2,3-Carbamate mannosamine glycosyl donors in glycosylation reactions of diacetone-D-glucose. An experimental and theoretical study

Article abstract of DOI:10.1016/j.carres.2021.108421

The role of the cyclic 2,3-N,O-carbamate protecting group in directing the selectivity of mannosylation reactions of diacetone-D-glucose, promoted by BSP/Tf₂O via α-triflate intermediates, has been investigated through a combined computational

Full text of DOI:10.1016/j.carres.2021.108421

Carbohydrate Research 509 (2021) 108421
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
2,3-Carbamate mannosamine glycosyl donors in glycosylation reactions of
diacetone-D-glucose. An experimental and theoretical study
Laura Morellia, Laura Legnanib, Silvia Ronchia, Laura Confalonieric, Daniela Imperioc,
,
,**
Lucio Tomab *, Federica Compostellaa
a
`
Dipartimento di Biotecnologie Mediche e Medicina Traslazionale, Universita degli Studi di Milano, Via Saldini 50, 20133 Milano, Italy
b
c
`
Dipartimento di Chimica, Universita di Pavia, Via Taramelli 12, 27100 Pavia, Italy
`
Dipartimento di Scienze del Farmaco, Universita del Piemonte Orientale, L.go Donegani 2, 28100 Novara, Italy
A R T I C L E I N F O
A B S T R A C T
Keywords:
The role of the cyclic 2,3-N,O-carbamate protecting group in directing the selectivity of mannosylation reactions
Carbohydrates
Glycosylation
1,2-Cis glycosides
β-mannosides
DFT calculations
of diacetone-D-glucose, promoted by BSP/Tf2O via
α-triflate intermediates, has been investigated through a
combined computational and experimental approach. DFT calculations were used to locate the transition states
leading to the or β anomers. These data indicate the preferential formation of the βꢀ adduct with mannosyl
donors either equipped with the 4,6-O-benzylidene protection or without it. The synthetic results confirmed this
preference, showing in both cases an /β selectivity of 4:6. This highlights a role for the 2,3-N,O-carbamate in
sharp contrast with what described in the case of 2,3-O-carbonate mannosyl donors.
α
α
1. Introduction
stereochemical course of the glycosylation is well known. For example,
the 4,6-O-benzylidene acetal, well explored by Crich, allows the
obtainment of βꢀ mannosides with noteworthy selectivity [11–15]. This
effect is completely reversed by the introduction of an additional cyclic
Glycosylation is still the most critical reaction in carbohydrate
chemistry. In spite of significant advances, there is not a universal
protocol that allows the formation of all types of glycosidic bonds, the
main problem being stereoselectivity [1,2]. The stereochemistry of a
glycosylation reaction is influenced by a large number of different fac-
tors, which can make it in some cases hard to predict and difficult to
control, especially when the synthetic targets are 1,2-cis glycosides [3].
An accurate design of the experimental conditions and of the structural
elements of the donor allow a certain control on the mechanism of these
reactions, which can range from a SN2-like to a SN1-like process, so
influencing in some respect the stereochemistry of the products [4–7].
Unfortunately, a complete control is far from reach. Similarly, several
concerns affect the computational approaches aimed to predict the
selectivity of the reaction or to rationalize the stereochemistry observed
experimentally [8,9].
protecting group such as 2,3-O-carbonate, which induces αꢀ selectivity
[14,16]. On the other hand, when the 3,4-O-carbonate cyclic protecting
group is used, a moderate βꢀ selectivity is again observed [17]. The ef-
fect of cyclic protecting groups in donors has been less studied in the
synthesis of β-mannosamine glycosides, where the more reliable proto-
col is still the indirect approach through β-glucosylation followed by
gluco to manno epimerization [18]. There is evidence that lower selec-
tivities are observed in glycosylation of 2-azidomannosyl donors under
Crich conditions, despite the presence of the benzylidene protecting
group [19,20]. Advances in β-mannosylation cannot be generalized to
mannosamine, and further studies are required to find new protocols
[21–23].
In this framework, we were interested in the possible role of the 2,3-
carbamate protecting group in directing the selectivity of the glycosyl-
ation of a mannosamine donor. Though an effect similar to that of 2,3-
Several well-established glycosylation procedures transform the
original glycosyl donor into the corresponding anomeric triflate inter-
mediate before the addition to the acceptor nucleophile. This strategy is
largely used for β-mannoside synthesis [10]. Focusing on the donor
structure, the strong influence of protecting groups on the
carbonate might be expected, with a preference for the
α product, we
hypothesized that differences in the reactants may translate into
different stereochemical outcomes, leading to the less easily accessible β
* Corresponding author.
** Corresponding author.
E-mail addresses: lucio.toma@unipv.it (L. Toma), federica.compostella@unimi.it (F. Compostella).
https://doi.org/10.1016/j.carres.2021.108421
Received 5 July 2021; Received in revised form 5 August 2021; Accepted 18 August 2021
Available online 21 August 2021
0008-6215/© 2021 Elsevier Ltd. All rights reserved.

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
anomer. In this paper, we have studied the selectivity of the glycosyla-
tions with two 2,3-carbamate protected mannoside donors, phenylthio
their main geometrical features. There is a small difference in the Gibbs
free energy of the two structures, 0.67 kcal/mol in favor of TS-4b, which
4,6-O-benzylidene-2,3-carbamate-
α-D-mannopyranoside
1
and its
corresponds to an α/β product ratio of 24:76 at room temperature and
analogue 2 lacking the benzylidene protecting group on positions 4 and
6, respectively protected as acetate and benzyl ether (Fig. 1). The
acceptor was diacetone-D-glucose 3, a known glucosyl acceptor with a
secondary unprotected hydroxyl group in position 3, commonly used to
test glycosylation reactions. Before starting the experimental work, we
have addressed the effect of the presence of the carbamate on the ster-
eoselectivity of the glycosylations with a computational approach, based
on the same theoretical model already used by Crich et al. to locate the
transition state structures for glycosylations performed through the
intermediacy of mannopyranosyl and glucopyranosyl triflates [8]. The
computational study was performed on compounds 4 and 5, corre-
17:83 at ꢀ 60 ◦C.
A higher selectivity was predicted in the case of the reaction of the
bicyclic glycosyl donor 5 with 3 which can result either in the
α anomer
7a or in the β anomer 7b of the disaccharide product. The two lowest
energy structures, TS-5a and TS-5b (Fig. 2) show a Gibbs free energy
difference higher than 2 kcal/mol with the latter one being favored. This
predicts a very selective reaction with an α/β product ratio of 3:97 at
room temperature and 1:99 at ꢀ 60 ◦C.
The pyranose ring of the mannosamine donors adopts the same ge-
ometry in all the four TSs, which is similar to the B2,5 conformation
found by Crich in the TS for the reaction of isopropanol with 4,6-O-
benzylidene mannosyl triflate leading to the β-glycoside product [8].
However, the presence of the additional ring determined by the 2,
sponding to the α-triflate intermediates of simplified forms of donors 1
and 2, better suited for computing (Scheme 1).
1
The results suggest a preference for the β adduct in the case of 4 and a
net preference for the same anomer in the case of 5. On this basis, we
performed the synthetic work, to validate the computational results.
Herein, we report all the theoretical and experimental investigations
that highlight the role of the 2,3-carbamate protecting group of man-
nosamine glycosyl donors in determining the outcome of the glycosyl-
ation reaction.
