Preparation of D-galactofuranosyl nitromethanes: A revision and a new approach

Article abstract of DOI:10.1016/j.carres.2011.02.007

Sodium methoxide-promoted methanolysis of 7-deoxy-7-nitro-l-glycero-l- galacto-heptitol peracetate rapidly and nearly quantitatively accumulates 7-deoxy-6-O-methyl-7-nitro-l-glycero-l-galacto-heptitol. The prolonged treatment then provides 76% of d-galactofuranosyl nitromethanes and finally results in the equilibrium of 77% of β-d-galactopyranosyl nitromethane and 7-9% of three other tautomeric d-galactosyl nitromethanes. Thermal treatment of 7-deoxy-7-nitro-l-glycero-l-galacto-heptitol in boiling water peaks at a 58% content of d-galactofuranosyl nitromethanes and ends in a similar equilibrium mixture of four d-galactosyl tautomers. The relevant kinetic parameters of the latter transformation are determined by a curve fitting using the nonlinear least-squares Marquardt-Levenberg algorithm.

Full text of DOI:10.1016/j.carres.2011.02.007

Carbohydrate Research 346 (2011) 715–721
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
Preparation of D-galactofuranosyl nitromethanes: a revision
and a new approach
a
a
b
a
a,
⇑
ˇ
Michal Vojtech , Mária Petrušová , Ivan Valent , Bozena Pribulová , Ladislav Petruš
a Institute of Chemistry, Center for Glycomics, GLYCOMED, Slovak Academy of Sciences, SK-84538 Bratislava, Slovakia
b Department of Physical and Theoretical Chemistry, Faculty of Natural Sciences, Comenius University, SK-84215 Bratislava, Slovakia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 March 2009
Received in revised form 29 January 2011
Accepted 4 February 2011
Available online 1 March 2011
Sodium methoxide-promoted methanolysis of 7-deoxy-7-nitro-
L
-glycero-
L
-galacto-heptitol peracetate
-glycero- -galacto-heptitol.
-galactofuranosyl nitromethanes and finally results in
-galactopyranosyl nitromethane and 7–9% of three other tautomeric
galactosyl nitromethanes. Thermal treatment of 7-deoxy-7-nitro- -glycero- -galacto-heptitol in boiling
water peaks at a 58% content of -galactofuranosyl nitromethanes and ends in a similar equilibrium
mixture of four -galactosyl tautomers. The relevant kinetic parameters of the latter transformation
rapidly and nearly quantitatively accumulates 7-deoxy-6-O-methyl-7-nitro-
The prolonged treatment then provides 76% of
the equilibrium of 77% of b-
L
L
D
D
D-
L
L
D
Keywords:
D
Anhydronitroalditol
Glycosyl nitromethane
1-Deoxy-2-O-methyl-1-nitroalditol beta-
elimination
are determined by a curve fitting using the nonlinear least-squares Marquardt–Levenberg algorithm.
Ó 2011 Elsevier Ltd. All rights reserved.
1-Deoxy-1-nitrohept-1-enitol
Tautomerization
Aldose mutarotation
1. Introduction
modynamically favored equatorial pyranosyl anomer 10. Even the
catalytically induced elimination of the pertinent C-2-OH group
Recently, efficient syntheses of C-glycofuranosyl compounds,
at ambient temperature in the
L-fucose-derived 1-deoxy-1-
especially those of the D-galacto and D-arabino configurations, have
nitroalditols has not been reported to provide the corresponding
become more desirable.1–3 The reason for this is that an extensive
search for new, specific agents that are active against mycobacte-
rial pathogens has emerged with the appearance of strains
resistant to classical drugs.4,5 Their cell wall polysaccharides, con-
6-deoxy-L-galactofuranosyl nitromethanes as primary cyclization
products.16 The prevalence of ribofuranosyl nitromethanes in the
near-equilibrium reaction mixture,17 due to obvious steric reasons
in the ribopyranosyl ring, is the only known exception among the
methods for the preparation of glycosyl nitromethanes. Using the
taining both D-galactofuranose and D-arabinofuranose, which are
endemic to the pathogenic microorganisms,6–8 have been identi-
fied as possible target structures for the development of new drugs
for treating these diseases.9–13 Easily available glycofuranosyl
nitromethanes could be a convenient starting material in this
respect as well.
desired
D-galacto model, which also implicitly encloses the homo-
morphic
L-arabino structure, we report a new, non-thermodynamic
approach to the synthesis of glycofuranosyl nitromethanes. We
also report a reinvestigation of the classic, thermally induced b-
elimination of nitroalditol 2, which, under the properly selected
reaction conditions, can also provide preparatively interesting
As exemplified by the D-galactosyl species, Figure 1 shows the
present state of synthetic methods for the preparation of all four
tautomeric forms of glycosyl nitromethanes 7–10, which are based
on the thermally induced b-elimination of the C-2-OH group from
the tautomeric, aci-nitro form of 1-deoxy-1-nitroalditols 4 and
5.14,15 However, these approaches have never been optimized for
capturing the first, primarily formed five-membered ring glycofu-
ranosyl nitromethanes 7 and 8 as the kinetic products of the cycli-
zation of the intermediate 1,2-dideoxy-1-nitroald-1-enitol 6 and,
until now, have been exploited mostly for preparation of the ther-
yields of D-galactofuranosyl nitromethanes 7 and 8.
2. Results and discussion
2.1. Et3N-catalyzed methanolysis of 1,2-dideoxy-1-nitro-D-
galacto-hept-1-enitol peracetate (11)
The recently described acid-catalyzed methanolysis of 1,2-
dideoxy-1-nitro- -arabino-hexenitol peracetate, which does not
L
stop at the stage of anticipated glycofuranosyl nitromethanes but
proceeds further with the formation of the corresponding
methanal dimethyl acetals,3 has primarily excluded the acidic
⇑
Corresponding author. Fax: +421 2 59410 222.
