A carbohydrate-based julia-kocienski reagent for syntheses of chain-extended and C-linked saccharides

Article abstract of DOI:10.1055/s-0033-1340851

The carbohydrate-derived Julia-Kocienski reagent 2-{[(3aS,4S,6R,6aR)-6- methoxy-2,2-dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methylsulfonyl}-1, 3-benzothiazole (6) was prepared from d-ribose and investigated in the eponymous olefination. The base-promoted generation of the Julia anion induced a rearrangement to the corresponding l-lyxose epimer 2-{[(3aS,4R,6R,6aR)-6- methoxy-2,2-dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methylsulfonyl}-1, 3-benzothiazole (12), which reacted readily with aldehydes and with a gluconolactone. The latter reaction furnished an exo-glycal-linked C-diglycoside. Georg Thieme Verlag Stuttgart, New York.

Full text of DOI:10.1055/s-0033-1340851

PAPER
▌1185
paper
A Carbohydrate-Based Julia–Kocienski Reagent for Syntheses of Chain-
Extended and C-Linked Saccharides
Carbohydrate-Based Julia–Kocienski Reagent
James A. Bull, Horst Kunz*
Institut für Organische Chemie, Johannes Gutenberg-Universität Mainz, Duesbergweg 10-14, 55128 Mainz, Germany
Fax +49(6131)3924786; E-mail: hokunz@uni-mainz.de
Received: 11.12.2013; Accepted after revision: 03.02.2014
ered through the ring system, offering production of the
carbanion through conjugate re-addition and subsequent
ring closure. A similar approach has been investigated by
Secrist et al., who prepared the corresponding triphe-
Abstract: The carbohydrate-derived Julia–Kocienski reagent 2-
{[(3aS,4S,6R,6aR)-6-methoxy-2,2-dimethyltetrahydrofuro[3,4-
d][1,3]dioxol-4-yl]methylsulfonyl}-1,3-benzothiazole (6) was pre-
pared from D-ribose and investigated in the eponymous olefination.
The base-promoted generation of the Julia anion induced a nylphosphosphonium salt 1 to use its ylide in the Wittig
olefination.7 The ylide reacted readily with aromatic and
aliphatic aldehydes to form the chain-extended carbohy-
rearrangement to the corresponding L-lyxose epimer 2-
{[(3aS,4R,6R,6aR)-6-methoxy-2,2-dimethyltetrahydrofuro[3,4-
d][1,3]dioxol-4-yl]methylsulfonyl}-1,3-benzothiazole (12), which
drate. Interestingly, in all cases the product had undergone
reacted readily with aldehydes and with a gluconolactone. The latter
a rearrangement to the L-lyxose epimer 2 via concomitant
reaction furnished an exo-glycal-linked C-diglycoside.
epimerization of the C4 position through the ring-opened
Key words: chain-extended carbohydrates, Julia–Kocienski olefi-
β-elimination product 37 (Scheme 1). Furthermore, when
nation, epimerization, lactone olefination, Smiles rearrangement
the ylide was presented with various ketones it was ob-
served to undergo a self-condensation reaction with the
intermediate aldehyde 4 to afford the alkene-linked deca-
furanose derivative 5. The fact that the self-condensed
product contains solely L-lyxose and that no self-conden-
As part of ongoing investigations into new methodologies
for the incorporation of C-linked carbohydrate units into
saccharides and glycopeptides,1 our attention turned to the
sation is observed when the coupling reaction succeeds in-
Julia–Kocienski olefination2 as a promising and underde-
dicates that the epimerization is a faster process than
veloped tool in the synthesis of glycoconjugates and long-
condensation with carbonyl compounds.
chain carbohydrates. As carbohydrate units integrated
into the backbone of larger structures are a common fea-
+
1. LiHMDS
2. ArCHO
OMe
O
ture in natural products, they present themselves as impor-
tant synthetic targets.3 The Kocienski–Lythgoe
modifications of the Julia olefination have already been
used to great effect in the preparation of such compounds,
however, to date research has focused on using the carbo-
hydrate fragment as the carbonyl-containing substrate
rather than the reagent itself. Furthermore, such protocols
are most commonly applied to the preparation of simple
exo-glycals from carbohydrate lactones.4 These exo-gly-
cals are typically the products of sterically undemanding
electron-poor reagents and simple carbonyl methylenea-
tions.5 The Julia–Kocienski reaction has thereby demon-
strated its synthetic compatibility with carbohydrate
chemistry.
Ph3P
OMe
O
Ar
O
O
O
O
2
2.
1
+
1.