3-carbamate protection makes the pyranose ring closer to a S5 skew
rather than to a B2,5 boat conformation. Moreover, the four TSs are
highly unsymmetrical as the bond of the anomeric carbon to the
incoming nucleophile is much shorter than that of the departing triflate
(Fig. 2). This suggests that they are late transition states where the bond
of the anomeric carbon with the triflate is almost completely broken
before the formation of the new bond with the nucleophile. When we
looked for geometries showing the alternative 4H3 conformation of the
2. Results and discussion
pyranose ring, described by Crich for the TS leading to the α-mannoside
product [8], we were unable to locate them as, during the optimizations,
the ring geometries always changed into the more stable 1S5
conformation.
2.1. Computational studies
Density functional theory calculations of possible transition state
structures for the reactions of glycosyl donors 4 and 5 and glycosyl
acceptor 3 were carried out in dichloromethane by optimizations with
the B3LYP functional at the 6-31G(d,p) level. The solvent effect was
taken into account by using a self-consistent reaction field (SCRF)
method, based on the polarisable continuum model (PCM), choosing
dichloromethane as the solvent. With the optimized geometries, single-
point energy calculations were performed using the B3LYP/6–311++G
(3df, 3pd) level still with the same solvent model. To avoid too long
computational times, calculations were performed with the simplified
models 4 and 5, instead of 1 and 2, respectively, in which the benzyl is
replaced by methyl groups and benzylidene by methylidene (Scheme 1)
[16].
2.2. Chemistry
The synthesis of mannosamine donor 1 (Scheme 2) started with the
peracetylation of N-acetyl-D-mannosamine, followed by glycosylation
with thiophenol, mediated by boron trifluoride diethyl etherate, to
obtain the corresponding
α
-phenylthio derivative 8 in 65% yield over
´
two steps. Deacetylation under Zemplen conditions of compound 8,
followed by 4,6-O-benzylidene protection gave 9. Then, the acetamido
group of 9 was hydrolyzed under basic conditions to allow the subse-
quent formation of the 2,3-carbamate by reaction with p-nitrophenyl
chloroformate and sodium hydrogen carbonate at reflux temperature.
The tricyclic compound 10 was obtained smoothly in 70% yield over 4
steps. Intermediate 10 was finally transformed into donor 1 through a
high yielding N-benzylation under standard conditions. The benzyl
group was introduced onto the carbamate to reduce side-reactions at the
carbamate nitrogen during glycosylation [24].
The reaction of the tricyclic glycosyl donor 4 with 3 was first
investigated and several transition state structures were located which
can give rise either to the
α anomer 6a or to the β anomer 6b of the
disaccharide product. The three-dimensional plots of the two lowest
energy structures, TS-4a and TS-4b, are reported in Fig. 2 together with
Mannosamine donor 2 was obtained starting from donor 1 through
the regioselective reductive opening of the benzylidene acetal with
triethylsilane/boron trifluoride-diethyl ether system to give compound
11 in 81% yield. Lastly, acetylation of the resultant free hydroxyl in
position 4 of 11 afforded donor 2 in 96% yield.
The overall procedure for the obtainment of the desired 2,3-carba-
mate mannosamine donors 1 and 2 is efficient and applicable to gram
scale synthesis.
Based on the predictions obtained from DFT calculations, we started
with the glycosylation between donor 2 and acceptor 3.
The 1-benzenesulfinyl piperidine (BSP)/triflic anhydride (Tf2O)
promotor system was initially evaluated. This system is known to readily
activate armed and disarmed thioglycosides via glycosyl triflates at
ꢀ 60 ◦C in the presence of 2,4,6-tri-tert-butylpyrimidine (TTBP) (Scheme
◦
3) [14,19,25]. We performed the reaction at ꢀ 40 C, due to the low
solubility of acceptor 3 at lower temperatures, pre-activating donor 2
before the addition of acceptor 3. Unfortunately, we observed pre-
dominantly the decomposition of the thiol donor into more polar com-
pounds, together with the formation of traceless amounts of the
glycosylation product 12, while most of acceptor 3 remained unaltered
(Table 1 –entry 1). We then turned to the NIS/AgOTf thiophilic activator
Fig. 1. Structures of the 2,3-carbamate protected mannosamine donors 1 and
2, and glucosyl acceptor 3.
2

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
Scheme 1. Computationally studied glycosylation reactions of diacetone-D-glucose 3 with model mannosamine triflates 4 and 5.
Fig. 2. Three-dimensional plots of the lowest energy transition states for the reaction of diacetone-D-glucose 3 with mannosamine triflates 4 and 5. For each
transition state the relative free energy of activation and the lengths of the partial bonds to the leaving group, r(C-OTf), and to the nucleophile glucosyl acceptor 3, r
(C-ONu), are listed.
system [26]. Donor 2 was activated in-situ with AgOTf/NIS at ꢀ 40 ◦C
and smoothly condensed with acceptor 3 to give disaccharide 12 in 70%
yields (Table 1 – entry 2). The glycosylation is highly reproducible, and
the donor consumption is indicated by the color of the reaction that
turns into a deep red. The analysis of the isolated products by 1H NMR
indicated the formation of a mixture of anomers 12a and 12b in a 4:6
ratio (Fig. 3A), in favor of the β-anomer. The NMR spectrum did not
allow an unequivocal attribution of the signals to the two anomers, as
the mannosidic anomeric protons were both broad singlets. The
configuration was tentatively established based on the assignments re-
ported in the literature for azido-mannosides [27] where the anomeric
proton of the alfa-anomer is in general downfield with respect to the
β-one. This was further validated with chemical correlation (see below).
Unfortunately, though the preferential formation of the β-anomer
predicted by theoretical calculations was confirmed by the experimental
results, the observed selectivity was much lower than what was
expected.
(Fig. 3C). This suggests that the former signal can be assigned to the
-anomer and the latter to the β-one. This was confirmed by computa-
α
tion of theoretical coupling constants of α- and β-model compounds,
reported in Fig. 4, for which J1,2 of 1.1 and 3.3 Hz, respectively, were
predicted (Fig. 4).
The stereochemical outcome of the glycosylation to disaccharide 13
mediated by BSP/Tf2O (α/β 40:60) is in good agreement with the
slightly more favourable data predicted by computing (α/β 24:76).
We wondered if the low selectivity was due to in-situ anomerization
in the reaction conditions, and stopped the reaction immediately after
the slow addition of the acceptor to the solution of the activated donor
(5′) or after 15’ (Table 1 – entry 5–6). However, we always found the
same 4:6 α/β anomeric ratio. The same when we performed the reaction
at ꢀ 60 ◦C: as expected, the yield was lower due to the poor solubility of
the acceptor at this temperature, but the anomeric ratio was unaffected
(Table 1 – entry 7).
Finally, we confirmed the configuration of the previously obtained
12a and 12b disaccharides by chemical correlation. After separation of
anomers 13a and 13b by flash chromatography and complete 1H NMR
characterization of the disaccharides, a pure amount of 13a was trans-
formed into 12a by reductive benzylidene ring-opening, followed by
acetylation of the 4′-OH. The 1H NMR of the crude product was
compared with the spectrum of the 12a/12b mixture, thus allowing to
confirm the previous assignments (Fig. 3B).
