Stereoelectronic factors in the stereoselective epoxidation of glycals and 4-deoxypentenosides

ARTICLE

pubs.acs.org/joc

Stereoelectronic Factors in the Stereoselective Epoxidation of Glycals

and 4-Deoxypentenosides

Laura Alberch, Gang Cheng, Seung-Kee Seo, Xuehua Li, Fabien P. Boulineau, and Alexander Wei*

Department of Chemistry, Purdue University, 560 Oval Drive, West Lafayette, Indiana 47907-2084, United States

S Supporting Information

b

ABSTRACT: Glycals and 4-deoxypentenosides (4-DPs), un-

saturated pyranosides with similar structures and reactivity

proﬁles, can exhibit a high degree of stereoselectivity upon

epoxidation with dimethyldioxirane (DMDO). In most cases,

the glycals and their corresponding 4-DP isosteres share the

same facioselectivity, implying that the pyran substituents are

largely responsible for the stereodirecting eﬀect. Fully substi-

tuted dihydropyrans are subject to a “majority rule”, in which

the epoxidation is directed toward the face opposite to two of

the three groups. Removing one of the substituents has a

variable eﬀect on the epoxidation outcome, depending on its position and also on the relative stereochemistry of the remaining

two groups. Overall, we observe that the greatest loss in facioselectivity for glycals and 4-DPs is caused by removal of the C3 oxygen,

followed by the C5/anomeric substituent, and least of all by the C4/C2 oxygen. DFT calculations based on polarized-π frontier

molecular orbital (PPFMO) theory support a stereoelectronic role for the oxygen substituents in 4-DP facioselectivity, but less

clearly so in the case of glycals. We conclude that the anomeric oxygen in 4-DPs contributes toward a stereoelectronic bias in

facioselectivity whereas the C5 alkoxymethyl in glycals imparts a steric bias, which at times can compete with the stereodirecting

eﬀects from the other oxygen substituents.

’ INTRODUCTION

pyranosides with an ^L-altro conﬁguration,⁹whereas 4R-EPs derived

from β-glucosides react with organozinc reagents for syn-selective

ring openings to produce pyranosides with ^L-ido conﬁguration.¹²

These studies show that the reactivity proﬁles of 4-EPs are similar to

that of glycal epoxides when treated with carbon nucleophiles, en

route to the stereoselective formation of C-glycosides.⁶

The class of 1,2-unsaturated sugar derivatives known as glycals

is a widely valued source of starting materials in the synthesis of

carbohydrates and their derivatives.^1ꢀ3Glycals can be converted

by chemical degradation or stereoselective rearrangement into a

variety of chiral synthons, but are especially useful as precursors

of 1,2-anhydropyranosides, often referred to as glycal epoxides.

The eﬃciency of this synthetic conversion has not always been so

eﬃcient: Aside from a few classic examples such as Brigl’s

anhydride (3,4,6-tri-O-acetyl-1,2-anhydroglucose),⁴a general

method for generating glycal epoxides was not available until

little more than 20 years ago, when Halcomb and Danishefsky

demonstrated the use of dimethyldioxirane (DMDO) as a mild

and stereoselective oxygen transfer agent.⁵The synthetic poten-

tial of glycal epoxides has grown considerably since then, and is

now considered as a major route toward the synthesis of O- and

C-glycosides^2,6and glycoconjugates,⁷as well as highly substi-

tuted tetrahydropyrans in natural product syntheses.⁸

4-Deoxypentenosides (4-DPs) are unsaturated pyranoside deriva-

tives bearing a strong resemblance to glycals, but the sp³carbon next

to the ring oxygen (i.e., C1 or C5) supports heteroatomic substituents

rather than carbon (Figure 1).⁹Like glycals, 4-DPs can be oxidized by

DMDO into either 4R- or 4β-epoxypyranosides (4-EPs) in a highly

facioselective manner.^10,11The 4-EPs are stable in solution, but can

react with carbon nucleophiles to generate rare or unnatural sugars:

For example, 4β-EPs derived from R-glucosides react with organo-

cuprates for anti-selective (S_N2) ring openings to generate novel

Our earlier studies have shown that the facioselectivity of

4-DP epoxidation can be predicted on the basis of a “majority

rule”, in which the oxygen is delivered anti to two of the three

substituents on the dihydropyran ring, regardless of their relative

position.¹⁰This selectivity is not readily explained by previously

described steric or stereoelectronic eﬀects,¹³and transition state

geometries derived from density functional theory (DFT) calcu-

lations do not suggest any torsional eﬀects or hydrogen bonding

interactions that might explain the high levels of stereoselectivity.

We have attributed the observed facioselectivities to the asym-

metric polarization of the π-bond itself, a notion that is

supported by polarized-π molecular orbital (PPFMO) analysis

using DFT for energy minimization.^10,14This has led us to

consider whether the facioselectivity of DMDO-mediated oxida-

tion of glycals is also attributable to a polarized-π eﬀect: while the

allylic C3 oxygen has been argued to have a major stereodirecting

inﬂuence,⁵the remaining substituents can also contribute toward

a stereoelectronic asymmetry in π-bond reactivity.

Received: December 1, 2010

Published: March 18, 2011

r

2011 American Chemical Society

2532

dx.doi.org/10.1021/jo102382r J. Org. Chem. 2011, 76, 2532–2547

|

The Journal of Organic Chemistry

ARTICLE

Scheme 1. Synthesis of D-Allal Derivatives^a

^aReagents and conditions: (a) PhSH, BF₃Et₂O, CH₂Cl₂, ꢀ78 °C

3

(69%); (b) NaOMe, MeOH, rt; (c) PhCH(OMe)₂, p-TsOH, rt; (d)

DMDO, CH₂Cl₂, ꢀ78 °C; (e) Et₂NH, THF, rt; (f) BnBr, Bu₄NI, NaH,

DMF, rt (36% over 5 steps); (g, h) same as steps (d, e) (59% over 2

steps); (i, j) same as steps (b, f) (92% over 2 steps); (k) NaH, CS₂, MeI,

THF, rt; (l) Bu₃SnH, AIBN, DMF, 120 °C; (m) NaOMe, MeOH, rt; (n)

BnBr, Bu₄NI, NaH, THF, rt (22% over 4 steps).

Figure 1. Four diastereomeric glycals, and their 4-deoxypentenoside

isosteres.

PPFMO analysis has been previously used to rationalize stereo-

selective additions to glycals, although with mixed results.^15,16This

may be due in part to a less than perfect match between theory and

experiment, which involved charged species and the generation of

cationic intermediates upon reaction with acid or electrophiles (e.g.,

halonium or sulfenium ions). Perturbation theories such as PPFMO

are typically based on energy-minimized structures in the absence of

other reactants,¹⁷and may be more appropriate for investigating

reactions with early transition states that do not require a large

disturbance in the electronic or conformational ground state. In this

regard, the DMDO oxidation of alkenes may be an ideal case because

the nucleophilic π-bond experiences minimum distortion in electro-

nic structure, as demonstrated in several transition state analyses.^10,18

Nevertheless, the facioselectivity of glycal oxidation by DMDO has

not yet been examined in the context of PPFMO analysis.

In this paper we describe the facioselective epoxidation of a

series of glycals and their isosteric 4-DPs, using DMDO at low

temperatures. This includes a set of glycals and 4-DP derivatives

having only two substituents, whose stereochemical outcomes

can no longer be predicted by a “majority rule”. The systematic

removal of substituents at various positions enables us to address

the relative impact of each on facioselectivity, and provides a

useful testing ground for comparing experimental and computa-

tional results using PPFMO analysis. In particular, we wished to

determine whether PPFMO theory could provide a reliable

method for predicting the facioselectivity of these chiral dihy-

dropyrans, based on the ground state electronic structures of

their π-bonds.

Scheme 2. Synthesis of D-Gulal Derivatives^a

^aReagents and conditions: (a) PhSH, SnCl₄, CH₂Cl₂, ꢀ20 °C (98%);

(b) DMDO, CH₂Cl₂, ꢀ78 °C; (c) Et₂NH, THF, rt; (d) NaOMe,

MeOH, rt; (e) BnBr, Bu₄NI, NaH, THF, rt (76% yield over 4 steps); (f)

PhCH(OMe)₂, p-TsOH, rt (70% over 2 steps); (g) same as steps (b),

(c), and (e) (45% over 3 steps).

followed by saponification, acetalization of the 4,6-diol, and

benzylation of the remaining C3 alcohol. Perbenzylated allal 2

and 4-deoxyallal 3 were both prepared from 3,6-di-O-acetyl allal,

which was generated from the Ferrier rearrangement intermedi-

ate in 59% yield by DMDO oxidation and Et₂NH-mediated

[2,3]-sigmatropic rearrangement with simultaneous acyl migra-

tion from O-4 to O-3. Deacetylation followed by perbenzylation

afforded 2 in 92% yield over two steps, whereas BartonꢀMc-

Combie deoxygenation in DMF followed by an exchange of

protecting groups yielded 3 in 22% yield over 4 steps. We note

that the low isolated yield of the latter sequence was due in part to

the volatile nature of 3,6-di-O-acetyl-4-deoxyallal, the product

immediately following BartonꢀMcCombie deoxygenation, and

can likely be improved by a judicious choice of organic solvent

during extractive workup.

’ RESULTS AND DISCUSSION

Synthesis of Unsaturated Pyranosides. Allal derivatives 1

and 3 and ^D-gulal derivatives 4 and 6 were prepared respectively

from tri-O-acetyl-^D-glucal and tri-O-acetyl-D-galactal, based on a

synthetic sequence developed by Danishefsky and co-workers

(Schemes 1 and 2).¹⁹In brief, compound 1 was prepared in 36%

overall yield from triacetyl glucal by a Ferrier rearrangement,

Gulal derivatives 4 and 6 were prepared from tri-O-acetyl galactal,

as attempts to epimerize the C4 stereocenter of 3,6 di-O-acetyl allal

proved unsuccessful. Treatment with benzenethiol in the presence

of SnCl₄at ꢀ78 °C resulted in a Ferrier rearrangement product in

2533

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Scheme 3. Synthesis of 3- and 4-Deoxyglucal Derivatives^a

Scheme 5. Synthesis of 2,4- and 3,4-Dideoxypentenosides^a

^aReagents and conditions: (a) NaH, CS₂, MeI, THF, rt; (b) Bu₃SnH,

AIBN, PhCH₃, reﬂux (54% isolated yield for 8, 82% for 11); (c)

BH₃THF, Bu₂BOTf, THF, ꢀ78 °C; (d) BnBr, Bu₄NI, NaH, THF,

3

rt (73% over 2 steps); (e) BnBr, Bu₄NI, NaH, DMF, rt; (f) iBu₂AlH,

CH₂Cl₂, 0 °C; (g) same as step (e); (h) DDQ, tBuOH, pH 7 phosphate

buﬀer, CH₂Cl₂, rt (61% over 4 steps). PMB = p-methoxybenzyl; PMP =

p-methoxyphenyl.

^aReagents and conditions: (a) NaH, CS₂, MeI, THF, rt; (b) Bu₃SnH,

AIBN, PhCH₃, reﬂux; (c) AcOH/THF/H₂O, 45 °C (81% over 3 steps

for 18, 56% for 19, 50% for 24, 87% for 25); (d) TEMPO, BAIB, H₂O/

CH₂Cl₂, rt; (e) DMFDNPA, DMF, 200 °C (58% isolated yield over 2

steps for 20, 60% for 21 and 26, 28% for 27).

Scheme 4. Synthesis of 3-Deoxygalactal Derivatives^a

careful attention paid to the reaction time and the stoichiometry

of Bu₃SnH to avoid reductive desulfurization. Compound 13 was

then cleanly converted into 3-deoxygalactal 14 in 78% yield by

reductive elimination with use of lithium naphthalenide. The same

chemistry was also applied toward the synthesis of di-O-benzyl

derivative 15; in this case, acetal hydrolysis and benzylation were

performed prior to reductive elimination, to yield the desired

3-deoxygalactal in 40% overall yield from 13. It is worth noting that

the reductive lithiation approach toward 3-deoxygalactals is much

more eﬃcient than methods based on the vinylogous reduction

^aReagents and conditions: (a) NaH, CS₂, MeI, THF, rt; (b) Bu₃SnH,

AIBN, PhCH₃, reﬂux (75% over 2 steps); (c) Li-naphthalenide (2.5

equiv), THF, ꢀ40 °C (78%); (d) AcOH/THF/H₂O, 45 °C; (e) BnBr,

Bu₄NI, NaH, DMF, rt (66% over 2 steps); (f) same as step (c) (60%).

23

98% yield (8:1 R:β mixture, Scheme 2), followed by DMDO

oxidation, Et₂NH-mediated [2,3]-sigmatropic rearrangement, and

chromatographic separation to aﬀord the desired 4,6 di-O-acetyl

gulal (83% isolated yield); an exchange of protecting groups led to

tri-O-benzyl gulal 4 in 76% yield over four steps. 4,6-Benzylidene

derivative 5 was obtained from the same Ferrier rearrangement

product in 70% yield (8:1 R:β mixture), followed by the tandem

oxidationꢀsigmatropic rearrangement and benzylation of the C3

alcohol to aﬀord 6 in high stereochemical purity (45% isolated yield

over three steps).

of hex-2-enopyranosides with LiAlH₄or the allylic reduction of

galactals with heterogeneous catalysts,²⁴but the stoichiometry of

Li-naphthalenide (2.5 equiv) and reaction temperature (ꢀ40 °C)

need to be carefully controlled to avoid debenzylation.

The synthesis of fully substituted 4-DP derivatives (β-Glc, β-

Gal, R-Man, R-Glc; see Figure1) has been reported previously,^9,10

and provides the basis for the synthesis of the 2,4- and

3,4-dideoxypentenosides (DDPs) described above (Scheme 5).

2,4-DDPs were prepared from ^D-glucosides 16 and 17 by C2

deoxygenation using BartonꢀMcCombie conditions,²⁰followed

by acetal hydrolysis to yield 4,6-diols 18 and 19 respectively in

high yields. Each of these was oxidized to a glucuronic acid using

catalytic tetramethyl-1-piperidine oxide (TEMPO)²⁵with bis-

(acetoxy)iodobenzene (BAIB) as the stoichiometric oxidant,²⁶

then heated under sealed-vessel conditions with N,N-dimethyl-

formamide dineopentyl acetal (DMFDNA), a reagent developed

by Eschenmoser for the decarboxylative elimination of

β-hydroxyacids.²⁷This two-step sequence cleanly produced 2,4-

DDPs 20 and 21 in 58% and 60% yields, respectively. 3,4-DDPs

were prepared from ^D-glucosides 22 and 23 in a nearly identical

fashion: C3 deoxygenation and acetal hydrolysis yielded 4,6-diols

24 and 25, which were subjected to TEMPO/BAIB oxida-

tionꢀdecarboxylative elimination to generate 3,4-DDPs 26 and

27 in 60% and 28% overallyields from 22 and 23. Again, the yields

were adversely aﬀected in the latter case by product volatility.

Lastly, pentopyranose derivatives 3,4-di-O-benzyl-^D-xylal and -

L-arabinal (28 and 29) were prepared by using the reductive

lithiation method described above. These compounds have been

Deoxyglucal derivatives 8, 9, and 11 were synthesized from the

common ^D-glucal precursor 7 (Scheme 3). Deoxygenation per-

formed under standard BartonꢀMcCombie conditions²⁰produced

3-deoxyglucal 8 in 54% isolated yield, as well as a C3 allylstannane

byproduct (36% yield). This unexpected byproduct is not without

precedent: glycal derivatives have been converted into allylstannanes

upon treatment with Bu₃SnH and photochemical activation.²¹

Regioselective cleavage of the 4,6-anisylidene acetal in 8 with dilute

Bu₂BOTf²²and subsequent benzylation produced 3-deoxyglucal

derivative 9 in 73% isolated yield. 4-Deoxyglucal 11 was prepared

straightforwardly from 7 in 50% overall yield by using standard

protecting group manipulations to obtain intermediate 10, followed

by BartonꢀMcCombie deoxygenation.

