Convergent de novo synthesis of vineomycinone B2 methyl ester

Article abstract of DOI:10.1039/c3cc44050h

An efficient de novo synthesis of vineomycinone B₂ methyl ester has been achieved. The longest linear route required only 14 steps from achiral commercially available starting materials (4.0% overall yield). The key transformations included the de novo asymmetric synthesis of two key fragments, which were joined by a convergent late stage Suzuki's glycosylation for the construction of the aryl β-C-glycoside. A subsequent BBr₃ one-pot debenzylation, demethylation and air oxidation provided vineomycinone B₂ methyl ester.

Full text of DOI:10.1039/c3cc44050h

ChemComm
View Article Online
View Journal
COMMUNICATION
Convergent de novo synthesis of vineomycinone B2
methyl ester†
Cite this: DOI: 10.1039/c3cc44050h
a
b
Qian Chen,z Yashan Zhongz and George A. O’Doherty*b
Received 29th May 2013,
Accepted 12th June 2013
DOI: 10.1039/c3cc44050h
www.rsc.org/chemcomm
An efficient de novo synthesis of vineomycinone B2 methyl ester has
Previously, we described the synthesis and anticancer activity of
been achieved. The longest linear route required only 14 steps from a sans-C-glycoside trisaccharide portion of vineomycin B2.6 More
achiral commercially available starting materials (4.0% overall yield). recently, we finished the asymmetric synthesis of both enantiomers
The key transformations included the de novo asymmetric synthesis of fridamycin E. These synthetic efforts enable the biological
of two key fragments, which were joined by a convergent late stage evaluation of these natural product structural motifs. Ultimately
Suzuki’s glycosylation for the construction of the aryl b-C-glycoside. we envision these efforts leading to a generalizable late stage
A subsequent BBr3 one-pot debenzylation, demethylation and air C-glycosylation strategy for the synthesis and structure activity
oxidation provided vineomycinone B2 methyl ester.
relationship studies of this class of C-glycoside natural products.7
Herein we describe our successful approach to vineomycinone
The anthraquinone containing vineomycin members of the B2 methyl ester, which is based upon our de novo Achmatowicz
angucycline family of antibiotics have been of particular interest approach to carbohydrates6 and our successful syntheses of
to both the synthetic and biological communities due to their fridamycin E8 from achiral starting materials (acyl furan and
unique structures and potent antitumor and antibacterial activities.1 anthrarufin, respectively). We believe this convergent de novo
Vineomycin B2 is a secondary metabolite of Streptomyces matensis asymmetric approach that utilizes a late stage Suzuki’s glycosyla-
subsp. vineus, which displays potent antitumor/antibiotic activity tion to install the b-C-glycoside provides synthetic material in a
with a pharmacologic profile similar to that of the clinically more efficient manner than the previous routes.
important anthracyclines.2,3 Acid-catalyzed methanolysis of
Our approach to vineomycinone B2 methyl ester (1) is
vineomycin B2 afforded its aglycon vineomycinone B2 methyl depicted in Scheme 1. We envisioned that 1 could be accessed
ester (1) which is the result of ortho-C-glycosylation (b-D-olivose) from a protected D-olivose or D-digitoxose sugar 2 as the glycosyl
of the methyl ester of fridamycin E (3).4
donor and the methyl ester of the natural product fridamycin E
The total synthesis of vineomycinone B2 methyl ester has been
achieved by several groups.5 While these approaches vary in terms of
overall efficiency, the routes tend to focus on the construction of the
aglycon portion, rather than b-C-glycoside. These methods varied
from the use of cycloadditions with siloxydienes, metalation–
stannylation reaction, Bradsher cycloaddition and Cp2HfCl2/AgClO4
promoted C-glycosylation to the use of tandem intramolecular
benzyne–furan cycloadditions.5 Despite the success and range of
these previous approaches, we felt there was room for a convergent
de novo asymmetric approach to this class of C-aryl glycoside natural
products. In additions to being able to provide access to enantio-
and diastereomeric isomers, a convergent de novo approach could
provide material in a reduced number of steps.
a Guangdong University of Technology, Guangzhou 510006, China
b Northeastern University, Boston, MA 02115, USA. E-mail: G.ODoherty@neu.edu
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3cc44050h
Scheme 1 Retrosynthesis of vineomycinone B2 methyl ester.
