Stereoselectivity in the Lewis acid mediated reduction of ketofuranoses

Article abstract of DOI:10.1021/acs.joc.5b00419

The Lewis acid mediated reduction of ribose-, arabinose-, xylose-, and lyxose-derived methyl and phenyl ketofuranoses with triethylsilane as nucleophile was found to proceed with good to excellent stereoselectivity to provide the 1,2-cis addition products

Full text of DOI:10.1021/acs.joc.5b00419

Article
pubs.acs.org/joc
Stereoselectivity in the Lewis Acid Mediated Reduction of
Ketofuranoses
Erwin R. van Rijssel,§ Pieter van Delft,§ Daan V. van Marle, Stefan M. Bijvoets, Gerrit Lodder,
Herman S. Overkleeft, Gijsbert A. van der Marel, Dmitri V. Filippov,* and Jeroen D. C. Codee*
́
Leiden Institute of Chemistry, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlands
S
* Supporting Information
ABSTRACT: The Lewis acid mediated reduction of ribose-,
arabinose-, xylose-, and lyxose-derived methyl and phenyl
ketofuranoses with triethylsilane as nucleophile was found to
proceed with good to excellent stereoselectivity to provide the
1,2-cis addition products. The methyl ketoses reacted in a more
stereoselective manner than their phenyl counterparts. The
stereochemical outcome of the reactions parallels the relative
stability of the oxocarbenium ion conformers involved, as
assessed by calculating the free energy surface maps of their complete conformational space. The Lewis acid mediated reduction
allows for a direct synthesis of C-glycosides with predictable stereochemistry.
Presumably, the stereoselectivity is the result of the structure
and particular reactivity of the intermediate in the process: the
furanosyl oxocarbenium ion. Woerpel and co-workers have
devised a two-conformer model in which the equilibrium
INTRODUCTION
■
C-Glycosides are carbohydrate-derived molecules in which the
anomeric exocyclic oxygen is substituted for a carbon
functionality. C-Glycosides are often used as metabolically
stable mimics of naturally occurring O- and N-glycosides, and
they can have appealing biological properties, ranging from
antibiotic and antiviral activity to anticancer activity.1 Examples
of naturally occurring C-glycosides are the C-nucleoside
pseudouridine1a−d and the C-arylglycoside dapagliflozin.1g
The latter compound is the C-glycoside analogue of the
naturally occurring O-glycoside phlorizin, and it is currently
used in the clinic for the treatment of type 2 diabetes.1g Because
of their attractive biological properties, various approaches have
been developed for their synthesis. A commonly used strategy
to synthesize C-glycosides entails the reaction of glycosyl
lactones with an alkyl or aryl nucleophile to give glycosyl
hemiketals that are subsequently reduced under Lewis acidic
conditions to provide the target compounds.1a,b,2 In this
process, a new stereogenic center is created and control over
the stereochemical course of the reaction is required to enable a
productive and effective overall transformation. A notable
example is the reduction of perbenzylated ribofuranose ketoses
that provides β-linked products in a highly stereoselective
fashion.2h,3 Scheme 1 depicts a synthesis of pseudouridine that
builds on the stereoselective reduction of perbenzylated
riboketosides. Reaction of ribonolactone 1 with lithiated
pyrimidine 2 gave hemiketal 3 that was reduced with
triethylsilane (TES) using BF3·OEt2 as the Lewis acid. After
removal of the tert-butyl protecting groups with trifluoroacetic
acid (TFA), the desired 1,2-trans compound 4 was obtained as
a single anomer. The stereoselectivity in the reduction is
striking because the nucleophile comes in cis with respect to the
substituents at C2 and C3.
3
between the E and E3 furanosyl oxocarbenium ion envelope
conformers is decisive in determining the stereoselectivity of
reactions involving these intermediates (Figure 1).4 Nucleo-
philic attack on either of the envelopes occurs on the inside to
avoid eclipsing interactions with the substituent at C2 and to
provide a product featuring a staggered C1−C2 conformation.
Reaction on the 3E conformer therefore occurs on the top face,
where the E3 envelope is approached from the bottom face.
The equilibrium between the two envelopes is determined by
the nature and orientation of the substituents on the
tetrahydrofuran ring. Alkoxy substituents at C2 preferentially
take up a pseudoequatorial position to allow hyperconjugative
stabilization of the oxocarbenium ion by the pseudoaxial C2−
H2 bond, while C3-alkoxy substituents are preferentially
positioned in a pseudoaxial fashion to allow the donation of
electron density into the electron-depleted anomeric center.4d,5
In our previous work, we established the relative stability of
fully decorated aldofuranosyl oxocarbenium ions.6 A complete
survey of the energy landscape of the entire conformational
space for the four possible pentoaldoses (ribose, arabinose,
xylose, and lyxose) provided a clear picture how the three-ring
substituents on the furanosyl oxocarbenium ion affect
oxocarbenium ion stability and, therefore, their reactivity in
addition reactions. Lewis acid mediated substitution reactions
of the four possible perbenzylated aldofuranosyl acetates using
triethylsilane-d as a nucleophile all proceeded in a 1,2-cis-
selective manner. These results parallel the theoretically
Received: February 23, 2015
Published: March 31, 2015
© 2015 American Chemical Society
4553
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Scheme 1. Diastereoselective Synthesis of Protected Pseudouridine 4
Figure 1. Two-conformer model. Inside attack on either of the two oxocarbenium ions leads to the two epimeric products.
Scheme 2. Synthesis of the Perbenzylated Ribose, Arabinose, and Xylofuranosyl Hemiketals, Lyxose-Configured Ketones, and
Their Ensuing Reduction Reactions
predicted relative stability of the intermediate furanosyl
oxocarbenium ion conformers.
Given the enhanced stability of tertiary cations over their
secondary counterparts, it is likely that ketofuranosyl
oxocarbenium ions are involved in the reduction of the C-
furanoses described above. To establish how a substituent at
the anomeric center of ketofuranoses influences the stability of
the corresponding oxocarbenium ions and their nucleophilic
addition reactions, we have investigated the reduction of the
methyl and phenyl C-furanoses of four possible pentose
configurations, and we have mapped the conformational energy
landscape of the intermediate ketofuranosyl oxocarbenium ions.
4554
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Table 1. Results of the Lewis Acid Mediated Reductions
a
Ratio determined by 1H NMR spectroscopy; stereochemistry was established using 2J coupling constants measured from HSQC-HECADE NMR
spectra.7 The 1,2-cis addition of TES-D to aldoses affords the cis product, while addition of TES in ketose gives the trans product because of
differing priorities of the anomeric substituents. Yield of isolated furanosides after column chromatography.
b
c
The six ketofuranoses 8−13 and the two linear chain lyxose-
derived ketones 16 and 17 were subjected to a Lewis acid
mediated reduction protocol using triethylsilane as reducing
agent and boron trifluoride diethyl etherate as Lewis acid in
dichloromethane at −78 °C. The results of these reactions are
summarized in Table 1, in combination with the results
obtained in the experiments with the corresponding aldoses
34−37 (see Figure 2), in which furanosyl acetates were treated
with TES-D and trimethylsilyl trifluoromethanesulfonate
(TMSOTf).6 In all cases, preferentially the 1,2-trans C-
furanosides were formed, corresponding to an attack of the
hydride cis with respect to the substituent at C2. These results
closely resemble those obtained for the aldoses 34−37. In the
case of ribose and lyxose, the reactions were completely
RESULTS AND DISCUSSION
■
The synthesis of the furanosyl lactols used in this study and the
Lewis acid mediated reduction reactions are depicted in
Scheme 2. The ribose-, arabinose-, and xylose-derived
perbenzylated methyl and phenyl ketofuranose starting
materials were prepared from the perbenzylated lactones.
Ribonolactone 1, arabinonolactone 5, and xylonolactone 6 were
reacted with methyl lithium or phenyl lithium to yield the
methyl and phenyl ketofuranoses that were directly used in the
reduction step. Subjection of perbenzylated lyxonolactone 7 to
methyl or phenyl lithium did not yield the desired
ketofuranoses but only led to furan 14, as the result of a
double β-elimination sequence. Because furan formation could
not be suppressed, we decided to generate the lyxose
oxocarbenium ions in situ in the projected reduction reactions
from their linear precursors 16 and 17. These were synthesized
from Weinreb amide 15 that was generated from lactone 7
using N,O-dimethylhydroxylamine together with trimethylalu-
minum, followed by capping of the alcohol function with
triethylsilyl chloride (TES-Cl). Reaction of Weinreb amide 15
with methyl or phenyl lithium delivered the open chain ketones
16 and 17.
Figure 2. Reference aldose furanosyl donors used in our previous
study.6
4555
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Figure 3. (a) Investigated furanosyl oxocarbenium ions. (b) Pseudorotational circle describing the conformational space that a five-membered ring
3
can occupy. (c) Possible rotamers around the C4−C5 bond in the E (top) and E3 (bottom) envelope conformers.
stereoselective and the products were obtained as a single
diastereomer, independent of the substituent at the anomeric
position (entries 1−3 and 10−12). In the arabinose series, both
the aldose 35 and methyl ketofuranose 10 yielded a single
product, whereas the reduction of phenyl ketofuranose 11
proceeded with a somewhat smaller stereoselectivity (entries
4−6). The xylo-aldofuranose 36 and xylo-methyl ketofuranose
12 yielded the 1,2-cis addition products with a similar
preference,2b where the phenyl ketofuranose 13 displayed the
lowest stereoselectivity (entries 7−9).
In order to clarify the stereoselectivity observed in the
reduction of the ketofuranoses and to assess the influence of
the substituent (R = CH3 or Ph) on the anomeric position on
the intermediate oxocarbenium ions, the energy landscapes of
the permethylated ketofuranosyl oxocarbenium ions 43, 44, 46,
47, 49, 50, 52, and 53 (Figure 3a) were calculated. For this
purpose, we used the free energy surface (FES) mapping
method, as introduced by Rhoad and co-workers8 and adapted
by us to interrogate aldofuranosyl oxocarbenium ions.6 In this
method, the energy associated with the complete conforma-
tional space is calculated, and mapped on a circular graph, the
pseudorotational circle (see Figure 3b).9 Each conformation is
described by a phase angle (P) that defines the shape of the ring
and by the puckering amplitude (τm) that indicates how far out
of the median plane the outlying atoms are positioned.10 The
energies of 81 fixed ring conformers were calculated with
Gaussian 03,11 employing the B3LYP density functional and the
6-311G** basis set, and these were corrected for the solvent
CH2Cl2 using the polarizable continuum model function.
