O- and N-sulfations of carbohydrates using sulfuryl imidazolium salts

Article abstract of DOI:10.1021/jo9014112

(Chemical Equation Presented) A series of sulfuryl imidazolium salts (SISs) were prepared and examined as reagents for incorporating trichloroethyl- protected sulfate esters into carbohydrates. The SIS that contained a 1,2-dimethylimidazolium moiety (SIS

Full text of DOI:10.1021/jo9014112

pubs.acs.org/joc
O- and N-Sulfations of Carbohydrates Using Sulfuryl Imidazolium Salts
Laura J. Ingram, Ahmed Desoky, Ahmed M. Ali, and Scott D. Taylor*
Department of Chemistry, University of Waterloo, 200 University Avenue West, Waterloo, Ontario,
Canada N2L 3G1
s5taylor@sciborg.uwaterloo.ca
Received July 3, 2009
A series of sulfuryl imidazolium salts (SISs) were prepared and examined as reagents for incorporat-
ing trichloroethyl-protected sulfate esters into carbohydrates. The SIS that contained a 1,2-di-
methylimidazolium moiety (SIS 9) proved to be a superior sulfating compared to SISs bearing no
alkyl groups or bulkier alkyl groups on the imidazolium ring. Difficult O-sulfations that required
prolonged reaction times and a large excess of the SIS bearing a 1-methylimidazolium group (SIS 5)
were achieved in high yield using less than half the amount of SIS 9 in less time. Certain N-sulfated
compounds that were practically inaccessible using SIS 5 were obtained in excellent yield using SIS 9.
Introduction
could be introduced at an early stage in the synthesis as a
protected, neutral sulfodiester(s).2,3 However, the sulfate
protecting groups that were initially designed for this pur-
pose had little impact on multistep organosulfate syntheses
mainly because of difficulties encountered with their intro-
duction and/or removal.2-5 More recently, two additional
sulfate protecting groups, with properties that make this
approach to organosulfate syntheses much more attractive
have been reported.6-8 We introduced one of these new
sulfate protecting groups, the 2,2,2-trichloroethyl (TCE)
group, in order to prepare aryl sulfates that could not be
obtained using the traditional methodology.6 The TCE-
protected sulfate group is incorporated using reagent 1 and
is readily removed under mild conditions using Zn or Pd/C in
the presence of ammonium formate (Scheme 1). Since our
initial report, this has become the most widely used sulfate
Biomolecules bearing one or more sulfate groups, such as
sulfated carbohydrates, nucleosides, steroids, and proteins,
have been known for many years and have been found to
play important roles in a wide range of crucial biochemical
processes. Consequently, new and improved methods for
preparing sulfated biomolecules and their derivatives are of
some importance. Organosulfates have traditionally been
prepared by treating their nonsulfated precursors with a
sulfating agent such as a sulfur trioxide-amine complex,
chlorosulfuric acid, or similar reagents. The sulfation step is
usually carried out at or near the end of the synthesis to
minimize purification and stability issues associated with the
highly polar and acid-sensitive sulfates. However, if the
target, such as carbohydrates, contains multiple reactive
groups but only one or a select few require sulfation then
intensive protecting group manipulation is required at the
end of the synthesis which can result in reduced yields.1 It has
been known for many years that the synthesis of organosul-
fates would be considerably improved if the sulfate group(s)
(2) Penney, C. L.; Perlin, A. S. Carbohydr. Res. 1981, 93, 241.
(3) Proud, A. D.; Prodger, J. C.; Flitsch, S. L. Tetrahedron Lett. 1997, 38,
7243.
(4) Karst, N. A.; Islam, T. F.; Linhardt, R. J. Org. Lett. 2003, 5, 4839.
(5) Karst, N. A.; Islam, T. F.; Avci, F. Y.; Linhardt, R. J. Tetrahedron
Lett. 2004, 45, 6433.
(1) For examples, see: (a) Orgueira, H. A.; Bartolozzi, A.; Schell, P.;
Litjens, R. E.; Palmacci, E. R.; Seeberger, P. H. Chem. Eur. J. 2003, 140.
(b) Nicolaou, K. C.; Bockovich, N. J.; Carcanague, D. R. J. Am. Chem. Soc.
1993, 115, 8843. (c) Singh, K.; Fernandez-Mayoralas, A.; Martin-Lomas, M.
J. Chem. Soc., Chem. Commun. 1994, 775. (d) Lubineau, A.; Lemoine, R.
Tetrahedron Lett. 1994, 35, 8795. (e) Stahl, W.; Sprengard, U.; Kretzschmar,
G.; DSchmidt, D. W.; Kunz, H. J. Prakt. Chem. 1995, 337, 441.
(6) Liu, Y.; Lien, I. F. F.; Ruttgaizer, S.; Dove, P.; Taylor, S. D. Org. Lett.
2004, 6, 209.
(7) Simpson, L. S.; Widlanski, T. S. J. Am. Chem. Soc. 2006, 128, 1605.