3-Deoxygalactal derivatives 14 and 15 were most eﬃciently

synthesized from 3-deoxygalactoside 13, as the corresponding

galactal was not amenable to various deoxygenation conditions

(Scheme 4). C3 deoxygenation of thiophenyl galactoside 12 was

achieved again by using BartonꢀMcCombie conditions, with

2534

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Scheme 6. Synthesis of D-Xylal and L-Arabinal^a

Table 2. Facioselectivities of 4-DP Epoxidation by DMDO

^aReagents and conditions: (a) Li-naphthalenide, THF, ꢀ40 °C

(quantitative).

Table 1. Facioselective Glycal Epoxidation by DMDO

^aReaction conditions: (A) DMDO (3 equiv), ꢀ55 °C, 2 days; (B)

DMDO (2 equiv), 0 °C, 1 h. ^bSee ref 10. ^cSee ref 9.

often be obtained in quantitative yields simply by concentrating

the reaction mixture, then used without further workup or

purification. Indeed, many of the epoxides generated by the

DMDO oxidation of glycals and 4-DPs (epoxyglycals and 4-EPs)

are stable at ambient temperatures, and react smoothly with good

nucleophiles under S_N2 conditions. However, several of the

disubstituted cases are susceptible to solvolysis and degrade at

an appreciable rate upon warming. Product degradation can be

minimized by preparing DMDO under “acetone-free” conditions

in CH₂Cl₂, followed by further drying and concentration (see

Experimental Section).³⁰Maximum stereoselectivity can be achieved

by conducting DMDO oxidations in CH₂Cl₂at ꢀ55 °C, followed

by concentration under reduced pressure at low temperatures

to avoid thermal decomposition. Epoxidation at ꢀ55 °C often

requires reaction times of 2ꢀ4 days to reach completion, but

ensures the highest possible facioselectivities for systematic

comparisons. It is also worth noting that benzylidene acetals

can be oxidized by DMDO above 0 °C, generating orthoesters

and benzoate esters as byproducts.^31,32

^aReaction conditions: (A) DMDO (3 equiv), ꢀ55 °C, 2 days; (B)

DMDO (2 equiv), 0 °C, 1 h. ^bSee ref 5. ^cRatio obtained with C3

silyl ether.

The facioselectivity of DMDO oxidation for the various glycals

and their isosteric 4-DPs were typically determined by peak

1

integration by using the epoxyacetal peaks in the H NMR

spectra (Tables 1ꢀ4). Facioselectivity was conﬁrmed in each

case by epoxide ring opening using strong nucleophiles such as

LiAlD₄and LiSEt (Tables 5 and 6, see the next section), and ¹H

NMR coupling constant analysis to establish the relative stereo-

chemistry of the S_N2 products. The high isolated yields of the

ring-opening products permitted us to validate the stereochem-

ical outcome of preceding DMDO addition. The tabulation of

the stereochemical outcomes enables us to establish the follow-

ing trends:

(i) All fully substituted glycals follow the empirical “majority

rule”, independent of the relationship among contiguous stereo-

centers (Table 1). The facioselectivities of most substrates are

10:1 or higher, the primary exceptions being gulal derivatives 4

previously prepared via reductive elimination of their peracety-

lated glycosyl halides, typically by treatment with zinc dust/

CuSO₄followed by benzylation.²⁸However, we found this

condition to be too harsh for the synthesis of 3,4-di-O-acetyl-^D-

arabinal, which was obtained in low yields and susceptible to

degradation upon storage at ꢀ20 °C. In contrast, subjecting

the thiophenyl glycosides of tribenzyl-^D-xylose and tribenzyl-L-

arabinose to Li-naphthalenide at ꢀ40 °C produced xylal 28 and

arabinal 29 in quantitative yields (Scheme 6), without undue

concern over their thermal stability.

Facioselectivity of Epoxidation. Oxidations with DMDO are

prized for their mildness and chemoselectivity,²⁹as products can

2535

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Table 3. Facioselectivities of Deoxyglycal Epoxidation by

DMDO

Table 4. Facioselectivities of Dideoxypentenoside Epoxida-

tion by DMDO

^aReaction conditions: (A) DMDO (3 equiv), ꢀ55 °C, 2 days; (B)

DMDO (2 equiv) 0 °C, 1 h. ^bReference 10. ^cReference 7.

1

DMDO (3 equiv), ꢀ55 °C, 1 day. ^bDetermined by H NMR peak

integration.

and 6, but even these exhibit modest selectivity and produce

epoxyglycals favoring the R-epoxide (3:2 and 3:1 ratio

respectively).³³It is worth mentioning that DMDO oxidation

of the gulals is sluggish compared with that of other diastereo-

mers and requires a reaction time of 4 days to reach completion

at ꢀ55 °C, implying a less electron-rich π-orbital.

perbenzylated congeners (2, 4, 9, and 15). However, gulal 6 and

3-deoxyglucal 8 have modestly higher facioselectivities than 4

and 9, respectively, whereas 3-deoxygalactal 14 has a lower

facioselectivity than 15.

It is worth pointing out that while the allylic C3 oxygen has the

strongest stereodirecting eﬀect, it is certainly not the sole determi-

nant in facioselectivity for either the glycals or 4-DPs, as the

remaining substituents can compensate for its absence in the case

of cis-disubstituted dihydropyrans. It is also interesting to compare

the facioselectivity of the glycals and 4-DPs with xylal derivative 28

(Table 4). This pentose-derived dihydropyran exhibits a lower

stereochemical bias in its reaction with DMDO, indicating signiﬁ-

cant competition between the C2 and C3 substituents. Further-

more, comparison of xylal 28 with dideoxypentenosides 21 and 26

(Table 4) and R-glc-4-DP (Table 2) shows that the C1 oxygen by

itself has a weaker inﬂuence than either the C3 or C2 oxygen, yet can

provide synergistic support to the C2 oxygen to completely override

the allylic stereodirecting eﬀect.

Stereochemical Assignments Based on Epoxide Ring

Opening. The facioselectivities of DMDO oxidations listed in

Tables 1ꢀ4 were established by subjecting the epoxyglycals or

4-EPs to ring-opening reactions under S_N2 conditions (Tables 5

and 6), followed by acetylation in some cases to assist ¹H NMR

coupling constant analysis. We found deuteride addition to be

ideal for this purpose as it does not introduce new peaks to the ¹H

NMR spectrum, and its low electronegativity (essentially that

of H) supports large coupling constants for diaxial vicinal protons

(ii) The facioselectivities of the glycals essentially mirror those

previously observed for the 4-DPs (Table 2).^9,10We note that the

facioselectivity of R-glc-4-DP, the isosteric equivalent of gulal, is

temperature dependent and is high only at ꢀ55 °C, whereas that

of β-glc-4-DP, the isosteric equivalent of glucal, remains high even

at 0 °C.

(iii) For deoxygenated (disubstituted) glycals with uncon-

strained cis substituents (11 and 15) the DMDO oxidation

remains highly anti-selective (Table 3). This selectivity is also

maintained for the isosteric 4-DPs (20 and 27) and arabinal 29

(Table 4).

(iv) For trans-disubstituted dihydropyrans with an allylic C3

substituent (4-deoxyallal 3, 2,4-dideoxypentenoside 21, and xylal

28), the DMDO oxidation is directed anti to the C3 oxygen. In

contrast, removal of the C3 group causes a loss in stereodirecting

eﬀect; facioselectivity is modest at best, slightly favoring anti to

the C2/C4 oxygen.

(v) Modifying the conformational behavior of the ring has a

variable eﬀect on facioselectivity. Overall, glycal derivatives that

are constrained by a 4,6-benzylidene acetal (1, 6, 8, and 14)

exhibit similar preferences toward DMDO oxidation as their

2536

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Table 5. S_N2 Ring-Opening Products of Epoxides Derived

from Glycals^a

Table 6. S_N2 Ring-Opening Products of Epoxides Derived

from Dideoxypentenosides^a

^aRelative stereochemistry of major products (facioselectivity g10:1,

unless otherwise noted) conﬁrmed by ¹H NMR coupling constant

analysis (see Table S2, Supporting Information). ^bSee ref 10. ^c3:2 ratio.

^d5:1 ratio.

We note that the remaining coupling constants of the pyrano-

sides listed in Tables 5 and 6 did not always clearly indicate a

preference for the ⁴C₁chair conformation, as most of these values

were small (see the Experimental Section). On the other hand,

the J values of pyranosides 30 and 32 (derived from allal 2 and

gulal 4, respectively) are very similar to those of 31 and 33, which

are conformationally constrained by 4,6-benzylidene acetals and

so likely to favor low-energy chair conformations.

Comparison between Facioselective DMDO Addition and

PPFMO Analysis. The PPFMO approach is based on a perturba-

tion method that desymmetrizes the 2p orbitals of alkenes,

following an energy minimization step.^14,17A qualitative analysis

of stereoelectronic bias is made possible by introducing addi-

tional 1s functions near the lobes of each 2p orbital; linear

combination of these wave functions produces facially dependent

coefficients (c_Rand c_β) that describe the relative polarization in

electron density. The polarization of each 2p orbital is calculated

as p = |c_Rꢀ c_β|, and is assigned R or β according to the larger of

the two coefficients. It should be noted that c_Ris negative and c_β

is positive by convention (see below), but p is presented as a

positive value regardless of the polarization direction.

^aRelative stereochemistry of major products (facioselectivity g10:1,

unless otherwise noted) conﬁrmed by ¹H NMR coupling constant

analysis (see Table S1, Supporting Information). ^b3:2 ratio. ^c3:1 ratio.

^d1:1 ratio. ^e2:1 ratio. ^f4:1 ratio.

2

based on the parametrized Karplus equation.³⁴However, some

disubstituted species were less compatible with strong reducing

agents such as LiAlD₄, so were subjected instead to LiSEt in THF

or to methanolysis at low temperature, all of which proceeded

with inversion of configuration according to the large J_1,2or J_4,5

values (>8 Hz; see Tables S1 and S2, Supporting Information).³⁵

Although both 2p orbitals contribute toward the overall

polarization of the π-bond their relative inﬂuences are unlikely

to be equal, as the coeﬃcients of the C2 orbital in glycals (or C4

orbital in 4-DPs) are larger in value than those at the other end of

2537

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Figure 3. Electron density maps of the 4-deoxyglucal derivative

(analogue of 11) with the C3 methyl ether in the tg conformation

(left) or the gt conformation (right). In the case of gt, the electron

density map reveals an incidental (but superﬂuous) hyperconjugation

between the hybrid orbital at C2 and a lone pair on O3 (extended green

lobe), creating an artiﬁcial polarization in the β direction.

Figure 2. PPFMO analysis of the dimethyl ether of 4-deoxyallal

(analogue of 3). Pairs of 1s functions are superimposed onto the 2p_y

orbitals at C1 and C2 to produce asymmetric wave functions (χ), whose

coeﬃcients c_Rand c_βare used to derive p, the net electronic polarization

per orbital (in purple). 2p orbitals and added s-functions are spatially

separated for clarity, and ( values refer to the sign of the coeﬃcients for

each lobe (open/ﬁlled).

We note that substituting each phenyl (C₆H₅) unitwithHgreatly

reduces computational time and can be expected to have a mini-

mum impact on PPFMO analysis, as determined by preliminary

studies on various O3-substituted glycals (not shown). On the other

hand, we observe that relatively small changes in the conformation

of exocyclic substituents can have a strong inﬂuence on the

polarization outcome of the 2p orbitals. The hybrid χ orbital at

C2 (C4 in the case of 4-DPs) is especially sensitive to the dihedral

angle of the C3ꢀO3 bond because it can overlap with a nonbonding

electron lone pair on O3, causing the value of c_Rto be artiﬁcially

high. As a case in point, PPFMO analysis of 4-deoxyglucal

11 indicates a signiﬁcant polarization in the R direction when

the O3 methyl group adopts a trans-gauche (tg) orientation

(C2ꢀC3ꢀO3ꢀCH₃dihedral angle = þ180°), but an unusually

large polarization in the β direction is observed when the O3 methyl

group adopts a gauche-trans (gt) orientation (dihedral angle =

þ60°). Inspection of electron density maps conﬁrms that the latter

case is grossly distorted by the incidental overlap between χ(C2)

and O3, which does not exist in the tg conformer (Figure 3). The

gas-phase conformational energy of gt is also higher than that of tg

(by ca. 3 kcal/mol), so should be less favored during DMDO

oxidation because of potential steric interactions between the C3

substituent and the incoming oxidant. This leads us to conclude that

hyperconjugative interactions between χ(C2) and O3 are not

stabilizing and have no meaningful impact on the PPFMO analysis.

PPFMO analysis was performed on all disubstituted glycals

and DDPs in their permethylated forms, starting from their

lowest energy half-chair conformations (⁴H₅and ²H₁, re-

spectively). The use of low-energy conformations in PPFMO

analysis is appropriate and does not violate the Curtinꢀ

Hammett principle, as their geometries are very close to

those observed in previous transition state analyses of DMDO

addition.^10,18Overall, these yielded polarization values that

correlated well with the experimental outcomes from DMDO

oxidation (Table 7). As expected, the polarized-π eﬀect for

structures corresponding to cis-disubstituted glycals (11, 14,

and 15), 4-DPs (20 and 27), and arabinal 29 is clearly in line

with the high anti selectivities observed after epoxidation. In the

case of trans-disubstituted derivatives, those having a C3 oxygen

the double bond. Nevertheless, it is unnecessary to introduce a

weighting factor for each 2p orbital because the p values scale

with their coeﬃcients. For example, if the relative diﬀerence

between c_Rand c_βis 5%, then the corresponding p value for that

2p orbital is approximately 0.1c_β. Therefore, we consider the

direct comparison of p values at C1 and C2 to be the simplest

method of evaluating the polarized-π eﬀect in glycals, analogous

to the comparison of p values at C4 and C5 in 4-DPs.¹⁰By the

same token, one should be mindful that such comparisons are

intended to be qualitative and are applicable for correlation with

facioselectivity, but less appropriate for quantitative outcomes.

Here we use PPFMO analysis to address the facioselectivities

of DMDO addition to disubstituted glycals and dideoxypenteno-

sides (DDPs), several of which cannot be predicted straightfor-

wardly by the empirical “majority rule”. The application of

PPFMO analysis toward 4-DP facioselectivity has been pre-

viously described,¹⁰and is concisely illustrated above by using

dimethyl 4-deoxyallal, the electronic analogue of 3 (Figure 2). In

brief, the structure is ﬁrst subjected to energy mimimization by

using DFT calculations (B3LYP/6-31þG(d,p)), starting from

the idealized half-chair (⁴H₅) conformation. A pair of 1s orbitals

is then introduced at set distances above and below the sym-

metric lobes of each 2p orbital as previously described,¹⁴followed

by PPFMO analysis to generate hybrid wave functions χ with

asymmetric lobes, labeled here as χ(1R), χ(1β), χ(2R), and

χ(2β). Each wave function contains the facially sensitive coeﬃ-

cients c_Rand c_β, which are used to calculate the ﬁnal p values for

the desymmetrized 2p orbitals. In this case, the hybrid orbitals at

C1 and C2 are polarized in diﬀerent directions (R and β

respectively); however, the value of p(2) is greater than that of

p(1) so the overall polarization of the π-bond is in the β

direction, which is in accord with the experimental result.

2538

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

Table 7. PPFMO Analysis of Disubstituted Glycals and

Dideoxypentenosides (Permethylated)^a

b

deoxyglycal

atom c_R

c_β

p^c

exptl^d

4-deoxyglucal (11)

C1 ꢀ0.182 0.169 0.005 (R) R (>20:1)

C2 ꢀ0.739 0.692 0.067 (R)

4-deoxyallal (3)

C1 ꢀ0.156 0.131 0.007 (β) β (>20:1)

C2 ꢀ0.737 0.746 0.014 (β)

3-deoxyglucal (9), ⁴H₅

3-deoxyglucal (9), ⁵H₄

C1 ꢀ0.213 0.180 0.013 (R) neither (1:1)

C2 ꢀ0.726 0.713 0.019 (R)

C1 ꢀ0.169 0.214 0.017 (β) neither (1:1)

C2 ꢀ0.651 0.711 0.082 (β)

3-deoxyglucal, 4,6-acetal (8) C1 ꢀ0.188 0.204 0.006 (β) β (2:1)

C2 ꢀ0.676 0.674 0.003 (R)

3-deoxygalactal (15)

C1 ꢀ0.217 0.183 0.013 (R) R (>20:1)

C2 ꢀ0.723 0.674 0.069 (R)

Figure 4. PPFMO analysis of dimethyl ether analogues of 3-deoxy-

glucal (9) and 3-deoxy-β-glc (26), starting from alternate half-chair

conformations. For each 2p orbital, p(n), the net electronic polarization

is presented as a ﬁlled lobe (purple).