Because rotation of the C4−C5 bond significantly influences
the stability of the furanosyl oxocarbenium ions,6 the FES of
the oxocarbenium ions was scanned for the three individual gg,
gt, and tg C4−C5 rotamers (Figure 3c). Thus, for each
furanosyl oxocarbenium ion, 243 (3 × 81) conformers were
optimized and their associated energies determined. For
comparison, also the FES maps of the permethylated
aldofuranosyl oxocarbenium ions 42, 45, 48, and 51 are
presented.6
Ribose. In Figure 4, the FES maps for the ribofuranosyl
(42), methyl ribofuranosyl (43), and phenyl ribofuranosyl (44)
oxocarbenium ions are depicted. In the columns from left to
right, the gg, gt, tg, and the global lowest FES maps are
displayed. The global FES map is composed by taking the
absolute lowest energy of the gg, gt, and tg rotamers for a
particular ring conformation and plotting these in a single map.
The maps indicate that the ribofuranosyl E3 oxocarbenium ion
3
is highly preferred over the E conformer, independent of the
nature of the anomeric substituent.12 In the E3 conformers, the
C2 and C3 methyloxy groups adopt a pseudoequatorial and
pseudoaxial position, respectively, maximizing their stabilizing
effect on the oxocarbenium ion. In all oxocarbenium ions (42−
44), the C4−C5 gg rotamer is more stable than its gt
counterpart, which in turn is preferred over the tg conformer.
Small differences are present in the maps for the three anomeric
substituents (R = H, Me, and Ph). The energy maps of the
methyl furanosyl oxocarbenium ion conformers are somewhat
steeper than the maps of its unsubstituted ion congener. The
phenyl ribofuranosyl oxocarbenium ion maps, on the other
3
hand, show a smaller energy difference between the E gg and
E3 gg conformer, but also in this case, the energy difference is
4556
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Figure 4. The gg, gt, tg, and global FES maps of the ribofuranosyl oxocarbenium ions 42−44.
Figure 5. Conformation of the anomeric substituents relative to the oxocarbenium ion for (a) methyl and (b) phenyl. (c) Rotation energy profile of
the anomeric methyl substituent for the ribofuranosyl oxocarbenium ion 43. (d) Steric interactions in two conformers of phenyl-substituted
oxocarbenium ion 44.
sufficiently large (∼2.7 kcal mol−1) to account for the selectivity
in the reaction (98:<2, Table 1, entry 3).
methyl substituent next to a carbonyl group.13 An analogous
analysis for the phenyl ketofuranose reveals that the phenyl
substituent is positioned parallel to the CO+ plane (Figure
5b), allowing effective conjugative stabilization of the positive
charge. In the E3 conformation, this planar constellation does
lead to an unfavorable 1,3-allylic interaction between the phenyl
ring and the substituent on C2 (Figure 5d). This allylic strain
destabilizes the E3 conformer and thereby decreases the energy
A closer look at the structures associated with the energies in
the FES maps of the methyl ketofuranosyl ion 43 reveals that
the methyl substituent is placed in an eclipsed conformation
with respect to the plane of the oxocarbenium ion [CO+]
function (Figure 5a). To investigate the influence of the
rotation of the methyl substituent around the C1−methyl
bond, the energy profile associated with this rotation was
calculated. As can be seen in Figure 5c, the eclipsed
conformation is the most stable one, and the perpendicular
structure (ϕ = 90°) that allows optimal hyperconjugative
stabilization from a β-C−H bond is 0.4 kcal mol−1 higher in
energy. This resembles the energy profile for the position of a
3
difference with the E conformers. This, in combination with
the conjugative stabilization by the phenyl substituent, leads to
flattening of the FES maps for the phenyl ketofuranoses.
The FES maps for the nonsubstituted (R = H) and the
methyl- and phenyl-substituted ribofuranosyl oxocarbenium
ions 42−44 show that a single envelope is preferred in all three
4557
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Figure 6. The gg, gt, tg, and global FES maps of the arabinofuranosyl oxocarbenium ions 45−47.
cases, that is, the E3 conformer. Inside attack of the nucleophile
(TES) on this conformer leads to the formation of 1,2-cis
addition products, providing an explanation for the stereo-
selectivity observed in the reaction of 34, 8, and 9 (Table 1,
entries 1−3).
with the 1,2-cis addition product as major components (Table
1, entries 4−6) originating from a E oxocarbenium ion.
3
Xylose. The reductions of the xylofuranosides proceed with
the least selectivity of the studied furanoses. The global FES
maps of xylofuranose oxocarbenium ions 48−50 (Figure 7)
show two energy minima, accounting for the formation of two
anomers in the experiments. The major conformer of ions 49
Arabinose. The FES maps of the arabinofuranosyl
oxocarbenium ions 45, 46, and 47 (R = H, Me, and Ph) are
displayed in Figure 6. The global FES maps indicate that all
three ions preferentially adopt an 3E conformation. The
individual C4−C5 rotamer maps reveal that the gg conformers
are the most stable. The conformational preference of the
arabinosyl oxocarbenium ions (45−47) is less profound than
the preference of the corresponding ribo-oxocarbenium ions
(42−44). This may be due to the fact that the C2 and C3
substituents cannot simultaneously adopt their most favorable
orientation in either of the envelope structures. As is the case of
the ribofuranosyl oxocarbenium ions, the FES maps for the
methyl-substituted arabinofuranosyl ion 46 are somewhat
steeper than for the aldose ion 45, while the maps for the
phenyl-substituted arabinofuranosyl ion 47 are somewhat less
steep. The global FES maps of ions 45−47 do not show a
distinct two conformer model; instead they show a gradual
3
and 50 is the gg E3 envelope, where the minor E conformer
places the C5 methoxy group in a gt position. For the aldose
4
ion 48, the lowest energy conformation is T3, a structure that
slightly deviates from the E3 conformer. This conformation
optimally positions the C5-OMe over the furanosyl ring,
providing the most stabilization. In the ketoxylofuranosyl FES
maps, this effect is less pronounced. As in the case of the ribo
3
and arabino ions, the energy difference between the E3 and E
conformers of the methyl ketofuranosyl ion 49 is somewhat
larger than the difference observed for aldofuranose ion 48 and
somewhat smaller for the phenyl furanosyl oxocarbenium ion
50. The larger energy difference between the E3 and 3E methyl
xylofuranosyl oxocarbenium ions compared to the aldose case
(Table 1, entries 7−9) is not reflected in the experimental
results. The phenyl ketose ion 50 shows a larger preference
toward the 3E envelope than the methyl ketofuranose ion 49. It
still maintains the E3 as the major conformer, but with a smaller
energy difference between the two envelopes. The shallower
FES map for the phenyl xylofuranosyl oxocarbenium ion 50
accounts for the lower stereoselectivity observed in the
reduction of phenyl ketose 13 (Table 1, entry 9).
3
increase in energy for the conformers around the E−E3 axis
(18−198°). On the basis of the flattened FES of the phenyl
ketose oxocarbenium ion 47, one predicts an erosion of
stereoselectivity in reactions involving this species. This is
confirmed in the experiments: the arabino aldose and methyl
ketofuranose both provide a single product upon reduction,
whereas arabino phenyl ketofuranose gives an anomeric mixture
Lyxose. In Figure 8, the FES maps for the lyxofuranosyl
oxocarbenium ions 51−53 are displayed. We previously found
4558
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Figure 7. The gg, gt, tg, and global FES maps of the xylofuranosyl oxocarbenium ions 48−50.
a strong preference for the 3E envelope of lyxofuranosyl
oxocarbenium ion 51 (R = H) and the associated remarkable
stereoselectivity in additions to this intermediate (cis with
respect to all ring substituents).6 This strong conformational
preference is maintained in the ketofuranosyl oxocarbenium
ions 52 and 53, as shown in the global FES maps. In all three 3E
oxocarbenium ions, the C2 substituent is in the preferred
pseudoequatorial position and the C3 substituent is in a
pseudoaxial position, allowing stabilization of the positive
charge at the anomeric position. The FES map of the individual
C4−C5 rotamer shows a preference for C5 gt, avoiding a
sterically and electronically unfavorable interaction with the
axial C3 substituent. Inside attack of the nucleophile on the
lyxofuranosyl 3E oxocarbenium ions leads to the 1,2-cis addition
product in both the aldose and ketoses.
those previously described for the aldoses. An anomeric methyl
substituent in the ketose ions makes the preference for one of
the two envelopes more distinct than in the case of the aldose
ions, and an anomeric phenyl substituent, on the other hand,
leads to smaller energy differences between the envelopes. The
latter effect can be partially attributed to the destabilizing 1,3-
allylic strain between the anomeric phenyl ring positioned
parallel to the CO+ plane and the pseudoequatorial C2
substituent. Also, the conjugative stabilization of the
oxocarbenium ion by the phenyl ring can lead to a smaller
energy difference between the different oxocarbenium ion
conformers. The calculated differences in energy between the
various oxocarbenium ion envelopes provide a reliable
prediction for the stereoselectivity of substitution reactions
involving addition to these intermediates. It appears that in the
addition reactions studied here a Curtin−Hammett kinetic
scenario in which a reaction on the higher energy intermediate
proceeds via a lower overall transition state, thereby providing
the prevalent reaction path, does not play a major role. Given
the fact that the oxocarbenium ion intermediates are high-
energy species, the transition states of the reactions proceeding
through these intermediates will be early and therefore
resemble the structures of the oxocarbenium ions. The relative
product ratios observed for the additions to the phenyl
ketofuranosides deviate most from the product ratios predicted
from the FES maps. The phenyl ketofuranosyl oxocarbenium
ions are the most stable intermediates of the three
oxocarbenium ion types studied, and therefore, the transition
states for reactions taking place on these ions will be the latest
of the series, departing most from the ground-state
oxocarbenium ion structure. Here, other factors, such as steric
CONCLUSION
■
Substitution reactions at the anomeric center of aldofuranoses
as well as ketofuranoses proceed with remarkable stereo-
selectivity to provide 1,2-cis addition products, independent of
the substitution pattern on the carbohydrate ring. The
stereoselectivity presumably is the result of a sequence of
reactions in which the most stable intermediate oxocarbenium
ion conformer is attacked from the inside by the nucleophile in
the product-forming step. Mapping of the FES of the complete
conformational space of the oxocarbenium ions involved
substantiates this mechanistic model: the calculated energy
differences between the envelopes of the differently configured
aldofuranoside and ketofuranoside ions parallel the observed
stereoselectivity in the reactions. Overall, significant similarities
exist between the FES maps for the ketoses studied here and
4559
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
Figure 8. The gg, gt, tg, and global FES maps of the lyxofuranosyl oxocarbenium ions 51−53.
trans = 65°, trans−gauche = 175°) and then generating the FES
through the above-mentioned method, generating a total of 243
optimized geometries. These three FES maps were graphed
individually and in a combined plot by comparing the corrected free
energies and, for each point, selecting the geometry of lowest energy
from the three entities.
interactions between the nucleophile and the electrophile, will
become more important, opening the way for a Curtin−
Hammett scenario described above. In all, the Lewis acid
mediated reduction of ketofuranosides provides efficient access
to a range of C-glycosides with predictable stereochemistry.
