Regioselectivity in Alkylation Reactions of 1,2-O-Stannylene Acetals of D-Arabinofuranose

Article abstract of DOI:10.1081/CAR-120026604

The synthesis of β-arabinofuranosides via alkylation of 1,2-O-stannylene acetal intermediates has been studied. With reactive alkyl halides (benzyl bromide, allyl bromide, and p-methoxybenzyl chloride), the method provides a mixture of β-arabinofuranosides and 2-O-alkylated lactols in ratios of 4:1 to 1:1.5. However, with carbohydrate-derived electrophiles, no alkylated products are produced. It appears, therefore, that the method is limited to the preparation of β-arabinofuranosides of simple alcohols. Through the use of computational chemistry, we have explored the conformational properties of one of these stannylene acetals and propose that these species exist in more than one conformation in solution and that this contributes to the relatively poor regioselectivity in these reactions.

Full text of DOI:10.1081/CAR-120026604

This article was downloaded by: [Case Western Reserve University]
On: 24 November 2014, At: 03:09
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Journal of Carbohydrate Chemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/lcar20
Regioselectivity in Alkylation Reactions of
1,2‐O‐Stannylene Acetals of d‐Arabinofuranose
Omar S. Darwish a , Christopher S. Callam a , Christopher M. Hadad a & Todd L. Lowary a b
a Department of Chemistry , The Ohio State University , 100 West 18th Ave., Columbus,
Ohio, 43210, USA
b Department of Chemistry and Alberta Ingenuity Centre for Carbohydrate Science , The
University of Alberta , Gunning‐Lemieux Chemistry Centre, Edmonton, Alberta, T6G 2G2,
Canada
Published online: 02 Feb 2007.
To cite this article: Omar S. Darwish , Christopher S. Callam , Christopher M. Hadad & Todd L. Lowary (2003) Regioselectivity
in Alkylation Reactions of 1,2‐O‐Stannylene Acetals of d‐Arabinofuranose, Journal of Carbohydrate Chemistry, 22:9, 963-981,
DOI: 10.1081/CAR-120026604
To link to this article: http://dx.doi.org/10.1081/CAR-120026604
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

JOURNAL OF CARBOHYDRATE CHEMISTRY
Vol. 22, No. 9, pp. 963–981, 2003
Regioselectivity in Alkylation Reactions of
1,2-O-Stannylene Acetals of
D-Arabinofuranose
Omar S. Darwish, Christopher S. Callam, Christopher M. Hadad,
*
and Todd L. Lowary
Department of Chemistry, The Ohio State University, Columbus, Ohio, USA
ABSTRACT
The synthesis of b-arabinofuranosides via alkylation of 1,2-O-stannylene acetal
intermediates has been studied. With reactive alkyl halides (benzyl bromide, allyl
bromide, and p-methoxybenzyl chloride), the method provides a mixture of b-arabi-
nofuranosides and 2-O-alkylated lactols in ratios of 4:1 to 1:1.5. However, with
carbohydrate-derived electrophiles, no alkylated products are produced. It appears,
therefore, that the method is limited to the preparation of b-arabinofuranosides of
simple alcohols. Through the use of computational chemistry, we have explored the
conformational properties of one of these stannylene acetals and propose that these
species exist in more than one conformation in solution and that this contributes to the
relatively poor regioselectivity in these reactions.
Key Words: Stannylene acetal; b-arabinofuranosides; Glycosylation; Alkylation.
*
Correspondence: Todd L. Lowary, Department of Chemistry and Alberta Ingenuity Centre for
Carbohydrate Science, The University of Alberta, Gunning-Lemieux Chemistry Centre, Edmonton,
Alberta, T6G 2G2, Canada; E-mail: tlowary@ualberta.ca.
963
DOI: 10.1081/CAR-120026604
0732-8303 (Print); 1532-2327 (Online)
www.dekker.com
Copyright D 2003 by Marcel Dekker, Inc.

964
Darwish et al.
Scheme 1. Representative examples of a b-arabinofuranoside (1), a b-fructofuranoside (2) and a
b-mannopyranoside (3).
INTRODUCTION
Polysaccharides containing b-arabinofuranosyl residues (1, Scheme 1) are impor-
tant structural components of the mycobacterial cell wall.[1,2] Recent interest[3,4] in
identifying inhibitors of the enzymes involved in the biosynthesis of these glycans has
prompted the development of new methods for the synthesis of b-arabinofurano-
sides,[5–10] as convenient access to fragments of these polysaccharides is needed for
fundamental biochemical investigations aimed ultimately at drug and vaccine devel-
opment.[11–15] The stereoselective synthesis of b-arabinofuranosides, as well as the
related b-fructofuranosides (2), is plagued by the same problems that complicate the
preparation of the stereochemically analogous b-mannopyranosides (3).[16,17] However,
in contrast to the significant body of work that has addressed the stereoselective
synthesis of b-mannopyranosides,[16,17] the preparation of 1,2-cis-b-furanosides has
received relatively little attention until recently.[5–10,18–20]
Shown in Figure 1 is a straightforward approach for the synthesis of b-manno-
pyranosides (or b-rhamnopyranosides) that was developed by Hodosi and Kovac.[21,22]
The method involves the treatment of a reducing sugar (4) with dibutyltin oxide
to produce the corresponding stannylene acetal (5), which is subsequently reacted
with an electrophile to generate the b-glycoside (6) in a stereoselective fashion.
This work extends earlier investigations on the alkylation of 1,2-O-stannylene
acetals with simple alkyl halides, which led to the production of b-mannopyrano-
sides,[23,24] a-glucopyranosides,[23] and a-ribofuranosides.[25] The simplicity of this
approach, coupled with the ready availability of 3,5-protected arabinofuranose de-
rivatives (e.g., 7, Figure 2), prompted us to explore this method for the synthesis of
b-arabinofuranosides.
Figure 1. Stannylene acetal approach to b-mannopyranosides developed by Hodosi and Kova´c.

1,2-O-Stannylene Acetals of D-Arabinofuranose
965
Figure 2. Proposed synthesis of b-arabinofuranosides via the stannylene acetal approach.
RESULTS AND DISCUSSION
In exploring the potential of the route shown in Figure 2 for the synthesis of b-
arabinofuranosides, we chose to probe the effect that the O-3 and O-5 protecting groups
had on the outcome of the reaction. We therefore chose diols 10–12 (Scheme 2) as
substrates. Compounds 10[26] and 11[27] were synthesized as previously reported. The
preparation of 12 (Figure 3) was achieved from the known thioglycoside 13,[28] upon
treatment with N-iodosuccinimide and silver triflate in wet acetonitrile. This reaction
afforded 12 in 78% yield as a 1:2.3 a:b mixture of isomers.
With all three substrates in hand, we explored the synthesis of simple alkyl gly-
cosides via the route outlined in Figure 2. A number of reaction conditions were
explored, and we found that the best results were obtained when the glycosylation was
carried out by heating a solution of the diol (e.g., 10) with dibutyltin oxide in toluene at
reflux for 30 minutes, followed by addition of the electrophile and n-Bu4NBr. The
progress of the reaction was monitored by TLC and was stopped after no change was
observed. In our initial series of reactions, we employed p-methoxybenzyl chloride,
benzyl bromide, or allyl bromide as the electrophile. The results of these reactions are
presented in Table 1; the structures of the products are provided in Scheme 3.
From the data presented in Table 1 it is clear that this approach is of limited utility for
the stereocontrolled synthesis of b-arabinofuranosides in high yield. As hoped, the reaction
produces the desired b-glycoside (e.g., 14) and none of the corresponding a-glycoside was
detected. However, in all cases, the reaction also produces significant quantities of the
isomeric 2-O-alkylated product (e.g., 15) as a byproduct. These lactol byproducts were
each isolated as an a/b mixture that was inseparable by chromatography.
Scheme 2. Glycosyl donors 10–12.

966
Darwish et al.
Figure 3. Synthesis of 12, (a) NIS, AgOTf, CH3CN, H2O, 0 °C, 78%.
1H and 13C NMR spectroscopy were used to determine the structures of the pro-
ducts. In the 1H NMR spectra of the b-glycosides, the resonance for the anomeric
hydrogen appeared as a doublet (J ꢀ 4.5 Hz) between 5.0 and 5.2 ppm. The 13C NMR
spectra of these compounds showed a resonance for the anomeric carbon in the range
of 99–102 ppm. Both of these features are diagnostic for b-arabinofuranosides.[29] Had
the a-glycosides been produced, the resonances for the anomeric hydrogens would have
appeared in the 1H NMR spectra as singlets, or doublets with a smaller coupling
constant (J ꢀ 1.5 Hz) and the resonances for the anomeric carbons would have been
found in the 13C NMR spectra above 105 ppm.[29] The identity of the lactols was
easily ascertained from 13C NMR spectra, which showed two resonances in the region
Table 1. Use of 1,2-O-stannylene acetals in the synthesis of b-arabinofuranosides.