E-mail address: chemlpet@savba.sk (L. Petruš).
0008-6215/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2011.02.007

716
M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
O H
O
OH
OH
O H
O
NO2
NO2
OH
N
N
O
HO
OH
OH
OH
HO
1. Nitromethane
addition
HO
HO
OH
HO
HO
HO
HO
+
+
HO
HO
HO
HO
2. Decationization
H2O
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
1
2
3
4
5
Parent aldose
Nitro forms
Hydrogen nitronate (aci-nitro) forms
of 1-deoxy-1-nitroalditols
of 1-deoxy-1-nitroalditols
HO
NO2
O
HO
Thermal treatment
or
acido-basic catalysis
OH
HO
OH
7
HO
HO
OH
Pseudoaxial furanosyl anomer
(minor)
O
NO2
O
HO
HO
HO
NO2
NO2
OH
9
Axial pyranosyl anomer
(minor)
OH
HO
OH
HO
HO
HO
OH
OH
NO2
O
O
OH
HO
HO
NO2
HO
6
OH
HO
NO2
Intermediate
1,2-dideoxy-1-nitroald-1-enitol
10
O
Equatorial pyranosyl anomer
(major)
HO
OH
HO
8
Pseudoequatorial furanosyl anomer
(minor)
Glycosyl nitromethanes
Figure 1. Schematic representation of present methods of preparing glycosyl nitromethanes, as exemplified with
D-galactosyl species.
and a small amount of D-galactopyranosyl nitromethanes 9 and
10, unwanted 2-O-methylated acyclic nitroalditols 14 and 15,
characteristic of the 13C NMR spectrum by their respective
O-methyl group signals at d 59.1 and d 59.6, (Scheme 2; Table
1, entry a). With a prolonged 7-day treatment of the reaction
mixture, methyl ethers 14 and 15 completely disappeared, and
the content of the furanosyl derivatives 7 and 8 increased to
70% (Table 1, entry b). Practically the same course in the methan-
NO2
OBn
OBn
NO2
O
Et3N
BnO
BnO
BnO
CH2Cl2-MeOH-H2O
OBn
OAc
OBn
OBn
olysis was observed, and final yields of the individual D-galactosyl
13
12
nitromethanes were obtained in a partly heterogeneous treat-
ment of 11 with Et3N in a starting MeOH–H2O suspension (Table
1, entries c and d).
Scheme 1.
The disappearance of 1-deoxy-2-O-methyl-1-nitroheptitols 14
and 15, which temporarily appeared in the reaction mixture and
formed by the Et3N-catalyzed addition of MeOH to the starting
nitroalkene 11, proceeds via Et3N-catalyzed b-elimination of their
2-O-methyl group from the respective nitronates 16 and 17 under
conditions for the de-O-acetylation of compound 11. Due to the
preferred Michael-like addition of the methoxide ion to 1,2-
dideoxy-1-nitro-ald-1-enitol peracetates, which is a part of a
well-established procedure for the preparation of 2-O-methyl
aldoses by the Sowden procedure,18 the classic Zemplén method
does not seem applicable to the de-O-acetylation of nitroalkene
11. Therefore, the basic reaction conditions reported by Martin
et al.19 for the de-O-acetylation of nitroalkene 12, the product of
which is cyclized into C-glycosyl nitro compound 13 (Scheme 1),
were attempted for this purpose.
final formation of D-galactosyl nitromethanes 7–10 via the compet-
itive ring-closure reactions of the intermediate nitroalkene 6
(Scheme 3). The newly observed elimination of the 2-O-methyl
group from the 1-deoxy-2-O-methyl-1-nitroalditol derivatives is
similar to the base-catalyzed elimination of the 3-O-substituent
from 3-O-substituted aldoses, which is a generally well-known
transformation in carbohydrate chemistry and is an another man-
ifestation of certain similarities in the reactivity of the compounds
containing nitro and carbonyl groups.
Despite the anticipation, a 3-day treatment of nitroalkene (11)
with Et3N as a base in a CH2Cl2–MeOH–H2O solution afforded, in
addition to the desired D-galactofuranosyl nitromethanes 7 and 8

M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
717
NO2
OMe
OH
NO2
OH
NO2
OAc
MeO
Et3N
MeOH
HO
HO
HO
HO
AcO
AcO
+
7
+
8
+
9
+
10
+
OH
OH
OH
OH
OAc
OAc
11
14
15
Scheme 2.
Table 1
rect intermediate of the anomeric and tautomeric
D
-galactosyl
Compositiona of the reaction mixture obtained upon treatment of 1,2-dideoxy-1-
nitromethanes 7–10 (Scheme 4). The first mole of methoxide was
used for the deprotonization of 18 to form nitronate 19. The gener-
ation of the conjugate base 19 was a prerequisite to the following
b-elimination of its 2-O-acetyl group (as sodium acetate) to pro-
vide the still per-O-acetylated nitroalkene 11, which was immedi-
ately able to consume the second mole of sodium methoxide via a
Michael-like addition, giving the sodium nitronate forms of the
acetylated O-methyl ethers 20 and 21. Apparently, all of these
transformations, due to the relatively high acidity of the nitroaldi-
tol CH2NO2 group (pKa = 8.8–9.2),20 are substantially faster than
the catalytic Zemplén de-O-acetylation of the other O-acetyl
groups of 18.
nitro-D-galacto-hept-1-enitol peracetate (11) with Et3N at rt
Entry Reaction time (d) Solvent mixture
Compound (%)
10 14 15
7
8
9
a
b
c
3
7
3
7
A
A
B
B
15 20
30 40 10 20
24 29
31 37 10 22
2
5
53
—
38
—
5
—
4
2
3
d
—
A, in a CH2Cl2–MeOH–H2O; B, in a MeOH–H2O mixture.
a
Determined from the 13C NMR spectra on the basis of the peak height mea-
surements of similar mixtures of known composition.