3-deoxygalactal, 4,6-acetal (14) C1 ꢀ0.134 0.095 0.009 (R) R (4:1)

C2 ꢀ0.454 0.430 0.021 (R)

dideoxypentenoside

atom

c_R

c_β

p^c

exptl^e

b

2-deoxy-β-glc (20)

C5

C4

C5

C4

ꢀ0.139 0.226 0.032 (β) R (>20:1)

ꢀ0.391 0.182 0.120 (R)

nor ⁴H₅is dominant under the DMDO oxidation conditions. A

parallel study was conducted on the analogue of the isosteric

2-deoxy-R-glc (21)

ꢀ0.201 0.180 0.008 (R) R (10:1)

ꢀ0.719 0.719 0.000

1

DDP, 3-deoxy-β-glc 26: again, PPFMO analysis of the H₂

conformation revealed the polarizations of the C4 and C5

orbitals to be essentially mirror images of those found in the

²H₁conformer (Figure 4, right).

3-deoxy-β-glc (26), ²H₁C5

ꢀ0.221 0.172 0.019 (R) β (3:2)

ꢀ0.679 0.690 0.015 (β)

C4

3-deoxy-β-glc (26), ¹H₂C5

C4

ꢀ0.074 0.162 0.021 (β) β (3:2)

ꢀ0.477 0.457 0.018 (R)

PPFMO analysis also yielded a strong correlation between π-bond

polarization and facioselective DMDO oxidation for fully substituted

4-DPs,¹⁰but was less successful in the case of fully substituted glycals.

In particular, we were unable to correlate the facioselective epoxida-

tion of ^D-glucal or D-allal derivatives (R or β respectively, see Table 1)

with the polarization values derived from their tri-O-methyl analogues

in their ⁴H₅conformations, even after taking into account the variable

eﬀects of exocyclic conformations. However, PPFMO analysis of

unprotected glucal and allal (i.e., 3,4,6-triols) produced net polariza-

tions in agreement with the experimental results, as well as with earlier

calculations derived from AM1 calculations.¹⁵This may indicate that

relatively remote eﬀects like O-alkylation can exert a signiﬁcant

electronic inﬂuence on the outcome of perturbation-based models

such as PPFMO, which limits its ability to predict stereochemical

outcomes for complex organic compounds. Nevertheless, we ﬁnd

PPFMO analysis to be a useful probe of stereoelectronic bias for

relatively simple molecules with multiple stereocenters, such as the

substituted dihydropyrans investigated here.

Glycals and 4-DPs may also be diﬀerent in their conforma-

tional behavior despite their apparent structural similarities,

which may have some subtle ramiﬁcations for the basis of their

facioselectivities. The C5 hydroxymethyl units in glycals are

sterically more demanding than the C1 alkoxy substituents in

4-DPs; the latter may even be predisposed toward a pseudoaxial

position due to the anomeric eﬀect.¹³The steric diﬀerence

implies that 4-DPs are more ﬂexible than glycals and experience

less torsional strain when adopting transition state geometries,

possibly reducing their facioselectivities. Recent transition state

analyses of the DMDO oxidation of simple glycal and 4-DP

derivatives support this notion: the R/β selectivity for glycals is

deﬁned by a diﬀerence in activation energy (ΔΔG^q) on the order

of 3 kcal/mol,^18dwhereas that for 4-DPs is closer to 2 kcal/mol.¹⁰

Nevertheless, glycals and 4-DPs are clearly comparable in their

facioselectivities (Tables 1 and 2), leading us to posit that the

3-deoxy-R-glc (27)

C5

C4

C1

C2

C1

C2

ꢀ0.230 0.184 0.019 (R) β (>20:1)

ꢀ0.639 0.691 0.069 (β)

D-xylal (28)

ꢀ0.077 0.129 0.011 (β) R (5:1)

ꢀ0.311 0.200 0.056 (R)

L-arabinal (29)

ꢀ0.174 0.221 0.019 (R) R (>20:1)

ꢀ0.709 0.753 0.064 (R)

^aAll structures optimized by DFT-B3LYP calculations (6-31þG(d,p))

prior to insertion of s-functions. Unless otherwise stated, glycals and

DDPs were optimized starting from their respective ⁴H₅and ²H₁

conformations. ^bEach coeﬃcient is calculated as the linear combination

of s-function and 2p_y; ( values refer to the sign of the coeﬃcients for

each lobe. ^cNet polarization of each orbital in parentheses. ^dR:β

selectivities from Table 3. ^eR:β selectivities from Table 4.

such as 4-deoxyallal 3, 2-deoxy-R-glc-4-DP 21, and xylal 28 also

exhibit polarizations that are well matched with the observed

facioselectivities. On the other hand, the p values nearly cancel

each other for analogues corresponding to 3-deoxyglucal 8 (4,6-

acetal derivative) and 3-deoxy-β-glc-4-DP 26, giving rise to

ambiguous interpretations with respect to facioselectivity. This

is also reﬂected by the experimental results, which indicate

facioselectivites of 2:1 or less.

In the case of 3-deoxyglucal 9 the experimental results do not

suggest any stereochemical preferences, but the ⁴H₅conforma-

tion of the 3,6-di-O-methyl analogue produces p values predict-

ing R facioselectivity. However, energy minimization and

5

PPFMO analysis of the alternate H₄half-chair conformation

indicates both C1 and C2 orbitals to be polarized in the

β direction, opposite that of the ⁴H₅conformer (Figure 4, left).

The energy diﬀerence between conformations (ΔΔG_conf) can be

expected to be less than 1 kcal/mol,^18dmeaning that neither ⁵H₄

2539

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

conformational ﬂexibility of 4-DPs may be compensated for by

the contribution of the C1 oxygen toward a stereoelectronic bias

in reactivity, namely the polarized-π eﬀect.

treated with potassium persulfate (120 g) in five portions at 3 min

intervals. After the last addition, a moderate vacuum (20ꢀ30 mmHg)

was applied and the DMDO solution was condensed in a recovery flask

(250 mL) at ꢀ78 °C for 2 h. The DMDO was decanted from excess ice

into a precooled flask, further dried over K₂CO₃for a minimum of 10

min, then filtered to obtain a DMDO solution as a pale yellow liquid.

The volume and concentration of the DMDO solution were typically

100 mL and 0.08 M respectively, as determined by NMR titration with

thioanisole.

DMDO in “acetone-free” CH₂Cl₂was prepared by diluting freshly

distilled DMDO as described above (50 mL) with cold deionized water

(50 mL), followed by extraction with CH₂Cl₂(2 ꢁ 5 mL).³⁰The

organic extracts were washed at 5 °C with a 0.01 M phosphate buﬀer, pH

7 (15 mL), then cooled to ꢀ78 °C for 1 h to produce a thin layer of ice at

the surface, which was removed by hand. The DMDO solution (0.4 M in

CH₂Cl₂as determined by thioanisole oxidation) was carefully trans-

ferred via syringe or cannula to a cold, dry ﬂask, and stored at ꢀ20 °C for

up to two weeks.

In summary, we ﬁnd that 4-DPs and their isosteric glycals have

similar facioselectivities, as demonstrated by DMDO oxidation at

low temperatures. Epoxidation is highly stereoselective for the

fully substituted 4-DPs and nearly all of the glycals, as well as for

many disubstituted derivatives. In most cases, the high facio-

selectivities correlate well with a polarization of the alkene 2p

orbitals by nearby oxygen substituents, as elucidated by PPFMO

analysis. We ﬁnd that while the allylic C3 oxygen provides the

strongest polarization eﬀect, the remaining exocyclic oxygens are

also signiﬁcant and can even override the allylic substituent. The

C5 hydroxymethyl unit in glycals is less likely to contribute

toward π-bond polarization and may even oppose it in some

cases, but may inﬂuence facioselectivity instead through diﬀer-

ential torsional strain. Conversely, the C1 alkoxy unit in 4-DPs is

sterically less demanding but contributes signiﬁcantly toward the

polarized-π eﬀect.

DMDO Epoxidation. In a typical reaction, a solution of 4-deoxy-

pentenoside (55.1 mg, 0.123 mmol) in CH₂Cl₂(1.5 mL) was treated

at ꢀ55 °C with a precooled solution of DMDO (0.92 mL, 0.4 M in

CH₂Cl₂), stirred for 2 days, then concentrated under reduced pressure

at ꢀ55 °C for 15 min. The cooling bath was removed, then further

concentrated under reduced pressure at 0 °C for 30 min. The epoxide

was used immediately without further purification.

’ EXPERIMENTAL SECTION

General Experimental Methods. See the Supporting

Information.

Epoxide S_N2 Ring Opening. Three methods were developed for

the nucleophilic ring-opening of epoxypyranosides, using LiAlD₄

(procedures 1 and 2) or MeOH (procedure 3). Procedure 1 was applied

toward derivatives bearing benzylidene acetals, whereas procedure 2 was

applied toward perbenzylated derivatives. Procedure 3 was applied

toward epoxypyranosides that were determined to be incompatible with

LiAlD₄treatment. Procedure 1: A solution of epoxide (41 mg, 0.11

mmol) in 4:1 Et₂O:THF (1.25 mL) was treated with LiAlD₄(43.2 mg,

0.99 mmol) at ꢀ78 °C, then warmed slowly to rt and stirred for 18 h.

The reaction was then cooled to ꢀ55 °C and diluted with cold Et₂O

(2 mL), prior to treatment with finely ground Glauber’s salt

Synthesis of Glycals by Reductive Elimination. In a typical

experiment, a 1.0 M lithium naphthalenide solution (100 mL) was

prepared by dissolving naphthalene (12.8 g, 0.1 mol) in anhydrous,

deoxygenated THF (100 mL), followed by the portionwise addition of

finely divided Li metal (690 mg, 98.6 mmol). The reaction mixture was

stirred at rt under an argon atmosphere for 16 h, until the lithium was

completely dissolved. We note that this dark green solution can be

maintained for at least two weeks at ꢀ20 °C, under an inert atmosphere.

A portion of Li-naphthalenide (2.4 mL, 1.0 M in THF) was added

dropwise via addition funnel to a solution of thiophenyl glycoside (0.25

g, 0.54 mmol) in anhydrous THF (11 mL) at ꢀ40 °C. Consumption of

the Li-naphthalenide was determined by the loss of color from the

reaction mixture, which was also monitored by TLC. The starting

material was completely consumed after several hours at ꢀ40 °C, and

the reaction mixture was neutralized by the dropwise addition of 4:1

THF:AcOH (5 mL), then diluted with CH₂Cl₂(10 mL) and washed

with 0.5 M NaOH (10 mL). The combined organic extracts were

washed with brine (20 mL) and dried over anhydrous Na₂SO₄to afford

the desired glycal.

(Na₂SO₄10H₂O) to precipitate the aluminum salts. After being stirred

3

for 15 min at ꢀ55 °C, the reaction mixture was warmed to rt and stirred

until a clear separation between the organic and aqueous layers was

observed. The combined organic layers were washed with brine and

dried over Na₂SO₄. Procedure 2: A solution of epoxide (42.1 mg, 0.123

mmol) in 4:1 Et₂O:THF (1.25 mL) was cooled to ꢀ78 °C, followed by

the portionwise addition of LiAlD₄(46.7 mg, 1.11 mmol). The mixture

was warmed to ꢀ10 °C and stirred for 7 h. The reaction was then cooled

to ꢀ55 °C and quenched with K,Na-tartrate solution, followed by

dilution with Et₂O (2 mL). The cooling bath was removed and the

solution was warmed to rt and stirred for 1 h. The organic layers were

extracted with EtOAc (3 ꢁ 20 mL), washed once with brine, and dried

over anhydrous Na₂SO₄. Procedure 3: A rb flask containing epoxide (9.5

mg, 0.036 mmol) was charged with anhydrous MeOH (15 mL),

precooled to ꢀ78 °C, then slowly warmed to rt over a period of 14 h.

MeOH was removed by rotary evaporation followed by azeotropic

distillation with toluene, then dried in vacuo to afford the desired methyl

glycoside.

Synthesis of 4-Deoxypentenosides by Decarboxylative

Elimination. In a typical experiment, a 4,6-diol (0.15 g, 0.56 mmol)

was dissolved at rt in CH₂Cl₂(20 mL) and H₂O (20 mL), then treated

with TEMPO (17.3 mg, 0.11 mmol) and BAIB (0.54 g, 1.67 mmol) with

vigorous stirring. After 1 h, the reaction mixture was quenched with satd

Na₂S₂O₃solution, extracted with EtOAc (3 ꢁ 20 mL), washed once

with brine, dried over Na₂SO₄, concentrated under reduced pressure,

and used without further purification. A solution of crude carboxylic acid

(46.8 mg, 0.17 mmol) in degassed DMF (5 mL) was treated with DMF

dineopentyl acetal (0.23 mL, 0.83 mmol) in a high-pressure reaction

vessel. This was placed in a preheated oil bath at 200 °C for 2 h, then

cooled to rt and concentrated under reduced pressure to afford a dark

brown oil. The residue was washed with H₂O (3 ꢁ 20 mL) and extracted

with Et₂O (3 ꢁ 20 mL). The combined organic extracts were washed

once with brine and dried over Na₂SO₄prior to purification by silica gel

chromatography.

3,6-Di-O-benzyl-4-deoxy-D-allal (3). A solution of 3,6-di-O-

acetyl-^D-allal³⁶(184 mg, 0.80 mmol) in 1:1 THF:CS₂(18 mL) was

treated with a 60% dispersion of NaH in mineral oil (97 mg, 2.40 mmol)

at 0 °C. After 15 min, the ice bath was removed and the solution was

stirred for a further 30 min at rt, after which MeI (500 μL, 7.98 mmol)

was added. The reaction mixture was stirred for 16 h or until TLC

indicated the disappearance of starting material, then quenched at 0 °C

with satd NH₄Cl (20 mL) and extracted with Et₂O (3 ꢁ 20 mL). The

combined organic extracts were washed with brine (20 mL), dried over

Na₂SO₄, and concentrated under reduced pressure. The resulting oil was

Preparation of DMDO.²⁹A 3-L, two-necked rb flask equipped

with a large stirring bar and a condenser for reduced pressure distillation

was charged with deionized H₂O (250 mL), NaHCO₃(58 g), and

acetone (200 mL). This mixture was stirred vigorously at 5 °C and

2540

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

purified by silica gel chromatography, using a 0ꢀ15% EtOAcꢀhexanes

gradient with 0.1% of Et₃N, to afford the xanthate as a white solid (232

mg, 91%). The xanthate was redissolved in degassed DMF at rt (7.2 mL)

and then treated with Bu₃SnH (0.97 mL, 4.15 mmol) and AIBN (5.9 mg,

0.36 mmol). The reaction mixture was heated at 120 °C in a sealed tube

for 30 min. The solution was allowed to cool to rt, and then diluted with

Et₂O (3 ꢁ 10 mL). The mixture was extracted, washed with H₂O (3 ꢁ

10 mL) and brine (10 mL), dried over Na₂SO₄, and concentrated.The

resulting residue was redissolved in CH₃CN (10 mL) and extracted with

hexane (3 ꢁ 10 mL). The acetonitrile extracts were concentrated under

reduced pressure to afford a volatile oil. The crude 3,6-di-O-acetyl-4-

deoxy-^D-allal was redissolved in MeOH (7.2 mL), treated at 0 °C with

NaOMe (1.4 mL, 1.0 M in MeOH, 1.44 mmol), and stirred for 15 min.