1,2-O-Stannylene Acetals of D-Arabinofuranose
967
Scheme 3. Structures of products produced in alkylation reactions.
between 94 and 102 ppm, corresponding to C-1 in the a and b reducing sugars.
The presence of the two isomers was also clearly visible in the H NMR spectrum;
1
however, due to spectral overlap it was sometimes difficult to assign the signals arising
from the anomeric hydrogen in both anomers. In the case of 17, further proof for the
1
structure was obtained by comparison with the H and 13C NMR data for the known
compound,[30] which is commercially available.
The overall alkylation yields ranged from 20–80% and the best results were
obtained when p-methoxybenzyl chloride was used as the electrophile. With regard to
the donor substrates, the benzylated and silylated donors 10 and 12 gave modest to
good combined yields of the alkylated products. The yields were slightly better with 12
as compared to 10; however, the regioselectivity in favor of the b-glycoside was
marginally higher when 10 was used as the donor. It is perhaps not unexpected that the
benzoylated substrate, 11, gave the lowest overall yields. The electron-withdrawing
ability of the ester protecting groups on 11 would be expected to reduce the nucleo-
philicity of the stannylene acetal oxygens through an inductive effect.
To rationalize the regioselectivity of these reactions, we considered first the factors
that govern alkylations of stannylene acetals. In reactions involving stannylene acetals
formed from non-anomeric cis-diols on pyranose rings, the equatorial oxygen is almost
always the site of predominant reaction.[31,32] Although these stannylene acetals are
known to aggregate in solution, when halide salts are added to promote the reaction (as
in the present case), it is believed that they react through their monomeric forms.[31] An
important consideration is that the relative rigidity of the pyranose ring ensures that the
molecule adopts a single major conformation in solution, at least with respect to the
relative orientation of O-1 and O-2, and thus the regioselectivity is generally high.
However, anomalous results have been obtained. For example, Srivastava and Schuerch
reported[23] that reaction of methyl iodide with stannylene acetal 32 (Scheme 4) pro-
vided a 7:3 mixture of a-glycoside 33 and the 2-O-alkylated lactol 34. It was subse-
quently shown[22] that treatment of triflate 35 with the maltose-derived stannylene

968
Darwish et al.
Scheme 4. Compounds 32–37.
acetal 36 yielded the 2-O alkylated product 37 in 75% yield and no mention was made
of the formation of the corresponding glycoside.
The paucity of previous investigations on organotin-mediated reactions on furanose
rings makes it difficult to draw substantially on literature precedence when proposing
an explanation for the regioselectivity observed in the alkylation reactions of these
arabinofuranose systems. Although some of the first demonstrated examples of
stannylene acetals in protecting group chemistry were carried out on nucleosides,[33]
relatively little additional work has been done on these systems to date.[25,34,35] In the
initial paper by Moffat and coworkers,[33] acylations of ribonucleosides (e.g., 38,
Scheme 5) via stanylene acetals were investigated. In general, alkylation of O-3 pre-
dominated (e.g., 39), however, determining the inherent regioselectivity of the process
was complicated by the tendency of the acyl group on O-2 to migrate to O-3. A paper
more directly relevant to the present work describes the alkylation of stannylene acetal
40.[36,37] When methyl iodide was used as the electrophile, a mixture of 2-O-methyl
lactol (41) and a-glycoside (42) were produced in 1:1 ratio, whereas benzyl bromide
afforded the a-glycoside as the major product over the corresponding lactol
(83%:13%).
Furanose rings are inherently more flexible than their pyranose counterparts, and
the substituents on a five membered ring adopt pseudo-axial or pseudo-equatorial
orientations rather than classical axial or equatorial arrangements. Mindful of this issue,
Scheme 5. Compounds 38–42.

1,2-O-Stannylene Acetals of D-Arabinofuranose
969
Scheme 6. Comparison of 43 and 44.
we envisioned two possible explanations for the poor regioselectivity observed in these
alkylation reactions. One possibility is that in the stannylene acetal intermediate the
pseudo-axial/pseudo-equatorial placement of both oxygens does not lead to sufficient
steric differentiation and therefore they compete for the electrophile. Another plausible
explanation is that the stannylene acetal adopts two or more low energy conformations
in which the relative orientations of O-1 and O-2 differ, e.g., O-1 is pseudo-axial in one
conformer and pseudo-equatorial in another. An erosion in regioselectivity would be
expected to result from the reaction of both conformers with the electrophile. Under
either scenario, it could be expected that changes in reaction conditions (e.g., protecting
groups on the donor, the electrophile used) would influence the regioselectivity of the
reaction, as is observed in the results presented in Table 1.
In an effort to probe these hypotheses we carried out density functional theory[36,37]
calculations on stannylene acetal, 43 (Scheme 6), the species formed upon reaction of
12 with dibutyltin oxide. This substrate was chosen as the yields of these reactions
were the best of the three donors explored. The general flexibility of the ring was of
concern and so we began by generating a diverse family of conformers using the
Systematic Pseudo Monte Carlo (SPMC) search protocol as implemented in Macro-
Model version 6.5.[38,39] A limitation in the application of MacroModel to this problem
is that it is not parameterized for tin and therefore the SPMC search was carried out on
an analog (44, Scheme 6) in which the tin atom was replaced with silicon. As the
purpose of the search was simply to generate diverse geometries for higher level
calculations, we viewed the tin/silicon replacement as a reasonable alternative to the
manual generation of a large number of initial starting geometries via an approach that
would necessarily be biased by intuition.
A total of 10,000 structures of 44 were generated by the SPMC search, which were
then minimized in the gas phase at the AM1[40] level of theory, yielding a family of
113 conformers, all of which were within 32 kcal/mol of the lowest energy structure.
Clustered around the global minimum was a group of four structures that all possessed
similar furanose ring conformations. A second family of six conformers, all possessing
a common furanose ring conformation, was present ꢀ 1.1 kcal/mol above the global
minimum (see Experimental Section for additional details). The other 103 conformers
obtained after AM1 optimization were more than 3.0 kcal/mol above the global min-
imum structure. The lowest AM1 energy structures from each of these two families of
conformers were then taken, the silicon atom connecting O-1 and O-2 replaced with
tin, and the resulting structures were subjected to higher level calculations as outlined
in the Experimental section. The optimized geometries for these conformers (43-A and

970
Darwish et al.
43-B) are shown in Scheme 7 (all hydrogens and the alkyl groups on the siloxane
moiety are omitted for clarity). The furanose ring in 43-A adopts an OE conformation,
3
which places O-1 pseudo-equatorial and O-2 pseudo-axial; in 43-B the ring is in a T2
conformation, in which O-2 is pseudo-equatorial and O-1 is pseudo-axial. Thus, the
orientation of O-1 and O-2 differ in these two structures. Based upon our calculations,
these two conformers differ by 2.7 kcal/mol, with 43-A being of lower energy.
These calculations suggest that the furanose ring in stannylene acetal 43 exists in (at
least) two conformations in which the relative pseudo-axial/pseudo-equatorial orienta-
tions of O-1 and O-2 differ. Moreover, the differences in energy between these con-
formers, 43-A and 43-B, are relatively small and, given the relatively low interconversion
barriers typically seen for furanose rings,[41] it is reasonable to expect that these isomers
interconvert rapidly, especially under the elevated temperatures at which these reactions
are carried out. Taken together, these data suggest that the low regioselectivity observed
in the alkylation reactions in these systems is the result of the presence of more than one
stannylene acetal conformer, which react with electrophiles at O-1 and O-2 to varying
degrees. We note that we have examined only the thermodynamically most stable con-
formers and it is possible (likely) that other reactive conformers may also be present.
Under these conditions, it is plausible that the regioselectivity of the reaction is influ-
enced not only by these conformational effects, but also by the donor substrates and
reaction conditions. For example, in the results presented in Table 1, the benzoylated
Scheme 7. B3LYP-optimized geometries of 43-A and 43-B. For clarity, all hydrogen atoms as
well as carbon atoms of the siloxane acetal and butyl groups have been deleted.

