CD and NMR assignment of the anomeric configuration of 4-(5-deoxy-α,β-l-arabinofuranosyl)-2-phenyl-2H-1,2,3-triazole C-nucleoside analogs

Article abstract of DOI:10.1016/j.carres.2009.11.002

Acid-catalyzed dehydrative cyclization of 5-deoxy-l-manno-pentitol-1-yl)-2-heptulose bisphenylhydrazone and subsequent reflux with copper sulfate gave an anomeric mixture of 4-(5-deoxy-α,β-l-arabinofuranosyl)-2-phenyl-2H-1,2,3-triazole C-nucleoside analog

Full text of DOI:10.1016/j.carres.2009.11.002

Carbohydrate Research 345 (2010) 341–345
Contents lists available at ScienceDirect
Carbohydrate Research
journal homepage: www.elsevier.com/locate/carres
CD and NMR assignment of the anomeric configuration of 4-(5-deoxy-a,b-L-ar-
abinofuranosyl)-2-phenyl-2H-1,2,3-triazole C-nucleoside analogs q
*
Mohammed A. E. Sallam
Chemistry Department, Faculty of Science, Alexandria University, Alexandria, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
Acid-catalyzed dehydrative cyclization of 5-deoxy-
one and subsequent reflux with copper sulfate gave an anomeric mixture of 4-(5-deoxy-
anosyl)-2-phenyl-2H-1,2,3-triazole C-nucleoside analogs. The mixture was separated by chromatography,
and the anomeric compositions configurations of the components were determined by CD, NMR, mass
spectroscopy, and acylation.
L
-manno-pentitol-1-yl)-2-heptulose bisphenylhydraz-
,b- -arabinofur-
Received 19 September 2009
Accepted 3 November 2009
Available online 27 November 2009
a
L
Keywords:
Ó 2010 Published by Elsevier Ltd.
4-(5-Deoxy-
phenyl-2H-1,2,3-triazole
5-Deoxy- -arabinofuranosyl triazoles
L-manno-pentitol-1-yl)-2-
L
C-Nucleosides
Circular dichroism
NMR spectroscopy
Triazole rule
1. Introduction
the C-nucleoside analogs, namely, 4-(5-deoxy-a- and b-L-arabino-
furanosyl)-2-phenyl-2H-1,2,3-triazoles (2) and (3), respectively
(Scheme 1). The two anomers were separated by column chroma-
tography on an anion-exchange resin with gradient elution using
aqueous methanol. The anomeric configuration of the products
was misassigned, on the basis of NMR spectroscopy from the
chemical shift of the methyl and H-40 signals, analogous to the pro-
cedure used for 6-deoxy pyranosides.10
Recently, we used circular dichroism for assigning the anomeric
configuration of C-glycosyl-2-phenyl-2H-1,2,3-triazoles. The CD
spectra of a series of 4-(tetrahydroxytetryl-1-yl)-2-phenyl-2H-
1,2,3-triazoles11,12 and 4-(pentahydroxypentyl-1-yl)-2-phenyl-
2H-1,2,3-triazoles8,13 were studied. A correlation between the sign
of the Cotton effect at the maximum UV absorption and the abso-
We have been interested recently in the synthesis of C-nucleo-
side analogs by the acid-catalyzed dehydrative cyclization of poly-
hydroxyalkyl heterocycles.1–7 This reaction gives an anomeric
mixture of C-nucleosides. The dehydrative cyclization of a tetra-
hydroxytetryl heterocycle side chain is stereospecific, affording
glycofuranosyl C-nucleoside anomers having in the cyclized prod-
uct a trans arrangement of the base moiety and the 20-OH group.