The ice bath was removed and the reaction mixture was stirred at rt for

an additional 16 h. The reaction mixture was concentrated under

reduced pressure and residual MeOH was removed by azeotropic

distillation with toluene (3 ꢁ 5 mL). The crude diol was then redissolved

in anhydrous DMF (7.2 mL), cooled to 0 °C under argon, and treated

with BnBr (428 μL, 3.60 mmol), TBAI (53 mg, 0.14 mmol), and a 60%

dispersion of NaH in mineral oil (144 mg, 3.60 mmol). The reaction was

stirred at rt overnight, quenched at 0 °C with satd NH₄Cl (10 mL), then

extracted with Et₂O (3 ꢁ 10 mL). The combined organic phase was

washed with H₂O (3 ꢁ 10 mL) and brine (10 mL), dried over Na₂SO₄,

and concentrated under reduced pressure. The residue was purified by

silica gel chromatography, using a 0ꢀ5% Et₂Oꢀpentanes gradient to

afford dibenzylated 4-deoxyallal 3 as a white solid (49.2 mg, 22% overall

yield). ¹H NMR (300 MHz, C₆D₆): δ 7.17ꢀ7.05 (m, 10 H), 6.49 (d, 1

H, J = 6.0 Hz), 4.87 (dt, 1 H, J = 1.8, 6.0 Hz), 4.44ꢀ4.26 (m, 5 H), 3.66

(m, 1 H), 3.43 (d, 2 H, J = 3.9 Hz), 1.94 (br dd, 1 H, J = 1.8, 14.4 Hz),

1.64 (ddd, 1 H, J = 3.9, 12.3, 14.7 Hz). ¹³C NMR (75 MHz, C₆D₆): δ

146.9, 139.4, 138.7, 128.2, 100.5, 73.1, 72.2, 71.1, 69.1, 66.0, 31.1. IR

(NaCl): 3447, 2971, 1454, 1377, 1103 cm^ꢀ1. [R]²⁵þ45.0 (c 0.2,

CH₂Cl₂). HRESI-MS: m/z calcd for C₂₀H₂₂O₃[M þ^DNa]^þ333.1467,

found 333.1465.

3.89ꢀ3.84 (m, 3 H), 3.81 (dd, 1 H, J = 1.8, 5.4 Hz). ¹³C NMR (75 MHz,

C₆D₆): δ 146.7, 139.0, 128.3, 128.3, 128.1, 98.5, 73.3, 73.1, 72.9, 72.0,

69.4, 68.7, 67.2. IR (NaCl): 2873, 1640, 1452, 1245, 1098, 1062 cm^ꢀ1

.

[R]²⁵ꢀ37 (c 0.7, CH₂Cl₂). HRESI-MS: m/z calcd for C₂₇H₂₈O₄

[M þ^DH]^þ439.1885, found 439.1891.

Phenyl 4,6-O-Benzylidene-2,3-dideoxy-1-thio-r-D-erythro-hex-2-

enopyranoside (5). Phenyl 4,6-di-O-acetyl-2,3-dideoxy-1-thio-R-D-threo-

hex-2-enopyranoside³⁷(514 mg, 1.59 mmol) was redissolved in MeOH

(4 mL) and treated with NaOMe in MeOH (1 M, 1.16 mmol) at 0 °C and

stirred for 15 min. The ice bath was removed and the reaction mixture was

stirred at rt for an additional 16 h. The reaction mixture was concentrated

under reduced pressure and the residual MeOH was removed by azeotropic

distillation with toluene (3 ꢁ 5 mL). The crude 4,6-diol (487 mg, 2.0 mmol)

was redissolved in DMF (20 mL) and treated with benzaldehyde dimethyl

acetal (0.92 mL, 12 mmol) and p-toluenesulfonic acid (78 mg, 0.4 mmol).

After being stirred at rt for 2 h, the reaction was heated to 30 °C for another

2 h with partial removal of solvent under reduced pressure (10 mmHg). The

reaction mixture was neutralized with satd NaHCO₃(20 mL) and then

extracted with Et₂O (3 ꢁ 20 mL). The combined organic phase was washed

with H₂O (3 ꢁ 20 mL) and brine (20 mL), dried over Na₂SO₄, and

concentrated under reduced pressure. The residue was purified by silica gel

chromatography, using a 10ꢀ50% EtOAcꢀhexanes gradient with 0.1%

Et₃N, to afford benzylidene acetal 5 as white crystals (468 mg, 70% over

2 steps). ¹H NMR (300 MHz, CDCl₃): δ 7.67ꢀ7.32 (m, 10 H), 6.33 (dd,

1H,J= 3.6, 9.9 Hz), 6.20 (ddd, 1 H, J= 1.8, 5.7, 9.9 Hz), 6.15 (dd, 1 H, J=2.1,

3.3 Hz), 5.71 (s, 1 H), 4.48 (d, 1 H, J = 12.9 Hz), 4.40ꢀ4.35 (m, 2 H), 4.30

(brs, 1H).¹³CNMR(75MHz, CDCl₃):δ137.9, 137.9, 135.5, 131.1, 130.8,

129.1, 128.3, 127.3, 126.3, 125.3, 100.9, 84.2, 70.0, 67.9, 62.7. IR (NaCl):

2865, 1583, 1478, 1383, 1330, 1140, 1058 cm^ꢀ1. [R]²⁵ꢀ0.02 (0.4,

D

CH₂Cl₂). HRESI-MS: m/z calcd for C₁₉H₁₈O₃S [M þ H]^þ349.0874,

found 349.0870.

3-O-Benzyl-4,6-O-benzylidene-D-gulal (6). Thiophenyl glyco-

side 5 (468 mg, 1.43 mmol) dissolved in CH₂Cl₂(14 mL) was treated with

a precooled solution of DMDO in acetone (0.08 M, 1 equiv) at ꢀ78 °C for

1 h. The resulting sulfoxide was concentrated under reduced pressure in an

ice bath to a colorless solid, redissolved in THF (29 mL), then treated with

diethylamine (0.7 mL, 7.15 mmol) and stirred at rt for 16 h. The reaction

mixture was concentrated under reduced pressure and purified by silica gel

chromatography, using a 10ꢀ50% EtOAcꢀhexanes gradient with 0.1%

Et₃N, to yield a D-gulal intermediate as white crystals (148 mg, 44%) along

with recovered thiophenyl glycoside (47 mg, 14%).

3,4,6-Tri-O-benzyl-D-gulal (3,4,6-Tri-O-benzyl-D-xylo-hex-1-enitol)

(4). Phenyl 4,6-di-O-acetyl-2,3-dideoxy-1-thio-R-D-threo-hex-2-enopyra-

noside³⁷(188 mg, 0.58 mmol) was dissolved in CH₂Cl₂(6 mL) and

treated with a precooled solution of DMDO in acetone (0.08 M, 1 equiv)

at ꢀ78 °C for 1 h. The reaction mixture was concentrated under reduced

pressure in an ice bath to a colorless solid, redissolved at rt in THF (12 mL),

then treated with diethylamine (0.3 mL, 3 mmol) and stirred for 16 h. The

reaction mixture was concentrated under reduced pressure and purified by

silica gel chromatography, using a 0ꢀ40% EtOAcꢀhexanes gradient with

0.1% Et₃N, to afford the desired 4,6-di-O-acetyl-D-gulal as a dark brown oil

(112 mg, 83%).

A portion of this intermediate (147 mg, 0.63 mmol) was dried by

azeotropic distillation with toluene, then redissolved in anhydrous THF

(6 mL), cooled to 0 °C under argon, and then treated with BnBr

(0.19 mL, 1.57 mmol) and TBAI (46 mg, 0.12 mmol), and a 60%

dispersion of NaH in mineral oil (75 mg, 1.89 mmol). The reaction was

stirred at rt overnight, then quenched at 0 °C with satd NH₄Cl (20 mL)

and extracted with Et₂O (3 ꢁ 20 mL). The combined organic extracts

were washed with H₂O (3 ꢁ 20 mL) and brine (20 mL), dried over

Na₂SO₄, and concentrated under reduced pressure. The residue was

puriﬁed by silica gel chromatography, using a 0ꢀ20% EtOAcꢀhexanes

gradient with 0.1% of Et₃N, to aﬀord 3-O-benzyl-D-gulal 6 as a white

solid (158 mg, 77%). ¹H NMR (300 MHz, C₆D₆): δ 7.67ꢀ7.15 (m, 10

H), 6.56 (d, 1 H, J = 6.6 Hz), 5.37 (s, 1 H), 4.86 (ddd, 1 H, J = 1.2, 4.8, 7.5

Hz), 4.51ꢀ4.30 (m, 2 H), 4.19 (dd, 1 H, J = 1.8, 12.3 Hz), 3.92 (br s, 1

H), 3.81 (dd, 1 H, J = 2.1, 4.8 Hz), 3.62 (br s, 1 H), 3.44 (dd, 1 H, J = 0.9,

12.3 Hz). ¹³C NMR (75 MHz, C₆D₆): δ 146.5, 138.6, 128.3, 128.1,

128.0, 127.8, 126.5, 100.9, 97.2, 73.8, 69.3, 69.0, 68.7, 65.6. IR (NaCl):

1643, 1451, 1259 cm^ꢀ1. [R]²⁵_Dþ197 (c 0.5, CH₂Cl₂). HRESI-MS: m/z

calcd for C₂₀H₂₀O₄[M þ H]^þ325.1440, found 325.1444.

A portion of this intermediate (52 mg, 0.22 mmol) was redissolved

in MeOH (2.5 mL), treated at 0 °C with NaOMe (1.0 M in MeOH,

0.11 mmol), and stirred for 15 min. The ice bath was removed and the

reaction mixture was stirred at rt for an additional 16 h. The reaction

mixture was concentrated under reduced pressure and residual MeOH

was removed by azeotropic distillation with toluene (3 ꢁ 5 mL). The

crude triol was dried by azeotropic distillation with toluene, then

redissolved in anhydrous DMF (2.5 mL), cooled to 0 °C under argon,

and treated with BnBr (0.12 mL, 0.99 mmol), TBAI (16 mg, 0.04

mmol), and a 60% dispersion of NaH in mineral oil (53 mg, 1.32 mmol).

The reaction was stirred at rt overnight, quenched at 0 °C with satd

NH₄Cl (10 mL), then extracted with Et₂O (3 ꢁ 10 mL). The combined

organic phase was washed with H₂O (3 ꢁ 10 mL) and brine (10 mL),

dried over Na₂SO₄, and concentrated under reduced pressure. The

residue was puriﬁed by silica gel chromatography, using a 0ꢀ15%

EtOAcꢀhexanes gradient with 0.1% Et₃N, to aﬀord tribenzyl-D-gulal 4

3-Deoxy-4,6-p-methoxybenzylidene-D-glucal (8). A solu-

tion of acetal 7 (75 mg, 0.28 mmol) in 1:1 THF:CS₂(7 mL) was

treated with a 60% dispersion of NaH in mineral oil (34 mg, 0.85 mmol)

at 0 °C. After 15 min, the ice bath was removed and the solution was

1

as a white solid (83.7 mg, 91% over 3 steps). H NMR (300 MHz,

C₆D₆): δ 7.24ꢀ7.08 (m, 15 H), 6.56 (d, 1 H, J = 6.6 Hz), 4.89 (dd, 1 H,

J = 1.8, 6.3 Hz), 4.51 (dt 1 H, J = 1.8, 5.1 Hz), 4.51ꢀ4.25 (m, 6 H),

2541

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

stirred for a further 30 min at rt, after which CH₃I (0.18 mL, 2.83 mmol)

was added. The reaction mixture was stirred for 16 h or until TLC

indicated the disappearance of starting material, then quenched at 0 °C

with satd NH₄Cl (10 mL) and extracted with CH₂Cl₂(3 ꢁ 15 mL). The

combined organic extracts were washed with brine (15 mL), dried over

Na₂SO₄, and concentrated under reduced pressure. The resulting

residue was purified by silica gel chromatography, using a 0ꢀ15%

EtOAcꢀhexanes gradient with 0.1% of Et₃N, to afford the C3 xanthate

as a white solid (82 mg, 81%).

mmol), and a 60% dispersion of NaH in mineral oil (0.41 g, 10.18

mmol). The reaction was stirred at rt overnight, then quenched at 0 °C

with satd NH₄Cl (50 mL) and extracted with Et₂O (3 ꢁ 50 mL). The

combined organic extracts were washed with H₂O (3 ꢁ 50 mL) and

brine (50 mL), dried over Na₂SO₄, and concentrated under reduced

pressure. The residue was purified by silica chromatography, using a

5ꢀ30% EtOAcꢀhexanes gradient with 0.1% of Et₃N, to afford the

corresponding 3-O-benzyl ether as white solid (2.21 g, 92%).

A portion of this protected glucal (770 mg, 2.17 mmol) was

redissolved in CH₂Cl₂(13 mL) and cooled to 0 °C, then treated with

iBu₂AlH (10 mL of a 1 M solution in hexanes) and stirred for 2 h, after

which the reaction mixture was quenched with satd NH₄Cl (25 mL) and

satd K,Na-tartrate (20 mL). The organic layers were extracted with

CHCl₃(3 ꢁ 50 mL), then dried over Na₂SO₄and puriﬁed by silica gel

chromatography, using a 30ꢀ50% EtOAcꢀhexanes gradient with 0.1%

Et₃N, to aﬀord the desired 4-O-PMB ether as a clear syrup (712 mg,

92%).

The intermediate xanthate (0.95 g, 2.68 mmol) was redissolved in

degassed toluene at rt (27 mL) and then treated with Bu₃SnH (3.6 mL,

13.40 mmol) and AIBN (22 mg, 1.34 mmol). The reaction mixture was

heated at reﬂux for 30 min. The solution was allowed to cool to rt, and

the volatiles were removed under reduced pressure. The residue was

puriﬁed by silica gel chromatography, using a 0ꢀ5% EtOAcꢀhexanes

gradient with 0.1% of Et₃N, to aﬀord 4,6-p-methoxybenzylidene acetal 8

as a white solid (360 mg, 54% over 2 steps) along with a C3 allylstannane

1

byproduct as a single diastereomer (611 mg). H NMR (400 MHz,

A portion of the 4-O-PMB ether (540 mg, 1.51 mmol) was dried by

azeotropic distillation with toluene, then redissolved in anhydrous DMF

(10 mL), cooled to 0 °C under argon, and treated with BnBr (0.63 mL,

5.29 mmol) and a 60% dispersion of NaH in mineral oil (220 mg, 5.29

mmol). The reaction was stirred at rt overnight, then quenched at 0 °C

with satd NH₄Cl (20 mL) and extracted with Et₂O (3 ꢁ 20 mL). The

combined organic extracts were washed with H₂O (3 ꢁ 20 mL) and

brine (20 mL), dried over Na₂SO₄, then concentrated and puriﬁed by

silica gel chromatography, using a 10ꢀ25% EtOAcꢀhexanes gradient

with 0.1% Et₃N, to aﬀord the corresponding 6-O-benzyl ether as a

colorless syrup (633 mg, 93%). This was redissolved in CH₂Cl₂

(40 mL), tBuOH (2.5 mL), and pH 7 phosphate buﬀer (7 mL), then

treated at 0 °C with DDQ (965 mg, 4.26 mmol). The reaction mixture

was stirred vigorously at rt for another 3 h, quenched with satd NaHCO₃

(20 mL), and extracted with CH₂Cl₂(3 ꢁ 20 mL). The organic layers

were washed with brine (25 mL), dried over Na₂SO₄, then concentrated

and puriﬁed by silica gel chromatography, using a 10ꢀ50% EtOAcꢀ

hexanes gradient with 0.1% Et₃N, to aﬀord 4,6-dibenzyl glucal 10 as a

colorless syrup (360 mg, 78%). ¹H NMR (300 MHz, C₆D₆): δ 7.30ꢀ

7.07 (m, 10 H), 6.20 (dd, 1 H, J = 1.4, 7.5 Hz), 4.67 (dd, 1 H, J = 2.3, 6.2

Hz), 4.52ꢀ4.42 (m, 2 H), 4.34ꢀ4.26 (m, 2 H), 4.08 (ddd, 1 H, J = 3.8,

6.7, 9.0 Hz), 3.98 (dt, 1 H, J = 6.7, 1.8 Hz), 3.90ꢀ3.85 (m, 1 H), 3.71 (dd,

1 H, J = 4.8, 10.6 Hz), 3.66 (dd, 1 H, J = 3.6, 10.6 Hz), 2.26 (d, 1 H, J = 3.9

Hz). ¹³C NMR (100 MHz, C₆D₆): δ 144.8, 139.4, 138.8, 128.1, 128.0,

127.9, 127.8, 100.6, 78.0, 76.8, 73.7, 70.8, 69.5, 68.8. IR (NaCl): 2861,

C₆D₆): δ 7.55ꢀ6.79 (m, 4 H), 6.16 (dt, 1 H, J = 1.2, 6.0 Hz), 5.36 (s, 1

H), 4.40 (dt, 1 H, J = 2.4, 5.6 Hz), 4.26 (dd, 1 H, J = 4.8, 10.4 Hz), 3.70

(dd, 1 H, J = 4.8, 9.2 Hz), 3.64 (dd, 1 H, J = 2.8, 9.2 Hz), 3.56 (m, 1 H),

3.25 (s, 3 H), 2.14 (m, 2 H). ¹³C NMR (100 MHz, C₆D₆): δ 160.2,

143.1, 130.7, 113.4, 113.4, 101.6, 98.4, 74.8, 70.2, 68.7, 54.5, 26.4. IR

(NaCl): 2875, 1641, 1609, 1517, 1248, 1085, 1000 cm^ꢀ1. [R]²⁵_Dþ40.9

(c 0.2, CH₂Cl₂). HR-EI-MS: m/z calcd for C₁₄H₁₆O₄[M]^þ248.1049,

found 248.1049.