–O
OMe
Ph3P
+
Ph3P
–O
OMe
O
O
3'
O
O
3
With the hope of utilizing the innate advantages of this
versatile methodology, we envisaged the benzothiazole-
2-sulfone 6 as a promising candidate for effecting the cou-
pling of a furanose with carbonyl compounds, being, to
our knowledge, the first example of a carbohydrate-based
Julia reagent.6 However, the formation of stabilized an-
ions on carbohydrate systems is known to be problematic
due to the invariable presence of a leaving group in the β-
position. In this case, however, the leaving anion is teth-
– OMe
H
OMe
O
+
Ph3P
+
Ph3P
O
O
O
O
O
O
O
5
4
Scheme 1 Carbohydrate carbon-chain extension through Wittig
reaction7
We theorized that as in the case of the 5-sulfonyl-5-de-
oxyriboside 6 (Scheme 2), an analogous β-elimination
would generate a Michael-type acceptor 7 that would en-
SYNTHESIS 2014, 46, 1185–1190
Advanced online publication: 18.02.2014
DOI: 10.1055/s-0033-1340851; Art ID: SS-2013-T0674-OP
© Georg Thieme Verlag Stuttgart · New York
0
0
3
9
-
7
8
8
1
1
4
3
7
-
2
1
0
X

1186
J. A. Bull, H. Kunz
PAPER
gender a greater stability to the system and reduce the In order to investigate the reactivity of the carbohydrate-
likelihood of eliminative self-condensation and/or derived Julia reagent, 6 was deprotonated with potassium
epimerization, shifting the ring opening–closure equilibri- hexamethyldisilazanide and the resultant anion reacted
um in favor of the desired Julia anion 6′. As a conse- with carbonyl compounds (Table 1). During these studies,
quence, Julia olefination with carbonyl compounds 1H NMR analysis of the olefination products 8 confirmed
including those less reactive than aldehydes should fur- that an epimerization had, indeed, occurred in all cases.
nish coupling products of type 8.
We were pleased to notice no self-condensation products
corresponding to 5.
R2
R1
Table 1 Julia–Kocienski Olefination Using a Ribose-Derived Julia
Anion
OMe
O
N
S
Btz
O
O
KHMDS
8'
S
R2
OMe
S
or LiHMDS
O
OMe
O
O
OMe
O
O
O
LiHMDS
O
R2
+
R1
R2
O
OMe
O
O
O
O
O
O
R2
O
6
R1
R1
8
6
O
O
O
R1
8
Carbonyl compound
Product
LiHMDS
OMe
O
H
O
N
Ph
O
O
S
N
S
OMe
O
O
S
OMe
O
S
8a
O
O
quant., E/Z 1:0.05a
O
Li
O
O
6'
7
O
O
OMe
O
H
O
Scheme 2 Julia–Kocienski olefination with a carbohydrate-derived
Julia anion
O
O
Ph
Compound 6 was readily prepared in high yield in four
steps from D-ribose (Scheme 3). D-Ribose was converted
into methyl 2,3-di-O-isopropylidene-ribofuranoside by
acid-catalyzed treatment with equimolar amounts of
methanol and acetone.8 Reaction of this compound with
triphenylphosphine, iodine, and imidazole9 afforded 5-
iodo-5-deoxy-riboside 9.10 Nucleophilic substitution of 9
with 1,3-benzothiazole-1,3-thiol under basic conditions
8b
90%, EE/ZE 1:0.3a
OMe
O
OBn
O
OBn
O
BnO
O
H
O
OBn
BnO
BnO
O
OBn
BnO
11
8c
34%, ratio of products 1:0.3a,b
furnished
5-[(1,3-benzothiazol-2-yl)thio]-substituted
a E/Z ratio calculated from 1H NMR signals.
b E and Z assignment ambiguous due to signal overlap with benzylic
protons.
furanoside 10. Its oxidation with magnesium
monoperoxyphthalate11 yielded the precursor 6 of the ri-
bofuranoside Julia reagent 6.
The optimum conditions for the condensation of 6 with al-
dehydes was found in this case not to be the standard
Barbier conditions.2 The conversion proceeds in higher
turnover when the Julia anion is generated first by the
treatment with lithium or potassium hexamethyldisilaza-
nide, followed by the addition of an aldehyde. Although
Barbier conditions lead to a poorer turnover, the fact that
the coupled products formed under these conditions had
undergone epimerization shows that the rate of reaction of
epimerization is considerably faster than that of the con-
densation itself.
N
S
I
HO
OMe
S
OMe
O
O
O
OH
a,b
c
d
6
HO
OH
O
O
O
O
10
9
Scheme 3 Synthesis of the ribose-derived Julia sulfone. Reagents
and conditions: (a) HCl, acetone–MeOH (1:1); (b) Ph3P, I2, imidaz-
ole; (c) 1,3-benzothiazole-2-thiol, K2CO3, acetone, 95%; (d) magne-
sium monoperoxyphthalate, CH2Cl2–EtOH (9:1), 57%.