1,2-O-Stannylene Acetals of D-Arabinofuranose
971
Scheme 8. Compounds 45–49.
donor 11 gave the highest regioselectivity in the alkylation reactions. The electron-
withdrawing ability of the benzoyloxy group at C-3 can be invoked to rationalize this
observation. The closer proximity of this protecting group to O-2 would be expected to
decrease the nucleophilicity of this oxygen relative to O-1.
Despite the relatively modest results in the alkylation reactions with the simple
electrophiles, we explored the possibility of synthesizing oligosaccharides via this
approach. As a simple model carbohydrate electrophile, we chose triflate 45
(Scheme 8), which was prepared from the known alcohol 46[42] upon treatment with
triflic anhydride and pyridine. However, upon reaction of 45 with the stannylidene
acetal generated from 12, no alkylated products could be isolated. Earlier reports[22]
have demonstrated that polar solvents can also be used to promote these alkylation
reactions. The reaction was therefore repeated by generating the stannylene acetal and
then adding 45 and DMF. However, when these conditions were used, no disaccharide
was formed; rather, the formate ester 47 was produced in modest (39%) yield. The
production of formate esters as byproducts in these reactions has been previously
reported[22] and has been minimized by carrying out the reaction at reduced tem-
peratures. However, in our hands, cooling the reaction mixture to room temperature
led to the formation of no alkylated products. As a final attempt to prepare a di-
saccharide, we synthesized the iodide 49 (via tosylate 48) and reacted it with stan-
nylene acetal 12 and n-Bu4NBr in toluene at reflux. However, no alkylated products
were produced.
CONCLUSIONS
In conclusion, the stereoselective synthesis of b-arabinofuranosides in high yields
was shown to be unsuccessful via of the 1,2-O-stannylene approach. Although the
formation of the a-glycoside was avoided, our results with simple electrophiles dem-
onstrated that the reactivity difference between the O-1 and O-2 is small. We propose
that in these furanose ring systems, more than one reactive stannylene acetal conformer
is present and they collectively contribute to the decrease in regioselectivity observed
in these alkylation reactions. This contrasts to the application of this method to the
synthesis of b-mannopyranoside or b-rhamnopyranosides, where the rigid nature of the
pyranose ring orients the tin acetal oxygen atoms in distinctly different steric envir-
onments. In the course of our investigations, Oscarsson and coworkers reported a study
in which the stereoselective synthesis of b-fructofuranosides (e.g., 2, Scheme 1) via
stannylene acetals was explored.[18] When a benzylated species analogous to 10 was

972
Darwish et al.
used as the donor and allyl bromide was employed as the electrophile, the desired
fructofuranosides were produced as a 1:4.6 a/b mixture. Interestingly, no products
arising from alkylation at the non-anomeric oxygen were reported; however, as in our
investigations, the use of carbohydrate electrophiles gave no disaccharide products.
These results, together with ours and earlier work by Klein[25] and coworkers, suggest
the 1,2-O-stannylene acetal glycosylation methodology is of limited utility for the
stereocontrolled synthesis of 1,2-cis-furanosides in high yield.
EXPERIMENTAL
Solvents were distilled from the appropriate drying agents before use. Unless stated
otherwise, all reactions were carried out under a positive pressure of argon and were
monitored by TLC on silica gel 60 F254 (0.25 mm, E. Merck). Spots were detected
under UV light, by charring with 10% H2SO4 in ethanol, or by charring with ani-
saldehyde in ethanol. Solvents were evaporated under reduced pressure and below 40°C
(bath). Column chromatography was performed on silica gel 60 (40–60 mM). The ratio
between silica gel and crude product ranged from 100 to 50:1 (w/w). Optical rotations
1
were measured at 21 2°C. H NMR spectra were recorded at 400 or 500 MHz, and
chemical shifts are referenced to TMS (0.0, CDCl3). 13C NMR spectra were recorded at
100 or 125 MHz, and 13C chemical shifts are referenced to CDCl3 (77.00, CDCl3).
Elemental analyses were performed by Atlantic Microlab Inc., Norcross, GA. Elec-
trospray mass spectra were recorded on samples suspended in THF or CH3OH with
added NaCl.
3,5-O-(Tetraisopropylsiloxane-1,3-diyl)-D-arabinofuranose (12). Thioglycoside
13 (990 mg, 2.0 mmol) was dissolved in acetonitrile (10 mL) and water (2 mL) and the
mixture was cooled to 0°C. N-iodosuccinimide (0.49 g, 2.2 mmol) and silver triflate
(0.13 g, 0.5 mmol) were added to the reaction mixture and, after stirring for 5 min,
triethylamine (1.0 mL) was added and the mixture was filtered through Celite. The
filtrate was diluted with CH2Cl2, washed with water, and the organic layer was dried
(MgSO4) and concentrated. The residue was purified by chromatography (hexane/
1
EtOAc, 2:1) to give 12 (600 mg, 78%) as a clear syrup: Rf 0.3 (hexane/EtOAc, 2:1); H
NMR (400 MHz, CDCl3, dH) 5.25 (d, 0.7 H, J = 4.7 Hz), 5.22 (d, 0.3 H, J = 2.8 Hz),
4.31–3.63 (m, 6 H), 2.70 (br s, 1 H), 1.10–0.94 (m, 28 H); 13C NMR (100 MHz,
CDCl3, dC) 101.9, 95.0, 82.8, 81.7, 78.6, 78.4, 78.1, 73.0, 66.3, 65.3, 17.9 (2), 17.80
(2), 17.7 (3), 17.5 (2), 17.4 (4), 13.9, 13.6, 13.2, 12.9 (2).
Anal. Calcd for C17H36O6Si2 (398.1366): C, 52.00; H, 9.24. Found: C, 51.98; H,
9.21.
General procedure for glycosylation reactions. A mixture of the diol (1.3
mmol), and dibutyltin oxide (1.5 mmol) was heated at reflux in toluene (15 mL) for 30
min. The electrophile (1.9 mmol) and n-Bu4NBr (1.2 mmol) were then added and
stirring with heating continued until there was no change in the progress of the reaction
as monitored by TLC. The reaction mixture was cooled, concentrated, and the products
were purified by column chromatography.