4-O-Benzyl-3-deoxy-6-O-p-methoxybenzylidene-D-glucal

(9). Acetal 8 (279 mg, 1.12 mmol) was dissolved in freshly distilled THF

(23 mL), cooled to ꢀ78 °C under argon, and then treated with

boraneꢀTHF complex (5.6 mL of a 1 M solution in THF). After 15

min, the mixture was treated with Bu₂BOTf (2.8 mL of a 1 M solution in

THF) and stirred for another 16 h at ꢀ78 °C. The reaction mixture was

quenched at this temperature with Et₃N (1.4 mL), followed by dropwise

addition of MeOH (5 mL). The cooling bath was removed and the

mixture was allowed to warm to rt and stirred for 30 min, then

concentrated under reduced pressure to yield a 4:1 mixture of 6- and

4-O-PMB ethers, which were separated by silica gel chromatograph,y

using a 5ꢀ20% EtOAcꢀhexanes gradient with 0.1% Et₃N. It is worth

noting that the alcohol can be contaminated by residual Bu₂BOH

(formed upon aqueous workup), but this impurity can be removed by

multiple rounds of azeotropic distillation with MeOH.

The 6-O-PMB ether (273 mg, 1.09 mmol) was dried by azeotropic

distillation with toluene, then dissolved in anhydrous THF (11 mL),

cooled to 0 °C under argon, and then treated with BnBr (0.32 mL, 2.72

mmol) and TBAI (81 mg, 0.22 mmol), and a 60% dispersion of NaH in

mineral oil (131 mg, 3.27 mmol). The reaction was stirred at rt

overnight, then quenched at 0 °C with satd NH₄Cl (20 mL) and

extracted with Et₂O (3 ꢁ 20 mL). The combined organic extracts were

washed with H₂O (3 ꢁ 20 mL) and brine (20 mL), dried over Na₂SO₄,

and concentrated under reduced pressure. The residue was puriﬁed by

silica gel chromatography, using a 0ꢀ10% EtOAcꢀhexanes gradient

with 0.1% of Et₃N, to aﬀord benzyl ether 9 as a white solid (270 mg, 73%

over 2 steps). ¹H NMR (300 MHz, (CD₃)₂CO): δ 7.41ꢀ7.01 (m, 9 H),

6.39 (d, 1 H, J = 5.7 Hz), 4.77ꢀ4.56 (m, 5 H), 3.92ꢀ3.82 (m, 4 H), 3.86

(s, 3 H), 2.49 (dt, 1 H, J = 16.8, 4.8 Hz), 2.11 (d, 1 H, J = 7.5 Hz). ¹³C

NMR (75 MHz, CDCl₃): δ 159.4, 143.3, 138.9, 130.7, 129.2, 128.3,

128.1, 113.8, 97.3, 77.3, 73.0, 70.8, 70.6, 68.8, 54.5, 26.7. IR (NaCl):

2907, 2865, 1654, 1609, 1512, 1240 cm^ꢀ1. [R]²⁵þ44.3 (c 1.1,

CH₂Cl₂). HR-EI-MS: m/z calcd for C₂₁H₂₄O₄[M]^þ^D340.1675, found

340.1679.

1642, 1453, 1235, 1095, 1069, 1028, 735, 697 cm^ꢀ1

.

3,6-Di-O-benzyl-4-deoxy-D-glucal (11). A solution of glucal 10

(60 mg, 0.18 mmol) in 1:1 THF:CS₂(5 mL) was treated with a 60%

dispersion of NaH in mineral oil (30 mg, 0.54 mmol) at 0 °C. After 15

min, the ice bath was removed and the solution was stirred for a further

30 min at rt, then treated with CH₃I (0.4 mL, 6.42 mmol) and stirred for

16 h or until TLC indicated the disappearance of starting material. This

reaction mixture was quenched with saturated NH₄Cl solution (5 mL),

extracted with CH₂Cl₂(3 ꢁ 25 mL), washed with brine (5 mL), dried

over anhydrous Na₂SO₄, and concentrated under reduced pressure. The

crude C4 xanthate was redissolved in degassed toluene (3 mL) and then

treated with Bu₃SnH (0.4 mL, 2.39 mmol) and AIBN (2 mg, 0.01 mmol)

at rt. After 5 min, the reaction mixture was heated at reflux and stirred for

a further 15 min. The solution was allowed to cool to rt, and the volatiles

were removed under reduced pressure. The residue was purified by silica

gel chromatography, using a 10ꢀ20% EtOAcꢀhexanes gradient with

0.1% of Et₃N, to afford 4-deoxyglucal 11 as a colorless oil (47 mg, 82%

over 2 steps). ¹H NMR (300 MHz, C₆D₆): δ 7.30ꢀ7.10 (m, 10 H), 6.32

(d, 1 H, J = 6.6 Hz), 4.82 (dt, 1 H, J = 6.3, 1.2 Hz), 4.32 (d, 4 H), 4.02 (m,

1 H), 3.93 (ddt, 1 H, J = 2.4, 1.2, 6.6 Hz), 3.54 (ddd, 1 H, J = 1.2, 6.0, 10.2

Hz), 3.38 (ddd, 1 H, J = 0.9, 6.0, 10.2 Hz), 1.96ꢀ1.78 (m, 2 H). ¹³C

NMR (75 MHz, C₆D₆): δ 144.8, 139.3, 138.7, 128.3, 128.2, 128.1, 127.7,

3,6-Di-O-benzyl-D-glucal (10). A solution of C3 alcohol 7 (1.79

g, 6.78 mmol) was dried by azeotropic distillation with toluene, then

redissolved in anhydrous DMF (66 mL), cooled to 0 °C under argon,

and then treated with BnBr (1.21 mL, 10.18 mmol), TBAI (0.50 g, 1.36

2542

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

127.5, 127.4, 127.3, 102.8, 73.7, 73.1, 72.0, 69.5, 68.7, 31.1. [R]²⁵_Dꢀ16.0

(c 1.0, CHCl₃).

4.38ꢀ4.18 (m, 5 H), 3.85ꢀ3.76 (m, 2 H), 3.65 (m, 1 H), 1.90ꢀ1.88 (m, 2

H). ¹³C NMR (100 MHz, C₆D₆): δ 143.0, 138.8, 128.1, 127.9, 127.7, 127.4,

127.3, 97.1, 75.2, 73.0, 70.7, 70.2, 68.6, 23.9. HRESI-MS: m/z calcd for

C₂₀H₂₂O₃[M þ Na]^þ333.1467, found 333.1470.

Thiophenyl 2-O-Benzyl-3-deoxy-4,6-p-methoxybenzylidene-

D-galactoside (13). A solution of alcohol 12 (0.40 g, 0.85 mmol) in 1:1

THF:CS₂(16 mL) was treated with a 60% dispersion of NaH in mineral

oil (61 mg, 2.55 mmol) at 0 °C. After 15 min, the ice bath was removed

and the reaction mixture was stirred for 30 min at rt, then treated with

CH₃I (0.26 mL, 4.26 mmol) and stirred for 16 h or until TLC indicated

the disappearance of starting material. The reaction mixture was

quenched with saturated NH₄Cl solution (5 mL), extracted with

CH₂Cl₂(3 ꢁ 25 mL), washed with brine (5 mL), dried over anhydrous

Na₂SO₄, and concentrated under reduced pressure. The resulting

residue was purified by silica gel chromatography, using 20% EtOAc

in hexanes with 0.1% of Et₃N, to afford the C3 xanthate as a white solid

(0.44 g, 91%). This was redissolved in degassed toluene (8 mL) and

treated with Bu₃SnH (1.04 mL, 3.88 mmol) and AIBN (64 mg, 0.39

mmol), then heated to reflux and stirred for another 30 min. The

reaction mixture was allowed to cool to rt, and the volatiles were

removed under reduced pressure. The residue was purified by silica

gel chromatography, using a 0ꢀ30% EtOAcꢀhexanes gradient with

0.1% Et₃N to afford 3-deoxygalactoside 13 as a white solid (0.26 g, 75%).

¹H NMR (400 MHz, C₆D₆): δ 7.88ꢀ6.84 (m, 14 H), 5.24 (s, 1 H), 4.61

(d, 1 H, J = 9.3 Hz), 4.52ꢀ4.17 (m, 2 H), 4.08 (d, 1 H, J = 11.7 Hz), 3.89

(dt, 1 H, J = 10.8, 5.4 Hz), 3.38ꢀ3.35 (m, 2 H), 3.26 (s, 3 H), 2.55 (s, 1

H), 2.31 (ddd, 1 H, J = 4.8, 7.8, 10.5 Hz), 1.37 (ddd, 1 H, J = 3.6, 10.4,

11.1 Hz). ¹³C NMR (100 MHz, C₆D₆): δ 138.8, 138.7, 135.7, 131.7,

128.7, 128.7, 128.2, 128.2, 128.1, 127.9, 127.7, 127.4, 126.7, 89.5, 79.1,

73.1, 72.9, 72.2, 71.8, 70.9, 69.3, 34.1. IR (NaCl): 2854, 1612, 1517,

1391, 1517, 1248, 1101 cm^ꢀ1. [R]²⁵ꢀ6.7 (c 0.7, CH₂Cl₂). HRESI-

MS: m/z calcd for C₂₇H₂₈O₅S [M þ^DH]^þ465.1736, found 465.1738.

3-Deoxy-4,6-p-methoxybenzylidene-D-galactal (14). A so-

lution of thiophenyl galactoside 13 (249 mg, 0.54 mmol) in THF

(11 mL) was subjected to reductive elimination conditions as previously

described, then purified by silica gel chromatography, using a 5ꢀ30%

EtOAcꢀhexanes gradient with 0.1% Et₃N, to afford the desired 3-deox-

ygalactal 14 as a white solid (104 mg, 78%). ¹H NMR (400 MHz, C₆D₆):

δ 7.62 (d, 2 H), 6.78 (d, 2 H), 6.43 (d, 1 H, J = 6.3 Hz), 5.37 (s, 1 H), 4.45

(m, 1 H), 4.16 (dd, 1 H, J = 1.8, 12.3 Hz), 3.52 (br s, 1 H), 3.43 (dd, 1 H,

J = 1.8, 12.3 Hz), 3.23 (s, 3 H), 3.09 (br s, 1 H), 2.00 (br s, 2 H). ¹³C NMR

(100 MHz, C₆D₆): δ 160.1, 143.0, 131.4, 127.9, 127.8, 127.7, 127.4, 113.3,

101.3, 96.0, 70.7, 69.3, 67.2, 54.4, 25.9. IR (NaCl): 2918, 1733, 1649, 1520,

Methyl 3-O-Benzyl-2,4-dideoxy-β-pent-4-enopyranoside

(20). Compound 18 (methyl 3-O-benzyl-2-deoxy-β-D-glucoside; 47 mg,

0.18 mmol) was subjected to the standard TEMPO oxidationꢀdecarboxyla-

tive elimination conditions as previously described. The residue was purified

by preparative TLC, using 60% EtOAc in hexanes with 0.1% Et₃N, to afford

2,4-dideoxy-4-pentenoside 20 as a volatile colorless syrup (22 mg, 58% over 2

steps). ¹H NMR (300 MHz, C₆D₆): δ 7.34ꢀ7.11 (m, 5 H), 6.25 (dd, 1 H,

J =1.2, 6.3 Hz), 4.89 (ddd, 1H, J= 0.9, 3.0, 6.3 Hz), 4.58 (dd, 1 H, J=2.7, 6.3

Hz), 4.36 (s, 2 H), 4.00 (ddt, 1 H, J= 1.2, 2.7, 7.5 Hz), 3.27 (s, 3 H), 2.16 (m,

1 H), 2.02 (dddd, 1 H, J=0.9, 2.4, 6.3, 12.9 Hz). ¹³CNMR(75MHz, C₆D₆):

δ 142.5, 139.2, 128.2, 128.1, 103.1, 99.4, 69.4, 68.0, 55.5, 34.3. IR (NaCl):

2923, 1644, 1454, 1388, 1227, 1190, 1090 cm^ꢀ1. [R]²⁵ꢀ10.0 (c 0.2,

D

CH₂Cl₂). HRESI-MS: m/z calcd for C₁₃H₁₆O₃[M þ H ꢀ CH₃OH]^þ

189.0916, found 189.0918.

Methyl 2-O-Benzyl-3,4-dideoxy-β-pent-4-enopyranoside

(26). Compound 24 (methyl 2-O-benzyl-3-deoxy-β-D-glucoside;³⁸68

mg, 0.25 mmol) was subjected to the standard TEMPO oxidationꢀ

decarboxylative elimination conditions as previously described. The

volatile residue was carefully purified by silica gel chromatography, using

a 0ꢀ5% diethyl etherꢀpentanes gradient, to afford the desired 3,4-

dideoxy-4-pentenoside 26 as a colorless syrup (37 mg, 66% over

1

2 steps). H NMR (300 MHz, C₆D₆): δ 7.29ꢀ7.07 (m, 5 H), 6.25

(dt, 1 H, J = 2.1, 6.3 Hz), 4.77 (d, 1 H, J = 3.9 Hz), 4.56 (m, 1 H),

4.43ꢀ4.34 (m, 2 H), 3.54 (m, 1 H), 3.23 (s, 3 H), 2.24 (ddt, 1 H, J = 17.4,

5.1, 2.7 Hz), 1.98 (dt, 1 H, J = 17.4, 4.5 Hz). ¹³C NMR (75 MHz, C₆D₆):

δ 140.3, 139.2, 128.5, 128.3, 128.0, 127.7, 127.7, 127.6, 127.6, 99.0, 98.9,

71.9, 71.2, 55.4, 23.0. IR (NaCl): 2913, 1658, 1453, 1231, 1190,

1092 cm^ꢀ1. [R]²⁵ꢀ93.0 (c 1.0, CHCl₃). HRESI-MS: m/z calcd for

D

C₁₃H₁₆O₃[M þ H]^þ221.1178, found 221.1184.

Methyl 2-O-Benzyl-3,4-dideoxy-r-pent-4-enopyranoside

(27). Compound 25 (methyl 2-O-benzyl-3-deoxy-R-D-glucoside;³⁹162

mg, 0.60 mmol) was subjected to the standard TEMPO oxidationꢀ

decarboxylative elimination conditions as previously described. The

volatile residue was carefully purified by silica gel chromatography, using

a 0ꢀ5% diethyl etherꢀpentanes gradient, to afford the desired 3,4-

dideoxy-4-pentenoside 27 as a colorless syrup (38 mg, 28% over 2 steps).

