Applying the Bent Bond/Antiperiplanar Hypothesis to the Stereoselective Glycosylation of Bicyclic Furanosides

Article abstract of DOI:10.1021/acs.joc.9b02791

The glycosylation stereoselectivities for a series of bicyclic furanoside models have been carried out in the presence of weak nucleophiles. These results were analyzed through the bent bond/antiperiplanar hypothesis (BBAH) orbital model to test its validity. According to the BBAH, incoming nucleophiles displace one of the two bent bonds of bicyclic oxocarbenium ion intermediates in an antiperiplanar fashion. The glycosylation stereoselectivity is then governed by the displacement of the weaker bent bond as determined by the presence of electron-withdrawing or -donating substituents at C₂. Overall, the BBAH analysis expands Woerpel's "inside/outside attack" glycosylation model by considering the stereoelectronic influence of neighboring electron-withdrawing and -donating groups on the nucleophilic addition to oxocarbenium ion intermediates.

Full text of DOI:10.1021/acs.joc.9b02791

Subscriber access provided by UNIV OF WESTERN ONTARIO
Article
Applying the Bent Bond / Antiperiplanar Hypothesis to
the Stereoselective Glycosylation of Bicyclic Furanosides
Jean-François Parent, Xavier Bertrand, Ghislain Deslongchamps, and Pierre Deslongchamps
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.9b02791 • Publication Date (Web): 10 Dec 2019
Downloaded from pubs.acs.org on December 10, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 54
The Journal of Organic Chemistry
1
/54
1
2
3
4
5
6
7
8
9
Applying the Bent Bond / Antiperiplanar Hypothesis to the Stereoselective
Glycosylation of Bicyclic Furanosides
†
†
‡
Jean-François Parent, Xavier Bertrand, Ghislain Deslongchamps and Pierre
Deslongchamps*
†
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
Département de Chimie, Université Laval, Québec City, QC, G1V 0A6, Canada
Department of Chemistry, University of New Brunswick, Fredericton, NB, E3B 5A3, Canada
‡
ABSTRACT
The glycosylation stereoselectivities for a series of bicyclic furanoside models have been carried
out in the presence of weak nucleophiles. These results were analyzed through the bent bond /
antiperiplanar hypothesis (BBAH) orbital model in order to test its validity. According to the
BBAH, incoming nucleophiles displace one of the two bent bonds of bicyclic oxocarbenium ion
intermediates in antiperiplanar fashion. The glycosylation stereoselectivity is then governed by
displacement of the weaker bentꢀbond as determined by the presence of electron-withdrawing or
donating substituents at C . Overall, the BBAH analysis expands Woerpel's "inside/outside attack"
2
glycosylation model by considering the stereoelectronic influence of neighbouring electron-
withdrawing and donating groups on the nucleophilic addition to oxocarbenium ion intermediates.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 54
2
/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
We have previously reported the development and application of the bent bond / antiperiplanar
hypothesis (BBAH), a conceptually novel orbital model for rationalizing the reactivity of various
types of unsaturated systems.1,2,3,4,5,6,7 This model was also applied to better understand the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
parameters that govern the O- and C-glycosylation of pyranosides. We wish to report that the O-
and C-glycosylation of furanosides can also be analyzed in a similar fashion using a series of
bicyclic furanoside model compounds.
9
Woerpel and co-workers previously reported the C-glycosylation of several furanosides with
allyltrimethylsilane (TMSallyl) in the presence of Lewis acid. For example, a 2-
benzyloxyfuranoside acetate gave the 1,2-cis product as a major isomer (cis:trans 85:15) (Scheme
1
).9e When the OBn group was replaced by a CH
group, the reaction became modestly trans
3
1
0
selective (cis-trans 32:68). Woerpel also reported the allylation of a conformationally rigid
bicyclic furanoside, which yielded mainly the β-isomer (27α:73β).9d,f,g From these results, Woerpel
postulated that the nucleophile preferably undergoes an "inside attack" on the corresponding
3
9
oxocarbenium ion intermediate in relation to its E conformation. As illustrated in Scheme 2, the
inside attack produces a five-membered product in an initial conformation where the substituents
at C
that eclipses the C
to account for the lower β stereoselectivity for C-allylation of the trans-fused cyclohexane model
73%) compared to its more conformationally mobile cyclooctane counterpart (93%) as shown in
Scheme 1.
2
and C
3
stagger each other. The alternative “outside attack" produces a five-membered product
2
–C bond. The inside attack also eclipses the C -C bond, a factor that was used
3
3
4
(
ACS Paragon Plus Environment

Page 3 of 54
The Journal of Organic Chemistry
3
/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Allylation of furanosides.
Scheme 2. Inside and outside attack.
As will be demonstrated herein, a straightforward application of the BBAH orbital model produces
the same general outcome as the Woerpel inside/outside attack model but also accounts for the
modulation of glycosylation stereoselectivities due to stereoelectronic interactions between the
bent bonds of the oxocarbenium ion intermediates and the substituents at C
2
.
Codée and colleagues11a,b reported the acid-catalyzed additions of deuterium (Et
SiD) and allyl
3
(TMSallyl) to benzylated pentafuranoside analogues of ribose, arabinose, xylose and lyxose
bearing different substituents at C (Scheme 3). The reactions all gave 1,2-cis adducts with good
2
(85:15) to excellent (98:2) stereoselectivity. These results were rationalized on the basis of
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 54
4
/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Woerpel's inside attack model by analyzing the most stable conformation of their respective
oxocarbenium ions by ab initio calculation. In all cases, the major product had the nucleophile cis
to the C
glycosylation of bicyclic 3,5-O-di-tert-butylsilane arabinose furanoside with several carbohydrate
nucleophiles (Scheme 4). Once again, the nucleophile added cis to the C EWG.
2
electron-withdrawing group (EWG). Boons and co-workers11c also reported the 1,2-cis
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
Scheme 3. 1,2-cis Deuteration or allylation of pentafuranoside analogues (Nu=D or allyl).
Scheme 4. 1,2-cis Glycosylation of arabinose furanoside.
Guindon and co-workers12 reported a highly diastereoselective route to 1ʹ,2ʹ-cis nucleoside
analogues in the D-ribo, D-lyxo, D-xylo and D-arabinoside series. These five-membered ring
lactols all underwent highly selective N-glycosylation reactions in the presence of dimethylboron
bromide and silylated pyrimidine nucleobases (Scheme 5). For example, the five-membered lactol
ACS Paragon Plus Environment