The first stages of the transformation of the starting compound
18 are supported by the diastereoisomeric excess (de) of >90% of
epimer 14, which later, after the de-O-acetylation of 3-O-acetyl
group, significantly dropped. The bulkier 3-O-acetyl group of nitro-
alkene 11 stereoselectively differentiates the methoxide addition
to its Si face more than the less bulky 3-OH group of nitroalkene
6 (i.e., from below the plane of the double bond of the respective
nitroalkenes 11 and 6, Scheme 4). Thus, in addition to the expected
2.2. Sodium methoxide-catalyzed methanolysis of 7-deoxy-7-
nitro-L-glycero-L-galacto-heptitol peracetate (18)
The finding from the above section substantiated a detailed
analysis of the formation of -galactofuranosyl nitromethanes 7
and 8 directly from 7-deoxy-7-nitro- -glycero- -galacto-heptitol
D
L
L
peracetate (18, Scheme 4) under treatment with sodium methox-
ide. Compound 18 is easily available by crystallization from an
aqueous solution of an epimeric 1-deoxy-1-nitroheptitol mixture
D
-galactofuranosyl nitromethanes 7 and 8 and the appearance of
-galactopyranosyl nitromethanes 9 and 10, methyl ethers 14
D
and 15, in their respective sodium nitronate forms 22 and 23, were
temporarily accumulated in the reaction mixture as well. A proba-
ble reason for this rather unexpected behavior is that the methox-
ide ion, despite its intermolecular action, efficiently competes with
a neutral, intramolecularly acting 5-OH group in their Michael-like
additions to the double bond of nitroalkene 6 due to its strong an-
ionic nucleophilicity. Thus, the addition interference of methoxide
considerably delays the kinetically preferred five-membered ring
obtained by the nitromethane synthesis from D-galactose, followed
by per-O-acetylation of the crystalline epimer 2.
Thus, the treatment of nitroalkene 18 with 1 M NaOMe in
MeOH at rt with a 3.2-mol excess of the base led to a fast dissolu-
tion of the starting compound. Remarkably, after 5 min, the reac-
tion mixture contained compound 14 only (Table 2, Fig. 2). Soon
after, the amount of compound 14 decreased due to the appear-
ance of O-methyl epimer 15 and D-galactofuranosyl nitromethanes
closure of nitroalkene 6 to
and 8 so that their maximum content in the reaction mixture is
reached only when considerable amounts of thermodynamically
D-galactofuranosyl nitromethanes 7
7 and 8. The highest contents, ꢀ34% and ꢀ42%, of the respective
compounds 7 and 8 in the reaction mixture were reached within
30–36 h. After a 30-day treatment, the distribution of the respec-
preferred D-galactopyranosyl nitromethanes 9 and 10 are already
tive D-galactosyl nitromethanes 7–10 in the reaction mixture be-
present. The observation of these complex transformations also ex-
plains only the moderate yields of 1-deoxy-2-O-methyl-1-nitroald-
itols in the original procedure developed by Sowden et al.,18 as
after the chosen 30-min treatment of the corresponding per-O-
acetylated nitroalkene under otherwise comparable reaction
came constant at a ratio of 7:7:9:77.
Two moles and a catalytic amount of NaOMe were necessary to
effect the conversion of the per-O-acetylated compound 18 to the
sodium nitronate forms of O-methyl derivatives 14 and 15, which
accompany and/or precede the formation of nitroalkene 6, the di-
-
+
+
O
O
Et3N-H
NO2
OH
N
-
OMe
OH
- Et3N
- MeOH
Et3N
HO
HO
HO
7 + 8 + 9 + 10
14 + 15
+ Et3N
+ MeOH
MeOH
HO
OH
OH
OH
OH
16, 17
6
Scheme 3.

718
M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
-
+
O
O
+
NO2
Na
N
NO2
-
NO2Na
NO2Na
OAc
OAc
OAc
OAc
OAc
OMe
OAc
MeO
AcO
AcO
+ NaOMe
- NaOMe
OAc
OAc
+ NaOMe
AcO
AcO
AcO
AcO
AcO
AcO
AcO
AcO
+
- NaOAc
- MeOH
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
OAc
11
Zemplén
deacetylation
NO2
18
19
21
Zemplén
deacetylation
20
NO2Na
OMe
OH
NO2Na
MeO
HO
HO
OH
+ NaOMe
- NaOMe
OH
OH
HO
HO
HO
HO
7
+
8
+
9
+
10
+
OH
OH
OH
OH
OH
22
23
Decationization
14 + 15
6
Scheme 4.
Table 2
Compositiona of the reaction mixture obtained upon treatment of 7-deoxy-7-nitro-L-
glycero-L-galacto-heptitol peracetate (18) with 3.2 equiv of NaOMe/MeOH at rt
Reaction time (h)
Compound (%)
7
8
9
10
14
15
1/12
1/4
1/2
6
24
30
36
48
72
168
720
—
—
—
—
—
5
9
9
10
10
13
15
9
—
—
—
5
13
15
16
17
22
50
77
>95
68
60
32
6
2
—
—
—
—
12
10
3
—
—
—
—
—
—
—
10
15
25
33
34
34
33
29
15
7
10
15
30
41
42
42
40
34
20
7
—
—
a
Determined from the 13C NMR spectra.
conditions, a part of the expected methoxy derivatives had to be
eliminated and cyclized to the pertinent glycosyl nitromethanes.