(8) Ethyl sulfite esters have also been used recently for preparing orga-
nosulfates. However, further optimization is required for this procedure to be
of general utility. See: Huiber, M.; Manuzi, A.; Rutjes, P. J. T.; van Delft, F.
L. J. Org. Chem. 2006, 71, 7473.
DOI: 10.1021/jo9014112
r
Published on Web 08/05/2009
J. Org. Chem. 2009, 74, 6479–6485 6479
2009 American Chemical Society

Ingram et al.
SCHEME 3. Sulfation of Carbohydrates Using Reagent 5
JOCFeatured Article
SCHEME 1. Introduction and Deprotection of TCE-Protected
Arylsulfate Esters
making a minor modification to the imidazolium ring of 5, a
more efficient sulfating agent can be obtained and certain sulfo-
protected compounds that were inaccessible or challenging to
obtain using reagent 5 can now be readily prepared.
SCHEME 2. Byproduct Produced from the Reaction of Carbo-
hydrate 2 with Reagent 1
Results and Discussion
We anticipated that the solubility, stability, and sulfating
properties of SIS 5could be altered by introducing alkyl groups
on the imidazolium ring. Toward this end, we prepared a series
of SISs, all of which contained the TCE group yet had different
alkyl groups at the 2- and 3-positions of the imidazolium ring.
The tetrafluoroborate counterion was also examined. In most
cases, the synthesis of these compounds was readily achieved
by reacting reagent 1 with the appropriate imidazole derivative
to give compounds 6-8 (Table 1). Reaction of 6-8with methyl
triflate or trimethyl- or triethyloxonium tetrafluoroborate
gave the SISs usually in very good yield (Table 1). When the
product precipitated out of the reaction mixture pure SISs were
obtained simply by filtration. It is possible that precipitation
during the reaction is required to drive SIS formation since
SISs 14, 16, and 17 did not precipitate out of the reaction
mixture irrespective of the solvent used (diethyl ether, THF,
CH2Cl2) and did not go to completion even after extended
reaction times. Attempts to selectively precipitate out the SISs
14, 16, and 17 using apolar solvents such as hexane or pentane
were unsuccessful as a semisolid was formed which consisted of
both the starting material and product. In general, SISs having
the triflate counterion were obtained in higher yields than
those having a tetrafluoroborate counterion. In one instance,
we were unable to obtain the BF4- salt (compound 14 which
did not precipitate during the reaction), yet the corresponding
TfO- salt (compound 11) was readily obtained. Several of
us encountered strong allergic reactions when we attempted to
prepare and isolate the TfO- analogues of compounds 15-17,
and so these triflate salts were not pursued any further. All of
the SISs that were obtained in pure form were white powders
and could be stored at 4 °C for months without any detectable
decomposition. Unlike compound 5, all of these SISs exhibited
good solubility in less polar solvents such as methylene chlor-
protecting group and has been used to prepare a broad range
of sulfated compounds.9a-j
Our interest in sulfated biomolecules lies primarily in the
synthesis and study of sulfated carbohydrates, a class of
compounds that play key roles in important biological pro-
cesses such as blot clotting, cell adhesion, and cell-cell com-
munication to name but a few.10 However, we encountered
problems when reagent 1 was used for preparing certain
sulfated carbohydrates. For example, the synthesis of com-
pound 3 from monosaccharide 2 proceeded in modest yield due
to the formation of chloro sugar 4 as a byproduct (Scheme 2).11
To solve this problem, we prepared reagent 5, a new class of
sulfating agent and the first known example of a sulfuryl
imidazolium salt (SIS) (Scheme 3).11 Using reagent 5, TCE-
protected sulfate esters could be introduced into monosacchar-
ides in high yield and withstand many of the conditions that are
commonly employed during the synthesis of complex carbohy-
drates.11 The sulfate group can be deprotected in high yields
under very mild conditions using Zn or Pd/C and ammonium
formate.6,11 Although SIS 5 works well for most sulfations it
does have some limitations. The sulfations must be carried out
in THF as 5 is poorly soluble in less polar solvents such as
CH2Cl2 and chloroform. In THF, its stability is limited and so
for certain difficult sulfations, which can take 24-48 h to
complete, 5 must be replenished during the reaction and,
therefore, a considerable excess of 5 is sometimes required for
high yields.11 Consequently we wish to design SISs with
improved stability, solubility, and sulfating properties. Toward
this end, we report the synthesis and sulfating properties of a
series of sulfuryl imidazolium salts. We demonstrate that by
-
ide and chloroform. SISs 12, 13, and 15, which had the BF4
counterion, were poorly soluble in THF. 1H NMR analysis of
solutions of the SISs in CDCl3 revealed no detectable decom-
position even after several days.
(9) (a) Bunschoten, A.; Kruijtzer, J. A. W.; Ippel, J. H.; de Haas, C. J. C.;
van Strijp, J. A. G.; Kemmink, J.; Liskamp, R. M. J. Chem. Commun. 2009,
2999. (b) Li, X. S.; Robertson, L. W.; Wehmler, H. J. Chem. Cent. J. 2009, 3,
5. (c) Ali, A. M.; Hill, B.; Taylor, S. D. J. Org. Chem. 2009, 74, 3583. (d) Chen,
J.; Yu, B. Tetrahedron Lett. 2008, 49, 1682. (e) Lu, C. P.; Ren, C. T.; Wu, S.