¹H NMR (300 MHz, C₆D₆): δ 7.30ꢀ7.09 (m, 5 H), 6.14 (ddd, 1 H, J =

1.2, 3.0, 5.7 Hz), 4.90 (d, 1 H, J = 2.4 Hz), 4.60 (dt, 1 H, J = 2.4, 6.0 Hz),

4.39ꢀ4.29 (m, 2 H), 3.63 (ddd, 1 H, J = 2.4, 6.0, 10.8 Hz), 3.29 (s, 3 H),

2.51 (ddd, 1 H, J = 2.1, 11.4, 16.1 Hz), 2.05 (dt, 1 H, J = 15.9, 5.7 Hz). ¹³C

NMR (100 MHz, C₆D₆): δ 139.3, 128.2, 127.9, 127.7, 127.4, 99.5, 97.4,

73.4, 70.7, 55.3, 22.3. IR (NaCl): 2918, 1649, 1451, 1232, 1177, 1093,

1029 cm^ꢀ1. [R]²⁵_Dþ96.6 (c 0.5, CH₂Cl₂). HRESI-MS: m/z calcd for

C₁₃H₁₆O₃[M þ H ꢀ CH₃OH]^þ189.0916, found 189.0917.

1396, 1248, 1074 cm^ꢀ1. [R]²⁵þ2.5 (c 0.2, CH₂Cl₂). HR-EI-MS: m/z

D

calcd for C₁₄H₁₆O₄[M]^þ248.1049, found 248.1044.

4,6-Di-O-benzyl-3-deoxy-D-galactal (15). A 50 mL rb flask

containing 13 (82 mg, 0.18 mmol) was dissolved in 8:1:1 AcOH:THF:

H₂O (1.8 mL) and heated to 45 °C for 5 h. The reaction mixture was

neutralized with NaHCO₃(20 mL), washed with brine (20 mL), dried over

Na₂SO₄, concentrated to dryness, and used without further purification. The

crude 4,6-diol was dried by azeotropic distillation with toluene, then

redissolved in anhydrous DMF (1.8 mL), cooled to 0 °C under argon,

and treated with BnBr (63 μL, 0.53 mmol), TBAI (13 mg, 0.04 mmol), and

a 60% dispersion of NaH in mineral oil (28 mg, 0.7 mmol). The reaction was

stirred at rt overnight, then quenched at 0 °C with satd NH₄Cl (20 mL) and

extracted with Et₂O (3 ꢁ 20 mL). The combined organic extracts were

washed with H₂O (3 ꢁ 20 mL) and saturated brine (20 mL), dried over

Na₂SO₄, and concentrated under reduced pressure. The residue was purified

by silica gel chromatography, using a 0ꢀ10% EtOAcꢀhexanes gradient, to

afford the corresponding 4,6-dibenzyl ether as a white solid (61 mg, 66%

over 2 steps). This intermediate (55 mg, 0.10 mmol) was dissolved in

anhydrous THF then subjected to the reductive elimination conditions

described above and purified by silica gel chromatography, using a 5ꢀ20%

EtOAcꢀhexanes gradient with 0.1% Et₃N, to afford the desired 3-deoxy-

galactal 15 as a white solid (20 mg, 60%). ¹H NMR (400 MHz, C₆D₆): δ

7.27ꢀ7.08 (m, 10 H), 6.39 (dt, 1 H, J = 4.7, 1.8 Hz), 4.44 (m, 1 H),

3,4,6-Tri-O-benzyl-1-deoxy-1R-deuterio-D-altrose (30).

Tribenzyl-D-allal 2 (40.5 mg, 0.10 mmol) was dissolved in CH₂Cl₂

(1.0 mL) and cooled to ꢀ55 °C, then treated for 2 d with a precooled

solution of DMDO (3.8 mL of a 0.4 M solution in CH₂Cl₂) followed by

a low-temperature workup as previously described to yield the corre-

sponding epoxyallal. The crude epoxide was redissolved in Et₂O

(2.5 mL) and subjected to S_N2 ring opening with LiAlD₄as previously

described (see procedure 1). The residue was purified by preparative

TLC, using 60% EtOAc in hexanes with 0.1% of Et₃N, to afford

1

deuteride adduct 30 as a white solid (35.5 mg, 84%). H NMR (300

MHz, C₆D₆): δ 7.31ꢀ7.08 (m, 15 H), 4.60 (d, 1 H, J = 9.0 Hz),

4.49ꢀ4.40 (m, 3 H), 4.42 (dd, 1 H, J = 4.2, 11.7 Hz), 4.35 (dd, 1 H, J =

8.0, 11.7 Hz), 4.05 (s, 2 H), 3.91 (d, 1 H, J = 4.8 Hz), 3.78 (d, 1 H, J = 7.9

Hz), 3.72 (d, 1 H, J = 7.7 Hz), 3.65 (m, 1 H), 3.57 (s, 1 H), 2.50 (s, 1 H).

¹³C NMR (75 MHz, C₆D₆): δ 159.4, 139.2, 138.9, 130.8, 129.4, 128.2,

113.8, 75.4, 75.0, 74.0, 73.0, 72.9, 71.3, 69.7, 68.7. IR (NaCl): 3416,

2543

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

2919, 1612, 1516, 1250, 1103 cm^ꢀ1. [R]²⁵þ47.3 (c 1.0, CH₂Cl₂).

126.1, 101.3, 76.4, 73.9, 73.2, 69.8, 66.2, 64.1. IR (NaCl): 3439, 2873,

D

HRESI-MS: m/z calcd for C₂₇H₂₉DO₅[M þ Na]^þ458.2064, found

1456, 1396, 1157, 1157, 1094 cm^ꢀ1. [R]²⁵þ8.41 (c 1.0, CH₂Cl₂).

D

458.2069.

HRESI-MS: m/z calcd for C₂₀H₂₁DO₅[M þ Na]^þ366.1428, found

2-O-Acetyl-3-O-benzyl-4,6-benzylidene-1-deoxy-1R-deu-

terio-^D-altrose (31). 4,6-Benzylidene-protected D-allal 1 (37.4 mg,

0.12 mmol) was converted to the corresponding epoxyallal as described

above (see compound 30), then redissolved in 4:1 Et₂O:THF (2.5 mL)

and subjected to S_N2 ring opening with LiAlD₄as previously described

(see procedure 2). Deuteride adduct 31 was characterized as the 2-O-

acetate by treatment with Ac₂O (1 mL) in pyridine (2 mL) at rt for 12 h,

concentrated to dryness with azeotropic distillation with toluene, and

purified by preparative TLC using 30% EtOAc in hexanes with 0.1%

Et₃N, to afford the acetate as a colorless oil (43.6 mg, 98%). 2-O-Acetyl-

366.1422. 3-O-Benzyl-4,6-benzylidene-1-deoxy-1R-deuterio-^D-idose

1

(33⁰): H NMR (400 MHz, C₆D₆): δ 7.49ꢀ7.32 (m, 10 H), 5.50 (s,

1 H), 4.70ꢀ4.66 (m, 2 H), 4.33 (dd, 1 H, J = 1.2, 12.4 Hz), 4.09 (br s, 1

H), 4.03 (dd, 1 H, J = 1.6, 12.4 Hz), 3.97 (s, 1 H), 3.79 (s, 1 H), 3.66 (s, 1

H), 3.64 (d, 1 H, J = 1.2 Hz), 1.50 (br s, 1 H). ¹³C NMR (75 MHz,

C₆D₆): δ 137.5, 129.1, 128.5, 128.3, 128.0, 127.5, 125.9, 101.4, 74.4,

73.9, 72.3, 70.3, 66.9, 65.6. IR (NaCl): 3505, 1648, 1451, 1399,

1127 cm^ꢀ1. [R]²⁵_Dþ9.64 (c 0.5, CH₂Cl₂). HRESI-MS: m/z calcd for

C₂₀H₂₁DO₅[M þ Na]^þ366.1428, found 366.1429.

Methyl 3,6-Di-O-benzyl-4-deoxy-1S-D-galactose (34). 4-

Deoxy-^D-glucal 11 (10.0 mg, 0.03 mmol) was dissolved in CH₂Cl₂

(0.5 mL) then converted to the corresponding epoxygalactal as pre-

viously described, but using a 0.08 M solution in acetone (0.09 mmol).

The crude epoxide was subjected to methanolysis as previously de-

scribed (see procedure 3). The residue was purified by preparative TLC

with 40% EtOAc in hexanes with 0.1% Et₃N to afford the methyl

glycoside 34 as a white solid (11.5 mg, quantitative yield). S_N2 ring

3-O-benzyl-4,6-benzylidene-1-deoxy-1R-deuterio-^D-altrose

(2-O-

1

acetyl 31): H NMR (400 MHz, C₆D₆): δ 7.31ꢀ7.18 (m, 10 H),

5.60 (s, 1H), 4.97 (d, 1 H, J = 2.1 Hz), 4.92ꢀ4.71 (m, 2 H), 4.34 (dd, 1

H, J = 5.1, 10.3 Hz), 4.06 (ddd, 1 H, J = 9.7, 10.0, 10.8 Hz), 3.95ꢀ3.92

(m, 2 H), 3.82 (br s, 1 H), 3.75 (t, 1 H, J = 10.4 Hz), 2.06 (s, 3 H). ¹³C

NMR (75 MHz, C₆D₆): δ 169.9, 138.1, 137.5, 129.0, 128.2, 128.2, 128.0,

126.1, 102.2, 78.0, 73.3, 72.6, 70.8, 69.1, 66.4, 21.1. IR (NaCl): 2873,

1736, 1371, 1234, 1141 cm^ꢀ1. [R]²⁵_Dþ14.8 (c 0.6, CH₂Cl₂). HRESI-

MS: m/z calcd for C₂₂H₂₃DO₆[M þ Na]^þ408.1433, found 408.1430.

2-O-Acetyl-3,4,6-tri-O-benzyl-1-deoxy-1R-deuterio-D-gu-

lose (32). Tribenzyl-D-gulal 4 (49.9 mg, 0.12 mmol) was converted to

the corresponding epoxygulal as described above (see compound 30),

then redissolved in 4:1 Et₂O:THF (2.5 mL) and subjected to S_N2 ring

opening with LiAlD₄as previously described (see procedure 2). The

deuteride adduct 32 and its diastereomer 32⁰were characterized as 2-O-

acetates by treatment with Ac₂O (1 mL) in pyridine (2 mL) at rt for 12 h,

concentration with azeotropic distillation with toluene, and separation

by preparative TLC using 25% EtOAc in hexanes with 0.1% Et₃N (42

mg, 2:3 R:β (C1), 81% combined yield). 2-O-Acetyl-3,4,6-tri-O-

benzyl-1-deoxy-1R-deuterio-^D-gulose (2-O-acetyl 32): ¹H NMR

(300 MHz, C₆D₆): δ 7.36ꢀ7.21 (m, 15 H), 5.28 (dd, 1 H, J = 3.0,

11.1 Hz), 4.66ꢀ4.46 (m, 6 H), 4.10ꢀ4.03 (m, 2 H), 3.84 (d, 1 H, J =

11.1 Hz), 3.70 (dd, 1 H, J = 6.3, 9.9 Hz), 3.62 (dd, 1 H, J = 1.8, 4.2 Hz),

3.54 (dd, 1 H, J = 6.0, 9.3 Hz), 2.03 (s, 3 H). ¹³C NMR (75 MHz, C₆D₆):

δ 170.0, 137.9, 137.5, 128.3, 128.3, 128.1, 127.9, 127.8, 127.7, 127.6,

74.4, 73.5, 73.4, 73.0, 72.6, 72.2, 69.1, 68.1, 20.9. IR (NaCl): 2870, 1739,

1454, 1367, 1235, 1093 cm^ꢀ1. [R]²⁵_Dꢀ16.9 (c 1.0, CH₂Cl₂). HRESI-

MS: m/z calcd for C₂₉H₃₁DO₆[M þ Na]^þ500.2159, found 500.2154.

2-O-Acetyl-3,4,6-tri-O-benzyl-1-deoxy-1S-deuterio-^D-idose (2-O-

1

opening was confirmed by coupling constant analysis. H NMR (300

MHz, C₆D₆): δ7.30ꢀ7.21 (m, 10 H), 4.66 (m, 4 H), 4.09 (d, 1 H,

J = 7.5 Hz), 3.57ꢀ3.30 (m, 8 H), 2.02 (dd, 1 H, J = 1.6, 12.6 Hz),

1.52ꢀ1.35 (m, 2 H). HRESI-MS: m/z calcd for C₂₁H₂₆O₅[M þ H]^þ

359.1859, found 359.1864.

Methyl 3,6-Di-O-benzyl-4-deoxy-S-D-altrose (35). 4-Deoxy-

D-allal 3 (10.3 mg, 0.03 mmol) was converted into the corresponding

epoxyallal as described above (see compound 34), then subjected to

methanolysis as previously described (see procedure 3) and purified by

preparative TLC developing twice with 40% EtOAc in hexanes with

0.1% Et₃N to afford the methyl glycoside 35 as a white solid (9.1 mg,

76%). S_N2 ring opening was confirmed by coupling constant analysis.

¹H NMR (300 MHz, C₆D₆): δ 7.34ꢀ7.13 (m, 10 H), 4.69 (d, 1 H,

J = 2.4 Hz), 4.54ꢀ4.40 (m, 5 H), 3.82 (br s, 1 H), 3.67 (m, 1 H,

J = 4.5 Hz), 3.44 (dd, 1 H, J = 6.0, 10.2 Hz), 3.36 (dd, 1 H, J = 4.2, 10.2

Hz), 3.30 (s, 3 H), 2.12 (br s, 1H), 1.95 (ddd, 1 H, J = 4.2, 9.9, 14.1 Hz),

1.64 (dt, 1 H, J = 13.5, 4.2 Hz). ¹³C NMR (75 MHz, C₆D₆): δ 139.2,

138.7, 128.2, 128.2, 102.5, 75.1, 73.1, 72.9, 70.8, 68.6, 64.9, 54.9, 28.8. IR

(NaCl): 3477, 2914, 1493, 1454, 1098, 1041 cm^ꢀ1. [R]²⁵_Dþ34.0 (c 0.5,

CH₂Cl₂). HRESI-MS: m/z calcd for C₂₁H₂₆O₅[M þ Na]^þ381.1678,

found 381.1676.

4-O-Benzyl-1,3-dideoxy-1R-deuterio-6-O-p-methoxyben-

zyl-^D-mannose (36). 3-Deoxyglucal 9 (39.4 mg, 0.11 mmol) was

converted to the corresponding epoxyglucal as described above (see

compound 30), then redissolved in 4:1 Et₂O:THF (1.3 mL) and

subjected to S_N2 ring opening with LiAlD₄as previously described

(see procedure 1) to afford deuteride adduct 36 and minor diastereomer

36⁰(30 mg, 2:1 R:β (C1), 72% combined yield). These were also

characterized as 2-O-acetates by treatment with Ac₂O (1 mL) in pyridine

(2 mL) at rt for 12 h, then concentrated with azeotropic distillation with

toluene. These were separated by preparative TLC, using 60% EtOAc in

hexanes with 0.1% Et₃N. 4-O-Benzyl-1R-deuterio-1,3-dideoxy-6-O-p-

1

acetyl 32⁰): H NMR (300 MHz, C₆D₆): δ 7.44ꢀ7.24 (m, 15 H),

4.86 (br s, 1 H), 4.78ꢀ4.38 (m, 6 H), 4.01 (d, 2 H, J = 1.8 Hz), 3.87 (br s,

1 H), 3.80 (dd, 1 H, J = 6.6, 9.3 Hz), 3.62 (dd, 1 H, J = 5.1, 9.3 Hz), 3.49

(d, 1 H, J = 1.2 Hz), 2.12 (s, 3 H). ¹³C NMR (75 MHz, C₆D₆): δ 170.6,

138.2, 137.9, 137.8, 128.3, 128.2, 128.0, 127.8, 127.8, 127.7, 127.6, 127.4,

74.3, 73.7, 73.4, 72.3, 70.9, 69.4, 68.1, 21.0. IR (NaCl): 2870, 1728, 1454,

1369, 1240, 1090 cm^ꢀ1. [R]²⁵_Dþ1.1 (c 1.1, CH₂Cl₂). HRESI-MS: m/z

calcd for C₂₉H₃₁DO₆[M þ Na]^þ500.2159, found 500.2155.