For preparation of the individual D-galactofuranosyl nitrome-
thanes 7 and 8, the reaction was quenched with water and a cation
Figure 2. Time dependence of the composition of the reaction mixture obtained by
the treatment of 7-deoxy-7-nitro- -glycero- -galacto-heptitol peracetate (18) with
1 M NaOMe in MeOH at rt under a 3.2-mol excess of the base, determined from the
13C NMR spectra. s, 7-deoxy-6-O-methyl-7-nitro-
-glycero- -galacto-heptitol (14)
and its epimer 15; 5, -galactofuranosyl nitromethane (7); 4, b- -galactofur-
anosyl nitromethane (8); +, -galac-
-galactopyranosyl nitromethane (9); ꢁ, b-
topyranosyl nitromethane (10).
L
L
L
L
a-
D
D
exchange resin after 36 h of treatment. Following per-O-acetyla-
a-
D
D
tion, chromatography indicated 15% and 38% yields of
a-D-galacto-
furanosyl nitromethane peracetate (24) and b- -galactofuranosyl
D
nitromethane peracetate (25). Fractional crystallization of the
decationized 36-h reaction mixture offered another possible value,
giving 25% of the crystalline compound 7.
data obtained for the individual components did not mutually dif-
fer by more than 3%, which justified the relevance of the 13C NMR
results discussed in the previous section. The desired D-galactofur-
2.3. Thermally induced b-elimination of 7-deoxy-7-nitro-
glycero- -galacto-heptitol (2)
L-
anosyl nitromethanes 7 and 8 were also preferentially formed in
this case, although to a lower total extent (maximum ca. 58% in
8–10 h, Fig. 3) when starting 2 was still present at <10%. The
L
A similar analysis was then performed for the classic, thermally
induced b-elimination of acyclic 1-deoxy-1-nitroalditols, namely
by heating compound 2 in water. In this case, due to a convenient
resolution of selected, non-overlapping signals of all the compo-
nents of the reaction mixture, a more exact 1H NMR analysis was
possible in addition to the cognate 13C NMR measurements. The
thermodynamic equilibrium of all four D-galactosyl nitromethanes
7–10 in the corresponding ratio 4:4:3:89 was then reached after
4 days of reflux and was different from that obtained under basic
conditions (7:7:9:77).
Assuming a steady-state approximation for the acyclic interme-
diate 6, kinetic Scheme 5 can be analytically solved by providing

M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
719
Figure 4. Graphical representation of the analytical solution of kinetic Scheme 5,
expressed as the time dependence of the composition of the reaction mixture; y(F)
is the sum of furanosyl compounds 7 and 8, and y(P) is the sum of pyranosyl
compounds 9 and 10, both expressed as molar fractions.
Figure 3. Time dependence of composition of reaction mixture obtained by
refluxing 7-deoxy-7-nitro-
L
-glycero-
L
-galacto-heptitol (2, s) in
a
5% aqueous
-galactofuranosyl nitrometh-
D-galactofuranosyl nitromethane (8); +, a-D-galactopyranosyl nitro-
solution, as determined from 1H NMR spectra; 5,
a-D
ane (7); 4, b-
methane (9); ꢁ, b-
D-galactopyranosyl nitromethane (10).
Because the intermediate of the mutual interconversion of both
anomeric glycopyranosyl nitromethanes and both anomeric glyco-
furanosyl nitromethanes, as demonstrated here by the D-galactosyl
expressions for the time dependence of the composition (molar
fraction) of furanosyl compounds y(F) as a sum of 7 and 8 and that
of pyranosyl compounds y(P) as a sum of 9 and 10:
nitromethanes, is 1,2-dideoxy-1-nitroald-1-enitol of the general
formula HOCH2–(CHOH)n–CH@CHNO2, the ring-chain tautomeric
behavior of glycosyl nitromethanes is analogous to the behavior
of aldoses. Comparison of the nitroalkene structure with the gen-
eral structure of acyclic aldoses HOCH2–(CHOH)n–CH@O indicates
that both structures, expressed by the generic formula HOCH2–
(CHOH)n–CH@X, differ only in the functional moiety @X, which is
@O for aldoses and @CHNO2 for glycosyl nitromethanes. The dis-
tinct characteristics of these moieties, which after cyclization of
the intermediates become the respective OH and CH2NO2 groups,
are responsible for the diverse proportions of the anomeric struc-
tures in the respective equilibrium mixtures of aldoses and glyco-
syl nitromethanes. The distinctions are striking, mainly in pyranoid
structures, where the anomeric and the other structural effects are
0ꢂt
KFP ꢂ ð1 ꢃ kÞ ꢂ kNA ꢃ k ꢂ k
1 ꢃ eꢃk
ꢃF
NAꢂt
0ꢂt
yðFÞ ¼
ðeꢃk ꢃ eꢃk Þ þ
KFP ꢂ ðk ꢂ k ꢃ kNAÞ þ k ꢂ k
1 þ KFP
ꢃF
ꢃF
yðPÞ ¼ 1 ꢃ eꢃk ꢃ yðFÞ
NAꢂt
ꢀ
ꢁ
kP
1
KFP
k ¼
;
k0 ¼ k ꢂ k
þ 1 ;
ꢃF
kF þ kP
where t is time, KFP denotes the equilibrium constant of the conver-
P
P
sion of furanosyl to pyranosyl compounds KFP
=
[P]/ [F], kNA is
the rate constant of the decomposition of the starting nitroalditol
2, kF and kP represent the rate constants of the furanosyl and pyran-
osyl compounds production from the acyclic intermediate, respec-
quite manifest. Thus, the proportions of
a- and b-pyranoid ano-
tively, and k and k are the rate constants of the corresponding
ꢃF
ꢃP
mers for all aldoses are within the same order of magnitude. On
the other hand, the percentages of the axial pyranosyl anomer in
the equilibrium mixtures of glycosyl nitromethanes are very low
in comparison with the pertinent concentrations of the equatorial
pyranosyl anomers, which are generally well above 50%. Addition-
ally, the proportions of furanoid anomers in the equilibrium mix-
tures of the most common glycosyl nitromethanes are usually
comparable with the proportion of the axial pyranosyl anomer. De-
spite the differences in the composition of the final equilibrium
mixtures of these two groups of compounds, the presented tauto-
reverse reactions.