H.; Chu, C. Y.; Lo, L. C. ChemBioChem 2007, 8, 2187. (f) Furstner, A.;
Domostoj, M. M.; Scheiper, B J. Am. Chem. Soc. 2006, 128, 8087.
(g) Yamaguichi, T.; Fukada, T.; Ishibashi, F.; Iwao, M. Tetrahedron Lett.
2006, 47, 3755. (h) Furstner, A.; Domostoj, M. M.; Scheiper, B. J. Am. Chem.
Soc. 2005, 127, 11620. (i) Ahmed, V.; Ispahany, M.; Ruttgaizer, S.; Guillemette,
G.; Taylor, S. D. Anal. Biochem. 2005, 340, 80. (j) Gunnarsson, G. T.; Riaz,
M.; Adams, J.; Desai, U. R. Bioorg. Med. Chem. 2005, 1783.
We chose carbohydrate 18 as a model substrate to test the
sulfating ability of the above SISs because we had previously
found that sulfation of the 4-OH was challenging using
reagent 5. In THF, 8.0 equiv of sulfating agent 5 and
8.5 equiv of 1-methylimidazole (1-MeIm) added over a
48 h period were required to obtain a 76% yield of carbohy-
drate 19. The yield (56%) was even poorer in CH2Cl2
possibly due to the very limited solubility of 5 in this solvent.
To evaluate the new SISs, compound 18 was reacted with
(10) Toida, T.; Chaidedgumjorn, A.; Linhardt, R. J. Trends Glycosci.
Glycotech. 2003, 15, 29.
(11) Ingram, L. J.; Taylor, S. D. Angew. Chem., Int. Ed. Engl. 2006, 3503.
6480 J. Org. Chem. Vol. 74, No. 17, 2009

Ingram et al.
JOCFeatured Article
TABLE 1. Preparation of Sulfuryl Imidazolium Salts 9-13 and 15
compd
R1
R2
X-
% yield
99a
9
Me
Et
iPr
Me
Et
iPr
Me
Et
Me
Me
Me
Me
Me
Me
Et
TfO-
TfO-
TfO-
10
11
12
13
14
15
16
17
98a
85a
BF4-
85b
-
BF4-
BF4-
BF4-
BF4-
BF4
79b
NDd
81c
Et
Et
NDd
NDd
IPr
aSolvent: dry diethyl ether. bSolvent: dry THF. cSolvent: dry CH2Cl2. dND = not determined. Mixture of starting material and product as semisolids
was obtained.
TABLE 2. Sulfation of Carbohydrate 18 with Reagents 5 and 9-13 and 15
1-methyl-2-ethylimidazole (1-Me-2-EtIm)) that was the
same as the leaving group of the respective SISs (entries 6,
8, 10, 12, and 15). However, if these reactions were per-
formed using 1,2-dMeIm as base then the yields improved
considerably (entries 7, 9, 11, 13, and 16). 1H NMR studies in
CDCl3 revealed that just 1 equiv of 1,2-dMeIm rapidly
displaced the 1-Me-2-EtIm or 1-Me-2-iPrIm from SISs 10
and 11 within minutes, thus forming SIS 9 in situ which is a
entry
SIS
base
1-MeIm
solvent
% yield of 19
1
better sulfating agent. Even after several hours, H NMR
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
5
9
9
9
DMF
THF
CH2Cl2
CH2Cl2
DMF
THF
THF
CH2Cl2
CH2Cl2
THF
35
65
80
53
40
18
58
21
68
18
54
30
66
79
38
78
45
75
provided no evidence that the reverse reaction back to SISs
10 and 11 was occurring, indicating that SIS 9 is more stable
than SISs 10 and 11 possibly due to steric hindrance between
the ethyl or isopropyl group at the 2-position and the
sulfonate moiety and methyl group. We also found that
a reduced yield of 19 was obtained using SIS 9 and 1-Me-2-
iPrIm in CH2Cl2 as the base (entry 4), revealing that the low
yields encountered with SISs 10, 11, and 13 with 1-Me-2-iPrIm
or 1-Me-2-EtIm may in part be due to the added imidazole
derivative itself, which may be acting as a general base during
the reaction. It is possible that there is greater steric crowding at
the transition state of the reaction with SISs 10, 11, and 13 and
1-Me-2-iPrIm or 1-Me-2-EtIm than with SISs 9 or 12 and 1,2-
dMeIm. SIS 15, which differs from SIS 12 only by the presence
of an ethyl rather than methyl group on one of the nitrogens,
gave lower yields than SIS 12 (entry 14) when 1-ethyl-2-methyl
imidazole (1-Et-2-MeIm) was used as base but almost the same
yield when 1,2-dMeIm was used as base (entries 17 and 18), again
suggesting that the 1-Et-2-MeIm group in SIS 15 was exchang-
ing with 1,2-dMeIm. We also examined DMF as a solvent for
SISs 5 and 9, but low yields were obtained (entries 1 and 5).