On the other hand, the cyclization of a pentahydroxypentyl side
chain2,7 in acid medium is not so stereospecific, with the formation
of anomeric pairs of glycofuranosyl and glycopyranosyl C-nucleo-
side analogs with and without inversion at C-10. In this work, the
anomeric configurations of the formed C-nucleosides were deter-
mined by CD measurements in light of the CD-triazole rule8 and
were confirmed by NMR spectroscopic measurements.
lute configuration of the hydroxyl group
ety was obtained. Compounds having hydroxyl groups
triazole base moiety and to the right in a Fischer projection for-
mula ( -glycero or S chirality) show a positive Cotton effect at
the maximum UV absorption. Those having the hydroxyl group
to the left in a Fischer projection ( -glycero or R chirality) show
a
to the triazole base moi-
a
to the
D
2. Results and discussion
L
Treatment of 4-(5-deoxy-L-manno-pentitol-1-yl)-2-heptulose
negative Cotton effect. This correlation has been stated as the
CD-triazole rule8 for anomeric configuration assignment of C-
nucleoside triazole analogs. This correlation was applied for glyco-
furanosyl,11,12 glycopyranosyl8,13, and 50-hydroxymethylfuranosyl
C-nucleoside triazole analogs8,13 without exception. In this work,
it is extended to 50-deoxyfuranosyl analogs. The CD spectrum of
1 (Fig. 1) showed a positive Cotton effect at the maximum UV
absorption, manifested by the right-hand configuration of the
bisphenylhydrazone (1a) with methanolic sulfuric acid and subse-
quent reflux with copper sulfate afforded an anomeric mixture of
q
Presented at the 11 International Symposium on Spin, magnetic field effects in
chemistry, related phenomena, Brook University, St. Catharines, Ontario, Canada, 9–
14 August, 2009.
* Tel./fax: +20 3 4812110.
E-mail address: maesallam@yahoo.com
0008-6215/$ - see front matter Ó 2010 Published by Elsevier Ltd.
doi:10.1016/j.carres.2009.11.002

342
M. A. E. Sallam / Carbohydrate Research 345 (2010) 341–345
configuration of the ring oxygen of its Fischer projection formula,
that is, the -configuration of the furanosyl group formed. The
a-L
CD spectrum of compound 3 showed a negative Cotton effect in
the same region in accord with the left-hand configuration of the
ring oxygen, that is, the b-L-configuration of the furanosyl group
formed. Compound 3 showed a higher spectrum amplitude than
did compound 2, which can be attributed to a different population
of conformations. This CD assignment of the anomeric configura-
tion of compounds 2 and 3 is opposite to the assignment obtained9
from NMR measurements. Similar correction for the NMR assign-
ment of the anomeric configuration by CD measurements was
obtained.8
The CD anomeric assignment of 2 and 3 was ascertained by
applying different NMR criteria. Compound 2 showed the anomeric
proton H-10 as a doublet at d 4.83 (J1 ,2 6.1 Hz) (Table 1). Compound
0
0
3 showed the anomeric proton H-10 as a doublet at d 5.09 (J1 ,2
0
0
5.6 Hz). The anomeric configuration cannot be obtained from the
0
0
coupling constant J1 ,2 values, since these values are not consis-
tently diagnostic. However, comparing the chemical shift values
of H-10, compound 3 has the anomeric proton at lower field (d
5.09) than compound 2 (d 4.83). This is in accord7 with the cis
arrangement of H-10 and H-20 for compound 3, that is, the 50-
deoxy-b-
the anomeric proton upfield (d 4.82) was assigned the trans
arrangement of H-10 and H-20, that is, the 5-deoxy-
-arabinofur-
L-arabinofuranosyl configuration. Its anomer 2 having
a-L
anosyl configuration. Analogously, the anomeric proton (H-10) for
the acetyl derivative 4 was shown upfield (d 5.28) to that of the
anomeric analogue 5 (d 5.39) in accord with the trans arrangement
of H-10 and H-20 for 4 and cis arrangement of H-10 and H-20 for 5.
Compound 2 was obtained from 1 without inversion in the config-
uration at C-10, and 3 was obtained from 1 with inversion in the
configuration at C-10.
The anomeric configurations of some rhamnopyranosyl ano-
mers14 were obtained from the chemical shifts of the methyl peaks.