A control experiment was carried out without the addition
of a carbonyl substrate in order to investigate the epimer-
ization and stability of the reagent when allowed to warm
Synthesis 2014, 46, 1185–1190
© Georg Thieme Verlag Stuttgart · New York

PAPER
Carbohydrate-Based Julia–Kocienski Reagent
1187
to ambient temperature (using LiHMDS as base at –78 °C found to induce the coupling of 6 with acetone or citral
in THF). This resulted in the prevailing conversion of 6 leading, in both cases, to quantitative recovery of the epi-
into the lyxose epimer 12 (6/12 1:9 by 1H NMR analysis) mer 12.
(Scheme 4).
As a more demanding test substrate the tetra-O-benzyl-D-
gluconopyranolactone 11 was prepared from glucose ac-
Btz
cording to a known procedure.14 On exposure to our Julia
LiHMDS, THF
S
conditions we observed the formation of the C-diglyco-
OMe
OMe
Btz
O
O
O
O
O
sidic exo-glycal 8c in trace quantities, along with multiple
decomposition products of the carbohydrate lactone. Op-
timization studies managed to elevate the yield of 8c to
34% (prolonged reaction times and stepwise addition of
base). Mass spectrometric studies of the crude reaction
mixture and certain complex mixed fractions after chro-
matography detected the persistence of a species at 923
m/z, believed to be the protonated form of the pre-Smiles
rearrangement intermediate 13 in the Julia mechanism. It
is conceivable that 13 is hindered in adopting the synclinal
conformation required for the Smiles rearrangement.
S
–78 °C to r.t.
O
O
O
O
O
6
12
Scheme 4 Epimerization of the 5-sulfonyl-5-deoxy-ribose 6
The curious fact that the intuitively more sterically de-
manding L-lyxose 12 is the major product was initially ra-
tionalized to be due to a chelation effect induced by
coordination to the lithium counterion residing on the α-
face of the ring ligated to the acetal protecting group
(Scheme 2). To investigate this effect the reagent 6 was
exposed to various bases and coordinating agents
(KHMDS, LiHMDS, DBU, HMPA, and MgBr2). In all
cases, no appreciable difference was observed, suggesting
that the driving force of this rearrangement is indeed ste-
ric.
That this normally unobserved intermediate in the Julia–
Kocienski reaction is stable to both the reaction conditions
and chromatography, albeit in trace amounts, may explain
the inherent poor turnover of this transformation. It is pre-
sumed that the intermediate anion is stabilized by opening
of the glucose ring to the keto compound 14, the ring clo-
sure back to 13 (which undergoes Smiles rearrangement
to 15) of which would also be sterically an uphill process,
due to the bulk of both the pendant lyxose moiety and the
benzyl-protecting groups along the glucose chain
(Scheme 5). Furthermore, the instability of the lactone
(and probably 14) to base over the required prolonged re-
action times also acts to reduce the turnover of this reac-
tion.
The C4 inversion of similarly substituted D-ribose deriva-
tives under basic conditions is already known in the liter-
ature with a handful of reports on similar systems.7,12 For
example, Holy showed that the corresponding 4-carboxyl-
ic acid of 6 on treatment with sodium methoxide efficient-
ly underwent epimerization.13 The fact that a 1-methyl-
2,3-isopropylidene protection pattern is a common struc-
tural motif in these rearrangements suggests that the steric
impetus for this rearrangement is a 1,3-diaxial interaction
between the sulfone and methoxy substituents, possibly
heightened by the torsional constraints of the bicyclic ac-
etal. The crystal structure of 8b (Figure 1) supports this
hypothesis.
