Communication
Organic & Biomolecular Chemistry
Table 1 Glycosylations of 1 and 6 under several conditions
Entry
DDQ (equiv.)
Boronic acid (4 or 5)
hva
% Yield (α : β)b
1
2
3
4
5
6
1.5
—
—
1.5
1.5
1.5
—
—
4
—
4
–
+
–
+
+
+
NR
NR
NR
38 (81 : 19)
74 (76 : 24)
57 (73 : 27)
Fig. 2 Boronic ester formation of 1 with 4. (a) 1H-NMR spectrum of 1 (10 mM)
in MeCN-d3. (b) 1H-NMR spectrum of complex 1’ resulting from 1 (10 mM) and
4 (10 mM) in MeCN-d3.
5
and a C4 secondary hydroxy group, would effectively reduce
the reactivity of the C3 hydroxy group in the 2-deoxyglycosyl
donor. In addition, the electron-donating property of the OMe
group of 4-methoxyphenylboronic acid (4) would reduce the
electron-withdrawing effect of the boronic ester compared to
phenylboronic acid (5), thus avoiding deactivation of the glyco-
syl donor by the temporary protecting group.
a Photo-irradiation was carried out using a UV lamp (365 nm, 100 W)
placed at 5 cm from the reaction mixture. b The α/β ratio was
determined using 1H-NMR analysis.
indicated the utility of temporary protection of the 4,6-diol of
1 by 4. On the other hand, when phenylboronic acid (5) was
used instead of 4, the yield of 7 decreased due to the low acti-
vation of 1 resulting from the stronger electron-withdrawing
nature of 5 compared to 4 (entry 6 in Table 1). Indeed,
1H-NMR analysis of 1′ and 1″ (the complex of 1 and 5) showed
that the low chemical shifts at the C4 and C6 hydrogens in 1′
are significantly smaller than those in 1″ (see Fig. S3 in the
ESI†). Furthermore, when 7α was exposed to these reaction
conditions, 7α was recovered in quantitative yield. This result
indicated that no epimerization of 7 occurred under these
reaction conditions, and glycosylation proceeded under kinetic
control, providing good α-stereoselectivity due to the kinetic
anomeric effect.16 In addition, the formation of disulfide 9
during the reaction was confirmed by quantitative isolation of
9 after the reaction.
We next examined the glycosylation of 2 and 3 with 6 under
similar conditions. These results are summarized in Table 2. A
similar tendency was observed with 2: the glycosylation of 2
and 6 using DDQ in the absence of 4 under photo-irradiation
conditions afforded 10 in moderate (57%) yield (entry 1 in
Table 2). In contrast, the glycosylation of 2 and 6 using DDQ in
the presence of 4 under photo-irradiation conditions afforded
10 in good (72%) yield (entry 2 in Table 2). On the other hand,
when 3 was used as a glycosyl donor, the opposite tendency
was observed: the glycosylation of 3 and 6 using DDQ in the
absence of 4 under photo-irradiation conditions afforded 11 in
high yield (82%, entry 3 in Table 2), while the presence of
4 gave 11 in low yield (32%, entry 4 in Table 2). These results
indicated that 3 selectively reacts with 6 even without 4 due to
We thus first investigated the boronic ester formation of
1
1–3 with 4 by H-NMR spectroscopy. These results are shown
in Fig. 2 (only the case for 1 is shown). For 1 or 2 (see Fig. S1
in the ESI†) with an equal amount of 4, 1 : 1 complex for-
mation of the glycosyl donor and the boronic acid was
observed. In addition, the chemical shift profiles in the
1H-NMR spectra (glycosyl donor 1 or 2 vs. complex 1′ or 2′,
respectively) confirmed that 4 bound to the 4,6-diol of 1 or 2,
as expected. On the other hand, no such boronic ester for-
mation was observed between 3 and 4 (see Fig. S2 in the ESI†).
We next examined the glycosylation of 1 and cyclohexyl-
methanol (6) using DDQ under photo-irradiation conditions
(365 nm, 100 W). These results are summarized in Table 1.
Neither photo-irradiation in the absence of DDQ, nor DDQ in
the absence of photo-irradiation, resulted in glycosylation, and
the glycosyl donor 1 was recovered in quantitative yield
(entries 1 and 2 in Table 1). In addition, the glycosyl donor
was not activated by 4-methoxyphenylboronic acid (4) alone
(entry 3 in Table 1). In contrast, the glycosylation of 1 and 6
using DDQ under photo-irradiation for 3 h provided the glyco-
side 7 in 38% yield (entry 4 in Table 1). These results clearly
indicated that the combination of DDQ and photo-irradiation
together triggers glycosylation. However, although a self-coup-
ling product(s) of 1 was not detected, 1,6-anhydro-2-deoxy-
galactoside 815 was produced as a by-product in 61% yield. In
sharp contrast, the glycosylation of 1 and 6 using DDQ in the
presence of 4 under photo-irradiation for 3 h smoothly pro-
ceeded to give the glycoside 7 in 74% yield, with no detectable
amount of 8 produced (entry 5 in Table 1). This result clearly
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2013
Nakanishi, Mayuka