Page 5 of 54
The Journal of Organic Chemistry
5
/54
1
2
3
4
5
6
7
8
9
having a C
group at C
So, the EWG at C
2
-OR group gave the 1ʹ,2ʹ-cis isomer in high yield (cis:trans ratio >20:1). When the OR
was replaced by a CH group, the selectivity was completely lost (cis:trans ratio 1:1).
is clearly a dominant factor in the stereoselectivity of furanoside oxocarbenium
2
3
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ion.9c-d,10b,12,13 These reactions were interpreted through an S
model.
N
1 or an “exploded” transition state
Scheme 5. Highly cis-selective N-glycosylation of lactols.
1
4
15
As opposed to the Hückel σ-π model, the bent bond (τ bond) model confers tetrahedral character
to the carbon of carbonyl groups and, by extension, to oxocarbenium ions. According to the
BBAH, nucleophilic addition to a carbonyl group is viewed as a S 2-like process in which the
N
nucleophile specifically displaces the anti bent bond, and necessarily generates an antiperiplanar
oxygen lone pair on the initial product conformation (Scheme 6).16 The reverse process,
corresponding to the elimination of the nucleophile, must occur by assistance of a strictly
antiperiplanar oxygen lone pair to reform the bent bonds of the oxocarbenium ion intermediate. In
3
the furanosyl series, nucleophilic displacement of one of the bent bonds on the E conformation of
the oxocarbenium ion yields a product conformation in which the new carbon-nucleophile bond
has no choice but to be antiperiplanar to the newly produced oxygen lone pair.
During the addition, a filled orbital on the nucleophile overlaps with the large outer lobe of the
3
antibonding τ* C-O orbital (Scheme 6), and release the sp oxygen lone pair in antiperiplanar
fashion.1-8 The electron density of each τ bond can vary independently depending on the
stereoelectronic alignment of neighbouring electron-withdrawing or donating groups through
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 54
6
/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
17
hyperconjugation; the BBAH further states that the nucleophile will preferably displace the
weaker (i.e. electron poorer) τ bond. As shown in Scheme 6, one of the oxocarbenium τ bonds is
electronically enriched by hyperconjugation with an antiperiplanar C-H bond (a σC-H→τ*
interaction), while the other τ bond is weakened by the antiperiplanar electron-withdrawing OR
group (a τ→σ*C-OR interaction). As a result, the nucleophile prefers to displace the weaker τ bond
and adds cis to the OR group. One should note that, in the σ-π model, the two π* lobes correspond
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
18
to the same orbital and cannot be differentiated in this way. The Cieplak effect and Inomata’s
1
9
syn effect can thus be easily rationalized using the τ bond model. Woerpel does mention possible
hyperconjugation between axial C
2
–H bonds in pseudoequatorial pyranosyl oxocarbenium ions.20
Scheme 6. Left: Orbital overlap between incoming Nu and antibonding τ* orbital of anti C–O τ
bond produces sp oxygen lone pair antiperiplanar to bound Nu. Right: facial selectivity modulated
3
by hyperconjugation between C
of the oxocarbenium ion.
2
-OR and C -H groups antiperiplanar to their respective τ bond
2
All the previous results presented in Schemes 1-5 are readily rationalized by the BBAH orbital
model. To further test this model, we synthesized and studied the glycosylation of a series of
conformationally rigid trans bicyclic furanosides 1-5 (Scheme 7). Each compound was prepared
with either OAc or SPh group at the anomeric position to allow for both C- and O-glycosylation
reactions to be analyzed. Four of these bicyclic models have a OCH
3
or a CH group located either
3
in a β-pseudoequatorial or α-pseudoaxial orientation at C and a trans bicyclic ring fusion that
2
confines their corresponding 5-membered oxocarbenium ions to basically exist in their 3E
ACS Paragon Plus Environment

Page 7 of 54
The Journal of Organic Chemistry
7
/54
1
2
3
4
5
6
7
8
9
conformation. This conformational restriction can potentially affect glycosylation selectivity. Our
model compounds do not perfectly reflect the reactivity of fully oxygenated carbohydrates, our
goal being to pinpoint the specific stereoelectronic contribution of the C -substituents within the
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
BBAH paradigm. This requires simplified carbohydrate donor surrogates in which the
oxocarbenium ion is unaffected by other conformational and stereoelectronic factors.
Nevertheless, the term glycosylation can still be applied as the conformational limitations of
furanoses and the particular reactivity of the anomeric center are two aspects that are shared in our
approach.
Scheme 7. Glycosylation donors. LG=OAc or SPh.
RESULTS AND DISCUSSION
Synthesis of bicyclic models
The starting trans-fused bicyclic furanosides were obtained by vinyl anion addition to cyclohexene
oxide.21 The non-substituted donors 1 (Scheme 8) were prepared by hydroboration and
deprotection of 6, which gave diol 7 followed by selective TEMPO oxidation to yield bicyclic
2
2
lactone 8. The C-glycosylation donor 1(OAc) was then prepared from 8 through a one-pot
9d
reduction-acetylation procedure. The O-glycosylation donor 1(SPh) was obtained from 1(OAc)
using a literature procedure.23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 54
8
/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
H
H
H
1
) BH -Me S
THF, 0 °C to r.t., 18 h
3 2
OH
OTBS
TEMPO, (AcO)
IPh
O
2
DCM, r.t., 3 h
2
3
) NaOH 3M, H O 30 %
2
2
OH
H
O
THF, 0 °C to r.t., 3 h
) TBAF (1M in THF)
THF, r.t., 1 h
85 %
8
6
7
50 % over two steps
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
) DIBAL-H (1M in heptanes)
Tol., -78 °C, 2 h
BF -Et O, PhSH
3
2
O
O
2
) Ac O, DMAP, Pyr.
DCM, -78 °C, 1 h
2
DCM, -78 °C to r.t., 18 h
SPh
OAc
1(SPh)
5a:85b
1
(OAc)
50 %
60a:40b
1
66 %
Scheme 8. Synthesis of type 1 donors.
The preparation of C
2
-OCH
3
donors 2 and 3 required a more complex synthetic route, similar to
8
that reported for the synthesis of substituted bicyclic pyranosides (Scheme 9). Dihydroxylation
of 6 gave a 2:1 mixture of diols 9 in which the primary alcohol was further oxidized to the
carboxylic acid using TEMPO followed by a Pinnick oxidation of the aldehyde intermediate.24
After allyl esterification to yield compound 10, the remaining secondary alcohol was methylated
using standard methodology. The resulting diastereoisomers 11/11ʹ were separated by
chromatography and submitted to a deprotection/cyclization sequence to provide lactones 12 and
1
3. The donors 2(OAc), 3(OAc), 2(SPh) and 3(SPh) were obtained using the same reaction
sequence described above from the corresponding lactone.
ACS Paragon Plus Environment

