Theoretical and experimental studies on methyl α-d-glucopyranoside derivatives

Article abstract of DOI:10.1016/j.molstruc.2006.10.057

Theoretical quantum mechanical calculations using density functional theory (DFT) at the B3LYP level have been carried out for synthesized esterified methyl α-d-glucopyranoside derivatives. The predicted NMR characteristics obtained with GIAO method are c

Full text of DOI:10.1016/j.molstruc.2006.10.057

Journal of Molecular Structure 834–836 (2007) 498–507
www.elsevier.com/locate/molstruc
Theoretical and experimental studies on methyl
a-D-glucopyranoside derivatives
a,
a
c
a
E. Chełmecka *, K. Pasterny , M. Gawlik-Je˛drysiak b,c, W. Szeja , R. Wrzalik
a
University of Silesia, Institute of Physics, Uniwersytecka 4, 40-007 Katowice, Poland
Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Weigla 12, 53-114 Wrocław, Poland
b
c
Silesian Technical University, Department of Chemistry, B. Krzywoustego 8, 44–100 Gliwice, Poland
Received 8 September 2006; accepted 17 October 2006
Available online 4 January 2007
Abstract
Theoretical quantum mechanical calculations using density functional theory (DFT) at the B3LYP level have been carried out for
synthesized esterified methyl a-D-glucopyranoside derivatives. The predicted NMR characteristics obtained with GIAO method are com-
pared with 1H, 13C and 31P experimental NMR data, whereas calculated vibrational frequencies have been confronted with experimental
Raman spectra. This comparison has permitted to state that the compounds under investigation do not occur in the form of a zwitterion
but in the neutral form as well as to determine the most favourable stereoisomer.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Phosphorylated saccharides; Glucose derivatives; DFT calculations; NMR and Raman
1. Introduction
ble isomer predominates. Understanding the biological role
of non-carbohydrate substituents in bacterial LPS is not
easy, because they are acid and basic-labile and are easily
lost during isolation of native LPS from bacterial mass.
For this reason chemical synthesis of compound with struc-
ture similar to native core oligosaccharide is necessary.
Synthetic oligosaccharides and glycoconjugates provide
materials for correlating structure with function [4]. The
aim of our research is to predict the structure of synthe-
sized sugar derivatives based on methyl a-D-glucopyrano-
side unit, using theoretical computational methods and to
compare the results with spectroscopic data.
The study of carbohydrates within biological systems
has illustrated that they are involved in a number of funda-
mental biological functions such as cell–cell recognition
and cell-external agent interactions. These interactions
can initiate beneficial biological events, such as immune
response, as well as detrimental disease processes, such as
inflammation, viral and bacterial infections [1]. It was
found that several core oligosaccharides of bacterial lipo-
polysaccharides (LPS), besides the common structural ele-
ments such as heptose, glucose, galactose, Kdo contain
also non-carbohydrate groups: phosphate, pyrophospho-
ethanolamine, phosphoethanolamine, phosphocholine,
acetate, and glycine [2]. The glycine is located at the sugar
unit, mainly gluco- or galactopyranose, usually phosphor-
ylated at primary hydroxyl group [3]. The ester groups can
migrate over the sugar ring and in equilibrium the most sta-
2. Experimental
2.1. Synthetic procedure (Fig. 1)
2.1.1. Methyl (a-D-glucopyranosid-1-yl)-6-O-
diphenylphosphate (1)
Methyl a-D-glucopyranosid was dissolved in a pyridine
in the presence of DMAP, and cooled for 0.5 h to ꢀ4 ꢁC,
in the next step, diphenyl phosphoryl chloride (1.1 equiv)
*
Corresponding author. Fax: +48 32 588431.
E-mail address: chelmeck@us.edu.pl (E. Chełmecka).
0022-2860/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2006.10.057

E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
499
was added. The reaction was stirred over night in temper-
ature range ꢀ4 ꢁC to room temperature, then poured into
water and extracted three times with methylene chloride.
The combined organic layers were washed twice with
water, dried over anhydrous Na2SO4 and concentrated in
vacuo with toluene. The residue was purified on silica gel
was measured using Peltie-cooled CCD detector
(1064 · 256 pixels).
3. Method of calculations
Optimized structures of all regarded isolated molecules
were calculated by the density functional three-parameter
hybrid method (DFT/B3LYP) using the 6-31G* basis set.
Computations were performed using the Gaussian 03 pro-
gram package [9]. The optimized geometries were used as
input for NMR and harmonic vibrational frequency calcu-
lations. The gauge-invariant atomic orbitals (GIAO)
method [10–12] was used for prediction of DFT nuclear
shielding. In NMR experiments, nuclear shielding r are
given as the chemical shift d. Calculations, on the other
20
(eluent: ethyl acetatae)[5]. ½aꢁD 62.1 (c 0.9, CHCl3).