Thus, we turned to the glycosylations with the mannosamine tricy-
clic donor 1 (Scheme 3), which was initially coupled with acceptor 3
under the activation of NIS/AgOTf (Table 1 – entry 3). The glycosylation
to disaccharide 13 proceeded smoothly in 55% yield, but without
selectivity. So, we evaluated the BSP/Tf2O promotor system which gave
compound 13 in comparable yield, around 60%, but with a slight
preference for one anomer (Table 1 – entry 4). Although both very small,
the vicinal mannose coupling constants (J1’-2’) of the two anomers 13a
and 13b show a different coupling pattern, being one signal a broad
singlet at 5.35 ppm and the other one a doublet (J = 3.0 Hz) at 5.02 ppm
In summary, our study reports the effect of cyclic 2,3-N,O-carbamate
protected systems on the stereoselectivity of mannosylations promoted
by BSP/Tf2O via α-triflate intermediates. DFT calculations on model
3

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
benzylidene. This is in sharp contrast with what was described in the
case of 2,3-O-carbonate mannosyl donors, where the presence of the 2,3-
cyclic protecting group on 4,6-O-benzylidene-mannosides induced the
exclusive formation of α-mannosides [14].
In addition, the β-selectivities herein described are similar to the
ones reported in the case of 2-azidomannoside donors: the glycosylation
of 4,6-O-benzylidene-2-azido-mannoside with acceptor 3 gives the
disaccharide product in
α/β ratio of 1:2 [27]. Our data support the
notion that cyclic carbamate protected mannosaminyl donors can be
considered a viable alternative to azido-mannosides in glycosylation
strategies leading to mannosamine-containing oligosaccharides. Indeed,
different naturally occurring oligosaccharides contain the 1,2-cis-linked
β-mannosaminyl residue, such as for example, the bacterial capsular
polysaccharides of Staphylococcus aureus type 5 [28] or Streptococcus
pneumoniae 19F and 19A [18]. The two new mannosamine donors
developed in this study can be suitable building blocks for the synthesis
of repeating unit oligomers of such bacterial species. In fact, besides the
stereodirecting effect of the carbamate protecting group, mannosamine
building blocks 1 and 2 can be regioselectively deprotected at C-4 to
allow elongation to oligomers of these capsular polysaccharide bacterial
species.
3. Experimental
3.1. General information
All chemicals were purchased from Sigma Aldrich (now owned by
Merck KGaA) and used without any further purification unless otherwise
described. All the reactions were performed under Argon atmosphere
and using dry solvents, unless otherwise indicated. Dichloromethane
(DCM), pyridine (Pyr) and triethylamine (TEA) were freshly distilled
from CaH2 prior to use. Dimethylformamide (DMF) and methanol were
dried over activated molecular sieves.
Scheme 2. Synthesis of donor 1. Reagents and conditions: a) Ac2O, Pyridine; b)
PhSH, BF3⋅OEt2, DCE, 60 ◦C, 65% over 2 steps; c) MeONa, MeOH; d) benzal-
dehyde dimethyl acetal, PTSA, CH3CN; e) KOH, EtOH/H2O, reflux; f) p-nitro-
phenyl chloroformate, NaHCO3, H2O, CH3CN, reflux, 70% over 4 steps; g) BnBr,
NaH, TBAI, DMF, 0 ◦C to r.t., 93%; h) Et3SiH, BF3⋅OEt2, MS 4 Å, CH2Cl2, 81%; i)
Ac2O, Pyridine, 96%. Abbreviations: DCE = dichloroethane; PTSA = p-Tolue-
Thin Layer Chromatography (TLC) was performed by using Silica gel
on Merck TLC-PET foils precoated with a fluorescent indicator. Flash
nesulfonic acid; TBAI
Dimethylformamide.
=
Tetrabutylammonium iodide; DMF
=
Table 1
Results of the glycosylations.
entry
Donor
Activation
system
Temp.
Time
Product
(yield)
α/β
ratioa
systems indicate that the relative energies of the transition states leading
to the alfa or beta anomers preferentially support the formation of the
β-adduct. β-Selectivity is expected for both donors 1 and 2, and the re-
sults of the experimental reactions confirm this preference, within the
limitation of model accuracy. Indeed, both benzylidene bearing donor 1
(min.)
1
2
3
4
5
6
7
2
2
1
1
1
1
1
BSP/Tf2O
NIS/AgOTf
NIS/AgOTf
BSP/Tf2O
BSP/Tf2O
BSP/Tf2O
BSP/Tf2O
- 40 ◦C
- 40 ◦C
- 40 ◦C
- 40 ◦C
- 40 ◦C
- 40 ◦C
- 60 ◦C
40′
90′
90′
40′
15′
5′
12 (ꢀ )
–
12 (70%)
13 (55%)
13 (57%)
13 (55%)
13 (59%)
13 (27%)
4:6
1:1
4:6
4:6
4:6
4:6
and bicyclic donor 2 gave glycosylation adducts in α/β ratio of 4:6. Even
if the β-selectivity is moderate, we show that the presence of the 2,3-
carbamate protecting group on mannosamine donor favors the forma-
tion of the β-adduct, either in the presence or not of the 4,6-O-
40′
a
Anomeric ratio determined from an anomeric mixture by 1H NMR analysis.
Scheme 3. Glycosylations of the 2,3-carba-
mate protected mannosamine donors 1 and
2 with acceptor 3. All reactions were con-
ducted in DCM with 1.2 eq of the donors.
Two different activation systems were used
for the glycosylations: a) 3, BSP (1.5 eq.)/
Tf2O (1.5 eq.)/TTBP (3 eq.); b) 3, NIS (1.5
eq.)/AgOTf (0.5 eq.). Other reaction condi-
tions: c) i) Et3SiH, BF3⋅OEt2, MS 4 Å, CH2Cl2;
ii) Ac2O, Pyridine. NIS = N-Iodosuccinimide,
BSP = 1-(Phenylsulfinyl)piperidine, TTBP =
2,4,6-Tri-tert-butylpyrimidine.
4

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
Fig. 3. A) 1H NMR spectrum of the
α
/β mixture of disaccharide 12; B) 1H NMR spectrum of the crude disaccharide 12a obtained starting from 13a after reductive
benzylidene opening and acetylation, which has been used for chemical correlation; C) 1H NMR spectrum of the
α/β mixture of disaccharide 13.
column chromatography was performed using a high-purity grade silica
gel (SiO2, high-purity grade (9385), pore size 60 Å, 230–400 mesh
particle size) by Merck. The purity of all synthetized compounds was
verified by nuclear magnetic resonance (NMR) analysis, using a 500
MHz Bruker FT-NMR AVANCE DRX500 spectrometer (pulsed field
gradient, reverse broadband probe, 1H at 500.13 MHz and 13C at 125.77
MHz) at a sample temperature of 298 K. High Resolution Mass Spec-
trometry (HRMS) was carried out on a high definition hybrid
quadrupole/time-of-flight (QTof) mass spectrometer (Synapt G2Si sys-
tem by Waters) equipped with electron-spry ionization (ESI) probe. A
Thermo Quest Finnigan LCQTMDECA ion trap mass spectrometer,
equipped with a Finnigan ESI interface, was used to perform Low Res-
olution Mass Spectrometry (LRMS). Optical rotation was measured at
room temperature with a PerkinElmer 241 polarimeter (589 nm, D line
from Na lamp). The synthetic sequence is reported at the best optimized
scale.