1,2-O-Stannylene Acetals of D-Arabinofuranose
973
Allyl 3,5-di-O-benzyl- -D-arabinofuranoside (14) and 2-O-Allyl-3,5-di-O-ben-
zyl-D-arabinofuranose (15). The crude mixture resulting from the reaction of 10 and
allyl bromide was purified by chromatography (hexane/EtOAc, 5:1) to give 14 (39%)
and 15 (24%) as clear oils. 14: Rf 0.5 (hexane/EtOAc, 2:1); [a]D ꢁ 34.9° (c 0.3,
1
CHCl3); H NMR (400 MHz, CDCl3, dH) 7.33–7.27 (m, 10 H), 5.88–5.81 (m, 1 H),
5.23 (ddd, 1 H, J = 17.2, 3.1, 1.6 Hz), 5.17 (ddd, 1 H, J = 10.4, 2.7, 1.3 Hz), 5.00 (d, 1
H, J = 4.7 Hz), 4.75 (d, 1 H, J = 11.9 Hz), 4.65 (d, 1 H, J = 12.3 Hz), 4.60–4.52 (m, 2
H), 4.26 (m, 2 H), 4.14 (dd, 1 H, J = 11.3, 5.6 Hz), 4.04–3.99 (m, 1 H), 3.86 (t, 1 H,
J = 5.8 Hz), 3.53 (d, 2 H, J = 5.7 Hz), 2.62 (br d, 1 H, J = 7.9 Hz); 13C NMR (100
MHz CDCl3, dC) 138.5 (2), 134.2, 129.0, 128.8. 128.2, 128.1, 118.1, 101.2, 85.2, 81.3,
78.5, 73.7, 72.5, 72.3, 69.1.
Anal. Calcd for C22H26O5 (370.18): C, 71.33; H, 7.07. Found: C, 71.39; H, 7.10.
1
15: Rf 0.4 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 7.35–7.25 (m, 10
H), 5.91–5.79 (m, 1 H), 5.34–5.16 (m, 3 H), 4.62–4.45 (m, 4 H), 4.11–3.89 (m, 5 H),
3.80 (d, 0.4 H, J = 6.0 Hz), 3.58–3.49 (m, 2 H), 3.32 (d, 0.6 H, J = 7.9 Hz); 13C NMR
(100 MHz, CDCl3, dC) 138.5, 138.3, 137.9, 137.7, 134.4, 134.3, 128.9, 128.8, 128.4,
128.3, 128.2 (2), 128.1, 118.3, 117.9, 101.5, 96.6, 86.8, 84.5, 83.1, 82.4, 82.2, 80.8,
74.0, 73.8, 72.5, 72.4, 71.7, 71.1, 70.9, 70.6. HRMS (ESI) Calcd for C22H26O5
[M + Na]+ 393.1672. Found 393.1709.
Benzyl 3,5-di-O-benzyl- -D-arabinofuranoside (16) and 2,3,5-Tri-O-benzyl-D-
arabinofuranose (17). The crude mixture resulting from the reaction of 10 and
benzyl bromide was purified by chromatography (hexane/EtOAc, 4:1) to give 16 (39%)
as a clear oil and 17[27] (22%) as a white solid. 16: Rf 0.5 (hexane/EtOAc, 2:1);
1
[a]D ꢁ 40.2° (c 0.02, CHCl3); H NMR (400 MHz, CDCl3, dH) 7.35–7.19 (m, 15 H),
5.07 (d, 1 H, J = 4.7 Hz), 4.78 (d, 1 H, J = 8.0 Hz), 4.75 (d, 1 H, J = 8.3 Hz), 4.63–
4.48 (m, 4 H), 4.27 (t, 1 H, J = 5.4 Hz), 4.17 (dd, 1 H, J = 11.3, 5.6 Hz), 3.89 (t, 1 H,
J = 5.9 Hz), 3.56 (d, 2 H, J = 5.7 Hz), 2.58 (br s, 1 H); 13C NMR (100 MHz, CDCl3,
dC) 138.0 (2), 137.1, 128.5, 128.4 (2), 128.2, 128.0, 127.8, 127.7 (2), 100.8, 84.7, 81.0,
78.1, 73.3, 72.0 (2), 69.7.
Anal. Calcd for C26H28O5 (420.19): C, 74.26; H, 6.71. Found: C, 74.05; H, 6.66.
1
17: Rf 0.3 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 7.34–7.24 (m, 15
H), 5.38 (d, 0.2 H, J = 7.7 Hz), 5.32 (dd, 0.8 H, J = 9.9, 4.3 Hz), 4.66–4.44 (m, 6 H),
4.17–4.14 (m, 0.8 H), 4.10–4.07 (m, 0.8 H), 4.02–3.97 (m, 1.4 H), 3.85 (d, 0.8 H,
J = 10.0 Hz), 3.62–3.49 (m, 2 H), 3.23 (d, 0.2 H, J = 7.8 Hz); 13C NMR (100 MHz,
CDCl3, dC) 138.5, 138.3, 137.86, 137.9, 137.7, 129.0, 128.9, 128.8, 128.5, 128.4, 128.3
(2), 128.2 (2), 128.1, 101.6, 96.7, 86.9, 84.5, 83.1, 82.5, 82.3, 81.0, 77.8, 74.0, 73.8,
72.7, 72.5, 72.4, 72.2, 71.0, 70.6. HRMS (ESI) Calcd for C26H28O5 [M + Na]+
443.1834. Found 443.1829.
p-Methoxybenzyl 3,5-di-O-benzyl- -D-arabinofuranoside (18) and 3,5-Di-O-
benzyl-2-O-p-methoxybenzyl-D-arabinofuranose (19). The crude mixture resulting
from the reaction of 10 and p-methoxybenzyl chloride was purified by chromatography
(hexane/EtOAc, 3:1) to give 18 (36%) and 19 (36%) both as white solids. 18: Rf 0.3

974
Darwish et al.
1
(hexane/EtOAc, 2:1); mp 41–44°C; [a]D ꢁ 25.5° (c 0.2, CHCl3); H NMR (400 MHz,
CDCl3, dH) 7.33–7.25 (m, 10 H), 7.18 (d, 2 H, J = 8.4 Hz), 6.83 (d, 2 H, J = 8.4 Hz),
5.04 (d, 1 H, J = 4.6 Hz), 4.75 (d, 1 H, J = 11.9 Hz), 4.70 (d, 1 H, J = 11.2 Hz),
4.62–4.55 (m, 3 H), 4.42 (d, 1 H, J = 11.2 Hz), 4.27–4.23 (m, 1 H), 4.17 (dd, 1 H,
J = 11.2, 5.6 Hz), 3.87 (t, 1 H, J = 5.9 Hz), 3.78 (s, 3 H), 3.56 (d, 2 H, J = 5.6 Hz),
2.62 (d, 1 H, J = 9.5 Hz); 13C NMR (100 MHz, CDCl3, dC) 159.9, 138.5 (2), 130.4,
129.6, 128.9, 128.8, 128.2, 128.1 (2), 114.3, 101.0, 85.1, 81.3, 78.5, 73.8, 72.5, 72.3,
69.8, 55.7.
Anal. Calcd for C27H30O6 (450.20): C, 71.98; H, 6.71. Found: C, 71.87; H,
6.68.
1
19: Rf 0.2 (hexane/EtOAc, 2:1); H NMR (500 MHz, CDCl3, dH) 7.40–7.30 (m, 10
H), 7.24 (d, 1 H, J = 6.9 Hz), 6.96–6.91 (m, 3 H), 5.40–5.35 (m, 1 H), 4.66–4.54 (m,
6 H), 4.52–4.42 (m, 1 H), 4.20–4.12 (m, 1 H), 4.06–3.95 (m, 1 H), 3.86 (s, 3 H),
3.66–3.52 (m, 2 H), 3.40 (br s, 1 H); 13C NMR (125 MHz, CDCl3, dC) 160.0, 159.9,
159.7, 138.5, 138.3, 137.9, 137.8, 133.6, 130.2, 129.9, 129.8, 129.1, 128.9, 128.9,
128.8, 128.4, 128.3 (3), 128.2, 128.20, 128.1, 114.4 (2), 101.6, 96.7, 86.7, 84.1, 83.2,
82.4, 82.3, 80.9, 74.0, 73.78, 72.4 (2), 71.9, 71.0, 70.7, 65.4, 55.7 (2). HRMS (ESI)
Calcd for C27H30O6 [M + Na]+ 473.1940. Found 473.1941.
Allyl 3,5-di-O-benzoyl- -D-arabinofuranoside (20) and 2-O-Allyl-3,5-di-O-
benzoyl-D-arabinofuranose (21). The crude mixture resulting from the reaction of
11 and allyl bromide was purified by chromatography (hexane/EtOAc, 5:1) to give 20
(13%) and 21 (7%) as white solids. 20: Rf 0.4 (hexane/EtOAc, 2:1); mp 60–64°C;
1
[a]D ꢁ 54.1° (c 0.03, CHCl3); H NMR (400 MHz, CDCl3, dH) 8.06–8.04 (m, 4 H),
7.61–7.37 (m, 2 H), 7.47–7.37 (m, 4 H), 5.94–5.84 (m, 1 H), 5.49 (t, 1 H, J = 6.0
Hz), 5.28 (ddd, 1 H, J = 17.2, 2.9, 1.4 Hz), 5.20 (dd, 1 H, J = 10.4 Hz), 5.15 (d, 1
H, J = 4.7 Hz), 4.67 (dd, 1 H, J = 11.7, 4.5 Hz), 4.55 (dd, 1 H, J = 11.7, 6.5 Hz),
4.50 (dd, 1 H, J = 6.5, 4.7 Hz), 4.43–4.39 (m, 1 H), 4.34–4.29 (m, 1 H), 4.12–4.07
(m, 1 H), 2.55 (br s, 1 H); 13C NMR (100 MHz, CDCl3, dC) 166.5, 166.2, 133.5,
133.4, 133.0, 129.9 (2), 129.8, 129.2, 128.5, 128.4, 118.2, 100.5, 79.9, 79.3, 76.7,
69.0, 65.6.
Anal. Calcd for C22H22O7 (398.1366): C, 66.32; H, 5.57. Found: C, 66.08; H, 5.70.
1
21: Rf 0.3 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 8.09–8.04 (m, 4
H), 7.59–7.40 (m, 6 H), 5.94–5.87 (m, 1 H), 5.57–5.20 (m, 4 H), 4.73–4.55 (m, 3 H),
4.26–4.12 (m, 3 H), 3.90 (d, 0.3 H, J = 9.6 Hz), 2.77 (d, 0.7 H, J = 4.2 Hz); 13C NMR
(100 MHz, CDCl3, dC) 166.2, 166.1, 134.1 (2), 133.7, 133.9, 133.5, 130.3, 130.2, 129.0
(2), 128.8 (2), 118.3, 102.1, 88.7, 88.5, 87.7, 87.4, 78.6, 71.4, 65.5, 64.9.
Anal. Calcd for C22H22O7 (398.1366): C, 66.32; H, 5.57. Found: C, 66.06; H, 5.70.
Benzyl 3,5-di-O-benzoyl- -D-arabinofuranoside (22) and 3,5-di-O-benzoyl-2-O-
benzyl-D-arabinofuranose (23). The crude mixture resulting from the reaction of 11
and benzyl bromide was purified by chromatography (hexane/EtOAc, 5:1) to give 22
(34%) and 23 (8%) both as white solids. 22: Rf 0.5 (hexane/EtOAc, 2:1); mp 85–87°C;
1
[a]D ꢁ 48.1° (c 0.04, CHCl3); H NMR (400 MHz, CDCl3, dH) 8.05–8.03 (m, 4 H),
7.59–7.29 (m, 11 H), 5.51 (t, 1 H, J = 6.1 Hz), 5.19 (d, 1 H, J = 4.7 Hz), 4.86 (d, 1

