Dye-sensitized photooxygenation of sugar-furans as synthetic strategy for novel C-nucleosides and functionalized exo-glycals

Article abstract of DOI:10.1016/j.tet.2006.07.109

The methylene blue-sensitized photooxygenation of β-ribofuranosyl furan 1e followed by in situ Et₂S treatment afforded the conformationally stable β-ribofuranoside 4e almost quantitatively. The latter was converted to pyridazine C-nucleoside 6e by cyclization with NH₂NH₂ and to pyrazoline 7e through a 1,3-dipolar cycloaddition with diazomethane. Attempts to epoxidize the double bond failed both by dimethyldioxirane (DMDO), which left 4e unchanged, and by NEt₃/t-BuOOH or NaOO-t-Bu which respectively afforded the new and unexpected exo-glycals E,Z-8e and the novel furan derivative 9.

Full text of DOI:10.1016/j.tet.2006.07.109

Tetrahedron 62 (2006) 10694–10699
Dye-sensitized photooxygenation of sugar-furans as synthetic
strategy for novel C-nucleosides and functionalized exo-glycals
*
Flavio Cermola and M. Rosaria Iesce
Dipartimento di Chimica Organica e Biochimica, Universitaꢀ di Napoli Federico II, Complesso Universitario di Monte Sant’Angelo,
Via Cinthia 4, 80126 Napoli, Italy
Received 5 June 2006; revised 26 July 2006; accepted 27 July 2006
Available online 18 September 2006
Abstract—The methylene blue-sensitized photooxygenation of b-ribofuranosyl furan 1e followed by in situ Et2S treatment afforded the con-
formationally stable b-ribofuranoside 4e almost quantitatively. The latter was converted to pyridazine C-nucleoside 6e by cyclization with
NH2NH2 and to pyrazoline 7e through a 1,3-dipolar cycloaddition with diazomethane. Attempts to epoxidize the double bond failed both
by dimethyldioxirane (DMDO), which left 4e unchanged, and by NEt3/t-BuOOH or NaOO-t-Bu which respectively afforded the new and
unexpected exo-glycals E,Z-8e and the novel furan derivative 9.
Ó 2006 Published by Elsevier Ltd.
1. Introduction and background
is the dienophile,5 as well as with good yields.6 [4+2] Cyclo-
addition of singlet oxygen to furans has been applied
with success in approaching highly functionalized mole-
cules, which in turn can serve as building blocks in construc-
tions of more complicated organic structures.4 The reaction
quantitatively affords 2,3,7-trioxabicyclo[2.2.1]hept-5-enes,
usually named endoperoxides (Fig. 1),4,5 which are ther-
mally unstable. The rearrangement pathways strictly depend
on the electronic nature of the substituents on the furan ring,
providing useful approaches to a wide range of functional-
ized cyclic or acyclic compounds.4,5,7
C-Glycoside synthesis represents a field of great interest due
to the potential biological relationships, which often charac-
terize these molecules.1 Many synthetic approaches have
been developed and others are being investigated for the pur-
pose of furnishing more efficient procedures.2 Two main
synthetic pathways provide C-nucleosides as well as glyco-
sides in general.3 The most common approach is based on
promoted coupling between a donor sugar and an acceptor,
the latter being the desired aglycone.2,3 The other approach
involves transforming a suitable pre-existent aglycone by
means of regio- and stereoselective reactions.3 Although
the first approach has wider applicability, it has definite
limitations including isomerization or decomposition of the
acceptor during the reaction due to the harsh conditions
required. When this approach fails, the second strategy could
represent the only possibility to achieve the target molecule.3
Previous studies aimed at controlling dye-sensitized photo-
oxygenation of furans bearing a sugar and investigated the
reaction of glucosyl furans a-1a,b (Scheme 1).8 The results
showed that the sugar ring at the a-position of a furan
strongly influences the chemical behaviour of the corre-
sponding endoperoxides 2a,b; they thermally rearranged
into the corresponding a-O-glycosides 3a,b through a
Baeyer—Villiger-type mechanism (Scheme 1). This uncom-
mon C- to O-glycoside transposition is promoted by the elec-
trophilicity of the anomeric carbon and does not depend on
the ring-sugar size. Indeed, similar results were obtained
starting from the arabinosyl furans 1c,d, which led quantita-
tively to the O-furanosides 3c,d (Scheme 1).9 Moreover, the
use of the b-anomer of 1c confirmed that the observed
Furans are versatile and useful starting materials in the field
of organic synthesis. For example, they are good diene-type
compounds in [4+2] cycloaddition with singlet oxygen
(1O2).4 This excited state of oxygen can be easily produced
starting from atmospheric oxygen, solar light and a sensi-
tizer, which together constitute a reaction system of very
low environmental impact. On the other hand, pericyclic
reactions are known to proceed with very high regio- and
stereoselectivity, the first of which fails when singlet oxygen
O
Keywords: Photooxygenation; [4+2] Cycloadditions; C-Nucleosides; 1,3-
Dipolar cycloadditions.