The C-50–CH3 peak for the
the b-anomer. For 5-deoxy-
compound 2 showed the C-40–CH3 peak at higher field (d 1.21) than
that for compound 3 (d 1.33), in accord with the -configuration
for 2 and b- -configuration for 3. The opposite correlation was ob-
tained for the H-40 signal for 2 and 3. The
-anomer 2 showed
deshielding of the H-40 peak (d 3.91) relative to that for the b-
a
-anomer is located at higher field than
L-arabinofuranosyl anomers 2 and 3,
Figure 1. CD spectrum of 4-(5-deoxy-
triazole 1b in methanol.
L-manno-pentitol-1-yl)-2-pheny1-2H-1,2,3-
a-L
L
a
-L
L-
anomer 3 (d 3.75). Analogously, the acetyl derivatives 4 and 5
showed the same correlation (Table 1). The C-40–CH3 for com-
pound 4 was shown upfield (d 1.45) to that for compound 5 (d
1.51), and the H-40 signal for compound 4 was deshielded (d
4.28) relative to that for compound 5 (d 4.07), in accord with the
a-L
-configuration for 4 and the b-L-configuration for 5.
The anomeric configuration of 2 and 3 can also be confirmed
13
from the C NMR data (Table 2). The anomeric carbon atom C-10
for compound 2 was shown with a downfield chemical shift (d
77.2) relative to that for 3 (d 76.6), in accord7 with the trans
arrangement of H-10 and H-20 for 2 and cis arrangement of H-10
and H-20 for 3. Similar observations were evident for compounds
4 and 5. Compound 4 showed a downfield shift of C-10 (d 77.1) rel-
ative to that for 5 (d 76.9). This is in accord with the
tion for and the b- -configuration for 5. The anomeric
configuration can be confirmed from the 13C chemical shift of the
CH3 peak for compounds 2 and 3. The -anomer 2 showed the
methyl peak at higher field (d 20.6) than that for the b- -anomer
3 (d 20.7). Similarly, the methyl peak of the acetyl anomer 4
was shown at higher field (d 20.0) than that for the b- acetyl ano-
mer 5 (d 20.8). The C chemical shift of C-40 showed the opposite
correlation. The C-40 of the
-anomers 2 (d 82.4) and 4 (d 79.5)
were more deshielded than the corresponding resonances for the
b- -anomers 3 (d 81.8) and 5 (d 77.4), respectively.
a-L configura-
4
L
a
-L
L
Figure 2. CD Spectra of 4-(5-deoxy-
azole (2) (—————) and 4-(5-deoxy-b-
azole (3) (—).
a
-
L
-arabinofuranosyl)-2-phenyl-2H-1,2,3- tri-
a-L
L-arabinofuranosyl)-2-phenyl-2H-1,2,3- tri-
L
13
10-OH group. The CD spectra of compounds 2 and 3 (Fig. 2) showed
opposite Cotton effects in the same region. Compound 2 showed a
positive Cotton effect identical to 1, indicating that the right-hand
a-L
L

M. A. E. Sallam / Carbohydrate Research 345 (2010) 341–345
343
Table 1
Chemical shift (d) and first-order coupling constants (J, Hz)a for compounds 1–5 at 500 MHz
Compd
Glycosyl moiety
H–40
2-Phenyl-2H-1,2,3-triazole moiety
H–10
H–20
H–30
OH
OAc
CH3
H-5
H-o
H-m
H-p
1bb
4.72d
3.93d
3.43d
3.58m
5.49d
J 5.4
4.46
J 5.4
4.27
J 8.4
4.22
J 7.7
1.11d
J 6.2
7.94s
7.96d
7.52t
7.36t
0
0
0
0
0
0
J1 ,2 9.2
J2 ,3 0.0
J3 ,4 7.7
2b
4.83d
4.16q
3.62d
2.47d
3.91m
3.75m
5.01d
J 5.4
5.31 d
J 3.8
1.21d
J 11.1
8.06s
7.90s
7.96d
7.53t
7.52t
7.38
0
0
J1 ,2 6.1
3b
5.09d
3.99m
5.33
J 3.9
5.12
J 5.4
1.33d
J 6.3
4.46d
J 8.6
7.37t
0
0
J1 ,2 5.6
4c
5c
5.28d
5.60t
J 2.9,
J 2.3
5.02t
J 2.4
4.28m
4.07m
2.14s
2.03s
1.45d
J 6.3
7.82s
7.83s
8.05d
8.02d
7.47t
7.46t
7.34t
7.34t
0
0
J1 ,2 3.5
5.39d
6.96d
J 3.1
4.94t
J 1.6
J 2.3
2.14s
1.96 s
1.51d
J 6.1
0
0
J1 ,2 3.9
a
b
c
J values after the addition of CD3CO2D.