2,3,5-Tri-O-benzyl-D-ribono-1,4-lactone (1). 2,3,5-Tri-O-ben-
zyl-D-ribofuranose (10.2 mmol) was dissolved in dimethylsulfoxide
(DMSO) (15 mL, 211 mmol) and cooled to 0 °C, and acetic
anhydride (10 mL, 106 mmol) was added and the mixture stirred
overnight. The reaction was poured onto ice water and extracted with
diethyl ether (3×). The combined organic layers were washed with
H2O (3×) and brine, dried over MgSO4, filtered, and concentrated
under reduced pressure. The residue was purified by silica gel column
chromatography (30% Et2O/pentane), yielding the title compound
(3.82 g, 9.1 mmol, 89%): Rf = 0.30 (30/70 Et2O/petroleum ether); 1H
NMR (400 MHz, CDCl3) δ 7.39−7.26 (m, 13H, CHAr Bn), 7.16 (dd, J
= 7.3, 2.2 Hz, 2H, CHAr Bn), 4.94 (d, J = 12.0 Hz, 1H, CHH Bn), 4.74
(d, J = 12.0 Hz, 1H, CHH Bn), 4.69 (d, J = 11.8 Hz, 1H, CHH Bn),
4.56−4.52 (m, 2H, CHH Bn, C-4), 4.49 (d, J = 11.9 Hz, 1H, CHH
Bn), 4.44−4.36 (m, 2H, CHH Bn, C-2), 4.09 (dd, J = 5.6, 2.0 Hz, 1H,
C-3), 3.66 (dd, J = 11.0, 2.9 Hz, 1H, C-5a), 3.55 (dd, J = 11.0, 2.7 Hz,
1H, C-5b). 13C NMR (101 MHz, CDCl3) δ 173.8 (CO), 137.4,
137.2, 137.1 (Cq Bn), 128.7, 128.6, 128.6, 128.4, 128.2, 128.2, 128.1,
127.7 (CHAr Bn), 81.9 (C-4), 75.5 (C-3), 73.8 (C-2), 73.8, 72.8, 72.5
(3× CH2 Bn), 68.9 (C-5); [α]D20 = 73 (c = 1, CHCl3); IR (neat) 602,
615, 652, 692, 733, 764, 781, 824, 854, 891, 908, 934, 957, 974, 995,
1011, 1022, 1084, 1098, 1155, 1182, 1215, 1233, 1246, 1333, 1362,
1400, 1454, 1497, 1773, 2874, 2909, 3030; HRMS (ESI) [M + H+]
calcd for C26H26O5 419.18530, found 419.18553.
EXPERIMENTAL SECTION
■
In all calculations, methyl ethers were used because of their reduced
calculation costs over benzyl ethers. All calculations were performed
with density functional theory ab initio calculations with the B3LYP
model. The starting conformer for the FES was optimized by starting
from a conformer distribution search option included in the Spartan
0414 program in the gas phase with the 6-31G* basis set. All generated
geometries were optimized with Gaussian 0311 at 6-311G**, and their
zero-point energy (ZPE) corrections were calculated and further
optimized with an incorporated polarizable continuum model to
correct for solvation in dichloromethane. The geometry with the
lowest ZPE-corrected, solvated free energy was selected as the starting
point for the FES. Two dihedral angles of the five-membered ring were
constrained, namely, C4−O4−C1−C2 (τ0, θ3) and C1−C2−C3−C4
(τ2, θ0), with angles from −40 to 40° over nine steps (10° per step),
giving a total of 81 conformers and dictating the entire pseudorota-
tional space within a maximum amplitude (τm) of 40°. All other
internal coordinates were unconstrained. The geometries were
optimized, and their ZPE was calculated and corrected for solvation
with Gaussian 03 at 6-311G** as above. The FES was visualized as a
polar contour plot through the Origin 8.5 graphing software by putting
the phase angle (P) as θ, the amplitude (τm) at r, and the energy
corrected for ZPE and optimized in solvent at the Z-axis. To
interrogate all rotamers of the C5 substituent, the starting conformer
was modified by rotating the O4−C4−C5−O5 dihedral angle to each
of the three staggered configurations (gauche−gauche = −65°, gauche−
1-C-(5′-(2′,3′-Di-tert-butoxy)pyrimidine)-2,3,5-tri-O-benzyl-
D-ribofuranose (3). To a stirred and cooled (−78 °C) solution of,
with toluene coevaporated (3×), 5-bromo-2,4-di-tert-butoxypyrimi-
4560
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
dine15 (2.4 g, 8 mmol) in anhydrous tetrahydrofuran (THF) (40 mL)
was added a solution of n-BuLi (5.5 mL, 8.8 mmol, 1.6 M in hexanes).
After 20 min at −78 °C, with toluene coevaporated (3×), 2,3,5-tri-O-
benzyl-D-arabino-1,4-lactone (5, 5.0 g, 11.8 mmol) in anhydrous THF
(40 mL) was added dropwise over 45 min. The mixture was allowed to
slowly warm to −55 °C over a period of 4 h after which thin layer
chromatography (TLC) indicated complete reaction. The reaction
mixture was cooled to −78 °C and was quenched by the addition of
NaH2PO4 (8 mL, 20% aq). The reaction mixture was allowed to warm
to room temperature, diluted with EtOAc, and washed with H2O (2×)
and brine, dried over Na2SO4, filtered, and concentrated under
reduced pressure. Silica gel purification (5−15% EtOAc/toluene)
yielded the title compound (2.98 g, 4.6 mmol, 58%), which was used
directly in the next step.
methyl-1-deoxy-2,3,5-tri-O-benzyl-D-ribofuranose (140 mg, 0.32
mmol, 64%) as a single diastereomer: Rf = 0.85 (20/80 EtOAc/
petroleum ether); 1H NMR (400 MHz, CDCl3) δ 7.35−7.23 (m, 15H,
CHAr Bn), 4.60 (d, J = 11.9 Hz, 1H, CHH Bn), 4.55 (s, 2H, CH2 Bn),
4.55 (d, J = 12.1 Hz, 1H, CHH Bn), 4.50 (d, J = 12.0 Hz, 1H, CHH
Bn), 4.48 (d, J = 11.9 Hz, 1H, CHH Bn), 4.19−4.14 (m, 1H, C-4),
4.13−4.06 (m, 1H, C-1), 3.87 (dd, J = 5.3, 4.3 Hz, 1H, C-3), 3.49 (d, J
= 4.5 Hz, 2H, C-5), 3.46 (dd, J = 6.7, 5.7 Hz, 1H, C-2), 1.25 (d, J = 6.3
Hz, 3H, CH3); 13C NMR (101 MHz, CDCl3) δ 138.2, 138.0, 138.0
(Cq Bn), 128.4, 128.4, 128.4, 128.2, 127.9, 127.8, 127.7, 127.6 (CHAr
Bn), 82.6 (C-2), 81.7 (C-4), 77.5 (C-3), 76.9 (C-1), 73.5, 72.1, 71.8
(3× CH2 Bn), 70.7 (C-5), 19.1 (CH3); [α]D20 = −10 (c = 1.1, CHCl3);
IR (neat) 697, 736, 1027, 1097, 1454, 1721, 2872; HRMS (ESI) [M +
Na]+ calcd for C27H30O4 441.20363, found 441.20330.
1-Phenyl-2,3,5-tri-O-benzyl-D-ribofuranose (9). To a stirred
and cooled (−78 °C) solution of bromobenzene (115 μL, 1.1 mmol)
in anhydrous THF (4 mL) was added n-BuLi (0.74 mL, 1.2 mmol, 1.6
M in hexanes). After 20 min at −78 °C, a solution of, with toluene
coevaporated (3×), 2,3,5-tri-O-benzyl-D-ribono-1,4-lactone (1, 382
mg, 0.9 mmol) in anhydrous THF (4 mL) was added dropwise. After
2 h, the reaction mixture was quenched by the addition of saturated aq
NH4Cl (4 mL) at −78 °C. The reaction mixture was allowed to warm
to room temperature, diluted with EtOAc, washed with H2O (2×) and
brine, dried over MgSO4, filtered, and concentrated under reduced
pressure. Silica gel purification (20−30% Et2O/petroleum ether)
allowed removal of any unreacted components and yielded the title
compound (218 mg, 0.44 mmol, 48%) which was used directly in the
next step: Rf = 0.30 (30/70 Et2O/petroleum ether).
2′,3′,5′-Tri-O-benzyl-pseudouridine (4). To a stirred and
cooled (−78 °C) solution of, with toluene coevaporated (3×), 1-C-
(5′-(2′,3′-di-tert-butoxy)pyrimidine)-2,3,5-tri-O-benzyl-D-ribofuranose
(3, 2.84 g, 4.4 mmol) in anhydrous dichloromethane (DCM) (15 mL)
were added triethylsilane (0.95 mL, 5.9 mmol) and BF3·Et2O (1.56
mL, 11 mmol). The reaction was allowed to slowly warm to 10 °C and
stirred overnight at this temperature. The reaction mixture was cooled
to −5 °C before TFA (15 mL) was added. The solution was stirred for
90 min, poured on ice water, and extracted with DCM. The organic
layer was washed with saturated aq NaHCO3, H2O (3×) and brine,
dried over Na2SO4, filtered, and concentrated under reduced pressure.
The residue was purified by silica gel column chromatography (5−15%
acetone/DCM), yielding 5-(2,3,5-tri-O-benzyl-β-D-ribofuranosyl)uracil
(1.53 g, 3.0 mmol, 68% yield over 2 steps) as a single diastereomer: Rf
1
= 0.15 (10/90 acetone/DCM); H NMR (400 MHz, CDCl3) δ 9.63
β-1-Phenyl-1-deoxy-2,3,5-tri-O-benzyl-D-ribofuranose (27).