The above expressions nicely fit the experimental data (Fig. 4)
using the measured values for KFP = 92/8 = 11.5 and kNA = 0.32 hꢃ1
(both at 100 °C). The nonlinear least-squares Marquardt–
Levenberg algorithm yields values for the parameters k and
k
ꢃF: k = 0.23 0.02 and kꢃF = 0.19 0.03 hꢃ1. These values can be
used to estimate the ratio of the rate constants for furanosyl and
pyranosyl productions from the acyclic intermediate, which
produce kF/kP = 1/k ꢃ 1 ꢄ 3.3; they can also be used to estimate
the reverse rate constant of pyranosyl formation, which produces
k
ꢃP ꢄ 4.9 ꢁ 10ꢃ3 hꢃ1
.
merization of 1-nitro-1,2-dideoxy-
D-galacto-hept-1-enitol and four
HO
O
NO2
NO2
OH
NO2
HO
kF
OH
HO
k-F
kP
HO
HO
HO
HO
kNA
7, 8
OH
OH
OH
OH
OH
H2O, reflux
OH
k-P
NO2
O
OH
HO
HO
2
6
OH
9, 10
Scheme 5.

720
M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
J3,4 = 9.6 Hz, H-4), 3.66–3.70 (m, 3H, H-1, H-10, H-3), 3.49 (s,
3H, CH3). 13C NMR (D2O) d: 79.45 (C-6), 76.86 (C-7), 71.01
(C-2), 70.17 (C-3), 69.32 (C-4), 69.19 (C-5), 64.18 (C-1), 59.12
(OCH3). MALDI-TOFMS: m/z 257.5 [M+H]+. Anal. Calcd for
C8H17NO8: C, 37.65; H, 6.71; N, 5.49. Found: C, 37.93; H, 6.57;
N, 5.74.
D
-galactosyl nitromethanes is a valuable model for the complete
mutarotation of aldoses and other reducing sugars, which still re-
mains to be studied in fundamental carbohydrate chemistry.
3. Experimental
3.1. General methods and materials
3.5. D-Galactofuranosyl nitromethane peracetates 24 and 25
Melting points were measured on a Kofler stage. Optical rota-
tions were measured with a Perkin–Elmer 141 polarimeter at
20 °C. Microanalyses were performed using a Fisons EA-1108
instrument. NMR spectra were recorded at 295 K on a Bruker
AVANCE DPX 300 spectrometer [300.13 MHz and internal sodium
(trimethylsilyl)propionate-2,2,3,3-d4, d 0.00 for 1H; 75.47 MHz
and internal MeOH, d 50.15 for 13C]. Mass spectra were obtained
with a Shimadzu-Kratos Analytical MALDI TOF IV instrument (ma-
trix 2,5-dihydroxybenzoic acid). Paper chromatography (PC) was
performed on a Whatman 1 using descending elution with the
upper layer of the S1 solvent mixture n-BuOH–EtOH–H2O 5:1:4
(v/v) and visualization with alkaline silver nitrate. TLC was run
on Merck 60 F254 silica gel precoated aluminum plates; the detec-
tion was effected by spraying the chromatograms with 10% etha-
nolic sulfuric acid and charring them on a hot plate. Flash
chromatography was performed using Acros silica gel (37–
Compound 18 (0.25 g, 0.5 mmol) was stirred in a mixture of
anhydrous MeOH (6 mL) and 1 M NaOMe in MeOH (1.6 mL) at rt
for 36 h. Then Amberlite IR 120, H+ form (5 mL) was added to the
mixture and filtered. The resin was washed with water
(3 ꢁ 5 mL). The combined filtrates were concentrated in a rotary
evaporator under reduced pressure. The foam-dry residue was dis-
solved in MeOH (2 mL) and added dropwise into Ac2O (15 mL) con-
taining concentrated H2SO4 (two drops) at 30 °C. After 24 h, the
reaction mixture was poured into ice and water (100 mL), and
the mixture was stirred for 5 h. The resulting solution was ex-
tracted with CHCl3 (3 ꢁ 30 mL), and the extract was washed with
water until neutral, dried (Na2SO4) and evaporated, yielding a mix-
ture of peracetates of compounds 7–10, which were separated by
flash chromatography.
75 lm) and the solvent mixture S2, hexanes–EtOAc 5:2 (v/v).
3.5.1. 1,2,4,5-Tetra-O-acetyl-3,6-anhydro-7-deoxy-7-nitro-
D-
glycero- -galacto-heptitol ( -galactofuranosyl nitromethane
L
a-D
3.2. Et3N-catalyzed methanolysis of 1,2-dideoxy-1-nitro-D-
galacto-hept-1-enitol peracetate (11)
1,2,4,5-tetraacetate, 24)
Yield 30 mg (15%), Rf = 0.31 (S2), ½a D20
ꢅ
+12.0 (c 1, CHCl3); Ref. 21:
½
a 2D0 +18.5 (c 1.5, CHCl3). 1H NMR (acetone-d6) d: 5.44 (dd, 1H,
ꢅ
(a) Compound 11 (0.1 g, 0.23 mmol) was dissolved in CH2Cl2
(1 mL), and an 8:2:1 (volume ratio) mixture of MeOH–H2O–Et3N
(2 mL) was added. The solution was kept at rt for 3 and 7 days,
respectively. Then the solvents were evaporated at diminished
pressure and temperature below 40 °C and submitted to 13C NMR
evaluation of the reaction mixtures against similar mixtures of
known composition (Table 1).
