Synthesis of tri-O-acetyl-D-allal from levoglucosenone

Article abstract of DOI:10.1021/ol302061a

Tri-O-acetyl-d-allal has been enantiospecifically synthesized in six steps from levoglucosenone in 55% overall yield. A key step in the synthesis is the anhydro bridge ring-opening with concomitant formation of a 1,3-oxathiolane-2- thione ring.

Full text of DOI:10.1021/ol302061a

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of Tri‑O‑acetyl‑D‑allal from
Levoglucosenone
ꢀ
Enrique D. V. Giordano, Agustina Frinchaboy, Alejandra G. Suarez, and
Rolando A. Spanevello*
´
´
ꢀ
Instituto de Quımica Rosario, Facultad de Ciencias Bioquımicas y Farmaceuticas,
Universidad Nacional de Rosario, CONICET Suipacha 531, S2002LRK Rosario,
Argentina
spanevello@iquir-conicet.gov.ar
Received July 25, 2012
ABSTRACT
Tri-O-acetyl-D-allal has been enantiospecifically synthesized in six steps from levoglucosenone in 55% overall yield. A key step in the synthesis is
the anhydro bridge ring-opening with concomitant formation of a 1,3-oxathiolane-2-thione ring.
Glycals are unsaturated derivatives of pentoses and
hexoses with a double bond between the anomeric carbon
and the adjacent one. They are very useful building
blocks in organic synthesis,1,2 as well as in carbohydrate
chemistry3 due to their enol ether functionality. Glycals
were discovered by Emil Fisher and Karl Zach4 at the
beginning of the twentieth century but, despite their long
existence, they remain a type of carbohydrate of great
interest because of the substitution,5 rearrangement,6 and
addition7 reactions they undergo and, more recently, they
have found uses in combinatorial chemistry.8 The glycals
most usually found in the literature are glucal and galactal,
while gulal and allal are not readily available, since they are
derived from rare sugars. It is noteworthy to mention that
rare sugars, in particular D-allose, have received much
attention because of their biological activity. In some cases
they can act as inhibitors of various glycosidases and cell
proliferation,9 to induce programmed cell death in hor-
mone refractory prostate cancer cells,10 or protective anti-
oxidative effects on ischemia-reperfusion damage in retina,11
brain12 and liver.13 However, regardless of their potential
importance in medicine and pharmacy, studies of their
biological effects are still limited because of their low
natural abundance.
(1) (a) Fraser-Reid, B., Tatsuta, K., Thiem, J., Eds. Glycoscience;
Springer-Verlag: Berlin, 2008. (b) Danishefsky, S. J.; Bilodeau, M. T. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1380. (c) Tolstikov, A. G.; Tolstikov,
G. A. Russ. Chem. Rev. 1993, 62, 579. (d) Moilanen, S. M.; Tan, D. S.
Org. Biomol. Chem. 2005, 3, 798.
(2) (a) Schmidt, R. R.; Vankar, Y. D. Acc. Chem. Res. 2008, 41, 1059.
(b) Ramesh, N. G.; Balasubramanian, K. K. Eur. J. Org. Chem. 2003,
4477.
(9) (a) Yamaguchi, F.; Kamitori, K.; Sanada, K.; Horii, M.; Dong,
Y.; Sui, L.; Tokuda, M. J. Biosci. Bioeng. 2008, 106, 248. (b) Yokohira,
M.; Hosokawa, K.; Keiko Yamakawa, K.; Saoo, K.; Matsuda, Y.;
Zeng, Y.; Kuno, T.; Imaida, K. J. Biosci. Bioeng. 2008, 105, 545. (c)
Tanaka, S.; Sakamoto, H. Cell Immunol. 2011, 271, 141.
(3) Seeberger, P. H.; Bilodeau, M. T.; Danishefsky, S. J. Aldrichimica
Acta 1997, 3, 75.
(4) Fischer, E.; Zach, K. Sitz. Ber. Kgl. Preuss. Akad. Wiss. 1913, 16,
311.
(5) Boons, G.-J.; Hale, K. J. Organic Synthesis with Carbohydrates;
Sheffield Academic Press: Shefffield, 2000.
(6) (a) Ferrier, R. J. Top. Curr. Chem. 2001, 215, 153. (b) Watanabe,
Y.; Itoh, T.; Sakakibara, T. Carbohydr. Res. 2009, 344, 516. (c) Godage,
H. Y.; Fairbanks, A. J. Tetrahedron Lett. 2003, 44, 3631.
(7) (a) Gammon, D. W.; Kinfe, H. H.; De Vos, D. E.; Jacobs, P. A.;
Sels, B. F. Tetrahedron Lett. 2004, 45, 9533. (b) Chun, K. H.; Yoon, T.;
Shin, J. E. N.; Jo, C.; Jo, Y. J.; Yun, H.; Oh, J. Bull. Korean Chem. Soc.