To a stirred and cooled (−78 °C) solution of, with toluene
coevaporated (3×), 1-phenyl-2,3,5-tri-O-benzyl-D-ribofuranose (9,
200 mg, 0.4 mmol) in anhydrous DCM (6 mL) were added
triethylsilane (192 μL, 1.2 mmol) and BF3·Et2O (303 μL, 2.4 mmol).
The reaction was stirred at −78 °C for 1 h, after which the reaction
was quenched by addition of saturated aq NaHCO3 (6 mL). The
reaction mixture was diluted with EtOAc and washed with saturated aq
NaHCO3 (2×), H2O, and brine, dried over MgSO4, filtered, and
concentrated under reduced pressure. The residue was purified by
silica gel column chromatography (20% Et2O/petroleum ether) and
yielded β-1-phenyl-1-deoxy-2,3,5-tri-O-benzyl-D-ribofuranose (150 mg,
0.31 mmol, 78%) as a single diastereomer: Rf = 0.60 (30/70 Et2O/
petroleum ether); 1H NMR (400 MHz, CDCl3) δ 7.42−7.37 (m, 3H,
CHAr Ph), 7.37−7.22 (m, 15H, CHAr Bn), 7.21−7.15 (m, 2H, CHAr
Ph), 5.03 (d, J = 6.6 Hz, 1H, C-1), 4.61 (d, J = 12.0 Hz, 1H, CHH Bn),
4.60 (d, J = 12.4 Hz, 1H, CHH Bn), 4.59−4.55 (m, 1H, CHH Bn),
4.55 (d, J = 11.9 Hz, 1H, CHH Bn), 4.49 (d, J = 12.6 Hz, 1H, CHH
Bn), 4.46 (d, J = 12.9 Hz, 1H, CHH Bn), 4.38−4.32 (m, 1H, C-4),
4.01 (dd, J = 4.9, 4.4 Hz, 1H, C-3), 3.81 (dd, J = 6.5, 5.3 Hz, 1H, C-2),
3.68 (dd, J = 10.4, 4.2 Hz, 1H,C-5a), 3.63 (dd, J = 10.4, 4.0 Hz, 1H, C-
5b); 13C NMR (101 MHz, CDCl3) δ 140.5 (Cq Ph), 138.3, 138.0,
137.9 (Cq Bn), 128.5, 128.4, 128.4, 128.2, 127.9, 127.9, 127.8, 127.8,
127.7, 126.4 (CHAr Ph, CHAr Bn), 83.8 (C-2), 82.8 (C-1), 81.8 (C-4),
77.6 (C-3), 73.6, 72.3, 72.0 (CH2 Bn), 70.5 (C-5); [α]D20 = −18.2 (c =
1, CHCl3); IR (neat) 696, 730, 1010, 1051, 1106, 1144, 1367, 1454,
1496, 2358, 2870; HRMS (ESI) [M + Na]+ calcd for C32H32O4
503.21928, found 503.21884.
(s, 1H, NH-3), 8.77 (d, J = 5.6 Hz, 1H, NH-1), 7.59 (d, J = 5.9 Hz,
1H, C-6), 7.50−7.38 (m, 2H, 2× CHAr Bn), 7.37−7.14 (m, 13H, CHAr
Bn), 5.04 (s, 1H, C-1′), 4.85 (d, J = 12.3 Hz, 1H, CHH Bn), 4.74 (d, J
= 12.3 Hz, 1H, CHH Bn), 4.51−4.45 (m, 2H, 2× CHH Bn), 4.42 (d, J
= 11.4 Hz, 1H, CHH Bn), 4.34−4.19 (m, 2H, CHH Bn, C-4′), 4.02
(d, J = 4.6 Hz, 1H, C-2′), 3.95 (dd, J = 8.6, 4.6 Hz, 1H, C-3′), 3.87
(dd, J = 10.7, 2.2 Hz, 1H, C-5′a), 3.63 (dd, J = 10.8, 2.7 Hz, 1H, C-
5′b); 13C NMR (101 MHz, CDCl3) δ 163.0 (C-4), 152.0 (C-2), 139.1
(C-5), 138.2, 138.2, 137.8 (Cq Bn), 128.6, 128.5, 128.5, 128.4, 128.0,
127.9, 127.8, 127.8 (CHAr Bn), 113.1 (C-6), 79.4 (C-4′), 78.6 (C-2′),
78.3 (C-1′), 75.7 (C-3′), 73.3, 72.0, 71.3 (CH2 Bn), 68.7 (C-5′); IR
(neat) 694, 731, 818, 1001, 1026, 1040, 1074, 1084, 1121, 1204, 1356,
20
1427, 1452, 1495, 1661, 1703, 2860, 2913; [α]D = 149.7 (c = 1.1,
CHCl3); HRMS (ESI) [M + H+] calcd for C30H30N2O6 515.21766,
found 515.21771.
1-Methyl-2,3,5-tri-O-benzyl-D-ribofuranose (8). To a stirred
and cooled (−78 °C) solution of, with toluene coevaporated (3×),
2,3,5-tri-O-benzyl-D-ribono-1,4-lactone (6, 0.9 mmol, 380 mg) in
anhydrous THF (3 mL) was added dropwise (20 min) methyl lithium
(0.69 mL, 1.1 mmol, 1.6 M in hexanes). After 2 h, an additional 0.2
mmol of methyl lithium (0.125 mmol) was added. The reaction was
quenched by the addition of saturated aq NH4Cl (0.8 mL), diluted
with EtOAc, and allowed to warm to room temperature. The organic
layer was washed with H2O (2×) and brine, dried over MgSO4,
filtered, and concentrated under reduced pressure. Silica gel
purification (40/60 Et2O/petroleum ether) allowed removal of any
unreacted components and yielded the title compound (310 mg, 0.7
mmol, 77%), which was used directly in the next step: Rf = 0.15 (30/
70 Et2O/petroleum ether).
β-1-Methyl-1-deoxy-2,3,5-tri-O-benzyl-D-ribofuranose (26).
To a stirred and cooled (−78 °C) solution of, with toluene
coevaporated (3×), 1-methyl-2,3,5-tri-O-benzyl-D-ribofuranose (8,
220 mg, 0.5 mmol) in anhydrous DCM (5 mL) were added
triethylsilane (242 μL, 1.5 mmol) and BF3·Et2O (380 μL, 3.0 mmol).
The reaction was stirred at −78 °C for 1.5 h, after which TLC analysis
indicated complete conversion. The reaction mixture was quenched by
addition of saturated aq NaHCO3 (1 mL). The reaction mixture was
diluted with EtOAc and washed with saturated aq NaHCO3 (2×),
H2O, and brine, dried over MgSO4, filtered, and concentrated under
reduced pressure. The residue was purified by silica gel column
chromatography (10% EtOAc/petroleum ether) and yielded β-1-
2,3,5-Tri-O-benzyl-D-arabino-1,4-lactone (5). 2,3,5-Tri-O-ben-
zyl-D-arabinofuranose (1.0 g, 2.4 mmol) was dissolved in DMSO (18.5
mL, 262 mmol); acetic anhydride (12.5 mL, 133 mmol) was added,
and the reaction mixture was stirred for 4 days. The mixture was
poured onto ice water and extracted with Et2O. The organic layer was
washed with H2O (2×) and brine, dried over Na2SO4, and
concentrated under reduced pressure. The residue was crystallized
from methanol, yielding the title compound (729 mg, 1.7 mmol, 73%):
1
Rf = 0.85 (25/75 EtOAc/pentane); H NMR (400 MHz, CDCl3) δ
7.44−7.25 (m, 13H, CHAr Bn), 7.25−7.18 (m, 2H, CHAr Bn), 5.06 (d,
J = 11.6 Hz, 1H, CHH Bn), 4.77 (d, J = 11.6 Hz, 1H, CHH Bn), 4.63
(d, J = 11.6 Hz, 1H, CHH Bn), 4.58−4.46 (m, 3H, 3× CHH Bn),
4.38−4.29 (m, 3H, C-2, C-3, C-4), 3.70 (dd, J = 11.4, 2.3 Hz, 1H, C-
5a), 3.60−3.54 (m, 1H, C-5b); 13C NMR (101 MHz, CDCl3) δ 172.6
4561
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
(CO-1), 137.6, 137.2, 136.9 (Cq Bn), 128.7, 128.6, 128.6, 128.6,
128.3, 128.2, 128.0, 128.0, 127.9 (CHAr Bn), 79.3, 79.2, 78.9 (C-2, C-3,
C-4), 73.6, 72.8, 72.6 (CH2 Bn), 68.0 (C-5); [α]D = 6 (c = 1,
CHCl3); IR (neat) 633, 644, 962, 731, 804, 872, 903, 949, 988, 997,
1016, 1028, 1038, 1049, 1070, 1086, 1113, 1130, 1146, 1180, 1227,
1265, 1321, 1369, 1452, 1495, 1778, 2866, 2911, 3030; HRMS (ESI)
[M + H+] calcd for C26H26O5 419.18530, found 419.18551.