Page 9 of 54
The Journal of Organic Chemistry
9
/54
1
2
3
4
5
6
7
8
9
H
H
H
1
) TEMPO, KBr, NaClO
DCM: NaHCO 0 °C, 1 h
2) NaOCl , NaH PO
2-methyl-2-butene
t-BuOH, 0 °C, 1 h
2
OTBS
OTBS
OH
OTBS
OH
OsO
THF:H O, r.t., 18 h
7 %
4
, NMO
3
2
2
2
4
H
6
H
H
9
OH
AllylO
O
9
d.r.2:1
3) Bu NI, AllylBr, K CO
4
2
3
1
0
DMF, r.t., 2 h
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
77 %
H
H
OTBS
OMe
1) HF-Pyridine
THF, r.t., 18 h
2
) NaH, MeI
THF, 0 °C to r.t., 18 h
7 %
O
X
2) TsOH
*
6
DCM, r.t., 18 h
O
Y
AllylO
O
1
2 X=OMe,Y=H 73 %
13 X=H, Y=OMe 82 %
11/11'
1) DIBAL-H (1M in heptanes)
Tol., -78 °C
O
BF
3
-Et
2
O, PhSH
O
2) Ac O, DMAP, Pyr.
X
X
2
DCM, -78 °C, 1 h
DCM, -78 °C to r.t., 18 h
OAc
SPh
Y
Y
2
3
(OAc) X=OMe,Y=H 89 % (33a:67b)
(OAc) X=H, Y=OMe 90 % (37a:63b)
2
3
(SPh) X=OMe,Y=H 59 % (33a:67b)
(SPh) X=H, Y=OMe 73 % (49a:51b)
Scheme 9. Synthesis of type 2 and 3 donors.
The syntheses of donors 4 and 5 having a methyl group at C
2
began by diethylmalonate anion
displacement of cyclohexene oxide, which gave a-carboxyethylbutyrolactone 14 (Scheme 10).
Marshall’s reduction protocol25 then yielded the desired allylic alcohol 15, which was
hydrogenated with catalytic platinum(IV) oxide, providing a ~1:1 mixture of diastereoisomeric
2
6
diols 16/16ʹ. Chromatographic separation followed by Stahl oxidation produced corresponding
lactones 17 and 18. As this reaction is known to use atmospheric oxygen as the external oxidation
26
source, larger scale reactions were speeded-up using pure oxygen gas. These two lactones were
obtained in good yields and converted to donors 4 and 5. To do so, 17 and 18 were reduced with
DIBAL-H and acetylated to yield 4(OAc) (13a:87b) and 5(OAc) (16a:84b) respectively,
providing 4(SPh) (90a:10b) and 5(SPh) (12a:88b) with PhSH and BF
3
·Et O.
2
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 54
1
0/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 10. Synthesis of type 4 and 5 donors.
Glycosylation experiments
Glycosylation experiments were carried out on bicyclic furanosides 1-5 (Scheme 11) having OAc
or SPh as anomeric leaving groups, allowing for both C- and O-glycosylation to be investigated.
For C-glycosylation, the OAc derivatives were reacted with BF
h at low temperature. For O-glycosylation, the SPh derivatives were reacted with NIS and 2,2,2-
trifluoroethanol (TFE) in CH Cl or CH CN for 2 h (Scheme 11). These conditions were based on
3
·Et
2
O and TMSallyl in CH
2
Cl
2
for
2
2
2
3
the work reported by Woerpel.23 The C-glycosylation conditions do not allow for product
epimerization due to the nature of the nucleophile. O-glycosylation reaction conditions can,
theoretically, lead to product epimerization. However, we showed that this does not occur under
8
the experimental conditions ; isolated products resubmitted to glycosylation conditions with a
different donor and nucleophile showed no change from their initial anomeric ratio.
Thermodynamic conditions were also performed on an isolated product, showing a change in the
anomeric ratio over time. This showed that the thioglycosides activated by NIS provided a clean,
ACS Paragon Plus Environment

Page 11 of 54
The Journal of Organic Chemistry
1
1/54
1
2
3
4
5
6
7
8
9
kinetic glycosylation reaction. All the SPh donors allowed for the reactions to be carried out
without any Lewis acid or stabilizing triflate source, which often can influence the reaction
selectivity (Scheme 11). All product ratios reported in Table 1 were duplicated and determined by
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
19
H NMR and/or F NMR using the crude mixture of reaction products. The relative configurations
at the anomeric center were established by 1D NOESY experiments. Any variation in the reported
isolated yields was attributed to the volatility and stability of the glycosylated adduct on silica gel.
1
27
Crude H NMR showed a complete conversion to the desired compound in almost every case.
Scheme 11. Reaction conditions for C and O-glycosylation described in Table 1.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 54
1
2/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 1. Glycosylation of bicyclic donors 1-5.
T °Cd
-78
-40
-40
-78
-40
-78
-40
-40
-78
-40
-78
-40
-78
-40
-40
-78
-40
-40
α:β ratio Yield
Solvent
CH Cl
''
CH CN
CH Cl
Entry
Donor
LG
OAc
Nu
e
1
2
1 (X=Y=H)
TMSallyl
2
2
30:70
51:49
27:73
34:66
34:66
98:2
N.D.
37 %
61 %
96 %
67 %
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
''
SPh (15α:85β)
''
TFE
3
''
''
3
4
2(X=OMe, Y=H)
OAc
TMSallyl
2
2
5a
6b
7
''
''
''
''
SPh (α only)
SPh (β only)
SPh (α only)
SPh (69α:31β)
TFE
''
''
c
''
N.D.
''
CH
3
CN
34:66
33:67
6:94
86 %
74 %
83 %
76 %
N.D.
58 %
98 %
73 %
56 %
90 %
73 %
56 %
8
''
''
9
3(X=H, Y=OMe) OAc
TMSallyl
CH
2
''
''
Cl
2
10a
''
SPh (49α:51β)
TFE
18:82
13:87
22:78
60:40
86:14
92:8
1
1b
''
''
''
1
2
3
''
''
''
CH
CH
''
CH CN
CH Cl
''
CH CN
3
CN
1
4(X=Me, Y=H)
OAc (13α:87β)
SPh (90α:10β)
''
TMSallyl
2
Cl
2
14
''
TFE
1
1
1
5
6
7
''
''
TMSallyl
TFE
3
5(X=H, Y=Me)
OAc (16α:84β)
SPh (12α:88β)
''
2
2
>3:97
7:93
''
''
18
''
3
9:91
a
The initial temperature was -40 °C and brought to approximately -15 °C. No conversion was
b
observed below -20°C. AgOTf was added as a triflate source and full conversion occurred at -78
c
1
19
d
°
C. No purification performed. Ratio determined using the crude H NMR and/or F NMR.
Initial temperature in CH
2
Cl for donor 2 and 3 was -40°C and slowly brought to -20 °C over the
2
e
2 h reaction period. N.D.: not determined.
ACS Paragon Plus Environment