2.1.2. Methyl (a-D-glucopyranosid-1-yl)-6-O-
phenylphosphate (2)
Methyl
(a-D-glucopyranosid-1-yl)-6-O-phenylphos-
phate was obtained using the new highly efficient
hydrogenolysis method with ammonium formate as a
hydrogen source and palladium on carbon as the cata-
20
lyst, which will be published elsewhere [6]. ½aꢁD 61.9 (c
0.5, H2O).
hand, produce the absolute tensors r and their traces riso.
The relationship d = rrefꢀr is a good approximation of the
relationship between chemical shifts and shieldings. In
order to compare 1H, 13C and 31P absolute isotropic shield-
ing with experimental chemical shift, we have used as rref
the value of the isotropic shielding tensor of the experimen-
2.1.3. Methyl (3-O-(N-Bocglicynyl)-a-D-glucopyranosid-1-
yl)-6-O-diphenylphosphate (3)
Methyl
(a-D-glucopyranosid-1-yl)-6-O-diphenylphos-
phate, Boc-glycine (2 equiv.) and DMAP were dissolved
in a methylene chloride and cooled for 0.5 h to ꢀ4 ꢁC. In
the next step, DCC (2 equiv.) was added, the reaction
was carried out 2 h at ꢀ4 ꢁC. The progress of reaction
was monitored by TLC (chloroform–methanol 15:1). After
filtration, the solution was concentrated in vacuo and the
residue was purified on silica gel (eluent: chloroform) [7].
1
tal standards, TMS (for H and 13C) and phosphoric acid
31
(for P), calculated at the same DFT/6-31G* level.
The atom numbering used for the present calculations is
shown in Fig. 2a; according to X,Y,Z substituents we have
to do with compounds (1), (2) or (4), shown on the synthe-
sis scheme (Fig. 1), which are the subject of our investiga-
tions. Oxygen atoms have the same number as carbon
atoms which they are linked to, except those atoms from
the phosphoric group that are denoted as O11, O12 and
O13. Carbon atoms in glycine radical are denoted as C20
and C21. Taking into account the space distribution of
the substituents at the phosphorus atom (Fig. 2b), for com-
pounds (2) and (4) two stereoisomers, denoted in the fol-
lowing as isomer I and II, can exist. Considering the
presence of the glycine group in (4), we have assumed the
migration possibility of the hydrogen atom from the phos-
phoric to amino group (Fig. 3); the emerging molecule has
a character of an inner salt (zwitterion, betaine).
2.1.4. Methyl (3-O-(N-Bocglicynyl)-a-D-glucopyranosid-1-
yl)-6-O-phenylphosphate
Methyl (3-O-(N-Bocglicynyl)-a-D-glucopyranosid-1-yl)-
6-O-phenylphosphate was obtained by the same method
as compound (2) [6].
2.1.5. Methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-
phenylphosphate (4)
The removal of Boc-group, protecting amino acid resi-
due, was carried out in 33% trifluoroacetic acid (TFA)
for 2 h. The progress of reaction was monitored by TLC
(toluene–ethyl acetate–ethanol 2:2:1). Finally, the mixture
was lyophilized [8].
4. Results and discussion
2.2. Techniques
Theoretically calculated energies for isomers I and II,
both in neutral and zwitterionic forms of compound (4),
differ very little [13,14] thus their coexistence is probable.
Therefore, possible isomers of all regarded compounds,
(1), (2) and (4), will be subjected to further analysis exploit-
ing theoretical and experimental data.
1H, 13C and 31P NMR spectra were recorded on a Var-
ian Inova 400 Spectrometer and measured in CDCl3 and
D2O solutions at room temperature. Both proton and car-
bon signals were referenced to TMS and phosphorus sig-
nals were referenced to H3PO4.
Raman measurements were made using Lab-Ram dis-
persive spectrometer with the sample at room temperature.
Argon-ion laser light 514.5 nm was used to excite the spec-
tra. The power at the sample was below 10 mW. The polar-
iser, half-wave plate and analyzer were used to control
incident and scattering polarizations. To eliminate the
Rayleigh line the scattered light was passed through a holo-
graphic notch filter with cut-off 100 cmꢀ1. The intensity
4.1. Molecular geometry
Selected DFT calculated structural parameters of inves-
tigated molecules are listed in Table 1. Both the bond
lengths and angles within the sugar ring for (1), (2) and
(4) are close each other and comparable to experimental
data for a-D-glucose [15]. For the neutral forms the inser-
tion of the glicydyl group causes noticeable elongation of

500
E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
OPh
OPh
PhO
P
O
(1)
(3)
PhO
P
O
O
O
HO
HO
O
O
O
(iii)
(i)
OMe
OMe
HO
OMe
HO
OH
O
O
OH
HO
OH
HO
H
Me
Me
O
N
Me
O
(ii)
(ii)
O
OH
(iv)
OH
PhO
P
PhO
P
O
(2)
(4)
O
O
O
O
OMe
HO
OMe
HO
OH
O
OH
HO
O
NH2
Fig. 1. Synthesis of methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylphosphate. Reagents and conditions: (i) diphenyl phosphoryl chloride,
pyridine; (ii) Pd/C, ammonium acetate; (iii) Boc-glycine, methylene chloride, DCC, DMAP; (iv) 33% TFA.