In Me2SO-d6.
In CDCl3.
Table 2
13C NMR data and chemical shifts (d) for compounds 1–5 at 125.8 MHz
Compd
Glycosyl moiety
2-Phenyl-2H-1,2,3-triazole moiety
C–10
C–20
C–30
C–40
CO
OCH3
CH3
C-4
C-5
C-a
C-o
C-m
C-p
1a
2a
3a
4b
66.3
77.2
76.6
77.1
71.8
79.4
78.8
82.0
73.8
82.8
83.0
82.5
66.7
82.4
81.4
79.5
21.4
20.6
20.7
153.8
135.2
139.7
139.8
139.9
139.7
136.8
134.0
135.4
151.1
148.3
147.7
118.5
118.8
130.1
119.0
130.2
130.2
118.6
129.3
127.8
128.1
127.9
127.7
170.3
170.1
170.0
169.3
21.0
18.5
21.0
19.1
20.00
5b
76.9
77.1
82.5
77.4
20.8
139.7
135.2
144.9
118.9
129.4
127.7
a
In Me2SO-d6 + CD3CO2D.
In CDCl3.
b
and the
a-L
-configuration of 2. Similar correlation12 was observed
-threofuranosyl ring. The b- -threofur-
anosyl anomer was more negative in its optical rotation than the
-threofuranosyl anomer.
for compounds having the
L
L
a-L
3. Conclusions
The anomeric configuration of 4-(5-deoxy-a- and b-L-arabino-
furanosyl)-2-phenyl-2H-1,2,3-triazole analogs was reinvestigated
in light of the CD-triazole rule for anomeric assignment. The ano-
mer having the ring oxygen at C-10 on the Fischer projection for-
Figure 3. NOE difference for 4-(2,3-di-O-acety1-5-deoxy-b-
phenyl-2H-1,2,3-triazole 5.
L-arabinofuranosyl)-2-
mula to the right (
Cotton effect at the maximum UV absorption in accord with the
-configuration. The anomer having the ring oxygen to the left
-glycero or R chirality) showed a negative Cotton effect in accord
with the b-
-configuration. This indicates that the CD-triazole rule8
is applicable for 5-deoxyl- -arabinofuranosyl anomers, in addition
D-glycero or S chirality) showed a positive
The NOE experiment is an even more reliable12 tool for anomer-
ic assignment. The NOE difference for 5 (Fig. 3) showed 5%
enhancement at H-40 upon irradiation at H-10. This indicates the
a
-L
(L
presence of H-10 and H-40 on the same face of the b-
L-furanosyl
L
ring.
L
to the glycofuranosyl, glycopyranosyl, and 50-hydroxymethyl
furanosyl analogs. This allows the correction of the anomeric con-
figurations obtained solely from NMR measurements and con-
firmed by different NMR criteria. Compound 2 was obtained from
The specific rotation at the D-line for compound 3 was more
negative ([ ꢀ28.7), indicating cis
ꢀ138.4) than that for 2 ([
relationship of the base moiety and the tail CH3 group at the 5-
deoxy- -arabinofuranosyl ring, that is, the b- -configuration of 3
a]
a]
D
D
L
L

344
M. A. E. Sallam / Carbohydrate Research 345 (2010) 341–345
1 without inversion in the configuration at the carbon atom
a
to
and a 500 lg/mL solution of sodium formate in 0.1% formic acid
the triazole base moiety, and compound 3 was obtained with
inversion. The acid-catalyzed dehydrative cyclization of 5-deoxy-
solution was used as the standard calibration solution. Combustion
analyses were carried out at the Chemistry Department, Faculty of
Science, Cairo University.
L-manno-pentitol-1-yl side chain is a nonstereospecific process. It
is therefore necessary to use more than one spectroscopic method
for anomeric assignment of C-nucleosides.