3.2. Phenyl 2-acetamido-2-deoxy-3,4,6-tri-O-acetyl-1-thio-
α-D-
mannopyranoside (8)
Ac2O (5 mL, 53 mmol) was added to a stirred suspension of N-acetyl-
◦
D-mannosamine (2.04 g, 9.23 mmol) in pyridine (10 mL) at 0 C. The
mixture was gradually warmed to room temperature and stirred for 24
h. The reaction, turned into a clear solution, was concentrated under
reduced pressure. The peracetylated intermediate (3.46 g, mixture of
anomers, Rf = 0.34 in ethyl acetate) was obtained as white solid, and
used in the next step without further purification.
Thiophenol (1.42 mL, 13.85 mmol) was added to a solution of the
crude peracetylated intermediate in dichloroethane (45 mL). BF3⋅Et2O
(2.30 mL, 18.46 mmol) was slowly added via dropping funnel, and the
resulting mixture was heated at 60 ◦C. After 24 h, the reaction (turned
into a deep purple solution) was cooled to room temperature, diluted
with dichloromethane (30 mL) and washed with satd NaHCO3 (3 × 50
mL, until basic pH). The aqueous phases were extracted with dichloro-
methane (80 mL) and the combined organic phases were washed with
5

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
3.4. Phenyl 2-amino-4,6-O-benzylidene-2,3-N,O-carbonyl-2-deoxy-1-
thio-α-D-mannopyranoside (10)
KOH (11.7 g, 209.0 mmol) was added to a stirred suspension of crude
9 (5.97 mmol) in aqueous ethanol (90 mL, EtOH/H2O 5:1). The reaction
mixture was heated under reflux for 24 h. The hot solution (turned into a
deep brown mixture) was carefully poured into hot water (50 mL). After
cooling to room temperature, the mixture was kept into an ice bath until
the formation of an off-white precipitate was observed. The solid was
filtered an a Hirsch funnel, washed with water and dried under vacuum.
The crude product de-acetylated intermediate (1.73 g, brown solid) was
used in the next step without any further purification. Rf = 0.49 (ethyl
acetate/methanol 6:1) and 0.10 (hexane/ethyl acetate 25:75), LRMS
(ESI+): m/z for C19H21NO4SNa calcd 382.11 [M+Na]+, found 360.0
[M+H]+ (100%); 741.0 [2 M + Na]+ (55%).
To a suspension of the crude intermediate in acetonitrile (35 mL), 1
M aqueous NaHCO3 (2.51 g, 29.85 mmol) was added, and the mixture
◦
was cooled to 0 C. A solution of p-nitrophenyl chloroformate (3.0 g,
Fig. 4. J1,2 coupling constant values of a tricyclic saccharidic model calculated
14.92 mmol) in acetonitrile (25 mL) was then added dropwise to the
reaction mixture. After 10 min, the reaction (turned from white into
yellow) was gradually warmed to room temperature and stirred for 30
by computational studies.
◦
brine (50 mL), dried over sodium sulfate, and concentrated at reduced
pressure. The crude (red oil) was purified by flash chromatography
(hexane/ethyl acetate gradient, 35:65 to 25:75) to give compound 8
(2.67 g, 65% two steps, Rf = 0.37 in hexane/ethyl acetate 3:7) as a white
solid. The spectroscopic data were in agreement with those reported in
literature [29].
min. Heating at 80 C for 4 h led to the disappearance of the starting
material as revealed by TLC (disappearance of the starting material was
followed in ethyl acetate/methanol 6:1, and product formation with
hexane/ethyl acetate 1:1). After cooling to room temperature, the re-
action mixture was diluted with ethyl acetate, washed with brine (2 ×
75 mL), and the combined aqueous layers were extracted with ethyl
acetate (100 mL). Then, the combined organic layers were dried over
sodium sulfate, filtered, and evaporated. The crude product was purified
by flash chromatography (hexane/ethyl acetate, 60:40) to give 10 (1.60
g, 70%, over 4 steps) as a white solid. Rf = 0.30 (hexane/ethyl acetate
3.3. Phenyl 2-acetamido-4,6-O-benzylidene-2-deoxy-1-thio-
mannopyranoside (9)
α-D-
20 = +169 (c 1, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.66–7.31
A 1 M solution of sodium methoxide in methanol (4.55 mL) was
slowly added to a solution of compound 8 (2.67 g, 6.07 mmol) in
methanol (60 mL). The reaction was stirred for 3 h at room temperature
(TLC monitoring: disappearance of the starting material with hexane/
ethyl acetate 3:7, product formation with ethyl acetate/methanol 6:1),
then neutralized with an ion exchange resin (Dowex® 50WX8, H+
form), filtered and concentrated. The deacetylated mannopyranoside
[29] (1.88 g, Rf = 0.49 in ethyl acetate/methanol 6:1) was recovered as
white foam, and used in the next step without any further purification.
To a solution of the crude mannopyranoside in acetonitrile (60 mL),
benzaldehyde dimethyl acetal (2.2 mL, 14.92 mmol) and p-toluene-
sulfonic acid (0.23 g, 1.19 mmol) were added. The disappearance of the
starting material was monitored by TLC, ethyl acetate/methanol 6:1.
After 10 min, a white precipitate was formed, and the reaction was
quenched by the addition of TEA and concentrated. Benzylidene 9 (4.3
g, crude) was recovered as an amorphous white solid and used directly
in the next step without any further purification. A small aliquot (50 mg)
of the crude was purified by flash chromatography for the spectroscopic
6:4). [
α]
D
(m, 10H, arom.), 6.53 (s, 1H, NH), 5.60 (s, 1H, CHPh), 5.57 (d, J1,2 = 1.9
Hz, 1H, H-1), 4.83 (t, J3,4 = J2,3 = 7.8 Hz, 1H, H-3), 4.36–4.22 (m, 3H, H-
2, H-5, H-6a), 4.06 (dd, J4,5 = 9.8, J3,4 = 7.8 Hz, 1H, H-4), 3.82–3.69 (m,
1H, H-6b). 13C NMR (CDCl3, 126 MHz) δ 158.74 (OCONH), 136.70 (1C,
quat.), 132.92 (2C, arom.), 131.52 (1C, quat.), 129.33–126.17 (8C,
arom.), 101.81 (CHPh), 83.75 (C-1), 78.80 (C-4), 75.02 (C-3), 68.57 (C-
6), 61.46 (C-5), 56.18 (C-2). HRMS (ESI+): m/z for C20H19NO5NaS calcd
408.0882 [M+Na]+, found 408.0878.
3.5. Phenyl 2-amino-2-N-benzyl-4,6-O-benzylidene-2,3-N,O-carbonyl-2-
deoxy-1-thio-α-D-mannopyranoside (1)
Benzyl bromide (1.5 mL, 12.3 mmol) was added to a stirred solution
of compound 10 (1.58 g, 4.10 mmol) and tetrabutylammonium iodide
TBAI (15 mg, 0.041 mmol) in DMF (40 mL). The mixture was cooled at
0 ◦C, and NaH (60% in mineral oil, 0.82 g, 20.50 mmol) was added. The
reaction was gradually warmed to room temperature, and monitored by
TLC analysis (disappearance of starting material was followed in hex-
ane/ethyl acetate 6:4, and product formation with toluene/ethyl acetate
95:5). After 90 min, the mixture was quenched by carefully addition of
methanol, then diluted with ethyl acetate and washed with water (1 ×
70 mL). The aqueous phase was extracted with ethyl acetate (3 × 40
mL), and the combined organics were washed with brine (70 mL), dried
over sodium sulfate, filtered, and evaporated. Flash chromatography of
the crude (toluene/ethyl acetate 95:5) gave compound 1 (1.81 g, 93%)
characterizations of compound 9. Rf = 0.76 (ethyl acetate/methanol
20
6:1) and 0.20 (hexane/ethyl acetate 25:75). [
α]
= +153 (c 0.2, Py).