1,2-dMeIm
1,2-dMeIm
1-Me-2-iPrIm
1,2-dMeIm
1-Me-2-EtIm
1,2-dMeIm
1-Me-2-EtIm
1,2-dMeIm
1-Me-2-iPrIm
1,2-dMeIm
1-Me-2-iPrIm
1,2-dMeIm
1,2-dMeIm
1-Me-2-EtIm
1,2-dMeIm
1-Et-2-MeIm
1,2-dMeIm
9
10
10
10
10
11
11
11
11
12
13
13
15
15
THF
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
2 equiv of SIS in the presence of 2.5 equiv of base in various
solvents for 20 h, the reaction was stopped, whether or not it
was complete, and the yield of 19 was determined after
purification (Table 2). For SISs 9-11, higher yields were
obtained in CH2Cl2 compared to THF. We were unable to
evaluate SISs 12-15 in THF due to their limited solubility in
this solvent. SISs 9 and 12, which bear a 1,2-dimethylimida-
zolium group, gave good yields using 1,2-dimethylimidazole
(1,2-dMeIm) as the base (entries 3 and 14). The counterion
(TfO- or BF4-) had no effect on yield. Low yields were
obtained when the SIS contained an ethyl or isopropyl
moiety at the 2-position of the imidazolium group (SISs 10,
11, and 13) but only if these reactions were performed using
a base (1-methyl-2-isopropylimidazole (1-Me-2-i-PrIm) or
SCHEME 4. Sulfation of Carbohydrate 20 in DMF and THF
J. Org. Chem. Vol. 74, No. 17, 2009 6481

Ingram et al.
JOCFeatured Article
However, we found that DMF can be used as a solvent for less
challenging sulfations. For example, 1:2,3:4-di-O-isopropylidene
galactose (20) is sulfated with 2 equiv of reagent 5 and 2 equiv of
1-MeIm in very good yields in either THF or DMF (Scheme 4).
Since SIS 9 can be readily prepared in very high yield and
the best yields in Table 1 were obtained with this compound,
we decided to use this reagent for subsequent studies.
Further studies with SIS 9 and carbohydrate 18 revealed
that an 88% yield of 19 could be obtained in 24 h using just
3 equiv of reagent 9 and 4.3 equiv of 1,2-dMeIm in CH2Cl2
(Scheme 5). We had previously prepared carbohydrate 23 in
90% yield after subjecting compound 22 to 10.5 equiv of SIS
5 and 11.6 equiv of 1-MeIm for 28 h in THF (Scheme 5).11
sulfate compound 24, which is the STol analogue of 22,
with either SIS 5 or 9, and we were unable to isolate the
desired sulfated compound 25. This was unexpected since we
had previously reported the sulfation of the 4-OH of p-tolyl
2,3-di-O-benzoyl-6-O-benzyl-1-thio-β-D-glucopyranoside in
high yield using reagent 5.11 It is possible that upon sulfation
of compound 24 intramolecular displacement of the TCE
sulfate group by the sulfur atom occurs resulting in the
formation of a reactive episulfonium ion.
Since N-sulfation is a common feature in heparin-like
glycosaminoglycans, we also examined the ability of SISs
to N-sulfate glycosamines. Recently, Chen and Yu reported
the synthesis of compound 27 (Table 3) in 82% yield by the
addition of a solution of compound 26 and 3 equiv of Et3N in
DMF over 1 h to a solution of 6 equiv of reagent 1 and
1 equiv of DMAP in DMF.9d The best yield of 27 that we
were able to obtain using these conditions was 45%, and
half of the product was always dimer 28, a byproduct not
mentioned in Chen and Yu’s original report (entry 1, Table 3).
We performed this reaction under these conditions numer-
ous times, and in each instance 27 and 28 were formed in
40-45% yield. Performing the reaction under the same
conditions in methylene chloride gave exclusively dimer 28
in a 70% yield. Attempts to prepare compound 27 using SIS
9 also gave disappointing results with the best yield of 27
being only 29% when performing the reaction with 6.0 equiv
of 9 in the presence of 3.5 equiv of 1,2-dMeIm in methylene
chloride (entry 3, Table 3). Performing the reaction in DMF
gave a lower yield (entry 4), though in neither case was dimer
28 formed probably because a stronger base such as Et3N is
necessary for elimination of trichloroethanol from 27 which
leads to dimer formation. Performing the reaction with
4 equiv of SIS 9 and 1,2-dMeIm in the presence of 1 equiv
of Et3N or in pyridine with 1 equiv of Et3N gave only trace
amounts of the desired product.