Page 13 of 54
The Journal of Organic Chemistry
1
3/54
1
2
3
4
5
6
7
8
9
8
23,28
As demonstrated in our previous work and in other studies,
stereoselectivity at the anomeric
2
9
carbon is highly affected by the nature of the nucleophile. Stronger nucleophiles generally exhibit
a lower stereoselectivity upon addition to oxocarbenium ion intermediates as a result of earlier
glycosylation transition state geometries that are less sensitive to steric and electronic asymmetry
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
near the electrophilic center. However, oxocarbenium ion intermediates may not form easily and,
13
as a result, glycosylations may proceed through an exploded S 2-like transition state, especially
N
3
0
in the presence of strong nucleophiles or by forming nitrilium ion intermediates. Glycosylations
with weak nucleophiles or stabilized glycosyl donors are expected to be more S 1-like; the
N
nucleophile addition must then proceed by a later transitions state, and be more heavily influenced
by steric hindrance and other factors as the nucleophile approaches the electrophilic center. To
tease out the factors guiding glycosylation reactions involving oxocarbenium ion intermediates,
only the weaker nucleophiles TMSallyl and TFE were considered for this study. Carbon based
29
nucleophiles are believed to react strictly through an S 1 mechanism, whereas TFE is generally
N
2
3
assumed not to be nucleophilic enough for an S 2 mechanism. It is thus anticipated that the
N
anomeric configuration of the donor moiety should not influence the glycosylation
31
stereoselectivity under the reaction conditions described in Scheme 11 as they are all assumed to
involve oxocarbenium ion intermediates. Results presented for donor 2 (entries 7 and 8) support
this statement as the initial anomeric orientation of the donor did not influence the selectivity
outcome of the reaction.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 54
1
4/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
As shown in Table 1, donors 1-3 and 5 preferentially formed β-glycosides with the TMSallyl
nucleophile (entries 1, 4, 9 and 16) whereas donor 4 produced the α-glycoside in slight excess
(entry 13). With the TFE nucleophile, O-glycosylation in CH
2
Cl
2
or CH CN showed a similar
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
trend in most cases. Donors 1-3 and 5 gave preferentially β-O-glycosides (entries 3; 5, 7-8; 10, 12
and 17-18). As above, donor 4 gave again mainly α-O-glycosides (entries 14-15).
Overall, the BBAH analysis expands Woerpel's "inside/outside attack" glycosylation model by
considering the stereoelectronic influence of neighbouring electron-withdrawing and donating
groups on the nucleophilic addition to oxocarbenium ion intermediates. As shown in Scheme 12,
3
inside attack on the fixed E conformation of the oxocarbenium ion directly furnishes a β-glycoside
in a conformation where the C
outside attack produces an α-glycoside in a conformation that eclipses the X and Y substituents at
with the axial H and C methylene at C . The BBAH analysis further considers any potential
hyperconjugation between ꢁ bonds in the oxocarbenium intermediates and the EWG or electron-
donating groups (EDG) at C in the glycosylation reactions.
2
substituents (X/Y) stagger the C
1
and C
3
groups. Conversely, the
C
2
8
3
2
Scheme 12. Steric and conformational factors to the S 1 glycosylation reaction of substituted
N
bicyclic furanosides.
The C-glycosylation of donors 1-5 with TMSallyl in CH
indicated in entry 1 (Table 1 or Scheme 13) using the carbon-based nucleophile, the bicyclic
furanoside 1 with no substituent at C gave a 30:70 ratio of the α- and β-anomers, both under
2
Cl
2
is illustrated in Scheme 13. As
2
stereoelectronic control. This is in accordance with the results previously reported by Woerpel in
ACS Paragon Plus Environment

Page 15 of 54
The Journal of Organic Chemistry
1
5/54
1
2
3
4
5
6
7
8
9
9c
which the b-isomer is favored for conformational reasons as a result of the inside attack. With
TFE in CH Cl (entry 2), there is an increase in the a-isomer as the stronger nucleophile predicates
an earlier transition state. In CH CN (entry 3), the oxocarbenium ion more stabilized so the
2
2
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
0
reaction may involve a somewhat later transition state or contact ion pair. The b selectivity is
thus observed as with TMSallyl.
Scheme 13. C-Glycosylation of donors 1-5 with TMSallyl and TFE in CH
2
Cl
2
and CH CN.
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 54
1
6/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In the case of bicyclic oxocarbenium ion 2 bearing a β-OCH
entry 4) favors the β-anomer (i.e. inside attack) by favorable displacement of the α-bent bond that
is weakened by hyperconjugation with the antiperiplanar β-OCH group; conversely, the β-bent
bond is instead stabilized as it is antiperiplanar to the donating C
-H bond.17,32 Thus, α-attack is
disfavored on stereoelectronic grounds. As for the steric effects, α-attack should be favored as
3
group, the addition of TMSallyl
(
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
trans addition avoids repulsion between the β-OCH
3
and the incoming nucleophile but is also
33
disfavored because of eclipsing of the C -C bond in the α-addition product. Overall, cis addition
2
3
to donor 2 is slightly favored, and the results with TFE as nucleophile (entries 5, 7-8) are similar
to those of TMSallyl (entry 4). The presence of a polar OCH group must produce a more reactive
oxocarbenium ion, but TFE is a stronger nucleophile than TMSallyl. As a result, the a:b selectivity
remains the same. Interestingly, comparing the reaction of donors 1 and 2 with TFE in CH Cl
entries 2 and 5) shows an increase of inside attack with donor 2 (17%). This indicates that the b-
3
2
2
(
OCH
3
hyperconjugation with the a-bent bond is more important than the steric repulsion between
during 1,2-cis addition. In the more polar CH CN, the inside
attack on donor 2 (entries 7-8) is slightly less that on donor 1 (entry 3).
the nucleophile and the β-OCH
3
3
In the case of donor 3 bearing an α-OCH
is stereoelectronically disfavored by a weakening of the β-bent bond (antiperiplanar to α-OCH
and concomitant enrichment of the α-bent bond (antiperiplanar to β-C
attack is sterically favored as it occurs trans to the α-OCH and the inside attack staggers the C
bond; α-attack would eclipse the C –C bond including severe repulsion between α-OCH and C
On the other hand, formation of the a-isomer is stereoelectronically favored by hyperconjugative
3
group (entry 9), the b-anomer through the inside attack
3
)
2
–H bond). However, the b-
3
2
-C
3
2
3
3
8
.
effects of the C
2
substituents but is quite sterically disfavored by the outside attack eclipsing the
ACS Paragon Plus Environment