(b) A mixture of powdered compound 11 (0.1 g, 0.23 mmol) in
an 8:2:1 mixture of MeOH–H2O–Et3N (3 mL) was stirred at rt for
3 and 7 days, respectively, and the final reaction solution was fur-
ther processed as described above.
J5,6 = 4.1 Hz, J4,5 = 1.4 Hz, H-5), 5.38 (ddd, 1H, J1,2 = 4.0 Hz,
0
J1 ,2 = 7.0 Hz, J2,3 = 5.8 Hz, H-2), 5.16 (dd, 1H, J3,4 = 3.2 Hz, H-4),
0
0
4.84–4.91 (m, H-6, H-7), 4.73 (dd, 1H, J6,7 = 9.4 Hz, J7,7 = 14.3 Hz,
H-70), 4.36 (dd, 1H, J1,2 = 4.0 Hz, J1,1 = 12.0 Hz, H-1), 4.09–4.16 (m,
0
2H, H-10, H-3), 1.98, 2.07, 2.09, 2.13 (4s, 12H, CH3). 13C NMR (ace-
tone-d6) d: 171.98, 171.74, 171.38, 171.22 (4CO), 84.56 (C-3), 79.76
(C-4), 78.57 (C-5), 78.46 (C-6), 76.48 (C-7), 71.76 (C-2), 64.55 (C-1),
21.80, 21.86, 21.90, 22.06 (4CH3).
3.5.2. 1,2,4,5-Tetra-O-acetyl-3,6-anhydro-7-deoxy-7-nitro-
L-
glycero- -galacto-heptitol (b- -galactofuranosyl nitromethane
L
D
1,2,4,5-tetraacetate, 25)
3.3. Sodium methoxide-catalyzed methanolysis of 7-deoxy-7-
nitro-L-glycero-L-galacto-heptitol peracetate (18)
Yield 74 mg (38%), Rf = 0,28 (S2), ½a D20
ꢅ
+6.0 (c 1, CHCl3); Ref. 21:
½
a 2D0 +6.0 (c 1.5, CHCl3). 1H NMR (acetone-d6) d: 5.36 (dt, 1H,
J1,2 = 7,0 Hz, J2,3 = 4,6 Hz, H-2), 5.23 (t, 1H, J3,4 = 2.8 Hz, H-4), 5.19
(t, 1H, J4,5 = 3.0 Hz, H-5), 4.92 (t, 2H, J6,7 = 4.3 Hz, H-7, H-70), 4.80
(dd, 1H, J5,6 = 5.2 Hz, H-6), 4.33 (dd, 1H, H-5), 4.28 (dd, 1H,
ꢅ
Compound 18 (0.1 g, 0.2 mmol) was stirred in a mixture of
anhydrous MeOH (2.4 mL) and 1 M NaOMe in MeOH (0.64 mL) at
rt. After the given time (Table 2), Amberlite IR 120, H+ form
(1 mL) was added to the mixture and filtered. The neutral filtrate
was evaporated, and the composition of the reaction mixture
was submitted to 13C NMR analysis.
J1,2 = 4.4 Hz, J1,1 = 11.9 Hz, H-1), 4.15 (dd, 1H, H-10), 1.98, 2.07,
0
2.09, 2.11 (4s, 12H, 4CH3). 13C NMR (acetone-d6) d: 171.92,
171.85, 171.66, 171.60 (4CO), 84.07 (C-3), 82.23 (C-6), 80.38
(C-4), 80.19 (C-5), 78.37 (C-7), 71.94 (C-2), 64.23 (C-1), 21.77,
21.89, 21.91, 22.07 (4CH3).
3.4. 7-Deoxy-6-O-methyl-7-nitro-L-glycero-L-galacto-heptitol
(14)
3.6. D-Galactopyranosyl nitromethane peracetates 26 and 27
Finely powdered compound 18 (1.05 g, 2.1 mmol) was sus-
pended by stirring in anhydrous MeOH (26 mL), and 1 M NaOMe
in MeOH (6.7 mL) was added. After 5 min of vigorous stirring at
rt, Amberlite IR 120, H+ form (10 mL) was added to the mixture
and filtered. The resin was washed with water (3 ꢁ 10 mL). The
combined neutral filtrates were concentrated on a rotary evapora-
tor under reduced pressure, and the residue was crystallized from
ethanol.
Yield 35 mg (18%, 1:2). Rf = 0.28 (S2).
3.6.1. 1,3,4,5-Tetra-O-acetyl-2,6-anhydro-7-deoxy-7-nitro-D-
glycero-
L-galacto-heptitol (a-D-galactopyranosyl nitromethane
1,2,4,5-tetraacetate, 26)
1H NMR (acetone-d6) d: 5.59–5.61 (m, 1H), 5.50–5.57 (m, 1H),
5.37–5.48 (m, overlap), 5.13–5.19 (m, 1H), 4.63–4.67 (m, overlap),
4.32–4.35 (m, 1H, H-7), 4.24 (dd, 1H, overlap, H-70), 2.14, 2.16,
2.26, 2.29 (4s, 12H, 4CH3). 13C NMR (acetone-d6) d: 171.82, 171.76,
171.44, 171.23 (4CO), 74.54, 71.80, 71.23, 69.43, 69.06, 68.57 (C-2–
C-7), 62.74 (C-1), 21.74, 21.83, 21.85, 21.92 (4CH3).