OBn
O
OBn
O
6, LiHMDS
BnO
BnO
BnO
BnO
OMe
O
O
OBn
BnO
11
LiO
O
O
S
Btz
O
O
Smiles
rearrangement
13
OBn
O
BnO
BnO
OMe
O
BnO
BtzO
LiO
O
OBn
OLi
O
S
15
O
BnO
BnO
OMe
O
BnO
O
O
O
S
Btz
O
14
O
8c
Scheme 5 Assumed mechanism of the Julia–Kocienski olefination
of the gluconolactone 11 supported by the detection of the primary al-
dol adduct 13
Figure 1 X-ray crystal structure analysis of carbon-chain-extended
carbohydrate 8b, showing theoretical proton positions for clarity
The condensation of 6 with benzaldehyde and cinnamal-
dehyde proceeded in high yields and high E selectivity
(Table 1). The high E selectivity is in keeping with the
current mechanistic understanding of the Julia reaction on
aromatic carbonyls.2 No reaction conditions could be
The corresponding 2-deoxylactone was also prepared and
exposed to the optimized olefination conditions, in the
hope that the sterically relaxed deoxylactone would allow
for the generation of the Julia product with a higher turn-
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 1185–1190

1188
J. A. Bull, H. Kunz
PAPER
2-{[(3aR,4R,6S,6aS)-4-Methoxy-2,2-dimethyltetrahydrofu-
ro[3,4-d][1,3]dioxol-6-yl]methylthio}-1,3-benzothiazole (10)
A mixture of methyl 5-iodo-5-deoxyribofuranoside10 9 (6.0 g, 19.1
mmol), 1,3-benzothiazole-2-thiol (4.5 g, 26.8 mmol), and K2CO3
(5.3 g, 38.2 mmol) in acetone (300 mL) was heated to reflux over-
over. Unfortunately, this lactone showed retarded reactiv-
ity towards the Julia anion, leading to a further prolonging
of the reaction time and thus decomposition of the lactone,
with only trace amounts of the Julia product observed by
mass spectrometric analysis of the reaction mixture. It can night. The mixture was cooled to r.t., diluted with H2O (100 mL),
and extracted with CH2Cl2 (3 × 100 mL). The combined organic ex-
tracts were the dried (MgSO4) and concentrated in vacuo to give a
yellow oil. The product was then extracted by column chromatog-
be concluded that the 2-deoxy-gluconolactone, like ace-
tone and citral, is too CH-acidic, because one of the two
CH bonds in 2-position is parallel to the π* orbital of the
raphy (EtOAc–cyclohexane, 1:3) to give 10 as a white crystalline
solid; yield: 6.2 g (95%); mp 102–103 °C, [α]D22 –36.6 (c 2.4, ace-
lactone carbonyl. The gluconolactone 11 obviously pre-
fers a half-chair conformation with a pseudoequatorial 2-
C–H bond that is not acidified by the carbonyl group.
Keeping these conformational prerequisites in mind, other
carbohydrate lactones and carbonyl compounds could
also be used in olefination reactions with the new carbo-
hydrate-based Julia–Kocienski reagent. Despite the low
turnover this reaction pathway offers a new approach for
the synthesis of carbohydrate mimics,15 in particular, C-
diglycosides,16 as well as offering a rare glimpse into the
mechanism17 and innate limitations of the Julia–Kocienski
olefination.
tone).
IR (KBr disk): 2990, 2936, 1461, 1429, 1160, 1104, 1000, 870, 728
cm–1.
1H NMR (400 MHz, CDCl3): δ = 7.87 (d, J = 8.0 Hz, 1 H, H4/7Btz),
7.73 (d, J = 8.0 Hz, 1 H, H4/7Btz), 7.40 (t, J = 8.3 Hz, 1 H, H5/6Btz),
7.28 (t, J = 7.5 Hz, 1 H, H5/6Btz), 5.03 (s, 1 H, H1rib), 4.78 (d, J = 5.9
Hz, 1 H, H3rib), 4.68 (d, J = 5.9 Hz, 1 H, H2rib), 4.58 (t, J = 7.8 Hz,
1 H, H4rib), 3.64–3.48 (m, 2 H, H5rib), 3.40 (s, 3 H, MeO), 1.47 (s, 3
H, Me), 1.31 (s, 3 H, Me).
13C NMR (100 MHz, CDCl3): δ = 165.42, 153.06, 135.34, 126.04,
124.32, 121.61, 120.98, 112.53, 109.89, 85.40, 85.05, 83.38, 55.17,
36.36, 26.39, 24.95.
A readily prepared reagent for the Julia–Kocienski olefi-
nation reaction is described that, in a role-reversal of cur-
rent literature president, allows for access to C-linked
glycosides using the carbohydrate part as the Julia anion.
Inherent β-elimination problems are overcome using ap-
propriate design of the reagent. Likewise possible self-
condensation is avoided using the innate anion-stabilizing
ability of the sulfonylbenzothiazole.
MS (TOF MS ES+): m/z (%) = 338 [M+ – CH3] (7), 376 [M] (100).
HRMS (TOF MS ES+): m/z [M + Na] calcd for C16H19NO4NaS2:
376.0653; found: 376.0662; m/z [M] C16H19NO4S2; found:
353.0755.
2-{[(3aS,4S,6R,6aR)-6-Methoxy-2,2-dimethyltetrahydrofu-
ro[3,4-d][1,3]dioxol-4-yl]methylsulfonyl}-1,3-benzothiazole (6)
Magnesium monoperoxyphthalate (38.9 g, 78.6 mmol) was added
in one portion to a solution of 10 (11.6 g. 32.7 mmol) in CH2Cl2–
EtOH (9:1, 1.25 L). The mixture was then heated to reflux for 3 h,
cooled to r.t., and diluted with aq NaHCO3 (50 mL). This mixture
was then extracted with EtOAc (3 × 50 mL); the combined organic
extracts were dried (MgSO4) and concentrated in vacuo. The resul-
tant residue was purified by column chromatography (EtOAc–cy-
clohexane, 1:3) to give 6 as a colorless crystalline solid; yield: 7.2 g
(57%); mp 121–122 °C; [α]D22 –32.6 (c 3, acetone).