Page 17 of 54
The Journal of Organic Chemistry
1
7/54
1
2
3
4
5
6
7
8
9
C
2
-C
the a-isomer is produced in only 6% with TMSallyl (entry 9). With TFE in CH
entries 10 and 12), an increase of the a-isomer is observed (~20%) which is probably due to the
3
bond, and important repulsion between the OCH
3
and the incoming nucleophile. As a result,
2
Cl or CH CN
2
3
(
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
stronger nucleophile, resulting in a slightly earlier transition state, and lessening the steric
selectivity factors.
Donor 4 (entry 13) bears a β-Me group at C and is the only one to produce the a-anomer as major
2
product (60%). In this case, stereoelectronics predict that the b-anomer should be somewhat
favored because the β-bent bond is strengthened by hyperconjugation with the antiperiplanar α-
C
2
-H bond (which is assumed to be somewhat more important than the C
but there is no longer an EWG at C to weaken the antiperiplanar α-bent bond so the
stereoelectronic steering is expected to be less than for entries 4,5,7,8. The a-anomer should be
sterically favored by adding trans to the β-CH group but disfavored by this outside attack that
bond in the α-anomer. As a result, reaction of donor 4 with the weaker
TMSallyl nucleophile (entry 13) should proceed by a later transition state, so the a-isomer prevails
60%) but formation of the BBAH-favored b-isomer is still important despite the strong steric
repulsion. With TFE in CH Cl (entry 14), the major a-isomer is increased to 86% while in CH CN
entry 15), it is raised to 92%. These two results indicate that increasing the nucleophile reactivity
2
-CH
3
hyperconjugation)
2
3
2
3
9
d
eclipses the C
2
-C
3
(
2
2
3
(
leads to an earlier transition state where the steric aspects prevails, yet not fully, over the
hyperconjugation. Again, the percentage of the a-isomer increases with a more polar solvent and
stabilizing solvent.
In the case of donor 5, the b-anomer is strongly favored by both the inside attack and the trans
addition and is only disfavored by the weak C -H hyperconjugation. Conversely, the a-anomer is
2
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 54
1
8/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
strongly disfavored by the 1,2-cis addition and by outside attack which eclipses the C
2
-C
. It is only
-H hyperconjugation. As a result, the b-anomer completely dominates with
3
bond,
including a severe interaction between the α-CH
3
group and the ring CH
2
group at C
8
favored by the C
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TMSallyl (entry 16) and is over 90% b-selective with TFE in both solvent (entries 17-18). Again,
there is a slight decrease of the β-isomer with the stronger nucleophile TFE, as the earlier transition
state is less sensitive to the steric environment. Thus, the BBAH can rationalize correctly all these
results at the molecular level using classic organic chemistry principles.
Interestingly, it was also noticed that the electron-donating effect of the CH group can be clearly
3
perceived through the reaction conditions of donors 4 and 5, where the glycosylation reaction could
be kept at -40 °C for the entire transformation without having to increase the temperature, which
was required for donors 2 and 3 bearing an OCH
3
EWG group. Consequently, the oxocarbenium
hyperconjugation. This
ion from donors 4 and 5 would be more stable because of the CH
3
observation led to a realization that the different stereoselectivities observed for donors 2-5 and
those of sugar-furanosides described in Schemes 1, 3-5 could be explained by the different
reactivity of their oxocarbenium ions. In carbohydrates, all the OR groups should destabilize their
corresponding oxocarbenium ions through electron-withdrawing inductive effects, rendering them
more reactive, and resulting in earlier transition states for nucleophilic addition. Thus, the favored
1
,2-cis addition described in Schemes 1,3-5 would be the result of an inside attack and the
hyperconjugation effect of the OCH group. In bicyclic donor models 1-5, the oxocarbenium ion
should be more stable, less reactive, resulting in later transition states for nucleophilic addition.
Indeed, for donors 1-5, both the C -C and C -C bonds are antiperiplanar to the C -O bond so
3
5
6
3
8
4
the oxocarbenium ion should be more stable and less reactive. This hyperconjugation effect can
be expressed by the resonance structures indicated in Scheme 14.32
ACS Paragon Plus Environment

Page 19 of 54
The Journal of Organic Chemistry
1
9/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 14. Resonance structures of a carbobicyclic oxocarbenium ion.
As indicated in entries 2 and 3 (Scheme 13 and Table 1), glycosylation of donor 1 with TFE occurs
even at a steady -40°C, which clearly indicates that the oxocarbenium ion intermediate is easily
formed and must be relatively stable. This result is further confirmed by the observation that, on
treatment of 3,5-O-di-tert-butylsilyl arabinose (Scheme 4) with our glycosylation conditions (NIS,
CH
3
CN, no Lewis acid), no reaction was observed even at 0°C because the oxocarbenium ion is
likely not produced. This highly oxygenated donor was synthesized using reported protocols by
1
3
34
Crich and Zhu.
Thus, the glycosylation of donors 1-5, compared to that of carbohydrate-derived donors, must
occur through later transition states because of their increased oxocarbenium ion stability.
Conversely, the glycosylation of carbohydrate pentafuranoside derivatives must occur through
earlier transition states because of the inductive effect of several neighbouring oxygens. As a result,
the known 1,2-cis addition observed in the glycosylation of the 4,6-O-benzylidene derivatives of
glucose and mannose must occur through earlier transition states in which the stereoselectivity is
essentially controlled by the orientation of the C
2
–OR group. Supporting the above analysis,
9
f
Woerpel and co-workers reported that, in the postulated 7-membered bicyclic furanoside
oxocarbenium intermediates, the α-outside attack is weak (8%) when X=CH
important (40%) when X=O (Scheme 15).
2
and become
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 54
2
0/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 15. Remote oxygen effect on selectivity.
The electron-withdrawing effect of benzylidene groups on oxocarbenium ion reactivity was
modeled by DFT calculations. B3LYP/6-31G* calculations were performed on (3aS,7aR)-1,4,6-
trioxahexahydroindene A (i.e. model of benzylidene-fused furanosyl oxocarbenium ion) and
(
3aS,7aR)-1-oxaoctahydroindene B (i.e. cyclohexane-fused furanosyl oxocarbenium ion) from
3
5
which Merz-Kollman atomic partial charges were derived (Figure 1). Partial charges on the
oxocarbenium oxygen were found to be -0.217 for the benzylidene model A and -0.285 for the
cyclohexane model B, supporting the remote electron-withdrawing effect of the benzylidene
oxygens on the oxocarbenium oxygen.
Figure 1. Merz-Kollman charges for oxocarbenium ion models from on B3LYP/6-31-G*
calculation.
Finally, entries 6 and 11 of Table 1 describe the glycosylation using conditions reported by Crich13
in which AgOTf was added in catalytic amount to the standard conditions, under the assumption
that catalytic triflate might provide insight into the nature of the reaction intermediate. To our
surprise, selectivity was inverted in entry 6 for donor 2 (98α:2β vs 34α:66β initially), whereas
ACS Paragon Plus Environment

Page 21 of 54
The Journal of Organic Chemistry
2
1/54
1
2
3
4
5
6
7
8
9
donor 3 produced a greater β-selective glycosylation (entry 11). The transformations occur rapidly
even when maintained at -78°C. One can argue that the triflate may preferentially form a α or β
covalent bond at the anomeric center depending on the orientation at C
2
. Considering the weak
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
nucleophilic nature of TFE, the covalent C –OTf bond would have to be highly labile. Thus,
1
formation of a covalent triflate intermediate could follow the bent bond approach by addition to
2-like displacement with TFE as illustrated in Scheme 16. 36
the weaker τ bond followed by S
N
A
contact ion pair can also be considered, which would stabilize the oxocarbenium ion. The presence
of a Lewis acid is also known to activate the NIS, which catalyses the reaction. This would explain
the high reactivity observed at lower temperature but not the inversion of selectivity.37
Scheme 16. Glycosylation in the presence of AgOTf/NIS with TFE.
CONCLUSION
This paper reports the C- and O-glycosylation of bicyclic furanosides 1-5 with allyltrimethylsilane
and trifluoroethanol, respectively. These reactions were carried out to test whether the bent bond /
antiperiplanar hypothesis (BBAH) orbital model could rationalize the stereoselectivity of
glycosylation reactions. The analysis expands Woerpel's "inside/outside attack" reaction model by
incorporating the stereoelectronic influence of electron-withdrawing (i.e. OCH ) and electron-
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 54
2
2/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
donating (i.e. H or CH
of the oxocarbenium ion intermediates.
Interestingly, the glycosylation stereoselectivities on bicyclic oxocarbenium glycosyl donors 1-5
Table 1) turned out to be quite different than those reported for five-membered carbohydrates
3
) groups at C
2
on the relative nucleophilic displacement of the two τ bonds
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
containing multiple OR groups as shown in Schemes 1, 3-5. The glycosylation stereoselectivities
for sugar-derived furanosides are rationalized by the Woerpel model with the BBAH-based
premise that nucleophilic addition to the oxocarbenium ion intermediates proceeds by early
transition states, and that the generally preferred 1,2-cis stereoselectivity is due largely to the
orientation of the C
In the case of the bicyclic donors 1-5, the glycosylation outcomes are different because the
oxocarbenium ion intermediates are stabilized by opposite hyperconjugative effects involving CH
groups at C and C . As a result, they form more easily at lower temperatures, whereas their
2
–OR group relative to the τ bonds of the oxocarbenium ion intermediates.
2
5
8
carbohydrate furanoside counterparts were not reactive at the same temperatures under our
glycosylation reaction conditions. Thus, nucleophilic addition to the stabilized oxocarbenium ions
1
-5, must proceed through later transition states than for their carbohydrate counterparts.
Consequently, the conformationally preferred inside attack on donors 1-3 and 5 becomes more
important, favoring the β-glycosylation. The electron-donating CH group at C plays both a steric
3
2
and a stereoelectronic role (stabilizing the adjacent antiperiplanar τ bond). Thus, the α-
glycosylation in bicyclic model 4 is now favored.
EXPERIMENTAL SECTION
General information
ACS Paragon Plus Environment