1H NMR (Py-d5, 500 MHz) δ 9.13 (d, J2,NH = 7.6 DHz, 1H, NHAc), 7.98 (d,
J3,OH = 4.1 Hz, 1H, OH), 7.68–7.14 (m, 10H, arom.), 6.15 (s, 1H, H-1),
5.53 (s, 1H, CHPh), 5.40 (dd, J2,NH = 7.2, J2,3 = 5.4 Hz, 1H, H-2),
4.83–4.76 (m, 1H, H-3), 4.71 (td, J4,5 = 9.8, J5,6a = 4.8 Hz, 1H, H-5),
4.41 (t, J4,5 = J3,4 = 9.8 Hz, 1H, H-4), 4.28 (dd, J6a,6b = 10.2, J5,6a = 4.8
Hz, 1H, H-6a), 3.66 (t, J6a,6b = 10.2 Hz, 1H, H-6b), 2.13 (s, 3H,
NHCOCH3). 13C NMR (Py-d5, 126 MHz) δ 170.73 (NHCOCH3), 138.29
(quat.), 134.64 (quat.), 129.28–126.86 (10C, arom.), 102.38 (CHPh),
88.89 (C-1), 80.38 (C-4), 68.51 (C-6), 66.87 (C-3), 65.28 (C-5), 56.08 (C-
2), 22.78 (NHCOCH3). HRMS (ESI+): m/z for C21H23NO5NaS calcd
424.1195 [M+Na]+, found 424.1190.
as a white foam. Rf = 0.30 (toluene/ethyl acetate 95:5). [
α
]
20 = +131
D
(c 1, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.57–7.25 (m, 15H), 5.70 (s,
1H, H-1), 5.61 (s, 1H, CHPh), 4.89 (d, J = 15.3 Hz, 1H, NCHHPh), 4.73
(t, J2,3 = J3,4 = 7.8 Hz, 1H, H-3), 4.37–4.29 (m, 1H, H-5), 4.27–4.18 (m,
2H, H-6a, NCHHPh), 4.00–3.92 (m, 2H, H-2, H-4), 3.76 (t, J6a,6b = J5,6b
= 10.2 Hz, 1H, H-6b). 13C NMR (CDCl3, 126 MHz) δ 157.84 (OCONBn),
136.68–131.97 (quat.), 129.40–126.22 (15C, arom.), 101.92 (CHPh),
82.03 (C-1), 79.33 (C-4), 72.12 (C-3), 68.42 (C-6), 61.19 (C-5), 58.54 (C-
2), 47.00 (NCH2Ph). HRMS (ESI+): m/z for C27H25NO5NaS calcd
6

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
20
498.1351 [M+Na]+, found 498.1351.
(toluene/ethyl acetate 7:3). [
α]
= ꢀ 2.2 (c 0.5, CHCl3). 1H NMR
D
(CDCl3, 500 MHz) δ 7.46–7.22 (m, 10H, arom.), 5.84 (d, J1,2 = 3.6 Hz,
1H, H-1), 5.32 (s, 1H, H-1′), 5.11 (dd, J3’,4’ = 7.3 Hz, J4’,5’ = 9.8 Hz, 1H,
H-4′), 4.75 (d, J = 15.4 Hz, 1H, NCHHPh), 4.71 (d, J1,2 = 3.6 Hz, 1H, H-
2), 4.63–4.54 (m, 3H, H-3′, OCH2Ph), 4.28 (d, J3,4 = 1.5 Hz, 1H, H-3),
4.25 (d, J = 15.4 Hz, 1H, NCHHPh), 4.11–4.05 (m, 4H, H-4, H-5, 2 H-6),
4.00–3.94 (m, 1H, H-5′), 3.80 (d, J2’,3’ = 7.8 Hz, 1H, H-2′), 3.65–3.55
(m, 2H, 2 H-6′), 2.05 (s, 3H, OCOCH3), 1.50 (s, 3H, CH3 isopropylidene),
1.43 (s, 3H, CH3 isopropylidene), 1.38 (s, 3H, CH3 isopropylidene), 1.22
(s, 3H, CH3 isopropylidene). 13C NMR (CDCl3, 126 MHz) δ 169.39
(OCOCH3), 157.65 (OCONBn), 137.58 (quat.), 135.04 (quat.),
128.87–127.78 (10C, arom.), 112.06 (CCH3), 109.63 (CCH3), 105.17 (C-
1), 95.76 (C-1′), 83.59 (C-2), 80.77 (C-3), 80.73 (C-4 or C-5), 73.73
(OCH2Ph), 73.23 (C-3), 72.52 (C-4 or C-5), 69.24 (C-6′), 69.05 (C-4′),
68.10 (C-5′), 67.50 (C-6), 57.76 (C-2′), 47.13 (NCH2Ph), 27.18 (CCH3),
26.78 (CCH3), 26.11 (CCH3), 25.59 (CCH3), 20.75 (OCOCH3). HRMS
(ESI+): m/z for C35H43NO12Na calcd 692.2683 [M+Na]+, found
692.2687. Beta anomer (12b): Rf β = 0.44 (toluene/ethyl acetate 7:3).
3.6. Phenyl 2-amino-2-N-benzyl-6-O-benzyl-2,3-N,O-carbonyl-2-deoxy-
1-thio-α-D-mannopyranoside (11)
A suspension of compound 1 (0.4 g, 0.84 mmol) and 4 Å MS (0.4 g) in
dichloromethane (17 mL) was stirred for 10 min at room temperature,
then triethylsilane (1.25 mL, 8.40 mmol) was added After 0.5 h, a 0.5 M
solution of BF3⋅Et2O (4.20 mL, 2.10 mmol) in dichloromethane was
slowly added via dropping funnel to the reaction mixture. After 2 h, the
reaction was quenched with 0.9 mL of TEA, diluted with dichloro-
methane, filtered over a Celite pad and concentrated in vacuo. The res-
idue was purified by flash chromatography (toluene/ethyl acetate
gradient, 85:15 to 70:30) to afford compound 11 (0.32 g, 81%) as a
colourless oil. Rf = 0.26 (toluene/ethyl acetate 8:2). [
α
]
20 = +51 (c 0.5,
D
CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.53–7.14 (m, 15H, arom), 5.67 (s,
1H, H-1), 4.86 (d, J = 15.3 Hz, NCHHPh), 4.60 (d, J = 11.9 Hz, 1H,
OCHHPh), 4.55 (t, J2,3 = J3,4 = 7.6 Hz, 1H, H-3), 4.50 (d, J = 11.9 Hz,
1H, OCHHPh), 4.32–4.24 (m, 1H, H-5), 4.18 (d, J = 15.3 Hz, 1H,
NCHHPh), 4.03–3.95 (m, 1H, H-4), 3.90 (dd, J1,2 = 1.1 Hz, J2,3 = 7.6 Hz,
1H, H-2), 3.73 (m, 2H, H-6), 3.17 (br s, 1H, OH). 13C NMR (CDCl3, 126
MHz) δ 158.12 (OCONBn), 137.65–132.35 (3C, quat.), 129.21–127.75
(15C, arom.), 81.74 (C-1), 76.48 (C-3), 73.60 (OCH2Ph), 69.67 (C-5),
69.40 (C-4), 69.29 (C-6), 58.22 (C-2), 46.75 (NCH2Ph). HRMS (ESI+):
m/z for C27H27NO5NaS calcd 500.1508 [M+Na]+, found 500.1520.