‡ These authors are listed alphabetically.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.

View Article Online
Communication
ChemComm
Scheme 3 Caption approach to glycosyl donor 17.
Scheme 2 Approach to glycosyl donor 12.
(3) as the aglycon. A Lewis acid promoted b-C-glycosylation could be
used to install the glycosidic bond of 1. While the olivose sugar is
more direct synthetically, the digitoxose diastereomer has the
potential for directing group effects that might be required for
selectivity in the C-glycosylation. Conveniently, our de novo
Achmatowicz strategy was envisioned to produce both diaster-
eomers of the glycosyl donor 2 from commercially available acyl
furan 4.5a So, the synthetic advantage to the olivose diastereomer
boils down to a step before versus after the point of convergence.
Building upon our fridamycin E work, we imagined an aglycon
subunit suitable for coupling (e.g., 3) could be prepared from
commercially available anthrarufin 5.8
Scheme 4 C-glycosylation attempts of fridamycin E methyl ester 3 with 12 and 17.
Previously, we have shown that the required Pd glycosyl donor
b-D-pyranone 6, as well as its enantiomer and a-diastereoisomers,
could be prepared from acyl furan 4 (Scheme 2). The stereodivergent
With the glycosyl donor in hand, we then turned to the synthesis
route employed an enantioselective Noyori reduction,9,10 an of the aglycon 3, the methyl ester of fridamycin E (Scheme 4), this
Achmatowicz oxidation, and diastereoselective tert-butyl carbo- was chosen as it would lead to the latest possible sight of conver-
nate formation.11,12 The protected b-D-digitoxose 9 was then gence. Previously, we have also developed the asymmetric synthesis
completed by a 4-step sequence (Scheme 2).6 This began with of fridamycin E.8 In this route, the b-lactone 19 was prepared in 7
Pd-catalyzed glycosylation of PMBOH and b-D-pyranone 6 to steps and in 19% overall yield. The synthesis began with the mono-
form the b-benzyloxy pyranone 7. A Luche reduction (NaBH4/ methallylation of anthrarufin, followed by a redox promoted Claisen
CeCl3) of 7 followed by the Myers’ reductive rearrangement to rearrangement and benzyl protection to afford ortho-methallylated
form 8 and the Upjohn dihydroxylation gave b-D-digitoxose product, was asymmetrically epoxidized in a 3-step protocol (Sharp-
sugar 9 (Scheme 2). The two hydroxyl groups were protected less dihydroxylation–tosylation–ring-closure) to give epoxide 18.
as acetate groups (Ac2O/Pyr) to form 10, which was followed by A subsequent Coates carbonylation regio- and stereospecifically
ceric ammonium nitrate (CAN) oxidative deprotection of the provided b-lactone 19, which after methanolysis (MeOH/K2CO3,
PMB-group to afford lactol 11. Finally the anomeric hydroxyl 96% yield) smoothly afforded b-hydroxyl carboxylic acid methyl ester
group was acylated to produce the desired glycosyl donor 12.
20 in excellent yield (96%) with the requisite tertiary alcohol. Finally,
We then turned to the preparation of the diastereomeric a reductive debenzylation of 20 with 1,4-cyclohexadiene and Pd/C in
D-digitoxose glycosyl donor 17 with and anomeric acetate ethanol gave the methyl ester of fridamycin E 3 in 86% yield.
(Scheme 3). This began with a highly regio- and stereospecific
With both glycosyl donor and aglycon in hand, we explored the
Mitsunobu like reaction (p-NO2PhCO2H/DIAD/PPh3) on cis-diol possibility of a Lewis acid promoted C-glycosylation of 3 with 12 or
9 to provide 13 in excellent overall yield. Hydrolysis of the nitro- 17 (Scheme 4). Unfortunately, our exhaustive survey of various
benzoate 13 with LiOH afforded the trans-diol 14 (88%). The Lewis acids and conditions (e.g., BF3ꢀEt2O, TMSOTf and
two hydroxyl groups were then protected as benzyl ethers Cp2HfCl2/AgClO4)5e met with no success. No signs of the desired
(BnBr, NaH/TBAI, 88% yield). A subsequent deprotection of C-glycoside product were detected.