3-O-Benzyl-4,6-benzylidene-1-deoxy-1S-deuterio-D-gulose

(33). 4,6-Benzylidene-protected D-gulal 6 (40.1 mg, 0.12 mmol) was

converted to the corresponding epoxygulal as described above (see

compound 30), then redissolved in 4:1 Et₂O:THF (1.3 mL) and

subjected to S_N2 ring opening with LiAlD₄as previously described (see

procedure 2) to afford deuteride adduct 33 and its diastereomer 33⁰

(30.6 mg, 2:3 R:β (C1), 72% combined yield). The diastereomers were

separated by preparative TLC, developing twice with 30% EtOAc in

hexanes with 0.1% Et₃N. 3-O-Benzyl-4,6-benzylidene-1-deoxy-1S-

1

methoxybenzyl-D-mannose (36): H NMR (400 MHz, CDCl₃): δ

7.33ꢀ7.25 (m, 7 H), 6.85 (d, 2 H), 4.57ꢀ4.36 (m, 4 H), 3.97 (br s, 1 H),

3.89 (s, 1 H), 3.76 (m, 1 H), 3.79 (s, 3 H), 3.74 (dd, 1 H, J = 2.0, 10.0

Hz), 3.73 (dd, 1 H, J = 5.2, 10.4 Hz), 3.39 (ddd, 1 H, J = 2.0, 5.2, 9.6 Hz),

2.46 (m, 1 H) 2.02 (br s, 1 H), 1.55 (m, 1 H). ¹³C NMR (100 MHz,

CDCl₃): δ 159.16, 138.07, 129.98, 129.55, 128.29, 127.63, 113.68,

80.44, 73.08, 71.07, 69.77, 69.07, 66.61, 55.16, 36.04. IR (NaCl): 3428,

1

deuterio-^D-gulose (33): H NMR (300 MHz, C₆D₆): δ 7.52ꢀ7.36

2917, 2862, 1610, 1511, 1248, 1094 cm^ꢀ1. [R]²⁵þ36.1 (c 0.8,

D

(m, 10 H), 5.53 (s, 1 H), 4.81ꢀ4.65 (m, 2 H), 4.30 (d, 1 H, J = 12.4 Hz),

4.18 (d, 1 H, J = 3.0 Hz), 4.16 (m, 1 H), 4.03 (dd, 1 H, J = 1.6, 12.6 Hz),

3.91 (t, 1 H, J = 3.0 Hz), 3.57 (d, 1 H, J = 11.9 Hz), 3.56 (s, 1 H), 2.05 (br

s, 1 H). ¹³C NMR (75 MHz, C₆D₆): δ 137.5, 129.0, 128.6, 128.2, 127.6,

CH₂Cl₂). HRESI-MS: m/z calcd for C₂₁H₂₅DO₅[M þ Na]^þ

382.1741, found 382.1740. 4-O-Benzyl-1S-deuterio-1,3-dideoxy-6-O-

1

p-methoxybenzyl-^D-glucose (36⁰): H NMR (400 MHz, CDCl₃): δ

7.32ꢀ7.21 (m, 8 H), 6.85 (d, 2 H), 4.58ꢀ4.39 (m, 4 H), 3.78 (s, 3 H),

2544

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

3.76 (dd, 1 H, J = 4.0, 10.0 Hz), 3.69 (dd, 1 H, J = 2.4, 10.4 Hz), 3.61 (dd,

1 H, J = 5.2, 10.8 Hz), 3.48 (ddd, 1 H, J = 4.8, 8.8, 12.2 Hz), 3.36 (ddd, 1

H, J = 2.4, 4.8, 8.4 Hz), 3.14 (d, 1 H, J = 9.6 Hz), 2.52 (dt, 1 H, J = 4.4, 1.6

Hz), 1.45 (d, 1 H, J = 10.4 Hz). ¹³C NMR (100 MHz, CDCl₃): δ 159.1,

137.9, 130.0, 129.5, 128.3, 127.7, 113.7, 79.0, 73.1, 72.1, 71.0, 68.6, 65.2,

(s, 3 H), 2.64 (d, 1 H, J = 1.2 Hz), 2.43 (ddd, 1 H, J = 2.7, 8.1, 13.5 Hz),

1.70 (s, 3 H), 1.37 (ddd, 1 H, J = 4.2, 11.7, 13.5 Hz). ¹³C NMR (100

MHz, C₆D₆): δ 169.5, 161.2, 132.0, 129.7, 128.9, 128.6, 128.4, 114.4,

109.6, 104.0, 102.2, 73.3, 69.7, 68.1, 55.9, 55.3, 34.8, 21.3. IR (NaCl):

2850, 1736, 1517, 1370, 1249, 1172, 1080, 827 cm^ꢀ1. [R]²⁵_Dꢀ61.8 (c

0.2, CH₂Cl₂). HRESI-MS: m/z calcd for C₁₇H₂₂O₇[M þ Na]^þ

361.1263, found 361.1268. Methyl 2-O-Acetyl-3-deoxy-4,6-p-methox-

ybenzylidene-r-^D-idose (2-O-acetyl 39⁰): ¹H NMR (400 MHz,

C₆D₆): δ 7.64 (d, 2 H), 6.79 (d, 2 H), 5.52 (ddd, 1 H, J = 3.6, 8.8,

11.6 Hz), 5.29 (s, 1 H), 5.11 (d, 1 H, J = 3.2 Hz), 4.09 (dd, 1 H, J = 0.8,

12.0 Hz), 3.50ꢀ3.47 (m, 2 H), 3.24 (s, 3 H), 3.11 (s, 3 H), 3.09 (d, 1 H,

J = 1.2 Hz), 2.18 (dd, 1 H, J = 3.2, 12.8 Hz), 2.13 (m, 1 H), 1.65 (s, 3 H).

Methyl 3-O-Benzyl-2-deoxy-5R-deuterio-D-xylose (40).

2-Deoxy-β-glc-4-DP 20 (37.4 mg, 0.17 mmol) was converted to the

corresponding 4,5-epoxypyranoside as described above (see compound

30), then subjected to S_N2 ring opening with LiAlD₄as previously

described (see procedure 1), and purified by preparative TLC using 40%

EtOAc in hexanes with 0.1% Et₃N, to afford deuteride adduct 40 as a

colorless oil (26 mg, 66%). ¹H NMR (300 MHz, C₆D₆): δ 7.35ꢀ7.15

(m, 5 H), 4.50ꢀ4.27 (m, 2 H), 4.03 (dd, 1 H, J = 2.4, 8.7 Hz), 3.62

(t, 1 H, J = 9.0 Hz), 3.31 (s, 3 H), 3.27 (ddd, 1 H, J = 4.8, 7.5, 10.5 Hz),

3.08 (d, 1 H, J = 9.6 Hz), 2.40 (br s, 1 H), 2.13 (ddd, 1 H, J = 2.4, 4.8, 12.9

Hz), 1.69 (dd, 1 H, J = 8.4, 12.9 Hz). ¹³C NMR (75 MHz, C₆D₆): δ

138.8, 128.3, 128.1, 127.7, 127.6, 127.4, 101.0, 78.1, 70.4, 70.0, 64.7, 55.7,

34.8. IR (NaCl): 3435, 2918, 1725, 1451, 1388, 1074 cm^ꢀ1. [R]²⁵_Dꢀ8.3

(c 0.5, CH₂Cl₂). HRESI-MS: m/z calcd for C₁₃H₁₇DO₄[M þ Na]^þ

262.1166, found 262.1169.

Methyl 2-O-Benzyl-3-deoxy-5R-thioethyl-D-xylose (42).

3-Deoxy-β-glc-4-DP 26 (5.0 mg, 0.02 mmol) was converted to the

corresponding 4,5-epoxypyranoside as described above (see compound

30), but without concentration. S_N2 epoxide ring opening was achieved

in situ by using LiSEt, prepared as a stock solution of EtSH (0.5 mL, 6.75

mmol) in anhydrous THF (8 mL) treated at 0 °C with n-BuLi (0.25 mL,

2.5 M in hexanes). The reaction mixture containing epoxides was treated

by the dropwise addition of LiSEt (3 mL of a 0.75 M solution) via

cannula at ꢀ55 °C, then warmed slowly to 0 °C over a period of 5 h. The

reaction was quenched with satd NaHCO₃, extracted with EtOAc, dried

over Na₂SO₄, and concentrated to yield a mixture of diastereomers (4.4 mg,

2:3 R:β (C5), 66% combined yield). Thioacetal 42 was characterized as

the 2-O-acetate by treatment with Ac₂O (1 mL) in pyridine (2 mL) at rt

for 12 h then concentrated with azeotropic distillation with toluene,

followed by preparative TLC, using a 0ꢀ20% EtOAcꢀhexanes gradient.

Methyl 4-O-Acetyl-2-O-benzyl-3-deoxy-5R-thioethyl-^D-xylose (2-O-

acetyl 42): ¹H NMR (300 MHz, C₆D₆): δ 7.40ꢀ6.99 (m, 5 H), 5.41 (m,

1 H), 5.18 (d, 1 H, J = 6.2 Hz), 4.82 (d, 1 H, J = 4.2 Hz), 4.63ꢀ4.47 (m, 2

H), 3.65 (m, 1 H), 3.19 (s, 3 H), 2.71ꢀ2.45 (m, 2 H), 2.22 (ddd, 1 H, J =

3.3, 4.8, 13.5 Hz), 2.02 (ddd, 1 H, J = 4.0, 7.6, 13.5 Hz), 1.60 (s, 3 H),

1.11 (t, 3 H, J = 9.0 Hz).

55.2, 38.0. IR (NaCl): 3401, 2923, 2862, 1607, 1511, 1248, 1056 cm^ꢀ1

.

[R]²⁵þ54 (c 0.7, CH₂Cl₂). HRESI-MS: m/z calcd for C₂₁H₂₅DO₅

[M þ^DNa]^þ382.1741, found 382.1739.

2-O-Acetyl-1,3-dideoxy-1R-deuterio-4,6-p-methoxybenzylidene-

D-mannose (37). Benzylidene-protected 3-deoxyglucal 8 (25.1 mg, 0.1

mmol) was converted to the corresponding epoxyglucal as described

above (see compound 30), then redissolved in 4:1 Et₂O:THF (2.5 mL)

and subjected to S_N2 ring opening with LiAlD₄as previously described

(see procedure 2). The deuteride adduct 37 and its diastereomer 37⁰

were characterized as 2-O-acetates by treatment with Ac₂O (1 mL) in

pyridine (2 mL) at rt for 12 h, concentrated with azeotropic distillation

with toluene, and purified by preparative TLC using 30% EtOAc in

hexanes with 0.1% Et₃N, gave an inseparable mixture (26 mg, 2:1 R:β

(C1), 83% combined yield). 2-O-Acetyl-1,3-dideoxy-1R-deuterio-4,6-

1

p-methoxybenzylidene-^D-mannose (2-O-acetyl 37): H NMR (300

MHz, CDCl₃): δ 7.42ꢀ7.39 (m, 2 H), 6.91ꢀ6.86 (m, 2 H), 5.56

(s, 1 H), 5.10 (m, 1 H), 4.27 (dd, 1 H, J = 4.8, 10.5 Hz), 3.99 (br s, 1 H),

3.92 (ddd, 1 H, J = 4.8, 9.3, 12.3 Hz), 3.75 (m, 1 H), 3.78 (s, 3 H), 3.38

(dt, 1 H, J = 9.9, 4.8 Hz), 2.30 (m, 1 H), 2.13 (s, 3 H), 1.88 (ddd, 1 H,

J = 3.6, 12.3, 13.5 Hz). 2-O-Acetyl-1,3-dideoxy-1S-deuterio-4,6-p-

methoxybenzylidene-^D-glucose (2-O-acetyl 37⁰): ¹H NMR (300

MHz, CDCl₃): δ 7.42ꢀ7.39 (m, 2 H), 6.91ꢀ6.86 (m, 2 H), 5.45

(s, 1 H), 4.77 (dt, 1 H, J = 5.1, 9.9 Hz), 4.36 (dd, 1 H, J = 5.1, 10.5 Hz),

4.06 (m, 1 H), 3.59 (t, 1 H, J = 10.5 Hz), 2.08 (s, 3 H), 1.75 (m, 1 H). ¹³C

NMR (diastereomeric mixture, 100 MHz, CDCl₃): δ 190.8, 170.3,

160.0, 131.9, 129.9, 127.3, 114.2, 113.6, 113.6, 101.8, 101.0, 81.1, 78.5,

74.1, 73.7, 69.3, 69.1, 68.9, 68.6, 67.9, 66.7, 64.0, 63.3, 55.5, 55.2, 36.1,

33.0, 28.0, 21.2, 20.9, 20.8. IR (NaCl): 2928, 1733, 1612, 1514, 1367,

1246 cm^ꢀ1. HRESI-MS: m/z calcd for C₁₆H₁₉DO₆[M þ H]^þ

310.1401, found 310.1399.

4,6-Di-O-benzyl-1,3-dideoxy-1S-deuterio-D-galactose (38).

3-Deoxygalactal 15 (19.7 mg, 0.06 mmol) was converted to the

corresponding epoxygalactal as described above (see compound 30),

then redissolved in 4:1 Et₂O:THF (1.3 mL) and subjected to S_N2 ring

opening with LiAlD₄as previously described (see procedure 1). The

residue was purified by preparative TLC using 70% EtOAc in hexanes

with 0.1% Et₃N, to afford deuteride adduct 38 as a white solid (19.0 mg,

91%). ¹H NMR (400 MHz, C₆D₆): δ 7.27ꢀ7.06 (m, 10 H), 4.40ꢀ4.16

(m, 4 H), 3.90 (m, 1 H), 3.80ꢀ3.71 (m, 2 H), 3.47 (m, 1 H), 3.45 (dt, 1

H, J = 1.6, 4.2 Hz), 2.98 (d, 1 H, J = 10.0 Hz), 2.19 (dt, 1 H, J = 13.2, 4.0

Hz), 1.08 (ddd, 1 H, J = 2.4, 11.2, 13.4, Hz), 0.72 (d, 1 H, J = 5.2 Hz). ¹³C

NMR (100 MHz, C₆D₆): δ 138.9, 138.8, 128.2, 127.9, 127.7, 127.4,

127.3, 77.8, 73.1, 71.0, 69.6, 62.4, 62.4, 35.5. IR (NaCl): 3435, 2923,

2860, 2364, 1449, 1101 cm^ꢀ1. [R]²⁵þ3.5 (c 0.3, CH₂Cl₂). HRESI-

MS: m/z calcd for C₂₀H₂₃DO₄[M þ^DNa]^þ352.1635, found 352.1637.

Methyl 2-O-Acetyl-3-deoxy-4,6-p-methoxybenzylidene-β-D-

galactose (39). 3-Deoxygalactal 14 (8.7 mg, 0.03 mmol) was con-

verted to the corresponding epoxygalactal as described above (see

compound 30), then subjected to methanolysis as previously described

(see Procedure 3). 3-Deoxygalactoside 39 and its diastereomer 39⁰were

characterized as their 2-O-acetates by treatment with Ac₂O (1 mL) in

pyridine (2 mL) at rt for 12 h, then concentrated with azeotropic

distillation with toluene. These were separated by preparative TLC

developing twice with 30% EtOAc in hexanes with 0.1% Et₃N (9 mg, 1:4

R:β (C1), 76% combined yield). Methyl 2-O-Acetyl-3-deoxy-4,6-p-

Methyl 2-O-Benzyl-3-deoxy-5S-deuterio-L-arabinose (43).