[
α
]
20 = +27.8 (c 0.5, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.50–7.15
(m,D10H, arom.), 6.02 (d, J1,2 = 3.6 Hz, 1H, H-1), 5.20 (dd, 1H, J3’,4’
=
7.5 Hz, J4’,5’ = 9.8 Hz, H-4′), 5.09 (br s, 1H, H-1′), 4.80 (d, J = 15.3 Hz,
1H, NCHHPh), 4.64–4.50 (m, 4H, H-2, H-3′, OCH2Ph), 4.32 (dd, J3,4
=
3.8 Hz, J4,5 = 7.1 Hz, 1H, H-4), 4.25–4.16 (m, 2H, H-3, NCHHPh),
3.98–3.91 (m, 1H, H-5′), 3.81 (dd, 1H, J5,6a = 6.3 Hz, J6a,6b = 11.4 Hz,
H-6a), 3.76 (br d, 1H, J2’,3’ = 8.0 Hz, H-2′), 3.71 (dd, J5,6b = 2.5 Hz,
J6a,6b = 11.4, 1H, H-6b), 3.68–3.62 (m, 1H, H-5), 3.62–3.51 (m, 2H, 2 H-
6′), 2.02 (s, 3H, OCOCH3), 1.51 (s, 3H, CH3 isopropylidene), 1.36 (s, 3H,
CH3 isopropylidene), 1.35 (s, 3H, CH3 isopropylidene), 1.32 (s, 3H, CH3
isopropylidene). 13C NMR (CDCl3, 126 MHz) δ 169.39 (OCOCH3),
157.69 (OCONBn), 137.78 (quat.), 134.82 (quat.), 128.92–127.70 (10C,
arom.), 112.25 (CCH3), 106.40 (C-1), 100.99 (CCH3), 94.78 (C-1′),
83.96 (C-2), 79.16 (C-4), 75.04 (C-3), 73.65, 73.52 (C-3′, OCH2Ph),
71.06 (C-5), 69.16 (C-6′), 69.02 (C-4′), 67.42 (2C, C-5′ and C-6), 57.39
(C-2’), 47.05 (NCH2Ph), 27.18 (CCH3), 26.51 (CCH3), 23.97 (CCH3),
23.95 (CCH3), 20.77 (OCOCH3). HRMS (ESI+): m/z for C35H43NO12Na
calcd 692.2683 [M+Na]+, found 692.2677.
3.7. Phenyl 4-O-acetyl-2-amino-2-N-benzyl-6-O-benzyl-2,3-N,O-
carbonyl-2-deoxy-1-thio-α-D-mannopyranoside (2)
Ac2O (1.5 mL, 15.70 mmol) was added to a stirred solution of 11
(0.30 g, 0.63 mmol) in pyridine (5 mL). After 24 h, the reaction was
concentrated under reduced pressure, and then dried under vacuum.
Flash chromatography of the crude (hexane/ethyl acetate gradient,
75:25 to 65:35) afforded compound 2 (313 mg, 96%) as a colourless oil.
Rf = 0.28 (hexane/ethyl acetate 75:25) and 0.49 (toluene/ethyl acetate
20
8:2). [
α
]
D
= +61 (c 1.04, CHCl3). 1H NMR (CDCl3, 500 MHz) δ
7.48–7.20 (m, 15H, arom.), 5.65 (brs, 1H, H-1), 5.20 (dd, J3,4 = 7.5 Hz,
J4,5 = 9.6 Hz, 1H), 4.88 (d, J = 15.3 Hz, 1H, NCHHPh), 4.63 (t, J2,3
=
3.9. Phenyl 2-amino-2-N-benzyl-4,6-O-benzylidene-2,3-N,O-carbonyl-2-
J3,4 = 7.5 Hz, 1H, H-3), 4.55–4.41 (m, 3H, OCH2Ph, H-5), 4.19 (d, J =
15.3 Hz, 1H, NCHHPh), 3.92 (dd, J1,2 = 0.9 Hz, J2,3 = 7.5 Hz, 1H, H-2),
3.62–3.53 (m, 2H, H-6a and H-6b), 2.08 (s, 3H, OCOCH3). 13C NMR
(CDCl3, 126 MHz) δ 169.43 (OCOCH3), 157.31 (OCONBn), 137.65
(quat.), 134.45 (quat.), 132.79 (arom.), 132.23 (quat.), 129.22–127.73
(14C, arom.), 81.87 (C-1), 73.54 (OCH2Ph), 73.39 (C-3), 69.28 (C-4),
69.18 (C-6), 68.88 (C-5), 58.03 (C-2), 46.76 (NCH2Ph), 20.82
(OCOCH3). HRMS (ESI+): m/z for C29H29NO6NaS calcd 542.1613
[M+Na]+, found 542.1612.
deoxy-1-thio-
α/β-D-mannopyranosyl-(1 → 3)-1,2:5,6-di-O-
isopropylidene-
α
-D-glucofuranoside (13)
A suspension of compound 1 (40 mg, 0.084 mmol), 1-(phenyl-
sulfinyl)piperidine (BSP, 22 mg, 0.10 mmol), 2,4,6-tri-tert-butylpyr-
imidine (TTBP, 52 mg, 0.21 mmol) and 4 Å MS (40 mg) in
dichloromethane (3 mL) was stirred for 5 min at room temperature and
then cooled to ꢀ 40 ◦C. After 10 min, Tf2O (0.018 mL, 0.10 mmol) was
added. Then, a solution of acceptor 3 (18 mg, 0.069 mmol) in
dichloromethane (2 mL) was added dropwise. The reaction was moni-
tored by TLC analysis (hexane/ethyl acetate 7:3, toluene/ethyl acetate
85:15). After 40 min, the reaction was quenched with TEA, diluted with
dichloromethane, filtered over Celite, and the solvent evaporated under
reduced pressure. Flash chromatography of the crude (toluene/ethyl
acetate 85:15) afforded disaccharide 13 (16 mg, 60%) as a colourless oil
3.8. 4-O-Acetyl-2-amino-2-N-benzyl-6-O-benzyl-2,3-N,O-carbonyl-2-
deoxy-1-thio-
α/β-D-mannopyranosyl-(1 → 3)-1,2:5,6-di-O-
isopropylidene-
α
-D-glucofuranoside (12)
D-Glucose diacetonide acceptor 3 (20 mg, 0.064 mmol), donor 2 (40
mg, 0.077 mmol) and freshly activated 4 Å molecular sieves (60 mg)
were suspended in dichloromethane (2 mL), stirred under argon for 10
min and then the mixture was cooled to ꢀ 40 ◦C. After 10 min, N-iodo-
succinimide (NIS, 22 mg, 0.096 mmol) and AgOTf (10 mg, 0.032 mmol)
were quickly added and the reaction was monitored by TLC analysis
(toluene/ethyl acetate 7:3). After 1.5 h, the reaction was quenched by
the addition of TEA, diluted with dichloromethane, filtered over Celite,
and the solvent evaporated under reduced pressure. The crude product
was purified by flash chromatography (toluene/ethyl acetate gradient,
80:20 to 70:30) to give product 12 (30 mg, 70%, colourless oil) as a 4:6
in a 4:6 α/β mixture of anomers. The two anomers can be separated by
flash chromatography (toluene/ethyl acetate gradient, 85:15). Alpha
anomer (13a): Rf α = 0.26 (toluene/ethyl acetate 85:15). [
α
]
20 = ꢀ 13.3
D
(c 0.2, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.58–7.26 (m, 10H, arom.),
5.86 (d, J1,2 = 3.7 Hz, 1H, H-1), 5.65 (s, 1H, CHPh), 5.35 (br s, 1H, H-1′),
4.77–4.66 (m, 2H, NCHHPh, H-3′), 4.50 (d, J1,2 = 3.7 Hz, 1H, H-2),
4.46–4.37 (m, 1H, H-6a′), 4.31 (d, 1H, J = 15 Hz, NCHHPh), 4.27 (br s,
1H, H-3), 4.17–4.03 (m, 4H, H-4, H-5, 2 H-6), 3.96–3.78 (m, 4H, H-2′, H-
4′, H-6b′, H-5′), 1.52 (s, 3H, CH3 isopropylidene), 1.45 (s, 3H, CH3 iso-
propylidene), 1.41 (s, 3H, CH3 isopropylidene), 1.33 (s, 3H, CH3 iso-
propylidene). 13C NMR (CDCl3, 126 MHz) δ 158.26 (OCONBn), 136.59
(quat.), 135.11 (quat.), 129.27–126.02 (10C, arom.), 112.26 (CCH3),
109.75 (CCH3), 105.17 (C-1), 101.79 (CHPh), 95.72 (C-1′), 84.11 (C-2),
mixture of
chromatography (toluene/ethyl acetate gradient, 80:20 to 70:30 or
hexane/ethyl acetate, 70:30). Alpha anomer (12a): Rf = 0.37
α/β anomers. The two anomers can be separated by flash
α
7

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
80.65 (C-4 or C-5), 79.60 (C-3), 78.87 (C-4′), 72.53 (C-4 or C-5), 72.02
(C-3′), 68.59 (C-6′), 67.50 (C-6), 60.53 (C-5′), 58.40 (C-2′), 47.38
(NCH2Ph), 27.29 (CCH3), 26.78 (CCH3), 26.28 (CCH3), 25.66 (CCH3).
HRMS (ESI+): m/z for C33H39NO11Na calcd 648.2421 [M+Na]+, found
commercial, or not-for-profit sectors.
Appendix A. Supplementary data
648.2416. Beta anomer (13b): Rf β = 0.20 (toluene/ethyl acetate 85:15).
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.carres.2021.108421.
20
[
α
]
= ꢀ 115 (c 0.2, CHCl3). 1H NMR (CDCl3, 500 MHz) δ 7.65–7.16
(m,D10H, arom.), 5.79 (d, J1,2 = 3.7 Hz, 1H, H-1), 5.69 (s, 1H, CHPh),
5.02 (d, 1H, J1’,2’ = 3.0 Hz, H-1′), 4.98 (dd, 1H, J3’,4’ = 7.1 Hz, J4’,5’
10.7 Hz, H-4′), 4.78 (d, J = 14.5 Hz, 1H, NCHHPh), 4.7 (dd, 1H, J2’,3’
=
=
Supporting Information
7.1 Hz, J3’,4’ = 9.0 Hz, H-3′), 4.46 (d, J = 3.4 Hz, 1H, H-3), 4.41–4.36 (m,
1H, H-5), 4.98 (dd, 1H, J5’,6’a = 5.0 Hz, J6’a’,6’b = 10.3 Hz, H-6’a),
4.22–4.16 (m, 2H, H-2, H-6a), 4.16–4.10 (m, 2H, H-4, H-6b), 4.00 (d, J
= 14.5 Hz, 1H, NCHHPh), 3.96 (dd, 1H, J1’,2’ = 3.0 Hz, J2’,3’ = 9.0 Hz, H-
2′), 3.92 (dd, 1H, J5’,6’b = J6’a’,6’b = 10.3 Hz, H-6’b), 3.71–3.64 (m, 1H,
H-5′), 1.54 (s, 3H, CH3 isopropylidene), 1.51 (s, 3H, CH3 iso-
propylidene), 1.43 (s, J = 8.5 Hz, 3H, CH3 isopropylidene), 1.36 (s, 3H,
CH3 isopropylidene). 13C NMR (CDCl3, 126 MHz) δ 157.58 (OCONBn),
136.78 (quat.), 135.33 (quat.), 129.21–125.94 (10C, arom.), 112.08
(CCH3), 109.48 (CCH3), 105.05 (C-1), 100.75 (CHPh), 90.89 (C-1′),
80.84 (C-2), 80.45 (C-4), 76.06 (C-3), 75.75 (C-4′), 72.86 (C-3′), 71.78
(C-5), 69.00 (C-6′), 67.70 (C-6), 64.48 (C-5′), 54.97 (C-2′), 47.32
(NCH2Ph), 27.63 (CCH3), 26.83 (CCH3), 26.38 (CCH3), 25.72 (CCH3).
HRMS (ESI+): m/z for C33H39NO11Na calcd 648.2421 [M+Na]+, found
648.2420.
Copies of the 1H and 13C NMR spectra of all new compounds are
given in the Supporting Information.
References
[1] D. Crich, Mechanism of a chemical glycosylation reaction, Acc. Chem. Res. 43
(2010) 1144–1153.
[2] L.K. Mydock, A.V. Demchenko, Mechanism of chemical O-glycosylation: from early
studies to recent discoveries, Org, Biomol. Chem. 8 (2010) 497–510.
[3] S.S. Nigudkar, A.V. Demchenko, Stereocontrolled 1,2-cis glycosylation as the
driving force of progress in synthetic carbohydrate chemistry, Chem. Sci. 6 (2015)
2687–2704.
´
[4] P.O. Adero, H. Amarasekara, P. Wen, L. Bohe, D. Crich, The experimental evidence
in support of glycosylation mechanisms at the SN1-SN2 interface, Chem. Rev. 118
(2018) 8242–8284.
[5] D. Crich, Methodology development and physical organic chemistry: a powerful
combination for the advancement of glycochemistry, J. Org. Chem. 76 (2011)
9193–9209.
[6] S.G. Pistorio, J.P. Yasomanee, A.V. Demchenko, Hydrogen-bond-mediated
aglycone delivery: focus on β-mannosylation, Org. Lett. 16 (2014) 716–719.
[7] A.A.H. Abdel-Rahman, S. Jonke, E.S.H. El Ashry, R.R. Schmidt, Stereoselective
synthesis of β-D-Mannopyranosides with reactive mannopyranosyl donors
possessing a neighboring electron-withdrawing group, Angew. Chem. Int. Ed. 41
(2002) 2972–2974.
3.10. Computational details
The Gaussian 09 program package was used for all the optimizations
of the transition states of the reaction between the glycosyl donors 4 and
5 and the glycosyl acceptor 3. The structures were optimized using the
BY3LYP functional at the 6-31G(d,p) level. All the optimization were
performed in dichloromethane solvent, using the PCM as solvation
model. Then, single-point energy calculations on the optimized geom-
etries were performed using the B3LYP/6–311++G(3df, 3pd) level still
with the same solvent model. Frequency calculations at the BY3LYP/6-
31G(d,p) gave the free energy correction which was added to the
B3LYP/6–311++G(3df, 3pd) single-point energy to give the free energy
values used to determine the relative free energy of the transition states.