An interesting C4-epimerization occurs concomitantly
with the Julia–Kocienski condensation forming substitut-
ed vinylic L-lyxose compounds. The rearrangement ap-
pears to be a sterically driven process, the presence of
coordinating additives has no influence on the epimeriza-
tion equilibrium. The methodology presented here is one
of the few synthetic routes for the preparation of the rare
L-lyxose sugars. L-Lyxose occurs sparsely in Nature and
IR (KBr disk): 2937, 1472, 1329, 1210, 1148, 1105, 1058, 868, 764
cm–1.
is of interest in the synthesis of analogues of natural com- 1H NMR (400 MHz, CDCl3): δ = 8.23 (d, J = 7.7 Hz, 1 H, HBtz),
pounds, such as nucleoside mimics. The methodology
shows promise in the preparation of C-diglycosides by
condensation with accordingly selected carbohydrate lac-
tones. Structural alterations to glyconolactone 6 may alter
the steric constraints of the system, such as to inhibit the
epimerization and/or give improved reactivity.
8.04 (d, J = 7.5 Hz, 1 H, HBtz), 7.68–7.59 (overlapping t, 2 H, HBtz),
4.95 (s, 1 H, H1rib), 4.86 (d, J = 5.9 Hz, 1 H, H3rib), 4.80 (t, J = 6.8
Hz, 1 H, H4rib), 4.63 (d, J = 5.9 Hz, 1 H, H2rib), 3.94–3.74 (m, 2 H,
H5rib), 3.29 (s, 3 H, MeO), 1.46 (s, 3 H, Me), 1.30 (s, 3 H, Me).
13C NMR (75 MHz,): δ = 165.59, 152.64, 136.88, 128.17, 127.75,
125.49, 122.39, 112.93, 110.13, 84.82, 83.88, 80.22, 58.54, 55.42,
26.29, 24.88.
MS (TOF MS ES+): m/z (%) = 392 [M – CH3 + Na] (6), 408 [M +
Na] (100).
Solvents for moisture-sensitive reactions (THF, MeOH, CH2Cl2)
were distilled and dried according to standard procedures.18 Reac-
tions were monitored by TLC with pre-coated silica gel 60 F254 alu-
minum plates (Merck). Flash column chromatography was
performed with silica gel (32–63 μm) from Merck. 1H, 13C, and 2D
NMR spectra were recorded with a Bruker WT-300 or a Bruker
AM-400 spectrometer relative to the signal of the deuterated sol-
vent. Assignment of proton and carbon signals was achieved by ad-
ditional COSY and HMQC experiments when noted; Btz =
benzothiazole, rib = ribose, lyx = lyxose. ESI-MS were measured on
a ThemoQuest Navigator or a Micromass-Q-TOF-Ultima-3 instru-
ment. Optical rotations [α]D were measured with a Perkin Elmer po-
larimeter 241.
HRMS (TOF MS ES+): m/z [M + Na] calcd for C16H19NO6NaS2:
408.0552; found: 408.0538; m/z [M] C16H19NO6S2; found:
385.0654.
2-{[(3aS,4R,6R,6aR)-6-Methoxy-2,2-dimethyltetrahydrofu-
ro[3,4-d][1,3]dioxol-4-yl]methylsulfonyl}-1,3-benzothiazole
(12)
A solution of 6 (100 mg, 0.26 mmol) in THF (8 mL) was treated
with KHMDS (177 mg, 0.89 mmol) at –78 °C under N2. The mix-
ture was allowed to return to r.t. over 30 min, and then it was passed
through a short silica plug and evaporated in vacuo to a residue. A
1:0.1 epimeric mixture of 12 and 6 was isolated quantitatively by
flash chromatography (EtOAc–cyclohexane, 1:3) as a white crystal-
line solid; yield: 100 mg (quant.). (The analogous experiment with
LiHMDS as the base gave an epimeric mixture 12/6 of 1:0.11.) An
Synthesis 2014, 46, 1185–1190
© Georg Thieme Verlag Stuttgart · New York

PAPER
Carbohydrate-Based Julia–Kocienski Reagent
1189
analytically pure sample of the major lyxose epimer 12 was pre-
pared by recrystallization (hot MeOH); mp 76–79 °C, [α]D22 –4.94
(c 2, acetone).
MS (MS-LCMS ES+, 0.3–0.6 min): m/z (%) = 296 [M – MeO + Na]
(100), 302 [M] (8).