* Corresponding author. E-mail: cermola@unina.it
O
O
Figure 1. 2,3,7-Trioxabicyclo[2.2.1]hept-5-ene.
0040–4020/$ - see front matter Ó 2006 Published by Elsevier Ltd.
doi:10.1016/j.tet.2006.07.109

F. Cermola, M. R. Iesce / Tetrahedron 62 (2006) 10694–10699
10695
stereoselectivity is due to retention of the configuration in
the sugar moiety from C- to O-migration (Scheme 1).9
Subsequently, we decided to apply the photooxygenation
procedure to b-ribofuranosides in the hope of producing
novel C-nucleosides of pharmacological interest. Thus we
synthesized the b-ribofuranosyl furan 1e,9 previously un-
known, to obtain functionalized furanosides with configura-
tionally stable unsaturated aglycones as well as to avoid
the C- to O-rearrangement. The photooxygenation of b-1e
in dichloromethane at ꢀ20 ꢁC was complete after 30 min
(TLC) and afforded the 6:1 diastereomeric mixture of the
endoperoxide 2e, which by Et2S reduction quantitatively
gave the cis-diketone 4e (Scheme 4). As expected, com-
pound 4e showed a configurational stability, and did not
isomerize at all into the corresponding trans-alkene.9 Here
we report some synthetic applications based on the use of the
b-ribofuranoside 4e as starting material for interesting new
C-nucleosides and exo-glycals. Compound 4e is easily pre-
pared by a one-pot procedure based on the dye photooxy-
genation of the furanosyl furan 1e followed by in situ Et2S
reduction (Scheme 4). Due to the high yield as well as the
stereoselectivity, compound 4e can be used without chroma-
tographic purification.
1O2
R
R
O
Sugar
O
CH2Cl2, -20 °C
O
Sugar
O
2
1
-20 °C
r.t.
O
R
Sugar
O O
3
O
BnO
BnO
a (b); R = H (Me); Sugar =
c (d); R = H (Me); Sugar =
OBn
OBn
O
BnO
BnO OBn
Scheme 1.
Although thermally unstable, the endoperoxides 2 could be
used in approaching novel and functionalized C-glycosides.
Indeed, when the photooxygenation mixtures of 2 were
treated with a precooled (ꢀ20 ꢁC) solution of Et2S, the
C-glycosides 4 were formed almost quantitatively (Scheme
2).8,9 They were obtained with complete cis-stereoselectiv-
ity, owing to the bicyclic structure of the parent endoper-
oxides, but the use of chloroform or other slightly acid
conditions promoted a complete isomerization into trans-
derivatives 5 (Scheme 2).
Me
1O2, CH2Cl2
O
O
2e
AcO
Me
(6:1 diastereomeric mixture)
Et2S
AcO
OAc
1e
1. 1O2, MeOH
70% 2. Et2S
Me
O
O
O
1O2
AcO
2
1
Me
CH2Cl2, -20 °C
AcO
OAc
Et2S
4e
O
Scheme 4.
R
R
CDCl3
Sugar
O
Sugar
O
O
2. Results and discussion
5
4
Scheme 2.
The cis-relationship of the two carbonyl groups of b-ribofur-
anoside 4e and its configurational stability led us to carry out
a cyclization into a pyrimidine-base system by the addition
of hydrazine hydrochloride. The reaction was performed us-
ing dry methanol as the solvent and led to the new pyridazine
C-nucleoside b-6e (Scheme 5). A one-pot process starting
The development of the one-pot procedure shown in
Scheme 3 is of considerable interest. It offers an easy
synthetic approach for b-C-glycosides such as the derivative
b-5b, starting from the corresponding a-glucosyl furans, for
example, a-1b, which can be more easily prepared than the
b-analogues.8
Me
N
N
O
.