reduced pressure. The residue was purified by silica gel column
chromatography (15% EtOAc/petroleum ether), yielding an anomeric
mixture (α/β = 85:15) of α/β-1-phenyl-1-deoxy-2,3,5-tri-O-benzyl-D-
arabinofuranose (173 mg, 0.36 mmol, 65%): Rf = 0.80 (25/75 Et2O/
20
1
petroleum ether). α-Anomer: H NMR (400 MHz, CDCl3) δ 7.45−
7.37 (m, 2H, CHAr Ph), 7.37−7.15 (m, 18H, CHAr Bn, CHAr Ph), 4.98
(d, J = 6.1 Hz, 1H, C-1), 4.58 (s, 2H, CH2 Bn), 4.55 (d, J = 11.8 Hz,
1H, CHH Bn), 4.51 (d, J = 11.9 Hz, 1H, CHH Bn), 4.45 (d, J = 11.7
Hz, 1H, CHH Bn), 4.43−4.39 (m, 2H, C-4, CHH Bn), 4.21 (t, J = 4.1
Hz, 1H, C-3), 4.13 (dd, J = 6.1, 4.0 Hz, 1H, C-2), 3.69−3.61 (m, 2H,
C-5); 13C NMR (101 MHz, CDCl3) δ 140.6 (Cq Ph), 138.2, 137.9,
137.8 (Cq Bn), 128.5, 128.5, 128.4, 128.2, 127.9, 127.9, 127.9, 127.8,
127.8, 127.7, 127.7, 127.7, 127.6, 127.6, 127.6, 126.5 (CHAr Bn, CHAr
Ph), 90.3 (C-2), 85.1 (C-3), 83.9 (C-1), 81.9 (C-4), 73.5, 72.3, 72.0
(3× CH2 Bn), 70.3 (C-5). β-Anomer: 1H NMR (400 MHz, CDCl3) δ
7.46−7.37 (m, 2H, CHAr Ph), 7.36−7.15 (m, 16H, CHAr Bn, CHAr
Ph), 6.91−6.85 (m, 2H, CHAr Ph), 5.10 (d, J = 3.7 Hz, 1H, C-1), 4.62
(d, J = 12.1 Hz, 1H, CHH Bn), 4.58 (d, J = 11.8 Hz, 1H, CHH Bn),
4.55 (d, J = 12.1 Hz, 1H, CHH Bn), 4.52−4.49 (m, 1H, CHH Bn),
4.27−4.22 (m, 1H, C-4), 4.07 (d, J = 12.2 Hz, 1H, CHH Bn), 4.04−
4.02 (m, 1H, C-3), 4.00−3.94 (m, 2H, CHH Bn, C-2), 3.79 (dd, J =
9.9, 5.9 Hz, 1H, C-5a), 3.71−3.67 (m, 1H, C-5b); 13C NMR (101
MHz, CDCl3) δ 138.4 (Cq Ph), 137.9, 137.8, 136.7 (Cq Bn), 128.5,
128.5, 128.4, 128.2, 127.9, 127.9, 127.8, 127.8, 127.7, 127.7, 127.7,
127.6, 127.6, 127.6, 126.5 (CHAr Bn, CHAr Ph), 85.1 (C-3), 84.2 (C-2),
83.3 (C-1), 82.5 (C-4), 73.4, 71.7, 71.6 (3× CH2 Bn), 70.6 (C-5); IR
(neat) 734, 1028, 1093, 1454, 1496, 2858; HRMS (ESI) [M + Na]+
calcd for C32H32O4 503.21928, found 503.21863.
1-Methyl-2,3,5-tri-O-benzyl-D-arabinofuranose (10). To a
stirred and cooled (−78 °C) solution of, with toluene coevaporated
(3×), 2,3,5-tri-O-benzyl-D-arabino-1,4-lactone (5, 418 mg, 1 mmol) in
anhydrous THF (4 mL) was added methyl lithium (0.69 mL, 1.1
mmol, 1.6 M in Et2O). After 6 h, the reaction was quenched by the
addition of saturated aq NH4Cl (4 mL), diluted with EtOAc, and
allowed to warm to room temperature. The layers were separated, and
the aqueous layer was extracted twice with EtOAc. The combined
organic layers were washed with brine, dried over MgSO4, filtered, and
concentrated under reduced pressure. Silica gel purification (15−25%
EtOAc/petroleum ether) allowed removal of any unreacted
components and yielded the title compound as a colorless oil (386
mg, 0.87 mmol, 86%) which was used directly in the next step: Rf =
0.30 (25/75 EtOAc/petroleum ether).
α-1-Methyl-1-deoxy-2,3,5-tri-O-benzyl-D-arabinofuranose
(28). To a stirred and cooled (−78 °C) solution of, with toluene
coevaporated (3×), 1-methyl-2,3,5-tri-O-benzyl-D-arabinofuranose
(10, 132 mg, 0.30 mmol) in anhydrous DCM (4.3 mL) were added
triethylsilane (63 μL, 0.40 mmol) and BF3·Et2O (50 μL, 0.4 mmol).
The reaction was stirred at −78 °C for 3 days at which TLC indicated
complete conversion. The reaction mixture was quenched by addition
of saturated aq NaHCO3 (4 mL). The reaction mixture was extracted
with EtOAc (3×), and the combined organic layers were washed with
H2O and brine, dried over MgSO4, filtered, and concentrated under
reduced pressure. The residue was purified by silica gel column
chromatography (8−10% EtOAc/petroleum ether) and yielded α-1-
methyl-1-deoxy-2,3,5-tri-O-benzyl-D-arabinofuranose (106 mg, 0.30
mmol, 83%) as a single diastereomer: Rf = 0.90 (25/75 EtOAc/
petroleum ether); 1H NMR (400 MHz, CDCl3) δ 7.35−7.23 (m, 15H,
CHAr Bn), 4.59−4.48 (m, 5H, 5× CHH Bn), 4.49 (d, J = 11.9 Hz, 1H,
CHH Bn), 4.25−4.19 (m, 1H, C-4), 4.17−4.09 (m, 1H, C-1), 4.02 (t,
J = 3.7 Hz, 1H, C-3), 3.76 (dd, J = 5.0, 3.3 Hz, 1H, C-2), 3.58 (dd, J =
9.9, 5.8 Hz, 1H, C-5a), 3.53 (dd, J = 10.0, 5.7 Hz, 1H, C-5b), 1.31 (d, J
= 6.4 Hz, 3H, CH3); 13C NMR (101 MHz, CDCl3) δ 138.2, 138.0 (Cq
Bn), 128.6, 128.5, 128.5, 127.9, 127.9, 127.9, 127.8, 127.7 (CHAr Bn),
89.4 (C-2), 85.6 (C-3), 81.2 (C-4), 78.3 (C-1), 73.5, 72.0, 72.0 (3×
CH2 Bn), 70.5 (C-5), 19.3 (CH3); [α]D20 = 11.9 (c = 1.1, CHCl3); IR
(neat) 696, 734, 1028, 1092, 1722, 2864; HRMS (ESI) [M + H+]
calcd for C27H30O4 419.22169, found 419.22165.
1-Phenyl-2,3,5-tri-O-benzyl-D-arabinofuranose (11). To a
stirred and cooled (−78 °C) solution of phenyl bromide (115 μL,
1.1 mmol) in anhydrous THF (1.5 mL) was added n-BuLi (0.74 mL,
1.1 mmol, 1.6 M in hexanes). After 20 min at −78 °C, with toluene
coevaporated (3×), 2,3,5-tri-O-benzyl-D-arabino-1,4-lactone (5, 420
mg, 1 mmol) in anhydrous THF (1.5 mL) was added dropwise. After
1 h, the reaction was quenched by the addition of saturated aq NH4Cl
(0.5 mL) at −78 °C. The reaction mixture was allowed to warm to
room temperature, diluted with EtOAc, and washed with H2O and
brine, dried over MgSO4, filtered, and concentrated under reduced
pressure. Silica gel purification (20% Et2O/petroleum ether) allowed
removal of any unreacted components and yielded the title compound
(348 mg, 0.7 mmol, 70%) which was used directly in the next step: Rf
= 0.70 (20/80 Et2O/petroleum ether).
α/β-1-Phenyl-1-deoxy-2,3,5-tri-O-benzyl-D-arabinofuranose
(29). To a stirred and cooled (−78 °C) solution of, with toluene
coevaporated (3×), 1-phenyl-2,3,5-tri-O-benzyl-D-arabinofuranose (11,
275 mg, 0.55 mmol) in anhydrous DCM (8 mL) were added
triethylsilane (233 μL, 1.46 mmol) and BF3·Et2O (400 μL, 3.2 mmol).
The reaction was stirred at −78 °C for 1.5 h at which TLC analysis
showed near complete conversion. The reaction was quenched by
addition of saturated aq NaHCO3 (1 mL). The reaction mixture was
diluted with EtOAc and washed with saturated aq NaHCO3 (2×),
H2O, and brine, dried over MgSO4, filtered, and concentrated under
2,3,5-Tri-O-benzyl-D-xylono-1,4-lactone (6). 2,3,5-Tri-O-ben-
zyl-D-xylofuranose (6.9 g, 16.4 mmol) was dissolved in DMSO (25
mL, 345 mmol). Ac2O (16 mL, 172 mmol) was added, and the
reaction was stirred at room temperature for 44 h after which TLC
analysis indicated full conversion. The reaction mixture was quenched
with ice water and extracted with Et2O, and the organic layer washed
with H2O and brine, dried over MgSO4, filtered, and concentrated.
The residue was purified by silica gel column chromatography (7.5−
12.5% EtOAc/petroleum ether) to provide the title compound (6.5 g,
15.5 mmol, 95% yield) as a white solid which was recrystallized from
MeOH to provide an analytical sample: Rf = 0.80 (25/75 EtOAc/
petroleum ether); 1H NMR (400 MHz, CDCl3) δ 7.44−7.17 (m, 15H,
CHAr Bn), 5.04 (d, J = 11.5 Hz, 1H, CHH Bn), 4.69 (d, J = 11.4 Hz,
1H, CHH Bn), 4.65 (d, J = 11.9 Hz, 1H, CHH Bn), 4.60−4.48 (m,
5H, C-4, C-2, 3× CHH Bn), 4.36 (t, J = 7.1 Hz, 1H, C-2), 3.76 (dd, J
= 10.9, 2.8 Hz, 1H, C-5a), 3.70 (dd, J = 10.9, 3.2 Hz, 1H, C-5b); 13C
NMR (101 MHz, CDCl3) δ 173.4 (CO), 137.7, 137.4, 137.2 (Cq
Bn), 128.6, 128.6, 128.5, 128.4, 128.2, 128.2, 127.8, 127.8, 127.7 (CHAr
Bn), 79.5 (C-3), 77.4 (C-2, C-4), 73.7, 72.8, 72.7 (3× CH2 Bn), 67.2
20
(C-5); [α]D = 91.9 (c = 1, CHCl3); IR (neat) 608, 625, 646, 671,
698, 743, 799, 835, 864, 912, 930, 972, 993, 1024, 1070, 1088, 1105,
1128, 1188, 1213, 1242, 1285, 1344, 1377, 1393, 1454, 1466, 1497,
1732, 1769, 2872, 2909; HRMS (ESI) [M + H+] calcd for C26H26O5
419.18530, found 419.18545.
1-Methyl-2,3,5-tri-O-benzyl-D-xylofuranose (12). 2,3,5-Tri-O-
benzyl-D-xylono-1,4-lactone (6, 252 mg, 0.60 mmol), three times
coevaporated with toluene, was dissolved in anhydrous THF (2.5 mL)
and cooled to −78 °C. Methyl lithium (0.41 mL, 0.66 mmol, 1.6 M in
Et2O) was added slowly. After 3 h, TLC analysis indicated full
conversion. The reaction was quenched by addition of saturated aq
NH4Cl (2.5 mL), and the mixture was allowed to warm to room
temperature. The suspension was extracted three times with EtOAc,
and the combined organic layers were washed with brine, dried over
MgSO4, filtered, and concentrated. The residue was purified by silica
gel column chromatography (20−30% EtOAc/pentane) to provide the
title compound (243 mg, 0.56 mmol, 93% yield) as a colorless oil
which was used directly in the next step: Rf = 0.55 (25/75 EtOAc/
pentane).