1,2-O-Stannylene Acetals of D-Arabinofuranose
975
H, J = 11.7 Hz), 4.68 (dd, 1 H, J = 11.7, 4.4 Hz), 4.60 (d, 1 H, J = 11.7 Hz), 4.58 (dd,
1 H, J = 11.7, 6.6 Hz), 4.50 (ddd, 1 H, J = 10.1, 5.4, 3.6 Hz), 4.44–4.40 (m, 1 H),
2.94 (d, 1 H, J = 10.3 Hz); 13C NMR (100 MHz, CDCl3, dC) 166.8, 166.6, 137.0,
133.9, 133.5, 130.3 (2), 129.7, 129.0, 128.9, 128.8, 128.7, 128.6, 100.9, 80.2, 79.8,
77.1, 70.5, 66.1.
Anal. Calcd for C26H24O7 (448.1522): C, 69.63; H, 5.39. Found: C, 69.37; H, 5.53.
23: Rf 0.4 (hexane/EtOAc, 2:1); 1H NMR (400 MHz, CDCl3, dH) 8.06–8.03
(m, 4 H), 7.48–7.25 (m, 11 H), 5.60–5.46 (m, 2 H), 4.89–4.56 (m, 5 H), 4.38–
4.34 (m, 0.3 H), 4.20 (br s, 0.4 H), 4.12 (dd, 0.3 H, J = 4.0, 2.7 Hz), 3.90 (br s,
0.3 H), 2.80 (br s, 0.7 H); 13C NMR (100 MHz, CDCl3, dC) 166.8, 166.3, 137.6,
137.0, 134.1, 134.0, 133.4, 130.3, 130.2 (2), 129.7, 129.1, 129.0 (2), 128.9, 128.8,
128.7, 128.6, 128.4, 128.3, 102.1, 97.7, 87.6, 81.9, 81.5, 79.5, 78.5, 77.1, 73.3,
72.5, 65.5, 64.8. HRMS (ESI) Calcd for C26H24O7 [M + Na]+ 471.1320. Found
471.1432.
p-Methoxybenzyl-3,5-di-O-benzoyl- -D-arabinofuranoside (24) and 3,5-Di-O-
benzoyl-2-O-p-methoxybenzyl-D-arabinofuranose (25). The crude mixture resulting
from the reaction of 11 and p-methoxybenzyl chloride was purified by chromatogra-
phy (hexane/EtOAc, 5:1) to give 24 (35%) and 25 (24%) both as white solids. 24: Rf
1
0.3 (hexane/EtOAc, 2:1); mp 60–64°C; [a]D ꢁ 42.2° (c 0.03, CHCl3); H NMR (500
MHz, CDCl3, dH) 8.11–8.09 (m, 4 H), 7.63 (t, 1 H, J = 7.4 Hz), 7.58 (t, 1 H, J = 7.4
Hz), 7.49 (t, 2 H, J = 7.8 Hz), 7.42 (t, 2 H, J = 7.8 Hz), 7.30 (d, 2 H, J = 8.6 Hz),
6.90 (d, 2 H, J = 8.5 Hz), 5.54 (t, 1 H, J = 6.0 Hz), 5.23 (d, 1 H, J = 4.7 Hz), 4.85 (d,
1 H, J = 11.3 Hz), 4.73 (dd, 1 H, J = 11.7, 4.3 Hz), 4.64 (dd, 1 H, J = 11.7, 6.7 Hz),
4.59 (d, 1 H, J = 11.3 Hz), 4.54 (dd, 1 H, J = 6.4, 4.8 Hz), 4.49–4.45 (m, 1 H), 3.84
(s, 3 H), 2.67 (br s, 1 H); 13C NMR (125 MHz, CDCl3, dC) 166.8, 166.6, 160.0, 133.9,
133.4 (2), 130.3, 130.2, 129.7, 129.1, 128.9, 128.8, 114.4, 100.6, 80.3, 79.9, 77.1,
70.2, 66.1, 55.7. HRMS (ESI) Calcd for C27H26O8 [M + Na]+ 501.1525. Found
501.1521.
1
25: Rf 0.2 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 8.06–8.03 (m, 4
H), 7.47–7.26 (m, 8 H), 6.86–6.83 (m, 2 H), 5.56–5.44 (m, 2 H), 4.81–4.55 (m, 4.5
H), 4.37–4.33 (m, 0.5 H), 4.17 (s, 0.6 H), 4.09 (dd, 0.5 H, J = 4.0, 2.6 Hz), 3.96 (br d,
0.4 H J = 8.2 Hz), 3.77 (s, 3 H), 2.92 (br s, 0.5 H); 13C NMR (100 MHz, CDCl3, dC)
166.3, 165.9, 159.4, 133.6, 133.5, 133.1, 133.0, 129.9 (2), 129.8, 129.6, 129.3 (2),
129.2, 128.7, 128.6 (2), 128.5, 128.4, 128.3, 114.1, 113.9, 101.8, 97.3, 86.9, 81.4, 80.7,
79.12, 78.17, 76.73, 72.62, 71.79, 65.1, 64.5, 55.3. HRMS (ESI) Calcd for C27H26O8
[M + Na]+ 501.1525. Found 501.1520.
Allyl 3,5-O-(tetraisopropylsiloxane-1,3-diyl)- -D-arabinofuranoside (26) and 3,5-
O-(tetraisopropylsiloxane-1,3-diyl)-2-O-benzyl-D-arabinofuranose (27). The crude
reaction mixture resulting from the reaction of 12 and allyl bromide was purified by
chromatography (hexane/EtOAc, 6:1) to give 26 (23%) and 27 (36%) as clear oils. 26:
Rf 0.8 (hexane/EtOAc, 2:1); [a]D ꢁ 58.1° (c 0.04, CHCl3); 1H NMR (400 MHz,
CDCl3, dH) 5.94–5.84 (m, 1 H), 5.27 (ddd, 1 H, J = 17.2, 3.1, 1.5 Hz), 5.20 (ddd, 1 H,
J = 10.4, 2.7, 1.3 Hz), 4.90 (d, 1 H, J = 4.7 Hz), 4.26–4.20 (m, 2 H), 4.16–4.10 (m, 1