2005, 26, 1104.
(10) (a) Naha, N.; Lee, H. Y.; Jo, M. J.; Chung, B. C.; Kim, S. H.;
Kim, M. O. Apoptosis 2009, 14, 926. (b) Naha, N.; Lee, H. Y.; Jo, M. J.;
Chung, B. C.; Kim, S. H.; Kim, M. O. Apoptosis 2008, 13, 1121.
(11) (a) Mizote, M.; Hirooka, K.; Fukuda, K.; Nakamura, T.; Itano,
T.; Shiraga, F. Jpn. J. Ophthalmol. 2011, 55, 294. (b) Hirooka, K.;
Miyamoto, O.; Jinming, P.; Du, Y.; Itano, T.; Baba, T.; Tokuda, M. y;
Shiraga, F. Invest. Ophth. Vis. Sci. 2006, 47, 1653.
(12) (a) Nakamura, T.; Tanaka, S.; Hirooka, K.; Toyoshima, T.;
Kawai, N.; Tamiya, T.; Shiraga, F.; Tokuda, M.; Keep, R.; Itano, T.;
Miyamoto, O. Neurosci. Lett. 2011, 487, 103. (b) Gao, D.; Kawai, N.;
Tamiya, T. Med. Hypotheses 2011, 76, 911.
(13) (a) Hossain, M. A.; Izuishi, K.; Maeta, H. J. Hepato-Biliary-Pan
Surg. 2003, 10, 218. (b) Hossain, M. A.; Izuishi, K.; Tokuda, M.;
Izumori, K.; Maeta, H. J. Hepato-Biliary-Pan Surg. 2004, 11, 181.
(8) Hotha, S.; Tripathi, A. J. Comb. Chem. 2005, 7, 968.
r
10.1021/ol302061a
XXXX American Chemical Society

Several methods have been reported for the synthesis of
D-allal14 (1,5-anhydro-2-deoxy-D-ribo-hex-1-enitol, 1), a
glycal derivative of D-allose, although only a few of these
are efficient, and to the best of our knowledge, none of the
reported methodologies have allowed this building block
to become commercially available. Therefore, the devel-
opment of a simple and efficient synthetic route toward
this deceptively simple, rare sugar is still a challenge for
those devoted to carbohydrate synthesis.
skeletal rigidity that allows control in the generation of
desired stereogenic centers. The steric hindrance produced
by the anhydro bridge on the β face of the molecule directs
the approach of any reagent from its opposite face. There-
fore, reduction of the carbonyl moiety of the anhydropyr-
anose can be easily obtained employing the Luche con-
ditions18 (Scheme 2). Treatment of 4 with the NaBH4À
CeCl3 7H2O in methanol at room temperature afforded
3
the allylic alcohol 5 in a chemo- and stereoselective manner
in 92% isolated material. Though the sp3-hybridized C-2
carbon will be converted into an sp2 carbon in the final
compound, the configuration of this carbon is important
to achieve a complete diastereoselection during the dihy-
droxylation process on the olefin. We observed that when
the allylic substituent at C-2 has the opposite configura-
tion; this selectivity is strongly affected.
Scheme 1. Retrosynthetic Strategy
Scheme 2. Synthesis of the Xanthate Intermediate 6
Herein we describe a new and efficient synthesis of tri-
O-acetyl-D-allal (3,4,6-tri-O-acetyl-1,5-anhydro-2-deoxy-
D-ribo-hex-1-enitol, 11), from levoglucosenone (4), a
biomass-derived starting material. Levoglucosenone (1,6-
anhydro-3,4-dideoxy-β-D-glycero-hex-3-enopyranos-2-ulose)
is a versatile and readily available chiral synthon of the
carbohydrate-derived chiral pool.15 Conventional pyroly-
sis of cellulose-containing materials such as waste paper is
typically used to generate4,16 but microwave irradiation of
microcrystalline cellulose was recently found to be an
effective alternative.17
Our initial retrosynthetic strategy (Scheme 1) suggested
construction of the cyclic enol ether system through re-
ductive ring-opening of the anhydro bridge, induced by
elimination of an appropiate group at the C-2. Dihydrox-
ylation of the double bond in 3 would precede this event.