SCHEME 5. Sulfation of Carbohydrates 18, 22, and 24 with
Reagents 5 and/or 9
However, when 4 equiv of SIS 9 and 4.6 equiv of 1,2-dMeIm
in CH2Cl2 were used, carbohydrate 23 was obtained in a 96%
yield after just 24 h. Performing the reaction under the same
conditions in THF gave a 40% yield of 23 with unreacted
22 still remaining after 24 h, demonstrating that these reac-
tions can be subject to a significant solvent effect. A complex
mixture of products was obtained when we attempted to
To determine whether other carbohydrates would under-
go N-sulfation more readily than carbohydrate 26, we ex-
amined carbohydrates 29 and 30 as substrates (Table 4)
which were prepared as stable free amines. Subjecting 29 to
reagent 1 and the conditions of Chen and Yu9d did not result
in the formation of N-sulfated product 31, although dimer
TABLE 3. N-Sulfation of Carbohydrate 26
entry
reagent (equiv)
base (equiv)
solvent
% yield of 27
% yield of 28
1
2
3
4
5
6
7
1 (6)
1 (6)
9 (6)
9 (6)
9 (6)
9 (4)
9 (4)
Et3N (3.5) DMAP (1)
Et3N (3.5) DMAP (1)
1,2-dMeIm (3.5)
1,2-dMeIm (3.5)
Et3N (2.5)
DMF
45a
0a
45a
70a
0b
CH2Cl2
CH2Cl2
DMF
CH2Cl2
CH2Cl2
pyridine
29b
18 b
0c
0b
68c
trace
trace
Et3N (1) 1,2-dMeIm (4)
Et3N (1)
trace
trace
a26 and NEt3 (3.5 equiv) in solvent (0.1 M wrt 26) were added at 0 °C over 1 h to a solution of 1 (6.0 equiv) and DMAP (1.0 equiv) in solvent (2.0 M wrt
1). b1,2-dMeIm (3.5 equiv) followed by 9 (6.0 equiv) were added to 26 in solvent (0.05 M). cEt3N (2.5 equiv) followed by 9 (3.0 equiv) were added to 26 in
CH2Cl2 (0.15 M). Additional 9 (3.0 equiv) was added after 12 h.
6482 J. Org. Chem. Vol. 74, No. 17, 2009

Ingram et al.
JOCFeatured Article
TABLE 4. N-Sulfation of Carbohydrates 29 and 30
entry
substrate
reagent (equiv)
base (equiv)
solvent
% yield of 31 or 32
% yield of 33
1
2
3
4
5
6
7
8
9
29
29
29
29
29
29
30
30
30
1 (6)
5 (2)
5 (6)
5 (6)
9 (6)
9 (6)
1 (2)
5 (6)
9 (4)
Et3N (3.5) DMAP (1)
1-MeIm (2.5)
1-MeIm (4)
DMF
THF
THF
CH2Cl2
THF
CH2Cl2
THF
THF
0a
70a
0b
traceb
traceb
traceb
94b
0b
1-MeIm (4)
0b
1,2-dMeIm (4)
1,2-dMeIm (4)
1,2-dMeIm (4)
1-MeIm (4)
0b
95b
0b
60c
NAd
NAd
NAd
9b
1,2-dMeIm (4)
THF
94b
aA solution of 29 and Et3N (3.5 equiv) in DMF (0.1 M wrt 29) was added dropwise over 1 h to a solution of 1 and DMAP (1.0 equiv) in DMF (1.8 M
wrt 1). b5 or 9 was added to a solution of 1-MeIm or 1,2-dMeIm and 29 or 30 in THF (0.23 M) at 0 °C. The reaction was gradually warmed to rt, 8-16 h.
cCompound 1 was added to a solution of 1,2-dMeIm and 30 in THF (0.23 M) at 0 °C. The reaction was gradually warmed to rt, 8 h. dNA: not applicable.
TABLE 5. Sulfation of Amines 34-37
SCHEME 6. Deprotection of Carbohydrates 31 and 32
substrate
reagent (equiv) product % yielda
34 (R0 = H, R = cyclohexyl)
9
5
9
5
9
9
38
38
39
39
40
41
99
13
95
14
90
98
35 (R0 = H, R = benzyl)
36 (R0 = Me, R = butyl)
37 (R = R0 = -(CH2)5-)
aTHF (0.23 M), 2.5 equiv of 1-MeIm or 1,2-dMeIm, and 2.0 equiv of
SCHEME 7. Reaction of Carbohydrates 44, 45, and 48 with
Reagent 9
5 or 9, 8-14 h.
33 was isolated in a 70% yield (entry 1). Sulfations using
reagent 5 and 1-MeIm were also unsuccessful in that a
complex mixture of products were formed (as determined
by TLC) and only trace quantities of the desired product 31
was formed under a variety of conditions (entries 2-4).
However, subjecting 29 to 6 equiv of reagent 9 and 4 equiv
of 1,2-dMeIm in either THF or CH2Cl2 gave compound 31 in
a 94-95% yield (entries 5 and 6). Reaction of 30 with 2 equiv
of reagent 1 in the presence of 4 equiv of 1,2-dMeIm resulted
in complete consumption of 30 within 6 h; however, product
32 was isolated in only a 60% yield (entry 7). Similarly,
sulfation of carbohydrate 30 with reagent 5 gave only 9% of
the desired product 32, while STS 9 under the same condi-
tions gave carbohydrate 32 in a 94% yield (entries 8 and 9).