4.2. 4-(5-Deoxy-L-manno-pentitol-1-yl)-2-phenyl-2H-1,2,3-triazole
(1b)
4. Experimental
Compound 1b was prepared from 7-deoxy-L-manno-pentitol-1-
yl-heptulosebisphenylhydrazone 1a by refluxing with copper sul-
fate as in the standard procedure. Compound 1a was prepared
from the precursor 7-deoxy heptoses15,16 by the action of phen-
ylhydrazine. Compound 1b was obtained as colorless needles:
4.1. General methods
Melting points were determined with a Fisher–Johns instru-
ment and are uncorrected. Evaporations were performed under
diminished pressure below 70 °C. Thin-layer chromatography
(TLC) was conducted on silica gel (Kieselgel G, E. Merck, Darmstadt,
Germany) with solvent A, 3:1 toluene–MeOH; B,1:3 EtOAc–n-hex-
ane. Compounds were detected under shortwave UV light at
254 nm. Optical rotations were recorded with Perkin–Elmer 141
and Jasco P1030 instruments (10 cm, 1 mL microcell). CD measure-
ments were recorded for solutions in MeOH on a Jasco 720 WI
spectropolarimeter at concentrations 0.25–0.3 mg/mL MeOH
(0.2 mL microcell). 1H NMR spectra were recorded with a Jeol Ex
400 MHz spectrometer using tetramethylsilane as the internal
standard. 13C NMR spectra were recorded with a Jeol/JNM ECA
500 at 125 MHz. Assignment of peaks was verified by 1H–13C COSY
experiments. Mass spectra were recorded with electrospray-ioniz-
aton Microtof instrument. The sample was dissolved in methanol,
mp 152–54 °C; Rf 0.44 (A); [
266.9 (log
, 4.3). For 1H NMR and 13C NMR data, see Tables 1 and
2; circular dichroism data in MeOH at 22 °C: observed CD; k ( ),
a]D +22.2 (c 1.01, MeOH); kmax (MeOH)
e
D
e
266.9 (+0.99). Anal. Calcd for C13H17N3O4: C, 55.91; H, 6.14; N,
15.04. Found: C, 56.13; H, 6.39; N, 14.94.
4.3. 4-(5-Deoxy-b-L-arabinofuranosyl)-2-phenyl-2H-1,2,3-trazole
(3)
Compound 3 was separated from the anomeric mixture on a col-
umn of Dowex-1ꢁ8 (OHꢀ) anion-exchange resin by elution with 60%
MeOH by a standard procedure.9 Recrystallization of the solid ob-
tained from water gave colorless needles: mp 120–122 °C. (lit.9
mp 118–120 °C); [
(log
4.3). For 1H NMR and 13C NMR data, see Tables 1 and 2; circular
dichroism data in MeOH at 22 °C; observed CD; k ( ), 265.4 (ꢀ2.9).
a]
D
ꢀ138.4 (c 1.5, MeOH); kmax (MeOH) 265.8
e
D
e
4.4. 4-(5-Deoxy-a-L-arabinofuranosyl)-2-phenyl-2H-1,2,3-triazole
(2)
Compound 2 was eluted from the column of Dowex-1ꢁ8 (OHꢀ)
ion-exchange resin with 90% MeOH. The solid residue was recrys-
tallized from benzene–n-hexane as colorless needles: mp 88 °C
(lit.9 mp 88 °C); [
a
]
ꢀ28.7 (c 3.0, MeOH); kmax 266.6 (log
e
4.3).
For 1H NMR and 13C NMR data see Tables 1 and 2; circular dichro-
ism data in MeOH at 22 °C; observed CD; k ( ), 260.2 (+0.4).
D
D
e
4.5. 4-(2,3-Di-O-acetyl-5-deoxy-a-L-arabinofuranosyl)-2-phenyl-
2H-1,2,3-triazole (4)
Compound 2 (10 mg, 0.02 mmol) in pyridine (2 mL) was treated
with Ac2O (2 mL) and kept at room temperature for 12 h. The mix-
ture was evaporated until dry, and traces of pyridine were re-
moved by spin coevaporation with toluene. The dry residue was
chromatographed on a short column (1 ꢁ 10 cm) of Silica Gel 60,
eluting with solvent B, giving a colorless syrup (yield 17 mg). For
1H and 13C NMR data, see Tables 1 and 2; HRMS of the molecular
ion peak calcd for C17H19N3O5Na: m/z 368.1222; found, m/z
368.1226.