Fig. 2. (a) Atom numbering used for all regarded molecules, (b) two possible stereoisomers at phosphorus atom.
˚
the C3–O3 bond (by about 0.025 A) and inconsiderable
elongation of C1–C2, C1–O1 and C2–C3 bonds, whereas
C4–O4 and O5–C1 distances are slightly shortened. For
both isomers of (4) the possibility of proton migration
has been considered that relates with the changes of
Oꢂ ꢂ ꢂHꢂ ꢂ ꢂN distances and affects the shape of optimized

E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
501
Fig. 3. The scheme for the reaction of the zwitterion formation.
molecules (see Fig. 4). In the case of Hꢂ ꢂ ꢂN distance we
lated and experimental proton and carbon chemical shifts
is quite satisfactory, somewhat greater differences appear
for carbon atoms of aromatic ring. An increase of the C6
chemical shift by about 6 ppm in comparison with methyl
a-D-glucopyranosid indicates that the phosphoric group is
attached to C6 atom.
˚
observe its decrease from about 12 A in the neutral form
˚
to 1 A in the zwitterionic form that means the proton trans-
fer from the acid to amino group. Besides these changes we
also observe for the betaine form the increase in the C21–
N, C5–C6, C3–O3 and C4–O4 and C20–C21 bond lengths
˚
by 0.005, 0.003, 0.002 and 0.018 A, respectively, whereas
The analysis of the fine structure of the spectra for (1)
2
˚
O3–C20 distance is shorter by about 0.012 A. The length
has given the coupling constants: JP,C(ipso) = 7.02 Hz,
3
3
of the ester bond P–O6 is sensitive to the ionization and
2JP,C6 = 5.85 Hz, JP,C5 = 6.04 Hz and JH1,H2 = 3.30 Hz.
˚
increases by 0.06 or 0.07 A according to the stereoisomer.
For (2) we succeeded in obtaining following coupling con-
3
Calculations for the neutral isomers have given the value
stants: 3JH4,H5 = 9.77 Hz, 3JH2,H3 = 9.53 Hz, JH3,H4
=
=
equal 1.600 A which is comparable with that reported by
9.28 Hz, 3JH1,H2 = 3.67 Hz, 2JP,C6 = 5.51 Hz, JP,C5
3
˚
Katti at al. [16] for G6P monobarium salt.
7.55 Hz. In the case of (4) the 3JH1,H2 value has been deter-
mined as 3.6 Hz. This value indicates that the configuration
of saccharides studied is the a one.
The angle differences between neutral isomers of (4) and
those of (2) and (1) are clear for O4–C4–C5, O3–C3–C4
and C2–C3–O3 only. In the case of (4) the first two men-
tioned angles are smaller by about 5ꢁ and 4ꢁ, respectively,
when comparing with corresponding angles for (2) and (1),
while the last angle is greater by about 3ꢁ; the differences
for other angles are negligible. The comparison of calculat-
ed angles for betaine and neutral forms of (4) shows more
significant differences, C3–C4–C5 angle for zwitterionic
form is greater by about 10ꢁ and O3–C3–C4 is smaller by
4ꢁ.
The C5–C6–O6–P dihedral angle indicates the differenc-
es originating from the stereoisomerism at the phosphorus
atom for (2) and (4); its calculated value for isomer I is
ꢀ178.2ꢁ and ꢀ180.0ꢁ for (4) and (2), respectively, and for
isomer II 137.2ꢁ and 138.2ꢁ. Considering the possible rota-
tion about the exocyclic C5–C6 bond, the O6–C6–C5–O5
angle is the most regarded dihedral angle for sugar deriva-
tives [17]. From our calculations follows that for (1), (2)
and neutral forms of (4) this angle takes the synclinal
(+sc) value (in the range 70–78ꢁ), whereas for zwitterionic
forms its value is antiplanar (ap, ꢀ163 to ꢀ168ꢁ). Consid-
erable differences between betaine and neutral forms of (4)
occur within the glicydyl group as a result of the proton
migration mentioned above.