NH2NH2 HCl
AcO
4e
1. 1O2
Me
MeOH, rt
AcO
70%
OAc
2. Et2S
3. SiO2
O
6e
R
R
Sugar
O
Sugar
O
1. 1O2, MeOH
2. Et2S
1b
5b
O
BnO
3. NH2NH2·HCl
b; R = Me; Sugar =
BnO
OBn
OBn
1e
Scheme 3.
Scheme 5.

10696
F. Cermola, M. R. Iesce / Tetrahedron 62 (2006) 10694–10699
from furan 1e was then developed and afforded compound
6e in 70% yield (based on furan 1e).9
Nevertheless, considering the cycloadducts similar to the
transition states, we tentatively assigned the 40R,50R config-
urations to 7e, on the basis of comparison of calculated
thermodynamical stabilities of the two isomers. Indeed,
a MM+ conformational analysis assigned a lower energy to
the cis-(RR)-7e of ca. 5 kcal/mol compared to the cis-(SS)-
pyrazoline isomer 7e0 (Fig. 2). All the spectroscopic data
were collected on the crude reaction mixture, and this shows
that the cycloaddition proceeds with very high yields and
with almost complete regio- and stereoselectivity.
The presence of the peracetylated sugar ring as well as of
two acetyl groups at the unsaturated carbons suggested the
use of 4e as a dipolarophile in [4+2] cycloadditions with
electron-rich diene compounds as well as in [3+2] reaction
with suitable 1,3-dipoles, providing six- and five-membered
cyclic compounds, respectively. As 1,3-dipoles, diazo-
alkanes are usually able to give cycloadditions, which are
HOMO-controlled by the dipole, the reactions proceeding
faster with electron-poor alkenes.10 Thus we chose diazo-
methane as the dipole and the reaction was carried out by
adding an excess of a freshly prepared ether solution of
CH2N2 to the ribofuranoside 4e (0.5 mmol), while stirring
at room temperature. After 30 min the reaction was com-
plete (TLC), and the 1H NMR spectrum showed the presence
of an unknown derivative as the main product. On the basis of
2D-NMR experiments, we assigned the C-pyrazoline struc-
ture 7e to the new product (Scheme 6).
We next tested the possibility of oxidizing the double bond
of 4e using diverse epoxidizing reagents in order to achieve
a new C-nucleoside characterized by an epoxidic aglycone.
Treatment at ꢀ20 ꢁC of a crude glycoside 4e with freshly
prepared dimethyldioxirane (DMDO) did not work, even
after a week. This was only to be expected considering the
electron-poor character of the double bond together with
the electrophilicity of the oxidant.11 We then attempted to
epoxide the alkene moiety by using a t-BuOOH/NEt3 mix-
ture, a nucleophilic oxidizing system usually used in the
epoxidation of alkenes substituted with two or three electron-
withdrawing groups.12 Unlike DMDO, the hydroperoxide/
triethylamine solution generally leads to a mixture of cis-
and trans-epoxides, the mechanism involved not being con-
certed.12 The reaction was carried out by adding 6 equiv of
the epoxidizing system to a stirred dichloromethane solution
of 4e at room temperature. Surprisingly, the work-up of the
reaction after 12 h did not afford trace of any stereoisomeric
epoxide; instead it produced the diastereomeric exo-glycals
E,Z-8e in very high yields (Scheme 7). The molar ratio of
E- and Z-8e changed from an initial 9:1 to 3:2 in the course
of time.
Ac
Ac
O
CH2N2
N
AcO
4e
N
Et2O, rt
AcO
OAc
cis-(RR)-7e
1. 1O2, CH2Cl2
2. Et2S
40%
3. CH2N2, Et2O, rt
1e
Scheme 6.
The structure of exo-glycals for the two observed products
was assigned on the basis of spectral data. The stereochem-
istry was firstly assigned by correlating the large differences
in chemical shifts of the AB system protons of CH2CO group
to the different spacial closeness to the sugar chiral centres
(ca. 0.5 ppm in E-8e vs 1 ppm in Z-8e). The assigned stereo-
chemistry was then confirmed by 2D-NOESY experiments
run on the diastereomeric mixture. In particular, the 2D-
spectrum showed a strong NOE correlation between the
methylene protons with the larger AB system and the
The correctness of the assigned regiochemistry was easily
established by the presence of the CH–CH2 system in
the H NMR spectrum, which has to be absent in the re-
1
gioisomer. The cis-facial relationship of the two acetyl
groups was given on the basis of the concerted mechanism,
which represents the usual pathway working in 1,3-dipolar
cycloaddition reactions.10 At present the configurations
of the two new chiral centres have still to be assigned.