the PMB-group in 15 with CAN smoothly converted it into lactol
Thus, we decided to target an earlier synthetic intermediate that
16 in 84% yield as an equilibrating mixture of diastereomers would (1) possess a more electron rich aromatic ring and (2) be
(a : b = 1.6 : 1). Finally, acylation of 16 gave b-D-olivose acetate 17 suitably protected to reveal the aglycon portion of the natural product
(99%, a : b = 2.8 : 1) in near quantitative yield.
in a single step.7e This led to the choice of the arene 23 with the
c
Chem. Commun.
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
In contrast, the C-glycosylation between 23 and 12 was less success-
ful, as only trace amount of the digitoxose isomer was observed.
In summary, the convergent de novo synthesis of vineomycinone
B2 methyl ester was accomplished in only 14 steps with 4.0% overall
yield in the longest linear sequence, which is a highly efficient route
comparing the shortest previous synthesis (16 steps) of 1.5 The route
shows the efficiency of the convergent de novo synthesis of coupling
partners for a late stage convergent C-glycosylation followed by a
single step deprotection–oxidation to yield the natural product. It is
interesting to note that while this is the first synthesis of vineomy-
cinone B2 methyl ester that uses asymmetric catalysis to install all of
the stereocenters, in terms of overall synthetic efficiency it compares
quite favorably with the elegant work of Martin and predecessors.5
Having this improved stereo-divergent assess to vineomycinone B2
methyl ester, its diastereomer and congeners should further enable
their biochemical/medicinal chemistry study. Further vineomycins
synthetic/biological investigations are ongoing and will be reported
in due course.
Scheme 5 Synthesis of the aglycon 23.
We are grateful to NIH (GM090259), NSF (CHE-1213596) and
the National Natural Science Foundation of China (QC, grant
21242002) for their financial support of this research.
Notes and references
1 (a) J. Rohr and R. Thiericke, Nat. Prod. Rep., 1992, 9, 103–137;
(b) K. Krohn and J. Rohr, Top. Curr. Chem., 1997, 188, 127–195;
(c) L. Andriy, V. Andreas and B. Andreas, Mol. BioSyst., 2005, 1, 117–126.
¯
¯
2 (a) S. Omura, H. Tanaka, R. Oiwa, L. Awaya, R. Masuma and K. Tanaka,
J. Antibiot., 1977, 30, 908–916; (b) N. Imamura, K. Kakinuma, N. Ikekawa,
H. Tanaka and S. Omura, J. Antibiot., 1981, 34, 1517–1578.
3 R. B. Weiss, G. Sarosy, K. Clagett-Carr, M. Russo and B. Leyland-
Jones, Cancer Chemother. Pharmacol., 1986, 18, 185–197.
¨
4 (a) P. Kricke, PhD thesis, University Gottingen, 1984; (b) K. Krohn and
Scheme 6 Synthesis of vineomycinone B2 methyl ester (1).
W. Baltus, Tetrahedron, 1988, 44, 49–54; (c) T. Matsumoto, H. Jona,
M. Katsuki and K. Suzuki, Tetrahedron Lett., 1991, 32, 5103–5106.
5 (a) D. J. Danishefsky, B. J. Uang and G. Quallich, J. Am. Chem. Soc.,
1985, 107, 1285–1293; (b) M. A. Tius, J. Gomez-Galeno, X. Gu and
J. H. Zaidi, J. Am. Chem. Soc., 1990, 112, 8188–8189; (c) M. A. Tius,
J. Gomez-Galeno, X. Gu and J. H. Zaidi, J. Am. Chem. Soc., 1991, 113,
5775–5783; (d) V. Bolitt, C. Mioskowski, R. O. Kollah, S. Manna,
D. Rajapaksa and J. R. Falck, J. Am. Chem. Soc., 1991, 113, 6320–6321;
(e) T. Matsumoto, M. Katsuki, H. Jona and K. Suzuki, J. Am. Chem.