For (4) we observe an increase in the chemical shift
for C3 to 77.18 ppm in comparison with 73.9 ppm for
(1). For H3 the chemical shift changes from 3.72 ppm
for (1) to 5.22 ppm for (4). For neighbouring C2 and
C4 atoms the chemical shift decreases, whereas for the
next ones these values are greater than in the case of
(1). These facts indicate that the glicydyl group is substi-
tuted to the C3 carbon atom. From calculation for (4)
significant difference of chemical shifts for C3 (about
10 ppm) between isomers in neutral and zwitterionic
forms is clearly seen and the theoretically predicted val-
ues for neutral form are in accordance with experimental
ones. Similar situation concerns chemical shifts for C4
although in this case the difference is smaller; zwitterionic
forms exhibit greater chemical shifts by about 5 ppm in
comparison with neutral ones. Practically the same can
be said with regard to the chemical shift for C20. In this
case, however, as opposed to C4 and C3, the chemical
shifts are greater for neutral forms by about 6–7 ppm
and are also in agreement with experiment.
In the proton NMR spectra for (4) the signal at
3.41 ppm was observed which has been assigned to the
PO/H acid proton. The signal originating from the NHH0
group in the vicinity of 0.4 ppm has not been observed,
probably it is vanishing among the noise signal. Simulta-
neously, the signal in the vicinity of 7 ppm which could
originate, according to the calculations, from the betaine
form (NHþ3 ) has not been registered either. Indeed, in the
neighbourhood of 7 ppm have been observed multiplets
but their integral analysis did not confirm the presence of
4.2. NMR studies
Experimental and calculated chemical shifts for all com-
pounds under investigations are presented in Table 2. For
(1) measurements have been carried out in CDCl3 solution
while for (2) and (4) in D2O. The agreement between calcu-

502
E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
Table 1
Selected DFT optimized structure parameters of methyl (a-D-glucopyranosid-1-yl)-6-O-diphenylophosphate (1), methyl (a-D-glucopyranosid-1-yl)-6-O-
phenylophosphate (2) and methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (4)
(1)
(2)
(4)
a-D-glucose
Isomer I
Isomer II
Isomer I
Isomer II
Isomer I ion
Isomer II ion
Exp. [15]
˚
Bond length (A)
C1–C2
C1–O1
C2–C3
C2–O2
C3–C4
C3–O3
C4–C5
C4–O4
C5–O5
O5–C1
C5–C6
C6–O6
O6–P
1.533
1.396
1.528
1.412
1.530
1.422
1.539
1.431
1.430
1.428
1.516
1.446
1.598
–
1.533
1.396
1.527
1.412
1.530
1.422
1.539
1.431
1.429
1.427
1.516
1.446
1.598
–
1.533
1.396
1.526
1.413
1.530
1.423
1.539
1.430
1.430
1.428
1.518
1.449
1.599
–
1.539
1.405
1.534
1.409
1.525
1.447
1.637
1.419
1.431
1.416
1.516
1.445
1.595
1.359
1.518
1.209
1.454
0.972
11.723
1.540
1.404
1.534
1.409
1.528
1.447
1.536
1.422
1.421
1.417
1.517
1.447
1.600
1.359
1.518
1.210
1.400
0.971
12.217
1.540
1.410
1.527
1.412
1.523
1.468
1.535
1.440
1.433
1.410
1.545
1.434
1.661
1.348
1.536
1.204
1.504
1.706
1.073
1.540
1.409
1.529
1.412
1.522
1.467
1.534
1.442
1.432
1.411
1.545
1.437
1.673
1.347
1.535
1.205
1.507
1.532
1.103
1.539
1.400
1.529
1.421
1.523
1.425
1.534
1.430
1.431
1.430
1.517
1.424
O3–C20