HRMS (TOF MS ES+): m/z [M + Na] calcd C18H22O4Na: 325.1416;
found: 325.1404; m/z [M] C18H22O4; found: 302.1518.
IR (KBr disk): 2990, 2935, 1703. 1471, 1332, 1149, 1107, 1022,
969, 863, 764 cm–1.
1H NMR (300 MHz, CDCl3): δ = 8.19 (d, J = 8.8 Hz, 1 H, HBtz),
7.99 (d, J = 7.9 Hz, 1 H, HBtz), 7.64–7.54 (m, overlapping t, 2 H,
(3aR,4R,6S,6aR)-4-Methoxy-2,2-dimethyl-6-{(E)-
[(3R,4S,5R,6R)-3,4,5-tris(benzyloxy)-6-(benzyloxymethyl)tetra-
hydro-2H-pyran-2-ylidene]methyl}tetrahydrofuro[3,4-
d][1,3]dioxole (8c) and (3aR,4R,6S,6aR)-4-Methoxy-2,2-dimeth-
yl-6-{(Z)-[(3R,4S,5R,6R)-3,4,5-tris(benzyloxy)-6-(benzyloxy-
methyl)tetrahydro-2H-pyran-2-ylidene]methyl}tetrahydro-
furo[3,4-d][1,3]dioxole (8c′)
HBtz), 4.73 (s, 1 H, H1lyx), 4.70 (dd, J = 5.8, 3.7 Hz, 1 H, H3lyx), 4.54
(q, J = 3.9 Hz, 1 H, H4lyx), 4.49 (d, J = 5.9 Hz, 1 H, H2lyx), 3.91 (d,
J = 6.0 Hz, 2 H, H5lyx), 3.20 (s, 3 H, MeO), 1.35 (s, 3 H, Me), 1.19
(s, 3 H, Me).
In a Schlenk flask under N2 the lactone 11 (200 mg, 0.37 mmol) and
6 (171 mg, 0.18 mmol) were dissolved in anhyd THF (3 mL) and
cooled to –75 °C. Then 2 M LiHMDS in THF (1.8 mL) was added
dropwise over 10 min. The mixture was stirred overnight at this
temperature, then passed through a short silica plug, then evaporat-
ed in vacuo to a residue. The resultant mixture was purified by col-
umn chromatography (EtOAc–cyclohexane, 1:3) to give a single
fraction of E/Z isomers; yield: 43 mg (34%); minor/major compo-
nent 1:2.8).
1H NMR (400 MHz, CDCl3): δ = 7.25–7.24 (m, 20 H, Harom), 5.20
(d, J = 8.3 Hz, 1 H, CH2Ph), 4.96 (s, 1 H, H1lyx), 4.83 – 4.43 (m, 17
H, overlapping, CH2Ph, Hglu, Hlyx, bridgehead-CH), 3.79–3.65 (m,
6 H, overlapping signals), 3.34 (s, 3 H, MeO, cis or trans, minor),
3.29 (s, 3 H, MeO, cis or trans, major), 1.47 (s, 3 H, Me, major),
1.41 (s, 3 H, Me, minor), 1.28 (s, 3 H, Me, major), 1.23 (s, 3 H, Me,
minor).
13C NMR (100 MHz, CDCl3): δ = 165.9, 152.6, 137.0, 128.0, 127.6,
124.0, 122.3, 113.0, 107.0, 84.5, 80.2, 73.5, 67.9, 54.8, 25.9, 24.7.
MS (TOF MS ES+): m/z (%) = 408 [M + Na] (100).
HRMS (TOF MS ES+): m/z [M + Na] calcd C16H19NO6NaS2:
408.0552; found: 408.0550; m/z [M] C16H19NO6S2; found:
385.0654.
(3aR,4R,6S,6aR)-4-Methoxy-2,2-dimethyl-6-styryltetrahydro-
furo[3,4-d][1,3]dioxole (8a)
A solution of 6 (171 mg, 0.44 mmol, 1 equiv) in THF (5 mL) at –78
°C was treated with KHMDS (212 mg, 1.07 mmol) under N2. To
this mixture benzaldehyde (0.2 mL, 1.78 mmol, 4 equiv) was intro-
duced via a syringe, and the reaction allowed to return to r.t. over 20
min. The mixture was then passed through a short silica plug, and
evaporated in vacuo to a residue. The product was isolated quanti-
tatively by flash chromatography (EtOAc–cyclohexane, 1:3) as a
colorless crystalline solid; yield: 120 mg (quant); E/Z 1:0.05. An an-
alytically pure sample prepared by recrystallization (hot MeOH)
was in agreement with literature values.7,12b
MS (TOF MS ES+): m/z (%) = 731.3 [M + Na] (100), 1439.7 [2 M
+ Na] (3); [M] C43H48O9: 708.8358.