Yield 0.5 g (92%), mp 170–172 °C (EtOH), ½a D20
ꢃ4.0 (c 1, H2O),
ꢅ
RGal 4.13 (S1). 1H NMR (D2O) d: 5.03 (dd, 1H, J6,7 = 3.5 Hz,
J7,7 = 13.2 Hz, H-7), 4.79 (dd, 1H, J6,7 = 7.1 Hz, H-70), 4.12 (dt,
1H, J5,6 = 7.4 Hz, H-6), 3.96–4.02 (m, 2H, H-2, H-5), 3.82 (d, 1H,
0
0

M. Vojtech et al. / Carbohydrate Research 346 (2011) 715–721
721
3.6.2. 1,3,4,5-Tetra-O-acetyl-2,6-anhydro-7-deoxy-7-nitro-
glycero- -galacto-heptitol (b- -galactopyranosyl nitromethane
1,2,4,5-tetraacetate, 27)
L
-
3.7.4. 2,6-Anhydro-7-deoxy-7-nitro-
L
-glycero-
L
-galacto-heptitol
L
D
(b-D-galactopyranosyl nitromethane, 10)
Yield 1.6 g (71%), RGal = 3.39 (S1), mp 198–200 °C (MeOH), ½a D20
ꢅ
1H NMR (acetone-d6) d: 5.63 (dd, 1H, J2,3 = 0.9 Hz, H-3), 5.45 (dd,
1H, J3,4 = 3.5 Hz, H-4), 5.28 (t, 1H, J4,5 = 9.9 Hz, H-5), 5.02 (dd, 1H,
+36 (c 1, H2O); Ref. 22: mp 199.5–200.5 °C, ½a D25
ꢅ
+36 (c 2.9. H2O).
1H NMR (D2O) d: 4.95 (dd, 1H, J7,7 = 13.5 Hz, J6,7 = 2.6 Hz, H-7),
0
J7,7 = 13.5 Hz, J6,7 = 2.3 Hz, H-7), 4.77 (dd, 1H, J6,7 = 9.4 Hz, H-70),
4.64 (dt, 1H, J5,6 = 9.6 Hz, H-6), 4.43 (dt, 1H, J1,2 = 6.5 Hz, H-2),
4.25 (dd, 2H, H-1, H-10), 2.11, 2.13, 2.23, 2.30 (4s, 12H, 4CH3). 13C
NMR (acetone-d6) d: 171.89, 171.76, 171.71, 171.40 (4CO), 78.21
(C-7), 76.58 (C-6), 76.24 (C-2), 73.45 (C-4), 69.74 (C-3), 68.57
(C-5), 63.31 (C-1), 21.74, 21.75, 21.77, 21.91 (4CH3).
4.67 (dd, 1H, J6,7 = 8.8 Hz, H-70), 4.02 (dt, 1H, J5,6 = 9.6 Hz, H-6),
0
0
0
3.96 (d, 1H, J2,3 = 3.3 Hz, H-3), 3.65–3.71 (m, 4H, H-1, H-10, H-2,
H-4), 3.57 (t, 1H, H-5). 13C NMR (D2O) d: 79.73 (C-2), 77.52 (C-6),
77.41 (C-7), 74.70 (C-4), 69.92 (C-3), 68.72 (C-5), 62.09 (C-1).
3.8. Thermally induced b-elimination of 7-deoxy-7-nitro-
glycero- -galacto-heptitol (2)
L-
L
3.7. D-Galactosyl nitromethanes 7–10
The starting compound 2 Ref. 23 was dissolved in deionized and
distilled water, and the solution was stirred and heated at an inter-
nal temperature of 100 °C. At selected time intervals (Fig. 3),
appropriate volume samples were taken, cooled, and concentrated
in a rotary evaporator under reduced pressure (0.5-mL samples
from 0.5 g (2.07 mmol) of 2 in 10 ml of H2O for 1H NMR spectros-
copy and 0.4-mL samples from 1.2 g (4.98 mmol) of 2 in 5 mL of
H2O for 13C NMR spectroscopy).
Compound 18 (4.93 g, 10 mmol) was stirred in a mixture of
anhydrous MeOH (120 mL) and 1 M NaOMe in MeOH (32 mL) at
rt for 36 h. Then Amberlite IR 120, H+ form (50 mL) was added to
the mixture and filtered. The resin was washed with water
(3 ꢁ 50 mL). The neutral combined filtrates were concentrated in
a rotary evaporator under reduced pressure and again dissolved
in water (100 mL) and treated with charcoal (2 g). The decolorized
filtrate was again concentrated, and the syrupy residue (2.2 g) was
submitted to fractional crystallization from MeOH. A part of the
mother liquor (0.5 g) was separated by the Whatman 3 preparative
PC and S1 solvent mixture.15
Acknowledgments
This work was supported by the APVV-51-046505 and VEGA-2/
0199/09 grants. The funds for the NMR measurements were pro-
vided by the Slovak State ProgrammeProject No. 2003SP200280203.
3.7.1. 3,6-Anhydro-7-deoxy-7-nitro-D-glycero-L-galacto-heptitol
(a-D
-galactofuranosyl nitromethane, 7)
Yield 0.3 g (15%), RGal = 4.79 (S1), mp 161–163 °C (MeOH–
References
Me2CO), ½a 2D0
ꢅ
ꢃ2 (c 1; H2O); Ref.21: mp 163–166 °C, ½a 2D0
ꢃ7
ꢅ
(c 0.8, H2O). 1H NMR (D2O) d: 4.87 (dd, 1H, J7,7 = 12.5 Hz,
J6,7 = 1,9 Hz, H-7), 4.66–4.77 (m, 2H, H-6, H-70), 4.28 (dd, 1H,
J4,5 = 2.9 Hz, H-5), 4.17 (t, 1H, J3,4 = 3.4 Hz, H-4), 3.82–3.86 (m,
2H, H-2, H-3), 3.57–3.71 (m, 2H, H-1, H-10). 13C NMR (D2O) d:
85.62 (C-3), 78.69 (C-4), 77.91 (C-6), 77.67 (C-5), 75.99 (C-7),
72.08 (C-2), 63.49 (C-1).