α/β-1-Methyl-1-deoxy-2,3,5-tri-O-benzyl-D-xylofuranose
(30). 1-Methyl-2,3,5-tri-O-benzyl-D-xylofuranose (12, 242 mg, 0.56
mmol), three times coevaporated with toluene, was dissolved in
anhydrous DCM (7.2 mL) and cooled to −78 °C. Triethylsilane (116
4562
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
μL, 0.73 mmol) was added before dropwise addition of BF3·OEt2 (92
μL, 0.73 mmol). The reaction was stirred for 6 days at −78 °C after
which TLC analysis indicated near full conversion. The reaction was
quenched with saturated aq NaHCO3 (8 mL), warmed to room
temperature, and the suspension extracted with EtOAc (3×). The
combined organic layers were washed with H2O and brine, dried over
MgSO4, filtered, and concentrated. The residue was purified by silica
gel column chromatography (8−12% EtOAc/pentane) to provide an
anomeric mixture (α/β = 15:85) of the title compound (203 mg, 0.49
mmol, 87% yield) as a colorless oil: Rf = 0.80 (20/80 EtOAc/
CDCl3) δ 7.44−7.13 (m, 18H, CHAr Ph, CHAr Bn), 6.93−6.86 (m, 2H,
CHAr Ph), 5.21 (d, J = 3.5 Hz, 1H, C-1), 4.67−4.60 (m, 1H, C-4),
4.59−4.39 (m, 4H, 4× CHH Bn), 4.13 (dd, J = 4.0, 1.2 Hz, 1H, C-2),
4.10 (d, J = 12.3 Hz, 1H, CHH Bn), 4.04−4.00 (m, 1H, CHH Bn),
3.96−3.93 (m, 1H, C-3), 3.84−3.75 (m, 2H, C-5); 13C NMR (101
MHz, CDCl3) δ 140.7 (Cq Ph), 138.4, 138.0, 137.8 (Cq Bn), 128.5,
128.5, 128.4, 128.4, 128.4, 128.3, 127.9, 127.9, 127.8, 127.6, 127.5,
127.5, 127.5 (CHAr Ph, CHAr Bn), 83.2 (C-3), 82.8 (C-2), 82.5 (C-1),
73.5, 72.5, 72.1 (3× CH2 Bn), 68.6 (C-5), 2JH1−C2 = +2.3 Hz; IR (neat)
604, 646, 694, 731, 887, 908, 951, 986, 1003, 1026, 1067, 1206, 1252,
1308, 1358, 1452, 1495, 2860, 2918, 3028; HRMS (ESI) [M + H+]
calcd for C32H32O4 481.23734, found 481.23748.
1
petroleum ether). β-Anomer: H NMR (400 MHz, CDCl3) δ 7.39−
7.21 (m, 15H, CHAr Bn), 4.63 (d, J = 12.0 Hz, 1H, CHH Bn), 4.56 (d,
J = 12.1 Hz, 1H, CHH Bn), 4.54−4.43 (m, 4H, 4× CHH Bn), 4.18
(ddd, J = 6.5, 5.3, 4.1 Hz, 1H, C-4), 3.99−3.91 (m, 2H, C-1, C-3),
3.81−3.76 (m, 1H, C-5a), 3.73 (dd, J = 10.0, 6.5 Hz, 1H, C-5b), 3.65
(dd, J = 4.1, 1.3 Hz, 1H, C-2), 1.35 (d, J = 6.5 Hz, 3H, CH3); 13C
NMR (101 MHz, CDCl3) δ 138.3, 138.1, 137.9 (3× Cq Bn), 128.6,
128.5, 128.5, 128.4, 128.0, 127.9, 127.9, 127.9, 127.8, 127.7, 127.7,
127.6 (CHAr Bn), 88.7 (C-2), 83.5 (C-3), 80.1 (C-4), 79.9 (C-1), 73.6,
2,3,5-Tri-O-benzyl-4-O-triethylsilyl-D-lyxonic N,O-Dimethyl-
hydroxylamide (15). N,O-dimethylhydroxylamine·HCl (506 mg,
5.2 mmol), three times coevaporated with toluene, was dissolved in
anhydrous THF (8 mL) and cooled to 0 °C. AlMe3 (2 M in toluene,
2.4 mL, 4.8 mmol) was added slowly and stirred for 30 min. 2,3,5-Tri-
O-benzyl-D-lyxono-1,4-lactone (7, 1.0 g, 2.40 mmol) in anhydrous
THF (8 mL) was slowly added, stirred for 5 min, and allowed to warm
to room temperature. After 1.5 h, TLC analysis indicated full
conversion. The reaction was quenched using EtOAc, and the mixture
was washed with potassium sodium tartrate (saturated aq). The
aqueous layer was extracted with EtOAc, and the combined organic
layers were washed with brine, dried over MgSO4, filtered, and
concentrated. The crude 2,3,5-tri-O-benzyl-D-lyxonic N,O-dimethyl-
hydroxylamide, three times coevaporated with toluene, was put under
argon, dissolved in anhydrous THF (12 mL), and cooled to 0 °C.
Imidazole (245 mg, 3.6 mmol) was added followed by dropwise
addition of triethylchlorosilane (1.0 mL, 6.0 mmol), and the mixture
was stirred for 5 min at 0 °C after which the mixture was allowed to
warm to room temperature. After being stirred overnight, TLC
analysis indicated full conversion and the reaction was quenched with
saturated aq NaHCO3. The suspension was extracted with Et2O, and
the combined organic layers were washed with brine. The combined
aqueous layers were extracted with Et2O, and the combined organic
layers were dried over MgSO4, filtered, and concentrated. The residue
was purified by silica gel column chromatography (14−16% EtOAc/
pentane) to provide the title compound (1.0 g, 1.8 mmol, 73%) as a
colorless oil: Rf = 0.60 (20/80 EtOAc/pentane); 1H NMR (400 MHz,
CDCl3) δ 7.38−7.17 (m, 15H, CHAr Bn), 4.88 (d, J = 8.8 Hz, 1H, C-
2), 4.56−4.43 (m, 5H, CHH Bn), 4.40 (d, J = 11.9 Hz, 1H, 5× CHH
Bn), 4.25−4.18 (m, 1H, C-4), 3.95 (dd, J = 8.8, 2.2 Hz, 1H, C-3),
3.61−3.46 (m, 5H, CH3N, C-5), 3.13 (s, 3H, CH3O), 0.92 (t, J = 7.9
Hz, 9H, 3× CH3 TES), 0.67−0.49 (m, 6H, 3× CH2 TES); 13C NMR
(101 MHz, CDCl3) δ 172.3 (CO), 138.7, 138.2, 137.7 (Cq Bn),
128.4, 128.3, 128.2, 128.2, 127.8, 127.8, 127.6, 127.5 (CHAr Bn), 80.0
(C-3), 74.9, 73.3 (2× CH2 Bn), 72.7 (C-2), 71.8 (CH2 Bn), 71.6 (C-
5), 70.8 (C-4), 61.7 (NCH3), 32.1 (OCH3), 7.1 (CH3 TES), 5.4 (CH2
TES); IR (neat) 613, 694, 731, 787, 956, 999, 1076, 1090, 1142, 1238,
1454, 1663, 2874, 2951; [α]D20 −3.1 (c = 1, CHCl3); HRMS (ESI) [M
+ H+] calcd for C34H47NO6Si 594.32454, found 594.32451.
2
71.8, 71.8 (3× CH2 Bn), 68.6 (C-5), 19.9 (CH3), JH1−C2 = −2.3 Hz.
α-Anomer: 1H NMR (400 MHz, CDCl3) δ 7.39−7.21 (m, 15H, CHAr
Bn), 4.62 (d, J = 12.1 Hz, 1H, CHH Bn), 4.57−4.44 (m, 5H, 5× CHH
Bn), 4.40−4.34 (m, 1H, C-4), 4.32−4.25 (m, 1H, C-1), 4.04 (dd, J =
4.3, 1.3 Hz, 1H, C-3), 3.78 (dd, J = 9.9, 5.3 Hz, 1H, C-2), 3.73−3.64
(m, 2H, C-5), 1.27 (d, J = 7.0 Hz, 3H, CH3); 13C NMR (101 MHz,
CDCl3) δ 138.4, 138.2, 138.1 (3× Cq Bn), 128.6, 128.4, 128.0, 127.9,
127.9, 127.9, 127.8, 127.7, 127.7, 127.6 (CHAr Bn), 82.7 (C-2), 82.1
(C-3), 78.5 (C-4), 76.2 (C-1), 73.5, 72.4, 72.0 (3× CH2 Bn), 68.7 (C-
5), 14.6 (CH3), 2JH1−C2 = +1.6 Hz; IR (neat) 606, 640, 673, 696, 733,
799, 849, 910, 930, 970, 993, 1026, 1069, 1088, 1190, 1206, 1312,
1344, 1393, 1454, 1497, 1771, 2862, 2913; HRMS (ESI) [M + H+]
calcd for C27H30O4 419.22169, found 419.22142.
2,3,5-Tri-O-benzyl-1-phenyl-D-xylofuranose (13). Bromoben-
zene (76 μL, 0.70 mmol) was dissolved in THF (1.5 mL) and cooled
to −78 °C. n-Butyl lithium (0.45 mL, 0.70 mmol, 1.6 M in hexanes)
was added slowly, and the mixture was stirred for 30 min. 2,3,5-Tri-O-
benzyl-D-xylono-1,4-lactone (6, 230 mg, 0.55 mmol), three times with
toluene coevaporated, in anhydrous THF (1.5 mL) was added
dropwise. After 2 h, TLC analysis indicated full conversion. The
reaction was quenched with saturated aq NH4Cl (3 mL), and the
suspension was extracted with EtOAc. The organic layer was washed
with H2O and brine, dried over MgSO4, filtered, and concentrated.
The residue was purified by silica gel column chromatography (10−
14% EtOAc/pentane) to provide the title compound (250 mg, 0.50
mmol, 91% yield) as a colorless oil which was used directly in the next
step: Rf = 0.55 (20/80 EtOAc/pentane).