Figure 2. Conformational analysis (MM+) for the cis-(RR)-7e and (SS)-7e0 pyrazolines.

F. Cermola, M. R. Iesce / Tetrahedron 62 (2006) 10694–10699
10697
Me
may be that, due to the absence of an acid–base equilibrium,
the fate of the anion 10e is different from that previously
observed in the presence of mild base NEt3. It undergoes
deacetylation of the sugar ring, the second elimination
being veryfast owing to the aromatization of the intermediate
dihydrofuran system (Scheme 8).
O
NaOO-t-Bu
DMDO
O
O
THF
acetone
AcO
4e
No reaction
28%
Me
- 20 °C
t-BuOOH/NEt3
9
35%
CH2Cl2, rt
O
O
It is noteworthy that deacetylation did not occur whether
starting from furan 1e or from 1,2,3,5-tetra-O-acetatyl-D-ri-
bofuranose. Thus the electron-poor aglycone in 4e should be
responsible for the two observed reactions in that it increases
the H-1 acidity, which represents the drawing force.
Me
Me
Me
Me
O
O
AcO
AcO
O
O
AcO
Z-8e
AcO OAc
E-8e
OAc
Although it was not possible to synthesize the planned epox-
ide, the above results illustrate the synthetic potential of the
photooxygenation procedure applied to the sugar-furans 1.
They also highlight the high reactivity of unsaturated ribo-
furanoside systems such as 4e, leading to a wide range of
differently functionalized compounds, which in turn can
be used as starting material for more complex glycosyl
derivatives.
DMDO = dimethyldioxirane
Scheme 7.
doublet at d 5.99 in agreement with the CH2CO group at the
same side of H-2, as occurs in the Z-configuration. Obvi-
ously, the same correlation was absent in the isomer E-8e.
This uncommon result shows that the acidity of the anomeric
proton in 4e is higher than that of the hydroperoxidic proton.
This explains how triethylamine promotes the anion 10e
shown in Scheme 8. Due to the high conjugation of the sys-
tem, subsequent re-protonation of the carbonyl oxygen
should afford the undetected enol E,Z-11e, which tautomer-
izes into Z-8e and E-8e, the latter being initially the major
isomer. The hypothesis was confirmed by control experi-
ments. Indeed, when only NEt3 was added to a solution of
4e, compounds Z-8e and E-8e were the only products in
the crude reaction mixture (1H NMR).
The synthesis of the new functionalized exo-glycals 8e is of
interest both for the potential biological activity14 and for
their synthetic application in producing pharmacologically
active derivatives.15
Moreover, the work demonstrated the use of glycosides with
a,b-unsaturated systems, such as compound 4e, in [3+2]
dipolar cycloadditions towards suitable 1,3-dipoles. This
provides an easy access to C-nucleosides of pharmacolo-
gical interest. Indeed, the reaction of 4e with diazomethane
leads regio- and stereoselectively to the novel pyrazoline
C-nucleoside 7e, which is structurally related to pyrazole
C-nucleosides, compounds widely used for their pharmaco-
logical properties.16
4e
NEt3
Et3NH
Me
Me
O
O
O
O
O
O
t-BuOO
AcO
3. Experimental
3.1. General
AcO
4e
Me
Me
AcO
NEt3
OAc
t-BuOOH
AcO
OAc
AcO
10e
Et3NH
Nuclear magnetic resonance (NMR) spectra were recorded
at 500 MHz for [1H] and 125 MHz for [13C] on a Fourier
Transform NMR Varian 500 Unity Inova spectrometer.
The carbon multiplicity was evidenced by DEPT experi-
ments. The proton couplings were evidenced by 1H–1H
COSY experiments. The heteronuclear chemical shift cor-
relations were determined by HMQC and HMBC pulse
sequences. 1H–1H proximities through space within a mole-
cule were determined by NOESY. Analytical TLC was
performed on precoated silica gel plates (Macherey-Nagel)
with 0.2 mm film thickness. Column chromatography was
performed on silica gel (Macherey-Nagel). Reagent-grade
commercially available reagents and solvents were used.
Me
Me
O
O
H
O
O
O
O
AcO
AcO
Me
Me
AcO
AcO
OAc
AcOH
E,Z-11e
9
E,Z-8e
Scheme 8.
