Stereoselective synthesis of methyl 7-dihydro-trioxacarcinoside B

Article abstract of DOI:10.1021/ol702078t

(Chemical Equation Presented) A stereoselective synthesis of 7-dihydro-triocacarcinose B, a branched octose from the quinocyclines, has been achieved. The biocatalytic resolution of a Baylis-Hillman adduct and a subsequent ring-closing metathesis were use

Full text of DOI:10.1021/ol702078t

ORGANIC
LETTERS
2
007
Vol. 9, No. 23
777-4779
Stereoselective Synthesis of Methyl
-Dihydro-trioxacarcinoside B
4
7
Christian M. K o1 nig, Klaus Harms, and Ulrich Koert*
Fachbereich Chemie, Philipps-UniVersity Marburg, Hans-Meerwein-Strasse,
D-35032 Marburg, Germany
koert@chemie.uni-marburg.de
Received August 23, 2007
ABSTRACT
A stereoselective synthesis of 7-dihydro-triocacarcinose B, a branched octose from the quinocyclines, has been achieved. The biocatalytic
resolution of a Baylis Hillman adduct and a subsequent ring-closing metathesis were used to assemble the molecular framework. Subsequent
key steps were a highly stereoselective epoxidation and a regio- and stereoselective opening of the epoxide by allyl alcohol and HClO to
introduce the C(3)-OH group in protected form. The 7-dihydro-triocacarcinose B could be converted into the corresponding 1,7-anhydrosugar.
−
4
Trioxacarcinose B (1) and 7-dihydro-trioxacarcinose B (2)
are structurally related γ-branched octoses found in the
quinocyclines (Figure 1). Quinocycline B and isoquinocy-
As part of a synthetic project toward the quinocyclines,
glycoconjugates of 1 and 2 have to be prepared. Glycocon-
jugates of 1 shall be accessible from glycoconjugates of 2
in a post-glycosylation step by selective oxidation of the C7
hydroxy group. These considerations led us first to develop
an efficient stereoselective synthesis of 7-dihydro-trioxa-
carcinose B (2).
1
cline B contain trioxacarcinose B (1), while quinocycline A
and isoquinocycline A are glycoconjugates of 7-dihydro-
2
trioxacarcinose B (2) and the anthracycline-type aglycone.
Quinocycline B was found to be identical with the recently
3
isolated kosinostatin. Trioxacarcinose B (1) is also part of
A retrosynthetic analysis reveals the regio- and stereose-
lective ring opening of an epoxide as a possible key step
(Figure 1). The properly protected target molecule 3 in its
low-energy chair conformation 4 has the 3,4-oxygen sub-
stituents in trans diaxial positions. This substructure should
be accessible by a regio- and stereoselective attack of a
4
5
trioxacarcin A and gutingimycin.
The antibiotic and cytotoxic activities of natural prod-
ucts containing the octoses 1 or 2 make these rare sugars
interesting to synthetic chemists. So far, a synthetic path
to 7-dihydro-trioxacarcinose B (2) has been reported by
6
7
2
8
Paulsen, and Suami has published a route to trioxacarcinose
B (1). Both made use of a chiral-pool approach starting from
sugar building blocks.
HOR -nucleophile at C3 of the epoxide 6, which leads to
9
the R-epoxide 5 as a key intermediate.
The starting point of the synthesis was a Baylis-Hillman
reaction of methylvinylketone and acetaldehyde followed by
1
0
(
1) (a) Celmer, W. D.; Murai, K.; Rao, K. V.; Tanner, F. W.; Marsh, W.
a biocatalytic resolution to obtain the R-methylene-â-
S. Antiobiot. Ann. 1957-58, 484-492. (b) Tulinsky, A. J. Am. Chem. Soc.
1
964, 86, 5368-5369.
2) Matern, U.; Grisebach, H.; Karl, W.; Achenbach, H. Eur. J. Biochem.
972, 29, 1-4.
3) (a) Furumai, T.; Igarashi, Y.; Higuchi, H.; Saito, N.; Oki, T. J.
(
(6) (a) Paulsen, H.; Sinnwell, V. Chem. Ber. 1978, 111, 869-878. (b)
Paulsen, H.; Sinnwell, V. Chem. Ber. 1978, 111, 879-889.
(7) (a) Suami, T.; Nakamura, K.; Hara, T. Chem. Lett. 1982, 1245-
1248. (b) Suami, T.; Nakamura, K.; Hara, T. Bull. Chem. Soc. Jpn. 1983,
56, 1431-1434.
(8) F u¨ rst, A.; Plattner, P. A. HelV. Chim. Acta 1949, 32, 275-283.
(9) The steroidal R,â-nomenclature is used at this point.
(10) (a) Burgess, K.; Jennings, L. D. J. Org. Chem. 1990, 55, 1138-
1139. (b) Barrett, A. G. M.; Kamimura, A. J. Chem. Soc., Chem. Commun.
1995, 1755.
1
(
Antibiot. 2002, 55, 128-133. (b) Igarashi, Y.; Higuchi, H.; Oki, T.; Furumai,
T. J. Antiobiot. 2002, 55, 134-140; Corrections J. Antiobiotics 2003, 56,
C1.
(
4) Suami, T.; Nakamura, K.; Hara, T. Bull. Chem. Soc. Jpn. 1983, 56,
431-1434.
5) Maskey, R. P.; Sevvana; Uson, M. I.; Helmke, E.; Laatsch, H. Angew.
Chem., Int. Ed. 2004, 43, 1281-1283.
1
(
1
0.1021/ol702078t CCC: $37.00
© 2007 American Chemical Society
Published on Web 10/13/2007

