A short access to the macrocyclic core of cycloviracin and glucolipsin

Article abstract of DOI:10.1016/j.tetlet.2003.10.047

The macrocyclic core of cycloviracin and glucolipsin has been synthesised in ten steps from levoglucosan and (S)-(-)-dimethyl malate. The limited number of steps to obtain this macrolide makes it a valuable procedure for the synthesis of analogues of cycl

Full text of DOI:10.1016/j.tetlet.2003.10.047

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 9151–9153
A short access to the macrocyclic core of cycloviracin
and glucolipsin
Vincent Bailliez, Renata M. de Figueiredo, Alain Olesker and Jeannine Cleophax*
Institut de Chimie des Substances Naturelles du CNRS, Avenue de la terrasse, 91198 Gif-sur-Yvette, France
Received 3 September 2003; revised 2 October 2003; accepted 7 October 2003
Abstract—The macrocyclic core of cycloviracin and glucolipsin has been synthesised in ten steps from levoglucosan and
(S)-(−)-dimethyl malate. The limited number of steps to obtain this macrolide makes it a valuable procedure for the synthesis of
analogues of cycloviracin and glucolipsin.
© 2003 Elsevier Ltd. All rights reserved.
A few years ago, cycloviracin1 was isolated from an
actinomycete strain found in a soil sample collected in
Mindanao Island in the Philippines (Scheme 1). This
natural product exhibits a very interesting antiviral
activity against herpes simplex virus type I. The devel-
opment of new antiviral agents is necessary for fighting
these diseases owing to the appearance of resistant
mutant strains. Two major components have been iso-
lated: cycloviracins B1 and B2. These natural products
are eighteen-membered macrolides which include two
lated from the culture broth of Streptomyces purpuro-
geniscleroticus and Nocardia vaccinii, respectively,
relieving the inhibition of glucokinase by long chain
acyl CoA esters (FAC).3 Glucokinase is one of the
principal enzymes of glucose phosphorylation in glu-
cose sensitive cells. Glucose phosphorylation influences
circulating blood glucose levels, making glucokinase
regulation one of the main points for possible therapeu-
tic intervention in diabetes.
molecules of
D
-glucose and carry two side chains. The
Due to the original structure and interesting biological
activity of these natural products, several groups have
attempted their total syntheses. A partial synthesis of
cycloviracin has been achieved by Velarde and co-
workers.4 Recently, the total synthesis of cycloviracin
B1 has been described by Fu¨rstner and co-workers.5 It
is also worthy of note that Nishizawa and co-workers
structural difference involves the constitution of their
side chains.
The closely related structure of glucolipsin2 can also be
connected to this family of molecules (Scheme 1). Two
substances, called glucolipsin A and B have been iso-
Scheme 1. Structures of cycloviracin and glucolipsin.
* Corresponding author. Fax: +33-1-6907-7247; e-mail: jeannine.cleophax@icsn.cnrs-gif.fr
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2003.10.047