4.6. 4-(2,3-Di-O-acetyl-5-deoxyl-b-L-arabinofuranosyl)-2-phenyl-
2H-1,2,3-triazole (5)
Compound 3 (13 mg, 0.037 mmol) in pyridine (2 mL) was trea-
ted with Ac2O (2 mL) and kept at room temperature for 24 h. The
mixture was poured into crushed ice, extracted with CH2Cl2, and
the organic layer was separated, washed with satd aq CuSO4, and
water, then dried over anhyd Na2SO4 and filtered. The filtrate
was evaporated to a syrup that was chromatographed on a short
column (1 ꢁ 10 cm) of Silica Gel 60, eluting with solvent B. It gave
a colorless syrup (20 mg); Rf 0.67 (B). For 1H and 13C NMR data, see
Tables 1 and 2; HRMS of the molecular ion peak calcd for
C17H19N3O5Na: m/z 368.1222; found, m/z 368.1217.
Scheme 1. Synthesis and anomeric assignment of 4-(5-deoxy-
furanosyl)-2-phenyl-2H-1,2,3-triazole C-nucleoside analogs.
a and b-L-arabino-

M. A. E. Sallam / Carbohydrate Research 345 (2010) 341–345
345
3. Sallam, M. A. E.; El-Nahas, H. M.; Abdel Megid, S. M. E.; Anthonsen, T.
Carbohydr. Res. 1996, 280, 127–138.
Acknowledgments
4. Sallam, M. A. E.; El-Shenawy, H. Carbohydr. Res. 1994, 261, 327–334.
5. Sallam, M. A. E. Nucleosides Nucleotides 1982, 1, 297–313.
6. Sallam, M. A. E.; Ibrahim, E. I.; El-Eter, K. A. A.; Cassady, J. M. Carbohydr. Res.
1997, 298, 93–104.
7. Sallam, M. A. E.; Abdel Megid, S. M. E.; Townsend, L. B. Carbohydr. Res. 2001,
330, 53–63.
This work was partially supported by the Ministry of Economic
and International Cooperation, Egypt. The author thanks Professor
Jonanthan R. Parquette, Department of Chemistry, The Ohio State
University, for the electrospray-ionization HRMS. Thanks also to
Professor N. Harada, Professor M. Watanabe, and Dr. S. Kuwahara,
Tohoku University for the J720 CD measurements; and Dr. W. R.
Browne, Organic Chemistry Laboratory, University of Gröningen,
Gröningen, The Netherlands, for refining some NMR and CD
measurements.
8. Sallam, M. A. E. Chirality 2006, 18, 790–798.
9. Sallam, M. A. E. Carbohydr. Res 1982, 106, 71–82.
10. Laffite, C.; Phuoc Du, A. N.; Wnternitz, F.; Wylde, R.; Sosa, F. P. Carbohydr. Res.
1978, 67, 91–103.
11. Sallam, M. A. E. Enantiomer 2002, 7, 283–286.
12. Sallam, M. A. E.; Louis, F. F. Chirality 2004, 16, 331–335.
13. Sallam, M. A. E. J. Carbohydr. Chem. 2009, 28, 498–505.
14. Sinclair, H. B.; Sleeter, R. T. Tetrahedron Lett. 1970, 11, 833–836.
15. Fischer, E.; Piloty, O. Ber. Deutsch. Chem. Ges. 1890, 23, 3102–3110.
16. Jackson, E. L.; Hudson, C. S. J. Am. Chem. Soc. 1953, 75, 3000–3002.
References
1. Sallam, M. A. E. Carbohydr. Res. 1978, 67, 79–89.
2. Sallam, M. A. E. J. Chem. Soc., Perkin Trans. 1 1982, 557–562. and references cited
therein.