976
Darwish et al.
H), 4.03–3.99 (m, 1 H), 3.95 (dd, 1 H, J = 10.7, 3.2 Hz), 3.88–3.83 (m, 1 H), 3.78
(dd, 1 H, J = 10.7, 8.8 Hz), 2.30 (d, 1 H, J = 10.3 Hz), 1.11–0.94 (m, 28 H); 13C NMR
(100 MHz, CDCl3, dC) 134.3, 117.9, 99.9, 82.6, 80.0, 79.1, 68.8, 66.4, 18.0, 17.9,
17.80, 17.5, 17.4 (2), 13.9, 13.7, 13.72, 13.3, 12.9.
Anal. Calcd for C20H40O6Si2 (432.24): C, 55.52; H, 9.32. Found: C, 55.67; H, 9.32.
1
27: Rf 0.7 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 5.96–5.85 (m, 1
H), 5.36–5.32 (m, 1 H), 5.32–5.28 (m, 1 H), 5.25–5.18 (m, 1 H), 4.34 (t, 0.5 H,
J = 6.2 Hz), 4.26 (dd, 0.5 H, J = 6.9, 5.0 Hz), 4.19–4.05 (m, 2.5 H), 4.03–3.98 (m, 1
H), 3.93–3.84 (m, 2 H), 3.81–3.77 (m, 0.5 H), 3.32 (d, 0.5, J = 6.2 Hz), 2.82 (d, 0.5
H, J = 5.2 Hz), 1.10–1.02 (m, 28 H); 13C NMR (100 MHz, CDCl3, dC) 134.3, 134.0,
117.9, 117.1, 100.5, 94.3, 89.5, 84.7, 81.8, 81.2, 76.7, 76.5, 72.0, 71.4, 65.5, 62.4, 17.4
(5), 17.1 (4), 17.0, 13.5 (2), 13.1, 13.0, 12.9, 12.6, 12.5. HRMS (ESI) Cald for
C20H40O6Si2 [M + Na]+ 455.2256. Found 455.2236.
Benzyl 3,5-O-(tetraisopropylsiloxane-1,3-diyl)- -D-arabinofuranoside (28) and
3,5-O-(Tetraisopropylsiloxane-1,3-diyl)-2-O-benzyl-D-arabinofuranose (29). The
crude mixture resulting from the reaction of 12 and benzyl bromide was purified by
chromatography (hexane/EtOAc, 6:1) to give 28[43] (36%) as a white solid and 29
(37%) as a clear oil. 28: Rf 0.8 (hexane/EtOAc, 2:1); [a]D ꢁ 113.2° (c 0.02, CHCl3);
1H NMR (400 MHz, CDCl3, dH) 7.38–7.29 (m, 5 H), 4.97 (d, 1 H, J = 4.7 Hz), 4.77
(d, 1 H, J = 11.5 Hz), 4.50 (d, 1 H, J = 11.5 Hz), 4.26 (dd, 1 H, J = 7.4, 6.0 Hz),
4.17–4.12 (m, 1 H), 3.97 (dd, 1 H, J = 10.8, 3.2 Hz), 3.90–3.86 (m, 1 H), 3.81 (dd, 1
H, J = 10.8, 8.7 Hz), 2.30 (d, 1 H, 10.5 Hz), 1.09–0.94 (m, 28 H); 13C NMR (100
MHz, CDCl3, dC) 137.1, 128.5, 128.2, 128.0, 99.6, 82.2, 79.6, 78.7, 69.6, 66.0, 17.6,
17.5, 17.4 (2), 17.2, 17.0 (3), 13.5, 13.3, 12.9, 12.5.
Anal. Calcd for C24H42O6Si2 (482.25): C, 59.71; H, 8.77. Found: C, 59.95; H, 8.77.
1
29: Rf 0.7 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 7.36–7.30 (m, 5
H), 5.34 (dd, 0.7 H, J = 6.6, 2.5 Hz), 5.23 (dd, 0.3 H, J = 7.7, 5.5 Hz), 4.75–4.60 (m, 2
H), 4.43 (t, 0.3 H, J = 7.6 Hz), 4.34 (dd, 0.7 H, J = 8.5, 6.0 Hz), 4.15–4.09 (m, 0.8 H),
4.06 (d, 0.3 H), 4.03–3.98 (m, 1.7 H), 3.92–3.86 (m, 1 H), 3.86–3.80 (m, 0.2 H), 3.33
(d, 0.3 H, J = 6.2 Hz), 2.83 (d, 0.7 H, J = 5.3 Hz). 1.10–0.91 (m, 28 H); 13C NMR
(100 MHz, CDCl3, dC) 138.2, 137.7, 128.9, 128.8, 128.5, 128.3, 128.1 (2), 100.9, 94.7,
89.9, 85.1, 82.5, 81.8, 77.8, 77.1, 73.2, 72.7, 66.0, 62.9, 17.9, 17.8 (3), 17.5 (4), 17.4
(2), 13.9 (2), 13.7, 13.5, 13.3, 13.0. HRMS (ESI) Calcd for C24H42O6Si2 [M + Na]+
505.2412. Found 505.2432.
p-Methoxybenzyl 3,5-O-(tetraisopropylsiloxane-1,3-diyl)- -D-arabinofuranoside
(30) and 3,5-O-(Tetraisopropylsiloxane-1,3-diyl)-2-O-p-methoxybenzyl-D-arabino-
furanose (31). The crude mixture resulting from the reaction of 12 and p-methox-
ybenxyl chloride was purified by chromatography (hexane/EtOAc, 6:1) to give 30
(45%) and 31 (36%) as clear oils. 30: Rf 0.7 (hexane/EtOAc, 2:1); [a]D ꢁ 80.9° (c
1
0.09, CHCl3); H NMR (400 MHz, CDCl3, dH) 7.26–7.23 (m, 2 H), 6.89–6.87 (m, 2
H), 4.95 (d, 1 H, J = 4.7 Hz), 4.70 (d, 1 H, J = 11.2 Hz), 4.43 (d, 1 H, J = 11.2 Hz),
4.23 (dd, 1 H, J = 7.3, 5.9 Hz), 4.15–4.09 (m, 1 H), 3.97 (dd, 1 H, J = 10.7, 3.2 Hz),

1,2-O-Stannylene Acetals of D-Arabinofuranose
977
3.89–3.85 (m, 1 H), 3.81 (s, 3 H), 3.80 (dd, 1 H, J = 10.7, 8.8 Hz), 2.28 (d, 1 H,
J = 10.4 Hz), 1.10–0.95 (m, 28 H); 13C NMR (100 MHz, CDCl3, dC) 159.9, 130.3,
129.6, 114.3, 99.7, 82.6, 80.1, 79.1, 69.7, 66.5, 55.7, 18.0, 17.8 (2), 17.4 (3), 13.9, 13.7,
13.3, 12.9.
Anal. Calcd for C25H44O7Si2 (512.26): C, 58.56; H, 8.65. Found: C, 58.38; H,
8.59.
1
31: Rf 0.5 (hexane/EtOAc, 2:1); H NMR (400 MHz, CDCl3, dH) 7.29–7.25 (m, 2
H), 6.89–6.86 (m, 2 H), 5.30 (dd, 0.6 H, J = 6.6, 2.4 Hz), 5.16 (t, 0.4 H, J = 3.8 Hz),
4.65–4.55 (m, 2 H), 4.37 (t, 0.4 H, J = 7.6 Hz), 4.29 (dd, 0.6 H, J = 8.4, 6.0 Hz),
4.09–4.03 (m, 1 H), 4.01–3.97 (m, 1 H), 3.95–3.93 (m, 1 H), 3.88–3.83 (m, 1 H),
3.81 (s, 3 H), 3.33 (d, 0.4 H, J = 6.2 Hz), 2.83 (d, 0.6 H, J = 5.4 Hz), 1.08–0.94 (m,
28 H); 13C NMR (100 MHz, CDCl3, dC) 160.0, 159.7, 130.2, 130.1, 129.8, 114.3,
114.2, 101.0, 94.8, 89.6, 84.9, 82.7, 81.8, 77.9, 77.2, 77.1, 73.0, 72.5, 65.9, 63.1, 55.7
(2), 18.0, 17.9, 17.8 (3), 17.6, 17.5 (2) 17.4, 13.9, 13.7 (2), 13.4, 13.0, 12.9. HRMS
(ESI) Calcd for C25H44O7Si2 [M + Na]+ 535.2518. Found 535.2494.
Methyl 2,3-anhydro-5-O-toluenesulfonyl- -D-lyxofuranoside (34). Alcohol 32
(0.90 g, 6.2 mmol) and p-TsCl (2.41 g, 12.7 mmol) were dissolved in pyridine and
stirred overnight at room temperature. The mixture was diluted with CH2Cl2, washed
with water, and the organic layer was dried (MgSO4) and concentrated. Column
chromatography (hexane/EtOAc, 2:1) of the residue gave 34 (1.55 g, 83%) as a white
solid: Rf 0.4 (hexane/EtOAc, 2:1); mp 74–76°C.[44] 1H NMR (400 MHz, CDCl3, dH)
7.81 (d, 2 H, J = 8.3 Hz), 7.36 (d, 2 H, J = 8.0 Hz), 4.88 (s, 1 H), 4.21 (t, 1 H, J = 6.3
Hz), 4.14–4.12 (m, 2 H), 3.72 (d, 1 H, J = 2.8 Hz), 3.64 (d, 1 H, J = 2.8 Hz), 3.37 (s,
3 H), 2.45 (s, 3 H); 13C NMR (100 MHz, CDCl3, dC) 145.1, 132.6, 130.0, 128.0, 102.3,
73.7, 67.5, 56.5, 55.7, 53.7, 21.7.
Methyl 2,3-anhydro-5-deoxy-5-iodo- -D-lyxofuranoside (35). Tosylate 34
(1.18 g, 3.9 mmol) and n-Bu4I (3.03 g, 8.2 mmol) were dissolved in toluene (20 mL)
and heated at reflux for 7 h. The mixture was concentrated and the residue was
dissolved in CH2Cl2, washed with water and a saturated aqueous solution of NaHSO3.
The organic layer was dried (MgSO4) and concentrated. Column chromatography
(hexane/EtOAc, 4:1) of the residue gave 35 (0.86 g, 86%) as a white solid: Rf 0.6
1
(hexane/EtOAc, 2:1). [a]D + 68.7° (c 0.2, CHCl3); mp 74–76°C; H NMR (400 MHz,
CDCl3, dH) 5.05 (s, 1 H), 4.28 (t, 1 H, J = 7.2 Hz), 3.90 (d, 1 H, J = 2.8 Hz), 3.68
(d, 1 H, J = 2.8 Hz), 3.42 (s, 3 H), 3.24 (s, 1 H), 3.23 (d, 1 H); 13C NMR (100 MHz,
CDCl3, dC) 102.3, 76.7, 56.3, 55.0, 54.4, 0.6. HRMS (ESI) Calcd for C6H9IO3
[M + Na]+ 278.9489. Found 278.9481.
Methyl 2,3-anhydro-5-O-formyl- -D-lyxofuranoside (37). A mixture of 12
(0.15 g, 0.38 mmol) and dibutyltin oxide (0.10 g, 0.41 mmol) in toluene (12 mL) was
heated at reflux for 20 min. Triflate 33 (0.31 g, 1.12 mmol) was added and stirring
continued at reflux for 6 h before DMF (2 mL) was added. After 1 h, a new product
was detected by TLC, and the reaction mixture was then cooled and concentrated. The
residue was dissolved in CH2Cl2, washed with water, and the organic layer was dried