Page 23 of 54
The Journal of Organic Chemistry
2
3/54
1
2
3
4
5
6
7
8
9
The following section includes the spectroscopic data of newly synthesized compounds.
Transformation requiring inert and anhydrous conditions were carried out within an argon
atmosphere with dry solvents and oven dried/flame dried glassware. Dry DCM, Et
THF was obtained from a Vacuum Atmosphere Inc; Solvent Purification System. CH
distilled over CaH . All commercially available reagents were used as received without further
2
O, toluene and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
CN was
2
purification. Thin-layer chromatography analysis of reaction mixtures were performed using
Silicycle’s Silica alumina backed TLC plates and analysed using UV light and/or by staining with
either potassium permanganate, cerium ammonium molybdate (CAM) or p-anysaldehyde. Flash
column chromatography purification was carried out on Silicycle silica gel 60 Å, 230–400 mesh.
1
13
19
NMR spectroscopic data for H, C, F and NOESY 1D were recorded at ambient temperature in
neutralized CDCl using an Agilent DD2 500, Bruker 300 MHz and Varian Inova 400
3
spectrometers. High-resolution mass spectra were obtained on a LC/MS–TOF Agilent 6210 using
electrospray ionization (ESI). Infrared spectra were obtained using an Aminco Bowmen Adrid
Zone infrared spectrometer with a NaCl crystal matrix. Melting point values were determined
using a Stanford Research System OptiMelt capillary melting point apparatus and are uncorrected.
C-Glycosylation: general procedure A
A solution of an acetate donor in DCM (0.1 M) under argon was brought to −78 °C and
allyltrimethylsilane (4 equiv) was added. The mixture was then treated with BF
3
·Et
2
O (1.2 equiv)
. The
and brought to ambient temperature over 2 h before quenching with saturated NaHCO
3
organic phase was separated and the aqueous one was washed with DCM (3×), dried over MgSO
4
1
and condensed in vacuo. Crude mixtures were analyzed by H NMR spectroscopy and purified as
described.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 54
2
4/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O-Glycosylation: general procedure B
A solution of a thiophenylacetal donor in dry DCM or CH
3
CN (0.1 M) and 4Å MS with
trifluoroethanol (4 equiv) under argon was brought to −40 °C. The mixture was then treated with
NIS (2 equiv) and brought to the required temperature over 1.5 h when needed before quenching
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with a saturated aqueous solution of Na
2
S
2
O
3
. Using CH
3
CN as the solvent requires flame dried
material and dropwise addition of a NIS/CH
3
CN solution in order to maintain an inert atmosphere.
NIS was added directly as a powder when DCM is the solvent. The organic phase was separated,
and the aqueous phase was washed with DCM (3×). Organic fractions were combined, dried over
MgSO
4
and condensed in vacuo. Special care was is to be taken for the removal of solvent with
1
those compounds as they appeared to be slightly volatile. Crude mixtures were analyzed by H
NMR spectroscopy and purified as described. To be noted that the characterization of O-
glycosylated products by any HRMS method available to us would often only provide the
corresponding oxocarbenium ion after the loss of the trifluoroethyl moiety.
tert-Butyldimethyl((2-vinylcyclohexyl)oxy)silane) (6): To the allyl alcohol (3.5 g, 27.7 mmol)
in dry DCM (30 mL) under argon was added imidazole (3.77 g, 55.5 mmol) followed by TBSCl
(
5.02 g, 33.3 mmol). The mixture was stirred overnight at ambient temperature. Silica was added
and the mixture condensed to a powdered blend. The desired compound was recovered by filtration
thought a fritted glass funnel filled with 10 cm of silica gel using hexanes to yield after evaporation
1
6.27 g (94 %) of 6 as colourless oil; H NMR (500 MHz, CDCl
3
) δ 5.82 (ddd, J = 17.6, 10.4, 7.4
Hz, 1H), 5.04 – 4.96 (m, 2H), 3.28 (td, J = 9.6, 4.2 Hz, 1H), 2.01 – 1.92 (m, 1H), 1.91 – 1.85 (m,
H), 1.78 – 1.70 (m, 2H), 1.67 – 1.61 (m, 1H), 1.37 – 1.13 (m, 4H), 0.88 (s, 8H), 0.04 (s, 3H), 0.03
1
1
3
1
(s, 3H); C{ H} NMR (101 MHz, CDCl ) δ 142.1, 113.9, 74.9, 49.8, 35.8, 30.5, 25.9, 25.0, 24.8,
3
ACS Paragon Plus Environment