This difference in reactivity between SISs 5 and 9 is not
limited to just carbohydrate N-sulfations. Sulfation of
amines 34 and 35 with reagent 9 gave the desired sulfated
products 38 and 39 in excellent yields, while very poor
yields were obtained using reagent 5 (Table 5). N-Sulfa-
tion of secondary amines with reagent 9 also proceeded
in very high yields (Table 5). We do not have a clear
understanding as to why SISs 5 and 9 behave so differently
with amines. A reaction between the amino group and
SIS 5 may be taking place such as attack of the amino
group at the 2-position of SIS 5 or deprotonation of the
proton at C-2 of SIS 5.12
The sulfate group in carbohydrate 31 was deprotected to
give carbohydrate 42 in 84% yield using Zn/ammonium
formate, while carbohydrate 32 was completely deprotected
using Pd(OH)2, H2, and ammonium formate (Scheme 6) to
give carbohydrate 43 in a 93% yield.
We also attempted to disulfate carbohydrates 44 and 45
using reagent 9 (Scheme 7). These reactions proceeded
(12) We were unable to examine the behavior of SIS 5 with amines by
NMR since SIS 5 is poorly soluble in CDCl3.
J. Org. Chem. Vol. 74, No. 17, 2009 6483

Ingram et al.
JOCFeatured Article
slowly, and a considerable excess of the SIS was required
before all of the 44 and 45 were consumed. Moreover, none
of the desired disulfated products were isolated, but instead
aziridines 46 and 47 were obtained in 55-60% yields.
Surprisingly, reaction of the Ν-benzyl derivative of 45,
compound 48, with 2.7 equiv of reagent 9 gave the O-sulfated
product 49 in a 65% yield, and no N-sulfated, disulfated, or
sulfated aziridine product was formed even after the addition
of a considerable excess of the sulfating agent (Scheme 7).
(0.100 g, 1.0 mmol) followed by reagent 9 (0.400 g, 0.874 mmol).
The reaction was gradually brought to rt and stirred for 16 h.
Analysis by TLC indicated remaining starting material; thus,
additional 1,2-diMeIm (0.100 g, 1.0 mmol) and sulfating agent
9 (0.4 g, 0.874 mmol) were added. The reaction was stirred for an
additional 8 h, diluted with dichloromethane (1.0 mL), and
quenched with water (1.0 mL). The organic layer was separated,
dried over MgSO4, and concentrated to a brown crude oil.
Purification by flash chromatography (25:75 EtOAc/hexanes)
afforded 23 as a white solid (0.278 g, 96%): mp 105-107 °C;
[R]25 = -34.6 (c 1.0, CHCl3);1H NMR (300 MHz, CDCl3) δ
D
Conclusions
3.57 (ddd, 1H, J5,6ax = 9.7, J5,4 = 9.5, J5,6eq = 5.0 Hz, H5), 3.80
(s, 3H, CH3), 3.86 (dd, 1H, J6eq,6ax = J6ax,5 =10.4 Hz, H6ax), 3.88
(dd, 1H, J4,5 = J4,3 = 9.2 Hz, H4), 3.98 (dd, 1H, J3,4 = J3,2 = 9.1
Hz, H3), 4.43 (dd, 1H, J6eq,6ax = 10.5, J6eq,5 = 5.0 Hz, H6eq),
4.68, 4.71 (AB, 2H, J = 11.1 Hz, CH2CCl3), 4.85 (m, 1H, H2),
4.87, 5.01 (AB, 2H, J = 11.2 Hz, CH2Ph), 5.02 (d, 1H, J1,2 = 7.8
Hz, H1), 5.61 (s, 1H, CHPh), 6.85-6.88 (m, 2H, ArH), 7.06-7.09
(m, 2H, ArH), 7.33-7.50 (m, 10H, ArH); 13C NMR (75 MHz,
CDCl3) δ 55.5, 66.2, 66.30, 74.6, 77.9, 79.9, 81.2, 83.9, 92.6, 99.9,
101.4, 114.6, 118.8, 125.9, 127.9, 128.1, 128.1, 128.3, 128.3, 128.3,
128.4, 129.1, 136.6, 137.1, 150.2, 156.0; HRMS (EIþ) m/z =
674.0546, C29H29Cl3O10S requires 674.0547.
In conclusion, we have prepared a series of sulfuryl imida-
zolium salts (SISs) and examined them as reagents for in-
corporating TCE-protected sulfate esters into carbohydrates.
We demonstrated that by incorporating a methyl group at the
2-position of the imidazolium ring a more efficient sulfating
agent, SIS 9 was obtained. O-Sulfations that required pro-
longed reaction times and a large excess of the original SIS,
compound 5, were more readily achieved using SIS 9. Certain
N-sulfated compounds that that were practically inaccessible
using SIS 5 couldbeobtained in excellent yieldusing SIS9. We
expect that this next generation of sulfating agent will find
widespread use in the preparation of sulfated carbohydrates
and other organosulfates.