3.1.1. Synthesis of 20,50-dimethyl-30-(2,3,5-tri-O-acetyl-D-
ribofuranosyl)furan b-1e. To a stirred solution of 2,3,5-
tri-O-acetyl-1-O-(acetyl)-D-ribofuranose (320 mg, 1 mmol)
in dry CH2Cl2 (10 mL) under argon and in the presence of
molecular sieves 2,5-dimethylfuran (96 mg, 1 mmol) and,
successively, a dichloromethane solution of SnCl4 (1 M,
The higher acidity of H-1 compared to t-BuOOH was
demonstrated by the use of NaOO-t-Bu, freshly prepared
by adding 1 equiv of NaH to the hydroperoxide at 0 ꢁC in
anhydrous THF.13 With this reagent, the epoxidation reac-
tion was still overcome by the acid–base reaction but surpris-
ingly the furan 9 was formed (Scheme 7). The explanation

10698
F. Cermola, M. R. Iesce / Tetrahedron 62 (2006) 10694–10699
1
16 mL) were added. The resulting mixture was stirred at
room temperature for 3 h. The reaction was then quenched
by saturated solution of NaHCO3 (5 mL). The organic layer
was separated and the aqueous layer was extracted with
CHCl3 (3ꢂ10 mL). The combined organic extracts were
washed with brine and dried over anhydrous MgSO4. After
filtration the solvent was removed under reduced pressure.
1H NMR spectrum showed that glycosyl furan 1e was pres-
ent as a mixture of a and b-anomers in ca. 1:8 molar ratio.17
Silica gel chromatography using n-hexane/EtOAc (85:15,
v/v) as eluent afforded glycosyl furan b-1c in 35% yield:
1H NMR (CDCl3) d 2.04 (s, 3H, CH3CO2), 2.10 (s, 6H,
2CH3CO2), 2.20 (s, 3H, CH3-20), 2.24 (s, 3H, CH3-50),
4.21 (m, 2H, H-4 and H-5A), 4.31 (dd, J¼13.4, 4.6 Hz,
1H, H-5B), 4.81 (d, J¼7.0 Hz, 1H, H-1), 5.08 (dd, J¼7.0,
5.7 Hz, 1H, H-2), 5.27 (dd, J¼5.7, 4.1 Hz, 1H, H-3), 5.86
stirring. After 1 h the reaction was complete and the H
NMR showed the presence of pyrazoline 7e as the only iden-
tifiable product (ca. 80%). After evaporation of the solvent,
TLC chromatography of the residue (n-hexane/Et2O, 1:4, v/v)
afforded pure 7e (50%; 40% based on starting furan 1e).
1
cis-(R,R)-7e: H NMR (CDCl3) d 2.06, 2.08, 2.14 (3s, 9H,
3CH3CO2), 2.25 and 2.41 (2s, 6H, 2CH3CO), 3.26 (dd,
J¼8.8, 3.8 Hz, 1H, H-50), 4.15 (m, 1H, H-4), 4.23 and 4.30
(2dd, J¼12.6, 4.4, 3.3 Hz, 2H, H2-5), 4.55 (dd, J¼18.1,
8.8 Hz, 1H, H-40A), 4.67 (d, J¼6.1 Hz, 1H, H-1), 4.76 (dd,
J¼18.1, 3.8 Hz, 1H, H-40B), 4.87 (t, J¼6.1 Hz, 1H, H-2),
5.05 (dd, J¼6.1, 5.5 Hz, 1H, H-3); 13C NMR (CDCl3)
d 20.5 (q, 2CH3CO2), 20.8 (q, CH3CO2), 31.1 (q, CH3CO),
32.4 (q, CH3CO), 46.2 (d, C-50), 62.5 (t, C-5), 70.5 (d,
C-2), 71.2 (d, C-3), 79.8 (d, C-4), 81.6 (t, C-40), 82.5 (d,
C-1), 110.9 (s, C-10), 169.1 (s, CO2), 169.3 (s, CO2), 173.4
(s, CO2), 203.9 (s, CO), 208.0 (s, CO).
13
(s, 1H, H-40); C NMR (CDCl3) d 15.9 (q, CH3-20), 16.8
(q, CH3-50), 20.6 (q, CH3CO2), 20.7 (q, CH3CO2), 20.8 (q,
CH3CO2), 63.8 (t, C-5), 71.7 (d, C-3), 74.8 (d, C-2), 75.2
(d, C-1), 79.6 (d, C-4), 104.2 (d, C-40), 116.5 (s, C-30),
148.4 (s, C-20), 150.5 (s, C-50), 169.6, 169.7 and 170.5 (3s,
3CO2).
No other isomer was detected either by careful NMR
analysis of the crude reaction mixture or by chromatography,
and only polymeric material was found in addition to
cis-(R,R)-7e.
3.1.2. One-pot synthesis of b-furanoside 4e by MB-sensi-
tized photooxygenation of b-1e and Et2S reduction. A
0.02 M solution of b-1e (0.25 mmol) in dry CH2Cl2 was
irradiated at ꢀ20 ꢁC with a halogen lamp (General Elec-
tric, 650 W) in the presence of methylene blue (MB,
1ꢂ10ꢀ3 mmol) while dry oxygen was bubbled through the
solution. The progress of the reaction was checked by peri-
3.1.4. Treatment of 4e with NEt3/t-BuOOH.12 Triethyl-
amine (6 equiv) and tert-butyl hydroperoxide (6 equiv)19
were added to a 0.1 M solution of crude ribofuranoside 4e
(1 mmol) in dry dichloromethane, and the resulting solution
was kept at room temperature under stirring. When the reac-
tion was complete (12 h, TLC), the mixture was partitioned
between water and dichloromethane and the aqueous layer
was extracted with further dichloromethane. The combined
organic layers were then dried over Na2SO4. After evapora-
tion of the solvents and unchanged reactants, the residue was
chromatographed on silica gel eluting with n-hexane/AcOEt
(2:3, v/v) and AcOEt, which afforded a mixture of E- and
Z-8e in ca. 9:1 molar ratio, respectively (35%; 23% based
1
odically monitoring the disappearance of 1e (TLC or H
NMR). When the reaction was complete (90 min), a pre-
cooled dichloromethane solution of Et2S (2 equiv) was
added to the photooxygenation mixture. The latter was
kept at ꢀ20 ꢁC for 1 h and then transferred at room temper-
ature. When the reduction was complete (2 h, 1H NMR), the
solvent and unreacted Et2S were removed under reduced
pressure. The residue was taken up in Et2O, the suspension
filtered to remove the insoluble sensitizer (MB) and the fil-
trate evaporated to give crude cis-b-4e (yield>90%). Silica
gel chromatography on a short column gave pure compound
4e in 70% yield.9,18
1
on the starting furan). The H NMR recorded on the same
mixture kept at room temperature for 24 h showed that the
molar ratio was changed to 3(E):2(Z).
E-8e: 1H NMR (CDCl3) d 2.05, 2.08, 2.13, 2.16 and 2.20 (5s,
15H, 5CH3CO), 3.48 and 3.55 (2d, J¼17.5 Hz, CH2CO),
4.09 (m, 1H, H-5A), 4.54 (m, 2H, H-4 and H-5B), 5.37 (dd,
J¼8.2, 6.0 Hz, 1H, H-3), 6.31 (d, J¼6.0 Hz, 1H, H-2); 13C
NMR (CDCl3) d 20.4 (q, CH3CO2), 20.5 (q, CH3CO2),
20.8 (q, CH3CO2), 28.7 (q, CH3CO), 29.5 (q, CH3CO),
41.5 (t, CH2CO), 61.8 (t, C-5), 69.3 (d, C-2), 69.4 (d, C-3),
79.4 (d, C-4), 111.9 (s, C]), 162.3 (s, O–C]), 169.1 (s,
CO2), 169.3 (s, CO2), 170.4 (s, CO2), 196.7 (s, CO), 205.0
(s, CO).
1
Compound 4e: H NMR (CDCl3) d 2.07 (s, 3H, CH3CO2),
2.08 (s, 3H, CH3CO2), 2.13 (s, 3H, CH3CO2), 2.26 (s, 3H,
CH3CO), 2.29 (s, 3H, CH3CO), 4.14 (dd, J¼12.1, 3.7 Hz,
1H, H-5A), 4.20 (m, 1H, H-4), 4.35 (dd, J¼12.1, 3.0 Hz,
1H, H-5B), 4.60 (dd, J¼6.0, 1.2 Hz, 1H, H-1), 5.22 (m,
2H, H-2 and H-3), 6.31 (d, J¼1.2 Hz, 1H, H-30); 13C NMR
(CDCl3) d 20.4 (q, CH3CO2), 20.5 (q, CH3CO2), 20.8 (q,
CH3CO2), 30.5 (2q, 2CH3CO), 62.8 (t, C-5), 71.4 (d, C-3),
73.7 (d, C-2), 80.5 (d, C-4), 80.9 (d, C-1), 124.4 (d, C-30),
153.9 (s, C-20), 169.3 (s, CO2), 169.6 (s, CO2), 170.6 (s,
CO2), 196.5 (s, C-40), 200.4 (s, C-10).
Z-8e (in 2:3 mixture with E-8e): 1H NMR (CDCl3) d 2.41 (s,
3H, CH3CO), 3.22 and 3.33 (2d, J¼17.5 Hz, CH2CO), 4.25
(dd, J¼12.4, 4.4 Hz, 1H, H-5A), 4.54 (partially overlapped
to the signal of E-8e, H-5B), 4.75 (m, 1H, H-4), 5.26 (dd,
J¼8.1, 5.9 Hz, 1H, H-3), 5.99 (d, J¼6.0 Hz, 1H, H-2); 13C
NMR (CDCl3) d 29.6 (q, CH3CO), 31.9 (q, CH3CO), 41.4
(t, CH2CO), 61.1 (t, C-5), 69.0 (d, C-2), 70.3 (d, C-3), 81.7
(d, C-4), 111.7 (s, C]), 161.0 (s, O–C]), 169.2 (s, CO2),
169.4 (s, CO2), 170.3 (s, CO2), 197.2 (s, CO), 205.7
(s, CO). The signals not reported are overlapped to those
of E-8e.
In the subsequent reactions, crude cis-b-4e was used without
further purification since control experiments showed that
the presence of non-volatile Et2SO was irrelevant.
3.1.3. Treatment of 4e with diazomethane. To the crude ri-
bofuranoside 4e (0.25 mmol) in dry Et2O was added a diethyl
ether solution of freshly prepared CH2N2 (ca. 1 mmol) and
the resulting mixture was kept at room temperature under

F. Cermola, M. R. Iesce / Tetrahedron 62 (2006) 10694–10699
10699
A. A., Ed.; CRC: Boca Raton, 1985; Vol. II, pp 205–272;
(c) Bloodworth, A. J.; Eggelte, H. J. Singlet Oxygen; Frimer,
A. A., Ed.; CRC: Boca Raton, 1985; Vol. II, pp 93–203.
5. Iesce, M. R. Synthetic Organic Photochemistry; Griesbeck,
A. G., Mattay, J., Eds.; Marcel Dekker: New York, NY, 2005;
Vol. 12, pp 299–363.
3.1.5. Treatment of 4e with NEt3. NEt3 (1 equiv) in CDCl3
(0.5 mL) was added to a solution of crude ribofuranoside 4e
in the same solvent (0.5 mL), and the resulting mixture was
monitored by 1H NMR. After 10 min, the spectrum showed
the presence of the E- and Z-8e mixture in ca. 9(E):1(Z)
molar ratio.
6. See for example: (a) Morgan, K. M. Ann. Rep. Prog. Chem.,
Sect B: Org. Chem. 2005, 101, 284; (b) Nicolaou, K. C.;
Snyder, S. A.; Montagnon, T.; Vassilikogiannakis, G. Angew.
Chem., Int. Ed. 2002, 41, 1668; (c) Wiest, O.; Houk, K. N.
Top. Curr. Chem. 1996, 183, 1.
7. Scarpati, R.; Iesce, M. R.; Cermola, F.; Guitto, A. Synlett
1998, 17; Gollnick, K.; Griesbeck, A. Tetrahedron 1985, 41,
2057.
8. Cermola, F.; Iesce, M. R.; Montella, S. Lett. Org. Chem. 2004,
1, 271.
9. Cermola, F.; Iesce, M. R.; Buonerba, G. J. Org. Chem. 2005, 70,
6503.
10. Regitz, M.; Heydt, H. 1,3-Dipolar Cycloaddition Chemistry;
Padwa, A., Ed.; Wiley: New York, NY, 1984; pp 393–558.
11. Singh, M.; Murray, R. W. J. Org. Chem. 1992, 57, 4263.
12. Graziano, M. L.; Iesce, M. R.; Scarpati, R. Synthesis 1984, 66.
13. Fernandez de la Pradilla, R.; Castro, S.; Manzano, P.; Martin-
Ortega, M.; Priego, J.; Viso, A.; Rodriguez, A.; Fonseca, I.
J. Org. Chem. 1998, 63, 4954.
14. For some recent examples, see: Lin, C.-H.; Lin, H.-C.; Yang,
W.-B. Curr. Top. Med. Chem. 2005, 5, 1431; Stolz, F.;
Reiner, M.; Blume, A.; Reutter, W.; Schmidt, R. R. J. Org.
Chem. 2004, 69, 665; Tatibouet, A.; Rollin, P.; Martin, O. R.
J. Carbohydr. Chem. 2000, 19, 641.
3.1.6. Treatment of 4e with NaOO-t-Bu.13 A suspension of
oil-free NaH (washed with hexane and dried) (50 mg,
2 mmol) in dry THF (20 mL) under the atmosphere of argon
was cooled to 0 ꢁC and 1.2 mL (6 equiv) of t-BuOOH (solu-
tion 5.0 M in decane) was added. The resulting mixture was
warmed under stirring to 25 ꢁC for 30 min, then cooled
to 0 ꢁC and a solution of the crude ribofuranoside 4e,
previously dried on P2O5 (95 mg, 0.25 mmol) in dry THF
(10 mL), was added dropwise. The reaction mixture was
stirred at 0 ꢁC until the disappearance of compound 4e
(16 h, by TLC). Then, the solvents were removed under re-
duced pressure and the crude mixture was chromatographed
on silica gel, using n-hexane/EtOAc (4:1) as eluent, to give
1
furan 9 (28%; 22% based on the starting furan): H NMR
(CDCl3) d 2.11 (3H, s, CH3CO2), 2.32 and 2.47 (6H, 2s,
2COCH3), 5.07 (2H, s, CH2), 6.48 (2H, br s, H-3 and
H-4), 6.61 (1H, s, CH); 13C NMR (CDCl3) d 20.5 (q,
CH3CO2), 30.1 (q, CH3CO), 30.9 (q, CH3CO), 67.2 (t,
CH2), 114.2 (d, C-3), 116.2 (d, C-4), 118.9 (d, CH), 148.5
and 148.5 (2s, C-5 and C]), 152.1 (s, C-2), 169.3 (s,
CO2), 197.4 (s, CO), 202.6 (s, CO).
15. Taillefumier, C.; Chapleur, Y. Chem. Rev. 2004, 104, 263;
Yang, W.-B.; Yang, Y.-Y.; Gu, Y.-F.; Wang, S.-H.; Chang,
C.-C.; Lin, C.-H. J. Org. Chem. 2002, 67, 3773; Lay, L.;
Nicotra, F.; Panza, L.; Russo, G.; Sello, G. J. Carbohydr.
Chem. 1998, 17, 1269; Link, J. T.; Gallant, M.; Danishefsky,
S. J.; Huber, S. J. Am. Chem. Soc. 1993, 115, 3782.
16. See for example Elgemeie, G. H.; Zaghary, W. A.; Amin,
K. M.; Nasr, T. M. Nucleosides Nucleotides Nucleic acids
2005, 24, 1227; Nishiyama, Y.; Nishimura, N.; Kuroyanagi,
N.; Maeba, I. Carbohydr. Res. 1997, 300, 183.
Acknowledgements
Financial support from MIUR (FIRB 2003–2005 and PRIN
2006–2008) is gratefully acknowledged. NMR experiments
were run at the Centro di Metodologie Chimico-Fisiche,
ꢀ
Universita di Napoli Federico II.
References and notes
17. The b-anomer 1e was the main product (b:a¼8:1) owing to the
neighbouring group of the acetyl at C-2 of the glycosyl donor.
18. Diacetylethylenes may undergo polymerization, mainly on
contact with chromatographic adsorbents (Graziano, M. L.;
Iesce, M. R.; Scarpati, R. Synthesis 1983, 125).
19. High equivalents of the reactants were used to balance
those oxidating diethyl sulfoxide present in the crude reaction
mixture.
1. Simons, C. Nucleoside Mimetics Their Chemistry and
Biological Properties; Gordon and Breach Science: Australia,
Canada, 2001; (b) Peasley, K. Med. Hypotheses 2000, 55, 408.
2. See for example: Wu, Q.; Simons, C. Synthesis 2004, 1533.
3. Postema, H. D. M. C-Glycoside Synthesis; CRC: London,
1995; (b) Du, Y.; Linhardt, R. J. Tetrahedron 1998, 54, 9913.
4. (a) Iesce, M. R.; Cermola, F.; Temussi, F. Curr. Chem. Org.
2005, 9, 109; (b) Matsumoto, M. Singlet Oxygen; Frimer,