1497
Glucosides of Forsythia
fruits
acetate exhibited
9 H ) and 7.26
substituted phenol moiety, and a signal at
type signals at 67.02
d,
This means that the aromatic and the aliphatic
moieties lie in a single metabolic route.
=
Z
d,
=
9 Hz), corresponding to a
s)
It is easy to predict that hydrolysis of 2 yields a
for an isolated methylene group next to a phenyl and a
carbonyl group. Further, the hydrogen signals of anoth-
alcohol which will cyclize spontaneously to rengyolone
(6) since the
acetate, hallerone, yields 6 as the sole
Compound 6 could then be
er methylene group at 3.65
= 7 Hz), a carbinyl
product on hydrolysis
methine group at 64.66
m,
Hz) and an
d,
transformed into rengyoxide (5) and rengyol(3) by a series
of biological processes. However, a more plausible bio-
genetic route to 5 and 3 is that the
axial anomeric methine group at 64.45
7 Hz)
were similar to the corresponding peaks of rengyoside A
pentaacetate (63.69, 4.70 and 4.49, respectively). Thus, it
was assumed that rengyoside C was an ester of rengyoside
A and a carboxylic acid having a p-substituted phenol
structure. Alkaline hydrolysis of rengyoside C
system of 2 is
reduced successively to form 8 and 7 which are then
hydrolysed to form 5 and 3 respectively. The character-
ization of rengyosides B (8) and A (7) support the latter
course, although the possibility that the glucosides are
derived from 5 and 3 cannot be ruled out. The character-
ization of p-hydroxyphenylacetic acid (1 1 ) as the ester,
rengyoside C (9) is also noteworthy because it links
ate with aqueous potassium carbonate afforded
side A (7) and p-hydroxyphenylacetic acid
the prod-
ucts being identified by TLC behaviour and
spectra. Further it was considered that the ester group in
rengyoside C was located at C-6 of the glucose moiety,
since only the chemical shifts of H-6 hydrogens of the
acetate of 9 shifted slightly but distinctly to lower-field by
0.02 ppm from those of the acetate of 7.
hydroxyphenylethanol, found in salidroside
presumed starting compound, phenylpyruvic acid (12).
to the
EXPERIMENTAL
In order to confirm the location of the ester group,
esterification at C-6 of the glucose moiety of 7 to form 9
was attempted. There are several methods available for
the introduction of an acyl group exclusively to the C-6
Extraction
Forsythia
for 56 hr at room temp. Removal of the solvent from the
combined gave the extract
(50 g). The extract was dissolved in
tioned with (100 ml). The
isolation. The crude drug (the air-dried fruits of
(500 g) was extracted x 3 with (2 1)
under red.
position of a glucose moiety, e.g. through a bromide
or
(300 ml) and parti-
by the stannylation reaction
The
phase was re-extracted with
phenylacetylglucose derivative has been obtained in good
yield (72%) on reaction of phenyl
ml) and removal of the solvent from the organic
phase under reduced pressure provided the extract
(9.4 g) which was subjected repeatedly to chromatography on
and p-hydroxyphenylacetic acid (1 1 ) with
zole
However, various experiments to obtain 9 by
silica gel with
with
systems, on Sephadex LH-20
to furnish
direct condensation of rengyoside A (7) and 11 did not
work and gave complete recovery of the starting mater-
ials. Enzymatic transglucosylation with phenyl glucoside
and rengyol (3) also failed if the substrate carried a
hydroxyphenylacetyl group at C-6 of the glucose
The successful instance where the ester 9 was obtained
from 7 and 1 1 was by the method which used
and on Toyopearl HW-40 with
fractions containing rengyosides A (7) (11 mg), B (8) (1.2 mg) and
C (9) (18 mg), along with salidroside (10) (7 mg). Final purifi-
cation of rengyoside B (8) was accomplished by HPLC (Toyo
Soda
7.5 mm i.d. x 30 cm;
0.5
RI detec-
tion). Rengyoside C (9) was purified by acetylation with
pyridine
followed by chromatography on silica gel
97: 3) to give rengyoside C pentaacetate.
dazole in DMF at
though the yield was low (12%).
These results indicated the existence of a severe steric
interaction between rengyol and the phenylacetyl groups
in rengyoside C (9). This may be the reason why 9 was
very easily hydrolysed even during the isolation and
purification procedures. It is also noteworthy that such a
steric interaction is not appreciable if the rengyol moiety
is exchanged with a phenoxyl group. Thus length rather
than the apparent bulkiness seems to be more important.
Rengyoside A (7). Amorphous solid;
11.0”
c
0.18);
NMR
1.2-2.0
H-3’. H-S and
1.67
Hz, H-7’), 3.4-4.1
H-2, H-3, H-4, H-5
H-8’), 4.24
d,
Hz, H-l); FDMS m/z: 344
A. A soln of 7 (2 mg) and
Acetylation
in
pyridine was allowed to stand at room temp. for 24 hr. The crude
product was chromatographed on silica gel to give the
The synthetic rengyoside C (9) exhibited in its
spectrum signals at 64.02 and 4.25 (each
NMR
11 Hz),
d,
=
acetate as a
Hz, H-7’),
oil;
61.68
s,
J
compatible to the esterified C-6 methylene group of the
glucose moiety. Further, this product was acetylated with
acetic anhydride and pyridine to afford the correspond-
ing pentaacetate identical to the acetylation product of
natural rengyoside C (TLC and spectral characteristics).
Salidroside (10) was identified by comparing its TLC
(each
3.6-3.8
m, H-5), 3.69
2 Hz, H-6), 4.23
Hz, H-8’), 4.16
dd, J= 12, 5 Hz, H-6), 4.49
dd,
d,
Hz,
H-l), 4.70
200
m,
169
Hz, H-4’); EIMS m/z
int.): 331
141
107
43 (100).
Enzymatic hydrolysis of rengyoside A. Crude hesperidinase
(1 mg) was added to a of 7 (2 mg) in citrate-phosphate buffer
4.1, 1 ml) and the mixture was incubated at 40” overnight
behaviour and
authentic sample prepared from acetobromoglucose and
p-hydroxyphenylethanol The co-existence of
sythosides cornoside (2) and rengyol (3) in the same
plant material indicates that the parts of these
spectrum with those of an
followed by
in
The residue was then subjected to
chromatography on silica gel
97: 3) affording
rengyol (3) as an amorphous solid; ‘HNMR
61.66
m,
are related biosynthetically. This view became
more certain with the characterization of salidroside (10)
which stood as a direct precursor of 2. The
system can be easily constructed by simple oxygenation
of the phenol group, and the successive reduction
products of 2 corresponding to rengyosides 8 and 7
Hz, H-7), 3.71
4); EIMS m/z (rel. int.): 142
98 (100); TLC:
Hz, H-8), 3.51
115
(silica
3: 1).
Glucose; TLC:
1).
0.32 (silica
Seya, Kazuhiko