Page 25 of 54
The Journal of Organic Chemistry
2
5/54
1
2
3
4
5
6
7
8
9
+
1
2
8.2, -0.02, -4.2, -4.5; HRMS (ESI-TOF) M/Z: [M+K] Calcd for C14
H
28OSiK 279.1546; Found;
79.1532.
2
-(2-Hydroxyethyl)cyclohexan-1-ol (7): To a solution of OTBS protected olefin 6 (1.3 g, 5.4
–Me S solution
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
mmol) in dry THF (10 mL) at 0 °C under an argon atmosphere was added 2 M BH
3
2
in THF (10.81 mmol, 5.41 mL) over 10 min. The reaction was then brought slowly to ambient
temperature and stirred for 18 h. The mixture was treated with 3 M NaOH at 0 °C until the pH was
basic and excess 30% H
diluted with water and DCM. The organic layer was separated, and the aqueous layer washed with
DCM (3×). The organic layers were combined, dried over MgSO and condensed in vacuo. The
2
O
2
was added. The reaction was stirred for 3 h until completion and was
4
crude residue was rapidly eluted on a silica gel pad with 50:50 EtOAc/hexanes, condensed in
vaccuo then diluted in dry THF (5 ml). 1 M TBAF solution in THF (5.4 mL, 5.4 mmol) was then
added. The reaction was stirred 1 h at ambient temperature and quenched with saturated NH
The reaction was diluted with DCM, the phases separated, and the aqueous portion washed with
DCM (3×), dried over MgSO , filtered and condensed in vacuo. The crude mixture was purified
4
Cl(aq).
4
by silica gel column chromatography (50% EtOAc/hexanes, TLC stained with p-anisaldehyde) to
1
obtain 0.430 g (50 % over two steps) of the diol 7 as a colourless thick oil; H NMR (400 MHz,
CDCl
3
) δ 4.21 (s, 1H), 3.75 (dt, J = 10.6, 4.7 Hz, 1H), 3.59 (ddd, J = 10.7, 9.5, 3.5 Hz, 1H), 3.19
(td, J = 9.8, 4.4 Hz, 1H), 1.99 – 1.91 (m, 1H), 1.77 – 1.64 (m, 3H), 1.64 – 1.56 (m, 1H), 1.51 (dtd,
1
3
1
J = 14.7, 4.8, 3.6 Hz, 1H), 1.36 – 1.09 (m, 4H), 1.07 – 0.97 (m, 1H); C{ H} NMR (126 MHz,
CDCl
3
) δ, 74.9, 61.4, 44.4, 44.37, 38.1, 35.5, 32.7, 25.6, 24.9; HRMS (ESI-TOF) M/Z:
+
[
M+H] Calcd for C
8
H
17
2
O 145.1223; Found 145.1212.
Hexahydrobenzofuran-2(3H)-one (8): To a solution of diol 7 (0.219 g, 1.51 mmol) in dry DCM
(3 mL) under argon was added [bis(acetoxy)iodo]benzene (1.46 g, 4.53 mmol) followed by
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 54
2
6/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TEMPO (0.070 g, 0.453 mmol). The reaction was stirred for 3 h at ambient temperature and
quenched with saturated aqueous Na . The organic layer was isolated and washed with DCM
(3×). Combined phases were dried over MgSO , filtered and condensed under reduce pressure.
2
S
2
O
3
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The crude mixture was purified by flash chromatography (20% EtOAc/hexanes, TLC stained with
1
CAM) to yield 0.180 g (85%) of a colourless oil; H NMR (500 MHz, CDCl
3
) δ 3.78 (td, J = 10.9,
3.8 Hz, 1H), 2.50 (ddd, J = 16.1, 6.8, 2.2 Hz, 1H), 2.22 (tt, J = 13.1, 4.8 Hz, 2H), 2.02 – 1.86 (m,
1
3
1
3
H), 1.82 – 1.76 (m, 1H), 1.53 (qd, J = 11.8, 3.4 Hz, 1H), 1.46 – 1.24 (m, 3H); C{ H} NMR (126
MHz, CDCl ) δ 176.6, 85.2, 44.8, 35.9, 30.2, 28.3, 25.3, 24.1; IR (NaCl) νmax 2939, 2865, 1781,
447, 1189, 1030, 931 cm ; HRMS (ESI-TOF) M/Z: [M+H] Calcd for C
141.0905.
3
-
1
+
1
8
H
13
O 141.0910; Found
2
Octahydrobenzofuran-2-yl acetate (1(OAc)): To a −78 °C solution of lactone 8 (0.900 g, 6.42
mmol) in dry toluene (20 mL) under an argon atmosphere was added dropwise 1 M DIBAL-H in
heptanes (7.70 mL, 7.70 mmol). The reaction was stirred at −78 °C for 2 h. Once the reaction
completed by TLC monitoring, pyridine (0.620 mL, 7.70 mmol), was slowly added at −78 °C,
followed by DMAP (0.941 g, 7.70 mmol) in 6 mL of dry DCM, stirred for 10 min and Ac
mL, 32.1 mmol) was added dropwise. The reaction was allowed to reach ambient temperature and
stirred for 12 h. The mixture was quenched with saturated NH Cl and diluted with EtOAc. The
extracted organic phase was washed with 1 N NaHSO (2×), saturated NaHCO (2×), brine (1×),
dried over MgSO and condensed in vacuo. The residue was purified by silica gel column
chromatography (10% EtOAc/hexanes with 2% Et
2
O (3.03
4
4
3
4
3
N, TLC stained with CAM) to obtain a mixture
1
of two diastereoisomeric acetates 1(OAc) (0.778 g, 60α:40β, 66 %); H NMR (400 MHz, CDCl
3
)
δ 6.24 (dd, J = 6.0, 4.7 Hz, 0.6H), 6.19 (d, J = 5.2 Hz, 0.4H), 3.37 (td, J = 10.4, 3.8 Hz, 0.6H), 3.19
(
td, J = 10.7, 3.8 Hz, 0.4H), 2.42 (ddd, J = 13.0, 7.1, 6.1 Hz, 0.6H), 2.19 – 2.10 (m, 1H), 2.05 (s,
ACS Paragon Plus Environment