The conformational constraint characteristic of levoglu-
cosenone isdue to the 1,6-anhydrobridge, which imposes a
Transformation of the allylic alcohol 5 into the corre-
sponding allylic xanthate 6 was envisioned as a possible
route toward a suitable intermediate which could form the
1,2-double bond through a radical elimination pathway.
For this reason, a THF solution of 5 was reacted with
carbon disulfide, methyl iodide, and sodium hydride at
room temperature to afford the protected allylic xanthate
6 in 94% yield. cis-Dihydroxylation of 6 with catalytic
amounts of osmium tetroxide in tert-butyl alcohol, and
N-methylmorpholine N-oxide as stoichiometric oxidant,
afforded 7 as the sole product in 94% yield. A stereoche-
mical assigment of this compound was possible by the use
of 1H NMR spin decoupling and NOE data.
As previously mentioned, complete control of the dia-
stereoselectivity in the dihydroxylation pathway was due
to the presence of both a 1,6-anhydro bridge and the
xanthate group on the β-face of the pyranoside fragment.
These structural features allowed us to generate two new
stereogenic centers (C-3 and C-4) with the absolute con-
figuration present in the D-allal. Further treatment of 7
with acetic anhydride, pyridine, and 4-(dimethylamino)-
pyridine produced the diacetate 8 in 96% yield.
(14) (a) Kugelman, M.; Mallams, A. K.; Vernay, H. F. J. Chem. Soc.,
Perkin Trans. 1 1976, 1113. (b) Guthrie, R. D.; Irvine, R. W. Carbohydr.
Res. 1979, 72, 285. (c) Wittman, M. D.; Halcomb, R. L; Danishefsky,
S. J.; Golik, J.; Vyas, D. J. Org. Chem. 1990, 55, 1979. (d) Halcomb,
R. L.; Danishefsky, S. J.; Wittman, M. D. U.S. Patent 5,104,982, 1992.
(e) Boutureira, O.; Rodríguez, M. A.; Matheu, M. I. Org. Lett. 2006, 8, 673.
(f) Fujiwara, T.; Hayashi, M. J. Org. Chem. 2008, 73, 9161and references
cited therein.
(15) (a) Witczak, Z. J., Ed. Levoglucosenone and Levoglucosans:
Chemistry and Applications; ATL Press: Mount Prospect, 1994. (b) Witczak,
Z. J., Tatsuta, K., Eds. Carbohydrate Synthons in Natural Products
Chemistry. Synthesis, Functionalization and Applications; ACS Sympo-
sium Series 841; American Chemical Society: Washington, DC, 2003. (c)
ꢀ
Sarotti, A. M.; Zanardi, M. M.; Spanevello, R. A.; Suarez, A. G. Curr. Org.
Synth. 2012, 9, 439.
(16) (a) Shafizadeh, F.; Furneaux, R. H.; Stevenson, T. T. Carbohydr.
Res. 1979, 71, 169. (b) Swenton, J. S.; Freskos, J. N.; Dalidowicz, P.;
Kerns, M. L. J. Org. Chem. 1996, 61, 459.
ꢀ
(17) Sarotti, A. M.; Spanevello, R. A.; Suarez, A. G. Green Chem.
2007, 9, 1137.
(18) Kurti, L.; Czabo, B. Strategic Applications of Named Reactions
in Organic Synthesis; Elsevier Academic Press: San Diego, 2005.
B
Org. Lett., Vol. XX, No. XX, XXXX

Cleavage of the 1,6-anhydro bridge of 8 was a crucial
event in this synthetic strategy. The difficulties of opening
the anhydro bridge system are well documented.19 After
several unsuccessful attempts employing different reaction
conditions to pursue the opening of the anhydro bridge
through a radical fragmentation, we focused our efforts on
a two-step alternative to achieve our goal. First, an acid-
mediated ring-opening of the acetal would be performed
followed by glycal formation.
acetyl sulfide20 at 15 °C, and formation of a single product
was observed by TLC (Scheme 4).
Scheme 4. Synthesis of Tri-O-acetyl-D-allal
Reaction of xanthate 8 with hydrobromic acid in acetic
acid cleaved the anhydro bridge with a migration/rearran-
gement transformation of the xanthate group into a
dithiocarbonate at the anomeric carbon and the concomi-
tant attackof anacetate atC-2 toaffordthe thioglycoside9
in 32% yield (Scheme 3).