Scheme 1. Synthesis of the Dihydropyrane 13
successful ring-closing olefin methathesis13 to produce the
dihydropyran 13.
The further synthesis is summarized in Scheme 2. A
DIBAH reduction converted the lactone 13 into the corre-
sponding lactol which was transformed into the methyl
glycoside 14 (mixture of R- and â-anomers). As pointed out
above, the synthetic plan required the stereoselective forma-
tion of an R-epoxide 5. By avoidance of syn-pentane
Figure 1. Structures of octoses 1, 2, and their glycoconjugates.
Epoxide opening as a synthetic key step.
14
interactions compound 14 is expected to adopt a conforma-
1
5
tion 15 with an R-oriented C7-OH group. This OH group
hydroxy ketone 7 (Scheme 1). The use of Pseudomonas AK
should allow a substrate-directed stereocontrolled epoxida-
2
0 (Amano) led to an ee >99%. Next in the synthetic plan
was an anti selective reduction of the R-methylene-â-
hydroxyketone. Me NBH(OAc) gives excellent anti-selec-
16
2
tion. Indeed, treatment of 14 with VO(acac) and tert-butyl
1
7
hydroperoxide yielded the R-epoxide 16 as the only
observed stereoisomer.
4
3
11
tivities for â-hydroxyketones but failed for this R-methylene-
â-hydroxyketone. After TBS-protection to 8 the reduction
with diisobutylaluminum 2,6-di-tert-butyl-4-methylphenox-
In the following acid-catalyzed epoxide opening the
possibility for the formation of an anhydrosugar exists if the
7
-OH remains unprotected (vide infra). To circumvent this
12
ide led to the formation of the anti alcohol 9 with a 7:1
diastereoselectivity. The assignment of the relative config-
uration of the reduction product was done via NMR analysis
of the corresponding benzylidene acetals (Supporting Infor-
mation [SI]). Esterfication of 9 with vinylacetic acid gave
the diene 10. All attempts to convert 10 into a dihydropyran
by ring-closure olefin methathesis resulted in the intermo-
lecular reaction with the formation of the triene 11. The bulky
substituents at the disubstituted double bond in 10 pre-
vented the desired intramolecular reaction. Therefore, the
bulky TBS-ether was converted into a smaller hydroxy
group via desilylation. The obtained dienol 12 underwent a
problem, the epoxyalcohol 16 was protected in form of the
benzylic ether 17. Initial attempts to open the epoxide in 17
directly to the corresponding trans-diol failed. For example,
4 2
the use of HClO in THF/H O led mainly to starting-material
decomposition.
Other ROH nucleophiles, which result in a protected form
of the C3-OH group such as benzylic alcohol or allyl alcohol,
(
13) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999, 1,
9
1
53-956.
(14) Hoffmann, R. W. Angew. Chem., Int. Ed. Engl. 1992, 31, 1124-
134.
(
15) The steroidal R,â-nomenclature is used within the stereochemical
discussion of the epoxidation.
(
11) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc.
988, 110, 3560-3578.
12) Iguchi, S.; Nakai, H.; Hayashi, M.; Yamamoto, H. J. Org. Chem.
979, 44, 1363-1364.
(16) Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. ReV. 1993, 93,
1
1
1307-1370.
(
(17) Sharpless, K. B.; Michaelson, R. C. J. Am. Chem. Soc. 1973, 95,
6136-6137.
4778
Org. Lett., Vol. 9, No. 23, 2007