Page 27 of 54
The Journal of Organic Chemistry
2
7/54
1
2
3
4
5
6
7
8
9
1
1
.8H), 2.03 (s, 1.2H), 2.00 (dd, J = 12.5, 5.4 Hz, 1H), 1.96 – 1.88 (m, 1H), 1.87– 1.80 (m, 1H),
.79 – 1.66 (m, 2H), 1.57 (ddd, J = 12.8, 11.9, 4.7 Hz, 1H), 1.47 – 1.34 (m, 1.4H), 1.33 – 1.02 (m,
1
3
1
3H); C{ H} NMR (101 MHz, CDCl
3
) δ 170.6, 170.4, 98.2, 97.5, 85.4, 82.9, 45.2, 41.9, 37.9,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
7.8, 31.5, 30.5, 28.5, 28.4, 25.6, 25.5, 24.2, 24.1, 21.4, 21.3.; IR (NaCl) νmax 2936, 2860, 1749,
-
1
+
1450, 1376, 1234, 1036, 963, 864 cm ; HRMS (ESI-TOF) M/Z: [M+Na] Calcd for C10
H
16
3
O Na:
2
07.0991; Found: 207.0992.
2
1
-(Phenylthio)octahydrobenzofuran (1(SPh)): To a solution of diastereoisomeric acetate
(OAc) (0.778. g, 4.22 mmol) in dry DCM (20 mL) under an argon atmosphere was added PhSH
(1.72 mL, 16.9 mmol). The mixture was cooled to −78 °C and BF
was slowly added. After 1 h of stirring at −78 °C, the reaction was quenched with saturated aqueous
NaHCO , brought to ambient temperature and diluted with DCM. The organic phase was isolated
and the aqueous one was washed with DCM (3×). The organic phases were combined, dried over
MgSO and condensed in vacuo. The residue was purified by silica gel column chromatography
2% EtOAc/hexanes to 10% EtOAc/hexanes, TLC revealed by UV light and stained with CAM)
3
·Et
2
O (0.570 mL, 4.64 mmol)
3
4
(
1
to give two diastereoisomeric thioacetals 1(SPh) (0.500 g, 50 %, 15α:85β) as colourless oils; H
NMR (300 MHz, CDCl ) δ 7.56 – 7.49 (m, 2H), 7.35 – 7.16 (m, 3H), 5.63 (t, J = 7.2 Hz, 0.85H),
.50 (dd, J = 7.8, 2.7 Hz, 0.15H), 3.39 (td, J = 10.2, 3.8 Hz, 0.85H), 3.17 (td, J = 10.6, 3.8 Hz,
.15H), 2.55 (dt, J = 12.2, 6.8 Hz, 1H), 2.20 – 2.10 (m, 1H), 1.98 – 1.89 (m, 1H), 1.88 – 1.79 (m,
3
5
0
1
3
1
1H), 1.73 – 1.67 (m, 1H), 1.53 (td, J = 11.9, 7.3 Hz, 1H), 1.45 – 1.03 (m, 5H); C{ H} NMR (75
MHz, CDCl
3
) δ 131.5, 130.2, 128.7, 126.9, 126.5, 85.6, 84.8, 81.4, 45.1, 44.2, 38.9, 38.5, 31.5,
+
3
2
0.6, 28.7, 25.6, 24.2; HRMS (ESI-TOF) M/Z: [M+H] Calcd for C14
H19OS: 235.1151; Found
35.1148.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 54
2
8/54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
-(2-((tert-Butyldimethylsilyl)oxy)cyclohexyl)ethane-1,2-diol (9): To a solution of 6 (6.27 g,
2
6.1 mmol) in THF (36 mL) and water (12 mL) at 0 °C was added N-methylmorpholine oxide
(6.41 g, 54.6 mmol) followed by catalytic OsO
8 h at ambient temperature. It was then diluted with water and EtOAc. The phases were separated,
and the aqueous mixture was washed with EtOAc (5×), dried over MgSO and concentrated in
4
(4% in H
2
O, 2 mL). The reaction was stirred for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
vacuo. The residue was purified by silica gel column chromatography (35% EtOAc/hexanes, TLC
stained with CAM) to give two diastereoisomeric diols 9 (6.9 g, 97%) isolated as a thick yellowish
1
oil (ratio ~2:1); H NMR (400 MHz, CDCl
3
) δ 3.69 – 3.53 (m, 4H), 1.98 – 1.87 (m, 1H), 1.80 –
1
.51 (m, 5H), 1.40 – 1.10 (m, 5H), 0.91 (s, 5.5H), 0.90 (s, 3.5H), 0.14 (d, J = 8.2 Hz, 3.5H), 0.11
1
3
1
(d, J = 4.1 Hz, 2.5H); C{ H} NMR (101 MHz, CDCl
3
) δ 78.6, 76.0, 73.6, 73.4, 64.7, 63.9, 47.6,
4
6.8, 36.4, 36.2, 26.6, 26.2, 25.87, 25.82, 25.78, 25.45, 25.11, 24.7, 24.6, 17.95, 17.87, -3.4, -3.6,
+
-4.69, -4.73; HRMS (ESI-TOF) M/Z [M+H] : Calcd for C14
H
31
O Si 275.2037; Found 275.2013.
3
Allyl 2-(2-((tert-butyldimethylsilyl)oxy)cyclohexyl)-2-hydroxyacetate (10): To a solution of
diol 9 (4.0 g, 14.6 mmol) in DCM (69 mL) and saturated NaHCO3(aq) (17 mL) was added TEMPO
(
0.09 g, 0.583 mmol) and KBr (1.91 g, 16.0 mmol). The reaction was stirred rapidly while a 0.435
M solution of NaOCl (36.8 mL, 16.0 mmol) was added at 0 °C. The dark-red reaction was stirred
for 1 h until its colour faded to a pale orange. Once completion reached, a Na aqueous
solution was added and the phases were separated and the aqueous one was washed with DCM
3×). The combine organic phases were dried over MgSO and condensed in vacuo. The crude
aldehyde was directly submitted to the next step. The intermediary aldehyde was put in solution
in t-BuOH (120 mL) and water (40 mL), NaH PO (3.49 g, 29.1 mmol). Once the salt completely
dissolved, the reaction was brought to 0 °C and 2-methyl-2-butene 2 M solution in THF (14.6 mL)
was added to the mixture followed by solid NaOCl pellets (80 % w/w, 3.29 g, 29.1 mmol) The
2
S
2
O
3
(
4
2
4
2
ACS Paragon Plus Environment

Page 29 of 54
The Journal of Organic Chemistry
2
9/54
1
2
3
4
5
6
7
8
9
reaction was stirred at 0 °C for 1 h. Following completion, the mixture was slowly acidified to pH
using 1 N HCl. The mixture was extracted using DCM (4×), dried with MgSO , filtered and
4
4
condensed under reduced pressure. The crude product was dissolved in minimal amount of DCM
and quickly purified through an acid/base workup using 1 N NaOH and 1 N HCl. After
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
reacidification, the desired acid was extracted using DCM (5×), dried over MgSO
The resulting thick yellow oil was put in solution in DMF (25 mL) followed by the addition of
CO (3.10 g, 22.49 mmol), tetrabutylammonium iodide (1.66 g, 4.49 mmol) and allyl bromide
1.94 mL, 22.49 mmol) dropwise. The reaction was stirred for 2 h at ambient temperature and then
diluted in water and Et O. The aqueous phase was washed with Et O (3×). The combined organic
phases were washed with 1 N HCl (1×) and saturated NaHCO (1×). The organic phase was dried
over MgSO and condensed in vacuo. The resulting crude was purified by silica gel column
4
and condensed.
K
2
3
(
2
2
3
4
chromatography (5% EtOAc/hexanes, TLC stained with CAM) to yield compound 10 (3.72 g, 77
1
% over three steps) as a yellowish oily mixture of the two diastereoisomers (2.3:1); H NMR (400
MHz, CDCl
.77H), 3.68 (ddd, J = 9.7, 3.5 Hz, 0.77H), 3.60 (ddd, J = 9.9, 3.8 Hz, 0.33H), 3.25 (d, J = 4.1 Hz,
0.77H), 2.73 (d, J = 5.5 Hz, 0.33H), 1.97 – 1.80 (m, 2H), 1.78 – 1.59 (m, 3H), 1.35 – 1.00 (m, 4H),
3
) δ 6.01 – 5.81 (m, 1H), 5.43 – 5.17 (m, 2H), 4.77 – 4.56 (m, 3H), 4.27 (q, J = 3.4 Hz,
0
1
3
1
0
.88 (s, 3H), 0.86 (s, 6H), 0.08, (s, 1H), 0.07, (s, 1H), 0.05 (s, 4H); C{ H} NMR (126 MHz,
CDCl ) δ 1757, 174.1, 131.7, 131.6, 119.1, 118.8, 69.8, 66.0, 65.9, 49.5, 48.7, 36.2, 36.1, 27.5,
6.0, 25.9, 25.4, 25.2, 24.8, 24.7, 23.8, 18.04, 18.03, -4.00, -4.1, -4.3, -5.0; IR (NaCl) νmax 3529,
3
2
-
1
+
2931, 2857, 1733, 1250, 1086 cm ; HRMS (ESI-TOF) M/Z [M+H] : Calcd for C17
H
33
4
O Si
3
29.2142; Found 329.2141.
Allyl 2-(2-((tert-butyldimethylsilyl)oxy)cyclohexyl)-2-methoxyacetate (11/11’): To a solution
of 10 (3.12 g, 9.48 mmol) in dry THF (11 mL) was added at 0 °C under argon NaH (60 % in
ACS Paragon Plus Environment