1H NMR (300 MHz, CDCl3): δ = 7.42 (dd, J = 8.2, 1.3 Hz, 2 H,
HPh), 7.33–7.20 (m, 3 H, HPh), 6.71 (d, J = 16.0 Hz, 1 H, CH=HPh),
6.35 (dd, J = 16.1, 7.8 Hz, 1 H, CH=HPh), 4.94 (s, 1 H, H1lyx), 4.70
(dd, J = 6.07, 3.37 Hz, 1 H, H3lyx), 4.61 (d, J = 5.8 Hz, 1 H, H2lyx),
4.54 (dd, J = 7.8, 3.6 Hz, 1 H, H4lyx), 3.36 (s, 3 H, MeO), 1.50 (s, 3
H, Me), 1.31 (s, 3 H, Me).
Acknowledgment
This work was supported by the Naturstoffsynthese-Zentrum at the
Johannes Gutenberg-Universität Mainz and by the Fonds der Che-
mischen Industrie. J.A.B. is gratefully for a Postdoctoral Fellowship
of the Alexander von Humboldt Foundation. The authors would
also like to thank Dr. Dieter Schollmeyer for X-ray crystallographic
measurements.
13C NMR (100 MHz, CDCl3): δ = 136.5, 134.4, 127.9, 112.6, 85.4,
81.0, 54.8, 26.2, 24.9.
(3aR,4R,6S,6aR)-4-Methoxy-2,2-dimethyl-6-[(1E,3E)-4-phenyl-
buta-1,3-dienyl]tetrahydrofuro[3,4-d][1,3]dioxole (8b)
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synthesis.SnoIufproi
m
tgioSrantnugIifoop
r
itmnatr
A solution of 6 (240 mg, 0.62 mmol, 1.2 equiv) in THF (6 mL) at
–78 °C was treated with KHMDS (298 mg, 1.5 mmol) under N2. To
this mixture cinnamaldehyde (0.65 mL, 0.52 mmol, 1 equiv) was in-
troduced via syringe and the reaction allowed to return to r.t. over
20 min. The mixture was then passed through a short silica plug, and
evaporated in vacuo to a residue. The resultant residue was purified
by column chromatography (EtOAc–cyclohexane, 1:3) to obtain 8b
as a colorless crystalline solid; yield: 142 mg (90%); E,E/E,Z 1:0.3).
An analytically pure sample of the major E,E-component was pre-
pared by recrystallization (EtOAc–cyclohexane). Absolute stereo-
chemistry confirmed by crystal structure analysis; mp 120–123 °C,
[α]D22 +25.9 (c 2, acetone).
References
(1) Koch, S.; Schollmeyer, D.; Löwe, H.; Kunz, H. Chem. Eur.
J. 2013, 19, 7020.
(2) (a) Blakemore, P. R. J. Chem. Soc., Perkin Trans. 1 2002,
2563. (b) Aïssa, C. Eur. J. Org. Chem. 2009, 1831.
(3) Emery, F.; Vogel, P. J. Org. Chem. 1995, 60, 5843.
(4) Illustrative examples: (a) Michelet, V.; Adiey, K.; Tanier,
S.; Dujardin, G.; Genêt, J.-P. Eur. J. Org. Chem. 2003, 2947.
(b) Bourdon, B.; Corbet, M.; Fontaine, P.; Goekjian, P. G.;
Gueyrard, D. Tetrahedron Lett. 2008, 49, 747.
(5) Gueyrard, D.; Haddoub, R.; Salem, A.; Bacar, N. S.;
Goekjian, P. G. Synlett 2005, 520.
(6) With one notable exception, vinyl sulfone carbohydrates
formed from carbohydrate lactones have been shown to
undergo Julia olefination with aromatic aldehydes to yield
semicyclic dienes: Aouadi, K.; Defaut, B.; Goekjian, P. G.;
Gueyrard, D. Synlett 2007, 2590.
IR (KBr disk): 2924, 1275, 1106, 1085, 1014, 972, 862, 753, 696
cm–1.
1H NMR (400 MHz, CDCl3): δ = 7.41 (d, J = 7.0 Hz, 2 H, HPh), 7.33
(t, J = 7.1 Hz, 2 H, HPh), 7.24 (t, J = 7.4 Hz, 1 H, HPh), 6.86 (dd,
J = 15.7, 15.7 Hz, 1 H, CH=CHPh), 6.62–6.51 (m, 2 H, overlap-
ping, CH=CHPh, lyx-CH=CH), 5.96 (dd, J = 15.3, 8.0 Hz, 1 H, lyx-
CH=CH), 4.94 (s, 1 H, H1lyx), 4.70 (dd, J = 5.8, 3.6 Hz, 1 H, H3lyx),
4.61 (d, J = 5.8 Hz, 1 H, H2lyx), 4.50 (dd, J = 8.0, 3.6 Hz, 1 H,
H4lyx), 3.38 (s, 3 H, MeO), 1.51 (s, 3 H, Me), 1.33 (s, 3 H, Me).