3-Deoxy-R-glc-4-DP 27 (34.7 mg, 0.16 mmol) was converted into the

corresponding 4,5-epoxypyranoside as described above (see compound

34), then subjected to S_N2 ring opening with LiAlD₄as previously

described (see procedure 1) and purified by preparative TLC using 80%

EtOAc in hexanes with 0.1% Et₃N, to afford deuteride adduct 43 as a

white solid (37 mg, 98%). ¹H NMR (400 MHz, C₆D₆): δ 7.36ꢀ7.12 (m,

5 H), 4.73 (d, 1 H, J = 2.9 Hz), 4.43 (s, 2 H), 3.96 (dt, 1 H, J = 11.4, 3.3),

3.59 (br s, 1 H), 3.26 (br s, 1 H), 3.24 (s, 3 H), 2.10 (dt, 1 H, J = 3.2, 11.7

Hz), 1.96 (m, 1 H), 1.71 (d, 1 H, J = 3.0 Hz). ¹³C NMR (100 MHz,

C₆D₆): δ 139.1, 128.1, 127.9, 127.9, 127.7, 127.4, 127.3, 98.2, 71.7, 70.6,

66.3, 54.7, 31.7. IR (NaCl): 3445, 2934, 1451, 1098, 1079, 1040,

992 cm^ꢀ1. [R]²⁵þ197 (c 1.1, CH₂Cl₂). HRESI-MS: m/z calcd for

1

methoxybenzylidene-β-^D-galactose (2-O-acetyl 39): H NMR (400

D

C₁₃H₁₇DO₄[M þ Na]^þ262.1163, found 262.1166.

MHz, C₆D₆): δ 7.58 (d, 2 H), 6.73 (d, 2 H), 5.55 (ddd, 1 H, J = 8.1, 11.7

Hz), 5.26 (s, 1 H), 4.29 (d, 1 H, J = 8.4 Hz), 4.08 (dd, 1 H, J = 1.2, 12.3

Hz), 3.44 (dd, 1 H, J = 1.8, 12.3 Hz), 3.39 (s, 3 H), 3.31 (br s, 1 H), 3.24

2,3-Di-O-benzyl-1-deoxy-1S-deuterio-D-xylose (44). Diben-

zyl-^D-xylal 28 (34.9 mg, 0.12 mmol) was converted to the corresponding

2545

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

epoxyxylal as described above (see compound 34), then redissolved in

4:1 Et₂O:THF (1.4 mL) and subjected to S_N2 ring opening with LiAlD₄

as previously described (see procedure 1) to afford deuteride adduct 44

and its minor diastereomer 44⁰as a white solid (20 mg, 1:5 R:β (C1),

76%). These were separated by preparative TLC using 60% EtOAc in

hexanes with 0.1% Et₃N. 2,3-Di-O-benzyl-1-deoxy-1S-deuterio-D-xy-

lose (44): ¹H NMR (400 MHz, C₆D₆): δ 7.23ꢀ7.07 (m, 10 H),

4.56ꢀ4.44 (m, 2 H), 4.27ꢀ4.26 (m, 2 H), 3.76 (dd, 1 H, J = 3.1, 11.7

Hz), 3.62 (m, 1 H), 3.51ꢀ3.45 (m, 2 H), 3.41 (dd, 1 H, J = 8.0, 11.6 Hz),

3.33 (m, 1 H), 2.96 (d, 1 H, J = 6.8 Hz). ¹³C NMR (100 MHz, C₆D₆): δ

138.8, 138.1, 128.3, 127.9, 27.7, 127.5, 127.4, 79.5, 76.4, 73.1, 71.5, 69.5,

69.3, 69.1, 68.5, 66.6. IR (NaCl): 3432, 2916, 2360, 1722, 1454,

1075 cm^ꢀ1. [R]²⁵_Dꢀ13.7 (c 0.4, CH₂Cl₂). HRESI-MS: m/z calcd for

C₁₉H₂₁DO₄[M þ Na]^þ338.1479, found 338.1482. 2,3-Di-O-benzyl-

1,5-deoxy-1R-deuterio-^D-mannose (44⁰): ¹H NMR (400 MHz, C₆D₆):

δ 7.31ꢀ6.99 (m, 10 H), 4.41ꢀ4.34 (m, 4 H), 3.82 (dd, 1 H, J = 3.2, 9.2

Hz), 3.78ꢀ3.75 (m, 2 H), 3.66 (m, 1H), 3.37 (dd, 1 H, J = 2.4, 5.6 Hz),

3.20 (dd, 1 H, J = 6.0, 9.2 Hz), 2.09 (d, 1 H, J = 3.6 Hz). HRESI-MS: m/z

calcd for C₁₉H₂₁DO₄[M þ Na]^þ338.1479, found 338.1477.

(2) (a) Danishefsky, S. J.; Bilodeau, M. T. Angew. Chem., Int. Ed. Engl.

1996, 35, 1380–1419. (b) Plante, O. J.; Andrade, R. B.; Seeberger, P. H.

Org. Lett. 1999, 1, 211–214. (c) Cheshev, P.; Marra, A.; Dondoni, A.

Carbohydr. Res. 2006, 341, 2714–2716. (d) Koester, D. C.; Leibeling, M.;

Neufeld, R.; Werz, D. B. Org. Lett. 2010, 12, 3934–3937.

(3) Examples of nonpyranoid glycals: (a) Singh, I.; Seitz, O. Org.

Lett. 2006, 8, 4319–4322. (b) Markad, S.; Xia, S.; Surana, B.; Morton,

M. D.; Hadad, C. M.; Peczuh, M. W. J. Org. Chem. 2008, 73, 6341–6354.

(c) Saha, J.; Peczuh, M. W. Org. Lett. 2009, 11, 4482–4484.

(4) Brigl, P. Hoppe-Seyler's Z. Physiol. Chem. 1921, 116, 1.

(5) Halcomb, R. L.; Danishefsky, S. J. J. Am. Chem. Soc. 1989,

111, 6661–6666.

(6) Selected examples: (a) Leeuwenburgh, M. A.; Timmers, C. M.;

::

van der Marel, G. A.; van Boom, J. H.; Mallet, J.-M.; Sinay, P. G.

Tetrahedron Lett. 1997, 38, 6251–6254. (b) Rainier, J. D.; Cox, J. M. Org.

Lett. 2000, 2, 2707–2709. (c) Parrish, J. D.; Little, R. D. Org. Lett. 2002,

4, 1439–1442. (d) Boulineau, F. P.; Wei, A. J. Org. Chem. 2004,

69, 3391–3399. (e) Wipf, P.; Pierce, J. G.; Zhuang, N. Org. Lett. 2005,

7, 483–485.

(7) Randolph, J. T.; Danishefsky, S. J. J. Am. Chem. Soc. 1995,

117, 5693–5700.

2,3-Di-O-benzyl-1-deoxy-1S-deuterio-L-arabinose (45).

Dibenzyl-D-arabinal 29 (46.7 mg, 0.158 mmol) was converted to the

corresponding epoxyarabinal as described above (see compound 30),

then redissolved in 4:1 Et₂O:THF (1.3 mL) and subjected to S_N2 ring

opening with LiAlD₄as previously described (see procedure 1), and

purified by preparative TLC with 50% EtOAc in hexanes with 0.5% Et₃N

to afford a colorless oil (36.4 mg, 73%). ¹H NMR (400 MHz, C₆D₆): δ

7.22ꢀ7.05 (m, 10 H), 4.36ꢀ4.23 (m, 2 H), 4.23ꢀ4.10 (m, 2 H), 4.10 (s,

1 H), 4.08 (q, 1 H, J = 2.8 Hz), 3.95 (s, 1 H), 3.94 (dd, 1 H, J = 0.9, 1.9

Hz), 3.91 (s, 1 H), 3.47 (dd, 1 H, J = 3.5, 10.9 Hz), 3.24 (dd, 1 H, J = 3.1,

10.2 Hz). ¹³C NMR (100 MHz, C₆D₆): δ 139.0, 138.3, 129.0, 128.0,

(8) Selected examples: (a) Hayward, M. M.; Roth, R. M.; Duﬀy, K. J.;

Dalko, P. I.; Stevens, K. L.; Guo, J.; Kishi, Y. Angew. Chem., Int. Ed. 1998,

37, 190–196. (b) Johnson, H. W. B.; Majumder, U.; Rainier, J. D. J. Am.

Chem. Soc. 2004, 127, 848–849. (c) Potuzak, J. S.; Moilanen, S. B.; Tan,

D. S. J. Am. Chem. Soc. 2005, 127, 13796–13797. (d) Akoto, C. O.;

Rainier, J. D. Angew. Chem., Int. Ed. 2008, 47, 8055–8058.

(9) Boulineau, F. P.; Wei, A. Org. Lett. 2002, 4, 2281–2283.

(10) Cheng, G.; Boulineau, F. P.; Liew, S.-T.; Shi, Q.; Wenthold,

P. G.; Wei, A. Org. Lett. 2006, 8, 4545–4548.

(11) It should be noted that protecting groups can aﬀect the

facioselectivity of DMDO oxidation (e.g., see ref 5), but this is beyond

the scope of the present study.

(12) Cheng, G.; Fan, R.-H.; Hernꢀandez-Torres, J.-M.; Boulineau,

F. P.; Wei, A. Org. Lett. 2007, 8, 4849–4852.

128.6, 128.4, 128.3, 128.2, 87.6, 84.1, 75.3, 75.1, 74.0, 72.0, 71.2. [R]²⁵

D

16.2 (c 0.5, CH₂Cl₂). HRESI-MS: m/z calcd for C₁₉H₂₁DO₄[M þ

Na]^þ338.1479, found 338.1476.

(13) Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereochemistry of

Organic Compounds; Wiley: New York, 1994.

(14) Huang, X. L.; Dannenberg, J. J.; Duran, M.; Bertrꢀan, J. J. Am.

Chem. Soc. 1993, 115, 4024–4030.

’ ASSOCIATED CONTENT

S

Supporting Information. Selected coupling constants

b

(15) Franck, R. W.; Kaila, N.; Blumenstein, M.; Geer, A.; Huang,

X. L.; Dannenberg, J. J. J. Org. Chem. 1993, 58, 5335–5337.

(16) Boschi, A.; Chiappe, C.; De Rubertis, A.; Ruasse, M. F. J. Org.

Chem. 2000, 65, 8470–8477.

(17) Dannenberg, J. J. Chem. Rev. 1999, 99, 1225–1241.

(18) (a) Houk, K. N.; Liu, J.; DeMello, N. C.; Condroski, K. R. J. Am.

Chem. Soc. 1997, 119, 10147–10152. (b) Jenson, C.; Liu, J.; Houk, K. N.;

Jorgensen, W. L. J. Am. Chem. Soc. 1997, 119, 12982–12983. (c) Bach,

R. D.; Dmitrenko, O.; Adam, W.; Schambony, S. J. Am. Chem. Soc. 2002,

125, 924–934. (d) Orendt, A. M.; Roberts, S. W.; Rainier, J. D. J. Org.

Chem. 2006, 71, 5565–5573. (e) D€ufert, A.; Werz, D. B. J. Org. Chem.

2008, 73, 5514–5519.

for pyranosides 30ꢀ45 and related diastereomers (Tables S1

and S2), and ¹H and ¹³C NMR spectra of new compounds 24,

26, 27, 30ꢀ40, and 42ꢀ45. This material is available free of

charge via the Internet at http://pubs.acs.org. Output ﬁles for

DFT and PPFMO calculations are available from the authors

upon request.

’ AUTHOR INFORMATION

Corresponding Author

*E-mail: alexwei@purdue.edu.

(19) (a) Jackel, C.; Dotz, H. J. Organomet. Chem. 2001,

624, 172–185. (b) Halcomb, R.; Boyer, S.; Wittman, M.; Olson, S.;

Denhart, D.; Liu, K.; Danishefsky, S. J. Am. Chem. Soc. 1995,

117, 5720–5749. (c) Wittman, M.; Halcomb, R.; Danishefsky, S.

J. Org. Chem. 1990, 55, 1979–1981.

(20) Barton, D. H. R.; McCombie, S. W. J. Chem. Soc., Perkin Trans. 1

1975, 16, 1574–1585.

(21) Braithwaite, D. H.; Holzapfel, C. W.; Williams, D. B. G.

’ ACKNOWLEDGMENT

This work was supported by the National Institutes of Health

(GM06982). The authors also gratefully acknowledge support

from the Purdue Center for Cancer Research for NMR and mass

spectrometry services, and Paul Wenthold and Panuwat Padun-

gros for helpful discussions on computational and synthetic

aspects of this work, respectively.

S. Afr. J. Chem. 1998, 51, 162–164.

(22) Hernꢀandez-Torres, J. M.; Achkar, J.; Wei, A. J. Org. Chem. 2004,

69, 7206–7211.

(23) (a) Fraser-Reid, B.; Radatus, B. J. Am. Chem. Soc. 1970,

92, 6661–6663. (b) Fraser-Reid, B.; Tam, S. Y.-K.; Radatus, B. Can. J. Chem.

1975, 53, 2005–2016.

(24) (a) Greenspoon, N.; Keinan, E. J. Org. Chem. 1988,

53, 3723–3731. (b) Sasaki, K.; Wakamatsu, T.; Matsumura, S.; Toshima,

K. Tetrahedron Lett. 2006, 47, 8271–8274.

’ REFERENCES

(1) (a) Tolstikov, A. G.; Tolstikov, G. A. Russ. Chem. Rev. 1993, 62

(6), 579–601.(b) Ferrier, R. J.; Hoberg, J. O. In Advances in Carbohydrate

Chemistry and Biochemistry; Horton, D., Ed.; Academic Press: New York,

2003; Vol. 58, pp 55ꢀ119.

2546

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

The Journal of Organic Chemistry

ARTICLE

(25) Davis, N. J.; Flitsch, S. L. Tetrahedron Lett. 1993, 34, 1181–1184.

(26) van den Bos, L.; Codꢀee, J. D. C.; van der Toorn, J. C.; Boltje,

T. J.; van Boom, J. H.; Overkleeft, H. S.; van der Marel, G. A. Org. Lett.

2004, 6, 2165–2168.

(27) Abdulla, R. F.; Brinkmeyer, R. S. Tetrahedron 1979, 35, 1675.

(28) (a) Whistler, R. L.; Wolfrom, M. L. In Methods in Carbohydrate

Chemistry; Academic Press: New York, 1962; Vol. I, pp 85 and 184. (b)

Tolstikov, A. G.; Khakhalina, N. V.; Spirikhin, L. V. Synthesis 1988, 221–222.

(c) Chmielewskia, M.; Foktb, I.; Grodnera, J.; Grynkiewiczc, G.; Szeja, W.

J. Carbohydr. Chem. 1989, 735–744.

(29) Adam, W.; Saha-M€oller, C. R.; Zhao, C. G. In Organic Reactions;

John Wiley & Sons, Inc.: New York, 2004; Vol. 61, pp 219ꢀ561.

(30) Gibert, M.; Ferrer, M.; Sꢀanchez-Baeza, F.; Messeguer, A.

Tetrahedron Lett. 1997, 53, 8643–8650.

(31) Stevenin, A.; Boyer, F.; Beau, J.-M. J. Org. Chem. 2010,

75, 1783–1786.

(32) Mycock, D. K.; Sherlock, A. E.; Glossop, P. A.; Hayes, C. J.

Tetrahedron Lett. 2008, 49, 6390–6392.

(33) DMDO oxidation of a gulal derivative similar to 6 was reported

to proceed with 1:1 facioselectivity at 0 °C; see ref 5.

(34) Haasnoot, C. A. G.; de Leeuw, F. A. A. M.; Altona, C.

Tetrahedron 1980, 36, 2783–2792.

(35) It should be noted that epoxide ring opening under protic

conditions (methanolysis) does not always proceed by an S_N2 mechanism.

(36) (a) Jackel, C.; Dotz, H. J. Organomet. Chem. 2001, 624, 172–185.

(b) Wittman, M.; Halcomb, R.; Danishefsky, S. J. Org. Chem. 1990,

55, 1979–1981.

(37) Halcomb, R.; Boyer, S.; Wittman, M.; Olson, S.; Denhart, D.;

Liu, K.; Danishefsky, S. J. Am. Chem. Soc. 1995, 117, 5720–5749.

(38) Kihlberg, J.; Frejd, T.; Jansson, K.; Magnusson, G. Carbohydr.

Res. 1986, 152, 113–130.

(39) Mori, Y.; Morishima, N. Bull. Chem. Soc. Jpn. 1993, 66, 2061–2067.

2547

dx.doi.org/10.1021/jo102382r |J. Org. Chem. 2011, 76, 2532–2547

Article Doi

Stereoelectronic factors in the stereoselective epoxidation of glycals and 4-deoxypentenosides

DOI: 10.1021/jo102382r

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jo102382r

Full text of DOI:10.1021/jo102382r

Products guided by the article

R&D Labs maybe for 149625-92-3

Relevant to this article

Hot Product