raphy. The spectral data and specific rotation for 19 and 20
were identical in all respects with those of the natural
products reported previously. An independent structural
Scheme 2. Synthesis of Methyl 7-Dihydro-trioxacarcinoside B
2
proof was possible by an X-ray crystal structure of the 3,7-
bis-(3,5-dinitrobenzoate) of 19. Structure 21 in Scheme 3 is
a structural representation of this X-ray structure.
Scheme 3. Synthesis of the Anhydrosugar 23
As pointed out already, the acid treatment of compounds
such as 16 with an unprotected 7-OH group led to the
formation of a 1,7-anhydrosugar (Scheme 3). The reaction
of 16 with allyl alcohol resulted in a combined epoxide
opening and anhydrosugar formation leading to the allyl ether
2
2. After a Pd-mediated cleavage of the allyl ether the
deprotected anhydrosugar 23 was obtained. The spectral data
and specific rotation for synthetic 23 were identical in all
respects with those of the natural product reported by Webb
1
9
et al.
In conclusion, an efficient, stereoselective route to methyl
-dihydro-trioxacarcinoside B as well as its 1,7-anhydrosugar
7
was developed. Key steps are a Baylis-Hillman/biocatalytic
resolution sequence, a ring-closing metathesis reaction, a
substrate-controlled epoxidation, and stereo- and regiocon-
trolled expoxide opening by allyl alcohol. The application
of this branched octose for the synthesis of quinocyclines is
currently under investigation.
were evaluated. It was found that use of the alcohol as the
solvent and 2-5 equiv of HClO as acid was optimal. Due
4
to its lower boiling point allyl alcohol was the reagent of
choice. The reaction temperature should not exceed 20 °C
to avoid decomposition of the starting material. The epoxide
opening was accompanied by a transacetalization at the
anomeric center leading to the allyl glycoside. A subsequent
treatment with CSA in MeOH gave the 3-allyloxy methyl
glycoside 18 (mixture of R- and â-anomers).
Acknowledgment. Generous support by the Deutsche
Forschungsgemeinschaft and the Fonds der Chemischen
Industrie is gratefully acknowledged.
Supporting Information Available: Experimental pro-
cedures and analytical data for all new compounds and
synthetic methyl glycosides 19, 20; single-X-ray crystal-
lographic data for 21. This material is available free of charge
via the Internet at http://pubs.acs.org.
The deprotection of the C3-allyl and C-7-benzyl ether
could be done in a stepwise manner. First, the cleavage of
18
the allyl ether with PdCl
hydrogenolysis of the benzyl ether gave the target compound
as methyl glycosides 19 (â-anomer) and 20 (R-anomer)
22% from 7). Both anomers were separated by chromotog-
2
in MeOH was possible. Second,
2
(
OL702078T
(
19) Webb, J. S.; Broschard, R. W.; Cosulich, D. B.; Mowat, J. H.;
(18) Bedini, E.; Carabellese, A. Tetrahedron 2005, 61, 5439-5448.
Lancaster, J. E. J. Am. Chem. Soc. 1962, 84, 3183-3184.
Org. Lett., Vol. 9, No. 23, 2007
4779