(7) Secrist, J. A. III.; Wu, S.-R. J. Org. Chem. 1977, 42, 4084.
(8) Barrett, A. G. M.; Lebold, S. A. J. Org. Chem. 1990, 55,
3853.
13C NMR (100 MHz, CDCl3): δ = 137.1, 134.8, 133.6, 128.6, 128.2,
127.7, 126.9, 126.5, 112.6, 107.2, 85.4, 81.6, 80.7, 54.8, 26.1, 24.9.
© Georg Thieme Verlag Stuttgart · New York
Synthesis 2014, 46, 1185–1190

1190
J. A. Bull, H. Kunz
PAPER
Scarano, F.; Bednarski, M. J. Am. Chem. Soc. 1992, 114,
10639. For 1,6-C-linked disaccharides, see, for example:
(c) Dondoni, A.; Marra, A.; Mizuno, M.; Giovanni, P. P.
J. Org. Chem. 2002, 67, 4186. (d) McAllister, G. D.;
Paterson, D. E. R.; Taylor, J. K. Angew. Chem. Int. Ed. 2003,
42, 1387. (e) Lubin-Germain, N.; Hallonet, A.; Huguenot,
F.; Palmier, S.; Uziel, J.; Augé, J. Org. Lett. 2007, 9, 3679.
(f) Chambers, D.; Evans, G. A.; Fairbanks, A. J.
(9) (a) Garegg, P. J.; Samuelsson, B. J. Chem. Soc., Perkin
Trans. 1 1980, 2866. (b) Horner, L.; Oediger, H.; Hoffmann,
H. Justus Liebigs Ann. Chem. 1959, 626, 26.
(10) (a) Mishra, G. P.; Kumar, B. S.; Venkateswara, R. B.
Tetrahedron: Asymmetry 2012, 23, 1161. (b) Fürstner, A.;
Jumbam, D.; Teslic, J.; Weidmann, H. J. Org. Chem. 1991,
56, 2213.
(11) (a) Cristau, H. J.; Torreilles, E.; Cassagne, T.; Lion, C.;
Desgranges, M.; Magnaud, G.; Delmas, G. Phosphorus,
Sulfur Silicon Relat. Elem. 1996, 111, 129. (b) Schenk, W.
A.; Kuhnert, N. Z. Naturforsch., B: Chem. Sci. 2000, 55b,
527. (c) Brown, D. L.; Talley, J. J.; Ludwig, C. L.; Graneto,
M. J. US 2002183362, 2002.
(12) (a) Sakaguchi, N.; Hirano, S.; Matsuda, A.; Shuto, S. Org.
Lett. 2006, 8, 3291. (b) Pal, T. K.; Pathak, T. Carbohydr.
Res. 2008, 343, 2826.
Tetrahedron 2005, 61, 7184. For 1,3-C-linked
disaccharides: (g) Zhu, Y. H.; Vogel, P. Synlett 2001, 82. For
2,3-C-linked sialyl saccharides: (h) Röhrig, C. H.; Takhi,
M.; Schmidt, R. R. Synlett 2001, 1170. (i) Notz, W.; Hartel,
C.; Waldscheck, B.; Schmidt, R. R. J. Org. Chem. 2001, 66,
4250. (j) Hirai, G.; Watanabe, T.; Yamaguchi, K.; Miyagi,
T.; Sodeoka, M. J. Am. Chem. Soc. 2007, 129, 15420.
(17) The cyclic sulfone intermediate in the Julia–Smiles
rearrangement has also been isolated: Xu, D. D.; Waykole,
L.; Calienni, J. V.; Ciszewski, L.; Lee, G. T.; Liu, W.;
Szewczyk, J.; Vargas, K.; Prasad, K.; Repič, O.; Blacklock,
T. J. Org. Process Res. Dev. 2003, 7, 856.
(13) Holy, A. Collect. Czech. Chem. Commun. 1982, 47, 2969.
(14) Kuzuhara, H.; Fletcher, H. G. J. Org. Chem. 1967, 32, 2531.
(15) Koester, D.; Holkenbrink, A.; Werz, D. B. Synthesis 2010,
3217.
(16) (a) Beau, J.-M.; Sinaÿ, P. Tetrahedron Lett. 1985, 26, 6185.
(b) Bertozzi, C. R.; Cook, D. G.; Kobertz, W. R.; Gonzalez-
(18) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon Press: Oxford, 1988.
Synthesis 2014, 46, 1185–1190
© Georg Thieme Verlag Stuttgart · New York