C20–C21
C20–O20
C21–N
Oꢂ ꢂ ꢂH
–
–
–
–
–
–
–
–
–
–
–
–
Hꢂ ꢂ ꢂN
–
–
–
Angles (deg)
O5–C1–O1
O5–C1–C2
O1–C1–C2
C1–C2–C3
C1–C2–O2
O2–C2–C3
C2–C3–O3
O3–C3–C4
C2–C3–C4
C3–C4–C5
C3–C4–O4
O4–C4–C5
C4–C5–O5
C5–O5–C1
C4–C5–C6
O5–C5–C6
C5–C6–O6
C6–O6–P
112.5
108.9
109.3
109.7
109.9
112.0
107.4
111.5
111.0
109.9
110.5
113.2
109.7
114.1
112.2
107.6
108.2
122.0
–
112.5
108.9
109.3
109.6
110.0
112.0
107.3
111.4
111.0
110.0
110.5
112.9
109.9
114.2
112.0
107.4
107.9
122.3
–
112.5
109.0
109.3
109.6
110.0
112.0
107.0
111.4
110.7
109.7
110.7
112.9
110.2
114.1
112.4
107.6
108.6
123.0
–
113.1
111.2
103.6
108.5
112.1
113.3
110.2
107.2
111.0
108.3
111.8
108.2
109.6
114.6
112.2
107.2
107.4
122.5
117.2
110.6
124.7
124.5
110.0
113.1
111.1
106.5
108.6
112.1
113.2
110.2
107.3
111.0
108.3
111.7
108.2
109.6
114.5
112.3
107.5
108.6
121.8
117.2
110.6
124.7
124.6
110.0
113.4
112.6
105.8
109.2
110.9
112.9
108.6
102.9
112.2
117.8
107.0
111.3
112.1
115.8
107.8
105.4
109.4
123.9
118.3
109.6
126.6
123.7
108.0
113.4
112.5
105.9
109.4
110.9
112.9
108.4
103.2
112.2
117.0
107.8
110.1
112.0
115.5
108.1
105.5
109.3
120.1
118.1
109.7
126.6
123.6
107.3
111.5
110.1
109.3
111.1
110.9
112.2
108.1
110.6
109.8
111.1
108.2
110.9
108.7
113.7
111.5
108.1
109.9
C6–O6–C20
O6–C20–C21
O6–C20–O20
O20–C20–C21
C20–C21–N
–
–
–
–
–
–
–
–
–
–
–
–
Dihedral angles (deg)
P–O6–C6–C5
162.0
72.0
ꢀ180.0
73.6
138.2
78.7
ꢀ178.1
70.3
137.2
72.0
105.7
ꢀ168.0
ꢀ48.0
59.9
ꢀ60.3
164.5
47.4
116.5
ꢀ163.5
ꢀ43.9
59.7
ꢀ60.4
166.7
49.3
O6–C6–C5–O5
O6–C6–C5–C4
C5–O5–C1–O1
C5–O5–C1–C2
C1–O5–C5–C6
C1–O5–C5–C4
O5–C1–C2–O2
O1–C1–C2–O2
O5–C1–C2–C3
O1–C1–C2–C3
C1–C2–C3–O3
O2–C2–C3–O3
C1–C2–C3–C4
O2–C2–C3–C4
70.2
ꢀ167.2
ꢀ165.5
ꢀ159.8
ꢀ169.2
ꢀ167.4
58.4
58.5
59.2
60.3
60.2
ꢀ62.9
ꢀ175.9
61.8
ꢀ62.9
ꢀ176.8
61.1
ꢀ62.9
ꢀ176.2
60.9
ꢀ59.4
ꢀ177.1
60.9
ꢀ59.6
ꢀ176.4
61.3
179.9
56.3
ꢀ60.9
ꢀ176.5
62.2
ꢀ179.3
ꢀ178.9
ꢀ178.9
ꢀ179.7
ꢀ176.8
ꢀ177.5
57.4
57.8
57.8
56.6
58.8
58.2
56.9
54.1
ꢀ68.7
57.1
57.5
57.5
54.3
54.1
58.1
57.3
ꢀ66.2
ꢀ176.0
61.6
ꢀ65.7
ꢀ176.2
61.5
ꢀ65.8
ꢀ176.7
61.0
ꢀ69.4
ꢀ173.3
61.4
ꢀ69.5
ꢀ173.3
61.5
ꢀ66.3
ꢀ159.6
76.5
ꢀ67.0
ꢀ159.8
76.1
63.2
ꢀ51.3
ꢀ53.9
ꢀ54.2
ꢀ54.8
ꢀ54.6
ꢀ54.5
ꢀ46.4
ꢀ46.5
ꢀ176.3
ꢀ176.6
ꢀ177.2
ꢀ179.9
ꢀ179.9
ꢀ170.4
ꢀ170.6

E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
503
Table 1 (continued)
(1)
(2)
(4)
a-D-glucose
Isomer I
Isomer II
Isomer I
Isomer II
Isomer I ion
Isomer II ion
Exp. [15]
C2–C3–C4–O4
O3–C3–C4–O4
C2–C3–C4–C5
O3–C3–C4–C5
C3–C4–C5–C6
O4–C4–C5–C6
C3–C4–C5–O5
O4–C4–C5–O5
C2–C3–O3–C20
C4–C3–O3–C20
C3–O3–C20–O20
C3–O3–C20–C21
O3–C20–C21–N
O20–C20–C21–N
178.1
ꢀ62.2
52.5
172.2
ꢀ174.1
61.8
ꢀ54.6
ꢀ178.7
–
–
–
–
–
177.6
ꢀ62.8
52.2
171.7
ꢀ173.1
62.8
ꢀ53.7
ꢀ177.7
–
–
–
–
–
178.2
ꢀ62.5
53.0
172.3
ꢀ174.4
61.7
ꢀ54.3
ꢀ178.2
–
–
–
–
–
175.7
ꢀ63.8
56.6
177.1
ꢀ176.5
62.2
175.7
ꢀ63.8
56.7
177.2
ꢀ177.2
61.5
164.6
ꢀ78.7
38.4
155.0
ꢀ152.3
83.5
ꢀ36.7
ꢀ160.8
ꢀ153.2
87.6
164.6
ꢀ78.8
39.9
156.4
ꢀ155.1
81.3
ꢀ39.2
ꢀ162.8
ꢀ153.1
87.8
ꢀ65.5
53.3
62.9
ꢀ57.5
ꢀ57.4
ꢀ178.7
ꢀ94.1
144.9
ꢀ3.5
ꢀ57.8
ꢀ179.1
ꢀ95.2
143.8
ꢀ3.4
43.8
42.5
179.9
ꢀ163.2
20.2
179.9
ꢀ163.8
19.5
ꢀ133.3
ꢀ133.2
67.4
73.6
–
–
–
ꢀ109.9
ꢀ102.2
Fig. 4. Optimized structures of methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (4) isomers.
additional three protons from NHþ3 group. It allowed us to
assign the analyzed signals to the protons of a phenyl
group that is in accordance with the calculated values for
these protons. The facts mentioned above show that the
compound (4) occurs in the neutral form.
For all compounds studied the 31P NMR measurements
have been performed. For (1) only one signal at
ꢀ10.33 ppm has been registered that agrees sufficiently well
with the calculated value (ꢀ12.03 ppm). For (2) three sig-
nals have been observed, namely ꢀ1.97, ꢀ3.25 and

Table 2
Comparison of experimental and theoretical chemical shifts [ppm] of methyl (a-D-glucopyranosid-1-yl)-6-O-diphenylophosphate (1), methyl (a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (2) and
methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (4)
(1)
(2)
(4)
Exp.
Calculated
Exp.
Calculated
Exp.
Calculated
Isomer I
Isomer II
Isomer I
Isomer II
Isomer I ion
Isomer II ion
13C
C1
C2
C3
C4
C5
C6
C20
C21
95.51
98.02
72.42
74.11
77.16
69.90
69.31
102.13
73.97
75.79
72.03
73.36
67.72
96.75
72.11
74.68
69.92
71.13
67.67
97.93
72.51
74.65
71.83
72.11
67.95
99.16
69.15
77.18
67.03
70.60
64.49
167.98
40.11
98.58
71.63
76.13
68.68
70.98
67.98
165.21
44.08
98.34
71.50
75.78
68.55
71.54
67.34
165.34
44.04
98.96
69.49
85.98
73.42
68.49
67.73
158.31
44.61
53.94
146.37
114.77
121.74
115.8
98.94
69.42
85.39
72.99
68.02
65.03
159.61
44.37
71.88
73.92
69.63
70.29
68.22
C30
55.27
150.45
120.12
129.84
125.52
55.20
145.25
114.95
122.55
118.02
57.94
154.58
123.08
132.63
127.21
53.15
145.10
113.81
122.40
116.91
53.70
144.39
114.15
122.61
117.88
55.06
53.57
53.27
53.90
C(ipso)
C(orto)
C(meta)
C(para)
1H
151.79
120.35
129.80
124.38
144.26
114.35
122.58
117.84
143.92
115.78
120.62
118.26
147.50
112.02
123.29
114.96
H1
H2
H3
H4
H5
H660
O2–H
O3–H
O4–H
C21/HH0
N/HH0(PO/H)
N/H0H00H000
C30/HH0H00
C(orto)/H
C(meta)/H
C(para)/H
31P
4.67
3.37
3.72
3.40
4.46
3.34
4.74
3.21
3.64
3.21
4.88
4.22
3.57
1.81
1.62
4.77
3.53
3.66
3.47
3.72–3.80
4.12–4.30
4.33
3.14
3.41
3.28
3.80
4.19
0.87
1.77
1.31
4.68
3.60
3.37
2.87
3.30
4.21
1.43
1.77
ꢀ0.60
4.97
3.67
5.22
3.40
3.75
4.02
4.66
3.27
5.13
3.38
4.13
4.30
0.87
0.99
4.54
3.20
5.11
3.35
3.91
4.33
0.80
1.18
4.68
3.65
4.98
5.37
3.79
4.16
1.04
3.53
4.65
3.59
4.97
5.12
3.80
4.07
1.01
4.03
2.05
1.26
3.86
(3.41)
3.41
0.38(2.91)
–
3.54
7.25
3.44
0.37(3.32)
–
3.20
7.41
3.16
2.97
(3.21)
(3.11)
7.21
3.45
7.23
7.06
6.48
7.17
3.43
7.09
7.13
6.83
3.35
7.14–7.22
7.27–7.33
7.14–7.22
3.80
7.13
7.21
7.06
3.40
3.10
7.34
7.16
6.95
3.37
7.29
7.20
7.02
3.43
7.22
7.42
7.22
7.22–7.24
7.39–7.44
7.22–7.24
7.20
7.04
7.20
7.03
P
ꢀ10.33
ꢀ12.03
ꢀ1.97(2.2)
ꢀ3.25(96.2)
ꢀ10.17(1.6)
ꢀ4.55
ꢀ3.97(5.74)
ꢀ4.37(1)
ꢀ2.60
ꢀ9.32
ꢀ3.46
ꢀ4.93
ꢀ5.90
For 31P data the integrated intensities are given in parentheses.

E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
505
Fig. 5. The correlation between calculated and experimental frequencies for methyl (a-D-glucopyranosid-1-yl)-6-O-diphenylophosphate (1) and methyl (3-
O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (4).
ꢀ10.17 ppm. The last signal has been assigned to (1) as to
an intermediate product in the synthesis and the small
amount of this product can be caused by the incomplete
reaction of the sample; the difference between ꢀ10.33 and
ꢀ10.17 ppm chemical shifts can arise, probably, from the
change of the solvent during the measurements in (1) and
(2). The two remaining signals have been assigned to the
The estimated ratio, isomer I: isomer II, gives this time
15:85. We see, therefore, that in the mixture the isomer II
predominates.
4.3. Raman studies
Raman measurements were made for (1) and (4); for (2),
because of the very strong fluorescence, the spectra have
been omitted in the further analysis.
Generally, the differences between calculated vibrational
frequencies, both for neutral and zwitterionic isomers, are
very small, except for the range below 2800 cmꢀ1 where
in the case of zwitterionic forms the vibrations arising from
the NHþ3 group appear. Raman spectroscopy is not sensi-
tive enough to the changes connected with the space distri-
two possible isomers, to the isomer
I the signal
ꢀ1.97 ppm and to the isomer II the signal ꢀ4.93 ppm.
Analyzing the integral intensities of these signals, the
ratio of both isomers has been estimated as 2:98 (isomer
I: isomer II). In the case of (4) we are dealing with the
similar situations. The ꢀ3.97 ppm signal has been assigned
to the neutral form of the isomer II, whereas the
ꢀ4.37 ppm signal to the neutral form of the isomer I.
non-scaled
3500
3000
1500
1000
500
scaled
experimental
3500
3000
1500
1000
500
0
wavenumber [cm-1]
Fig. 6. Comparison of the calculated non-scaled and scaled Raman spectrum of methyl (a-D-glucopyranosid-1-yl)-6-O-diphenylophosphate (1) with the
experiment.

506
E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
non-scaled
3500
3000
1500
1000
500
scaled
experimental
3500
3000
1500
1000
500
wavenumber [cm-1]
Fig. 7. Comparison of the calculated non-scaled and scaled Raman spectrum of methyl (3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylophosphate (4)
with the experiment.
bution of the substituents at the phosphorus atom and it do
not allow to distinguish which isomer has been obtained.
Therefore, for (4) we have used for the comparison with
experimental spectra the theoretically predicted results for
the neutral isomer II, i.e., for that isomer which, according
to the NMR analysis, predominates.
vibrations should appear at 2715 and 2309 cmꢀ1 for beta-
ine forms of isomers I and II, respectively. This band has
not been found on the registered spectrum that confirms
conclusions drawn from NMR studies about the existence
of the compounds considered in the neutral form.
To calculated Raman spectra the scaling procedure has
been applied. The common scaling factor f(1) = 0.97 for (1)
and f(4) = 0.92 for (4) for all frequencies studied was taken
from the calibration of the corresponding theoretical DFT
results with the most intense experimental bands from the
range 3500–500 cmꢀ1. Vibrational assignments were made
using the visualization of vibrations by Molekel program
and basing on the assignment for D-glucose-6-phosphate
[17]. In Fig. 5, as an example, the correlation between cal-
culated and experimental frequencies for (1) and (4) is
presented.
The comparison of the theoretical scaled Raman spectra
with the experimental spectra of compounds (1) and (4) are
shown in Figs. 6 and 7, respectively. In the case of both
spectra the additional theoretical non-scaled spectra have
been inserted. Although the very simple scaling procedure
with the only one scaling factor has been applied, the sat-
isfactory agreement between theory and experiment is
observed.
We would like to focus here on the most characteristic
features of the investigated spectra only. In the phosphor-
ylated saccharides one observes pronounced band at
1200 cmꢀ1 originating from the stretching P@O mode
[17,18]. For (1) this mode correspond to 1229 cmꢀ1 and
for (4) to 1250 cmꢀ1. For (4) in opposition to (1), without
the glicydyl group, the band at 1761 cmꢀ1 is observed. The
presence of this band can suggest the existence both of
C@O and NH2 groups. It follows from the performed cal-
culations that the band corresponding to NHþ3 stretching
5. Concluding remarks
Methyl
(3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-
phenylphosphate has been obtained using the new highly
efficient method of hydrogenolysis with ammonium for-
mate as a hydrogen source and palladium on carbon as a
catalyst. This compound and two intermediate products
of synthesis have been studied by theoretical DFT method
as well as by NMR and Raman spectroscopy. Two types of
the optimized stereoisomers can be distinguished according
to the space distribution of the substituents at the phospho-
rus atom. The presence of the glycine group in methyl
(3-O-glicynyl-a-D-glucopyranosid-1-yl)-6-O-phenylphosphate
has caused the necessity to take into account the studied
molecule in the form of a zwitterion. DFT calculations,
using the B3LYP exchange-correlation functional and the
6-31G* basing set, have shown to be a powerful help in
the interpretation of NMR and Raman experiments. The
comparison of predicted chemical shifts for H, 13C and
1
31P NMR with experiment has permitted to find that the
phosphoric and glicydyl groups are attached to C6 and
C3 carbon atoms, respectively. The agreement between cal-
culated vibrational frequencies scaled by one factor only
and experimental ones is satisfactory. The calculations
have also shown that from among two possible stereoiso-
mers one predominates and that the examined compounds
occur in the neutral form. This last conclusion is certainly
important in searching of common epitopes suitable for

E. Chełmecka et al. / Journal of Molecular Structure 834–836 (2007) 498–507
507
Burant, J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B.
Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A. Petersson, H.
Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X.
Li, J.E. Knox, H.P. Hratchian, J.B. Cross, C. Adamo, J. Jaramillo, R.
Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P.
Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D.
Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-
Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill,
B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, and J.A. Pople,
Gaussian 03, Revision B.04, Gaussian, Inc., Pittsburgh PA, 2003.
[10] K. Wolinski, J.F. Hilton, P. Pulay, J. Am. Chem. Soc. 112 (1990)
8251.
construction of antimicrobial vaccine with desired broad
specificity.
Acknowledgements
This work was supported by the Polish Ministry of Sci-
entific Research and Information Technology under Grant
No. 3T09B 11229.
References
[1] H.M.I. Osborn (Ed.), Carbohydrates, Academic Press, Elsevier
Science, London, 2003.
[2] E.K.H. Schweda, J. Li, E.R. Moxon, J.C. Richards, Carbohydr. Res.
337 (2002) 409.
[11] J.L. Dodds, R. McWeeny, A. Sadlej, Mol. Phys. 41 (1980) 1419.
[12] R. Ditchfield, Mol. Phys. 27 (1974) 789.
´
[3] T. Lipinski, Ph.D. thesis. Institute of Immunology and Experimental
Therapy. Wrocław, 2000.
[13] E. Chełmecka, K. Pasterny, W. Szeja, A. Gamian, Physic. Med. XX
(Suppl. 1) (2004) 89.
[14] E. Chełmecka, Ph.D. Thesis, Institute of Physics, University of
Silesia, Katowice, 2006.
[15] G.M. Brown, H.A. Levy, Acta Crystallogr. B35 (1979) 656.
[16] S. Katti, T.P. Seshardi, M.A. Viswamitra, Acta Crystallogr. B38
(1982) 1136.
[4] H. Lis, N. Sharon, Chem. Rev. 98 (1998) 637.
[5] N.K. de Vries, H.M. Buck, Carbohydr. Res. 165 (1987) 1.
[6] W. Szeja, M. Gawlik-Je˛drysiak, A. Gamian, Deprotection of diphenyl
phosphate. The convenient synthesis of sugar phenyl phosphates.
Tetrahedron Lett, Submitted for publication.
[7] M. Tamura, T. Miyoshi, N. Mori, K. Kinomura, M. Kawaguchi,
et al., Agr. Biol. Chem. 54 (1990) 1401.
[17] K. Pasterny, E. Chełmecka, R. Wrzalik, W. Szeja, J. Mol. Struct.
744–747 (2005) 929.
[8] T.W. Greene, P.G.M. Wuts (Eds.), Protective Groups in Organic
Synthesis, John Wiley & Sons, New York, 1999.
[18] G. Socrates, Infrared and Raman Group Frequencies: Tables and
Charts, third ed., Wiley, New York, 2004.
[9] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C.