Scheme 3. Migration/Rearrangement of the Xanthate Group
To our surprise, the 300 MHz NMR analysis in CDCl3
revealed that we had achieved the ring-opening of the 1,6-
anhydro bridge with concomitant formation of a five-
membered ring 1,3-oxathiolane-2-thione, and acetylation
of the primary alcohol to afford 10 in 94% yield. The 300
MHz 1H NMR spectrum in CDCl3 showed the presence of
three methyl singlets attributed to the acetate groups but
the xanthate’s methyl group signal was not detected. The
75 MHz 13C NMR spectrum in CDCl3 showed an upfield
shift (14 ppm) for the C-1 and a downfield shift for C-2
(10 ppm) compared to its precursor, as well as the presence
of a signal for a thiocarbonyl group at 207.2 ppm. Also, in
this case, the ring junction of the bicyclic system was clearly
established by 1H NMR spin decoupling and NOE data.
We assumed that compound 10 could be a feasible
precursor for our target molecule if submitted to the
CoreyÀWinter reaction.18 Treatment of 10 with trimethyl
phosphite at 150 °C afforded cleanly the tri-O-acetyl-D-
allal (11) in 75% isolated yield.
This trans-cycloacetalization reaction involved once
again the participation of the xanthate group in the ring
opening (Scheme 5), but in this case probably a further
attack by the thioacetate generated from the acetyl sulfide
on the methyl group of the xanthate affords the cyclic
system.
The result of this ring-opening process differs consider-
ablyfrom theoutcomeof the acetolysis reaction performed
by treating 8 with acetic anhydride and trimethylsilyl
triflate, which afforded an unseparable mixture of anome-
ric acetate isomers in modest yields.
The 300 MHz 1H NMR spectrum in CDCl3 showed five
methyl singlets. Four of these were due to acetate groups,
plus there was methyl attached to a sulfur atom which
suffered an upfield shift compared to the starting material.
The H-1 signal also showed a downfield shift to 5.69 ppm
with a strong coupling (J = 10.7 Hz) that suggested the
antiperiplanar relationship between the substituents at-
tached at C-1 and C-2. HMBC experiments allowed assig-
ment of each of the signals in the spectrum, and irradiation
ofH-5 showed a NOE enhancement of H-1, confirming the
β configuration of the anomeric carbon in a D-allopyrano-
side derivative. The 75 MHz 13C NMR spectrum in CDCl3
showed no signal at 215 ppm characteristic of the xanthate
group. Instead, we observed an extra signal corresponding
to the primary acetate group and a new carbonyl signal at
187.2 ppm, which was consistent with rearrangement of
the xanthate group into a dithiocarbonate. The IR spec-
trum was also in agreement with the proposed structure,
since it showed an absorption band at 1748 cmÀ1 assign-
able to the acetate groups and another band at 1651 cmÀ1
attributable to the dithiocarbonate group.
Unfortunately, even though it was an interesting trans-
formation, neither the structure nor the yield were satis-
factory. Other attemps to obtain a thioglycoside by
treating the diacetate with thiophenol or thioanisole and
different Lewis acid were also disappointing.
Based on these results, we decided to try a thioacetolysis
procedure in order to obtain the anomeric thioacetate
without affecting the xanthate group for further manip-
ulation. Hence, we treated 8 with trimethylsilyl triflate and
A possible explanation for the different behavior can be
found by taking into account the characteristics of the
sulfur atom, which is an element of the third period with a
wider atomic radius, higher polarizability and nucleophi-
licity which could permit transition states that are not
feasible for smaller elements of the second period like
oxygen.
Inconclusion, we havesynthesizedtri-O-acetyl-D-allalin
six steps and 55% overall yield from levoglucosenone, a
biomass derived starting material. The key step in the
synthesis was the novel trans-cycloacetalization pathway
(19) See ref 15a, Chapter 12, p 173 and references cited therein.
(20) Bonner, W. A. J. Am. Chem. Soc. 1950, 72, 4270.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 5. Proposed Mechanism for Generation of the 1,3-Oxathiolane-2-thione Ring
ꢀ ꢀ
Agencia Nacional de Promocion Cientıfica y Tecnologica
´
(ANPCyT). E.D.G. thanks CONICET for the award of a
fellowship.
from the xanthate intermediate into the 1,3-oxathiolane-2-
thione ring. This new procedure for the preparation of
substituted D-allal is simple and efficient with an overall yield
that is significantly higher than the methodologies previously
reported. We also found a new application for acetyl sulfide,
which could have a more general use in the future.
Supporting Information Available. Experimental pro-
cedures for the synthesis of all compounds, characteriza-
tion data, and copies of 1H and 13C NMR spectra of new
products. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. This research was supported by grants
from Universidad Nacional de Rosario, Consejo Nacional de
ꢀ
´
Investigaciones Cientıficas y Tecnicas (CONICET), and
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX