Sulfur-containing functionalized carbohydrates are a remarkably important class of
compounds that find widespread use as pharmaceuticals.[1 ] Amongst them, thioglycosides are an important class of sugar derivatives that are
considered as useful simulants of biologically relevant O-glycosides, because they
are known to resist enzymatic hydrolysis[2 ] and can consequently be used as enzyme inhibitors in various biochemical studies.[3 ]
S -Glycosides are valuable glycomimetic derivatives that find a wide range of applications
in pharmaceutical science.[1a ]
[c ]
,
[4 ] Protected thioglycosides are often used in synthetic carbohydrate chemistry as versatile
glycosyl donors[5 ] because of their stability during various chemical transformations and the ease
with which they can be converted into a variety of other functionalities.[6 ] In addition, these derivatives serve as versatile intermediates in organic synthesis.[7 ]
Among the several glycosyl donors, thioglycosides are considered to be versatile intermediates
because of their high degree of stability in many organic reactions. These derivatives
are routinely used as glycosyl donor building blocks in producing a variety of glycosidic
linkages;[7a ]
[8 ] moreover, they are also found in various drugs, natural products, and medicinally
active agents,[9 ] including an hSGLT1 inhibitor, a ligand of lectin A, a cytotoxic Hsp90 inhibitor,
inhibitors of galactosidase and glycosidase,[10 ] and various antimicrobial agents (Figure [1 ]). Interestingly, another representative natural product bearing a C–S glycosidic
bond is the clinically used antibiotic lincomycin A[11 ] (Figure [1 ]). These derivatives clearly hold great potential in medicinal chemistry as well
as in carbohydrate chemistry. Therefore, the development of highly efficient synthetic
methods to access functionalized sulfur-containing carbohydrates is particularly appealing.
Thioglycoside derivatives are commonly prepared treating per-O-acetylated glycosyl
precursors with a thiophenol in the presence of a Lewis acid,[12 ] by substituting the halogen atom of an acetohaloglycoside with a thiolate anion,[13 ] or by a reaction of a 1-thioglycopyranose with an alkyl halide (Scheme [1a ]).[14 ]
Figure 1 Some examples of biologically active thioglycosides
Scheme 1 Conventional methods and the strategic use of para -quinone methides to access diarylmethyl thioglycosides in one-pot. P = protecting
group.
An alternative route to access (het)arylthioglycosides involves the use of thioglycosides
as nucleophiles under transition-metal catalysis (Scheme [1b ]).[15 ]
[16 ] These derivatives have also been prepared from carbohydrate sulfenates[17 ] or by free-radical addition of 1-thiosugars to alkenes.[18 ]
Despite these significant recent improvements, it is surprising to find that there
have been no reports of the preparation of diarylmethyl thioglycosides from para -quinone methides (p -QMs) through the generation in situ of glycosyl isothiouronium salts from per-O-acetylated
sugars (Scheme [1c ]).
In recent years, p -QMs have been explored extensively because of their unique ability to act as powerful
Michael acceptors with a variety of nucleophiles to generate highly functionalized
diarylmethane derivatives.[19 ]
[20 ] Anand’s group[19d,21 ] and others[22 ] have reported that a diversity of nucleophiles, including cyanides, malonates, glycine
Schiff bases, dicyanoolefins, the Seyferth–Gilbert reagent, allenic esters, styrenes,
and β-naphthols, can be used in 1,6-conjugate additions to p -QMs.
Despite these tremendous achievements, the development of new nucleophiles, such as
various S -glycosides, for the 1,6-conjugate addition of p -QMs is still in great demand. The chemistry of p -QMs is well explored in organic synthesis and physical organic chemistry, but less
so in carbohydrate chemistry. Keeping in mind the high reactivity of p -QMs toward nucleophiles, we surmised that these units might server as versatile diarylmethylating
agent for the synthesis of diarylmethane S -glycosides.
The direct one-pot conversion of glycosyl isothiouronium salts derivatives into glycosyl
thiol derivatives under phase-transfer conditions is a well-established process.[23 ] Therefore, glycosyl isothiouronium salts might be used for 1,6-conjugate addition
of p -QMs in the presence of base to generate glycosyl thiols in situ. However, to the
best of our knowledge, there are no reports that describe the synthesis of diarylmethyl
S -glycosides by nucleophilic addition of glycosyl thiols generate in situ with p -QMs. With this method diarylmethyl thioglycosides could be synthesized under mild
conditions without the use of malodorous and toxic mercaptans such as alkyl or aryl
thiols or expensive alkyl or aryl thiotrimethylsilanes.
Generally, the most-often-employed approaches involve cleavage of a glycosyl isothiouronium
salt by treatment with potassium carbonate and sodium hydrogen sulfite or metabisulfite
in water–acetone media, followed by reaction of the resulting acetylated 1-thioglycose
with an alkyl or aryl halide.[24 ] In most cases, the presence of water in the reaction mixture does not permit the
use of water-sensitive alkylation and acylation agents. Consequently, the 1-thiosugar
formed has to be isolated before alkylation, making the approach more laborious. The
synthesis of alkyl- or arylthioglycosides and glycosyl thioesters via isothiouronium
salts by using a base at room temperature has been recently described.[25 ]
We attempted to develop a one-pot reaction protocol for the direct preparation of
diarylmethyl thioglycoside from per-O-acetylated sugars via glycosyl isothiouronium
derivatives through the strategic use of p -QMs in the presence of alkylamines in acetonitrile. Reports in the literature suggest
that alkylamines are stronger bases but weaker nucleophiles than thiols; therefore,
by converting the isothiouronium salts into more-nucleophilic thiolate anions, they
should react readily with thioaldoses.
To standardize the reaction protocol, 2,3,4,6-tetra-O -acetyl-β-d -galactopyranose (1 ; 1.0 mmol) was treated with a mixture of thiourea (1.1 mmol) and boron trifluoride
etherate (1.1 mmol) in acetonitrile, and the mixture was heated to 80 °C for 30 minutes.
TLC showed complete conversion of the sugar per-O -acetate into slower-moving S -glycosyl isothiouronium salts. The mixture was then cooled to room temperature, and
the p -QM derivative 2a (1.1 mmol) and excess alkylamine were added sequentially. The mixture was then stirred
at room temperature for a further one hour. Various organic bases, such mono-, di-,
or trialkylamines, were found to be suitable for use in synthesis of the thioglycoside
by the described procedure (Table [1 ]). In the case of triethylamine, the reaction proceeded smoothly at room temperature
to give the diarylmethyl S -glycoside 3a
[26 ] as a diastereomeric mixture (dr 1:1) in 83% yield (Table [1 ], entry 1); no improvement in the yield of 3a was observed in the presence of diethylamine, DIPEA, DBU, or DABCO (Table [1 ], entries 2–6), even with longer reaction times. Less-basic amines, for example pyridine,
were unable to cleave the glycosyl isothiouronium salt (entry 7).
Table 1 One-Pot Protocol for the Synthesis of Diarylmethyl Thioglycoside 3a , and Screening of Various Aminesa
Entry
Base
Equiv
Time (h)
Yieldb (%)
1
Et3 N
2
2
83
2
DIPEA
2
2
77
3
DBU
2
2
69
4
DABCO
2
2
68
5
MeNH2
2
2
68
6
Et2 NH
2
2
78
7
py
2
2
NDc
8
Et3 N
4
6
84
a Reaction conditions: 1a (1.0 mmol), thiourea (1.1 mmol), BF3 ·Et2 O (1.1 mmol), MeCN (5 mL), 80 °C then 2a (1.1 mmol), base (2 mmol), rt.
b Isolated yield after silica gel chromatography.
c ND = not detected.
After completion of the reaction, as monitored by TLC (hexane: EtOAc 1:1), the solvent
was evaporated and the resulting syrup was diluted with CH2 Cl2 and washed with water, which on evaporation gave the pure diaryl thioglycoside 3a in 83% yield as a diastereomeric mixture in a one-pot process starting from the per-O -acetylated sugar.
After identifying the optimal one-pot conditions for this transformation (Table [1 ], entry 1), we examined the substrate scope of this transformation by using a wide
range of p -QMs 2a –n
[27 ] bearing various substituents on the phenyl ring together with various easily accessible
per-O -acetylated sugars (Table [2 ]). Most of the p -QMs employed in the reaction with various per-O -acetylated sugars delivered the corresponding diarylmethyl thioglycoside 3a –v in moderate to good yields (69–83%) in short reaction times.
To elaborate the substrate scope further, this one-pot method was extended to 1,6-conjugate
addition of several per-O -acetylated disaccharide moieties with wide range of p -QMs 2 to give corresponding desired diarylmethyl thioglycosides 4a –f
[28 ] (Table 3). The reaction conditions were successfully applied to the preparation
of diarylthioglycosides from d - or l -sugars as well as from disaccharides. The reaction proceeded smoothly at room temperature
affording the products in high yields. A series of thioglycosides of mono- and disaccharides
were prepared by using this one-pot procedure in short reaction times not exceeding
two hours.
Based on the above results and previously reports,[29 ] a plausible mechanism for the formation of 3 and 4 is proposed (Scheme [2 ]). The reaction is probably initiated by BF3 -induced formation of the 1,2-acyloxonium ion as an intermediate that reacts with
nucleophiles in only one possible way, leading exclusively to 1,2-trans -substitution products. The isothiourea intermediates obtained in 30 minutes without
isolation can be directly converted into thioglycosides by 1,6-conjugate addition
under basic conditions.
Scheme 2 Plausible mechanism of the reaction.
In conclusion, we have demonstrated the first strategic use of p -QMs to access diarylmethyl S -glycosides from the per-O -acetyl sugars through sequential thioglycosidation via S -glycosyl isothiouronium salts. The reaction conditions are operationally simple,
mild, reproducible, high-yielding, and can be scaled up for large-scale preparation.
We therefore believe that this method will open a route to novel diarylmethyl thioglycosides
as a novel and largely unexplored structural class of glycomimetic compounds.