Representative N-Sulfation Using Reagent 9 (Compound 32,
Table 4, Entry 9). To a solution of 30 (0.10 g, 0.252 mmol) in
THF (1.1 mL, 0.23 M) at 0 °C was added 1,2-dMeIm (0.06 g,
0.62 mmol) followed by reagent 9 (0.460 g, 1.01 mmol). The
reaction was stirred at 0 °C, gradually warmed to room tempera-
ture by allowing the ice bath to melt, and then stirred overnight.
After 14 h, the mixture was applied directly to a silica gel column.
Flash chromatography (33:67 EtOAc/hexanes) gave pure 32 as an
Experimental Section
Representative Procedure for the Preparation of Compounds
6-8 (Table 1, Compound 6). To a solution of 2-methylimidazole
(5.9 g, 0.072 mol, 3.60 equiv) in dry THF (40 mL) at 0 °C was
added dropwise a solution of reagent 1 (5.0 g, 0.02 mol, 1.0
equiv) in THF (50 mL). The reaction was stirred at 0 °C for 1 h,
warmed to room temperature, and stirred for an additional
1 h. The reaction mixture was filtered, the residue was washed
with THF, and the filtrate was concentrated under vacuum. The
crude residue was purified by flash chromatography (33:67
EtOAc/hexanes) to give 6 as a white solid (5.2 g, 88%): mp 53-
55 °C; 1H NMR (300 MHz, CDCl3) δ 7.32 (s, 1H, Himi), 6.94 (s,
1H, Himi), 4.65 (s, 2H, CH2), 2.67 (s, 3H, CH3-imi); 13C NMR
(75 MHz, CDCl3) δ 146.4, 128.2, 120.1, 91.7, 80.0, 14.9; HRMS
(EIþ) calcd for C6H7Cl3N2O3S (M)þ 291.9243, found 291.9244.
Representative Procedure for the Preparation of Sulfuryl Imida-
zolium Triflate Salts (Table 1 Compound 9). To a solution of
compound 6 (4.4 g, 15 mmol, 1.0 equiv) in dry Et2O (70 mL) at
0 °C was added methyl triflate (1.8 mL, 15 mmol, 1.0 equiv)
dropwise . The reaction was stirred for 3 h at 0 °C during which
time a white precipitate formed. The mixture was filtered. The
filter cake was washed with cold ether, and the filtrate was cooled
to - 20 °C and filtered. This second precipitate was washed with
cold ether and then combined with the first precipitate, which
afforded compound 9 as a white solid (6.8 g) in 99% yield. We
have prepared compound 9 in batches up to 75 g in excellent yield.
We typically store compound 9 at 4 or -20 °C and have never
detected any decomposition even after 6 months. We have also
stored reagent 9 on the benchtop at room temperature for a
month and not detected any decomposition: 1H NMR (300 MHz,
CD3OD) δ 8.09 (d, 1H, J = 2.1 Hz, Himi), 7.74 (d, 1H, J = 1.8Hz,
amorphous white solid (0.144 g, 94%): [R]25 = þ71.5 (c 1.0,
D
CHCl3); 1H NMR (300 MHz, CDCl3) δ 3.68 (br-ddd, 1H,
J2,NH = J2,3 = 8.8 Hz, J2,1 = 3.4 Hz, H2), 3.73-3.94 (m, 4H,
H3, H4, H5, H6ax), 4.08 (dd, 1H, JH,H = 12.6, 6.4 Hz, OCH2-
CHCH2), 4.25 (dd, 1H, JH,H = 12.6, 5.5 Hz, OCH2CHCH2), 4.32
(dd, 1H, J6eq,6ax = 9.9 Hz, J6eq,5 = 4.3 Hz, H6eq), 4.58, 4.62 (AB,
2H, J = 10.8 Hz, CH2CCl3), 4.76, 4.99 (AB, 2H, J = 11.5 Hz,
CH2Ph), 5.00 (br-d, JNH,2 = 8.8 Hz, NH), 5.15 (d, 1H, J1,2 = 3.7
Hz, H1), 5.27-5.37 (m, 2H, OCH2CHCH2), 5.62 (s, 1H, CHPh),
5.88-6.01 (m, 1H, OCH2CHCH2), 7.28-7.52 (m, 10H, ArH); 13
C
NMR (75 MHz, CDCl3) δ 57.8, 62.7, 68.8, 69.0, 75.0, 75.5,
78.3, 82.8, 92.9, 96.9, 101.37, 118.8, 126.0, 128.0, 128.2, 128.3,
128.6, 129.1, 132.9, 137.1, 137.7; HRMS (ESIþ) m/z = 608.0665,
C25H29NO8Cl3S (M þ H) requires 608.0679.