1. Davis, C. B.; Hartnell, R. D.; Madge, P. D.; Owen, D. J.; Thomson, R. J.; Chong, A. K.
J.; Coppel, R. L.; von Itzstein, M. Carbohydr. Res. 2007, 342, 1773–1780.
2. Owen, D. J.; Thomson, R. J.; von Itzstein, M. Carbohydr. Res. 2002, 337, 2017–
2022.
0
3. Petrušová, M.; Vojtech, M.; Pribulová, M.; Lattová, E.; Matulová, M.; Poláková,
ˇ
M.; BeMiller, J. N.; Kren, V.; Petruš, L. Carbohydr. Res. 2006, 341,
2019–2025.
4. Wade, M. M.; Zhang, Y. Front. Biosci. 2004, 9, 975–994.
5. Nachega, J. B.; Chaisson, R. E. Clin. Infect. Dis. 2003, 36, S24–S30.
6. Pan, F.; Jackson, M.; Ma, F.; McNeil, M. J. Bacteriol. 2001, 183, 3991–3998.
7. Suzuki, E.; Tanaka, A. K.; Toledo, M. S.; Takahashi, H. K.; Straus, A. H. Infect.
Immun. 2002, 70, 6592–6596.
8. Beverley, S. M.; Owens, K. L.; Showalter, M.; Griffith, C. L.; Doering, T. L.; Jones,
V. C.; McNeil, M. R. Eukaryot. Cell 2005, 4, 1147–1154.
9. Lowary, T. L. Mini-Rev. Med. Chem. 2003, 3, 694–707.
10. Jachak, S. M.; Jain, R. Anti-Infect. Agents Med. Chem. 2006, 5, 123–133.
11. Biava, M.; Porretta, G. C.; Deidda, D.; Pompei, R. Infect. Disord. Drug Targets
2006, 6, 159–172.
12. Dietrich, J.; Lundberg, C. V.; Andersen, P. Tuberculosis 2006, 86, 163–168.
13. Skeiky, Y. A. W.; Sadoff, J. C. Nat. Rev. Microbiol. 2006, 4, 469–476.
14. Sowden, J. C.; Oftedahl, M. L. J. Org. Chem. 1961, 26, 1974–1977.
15. Petruš, L.; Bystricky´, S.; Sticzay, T.; Bílik, V. Chem. Zvesti 1982, 36,
103–110.
3.7.2. 3,6-Anhydro-7-deoxy-7-nitro-L-glycero-L-galacto-heptitol
(b-D-galactofuranosyl nitromethane, 8)
Yield 0.1 g (4.5%, by PC), RGal = 4.13 (S1), ½a D20
ꢃ48 (c 1, H2O);
ꢅ
Ref.21: mp 99–100 °C (EtOH), ½a 2D0
ꢅ
ꢃ63.2 (c 0.7, H2O). 1H NMR
(D2O) d: 4.75–4.80 (m, 2H, H-7, H-70), 4.48 (dt, 1H, J6,7 = 7.3 Hz,
J5,6 = 3.7 Hz, H-6), 4.05 (t, 1H, J3,4 = 6.4 Hz, H-4), 3.97 (t, 1H,
J4,5 = 6.3 Hz, H-5), 3.88 (dd, 1H, J3,4 = 6,4 Hz, H-3), 3.76–3.83 (m,
1H, H-2), 3.60 (dt, 2H, J1,2 = 4.5 Hz, J1,1 = 12.0 Hz, H-1, H-10). 13C
0
NMR (D2O) d: 83.17 (C-3), 79.89 (C-6), 78.22 (C-5), 77.71 (C-7),
77.30 (C-4), 71.80 (C-2), 63.53 (C-1).
16. Phiasivonga, P.; Samoshin, V. V.; Gross, P. H. Tetrahedron Lett. 2003, 44, 5495–
5498.
17. Sakakibara, T.; Takamoto, T.; Matsuzaki, T.; Omi, H.; Maung, U. W.; Sudoh, R.
Carbohydr. Res. 1981, 95, 291–298.
18. Sowden, J. C.; Oftedahl, M. L.; Kirkland, A. J. Org. Chem. 1962, 27, 1791–1793.
19. Martin, O. R.; Khamis, F. E.; Rao, S. P. Tetrahedron Lett. 1989, 30, 6139–6142.
20. Hough, L.; Shute, S. H. J. Chem. Soc. 1962, 4633–4637.
21. Köll, P.; Kopf, J.; Wess, D.; Brandenburg, H. Liebigs Ann. Chem. 1988, 685–693.
22. Sowden, J. C.; Bowers, C. H.; Lloyd, K. O. J. Org. Chem. 1964, 29, 130–132.
23. Köll, P.; Stenns, C.; Seelhorst, W.; Brandenburg, H. Liebigs Ann. Chem. 1991,
201–206.
3.7.3. 2,6-Anhydro-7-deoxy-7-nitro-
D
-glycero-
L
-galacto-heptitol
(a-D-galactopyranosyl nitromethane, 9)
Yield 0.05 g (2.2%, by PC), RGal = 2.63 (S1), ½a D20
ꢅ
+145 (c 1, H2O);
Ref.20: mp 171–181 °C, ½a 2D4
ꢅ
+88.5 (c 2, H2O). 1H NMR (D2O) d:
5.05 (dd, 1H, J7,7 = 11.7 Hz, H-7), 4.80–4.88 (m, 2H, H-6, H-70),
4.12 (dd, 1H, H-5), 3.91–3.93 (m, 1H, H-2), 3.67–3.73 (m, 4H, H-
1, H-10, H-3, H-4). 13C NMR (D2O) d: 74.34 (C-2), 74.07, 73.21
(C-6, C-7), 70.65, 69.65 (C-3, C-4), 67.74 (C-5), 61.89 (C-1).
0