Starting geometries were built taking into account the calculated TSs
reported in the literature for the reaction of isopropanol with 4,6-O-
methylidene protected mannosyl triflates [8] and introducing the
appropriate structural modifications on both the glycosyl donor and
acceptor. After a first optimization maintaining a fixed length (opt =
modredundant option) for the partial bonds of the anomeric carbon to
the leaving group and to the nucleophile, the free optimization of the TSs
was performed. An extensive search of the conformational space was
performed taking into account the orientation of the triflate as well as
the nucleophile with respect to the pyranose ring and, in the case of 5,
the gg, gt, and tg orientation at the C5–C6 bond. In each of the four
cases, a dozen of transition states were located and those reported in
Fig. 2 are the lowest energy structures. The calculated J1,2 coupling
constants were obtained using the nmr = spinspin option.
´
[8] M. Huang, G.E. Garrett, N. Birlirakis, L. Bohe, D.A. Pratt, D. Crich, Dissecting the
mechanisms of a class of chemical glycosylation using primary 13C kinetic isotope
effects, Nat. Chem. 4 (2012) 663–667.
´
´
´
[9] A. Franconetti, A. Arda, J.L. Asensio, Y. Bleriot, S. Thibaudeau, J. Jimenez-Barbero,
Glycosyl oxocarbenium ions: structure, conformation, reactivity, and interactions,
Acc. Chem. Res. 54 (2021) 2552–2564.
[10] D. Crich, S. Sun, Direct chemical synthesis of β-mannopyranosides and other
glycosides via glycosyl triflates, Tetrahedron 54 (1998) 8321–8348.
[11] D. Crich, S. Sun, formation of β-mannopyranosides of primary alcohols using the
sulfoxide method, J. Org. Chem. 61 (1996) 4506–4507.
[12] D. Crich, S. Sun, Direct synthesis of β-mannopyranosides by the sulfoxide method,
J. Org. Chem. 62 (1997) 1198–1199.
[13] D. Crich, S. Sun, Direct formation of β-mannopyranosides and other hindered
glycosides from thioglycosides, J. Am. Chem. Soc. 120 (1998) 435–436.
[14] D. Crich, M. Smith, 1-Benzenesulfinyl piperidine/trifluoromethanesulfonic
Anhydride: A potent combination of shelf-stable reagents for the low-temperature
conversion of thioglycosides to glycosyl triflates and for the formation of diverse
glycosidic linkages, J. Am. Chem. Soc. 123 (2001) 9015–9020.
[15] D. Crich, M. Smith, S-(4-Methoxyphenyl) benzenethiosulfinate (MPBT)/
Trifluoromethanesulfonic Anhydride: A convenient system for the generation of
glycosyl triflates from thioglycosides, Org. Lett. 2 (2000) 4067–4069.
[16] D. Crich, W. Cai, Z. Dai, Highly diastereoselective α-mannopyranosylation in the
absence of participating protecting groups, J. Org. Chem. 65 (2000) 1291–1297.
[17] D. Crich, A.U. Vinod, J. Picione, The 3,4-O-carbonate protecting group as a
β-directing group in rhamnopyranosylation in both homogeneous and
heterogeneous glycosylations as compared to the chameleon-like 2,3-O-carbonates,
J. Org. Chem. 68 (2003) 8453–8458.
[18] L. Morelli, S. Fallarini, G. Lombardi, C. Colombo, L. Lay, F. Compostella, Synthesis
and biological evaluation of a trisaccharide repeating unit derivative of
Streptococcus pneumoniae 19A capsular polysaccharide, Bioorg. Med. Chem. 26
(2018) 5682–5690.
´
[19] J.D.C. Codee, R.E.J.N. Litjens, R. den Heeten, H.S. Overkleeft, J.H. van Boom, G.
A. van der Marel, Ph2SO/Tf2O: a powerful promotor system in chemoselective
glycosylations using thioglycosides, Org. Lett. 5 (2003) 1519–1522.
[20] R.E.J.N. Litjens, M.A. Leeuwenburgh, G.A. van der Marel, J.H. van Boom, A novel
approach towards the stereoselective synthesis of 2-azido-2-deoxy-β-d-mannosides,
Tetrahedron Lett. 42 (2001) 8693–8696.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
[21] C. Alex, S. Visansirikul, A.V. Demchenko, A versatile approach to the synthesis of
mannosamine glycosides, Org. Biomol. Chem. 18 (2020) 6682–6695.
[22] S. Nakamura, T. Tsuda, N. Suzuki, S. Hashimoto, Direct and stereoselective
synthesis of 2-azido-2-deoxy-β-d-mannosides using the phosphate method,
Heterocycles 77 (2009) 843–848.
Acknowledgements
[23] B.R. Bhetuwal, F. Wu, S. Meng, J. Zhu, Stereoselective synthesis of 2-Azido-2-
deoxy-β-d-mannosides via Cs2CO3-mediated anomeric O-alkylation with primary
triflates: synthesis of a tetrasaccharide fragment of micrococcus luteus teichuronic
acid, J. Org. Chem. 85 (2020) 16196–16206.
The authors gratefully thank COST Action CA18013 - “Innovation
with Glycans: new frontiers from synthesis to new biological targets”.
The HRMS spectra have been obtained from UNITECH (Piattaforme
[24] S. Manabe, K. Ishii, Y. Ito, N-Benzyl-2,3-oxazolidinone as a glycosyl donor for
`
Tecnologiche di Ateneo, Universita degli Studi di Milano). This research
selective α-glycosylation and one-pot oligosaccharide synthesis involving 1,2-cis-
did not receive any specific grant from funding agencies in the public,
Glycosylation, J. Am. Chem. Soc. 128 (2006) 10666–10667.
8

L. Morelli et al.
Carbohydrate Research 509 (2021) 108421
[25] P. Wei, R.J. Kerns, Factors affecting stereocontrol during glycosidation of 2,3-
oxazolidinone-protected 1-tolylthio-N-acetyl-d-glucosamine, J. Org. Chem. 70
(2005) 4195–4198.
[28] C. Jones, Revised structures for the capsular polysaccharides from Staphylococcus
aureus Types 5 and 8, components of novel glycoconjugate vaccines, Carbohydr.
Res. 340 (2005) 1097–1106.
[26] J.D.M. Olsson, L. Eriksson, M. Lahmann, S. Oscarson, Investigations of
glycosylation reactions with 2-N-Acetyl-2N,3O-oxazolidinone-Protected
glucosamine donors, J. Org. Chem. 73 (2008) 7181–7188.
[29] T. Buskas, P.J. Garegg, P. Konradsson, J.-L. Maloisel, Facile preparation of glycosyl
donors for oligosaccharide synthesis: 2-azido-2-deoxyhexopyranosyl building
blocks, Tetrahedron: Asymmetry 5 (1994) 2187–2194.
´
[27] R.E.J.N. Litjens, L.J. van den Bos, J.D.C. Codee, R.J.B.H.N. van den Berg, H.
S. Overkleeft, G.A. van der Marel, Sulfonium triflate mediated glycosidations of
aryl 2-Azido-2-deoxy-1-thio-d-mannosides, Eur. J. Org Chem. (2005) 918–924.
9