Diastereomers of lithospermic acid and lithospermic acid B from Monarda fistulosa and Lithospermum erythrorhizon

Article abstract of DOI:10.1016/j.fitote.2013.08.009

Monardic acids A (1) and B (2), which are (7R,8R) diastereomers of lithospermic acid (LA) and lithospermic acid B, respectively, were isolated from Monarda fistulosa. A (7S,8R) isomer (3) of LA was also isolated from this plant, and a (7R,8S) isomer (7) of LA was obtained from Lithospermum erythrorhizon. The absolute configuration of 1 was confirmed by analysis of its hydrolysates, 7-epiblechnic acid and 2R-3-(3,4-dihydroxyphenyl)-2- hydroxypropanoic acid. The configuration in the dihydrobenzofuran moieties of 2, 3, and 7 was extrapolated by using the phenylglycine methyl ester method and a Cotton effect at approximately 250-260 nm in their electronic circular dichroism spectra. Diastereomers (1-3 and 7) displayed moderate hyaluronidase inhibitory and histamine release inhibitory activities.

Full text of DOI:10.1016/j.fitote.2013.08.009

Fitoterapia 91 (2013) 51–59
Contents lists available at ScienceDirect
Fitoterapia
journal homepage: www.elsevier.com/locate/fitote
Diastereomers of lithospermic acid and lithospermic acid B
from Monarda fistulosa and Lithospermum erythrorhizon
a
b
b
b
Toshihiro Murata a, , Kanae Oyama , Minami Fujiyama , Bunmei Oobayashi , Kaoru Umehara ,
⁎
Toshio Miyase b, Fumihiko Yoshizaki a
Tohoku Pharmaceutical University, 4-4-1 Komatsushima, Aoba-ku, Sendai 981–8558, Japan
a
b
School of Pharmaceutical Sciences, University of Shizuoka, 52–1, Yada, Suruga-ku, Shizuoka 422–8526, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 July 2013
Accepted in revised form 14 August 2013
Available online 23 August 2013
Monardic acids A (1) and B (2), which are (7R,8R) diastereomers of lithospermic acid (LA) and
lithospermic acid B, respectively, were isolated from Monarda fistulosa. A (7S,8R) isomer (3) of
LA was also isolated from this plant, and a (7R,8S) isomer (7) of LA was obtained from
Lithospermum erythrorhizon. The absolute configuration of 1 was confirmed by analysis of its
hydrolysates, 7-epiblechnic acid and 2R-3-(3,4-dihydroxyphenyl)-2-hydroxypropanoic acid.
The configuration in the dihydrobenzofuran moieties of 2, 3, and 7 was extrapolated by using
the phenylglycine methyl ester method and a Cotton effect at approximately 250–260 nm in
their electronic circular dichroism spectra. Diastereomers (1–3 and 7) displayed moderate
hyaluronidase inhibitory and histamine release inhibitory activities.
Chemical compounds:
Lithospermic acid (PubChem CID: 6441498)
Lithospermic acid B (PubChem CID: 6451084)
Rosmarinic acid (PubChem CID: 5281792)
© 2013 Elsevier B.V. All rights reserved.
Keywords:
Monarda fistulosa
Lithospermum erythrorhizon
Phenylpropanoid oligomers
Hyaluronidase inhibitors
Histamine release inhibitors
1. Introduction
[5–8]. Furthermore, the entire synthesis route of LA has been
previously demonstrated [9–12].
Lithospermic acid (LA) [1,2] and lithospermic acid B (LAB)
[3,4] are major phenylpropanoid oligomers present in various
Lamiaceae and Boraginaceae plants. LA and LAB are the
bioactive components in medicinal herbs, including sage
(Salvia officinalis L.), danshen (S. miltiorrhiza Bunge), and
shikon (Lithospermum erythrorhizon Siebold et Zuccarini).
Previously, LA and LAB have been shown to possess
antioxidative properties and biological activities, including
HIV-1 integrase inhibition, hyaluronidase inhibition, aldose
reductase inhibition, and improvement of uremic symptoms,
resulting in significant decreases in blood urea nitrogen,
creatinine, methylguanidine, and guanidinosuccinic acid levels
The dihydrobenzofuran moiety, with the (7S,8S) config-
uration, is a common structural feature in both LA and LAB
[1,2,4]. To date, the (7R,8R) isomers of LA and LAB have not
been reported as natural products. Although salvianolic acid
B was initially assigned a (7R,8R) configuration [3], Watzke
et al. reassigned its absolute configuration as (7S,8S) and
demonstrated that it was identical to LAB [13].
Monarda fistulosa L. (Lamiaceae) is used in traditional
medicines, and Native Americans have been known to use it
for treating vomiting, headaches, and rheumatic pains
[14,15]. Herein, we report the isolation of monardic acids A,
B, and C (1–3) together with rosmarinic acid (4) from the
acetone–H2O (8:2) extract of whole M. fistulosa plants. In
order to compare the steric structures of the isolates (1–3),
we isolated LA (5), LAB (6), and compound 7 from the
acetone–H2O (8:2) extract of the roots of L. erythrorhizon
⁎
Corresponding author. Tel./fax: +81 22 727 0220.
E-mail address: murata-t@tohoku-pharm.ac.jp (T. Murata).
0367-326X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.fitote.2013.08.009

52
T. Murata et al. / Fitoterapia 91 (2013) 51–59
(Boraginaceae). Compounds 1 and 2 have a trans-oriented
dihydrobenzofuran moiety similar to that in LA and LAB; 3
and 7 are cis isomers of LA (Fig. 1).
were obtained on a JEOL JMS700 mass spectrometer by using
m-nitrobenzyl alcohol or glycerol matrix. Preparative HPLC
was performed on a JASCO 2089 instrument (detector: UV
210 nm).
Alkaline hydrolysis of 1 gave 8 and 9, and the absolute
configuration of 1 was determined by identification of the
hydrolysates. On the other hand, it is known that the
phenylglycine methyl ester (PGME) method and electronic
circular dichroism (ECD) are available for determining the
absolute configuration of chiral compounds. It was confirmed
that the PGME method was applicable for determinating
the configuration of C-8 in the dihydrobenzofuran based
on examples applied to 1 (8R) and 5 (8S). The absolute
configurations at C-7 and C-8 in the dihydrobenzofuran
moieties of 2, 3, and 7 were extrapolated using the PGME
method and ECD spectra.
Isolated diastereomers (1–3 and 7) were expected to show
biological activities similar to those of LA and LAB. The
hyaluronidase inhibitory activity and histamine release inhib-
itory activity of these diastereomers were evaluated in order to
identify potential anti-allergic and anti-inflammatory agents
[6].
2.2. Plant materials
Monarda fistulosa L. was harvested from the medicinal
plant garden of Tohoku Pharmaceutical University, Sendai,
Japan. The young plants were purchased from Tamagawa engei,
Nagano, Japan. The plant was identified according to advice
from Dr. Goro Kokata, the Plant Identification Unit of the
National Museum of Nature and Science, Tsukuba Botanical
Garden, Tsukuba, Japan [14–16]. A voucher specimen was
deposited at the herbarium of Tohoku Pharmaceutical Univer-
sity, No. 20081101. Dried roots of Lithospermum erythrorhizon
Siebold et Zuccarini were purchased from Tochimoto Tenkaido
Co., Ltd., Osaka, Japan, as a medicinal Shikon, which was
identified according to microscopic observation of the roots by
the company [17]. A voucher specimen was deposited at the
herbarium of Tohoku Pharmaceutical University, No. 20120803.
2. Experimental
2.3. Extraction and isolation
2.1. General experimental procedures
Whole plants (980 g) of M. fistulosa were extracted with
acetone–H2O (8:2) (2 × 10 L) at 60 °C. The extract (124.5 g)
was concentrated at reduced pressure, suspended in H2O (1.5 L),
and extracted with Et2O (3 × 1 L). The aqueous extract (78.4 g)
was a red–brown syrup. It was dissolved in H2O, and the aqueous
solution was passed through a porous polymer gel column
(Mitsubishi Diaion HP-20, 70 × 180 mm) and eluted with H2O,
10, 80% MeOH, and MeOH. The 80% MeOH eluate (22.4 g) was
chromatographed on a reversed-phase column using ODS
(Cosmosil 140C18-OPN, 150 g; Nacalai Tesque) and was eluted
with 20, 40, 60, and 80% MeOH (fractions 1A–1D). Fraction 1A
(16.4 g) was subjected to preparative column chromatography
[Yamazen, Ultra Pack ODS-SM-50C-M, 37 × 100 mm; solvent,
MeOH–H2O containing 0.2% TFA (20:80)–(40:60); detector, UV
210 nm] to give 16 fractions (Frs. 2A–2P). Fractions 2 F, 2G, and
2H (755 mg) were subjected to preparative HPLC [columns,
Cosmosil, AR-II, 20 × 250 mm; solvent, CH3CN–H2O containing
0.2% TFA; Cosmosil, 5PE-MS, 20 × 250 mm; solvent, CH3CN–
H2O containing 0.2% TFA (25:75)] to yield compounds 1
(261.2 mg) and 4 (23.1 mg). Fractions 2I and 2 J (767 mg)
were subjected to preparative HPLC [columns, Cosmosil,
AR-II, 20 × 250 mm; solvent, CH3CN–H2O containing 0.2% TFA
(20:80); Cosmosil, 5PE-MS, 20 × 250 mm; solvent, CH3CN–H2O
containing 0.2% TFA (25:75); and Kanto Chemical, Mightysil
RP-18 GP, 10 × 250 mm; solvent, CH3CN–H2O containing 0.2%
TFA (25:75)] to yield compounds 2 (5.3 mg) and 3 (11.2 mg).
The roots (480 g) of L. erythrorhizon were extracted
with acetone–H2O (8:2) (4 × 3 L) at room temperature for
1 week. The extract (236.0 g) was concentrated at reduced
pressure, suspended in H2O (1.5 L), and extracted with Et2O
(3 × 1 L). The aqueous extract (206.6 g) was a black syrup. It
was dissolved in H2O, and the aqueous solution was passed
through a porous polymer gel column (Mitsubishi Diaion
HP-20, 70 × 180 mm) and eluted with H2O, 10, 20, 40, 60,
80% MeOH, and MeOH. The 80% MeOH eluate (1.38 g) was
subjected to preparative column chromatography [Yamazen,
Ultra Pack ODS-SM-50C-M, 37 × 100 mm; solvent, MeOH–
Optical rotations were measured on a JASCO P-2300
polarimeter. ECD spectra were recorded on a JASCO J-700
spectropolarimeter. A JEOL JNM-AL400 spectrometer was
used to record 1H NMR (400 MHz) and 13C NMR (100 MHz)
spectra, and chemical shifts are given as δ values with TMS as
an internal standard at 30 °C. Inverse-detected heteronuclear
correlations were measured using HMQC (optimized for
1JC–H = 145 Hz) and HMBC (optimized for JC–H = 8 Hz)
pulse sequences with a pulsed field gradient. HRFABMS data
n
3'''
3
5
9'''
5'''
7'''
1'''
8'''
1
9
7
8
3'
1'
7'
5'
9'
8''
3''
9''
1''
6'' 5''
Fig. 1. Structures of 1–3 and 7.

T. Murata et al. / Fitoterapia 91 (2013) 51–59
53
H2O containing 0.2% TFA (20:80)–(30:70); detector, UV
210 nm] to give nine fractions (Frs. 3A–3I). Fractions 3D, 3E,
and 3 F (375 mg), which were subjected to preparative HPLC
[columns, Cosmosil, AR-II, 20 × 250 mm; solvent, CH3CN–H2O
containing 0.2% TFA (25:75); Cosmosil, 5PE-MS, 20 × 250 mm;
solvent, CH3CN–H2O containing 0.2% TFA (25:75); and Kanto
Chemical, Mightysil RP-18 GP, 10 × 250 mm; solvent, CH3CN–
H2O containing 0.2% TFA (25:75)] to yield compounds 5
(234.6 mg), 6 (27.6 mg), and 7 (5.3 mg).
hydroxybenzotriazole (HOBT), and N-methylmorpholine were
added as reported previously [18]. The mixtures were then
stirred for 10 h at room temperature to give (S)-PGME amide;
tR = 33.9 min in the HPLC analysis [column, Mightysil
RP18-GP (6.0 × 250 mm); solvent, CH3CN–H2O containing
0.2% TFA (25:75); flow rate, 0.8 mL/min; detector, UV
210 nm]. The retention time of (S)-PGME amide of (2R)-3-
(3,4-dihydroxyphenyl)-2-hydroxypropanoic acid was 33.9 min
and that of (R)-PGME amide of (2R)-3-(3,4-dihydroxyphenyl)-
2-hydroxypropanoic acid was 34.7 min, which corresponded
with authentic samples.
2.3.1. Monardic acid A (1)
Colorless amorphous solid; [α]25 −86.2 (c 2.49, MeOH);
D
UV (MeOH) λmax (log ε) 256 (4.28), 291 (4.24), 311 (4.26);
ECD (c 0.05, MeOH) nm ([θ]) 230 (−52,700), 252 (−75,100),
288 (5700), 333 (−5400) nm; 1H NMR (CD3OD, 400 MHz)
and 13C NMR: (CD3OD, 100 MHz) see Table 1; HRFABMS
2.4.1. 7-Epiblechnic acid (8)
Colorless amorphous solid; [α]24 −144.0(c 0.18, MeOH);
D
ECD (c 0.02, MeOH) nm ([θ]) 230 (−32,000), 250 (−44,400),
283 (2100), 326 (−9900) nm; 1H NMR (CD3OD, 400 MHz) δ
7.77 (1H, d, J = 16.0 Hz), 7.16 (1H, d, J = 9.0 Hz), 6.79 (1H, d,
J = 2.0 Hz), 6.79 (1H, d, J = 9.0 Hz), 6.75 (1H, d, J = 8.0 Hz),
6.72 (1H, dd, J = 8.0, 2.0 Hz), 6.26 (1H, d, J = 16.0 Hz), 5.87
(1H, d, J = 5.0 Hz), 4.30 (1H, d, J = 5.0 Hz); FABMS (positive)
m/z 359 [M + H]+.
(positive) m/z 539.1196 [M + H]+ (Calcd for C27H23O12
:
539.1189).
2.3.2. Monardic acid B (2)
Colorless amorphous solid; [α]24 −43.5 (c 0.46, MeOH);
D
UV (MeOH) λmax (log ε) 254 (4.25), 288 (4.26), 307 (4.19);
ECD (c 0.05, MeOH) nm ([θ]) 234 (−49,100), 252 (−67,000),
289 (20,100), 338 (−3500) nm; 1H NMR (CD3OD, 400 MHz)
and 13C NMR (CD3OD, 100 MHz) see Table 1; HRFABMS
(positive) m/z 741.1429 [M + Na]+ (Calcd for C36H30O16Na:
741.1430).
2.5. (S)-PGME and (R)-PGME amides of compounds 1, 3, 5, and
7 for determining the configuration of C-8
Separately, (S)- or (R)-PGME (20 mg) was added to 1 and
5 (10 mg each) in DMF (0.8 mL), and then PyBOP (20 mg),
HOBT (20 mg), and N-methylmorpholine (20 μL) were
added, and the mixtures were stirred for 10 h at room
temperature. The reactions produced the (S)-PGME amide of
1 (10: 11.5 mg) and the (R)-PGME amide of 1 (11: 8.1 mg)
and the (S)-PGME amide of 5: (12: 7.0 mg) and the
(R)-PGME amide of 5 (13: 7.0 mg), respectively. Separately,
(S)- or (R)-PGME (20 mg) was added to 3 (2.0 mg) and 7
(1.0 mg) in DMF (0.8 mL), and then PyBOP (8 mg), HOBT
(8 mg), and N-methylmorpholine (20 μL) were added, and
the mixtures were stirred for 10 h at room temperature. The
reactions of 3 produced the (S)-PGME amide: 12 (1.2 mg)
and 14 (0.8 mg) and the (R)-PGME amide: 13 (1.2 mg) and
15 (0.4 mg), respectively. The reactions of 7 produced the
(S)-PGME amide (10: 0.6 mg) and the (R)-PGME amide (11:
0.4 mg), respectively.
2.3.3. Monardic acid C (3)
Colorless amorphous solid; [α]25 +41.4 (c 0.88, MeOH);
D
UV (MeOH) λmax (log ε) 256 (4.19), 290 (4.18), 313 (4.21);
ECD (c 0.05, MeOH) nm ([θ]) 233 (−72,500), 254 (61,900),
278 (6700), 298 (39,900) nm; 1H NMR (CD3OD, 400 MHz)
and 13C NMR (CD3OD, 100 MHz) see Table 1; HRFABMS
(positive) m/z 561.1012 [M + Na]+ (Calcd for C27H22O12Na:
561.1008).
2.3.4. Lithospermic acid C (7)
Colorless amorphous solid; [α]25 +40.7 (c 0.54, MeOH);
D
UV (MeOH) λmax (log ε) 257 (4.21), 290 (4.19), 311 (4.23);
ECD (c 0.05, MeOH) nm ([θ]) 230 (99,800), 255 (−46,600),
291 (−7700), 336 (11,000 ) nm; 1H NMR (CD3OD, 400 MHz)
and 13C NMR (CD3OD, 100 MHz) see Table 1; HRFABMS
(positive) m/z 561.1012 [M + Na]+ (Calcd for C27H22O12Na:
561.1008).
2.5.1. (S)-PGME amide of 1 (10)
Colorless amorphous solid; [α]24 −63.5 (c 0.34, MeOH);
D
UV (MeOH) λmax (log ε) 256 (4.31), 290 (4.28), 314 (4.29);
ECD (c 0.02, MeOH) nm ([θ]) 218 (96,500), 253 (−43,300), 344
(–7600) nm; 1H NMR (CD3OD, 400 MHz) δ 7.77 (1H, d, J =
16.0 Hz), 7.27–7.40 (8H, overlapping), 7.21 (2H, overlapping),
7.19 (1H, d, J = 8.5 Hz), 6.81 (1H, d, J = 8.5 Hz), 6.78 (1H, d,
J = 2.0 Hz), 6.72 (1H, d, J = 8.0 Hz), 6.70 (1H, d, J = 2.0 Hz),
6.61 (1H, d, J = 8.0 Hz), 6.61 (1H, dd, J = 8.0, 2.0 Hz), 6.53 (1H,
dd, J = 8.0, 2.0 Hz), 6.39 (1H, d, J = 16.0 Hz), 5.64 (1H, d, J =
6.0 Hz), 5.48 (1H, dd, J = 3.5, 3.0 Hz), 5.45 (1H, dd, J = 3.5,
3.0 Hz), 5.35 (1H, t, J = 6.5 Hz), 4.53 (1H, d, J = 6.0 Hz), 3.66
(3H, s), 3.54 (3H, s), 2.98 (2H, d, J = 6.5 Hz); FABMS (positive)
m/z 833 [M + H]+, 855 [M + Na]+.
2.4. Alkaline hydrolysis of compounds 1–3 and 7, and condensation
with (S)-PGME
Each compound [compounds 1 (20.0 mg), 2, 3, and 7
(each 1.0 mg)] was dissolved in 10% NaOH (0.5 mL) and
stirred for 2 h. The reaction mixtures were passed through a
Dowex 50 W × 2 column (1.0 × 7.0 cm) and eluted with H2O
(100 mL). The residue from 1 was subjected to HPLC [column,
Mightysil RP18-GP (10 × 250 mm); solvent, CH3CN–H2O
containing 0.2% TFA (20:80); flow rate, 0.8 mL/min; detector,
UV 210 nm] to yield 7-epiblechnic acid (8: 1.9 mg) and
(2R)-3-(3,4-dihydroxyphenyl)-2-hydroxypropanoic acid (9:
4.6 mg) [18,19]. The residues from 2, 3, 7 and 9 from 1 were
dissolved in DMF and (S)-PGME, benzotriazol-1-yl-oxy-tris-
pyrrolidinophosphonium hexafluorophosphate (PyBOP), 1-
2.5.2. (R)-PGME amide of 1 (11)
Colorless amorphous solid; [α]24 −126.9 (c 0.35,
D
MeOH); UV (MeOH) λmax (log ε) 257 (4.25), 289 (4.39),

Table 1
1H and 13C NMR spectroscopic data (methanol-d4) for compounds 1–3 and 7.
1
2
3
7
Position
δ
H (J in Hz)
δc
HMBC
δH (J in Hz)
δc
HMBC
δH (J in Hz)
δc
HMBC
δ
H (J in Hz)
δc
HMBC
(H to C)
(H to C)
(H to C)
(H to C)
1
2
3
4
5
6
7
8
133.7
113.5
146.6
146.7
116.4
118.3
88.8
133.9
113.3
146.2
146.7
116.4
118.4
88.2
129.3
115.1
146.1
146.5
115.9
119.6
88.4
129.4
115.1
146.0
146.5
115.9
119.6
88.4
6.80 (d, 2.0)
7
6.75 (d, 2.0)
4,6,7
6.98 (d, 2.0)
1,6,7
6.98 (d, 2.0)
4,7
6.77 (d, 8.5)
6.73 (dd, 8.5, 2.0)
5.90 (d, 4.5)
3
4,7
6.74 (d, 8.0)
6.62 (dd, 8.0 2.0)
5.85 (d, 4.5)
6.76 (d, 8.0)
6.86 (dd, 8.0, 2.0)
5.94 (d, 9.5)
1,3
4,7
6.76 (d, 8.0)
6.86 (dd, 8.0, 2.0)
5.94 (d, 9.5)
1,3
4,7
2,7
1,2,6,8,9,2',3'
1,7,9,1',2',3'
1,2,6,8,9,2',3'
1,7,9,1',2',3'
1,2,6,8,9,2',3'
1,7,9,2',3'
1,2,6,8,9,2',3'
1,7,9,2',3'
4.36 (d, 4.5)
57.5
4.42 (d, 4.5)
57.3
4.61 (d, 9.5)
55.2
4.63 (d, 9.5)
55.4
9
175.1
124.6
127.6
148.8
145.2
118.3
121.7
144.0
116.4
168.2
129.3
117.5
146.1
145.2
116.3
121.9
37.9
172.5
124.9
126.3
149.1
145.3
118.5
122.0
144.3
116.6
168.2
129.4
117.6
146.6
145.3
116.4
122.0
37.9
173.6
124.3
129.3
149.6
145.3
118.1
122.6
143.8
116.4
168.1
129.3
117.6
146.0
145.4
116.5
121.9
37.9
173.8
124.3
129.4
149.6
145.2
118.1
122.7
143.8
116.4
168.0
129.3
117.7
146.0
145.4
116.7
122.0
38.0
1'
2'
3'
4'
5'
6
7'
8'
9'
1''
2''
3''
4''
5''
6''
7''
6.81 (d, 8.5)
7.21 (d, 8.5)
7.81 (d, 16.0)
6.34 (d, 16.0)
1'
6.82 (d, 8.5)
7.19 (d, 8.5)
7.69 (d, 16.0)
6.30 (d, 16.0)
1',3'
6.81 (d, 8.5)
7.15 (d, 8.5)
7.58 (d, 16.0)
6.30 (d, 16.0)
1',3'
6.81 (d, 8.5)
7.15 (d, 8.5)
7.61 (d, 16.0)
6.32 (d, 16.0)
1',3'
2',4',7'
1',2',6',9'
1',9'
2',4',7'
1',2',6',9'
1',9'
2',4',7'
1',2',6',9'
1',9'
2',4',7'
1',2',6',9'
1',9'
6.76 (d, 2.0)
7''
6.74 (d, 2.0)
4'',6''
6.76 (d, 2.0)
1''
6.75 (d, 2.0)
4'',7''
6.68 (d, 8.0)
3''
4'',7''
1'',2'',6'',8'',9''
1'',2'',6'',8'',9''
9',1'',7'',9''
6.65 (d, 8.0)
6.70 (d, 8.0)
1''
4''
6.68 (d, 8.0)
3''
2'',4'',7''
1'',2'',6'',8'',9''
1'',2'',6'',8'',9''
1'',7'',9''
6.62 (dd, 8.0, 2.0)
2.99 (dd, 14.0, 8.0)
3.08 (dd, 14.0, 4.5)
5.14 (dd, 8.0, 4.5)
6.60 (dd, 8.0, 2.0)
2.99 (dd, 14.0, 8.5)
3.07 (dd, 14.0, 4.5)
5.15 (dd, 8.5, 4.5)
4''
6.62 (dd, 8.0, 2.0)
2.97 (dd, 14.0, 8.5)
3.08 (dd, 14.0, 4.0)
5.11 (dd, 8.5, 4.0)
6.60 (dd, 8.0, 2.0)
2.98 (dd, 14.0, 8.0)
3.06 (dd, 14.0, 4.5)
5.16 (dd, 8.0, 4.5)
1'',2'',6'',8'',9''
1'',2'',6'',8'',9''
1'',2'',6'',8'',9''
1'',2'',6'',8'',9''
9',1'',7'',9''
8''
9''
74.9
173.4
74.9
173.5
74.9
173.3
74.9
173.5

T. Murata et al. / Fitoterapia 91 (2013) 51–59
55
315 (4.39); ECD (c 0.02, MeOH) nm ([θ]) 221 ( 101,600), 256
(–37000), 337 (–3700) nm; 1H NMR (CD3OD, 400 MHz) δ
7.59 (1H, d, J = 16.0 Hz), 7.25–7.40 (8H, overlapping), 7.18
(2H, overlapping), 7.14 (1H, d, J = 8.5 Hz), 6.86 (1H, d, J =
2.0 Hz), 6.78 (1H, d, J = 8.5 Hz), 6.78 (1H, d, J = 8.0 Hz),
6.78 (1H, dd, J = 8.0, 2.0 Hz), 6.70 (1H, d, J = 2.0 Hz), 6.65
(1H, d, J = 8.0 Hz), 6.55 (1H, dd, J = 8.0, 2.0 Hz), 6.30 (1H,
d, J = 16.0 Hz), 5.71 (1H, d, J = 5.5 Hz), 5.45 (1H, dd, J =
3.5, 3.0 Hz), 5.43 (1H, dd, J = 3.5, 3.0 Hz), 5.27 (1H, dd,
J = 8, 4.5 Hz), 4.51 (1H, d, J = 5.5 Hz), 3.69 (3H, s), 3.67
(3H, s), 3.01 (1H, dd, J = 14.0, 4.5 Hz), 2.95 (1H, dd, J =
14.0, 8.0 Hz); FABMS (positive) m/z 833 [M + H]+, 855
ECD (c 0.02, MeOH) nm ([θ]) 223 (−130,700), 256 (47,400),
299 (26,700) nm; 1H NMR (CD3OD, 400 MHz) δ 7.57 (1H, d,
J = 16.0 Hz), 7.10–7.40 (10H, overlapping), 7.14 (1H, d, J =
8.5 Hz), 6.93 (1H, d, J = 2.0 Hz), 6.81 (1H, d, J = 8.0 Hz), 6.78
(1H, dd, J = 8.5, 2.0 Hz), 6.73 (1H, d, J = 8.0 Hz), 6.68 (1H, d,
J = 2.0 Hz), 6.64 (1H, d, J = 8.0 Hz), 6.53 (1H, dd, J = 8.0,
2.0 Hz), 6.32 (1H, d, J = 16.0 Hz), 5.88 (1H, d, J = 9.0 Hz), 5.48
(1H, m), 5.27 (1H, dd, J = 8.0, 5.0 Hz), 5.08 (1H, m), 4.73
(1H, d, J = 9.0 Hz), 3.71 (3H, s), 3.58 (3H, s), 2.95 (1H, dd, J =
14.0, 8.0 Hz), 3.02 (1H, dd, J = 14.0, 5.0 Hz); FABMS (positive)
m/z 833 [M + H]+, 855 [M + Na]+.
[M + Na]+
.
2.6. Hyaluronidase inhibitory assay
2.5.3. (S)-PGME amide of 5 (12)
The inhibitory activity of hyaluronidase was determined by
the Morgan–Elson method, which was modified by Davidson
and Aronson. The assay was carried out in accordance with the
procedure reported previously [6]. Samples dissolved in 0.1 M
acetate buffer (0.2 mL) and hyaluronidase (Type IV-S: from
bovine testes, Sigma Chemical Co., St. Louis, USA) in buffer
(final concentration: 400 unit/mL, 0.1 mL) were mixed and the
mixture was incubated at 37 °C for 20 min. Then, compound
48/80 (Sigma Chemical Co.) in buffer (final concentration:
0.3 mg/mL, 0.2 mL) was added and incubated at 37 °C for
20 min. After hyaluronic acid potassium salt, from rooster
comb (Sigma Chemical Co.) in buffer (final concentration:
0.4 mg/mL, 0.5 mL) had been added, the mixture was incubat-
ed at 37 °C for 40 min. Then, the reaction was stopped by
adding 0.4 M NaOH and borate solution and then boiling the
mixture in a H2O bath for 3 min. An acetone solution of
dimethylaminobenzaldehyde (Wako Pure Chemical Industries
Ltd., Osaka, Japan) (6 mL) was then added and incubated at
37 °C for 20 min. Acetate buffer was added in place of the
sample as a control, and the buffer was added in place of
hyaluronidase in buffer as a blank. The enzyme inhibitory
activity (%) was calculated as follows: Inhibitory activity (%) =
Colorless amorphous solid; [α]25 + 134.5 (c 0.84,
D
MeOH); UV (MeOH) λmax (log ε) 257 (4.34), 291 (4.28),
314 (4.31); ECD (c 0.02, MeOH) nm ([θ]) 221 (116,500), 253
(47,000), 334 (9900) nm; 1H NMR (CD3OD, 400 MHz) δ 7.63
(1H, d, J = 16.0 Hz), 7.23–7.35 (8H, overlapping), 7.15 (2H,
overlapping), 7.15 (1H, d, J = 8.5 Hz), 6.88 (1H, d, J =
2.0 Hz), 6.79 (1H, d, J = 8.5 Hz), 6.79 (1H, d, J = 8.0 Hz),
6.79 (1H, dd, J = 8.0, 2.0 Hz), 6.62 (1H, d, J = 2.0 Hz), 6.61
(1H, d, J = 8.0 Hz), 6.46 (1H, dd, J = 8.0, 2.0 Hz), 6.30 (1H,
d, J = 16.0 Hz), 5.72 (1H, d, J = 6.0 Hz), 5.42 (1H, br s), 5.40
(1H, br s), 5.29 (1H, t, J = 7.5 Hz), 4.54 (1H, d, J = 6.0 Hz),
3.72 (3H, s), 3.65 (3H, s), 2.82 (2H, d, J = 7.5 Hz); FABMS
(positive) m/z 833 [M + H]+, 855 [M + Na]+
.
2.5.4. (R)-PGME amide of 5 (13)
Colorless amorphous solid; [α]25 + 55.5 (c 0.84, MeOH);
D
UV (MeOH) λmax (log ε) 257 (4.33), 290 (4.27), 314 (4.29); ECD
(c 0.02, MeOH) nm ([θ]) 219 (−82,500), 254 (46,200), 334
(12500) nm; 1H NMR (CD3OD, 400 MHz) δ 7.77 (1H, d, J =
16.0 Hz), 7.28–7.37 (10H, overlapping), 7.20 (1H, d, J = 8.5 Hz),
6.81 (1H, d, J = 8.5 Hz), 6.77 (1H, d, J = 2.0 Hz), 6.71 (1H, d,
J = 8.0 Hz), 6.71 (1H, d, J = 2.0 Hz), 6.65 (1H, d, J = 8.0 Hz),
6.61 (1H, dd, J = 8.0, 2.0 Hz), 6.55 (1H, dd, J = 8.0, 2.0 Hz), 6.37
(1H, d, J = 16.0 Hz), 5.63 (1H, d, J = 6.0 Hz), 5.48 (1H, br s),
5.47 (1H, br s), 5.34 (1H, t, J = 6.5 Hz), 4.52 (1H, d, J = 6.0 Hz),
3.64 (3H, s), 3.54 (3H, s), 3.02 (2H, d, J = 6.5 Hz); FABMS
(positive) m/z 833 [M + H]+, 855 [M + Na]+.
[(Control Abs
− Control
Abs 600 nm) − (Sample
600 nm
blank
Abs 600 nm − Sample blank Abs 600 nm)]/(Control Abs 600 nm
Control blank Abs 600 nm) × 100.
−
2.7. Histamine release inhibitory assay
Degranulation of KU812F cells was monitored by mea-
suring the release of histamine, which was carried out in
accordance with the reported procedure [20]. KU812F cells
(Japan Human Sciences Foundation, Tokyo, Japan) were
stimulated with 1 ng/mL of IL-4 for 2 weeks to prepare
mature basophils. KU812F cells (6.25 × 106 cells/mL) were
resuspended in 400 μL Tyrode buffer at pH 7.2. Then, 50 μM
A23187 (BioVision, Inc., California, USA) (50 μL) was incu-
bated with each sample (50 μL) as the test solution and was
then added to the cell suspension (Test Sample). Tyrode
buffer (50 μL) was added in place of the sample as a positive
control (PC), buffer (100 μL) was added in place of the
sample and A23187 as a negative control (NC), and buffer
(50 μL) was added in place of the A23187 as a negative
control-2 (NC-2). After adding the mixture to the cells, the
mixture was incubated at 37 °C for 20 min, and reaction was
terminated by cooling at 4 °C for 15 min. The cell suspension
was then centrifuged at 1000 rpm for 3 min, and the
supernatant moved to a 50 μL microtiter plate. It was mixed
2.5.5. (S)-PGME amide of 3 (14)
Colorless amorphous solid, [α]21D +46.3 (c 0.06, MeOH); UV
(MeOH) λmax (log ε) 257 (4.30), 290 (4.19), 311 (4.20); ECD
(c 0.02, MeOH) nm ([θ]) 219 (67,700), 234 (−35,800), 254
(34,400), 297 (21,900) nm; 1H NMR (CD3OD, 400 MHz) δ 7.64
(1H, d, J = 16.0 Hz), 7.22–7.32 (8H, overlapping), 7.16 (1H, d,
J = 8.5 Hz), 6.93 (2H, overlapping), 6.89 (1H, d, J = 2.0 Hz),
6.81 (1H, d, J = 8.5 Hz), 6.74 (1H, d, J = 2.0 Hz), 6.73 (1H, dd,
J = 8.0, 2.0 Hz), 6.64 (1H, d, J = 8.0 Hz), 6.63 (1H, d, J =
8.0 Hz), 6.56 (1H, dd, J = 8.0, 2.0 Hz), 6.37 (1H, d, J = 16.0 Hz),
5.82 (1H, d, J = 9.0 Hz), 5.50 (1H, m), 5.33 (1H, t, J = 7.0 Hz),
4.61 (1H, d, J = 9.0 Hz), 3.66 (3H, s), 3.50 (3H, s), 3.01 (2H, d,
J = 7.0 Hz); FABMS (positive) m/z 833 [M + H]+, 855
[M + Na]+
.
2.5.6. (R)-PGME amide of 3 (15)
Colorless amorphous solid; [α]22 –40.0(c 0.34, MeOH);
D
UV (MeOH) λmax (log ε) 257 (4.30), 290 (4.29), 314 (4.30);

56
T. Murata et al. / Fitoterapia 91 (2013) 51–59
with 50 μL of 1 M NaOH and 50 μL of 0.2% o-phthalaldehyde
(Wako, Osaka, Japan), and kept for 5 min at room temper-
ature. The reaction was terminated by adding 50 μL of 3 N
HCl, and then the fluorescence intensity was measured using
a spectrofluorophotometer with excitation at 355 nm and
emission at 460 nm. The inhibitory percentage histamine
release was calculated as follows: inhibitory activity (%) =
[1 − (Test Sample − NC-2)/(PC − NC)] × 100. The significance
of the differences was examined by Tukey's test.
The PGME method has been used to define the absolute
configuration of natural products [22]. In this study, the
PGME method was employed to determine the absolute
configuration at C-8 of
treatment of 1 with (S)-PGME and (R)-PGME produced 10
and 11, respectively. Similarly, treatment of with
1 and its analogues. Separate
5
(S)-PGME and (R)-PGME produced 12 and 13, respectively.
For 1, the values of the 1H NMR chemical shift differences [Δδ
(ppm) = δ(S)-PGME amide (10) – δ(R)-PGME amide (11)] adjacent to
C-8 were consistent with the (8R) absolute configuration [22],
except for the negative values of H-5′′ and H-6′′ affected by the
second PGME moiety at C-8′′ (Fig. 2). Furthermore, the values
around C-8 had signs opposite to those of 5 suggesting (8S)
configuration [Δδ (ppm) = δ(S)-PGME amide (12) – δ(R)-PGME amide
(13)] (Fig. 2). These data suggested that the PGME method is
applicable to the analysis of the dihydrobenzofuran moiety and
supported the (8R) configuration of 1.
2.8. HPLC analysis of compounds 1, 2, 5, and 6 using reversed-
phase columns
The methanol solutions of compounds 1, 2, 5, and 6 were
subjected to HPLC using a C18 column [column, Cosmosil
5C18-AR-II (4.6 × 250 mm); solvent, CH3CN–H2O contain-
ing 0.2% TFA (22.5:77.5); flow rate, 1.0 mL/min; detector,
UV 310 nm]. The retention times were as follows:
1
The molecular formula of 2 was established as C36H30O16
on the basis of HRFABMS (m/z 741.1429 [M + Na]+),
which, when compared with 1, corresponded to an additional
phenylpropanoid unit of formula C9H8O4. The molecular
formula revealed 2 to be a phenylpropanoid tetramer. The 1H
and 13C NMR resonances of 2 (Table 1) were close to those of 6
[4,13]. The resonances of the dihydrobenzofuran moiety
(δH 5.85, d, J = 4.5 Hz, H-7; 4.42, d, J = 4.5 Hz, H-8; δC 88.2,
C-7; 57.3, C-8) suggested that the methine protons of 2 were
in a trans configuration (7S,8S or 7R,8R) [19,21]. The ECD
spectrum of 2 showed that its Cotton effects (CEs) were similar
to those of 1 and opposite to those of 5 and 6 [13], suggesting
that the absolute configuration of the dihydrobenzofuran
moiety of 2 was also (7R,8R) (Fig. 3) [18]. In the 1H NMR
spectrum, two –CH(O–)–CH2– units [δH 2.99 (dd, J = 14.0,
8.5 Hz, H-7′′), 3.07 (dd, J = 14.0, 4.5 Hz, H-7′′), 5.15 (dd, J =
8.5, 4.5 Hz, H-8′′)] and [δH 3.00 (dd, J = 14.0, 7.5 Hz, H-7′′′),
3.09 (dd, J = 14.0, 4.5 Hz, H-7′′′), 5.20 (dd, J = 7.5, 4.5 Hz,
H-8′′′)] and two aromatic spin systems [δH 6.74 (d, J = 2.0 Hz,
H-2′′), 6.65 (d, J = 8.0 Hz, H-5′′), 6.60 (dd, J = 8.0, 2.0 Hz,
H-6′′)] and [δH 6.67 (d, J = 2.0 Hz, H-2′′′), 6.62 (d, J = 8.0 Hz,
H-5′′′), 6.48 (dd, J = 8.0, 2.0 Hz, H-6′′′)] suggested the presence
of two 3-(3,4-dihydroxyphenyl)-2-hydroxypropanoyl moieties.
The configurations of C-8′′ and C-8′′′ of 2 were determined to
be (R) from the retention time by performing HPLC analysis
of the (S)-PGME amide of 3-(3,4-dihydroxyphenyl)-2-
hydroxypropanoic acid from the alkaline hydrolysis of 2
[18]. From these data, the structure of 2 was determined as
monardic acid B, which is the (7R,8R) isomer of LAB.
Compounds 3 and 7 also had the molecular formula
(15.3 min), 2 (26.5 min), 5 (16.3 min), and 6 (24.0 min).
When the C30 column was used [column, Develosil C30-UG-5
(4.6 × 250 mm); solvent, CH3CN–H2O containing 0.2% TFA
(25:75); flow rate, 1.0 mL/min; detector, UV 310 nm], the
retention times were as follows: 1 (14.1 min), 2 (23.1 min), 5
(15.4 min), and 6 (22.3 min).
3. Results and discussion
Whole plants of M. fistulosa were extracted with acetone–
H2O (8:2). The extract was concentrated at reduced pressure,
suspended in H2O, and extracted with Et2O. The aqueous
extract was subjected to column chromatography and prepar-
ative HPLC to afford compounds 1–4. Similarly, compounds
5–7 were isolated from the acetone–H2O (8:2) extract of the
roots of L. erythrorhizon. Rosmarinic acid (4) [18], LA (5) [9],
and LAB (6) [4] were identified by comparing their observed
and reported spectroscopic data.
The molecular formula of 1 (C27H22O12, m/z 539.1196
[M + H]+) was established using HRFABMS. In the 13C NMR
spectra (Table 1), 27 carbons, including three carbonyl carbons
(δc 175.1, C-9; 168.2, C-9′; 173.4, C-9′′), suggested that 1 was a
phenylpropanoid trimer. In the 1H NMR spectrum (Table 1),
a –CH(O–)–CH2– unit [δH 2.99 (dd, J = 14.0, 8.0 Hz, H-7′′), 3.08
(dd, J = 14.0, 4.5 Hz, H-7′′), 5.14 (dd, J = 8.0, 4.5 Hz, H-8′′)]
and the aromatic spin system [δH 6.76 (d, J = 2.0 Hz, H-2′′), 6.68
(d, J = 8.0 Hz, H-5′′), 6.62 (dd, J = 8.0, 2.0 Hz, H-6′′)] suggested
the presence of a 3-(3,4-dihydroxyphenyl)-2-hydroxypropanoyl
moiety. The NMR spin system at δH 5.90, d, J = 4.5 Hz, H-7; 4.36,
d, J = 4.5 Hz, H-8; δC 88.8, C-7; 57.5 C-8 indicated the presence
of a dihydrobenzofuran moiety. The 1H and 13C NMR resonances
of 1 were close to those of 5 [9]. In fact, 2D NMR data suggested
that the molecular structure of 1 was the same as that of 5. The
C
27H22O12 (HRFABMS; 3: m/z 561.1012 [M + Na]+ and 7: m/z
561.1012 [M + Na]+) suggesting a phenylpropanoid trimer.
In their 1H NMR spectra (Table 1), the coupling constants
between H-7 and H-8 (J = 9.5 Hz) suggested that they were
cis-related (7S,8R or 7R,8S) [19,23]. The resonances of H-8 (3: δH
4.61, 7: δH 4.63) were downfield shifted relative to those of the
trans isomers, which also supported the cis-dihydrobenzofuran
moiety [10]. These data indicated that one has (7S,8R)
configuration and the other a (7R,8S) configuration. Some
cis-oriented LA analogues have been reported from natural
sources, e.g., cis-lithospermic acid from S. yunnanensis C. H.
Wright [23] and clinopodic acid N from Clinopodium gracile
(Benth.) O. Kuntze [24]. However, there was insufficient
discussion about the absolute configuration, and ECD data
3
J7,8 coupling constant (4.5 Hz) suggested that the methine
protons were in a trans configuration (7S,8S or 7R,8R) [19,21].
Alkaline hydrolysis of 1 gave 7-epiblechnic acid (8) [19],
indicating the (7R,8R) configuration of the dihydrobenzofuran
moiety (Scheme 1). The absolute configuration of C-8′′ of 1 was
R, as determined from its hydrolysate (9), which was identical to
that of (2R)-3-(3,4-dihydroxyphenyl)-2-hydroxypropanoic acid
in our previous report (Scheme 1) [18]. From these data, the
structure of 1 was determined to be monardic acid A, which is
the (7R,8R) isomer of LA.

T. Murata et al. / Fitoterapia 91 (2013) 51–59
57
Scheme 1. PGME amidations and alkaline hydrolysis of compound 1a.
from these analogues in previous reports. Compounds 3 and 7
showed nearly identical specific optical rotations and NMR
resonances, and it is difficult to distinguish these compounds
using these data alone. Although the PGME method is an
empirical method similar to Mosher's methods, the applied
models of 1 and 5 suggested that the method is applicable to
compounds comprising dihydrobenzofurans. According to this
method, the (S)-PGME and (R)-PGME amidation of 3 produced
14 and 15, and the values of the 1H NMR chemical shift
indicated that C-8 was in the (R) configuration (Fig. 2) [22]. That
is, 3 had the (7S,8R) configuration of the dihydrobenzofuran
moiety and 7 had the (7R,8S) configuration of this moiety. The
absolute configuration of C-8′′ in 3 and 7 was determined to be
(R) using the same HPLC method of the (S)-PGME amide as used
for 2. Thus, 3 was identified as the (7S,8R) cis isomer of LA, and 7
was determined to be the (7R,8S) cis isomer of LA, hereafter
referred to as monardic acid C and lithospermic acid C,
respectively.
differences [Δδ (ppm) = δ(S)-PGME amide (14) − δ(R)-PGME amide (15)
]
Using the PGME method, amidation of 3 with (S)-PGME
and (R)-PGME produced not only 14 and 15 but also 12
and 13. Each of these products (12 and 13) had a C-8
configuration that was opposite to the C-8 configuration in 3.
These products revealed that the cis dihydrobenzofuran
moiety of 3 changed to the trans dihydrobenzofuran moiety
during the reaction. Similarly, the main products of the
(S)-PGME and (R)-PGME amidation of 7 were 10 and 11. One
hypothesis to explain these unexpected products was their
epimerization at C-8 through enolization via a sterically
hindered condensation reaction during PGME amidation.
Thus, attention to epimerization is required when the PGME
method is used.
+
+
0.06
0.17
0.08
0.18
+
0.08
0.07
0.11
+
0.09
+
+
0.02
0.14
0.02 0.05 0.07
0.02
+
0.18
+
+
+
0.09
0.03 0.05
0.09
0.09
0
+
0.05
0.20
0.08
0
0.04
0.04
0.02
In the ECD spectrum of 3, a negative CE at 230 nm and
a positive CE at 254 nm were observed, and these effects
were opposite for 7 (Fig. 3). In many compounds with
(S)-amide (10)
(R)-amide (11)
(S)-amide (12)
(R)-amide (13)
0.17
0
0.04
100
50
0
0.06
0.13
+
0.07
+
+
+
0.05
0.03 0.02
+
0.06
1
2
+
0.05
(S)-PGME amide
(R)-PGME amide
3
−50
+
0.01
5
7
0.01
+
0.03
−100
200
250
300
350
(S)-amide (14)
(R)-amide (15)
Wavelength (nm)
Fig. 2. The values (in ppm) of the 1H NMR chemical shift differences for 1, 3,
and 5 and the model of the PGME method.
Fig. 3. ECD spectra of 1–3, 5, and 7.

58
T. Murata et al. / Fitoterapia 91 (2013) 51–59
b
: 1 mg/mL
ab
dihydrobenzofuran moieties, the configurations at C-7 were
determined from the 220–240 nm region [24,25]. However,
it has not been demonstrated whether this empirical rule is
applicable to the C-7 configuration in the dihydrobenzofuran
moieties of isolates (1–3 and 7). Specific CEs were observed
around 220–240 nm and 250–260 nm in the ECD spectra of
derivatives obtained here. In the region around 250–260 nm,
compounds with the (7R) configuration (1, 2, 7, 8, 10,
and 11) had negative values, and in contrast, compounds
with the (7S) configuration (3, 5, 6, and 12–15) had positive
values. On the other hand, signals around 220–240 nm
were not related to the configuration of C-7 or were not
observed. These ECD data suggested that the values around
250–260 nm were responsible for the absolute configuration
at C-7 in the dihydrobenzofuran moiety [3,4,13].
Monardic acid A (1)
Rosmarinic acid (4)
Lithospermic acid (5)
Epigallocatechin-
3-O-(3-O-methyl)gallate
a
: 10 mg/mL
: 100 mg/mL
b
b
a
b
b
a
b
b
a
0
20
40
60
80
100
Inhibitory activity of histamine release (%)
Fig. 4. Inhibitory effect of 1, 4, and 5 on the degradation of KU812F cells.
SD for n = 5. Values not sharing a common letter
(a, ab, or b) are significantly different between groups (p b 0.05).
Results are means
Acknowledgments
We thank Mr. S. Sato and Mr. T. Matsuki of Tohoku
Pharmaceutical University for assisting with MS measurements.
The diastereomers isolated in this research are expected to
have a variety of biological activities similar to LA and LAB.
To identify anti-allergic and anti-inflammatory agents [6], the
hyaluronidase inhibitory activity of these diastereomers was
measured (Table 2). The compounds (1–3 and 7) showed
potent activities (IC50 381–406 μM) similar to 5 and 6. The
positive control was disodium cromoglycate (DSCG). Moreover,
1, 4, and 5 had moderate inhibitory effects on histamine release
on the degranulation process in a human basophilic cell line
(KU812F) similar to the positive control epigallocatechin-3-
O-(3-O-methyl)gallate (Fig. 4) [20]. These results suggested that
these compounds possibly have anti-allergic effects.
The biosynthetic pathways of LA and LAB are unknown, but
it is expected that they are produced from rosmarinic acid [26].
Compound 1 was obtained as the main phenylpropanoid
oligomer, and 5 and 6 were not detected in our investigation
of M. fistulosa. Compounds 1, 2, 5, and 6 showed different
retention times of the observed peaks by HPLC using reversed-
phase columns (Experimental Section). For the four diastereo-
mers, only peaks of the (7R,8R) diastereomers (1 and 2)
were detected in the chromatograms of the Et2O and aqueous
extracts of M. fistulosa, and only peaks of the (7S,8S)
diastereomers (5 and 6) were detected in the chromatograms
of the Et2O and aqueous extracts of L. erythrorhizon. From these
observations, each plant species may have a stereoselective
biosynthetic pathway to produce either the (7R,8R) or (7S,8S)
dihydrobenzofuran moieties. The cis isomers (3 and 7) were C-7
epimers of 1 and 5, respectively. It is assumed that 3 and 7 are
epimerized from 1 and 5 through the ring-opening intermedi-
ate [19].
References
[1] Kelly CJ, Mahajan JR, Brooks LC, Neubert LA, Breneman WR, Carmack
M. Polyphenolic acids of Lithospermum ruderale Dougl. ex Lehm.
(Boraginaceae). 1. Isolation and structure determination of lithospermic
acid. J Org Chem 1975;40:1804–15.
[2] Kelly CJ, Harruff RC, Carmack M. The polyphenolic acids of
Lithospermum ruderale. II. Carbon-13 nuclear magnetic resonance of
lithospermic and rosmarinic acids. J Org Chem 1976;41:449–55.
[3] Ai CB, Li LN. Stereostructure of salvianolic acid B and isolation of
salvianolic acid C from Salvia miltiorrhiza. J Nat Prod 1988;51:145–9.
[4] Tanaka T, Morimoto S, Nonaka G, Nishioka I, Yokozawa T, Chung HY,
Oura H. Magnesium and ammonium-potassium lithospermates B,
the active principles having a uremia-protective effect from Salvia
miltiorrhiza. Chem Pharm Bull 1989;37:340–4.
[5] Abd-Elazen IS, Chen HS, Bates RB, Huang RC. Isolation of two highly potent
and non-toxic inhibitors of human immunodeficiency virus type
1
(HIV-1) integrase from Salvia miltiorrhiza. Antiviral Res 2002;55:91–106.
[6] Murata T, Miyase T, Yoshizaki F. Cyclic spermidine alkaloids and
flavone glycosides from Meehania fargesii. Chem Pharm Bull 2010;58:
696–702.
[7] Yokozawa T, Chung HY, Oura H, Nonaka G, Nishioka T. Isolation of the
active component having the uremia-preventive effect from Salvia
miltiorrhizae radix extract. Chem Pharm Bull 1988;36:316–20.
[8] Yamamoto H, Zhao P, Yazaki K, Inoue K. Regulation of lithospermic
acid B and shikonin production in Lithospermum erythrorhizon cell
suspension cultures. Chem Pharm Bull 2002;50:1086–90.
[9] O’Malley SJ, Tan KL, Watzke A, Bergman RG, Ellman JA. Total synthesis
of (+)-lithospermic acid by asymmetric intramolecular alkylation via
catalytic C-H bond activation. J Am Chem Soc 2005;127:13496–7.
[10] Fischer J, Savage GP, Coster MJ. A concise route to dihydrobenzo[b]furans:
formal total synthesis of (+)-lithospermic acid. Org Lett 2011;13:3376–9.
[11] Wang DH, Yu JQ. Highly convergent total synthesis of (+)-lithospermic
acid via a late-stage intermolecular C-H olefination. J Am Chem Soc
2011;133:5767–9.
[12] Ghosh AK, Cheng X, Zhou B. Enantioselective total synthesis of
(+)-lithospermic acid. Org Lett 2012;14:5046–9.
[13] Watzke A, O’Malley SJ, Bergman RG, Ellman JA. Reassignment of the
configuration of Salvianolic acid B and establishment of its identity
with lithospermic acid B. J Nat Prod 2006;69:1231–3.
[14] Chevallier A, Nanba T. Sekai Yakuyou Syokubutu Hyakka Jiten, Original
title: The encyclopedia of medicinal plants, first published in Great
Britain in 1996 by Dorling Kindersley Ltd., London. Tokyo: Seibundo
Shinkosya Publishing Co., Ltd; 2000 234.
Table 2
Hyaluronidase inhibitory activity of 1–7 and DSCG.
Compound
Hyaluronidase inhibition
[15] Stuart M. Dr. Stuart’s encyclopedia of herbs and herbalism. U.K:
Edgerton International Ltd; 1994 225.
IC50 (μ M)
1
2
3
4
5
6
7
406
398
404
1360
379
373
381
1020
[16] Prather LA, Monfils AK, Posto AL, Williams RA. Monophyly and phylogeny
of Monarda (Lamiaceae): evidence from the internal transcribed spacer
(ITS) region of nuclear ribosomal DNA. Syst Bot 2002;27:127–37.
[17] Fujita N, Yoshida Y. Ueber die anatomie von radix Lithospermi. J Pharm
Soc Jpn 1937;57:368–91 [in Japanese].
[18] Murata T, Sasaki K, Sato K, Yoshizaki F, Yamada H, Mutoh H, Umehara K,
Miyase T, Warashina T, Aoshima H, Tabata H, Matsubara K. Matrix
metalloproteinase-2 inhibitors from Clinopodium chinense var. parviflorum.
J Nat Prod 2009;72:1379–84.
DSCG

T. Murata et al. / Fitoterapia 91 (2013) 51–59
59
[19] Wada H, Kido T, Tanaka N, Murakami T, Saiki Y, Chen CM. Chemical and
chemotaxonomical studies of ferns. LXXXI. Characteristic lignans of
Blechnaceous ferns. Chem Pharm Bull 1992;40:2099–101.
[23] Zhang ZF, Peng ZG, Gao L, Dong B, Li JR, Li ZY, Chen HS. Three new
derivatives of anti-HIV-1 polyphenols isolated from Salvia yunnanensis.
J Asian Nat Prod Res 2008;10:391–6.
[20] Tachibana H, Sunada Y, Miyase T, Sano M, Maeda-Yamamoto M,
Yamada K. Identification of a methylated tea catechin as an inhibitor of
degranulation in human basophilic KU812 cells. Biosci Biotechnol
Biochem 2000;64:452–4.
[21] Koukoulitsa C, Karioti A, Bergonzi MC, Pescitelli G, Bari LD, Skaltsa H.
Polar constituents from the aerial parts of Origanum vulgare L. ssp.
hirtum growing wild in Greece. J Agric Food Chem 2006;54:5388–92.
[22] Yabuuchi T, Kusumi T. Phenylglycine methyl ester, a useful tool for
absolute configuration determination of various chiral carboxylic acids.
J Org Chem 2000;65:397–404.
[24] Aoshima H, Miyase T, Warashina T. Caffeic acid oligomers with
hyaluronidase inhibitory activity from Clinopodium gracile. Chem Pharm
Bull 2012;60:499–507.
[25] Lemière G, Gao M, Groot AD, Dommisse R, Lepoivre J, Pieters L, Buss VJ.
3′,4-Di-O-methylcedrusin: synthesis, resolution and absolute configu-
ration. Chem Soc Perkin Trans 1995;1:1775–9.
[26] Xiao Y, Zhang L, Gao S, Saechao S, Di P, Chen J, Chen W. The c4h,
tat, hppr and hppd genes prompted engineering of rosmarinic acid
biosynthetic pathway in Salvia miltiorrhiza hairy root cultures. PLoS
One 2011;6:e29713.