p-Tolyl 3-O-Benzyl-4,6-O-benzylidene-2-deoxy-2-sulfoxyami-
no-1-thio-β-D-glucopyranoside (42). To a suspension of ammo-
nium formate (0.083 g, 1.3 mmol) in HPLC-grade MeOH (1.3 mL,
1.0 M) was added 31 (0.15 g, 0.22 mmol) followed by zinc dust
(0.1 g, 1.55 mmol). The reaction was stirred for 7 h at room
temperature, at which point no starting material was detected using
TLC. The reaction was filtered through Celite and concentrated to
crude product. Flash chromatography (20:4:1 CH2Cl2/MeOH/
NH4OH) afforded a white solid, which was lyophilized (3ꢀ) from
water to yield 42 as a white powder (0.105 g, 84%): [R]25D = -78.0
(c 0.85, DMSO); 1H NMR (500 MHz, DMSO-d6) δ 2.29 (s, 3H,
CH3), 3.01 (dd, 1H, J2,3 þ J2,1= 17.1 Hz, H2), 3.39-3.46 (m, 1H,
H5), 3.65 (dd, 1H, J6ax,6eq þ J6ax,5 = 10.2 Hz, H6ax), 3.71 (dd, J4,5
þ J4,3= 18.3 Hz, H4), 4.20 (dd, 1H, J6eq,6ax= 10.2, J6eq,5 = 5.1 Hz,
H6eq), 4.39 (dd, J3,4 =J3,2 = 8.5 Hz, H3), 4.74 (d, 1H, J = 11.5 Hz,
1/2 CH2Ph), 4.97 (d, 1H, J = 11.5 Hz, 1/2 CH2Ph), 5.59 (d, 1H,
J = 9.1 Hz, H1), 5.66 (s, 1H, CHPh), 5.76 (s, 1H, NH), 7.14-7.40
(m, 18 H, 14 ArH þ NH4); 13C NMR (125 MHz, DMSO-d6) δ
21.1, 59.8, 68.5. 68.9, 74.0, 78.9, 81.7, 87.0, 100.6, 126.4, 126.4,
127.4, 128.2, 128.3, 128.5, 130.0, 131.2, 131.8, 138.3, 140.1; HRMS
(ESI-) m/z = 542.1306, C27H28NO7S2 requires 542.1307.
H
imi), 5.35 (s, 2H, CH2), 3.92 (s, 3H, CH3-imi), 2.91 (s, 3H, CH3-imi);
13C NMR (75 MHz, CD3OD) δ 148.6, 123.52, 120.8, 120.4 (q,
J
CF = 316.5 Hz, CF3), 91.6, 82.0, 35.3, 10.5; 19F NMR (282 MHz,
CD3OD) δ -79.8; HRMS (þESI) calcd for (M - OTf)þ C7H10-
Cl3N2O3Sþ 306.9478, found 306.9469.
Representative O-Sulfation Using Reagent 9 (Scheme 5, Com-
pound 23). To carbohydrate 22 (0.200 g, 0.431 mmol) in dry
dichloromethane (2.7 mL) at 0 °C was added 1,2-diMeIm
Propyl 2-Deoxy-2-sulfoxyamino-r-D-glucopyranoside (43).
To a solution of compound 42 (61 mg, 0.1 mmol) in MeOH
6484 J. Org. Chem. Vol. 74, No. 17, 2009

Ingram et al.
JOCFeatured Article
(2 mL) was added ammonium formate (38 mg, 0.6 mmol)
and the mixture stirred until all of the ammonium formate
dissolved. Pd(OH)2 (20 mg, 33 wt %) was added and the mix-
ture stirred over an atmosphere of H2 (balloon) for 16 h. The
mixture was filtered through Celite and the filtrate concen-
trated. The filtrate was dissolved in MeOH and concentra-
ted again, and this was repeated two more times. The residue
was dissolved in MeOH (2 mL), 10 mg of Pd(OH)2 was added,
and the mixture was stirred over an atmosphere of H2 (balloon)
for 16 h. The mixture was filtered through Celite and the filt-
rate concentrated. The residue was dissolved in water and
passed through a small Dowex 50 Naþ ion-exchange column
which gave compound 43 as a white powder (28 mg, 93%):
71.6, 97.2; HRMS (ESI-) m/z = 300.0753, C9H18NO8S requires
300.0755.
Acknowledgment. This research was supported by a Dis-
covery Grant from the Natural Sciences and Engineering
Research Council (NSERC) of Canada to S.D.T. We thank
NSERC for a postgraduate scholarship for L.J.I. and the
Egyptian Ministry of Higher Education for scholarships to
A.D. and A.M.A. We thank J. Chen and B. Yu for sending us
a detailed procedure for their preparation of compound 27
using reagent 1.
[R]25 = þ89.3 (c 1.4, H2O); 1H NMR (300 MHz, D2O) δ
Supporting Information Available: Experimental proce-
dures and characterization data for all novel compounds except
compounds 6, 9, 23, 32, 42, and 43. 1H and 13C NMR spectra for
all compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
D
0.81 (t, 3H, J = 7.4 Hz, CH3), 1.52 (m, 2H, CH2CH3) 3.11 (dd,
1H, J1,2= 3.0 Hz, J2,3 = 10.2 Hz, H2), 3.53 (m, 7H, H3, H4, H5,
H6, H60, OCH2-), 5.01 (d, 1H, J1,2 = 3.0 Hz, H1); 13C NMR
(125 MHz, D2O) δ 10.0, 22.1, 57.8, 60.6, 70.1, 70.2, 71.5,
J. Org. Chem. Vol. 74, No. 17, 2009 6485