978
Darwish et al.
(Na2SO4), and concentrated. The residue was purified by chromatography (hexane/
EtOAc, 6:1) to give 38 (77 mg, 39%) as a white solid: Rf 0.4 (hexane/EtOAc, 2:1);
1
[a]D + 13.8° (c 0.02, CHCl3); H NMR (400 MHz, CDCl3, dH) 8.10 (d, 1 H, J = 0.7
Hz,), 4.98 (s, 1 H), 4.37 (d, 2 H, J = 6.2 Hz), 4.23 (t, 1 H, J = 5.9 Hz), 3.74 (dd, 1 H,
J = 2.8, 0.7 Hz), 3.68 (d, 1 H, J = 2.8 Hz), 3.43 (s, 3 H); 13C NMR (100 MHz, CDCl3,
dC) 160.9, 102.8, 74.2, 62.5, 56.6, 56.2, 54.2. HRMS (ESI) Calcd for C7H10O5
[M + Na]+ 197.0420. Found 197.0424.
Computational studies on 43. Because many force fields for stochastic searching
do not effectively handle tin as an element, a surrogate approach using silicon as a mimic
was employed. Thus, the SPMC search protocol available in MacroModel Version 6.5
was used to generate an initial family of 10,000 conformers of 44. Each conformer was
then minimized in the gas phase using AM1,[40] which generated 113 unique conformers,
all of which were within 32 kcal/mol of the global minimum. Two families of structures
were present. One family (four conformers) was clustered at the global minimum and the
other (6 conformers) ꢀ 1.1 kcal/mol above the global minimum. Within each family of
structures, the furanose ring adopted the same conformation, and the geometrical dif-
ferences were only in the orientations about the bonds in the siloxane moeity. The two
lowest energy conformers in each family were subsequently optimized at the B3LYP
level of theory after replacement of the surrogate silicon atom with tin. The basis set
used for the full geometry optimization was 3-21G* for silicon, oxygen, carbon, and
hydrogen and the basis set for tin was the SDD effective core potential.[45] Single-point
energy calculations were subsequently performed at the B3LYP level of theory[46–48]
using a 6-31G* basis set for silicon, oxygen, carbon, and hydrogen and the SDD ef-
fective core potential for tin. All DFT calculations utilized 6 Cartesian d functions and
were performed using Gaussian 98.[49]
ACKNOWLEDGMENTS
This work was supported by The National Institutes of Health and the Ohio Su-
percomputer Center. OSD thanks the Ohio State University College of Arts and Science
for a senior Honors Thesis Scholarship.
REFERENCES
1. Brennan, P.J.; Nikaido, H. The envelope of mycobacteria. Ann. Rev. Biochem.
1995, 64, 29.
2. Lowary, T.L. Mycobacterial cell wall components. In Glycoscience: Chemistry and
Chemical Biology; Fraser-Reid, B., Tatsuta, K., Thiem, J., Eds.; Springer-Verlag:
Berlin, 2001; 2005–2080.
3. Lowary, T.L. Recent progress towards the identification of inhibitors of
mycobacterial cell wall polysaccharide biosynthesis. Mini-Rev. Med. Chem.
2003, 3, 694.

1,2-O-Stannylene Acetals of D-Arabinofuranose
979
4. Crick, D.C.; Mahapatra, S.; Brennan, P.J. Biosynthesis of the arabinogalactan–
peptidoglycan complex of Mycobacterium tuberculosis. Glycobiology 2001, 11,
107R.
5. Mereyala, H.B.; Hotha, S.; Gurjar, M.K. Synthesis of pentaarabinofuranosyl
structure motif A of Mycobacterium tuberculosis. J. Chem. Soc., Chem. Commun.
1998, 685.
6. De´sire´, J.; Prandi, J. Synthesis of methyl b-D-arabinofuranoside 5-[1-D (and L)-myo-
inositol 1-phosphate], the capping motif of the lipoarabinomannan of Mycobacte-
rium smegmatis. Carbohydr. Res. 1999, 317, 110.
7. Sanchez, S.; Bamhaoud, T.; Prandi, J. Elaboration of monoarabinofuranosidic
building blocks. Eur. J. Org. Chem. 2002, 22, 3864.
8. Subramaniam, V.; Lowary, T.L. Synthesis of oligosaccharide fragments of
mannosylated lipoarabinomannan from Mycobacterium tuberculosis. Tetrahedron
1999, 55, 5965.
9. Yin, H.; D’Souza, F.W.; Lowary, T.L. Arabinofuranosides from mycobacteria:
synthesis of a highly branched hexasaccharide and related fragments containing b-
arabinofuranosyl residues. J. Org. Chem. 2002, 67, 892.
10. Gadikota, R.R.; Callam, C.S.; Wagner, T.; Del Fraino, B.; Lowary, T.L. 2,3-
Anhydrosugars in glycoside bond synthesis. Highly stereocontrolled syntheses of
oligosaccharides containing a- and b-arabinofuranosyl linkages. J. Am. Chem. Soc.
2003, 125, 4155.
11. Ayers, J.D.; Lowary, T.L.; Morehouse, C.B.; Besra, G.S. Synthetic arabinofur-
anosyl oligosaccharides as mycobacterial arabinosyltransferase substrates. Bioorg.
Med. Chem. Lett. 1998, 8, 437.
12. Kaur, D.; Lowary, T.L.; Vissa, V.D.; Crick, D.C.; Brennan, P.J. Characterization of
the epitope of anti-lipoarabinomannan antibodies as the terminal hexaarabinofur-
anosyl motif of mycobacterial arabinans. Microbiology UK 2002, 148, 3059.
13. Pathak, A.K.; Pathak, V.; Suling, W.J.; Gurcha, S.S.; Morehouse, C.B.; Besra, G.S.;
Maddry, J.A.; Reynolds, R.C. Studies on n-octyl-5-(a-D-arabinofuranosyl)-b-D-
galactofuranosides for mycobacterial glycosyltransferase activity. Bioorg. Med.
Chem. 2002, 10, 923.
14. Pathak, A.K.; Pathak, V.; Maddry, J.A.; Suling, W.J.; Gurcha, S.S.; Besra, G.S.;
Reynolds, R.C. Studies on a-(1 ! 5) linked octyl arabinofuranosyl disaccharides
for mycobacterial arabinosyltransferase activity. Bioorg. Med. Chem. 2001, 9,
3145.
15. Pathak, A.K.; Pathak, V.; Seitz, L.; Maddry, J.A.; Gurcha, S.S.; Besra, G.S.; Suling,
W.J.; Reynolds, R.C. Studies on b-(1 ! 5) and b-(1 ! 6) linked octyl Galf
disaccharides as substrates for mycobacterial galactosyltransferase activity. Bioorg.
Med. Chem. 2001, 9, 3129.
16. Gridley, J.J.; Osborn, H.M.I. Recent advances in the construction of b-D-mannose
and b-D-mannosamine linkages. J. Chem. Soc., Perkin Trans. 1 2000, 1471.
17. Pozsgay, V. Stereoselective synthesis of b-D-mannosides. In Carbohydrates in
Chemistry and Biology; Ernst, B., Hart, G.W., Sinay¨, P., Eds.; Wiley-VCH:
Weinheim, 2000.
18. Oscarson, S.; Sehgelmeble, F.W. Chemical syntheses of inulin and levan structures.
J. Org. Chem. 2002, 67, 8457.

