but with poor regioselectivity (17 to
18 in a 1:2 ratio). Dess-Martin oxidation of minor product 17
cleanly gave an enone, which after TBS deprotection (5 mol
% HF/CH3CN) gave 19, one of the two possible diastereomers
of the purported cladospolide D structure. Similar mono-TBS
protection of diol 2 occurred when exposed to TBSCl/imid. in
CH2Cl2 (90%), to give a separable mixture of regioisomers, with
better regioselectivity (20 to 21 in a 2:1 ratio). Dess-Martin
oxidation and TBS deprotection (5 mol % HF/CH3CN) of 20
gave 22, which is the other possible diastereomer of the
purported cladospolide D structure. To our surprise, comparison
of the 1H NMR and 13C NMR spectra of synthetic 19 and 22
with that reported for cladospolide D clearly indicated that these
two isomers were different structures than cladospolide D.15
Particularly diagnostic details were coupling constants and
chemical shifts for the R,ꢀ-enone protons.
Thus, we turned to the synthesis of the two diastereomers
of the cis-double bond isomer of the cladospolide D, which
Scheme 7. Improved Synthesis of Cladospolide D
Acknowledgment. We are grateful to NSF (CHE-
0749451) and the ACS-PRF (47094-AC1) for the support
of our research program and NSF-EPSCoR (0314742) for a
600 MHz NMR at WVU.
Supporting Information Available: Complete experi-
mental procedures and spectral data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL9000119
1110
Org. Lett., Vol. 11, No. 5, 2009