9152
V. Bailliez et al. / Tetrahedron Letters 44 (2003) 9151–9153
have obtained a closely related structure during their
total synthesis of arthrobacilin A.6 The absolute stereo-
chemistry of the 9 and 9% carbons was not elucidated
when these natural products were isolated. The com-
pleted syntheses have established the (R,R) configura-
tion of these carbons.4–6 The results described by
Fu¨rstner and co-workers, prompted us to disclose our
own results based on an approach allowing ready
access to analogues.
sis to afford 9 (Scheme 3). Subsequent selective cleav-
age of the anomeric acyl group with benzylamine,9
followed by formation of the corresponding
trichloroacetimidate gave the desired compound 5 in
56% overall yield (from 7). Starting from commercially
available (S)-(−)-dimethyl malate 8, we have synthe-
sised compound 6 in two steps and in 67% overall yield,
according to the procedure described by Moriwake.10
The glycosyl donor 5 (1.5 equiv.) was reacted with
alcohol 6 in acetonitrile: dichloromethane (1:1) at −
40°C, in the presence of trimethylsilyl triflate, to give 4
in 85% yield (a:b=1:5) as an inseparable mixture. The
We were interested in finding a simple and rapid access
to macrolide 3 allowing the easy introduction of vari-
ous types of side-chains (Scheme 2). In fact, the macro-
cyclisation step was envisaged by cyclodimerisation of
intermediate 4, prepared from the building blocks 5 and
6 by Schmidt glycosidation.7 Compounds 5 and 6 could
easily be obtained from levoglucosan 7 and commer-
cially available (S)-(−)-dimethyl malate 8, respectively.
Recently, we described an easy preparation of levoglu-
cosan 7 via a solid-supported, solvent-free, microwave-
assisted procedure.8
deacetylation of
4 was achieved under Zemplen
conditions11 to give 10 in 86% yield. At this stage, a
partial separation of the anomeric mixture by flash
chromatography on silica gel was accomplished. After
saponification of 10, 11 was obtained in 83% yield. The
cyclodimerisation was performed under Fu¨rstner’s con-
ditions from 11; we obtained the macrolide 312 in 37%
yield along with the cyclic monomer 12 in 56% yield
(and cyclic oligomers in 6% yield). These compounds
could be separated by flash chromatography. The over-
all yield could be improved by recycling 12 to 11 by
saponification (Scheme 4).
The synthesis of 5 began with the benzylation of the
hydroxy groups in levoglucosan 7, followed by acetoly-
Scheme 2. Retrosynthetic scheme for macrolide 3.
Scheme 3. Reagents and conditions: (A) NaH, BnBr, DMF, rt, 4 h, 90%; (B) CF3COOH, Ac2O, rt, 5 h, 92%; (C) BnNH2, rt, 3
h, 90%; (D) NaH (0.2 equiv.), Cl3CCN, CH2Cl2, rt, 1 h then NaH (0.2 equiv.), 2 h, 75%; (E) BH3·Me2S, THF then NaBH4, 4
h, 75%; (F) TBDPSCl, imidazole, DMF, rt, 12 h, 90%; (G) 6 (1 equiv.), 5 (1.5 equiv.), TMSOTf (0.1 equiv.), CH3CN: CH2Cl2
(1:1), −50°C, 1.5 h then rt, 30 min, 85%; (H) MeONa (20 mol%), MeOH, CH2Cl2, rt, 3 h, 86%; (I) LiOH aq., Et2O, MeOH, reflux,
1 h, 83%; (J) KH (2 equiv.), DMC (2.5 equiv.), DMAP (4 equiv.), CH2Cl2, rt, 15 h, 37%.