α/β-1-Phenyl-1-deoxy-2,3,5-tri-O-benzyl-D-xylofuranose
(31). 1-Phenyl-2,3,5-tri-O-benzyl-D-xylofuranose (13, 238 mg, 0.48
mmol), three times coevaporated with toluene, was dissolved in
anhydrous DCM (6.9 mL) and cooled to −78 °C. Triethylsilane (103
μL, 0.65 mmol) was added followed by dropwise addition of BF3·OEt2
(82 μL, 0.65 mmol). After being stirred for 3 days at this temperature,
TLC analysis indicated full conversion. The reaction was quenched
with saturated aq NaHCO3 (7 mL), allowed to warm to room
temperature, and the suspension extracted with EtOAc (3×). The
combined organic layers were washed with H2O and brine, dried over
MgSO4, filtered, and concentrated. The residue was purified by silica
gel column chromatography (6−8% EtOAc/pentane) to provide an
anomeric mixture (α/β = 25:75) of the title compound (183 mg, 0.38
mmol, 79% yield) as a colorless oil: Rf = 0.76 (20/80 EtOAc/
1-Methyl-2,3,5-tri-O-benzyl-4-O-triethylsilyl-D-lyxose (16).
2,3,5-Tri-O-benzyl-4-O-triethylsilyl-D-lyxonic N,O-dimethylhydroxyl-
amide (15, 416 mg, 0.7 mmol), three times coevaporated with
toluene, was dissolved in anhydrous THF (4 mL) and cooled to −78
°C. Methyl lithium (0.70 mL, 1.12 mmol, 1.6 M in Et2O) was added
dropwise. After being stirred for 5 h at this temperature, TLC analysis
indicated complete conversion. The reaction was quenched with
saturated aq NH4Cl, and the suspension extracted with Et2O. The
organic layer was washed with water, and the combined aqueous layers
were extracted with Et2O. The combined organic layers were washed
with brine, dried over MgSO4, filtered, and concentrated. The residue
was purified by silica gel column chromatography (4−5% EtOAc/
pentane), yielding the title compound (278 mg, 0.51 mmol, 72% yield)
1
petroleum ether). β-Anomer: H NMR (400 MHz, CDCl3) δ 7.49−
7.11 (m, 20H, CHAr Ph, CHAr Bn), 4.85 (d, J = 4.0 Hz, 1H, C-1),
4.67−4.39 (m, 6H, 6× CHH Bn), 4.40−4.35 (m, 1H, C-4), 4.09−4.05
(m, 1H, C-3), 3.99 (dd, J = 4.1, 1.4 Hz, 1H, C-2), 3.92 (dd, J = 10.0,
5.3 Hz, 1H, C-5a), 3.87 (dd, J = 9.9, 6.2 Hz, 1H, C-5b); 13C NMR
(101 MHz, CDCl3) δ 140.7 (Cq Ph), 138.3, 138.0, 137.7 (Cq Bn),
128.5, 128.5, 128.4, 128.4, 128.4, 128.3, 127.9, 127.9, 127.8, 127.7,
127.7, 127.6, 127.5, 127.5, 127.5, 126.7 (CHAr Ph, CHAr Bn), 89.6 (C-
2), 86.1 (C-1), 83.3 (C-3), 80.7 (C-4), 73.6, 72.0, 71.5 (3× CH2 Bn),
1
as a colorless oil: Rf = 0.7 (10/90 EtOAc/pentane); H NMR (400
MHz, CDCl3) δ 7.36−7.23 (m, 15H, CHAr Bn), 4.64 (d, J = 11.7 Hz,
1H, CHH Bn), 4.60 (d, J = 11.7 Hz, 1H, CHH Bn), 4.55 (d, J = 11.9
Hz, 1H, CHH Bn), 4.51 (d, J = 12.4 Hz, 1H, CHH Bn), 4.48 (d, J =
12.1 Hz, 1H, CHH Bn), 4.44 (d, J = 12.0 Hz, 1H, CHH Bn), 4.11−
4.07 (m, 1H, C-4), 4.07 (d, J = 3.7 Hz, 1H, C-2), 3.94 (dd, J = 5.6, 3.7
2
68.4 (C-5), JH1−C2 = −2.3 Hz. α-Anomer: 1H NMR (400 MHz,
4563
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
20
Hz, 1H, C-3), 3.68 (dd, J = 10.3, 3.7 Hz, 1H, C-5a), 3.64 (dd, J = 10.3,
5.6 Hz, 1H, C-5b), 2.17 (s, 3H, COCH3), 0.89 (t, J = 7.9 Hz, 9H, CH3
TES), 0.61−0.48 (m, 6H, CH2 TES); 13C NMR (101 MHz, CDCl3) δ
210.0 (CO), 138.3, 138.3, 137.7 (Cq Bn), 128.5, 128.4, 127.9, 127.8,
127.7, 127.6 (CHAr Bn), 84.6 (C-2), 81.8 (C-3), 73.6, 73.3, 72.9 (3×
CH2 Bn), 72.6 (C-4), 72.3 (C-5), 27.9 (COCH3), 7.0 (CH3 TES),
4.9 (CH2 TES); IR (neat) 606, 694, 731, 783, 908, 980, 1003, 1026,
1074, 1088, 1207, 1283, 1350, 1414, 1454, 1715, 2874, 2911, 2951;
2951; [α]D 3.2 (c = 1, CHCl3); HRMS (ESI) [M + H+] calcd for
C38H46O5Si 611.31873, found 611.31904.
2,3,5-Tri-O-benzyl-1-deoxy-α-1-phenyl-D-lyxofuranose (33).
2,3,5-Tri-O-benzyl-4-O-triethylsilyl-1-phenyl-D-lyxose (17, 70 mg, 0.12
mmol), three times coevaporated with toluene, was dissolved in
anhydrous DCM (1.4 mL) and cooled to −78 °C. Triethylsilane (24
μL, 0.15 mmol) was added before dropwise addition of BF3·OEt2 (38
μL, 0.30 mmol). After being stirred for 1 week at this temperature,
TLC analysis showed partial conversion, with the remainder still being
the starting material. The reaction was quenched saturated aq
NaHCO3, and the suspension was extracted with Et2O. The organic
layer was washed with water and brine, and the combined aqueous
layers were extracted with Et2O. The combined organic layers were
dried over MgSO4, filtered, and concentrated. The residue was purified
by silica gel column chromatography (8−10% EtOAc/pentane),
yielding the title compound (43 mg, 0.09 mmol, 77% yield) as a
single diastereomer: Rf = 0.60 (10/90 EtOAc/pentane); 1H NMR
(400 MHz, CDCl3) δ 7.40−7.21 (m, 18H, CHAr Bn, CHAr Ph), 7.18−
7.09 (m, 2H, 2× CHAr Ph), 5.05 (d, J = 7.7 Hz, 1H, C-2), 4.79 (d, J =
11.8 Hz, 1H, CHH Bn), 4.63 (d, J = 11.7 Hz, 1H, CHH Bn), 4.61 (d, J
= 11.9 Hz, 1H, CHH Bn), 4.53 (d, J = 11.9 Hz, 1H, CHH Bn), 4.50−
4.42 (m, 2H, C-4, CHH Bn), 4.40 (d, J = 12.1 Hz, 1H, CHH Bn), 4.17
(t, J = 4.1 Hz, 1H, C-3), 3.91−3.83 (m, 2H, C-2, C-5a), 3.78 (dd, J =
9.8, 6.4 Hz, 1H, C-5b); 13C NMR (101 MHz, CDCl3) δ 141.1 (Cq
Ph), 138.4, 138.3, 137.8 (Cq Bn), 128.5, 128.5, 128.5, 128.4, 127.8,
20
[α]D 0.8 (c = 1, CHCl3); HRMS (ESI) [M + H+] calcd for
C33H44O5Si 549.30308, found 549.30337.
α-1-Methyl-1-deoxy-2,3,5-tri-O-benzyl-D-lyxofuranose (32).
1-Methyl-2,3,5-tri-O-benzyl-4-O-triethylsilyl-D-lyxose (16, 137 mg,
0.25 mmol), three times coevaporated with toluene, was dissolved in
anhydrous DCM (3.1 mL) and cooled to −78 °C. Triethylsilane (52
μL, 0.33 mmol) was added before dropwise addition of BF3·OEt2 (82
μL, 0.65 mmol). After being stirred for 1 week at this temperature,
TLC analysis showed partial conversion, with the remainder still being
the starting material. The reaction was quenched with saturated aq
NaHCO3, allowed to warm to room temperature, and extracted with
Et2O. The organic layer was washed with water and brine, and the
combined aqueous layers were extracted with Et2O. The combined
organic layers dried over MgSO4, filtered, and concentrated. The
residue was purified by silica gel column chromatography (8−13%
EtOAc/pentane), yielding the title compound (46 mg, 0.11 mmol,
44% yield) as a single diastereomer: Rf = 0.55 (20/80 EtOAc/
1
127.7, 127.6, 126.2 (CHAr Ph, CHAr Bn), 86.2 (C-2), 81.3 (C-1), 79.6
pentane); H NMR (400 MHz, CDCl3) δ 7.37−7.24 (m, 15H, CHAr
2
(C-4), 77.5 (C-3), 73.7, 73.6, 72.6 (CH2 Bn), 69.2 (C-5), JH1−C2
=
Bn), 4.72 (d, J = 11.8 Hz, 1H, CHH Bn), 4.64−4.56 (m, 3H, 3× CHH
Bn), 4.50 (d, J = 11.9 Hz, 1H, CHH Bn), 4.47 (d, J = 12.0 Hz, 1H,
CHH Bn), 4.27−4.21 (m, 1H, C-4), 4.19−4.11 (m, 1H, C-1), 4.08 (t,
J = 4.2 Hz, 1H, C-4), 3.74 (dd, J = 9.8, 5.8 Hz, 1H, C-5a), 3.67 (dd, J =
9.7, 6.7 Hz, 1H, C-5b), 3.55 (dd, J = 7.7, 4.2 Hz, 1H, C-2), 1.24 (d, J =
6.3 Hz, 3H, CH3); 13C NMR (101 MHz, CDCl3) δ 138.5, 138.3, 138.0
(Cq Bn), 128.5, 128.4, 128.4, 128.0, 127.9, 127.9, 127.8, 127.7, 127.7
(CHAr Bn), 85.1 (C-2), 78.6 (C-4), 77.2 (C-3), 75.6 (C-1), 73.6, 73.5,
−5.6 Hz; [α]D20 = 25.8 (c = 1, CHCl3); IR (neat) 652, 694, 729, 752,
833, 887, 920, 961, 984, 1007, 1026, 1053, 1084, 1101, 1144, 1159,
1204, 1360, 1452, 1493, 2859, 2884, 2899, 2949, 3030; HRMS (ESI)
[M + H+] calcd for C32H32O4 481.23734, found 481.23746.