980
Darwish et al.
19. Oscarson, S.; Sehgelmeble, F.W. A novel b-directing fructofuranosyl donor
concept. Stereospecific synthesis of sucrose. J. Am. Chem. Soc. 2000, 122,
8869.
20. Krog-Jensen, C.; Oscarson, S. Stereospecific synthesis of b-D-fructofuranosides
using thioglycoside donors and internal aglycon delivery. J. Org. Chem. 1998, 63,
1780.
21. Hodosi, G.; Kova´c, P. A fundamentally new, simple, stereospecific synthesis of
oligosaccharides containing the b-mannopyranosyl and b-rhamnopyranosyl linkage.
J. Am. Chem. Soc. 1997, 119, 2335.
22. Hodosi, G.; Kova´c, P. Glycosylation via locked anomeric configuration:
stereospecific synthesis of oligosaccharides containing the b-D-mannopyranosyl
and b-L-rhamnopyranosyl linkage. Carbohydr. Res. 1998, 308, 63.
23. Srivastava, V.K.; Schuerch, C. Synthesis of b-D-mannopyrabnosides and regio-
selective O-alkylation of dibutylstannylene complexes. Tetrahedron Lett. 1979,
3269.
24. Dessinges, A.; Olesker, A.; Lukacs, G.; Thang, T.T. Carbohydrate synthesis for
nuclear medicine—a new, rapid, and stereospecific route to 2-deoxy-2-fluoro-D-
glucose. Carbohydr. Res 1984, 126, C6.
25. Su, T.-S.; Klein, R.A.; Fox, J.J. Improved synthesis of a-D-ribofuranosides via
stereoselective alkylation of a dibutylstannylene derivative for ready access to the
2-substituted 2-deoxyarabinofuranosides. J. Org. Chem. 1982, 47, 1506.
26. Ning, J.; Kong, F. Synthesis and glycosidic coupling reaction of substituted 2,6-
dioxabicyclo[3.1.0]hexanes: 1,2-anhydro-3,5-di-O-benzyl-a-D-ribofuranose. Carbo-
hydr. Res. 1997, 300, 355.
27. Reist, E.J.; Hart, P.A.; Goodman, L.; Baker, B.R. Potential anticancer agents.
XXIII. The anomeric ethyl 1-thio-D-arabinofuranosides. J. Am. Chem. Soc. 1959,
81, 5176.
28. D’Souza, F.W.; Ayers, J.D.; McCarren, P.R.; Lowary, T.L. Arabinofuranosyl
oligosaccharides from mycobacteria: synthesis and effect of glycosylation on ring
conformation and hydroxymethyl group rotamer populations. J. Am. Chem. Soc.
2000, 122, 1251.
29. Mizutani, K.; Kasai, R.; Nakamura, M.; Tanaka, O.; Matsuura, H. NMR spectral
study of a- and b-L-arabinofuranosides. Carbohydr. Res. 1989, 185, 27.
30. Gent, P.A.; Gigg, R. The allyl ether as a protecting group in carbohydrate
chemistry: part VIII. Synthesis of benzyl and allyl ethers of D-glucopyranose.
Carbohydr. Res. 1976, 49, 325.
31. Grindley, T.B. Applications of tin-containing intermediates to carbohydrate
chemistry. Adv. Carbohydr. Chem. Biochem. 1998, 53, 17.
32. David, S.; Hanessian, S. Regioselective manipulation of hydroxyl groups via
organotin derivatives. Tetrahedron 1985, 41, 643.
33. Wagner, D.; Verheyden, J.P.H.; Moffatt, J.G. Preparation and synthetic utility of
some organotin derivatives of nucleosides. J. Org. Chem. 1974, 39, 24.
34. Grouiller, A.; Buet, V.; Uteza, V.; Descotes, G. Novel p-toluenesulfonylation and
thiocarbonylation of unprotected thymine nucleosides. Synlett 1993, 221.
35. Gough, G.R.; Miller, T.J.; Mantick, N.A. p-Nitrobenzyloxymethyl: a new fluoride-
removable protecting group for ribonucleoside 2’-hydroxyls. Tetrahedron Lett.
1996, 37, 981.

1,2-O-Stannylene Acetals of D-Arabinofuranose
981
36. Labanowski, J.W.; Andzelm, J. Density Functional Methods in Chemistry;
Springer: New York, 1991.
37. Parr, R.G.; Yang, W. Density Functional Theory in Atoms and Molecules; Oxford
University Press: New York, 1989.
38. Mohamadi, F.; Richards, N.G.J.; Guida, W.C.; Liskamp, R.; Lipton, M.; Caulfield,
C.; Chang, G.; Hendrickson, T.; Still, W.C. Macromodel—an integrated software
system for molecular modeling organic and bioorganic molecules using molecular
mechanics. J. Comput. Chem. 1990, 11, 440.
39. Goodman, J.; Still, W.C. An unbounded systematic search of conformational space.
J. Comput. Chem. 1991, 12, 1110.
40. Hehre, W.J.; Radom, L.; Schleyer, P.v.R.; Pople, J.A. Ab initio Molecular Orbital
Theory; John Wiley & Sons: New York, 1986.
41. Altona, C.; Sundaralingam, M. Conformational analysis of the sugar ring in
nucleosides and nucelotides. A new description using the concept of pseudorota-
tion. J. Am. Chem. Soc. 1972, 94, 8205.
42. Martin, M.G.; Ganem, B.; Rasmussen, J.R. An improved synthesis of methyl 2,3-
anhydro-a-D-lyxofuranoside. Carbohydr. Res. 1983, 123, 332.
43. Kakinuma, H.; Yuasa, H.; Hashimoto, H. Glycosyltransfer mechanism of a-
glucosyltransferase from Protaminobacter rubrum. Carbohydr. Res. 1998, 312,
103.
44. Unger, F.M.; Christian, R.; Waldstaetten, P. Regiospecific syntheses of azidodeoxy
and diazidodideoxy derivatives of methyl a-D-arabinofuranoside. Carbohydr. Res.
1979, 69, 71.
45. Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Energy-adjusted ab
initio pseudopotentials for the second and third row transition elements. Theor.
Chim. Acta 1990, 77, 123.
46. Becke, A.D. Density functional exchange energy approximation with correct
asymptotic behavior. Phys. Rev., A 1988, 38, 3098.
47. Becke, A.D. Density functional thermochemistry. 3. The role of exact exchange. J.
Chem. Phys. 1993, 98, 5648.
48. Lee, C.; Yang, W.; Parr, R.G. Development of the Colle–Salvetti correlation
energy formula into a functional of the electron density. Phys. Rev., B 1988, 37,
785.
49. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheese-
man, J.R.; Zakrzewski, V.G.; Montgomery, J.A., Jr.; Stratmann, R.E.; Burant, J.C.;
Dapprich, S.; Millam, J.M.; Daniels, A.D.; Kudin, K.N.; Strain, M.C.; Farkas, O.;
Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J.; Petersson, G.A.; Ayala, P.Y.; Cui, Q.; Morokuma, K.;
Malick, D.K.; Rabuck, A.D.; Raghavachari, K.; Foresman, J.B.; Cioslowski, J.;
Ortiz, J.V.; Stefanov, B.B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R.L.; Fox, D.J.; Keith, T.; Al-Laham, M.A.; Peng, C.Y.;
Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P.M.W.; Johnson, B.; Chen,
W.; Wong, M.W.; Andres, J.L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E.S.;
Pople, J.A. Gaussian 98, Revision A.9; Gaussian, Inc.: Pittsburgh, PA, 1998.
Received May 20, 2003
Accepted August 7, 2003