V. Bailliez et al. / Tetrahedron Letters 44 (2003) 9151–9153
9153
Mlynarski, J.; Albert, M. J. Am. Chem. Soc. 2002, 124,
10274.
6. Garcia, D. M.; Yamada, H.; Hatakeyama, S.; Nishizawa,
M. Tetrahedron Lett. 1994, 35, 3325.
7. (a) Schmidt, R. R. Angew. Chem., Int. Ed. Engl. 1986, 25,
212; (b) Schmidt, R. R.; Kinzy, W. Adv. Carbohydr.
Chem. Biochem. 1994, 50, 21.
Scheme 4.
8. Bailliez, V.; de Figueiredo, R. M.; Olesker, A.; Cleophax,
J. Synthesis 2003, 7, 1015.
9. Bourke, D. G.; Collins, D. J.; Hibberd, A. I. Aust. J.
Chem. 1996, 49, 425.
In conclusion, we have synthesised the macrocyclic core
of cycloviracin and glucolipsin in ten steps. This short
access makes it a valuable procedure for the synthesis
of analogues of cycloviracin and glucolipsin.
10. Moriwake, T. Chem. Lett. 1984, 1389.
11. Zemplen, G.; Gerecs, A.; Hadacsy, I. Ber. 1936, 69B,
1827.
Acknowledgements
12. Characteristic data for macrolide 3: colorless oil; Rf 0.52
(30% ethyl acetate in heptane); [h]2D0 +31 (c 1.9, CHCl3);
1H NMR (600 MHz, CDCl3), l=7.58–7.65 (m, 8H,
H-Ar), 7.12–7.40 (m, 42H, H-Ar), 4.90 (d, 2H, H-b, H-b%,
J=11 Hz), 4.83 (d, 2H, H-c, H-c%, J=11 Hz), 4.82 (d,
2H, H-a, H-a%, J=11 Hz), 4.76 (d, 2H, H-b, H-b%, J=11
Hz), 4.63 (d, 2H, H-a, H-a%, J=11 Hz), 4.50 (d, 2H, H-c,
H-c%, J=11 Hz), 4.45 (d, 2H, H-1, H-1%, J=7.8 Hz), 4.34
(dd, 2H, H-6, H-6%, J=2.2, 11.8 Hz), 4.19 (qt, 2H, H-9,
H-9%, J=5.3 Hz), 4.01 (dd, 2H, H-6, H-6%, J=4.9, 11.8
Hz), 3.80 (dd, 2H, H-10, H-10%, J=5.2, 10.5 Hz), 3.67
(dd, 2H, H-10, H-10%, J=6.2, 10.5 Hz), 3.59 (t, 2H, H-3,
H-3%, J=9 Hz), 3.51 (m, 2H, H-4, H-4%), 3.39 (m, 2H,
H-5, H-5%), 3.37 (dd, 2H, H-2, H-2%, J=7.8, 9 Hz), 2.78
(dd, 2H, H-8, H-8%, J=6.3, 16.4 Hz), 2.69 (dd, 2H, H-8,
H-8%, J=5, 16.4 Hz), 1.00 (s, 18H, 2×t-Bu); (qt: quintu-
plet). 13C NMR (150 MHz, CDCl3), l=171.3 (C-7, C-7%),
138.5, 138.2, 137.8, 135.6, 135.5, 129.8, 129.7, 128.4,
128.3, 128.2, 128.0, 127.9, 127.8, 127.7, 127.6, 127.5 (C-
Ar), 104.6 (C-1, C-1%), 84.5 (C-3, C-3%), 82.1 (C-2, C-2%),
78.0 (C-9, C-9%), 77.5 (C-4, C-4%), 75.6 (C-b, C-b%), 75.0
(C-c, C-c%), 74.7 (C-a, C-a%), 72.2 (C-5, C-5%), 65.3 (C-10,
C-10%), 63.2 (C-6, C-6%), 38.0 (C-8, C-8%), 26.8 (CH3, tBu),
19.2 (C-quaternary, tBu). HRMS (ESI-pos.) calcd for
C94H104O16NaSi2 (M+Na)+ 1567.6777. Found: 1567.6761.
We thank the CNRS for financial support and
MENRT for a grant (V.B.). The authors are grateful to
Marie-The´re`se Martin for the NMR spectra.
References
1. (a) Tsunakawa, M.; Komiyama, N.; Tenmyo, O.;
Tomita, K.; Kawano, K.; Kotake, C.; Konishi, M.; Oki,
T. J. Antibiot. 1992, 45, 1467; (b) Tsunakawa, M.;
Kotake, C.; Yamasaki, T.; Moriyama, T.; Konishi, M.;
Oki, T. J. Antibiot. 1992, 45, 1472.
2. Qian-Cutrone, J.; Ueki, T.; Huang, S.; Mookhtiar, K. A.;
Ezekiel, R.; Kalinowski, S. S.; Brown, K. S.; Golik, J.;
Lowe, S.; Pirnik, D. M.; Hugill, R.; Veitch, J. A.; Klohr,
S. E.; Whitney, J. L.; Manly, S. P. J. Antibiot. 1999, 52,
245.
3. Storer, A. C.; Cornish-Bowden, A. Biochem. J. 1977, 165,
61.
4. Velarde, S.; Urbina, J.; Pena, M. R. J. Org. Chem. 1996,
61, 9541.
5. (a) Fu¨rstner, A.; Albert, M.; Mlynarski, J.; Matheu, M.
J. Am. Chem. Soc. 2002, 124, 1168; (b) Fu¨rstner, A.;