ASSOCIATED CONTENT
■
S
* Supporting Information
2
20
72.7 (CH2 Bn), 69.1 (C-5), 19.4 (CH3), JH1−C2 = −3.7 Hz; [α]D
=
General experimental information and H, 13C, and 2D NMR
1
26.6 (c = 1.1, CHCl3); IR (neat) 696, 735, 810, 845, 914, 932, 1003,
1026, 1061, 1086, 1146, 1206, 1273, 1312, 1344, 1366, 1452, 1497,
1722, 2866, 2924, 2968; HRMS (ESI) [M + H+] calcd for C27H30O4
419.22169, found 419.22149.
spectra of all new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: filippov@chem.leidenuniv.nl.
*E-mail: jcodee@chem.leidenuniv.nl.
2,3,5-Tri-O-benzyl-4-O-triethylsilyl-1-phenyl-D-lyxose (17).
Bromobenzene (76 μL, 0.72 mmol) was dissolved in anhydrous
THF (1 mL) and cooled to −78 °C. n-Butyl lithium (0.45 mL, 0.72
mmol, 1.6 M in hexanes) was added dropwise, and the mixture was
stirred at this temperature for 30 min. 2,3,5-Tri-O-benzyl-4-O-
triethylsilyl-D-lyxonic N,O-dimethylhydroxylamide (15, 342 mg, 0.58
mmol), three times coevaporated with toluene, in dry THF (4 mL)
was added slowly, and the reaction mixture was stirred for 18 h. TLC
analysis indicated full consumption of the starting material, and the
reaction was quenched with saturated aq NH4Cl. The suspension was
extracted with Et2O, and the organic layers were washed with water
and brine. The combined aqueous layers were extracted with Et2O,
and the combined organic layers were dried over MgSO4, filtered, and
concentrated. The residue was purified by silica gel column
chromatography (4−7% EtOAc/pentane) to provide the title
compound (210 mg, 0.34 mmol, 60% yield) as a colorless oil: Rf =
■
Author Contributions
§E.R.v.R. and P.v.D. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was financially supported by The Netherlands
Organization for Scientific Research (NWO). Grid computing
resources have been provided by BigGrid.nl, NIKHEF, and
LGI. We kindly acknowledge Dr. Mark Somers for technical
support.
1
0.80 (10/90 EtOAc/pentane); H NMR (400 MHz, CDCl3) δ 8.06−
7.98 (m, 2H, m-CHAr Ph), 7.53−7.45 (m, 1H, p-CHAr Ph), 7.39−7.33
(m, 2H, o-CHAr Ph), 7.32−7.21 (m, 10H, CHAr Bn), 7.16−7.09 (m,
3H, CHAr Bn), 6.95−6.89 (m, 2H, CHAr Bn), 5.14 (d, J = 7.4 Hz, 1H,
C-2), 4.56 (d, J = 11.2 Hz, 1H, CHH Bn), 4.45 (d, J = 12.0 Hz, 1H,
CHH Bn), 4.42−4.34 (m, 3H, 3× CHH Bn), 4.29 (d, J = 11.4 Hz, 1H,
CHH Bn), 4.27−4.20 (m, 1H, C-4), 4.04 (dd, J = 7.4, 3.1 Hz, 1H, C-
3), 3.59−3.50 (m, 2H,C-5), 0.91 (t, J = 7.9 Hz, 9H,CH3 TES), 0.63−
0.52 (m, 6H, CH2 TES); 13C NMR (101 MHz, CDCl3) δ 200.4 (C
O), 138.1, 137.8, 137.5, 137.1 (Cq Ph, Cq Bn), 133.2 (p-CHAr Ph),
129.0 (m-CHAr Ph), 128.4, 128.3, 128.1, 128.0, 127.8, 127.8, 127.6,
127.4 (CHAr Bn, o-CHAr Ph), 80.9 (C-3), 79.7 (C-2), 74.2, 73.2, 72.0
(CH2 Bn), 71.7 (C-4), 71.5 (C-5), 7.1 (CH3 TES), 5.2 (CH2 TES); IR
(neat) 652, 694, 735, 802, 824, 845, 812, 1001, 1026, 1070, 1090,
1177, 1207, 1248, 1269, 1315, 1362, 1395, 1452, 1688, 1722, 2874,
REFERENCES
■
(1) (a) Wu, Q. P.; Simons, C. Synthesis 2004, 1533−1553.
(b) Stambasky, J.; Hocek, M.; Kocovsky, P. Chem. Rev. 2009, 109,
6729−6764. (c) Wellington, K. W.; Benner, S. A. Nucleosides,
Nucleotides Nucleic Acids 2006, 25, 1309−1333. (d) Charette, M.;
Gray, M. W. IUBMB Life 2000, 49, 341−351. (e) Levy, D. E.
Strategies towards C-Glycosides. In The Organic Chemistry of Sugars;
Levy, D. E., Fugedi, P., Eds.; CRC Press: Boca Raton, FL, 2006; pp
̈
269−348. (f) C-Glycoside Synthesis; Postema, M. H. D., Ed.; CRC
Press: Boca Raton, FL, 1995. (g) Meng, W.; Ellsworth, B. A.; Nirschl,
A. A.; McCann, P. J.; Patel, M.; Girotra, R. N.; Wu, G.; Sher, P. M.;
Morrison, E. P.; Biller, S. A.; Zahler, R.; Deshpande, P. P.; Pullockaran,
4564
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565

The Journal of Organic Chemistry
Article
A.; Hagan, D. L.; Morgan, N.; Taylor, J. R.; Obermeier, M. T.;
Humphreys, W. G.; Khanna, A.; Discenza, L.; Robertson, J. G.; Wang,
A.; Hang, S.; Wetterau, J. R.; Janovitz, E. B.; Flint, O. P.; Whaley, J. M.;
Washburn, W. N. J. Med. Chem. 2008, 51, 1145−1149. (h) Roy-
Burman, S.; Roy-Burman, P.; Visser, D. W. Cancer Res. 1968, 28,
1605−1610. (i) Hosoya, T.; Takashiro, E.; Matsumoto, T.; Suzuki, K.
J. Am. Chem. Soc. 1994, 116, 1004−1015.
(2) (a) Cortezano-Arellano, O.; Melendez-Becerra, C. A.; Cortes, F.;
Sartillo-Piscil, F.; Cordero-Vargas, A. Carbohydr. Res. 2014, 393, 51−
59. (b) Calzada, E.; Clarke, C. A.; Roussinbouchard, C.; Wightman, R.
H. J. Chem. Soc., Perkin Trans. 1 1995, 517−518. (c) Liu, W. M.; Wise,
D. S.; Townsend, L. B. J. Org. Chem. 2001, 66, 4783−4786.
(d) Centrone, C. A.; Lowary, T. L. J. Org. Chem. 2002, 67, 8862−
8870. (e) Sollogoub, M.; Fox, K. R.; Powers, V. E. C.; Brown, T.
Tetrahedron Lett. 2002, 43, 3121−3123. (f) Terauchi, M.; Abe, H.;
Matsuda, A.; Shuto, S. Org. Lett. 2004, 6, 3751−3754. (g) Peifer, M.;
Berger, R.; Shurtleff, V. W.; Conrad, J. C.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2014, 136, 5900−5903. (h) Metobo, S. E.; Xu, J.; Saunders,
O. L.; Butler, T.; Aktoudianakis, E.; Cho, A.; Kim, C. U. Tetrahedron
Lett. 2012, 53, 484−486.
́
(3) Hernandez, A. R.; Kool, E. T. Org. Lett. 2011, 13, 676−679.
(4) (a) Larsen, C. H.; Ridgway, B. H.; Shaw, J. T.; Woerpel, K. A. J.
Am. Chem. Soc. 1999, 121, 12208−12209. (b) Smith, D. M.; Tran, M.
B.; Woerpel, K. A. J. Am. Chem. Soc. 2003, 125, 14149−14152.
(c) Smith, D. M.; Woerpel, K. A. Org. Lett. 2004, 6, 2063−2066.
(d) Larsen, C. H.; Ridgway, B. H.; Shaw, J. T.; Smith, D. M.; Woerpel,
K. A. J. Am. Chem. Soc. 2005, 127, 10879−10884.
(5) Jensen, H. H.; Lyngbye, L.; Jensen, A.; Bols, M. Chem.Eur. J.
2002, 8, 1218−1226.
(6) van Rijssel, E. R.; van Delft, P.; Lodder, G.; Overkleeft, H. S.; van
der Marel, G. A.; Filippov, D. V.; Codee
2014, 53, 10381−10385.
́
, J. D. C. Angew. Chem., Int. Ed.
(7) Daranas, A. H.; Napolitano, J. G.; Gavin, J. A.; Garcia, C.; Norte,
M.; Fernandez, J. J. Chem.Eur. J. 2011, 17, 6338−6347.
(8) Rhoad, J. S.; Cagg, B. A.; Carver, P. W. J. Phys. Chem. A 2010,
114, 5180−5186.
(9) Altona, C.; Sundaralingan, M. J. Am. Chem. Soc. 1972, 94, 8205−
8212.
(10) Cremer, D.; Pople, J. A. J. Am. Chem. Soc. 1975, 97, 1354−1358.
(11) Frisch, M. J.; et al. Gaussian 03, revision E.01; Gaussian, Inc.:
Wallingford, CT, 2004.
(12) The graphs of the aldofuranosyl, methyl ketofuranosyl, and
phenyl ketofuranosyl oxocarbenium ions have not been normalized
with respect to each other and can therefore not be compared to each
other.
(13) (a) Allinger, N. L.; Hirsch, J. A.; Miller, M. A.; Tyminski, I. J. J.
Am. Chem. Soc. 1969, 91, 337−343. (b) Dorigo, A. E.; Pratt, D. W.;
Houk, K. N. J. Am. Chem. Soc. 1987, 109, 6591−6600.
(14) Kong, J.; et al. Spartan ’04; Wavefunction Inc.: Irvine, CA, 2004.
(15) Kofink, C. C.; Knochel, P. Org. Lett. 2006, 8, 4121−4124.
4565
DOI: 10.1021/acs.joc.5b00419
J. Org. Chem. 2015, 80, 4553−4565