Soc., 1991, 113, 6982–6992; ( f ) C.-L. Chen, S. M. Sparks and
S. F. Martin, J. Am. Chem. Soc., 2006, 128, 13696–13697.
6 (a) X. Yu and G. A. O’Doherty, Org. Lett., 2008, 10, 4529–4532. For
related approaches, see: (b) X. Yu, M. Li and G. A. O’Doherty,
Heterocycles, 2011, 82, 1577–1584; (c) A. Sobti, K. K. Kim and
G. A. Sulikowski, J. Org. Chem., 1996, 61, 6–7; (d) K. Sasaki,
S. Matsumura and K. Toshima, Tetrahedron Lett., 2007, 48, 6982–6986;
(e) M. Zhou and G. A. O’Doherty, Org. Lett., 2008, 10, 2283–2286;
( f ) M. Zhou and G. A. O’Doherty, J. Org. Chem., 2007, 72, 2485–2493;
(g) M. Zhou and G. A. O’Doherty, Org. Lett., 2006, 8, 4339–4342.
7 (a) Md. M. Ahmed and G. A. O’Doherty, Tetrahedron Lett., 2005, 46,
4151–4155; (b) D. Balachari and G. A. O’Doherty, Org. Lett., 2000, 2,
863–866; (c) D. Balachari and G. A. O’Doherty, Org. Lett., 2000, 2,
4033–4036; (d) D. Balachari, L. Quinn and G. A. O’Doherty, Tetra-
hedron Lett., 1999, 2, 4769–4773.
8 Q. Chen, M. Mulzer, P. Shi, P. J. Beuning, G. W. Coates and
G. A. O’Doherty, Org. Lett., 2011, 13, 6592–6595.
9 A. Fujii, S. Hashiguchi, N. Uematsu, T. Ikariya and R. Noyori, J. Am.
Chem. Soc., 1996, 118, 2521–2522.
10 (a) M. Li, J. G. Scott and G. A. O’Doherty, Tetrahedron Lett., 2004, 45,
1005–1009; (b) M. Li and G. A. O’Doherty, Tetrahedron Lett., 2004, 45,
6407–6411.
11 H. Guo and G. A. O’Doherty, J. Org. Chem., 2008, 73, 5211–5220.
12 (a) R. S. Babu and G. A. O’Doherty, J. Carbohydr. Chem., 2005, 24,
169–177; (b) H. Guo and G. A. O’Doherty, Org. Lett., 2005, 7, 3921–3924.
reduced quinone ring protected as a bis-methyl ether (Scheme 5). The
revised synthesis began with an MOM protection of 3, then in situ
dithionate reduction of 21 to an anthraquinol intermediate, followed
by methylation (KOH/Me2SO4) to give 22 in 63% yield. Finally
deprotection of 22 gave the bis-phenol 23 in 90% yield.
With the second-generation C-glycosylation coupling partner in
hand, we turned our focused on the C-glycosylation of 23 with 12 and
17 (Scheme 6). To our delight, treatment of a 1 : 1 mixture of 17 and
23 under the typical Suzuki conditions (Cp2HfCl2/AgClO4) led to
formation of the desired b-C-glycoside 24 as a single regio- and
diastereo-isomer in 48% yield. While the reaction is envisaged
as proceeding through an O- to C-glycoside rearrangement,5 no
evidence of an O-glycoside product was detected. A clue as to how
to best complete the synthesis came from careful observation of the
crude reaction mixture. Specifically, under the Lewis acid conditions
of the C-glycosylation reaction one can observe the conversion of the
acid sensitive starting material 23 into a corresponding anthraqui-
none. As a result, a one-pot per-deprotection–oxidation was realized
with the exposure of 24 to excess BBr3. The bis-debenzylation, bis-
demethylation and subsequent air oxidation of an anthraquinol
intermediate afforded vineomycinone B2 methyl ester 1 in 88% yield.
This synthetic material had physical and spectroscopic data in good
agreement with literature values (e.g., [a]2D0 = +109 (c = 0.27, dioxane),
[a]D29 = +118 (c = 1.05, dioxane),5e [a]D24 = +109.8 (c = 0.00091, CDCl3)).5f
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun.