Tellimagrandin I, HCV invasion inhibitor from Rosae Rugosae Flos

Article abstract of DOI:10.1016/j.bmcl.2010.01.084

By use of the model virus, expressing the HCV envelope proteins E1 and E2, bioassay guided separation of the MeOH extract from Rosa rugosa Thunb. disclosed tellimagrandin I (1) together with eugeniin (2) and casuarictin (3) as the potent HCV invasion inhibitors. Furthermore, structure-activity relationship analysis of some relative tannins including the synthesized analogs elucidated the partial structures crucial for potent activity of 1.

Full text of DOI:10.1016/j.bmcl.2010.01.084

Bioorganic & Medicinal Chemistry Letters 20 (2010) 1598–1600
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Tellimagrandin I, HCV invasion inhibitor from Rosae Rugosae Flos
Satoru Tamura a, Gang-Ming Yang a, Natsuko Yasueda a, Yoshiharu Matsuura b,
Yasumasa Komoda b, Nobutoshi Murakami a,
*
a Graduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamada-oka Suita, Osaka 565-0871, Japan
b Research Center for Emerging Infectious Diseases and Research Institute for Microbial Diseases, Osaka University, 3-1 Yamada-oka Suita, Osaka 565-0871, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
By use of the model virus, expressing the HCV envelope proteins E1 and E2, bioassay guided separation of
the MeOH extract from Rosa rugosa Thunb. disclosed tellimagrandin I (1) together with eugeniin (2) and
casuarictin (3) as the potent HCV invasion inhibitors. Furthermore, structure–activity relationship anal-
ysis of some relative tannins including the synthesized analogs elucidated the partial structures crucial
for potent activity of 1.
Received 1 December 2009
Revised 8 January 2010
Accepted 13 January 2010
Available online 21 January 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
HCV
Envelope protein E1, E2
Rosa rugosa Thunb.
Tellimagrandin I
Invasion inhibitor
Hepatitis C virus (HCV) is the most important causative agent of
posttransfusion and sporadic non-A, non-B hepatitis, infecting more
than 170 million people worldwide.1 HCV infection becomes chronic
in most cases and may eventually result in hepatitis, liver cirrhosis,
hepatic failure and hepatocellular carcinoma.2 However, the current
therapy based on combination of pegylated interferon with ribavirin
is far from ideal, it is effective in about 50% of patients, dependent on
the virus genotype, and has several serious side effects.3 Thus, it has
been urgently demanded to search for promising anti-HCV candi-
dates. However, no establishment for an efficient in vitro culture of
HCV prevented exploration for new anti-HCV seed principles by
cell-based biochemical assay for a long period. Therefore, virus de-
rived protease and RNA-polymerase from sub-genomic replicon sys-
tem were considered as the most prominent targets up to recently.4
Inthiscontext, several researchers triedtodevelopsurrogatemodels
to examine HCV entry. Recently, Matsuura et al. have established the
model virus (E1E2 virus) encoding two envelope glycoproteins
responsible for binding to receptors in virus invasion.5 In addition,
deficiency of either the E1- or E2-envelope protein was revealed to
result in significant reduction of invasion efficacy into hepatocytes.
Based on these biological outcomes, the envelope proteins, E1 and
E2, wereclarified to be essentialforinvasionof HCV. In spiteof estab-
lishment of the HCV model virus, search for anti-HCV principles,
interrupting the virus invasion process, have never been investi-
gated by use of the model E1E2 virus. This circumstance prompted
us to be engaged in exploring HCV invasion inhibitors from medici-
nal plants. Here, we deal with the HCV invasion inhibitor, tellima-
grandin I (1) along with its structural requirement for biological
potency of 1 from structure–activity relationship in the natural
and synthesized relatives.
The assay to search for HCV invasion inhibitory principles was
carried out by use of the E1E2 virus expressing not only the enve-
lope proteins E1 and E2 but secretory alkaline phosphatase
(SEAP).6 After the virus invaded into the hepatocytes, SEAP gener-
ated in the host cells was secreted extracellularly. Thus, activity of
the secreted SEAP is recognized as indication of HCV invasion.
However, the samples, inhibiting directly SEAP and the process of
protein production, must be regarded as effective in the assay
using only the E1E2 virus. To eliminate the pseudo-positive sam-
ples, we also utilized the SEAP transformed G* virus possessing
the G envelope instead of the E1 and E2. Namely, the samples, giv-
ing rise to low SEAP activity by only infection of the E1E2 virus, are
considered to inhibit HCV invasion genuinely.
As a result of screening the extracts from about 400 medicinal
plants in combination with both viruses, the MeOH extract from
Rosae Rugosae Flos (flower buds of Rosa rugosa Thunb.) was dis-
closed as a promising candidate. The extract inhibit 77.2% of
E1E2 and 24.6% of G* invasion at the concentration of 10 lg/mL,
* Corresponding author at present address: Kyoto Pharmaceutical University,
Japan. Tel.: +81 75 595 4634; fax: +81 75 595 4768.
E-mail addresses: murakami@phs.osaka-u.ac.jp, murakami@poppy.kyoto-phu.
ac.jp (N. Murakami).
respectively. After this extract was successively partitioned be-
tween EtOAc and H2O, n-BuOH and H2O, the resulting EtOAc ex-
tract exhibited the strongest efficacy among the three extracts.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.01.084

S. Tamura et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1598–1600
1599
Figure 1. Inhibitory activity of tellimagrandin I (1) for invasion of HCV model virus.
Bioassay guide separation of the EtOAc extract by successive ODS
column chromatography and reversed phase HPLC led to isolation
of tellimagrandin I (1, 0.036% from the crude drug) as a responsible
principle.
Scheme 1. Preparation for relatives of tellimagrandin I without HHDP function.
Reagents and conditions: (a) HCl gas, nBuOH, 82%; (b) 8, EDCIÁHCl, DMAP, DMF, 80%
for 9; (c) Pd–C, H2, EtOAc–MeOH, two steps 78% for 5, 99% for 6.
The IR spectrum of 1 showed the absorption bands due to the hy-
droxyl (3189 cmÀ1) and ester carbonyl groups (1722, 1718 cmÀ1
)
along with the aromatic rings (1611 cmÀ1). The positive-ion FAB-
the signals due to one galloyl and two HHDP groups appeared in
the 1H and 13C NMR spectra of 3. Intensive analysis of their CD
and IR spectra, FAB-MS, HR FAB-MS as well as their 1H and 13C
NMR data revealed them to be eugeniin (2, 0.078%)10,11 and casu-
arictin (3, 0.069%),8,12 respectively (Fig. 2).
Since the disclosed active principles (1–3) contained the HHDP
and galloyl groups, participation of these functionalities in HCV
invasion inhibitory activity was examined. Thus, the hydrolysable
tannin, pedunculagin (4), was isolated from Juglans regia13 and
the two galloyl glucose analogs (5 and 6) were prepared (Fig. 2).
The syntheses of the two analogs (5 and 6) were conducted as de-
MS gave the pseudomolecular ion peak at m/z 787 [M+H]+, while
the FABHRMS revealed the molecular formula of 1 to be C34H26O22
.
The H–H correlation spectroscopy (COSY) spectrum indicated the
presence of glucose residue of which 2-, 3-, 4-, and 6-hydroxyl
groups were esterified. Furthermore, detailed analysis of the 1H,
13C NMR, and heteronuclear multiple quantum coherence (HMQC)
spectra clarified the presence of one hexahydroxydiphenyl (HHDP)
and two galloyl groups. In the CD spectrum, the negative split Cotton
ascribable to the diphenyl functionality was observed around
250 nm [
D
e
: À4.3 (265 nm), +10.1 (238 nm)], indicating S configura-
tion of the HHDP moiety.7 Taking these physicochemical properties
into account, the active principle was unambiguously identified as
tellimagrandin I (1) in comparison with the previously reported
spectroscopic data.8,9 Tellimagrandin I (1) inhibited invasion of the
picted in Scheme 1. Treatment of D-glucose with HCl gas in benzyl
alcohol afforded 1-O-benzyl ether 7 in 82% yield. Condensation of 7
with 3,4,5-O-tribenzylgallic acid (8) in the presence of 1-ethyl-3-
(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDCIÁHCl)
and 4-dimethylaminopyridine (DMAP) gave the corresponding es-
ter. The ester was submitted to debenzylation by using Pd–C under
E1E2 virus with IC50 of 1.7 lM in a concentration dependent manner
as illustrated in Figure 1. In addition, 1 showed approximate 10-fold
selectivity against the G* virus from the IC50 values together with a
little cytotoxicity, 15.6% of growth inhibition, against the host
a hydrogen atmosphere to provide 2,3,4,6-O-tetragalloyl-
(5)14 as a mixture of
- and b-anomer in a ratio of 2:1 in 78% yield
for two steps. On the other hand, coupling between -glucose and 8
D-glucose
a
D
HepG2 cells at the concentration of 10 lM.
by EDCIÁHCl and DMAP afforded pentaester 9 in 80% yield. Removal
Extensive search for active principles from the MeOH extract of
Rosa rugosa Thunb. afforded other two active principles showing
nearly similar potency to tellimagrandin I (1). Analysis of the 1H
and 13C NMR data indicated the two principles (2 and 3) to be
the congeners of 1. The 1H and 13C NMR spectra of 2 showed the
presence of three galloyl and one HHDP groups. On the other hand,
of the benzyl groups in 9 by Pd–C under a hydrogen atmosphere
furnished 1,2,3,4,6-O-pentagalloyl-b-D
-glucose (6)8,14 in 99% yield
(Scheme 1).
Inhibitory activity against HCV invasion of tellimagrandin I (1)
and its relatives is summarized in Table 1. In respective compari-
Figure 2. Tellimagrandin I and relatives evaluated for inhibition for HCV invasion.

1600
S. Tamura et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1598–1600
Table 1
Inhibitory activity for HCV invasion of tellimagrandin I (1) and relatives (2–6)
E1E2 virusa
G* virusa
10
l
M
3
lM
1
l
M
10
lM
3
lM
1 lM
Tellimagrandin I (1)
Eugeniin (2)
Casuarictin (3)
Pedunculagin (4)
Tetragalloylglucose (5)
Pentagalloylglucose (6)
93.8 6.7
92.8 1.3
87.4 6.3
81.1 4.3
43.7 8.3
72.3 11.1
74.3 8.2
51.3 8.3
49.6 2.3
58.8 4.4
20.8 2.1
28.9 6.3
28.3 9.8
18.4 7.3
12.5 3.7
14.6 2.8
9.1 2.2
45.6 4.3
35.7 4.1
45.1 4.8
30.9 4.1
51.2 9.3
40.7 1.8
25.4 8.3
21.2 4.2
28.0 10.2
14.8 5.8
40.7 10.3
22.8 4.5
6.4 6.0
17.2 8.4
15.5 7.0
9.4 2.2
16.0 3.1
5.0 4.7
4.2 4.7
a
All values are expressed as mean SD of three experiments.
3. Fried, M. W.; Shiffman, M. L.; Reddy, K. R.; Smith, C.; Marinos, G.; Goncales, F. L.,
Jr.; Haussinger, D.; Daigo, M.; Carosi, G.; Dhumeaux, D.; Craxi, A.; Lin, A.;
Hoffman, J.; Yu, J. N. Eng. J. Med. 2002, 347, 975.
4. Thompson, A. J. V.; McHutchison, J. G. Aliment. Pharmacol. Ther. 2009, 29, 689.
5. Matsuura, Y.; Tani, H.; Suzuki, K.; Kimura-Someya, T.; Suzuki, R.; Aizaki, H.;
Ishii, K.; Moriishi, K.; Robison, C. S.; Whitt, M. A.; Miyamura, T. Virology 2001,
286, 263.
6. The coding region of G protein gene in the full-length cDNA clone of Vesicular
stomatitis virus (VSV) genome was replaced with the coding region of the SEAP
gene, clipped from the plasmid, pNiFty2-56 K-SEAP (Nacalai). This plasmid was
son of inhibitory activity between 1 and 5, 2 and 6, the HHDP
groups located between C-4 and C-6 with S-configuration were
shown to enhance the activity. Respective comparison of inhibitory
potency between 1 and 4, 2 and 3, the galloyl groups on C-2 and C-
3 instead of the HHDP groups were revealed to be crucial for the
bioactivity. In the case of the galloyl esters (5 and 6) without the
HHDP group, the penta-O-galloyl ester 6 exhibited more potent
activity than the tetraester 5.
In conclusion, we disclosed tellimagrandin I (1) along with
eugeniin (2) and casuarictin (3) as the HCV invasion inhibitors
using the model virus, expressing the HCV envelope proteins E1
and E2, through bioassay guided separation of the MeOH extract
from the medicinal plant, Rosae Rugosae Flos. Additionally, struc-
ture–activity relationship analysis of the hydrolysable tannins
including the natural and synthesized relatives clarified that the
HHDP group bridged between C-4 and C-6 and the galloyl groups
on C-2 and C-3 enhanced inhibitory activity for HCV invasion. Up
to date, only iridoid monoterpenes, quinolylamines, and tricyclic
diphenylamines have been found out as the HCV invasion inhibi-
tors by using other model systems for HCV invasion.15–17 It should
be noteworthy for the active principles in the present investigation
to possess a considerably different structural feature from the
known inhibitors.
designated as pVSVDG*. The G* virus was prepared by transfection of pVSVDG*
and the plasmid of G protein into CHO cells according to Takada’s protocol.18
The E1E2 virus was provided by infection of the G* virus to CHO cells
expressing chimeric E1 and E2 proteins agreeably to Matsuura’s procedure.5
The resulting suspensions of both viruses were diluted to moderate titer:
inoculation of 10
lL of suspension in the following culture exhibited about 20-
fold SEAP activity.
In the 96-well microculture plates, HepG2 cells (9.0 Â 104 cells/mL) were
cultured in 90
l
L
of Dulbecco’s Modified Eagle medium (Nissui
Pharmaceutical) containing 10% fetal bovine serum (Wako) at 37 °C under a
5% CO2 atmosphere for 24 h. The test samples were dissolved in DMSO and
diluted to appropriate concentrations using the medium, then 10 lL of each
sample solution was inoculated. The final concentration of DMSO in the culture
is 1.0%. After incubation at 37 °C under a 5% CO2 atmosphere for 1 h, the E1E2
virus or G* virus suspension (10 lL) was inoculated. The cells were further
incubated for 24 h, then SEAP activities in supernatants were evaluated by use
of Great EscAPe SEAP Fluorescent Detection kit (BD Biosciences). Lactoferrin
was used as the positive control for the assay and showed about 45% inhibition
at the concentration of 30 lg/mL.
7. Okuda, T.; Yoshida, T.; Hatano, T.; Koga, T.; Toh, N.; Kuriyama, K. Tetrahedron
Lett. 1982, 23, 3937.
8. Lee, S.; Tanaka, T.; Nonaka, G.; Nishioka, I. Phytochemistry 1990, 29, 3621.
9. Feldman, K.; Ensel, S.; Minard, R. J. Am. Chem. Soc. 1994, 116, 1742.
10. Yoshida, T.; Haba, K.; Arata, R.; Nakato, F.; Singu, T.; Okuda, T. Chem. Pharm.
Bull. 1995, 43, 1101.
11. Nishio, T.; Omote, Y. J. Chem. Soc., Perkin Trans. I 1983, 1765.
12. Spencer, C. M.; Cai, Y.; Martin, R.; Lilley, T. H.; Haslam, E. J. Chem. Soc., Perkin
Trans. II 1990, 4, 651.
13. Fukuda, T.; Ito, H.; Yoshida, T. Phytochemistry 2003, 63, 795.
14. Karamali, K.; Kerstin, L. Tetrahedron 1997, 53, 10725.
15. Zang, H.; Rothwangl, K.; Mesecar, A. D.; Sabahi, A.; Rong, L.; Fong, H. H. S. J. Nat.
Prod. 2009, 72, 2158.
Acknowledgments
This work was supported in part by Grants-in-Aid for Scientific
Research from the Ministry of Education, Science, Culture and
Sports, and Research funds from San-Ei Gen F. F. I. Inc. The authors
are grateful to the Kampou Science Foundation for financial
support.
16. Takebe. Y.; Hakamada, W.; Uenishi, R. Jpn. Kokai Tokkyo Koho JP2009215280,
2009.
References and notes
17. Cuthbertson, T. J.; Ibamez, M.; Rijnbrand, C. A.; Jackson, A. J.; Mittapalli, G. K.;
Zhao, F.; MacDonald, J. E.; Wong-Staal, F. PCT Int. Appl. WO2008021745, 2008.
18. Takada, A.; Robison, C.; Goto, H.; Sanchez, A.; Murti, K. G.; Whitt, M. A.;
Kawaoka, Y. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 14764.
1. Lohmann, V.; Korner, F.; Koch, J.-O.; Herian, U.; Theilmann, L.; Bartenschlager,
R. Science 1999, 285, 110.
2. Shepard, C. W.; Finelli, L.; Alter, M. J. Lancet Infect. Dis. 2005, 5, 558.