(-)-Ternatin, a highly N-methylated cyclic heptapeptide that inhibits fat accumulation: structure and synthesis

Article abstract of DOI:10.1016/j.tetlet.2006.04.073

A highly N-methylated cyclic heptapeptide, (-)-ternatin, was isolated from the mushroom Coriolus versicolor, which significantly suppressed fat accumulation against 3T3-L1 murine adipocytes (EC₅₀ = 0.14 μg/mL). Although ternatin was previously

Full text of DOI:10.1016/j.tetlet.2006.04.073

Tetrahedron Letters 47 (2006) 4445–4448
(ꢀ)-Ternatin, a highly N-methylated cyclic heptapeptide
that inhibits fat accumulation: structure and synthesis
Kenichiro Shimokawa,a Itsuka Mashima,a Akiko Asai,a Kaoru Yamada,a
Masaki Kitab and Daisuke Uemuraa,c,
*
aDepartment of Chemistry, Graduate School of Science, Nagoya University, Furo-cho, Chikusa, Nagoya 464-8602, Japan
bResearch Center for Materials Science, Nagoya University, Furo-cho, Chikusa, Nagoya 464-8602, Japan
cInstitute for Advanced Research, Nagoya University, Furo-cho, Chikusa, Nagoya 464-8602, Japan
Received 20 March 2006; revised 10 April 2006; accepted 11 April 2006
Available online 12 May 2006
Abstract—A highly N-methylated cyclic heptapeptide, (ꢀ)-ternatin, was isolated from the mushroom Coriolus versicolor, which sig-
nificantly suppressed fat accumulation against 3T3-L1 murine adipocytes (EC50 = 0.14 lg/mL). Although ternatin was previously
reported to be antibacterial or antimicrobial compound, its inhibitory effect on fat accumulation has been first shown. The structure
of (ꢀ)-ternatin was revised to be a cyclo [D-allo-Ile1-L-(NMe)Ala2-L-(NMe)Leu3-L-Leu4-L-(NMe)Ala5-D-(NMe)Ala6-(2R,3R)-3-
hydroxy-Leu7] by spectroscopic analysis and chemical synthesis.
Ó 2006 Elsevier Ltd. All rights reserved.
Obesity caused by a modern lifestyle, such as an exces-
sive intake of food and a lack of sleep and exercise, is
a serious problem worldwide.1 WHO has reported that
at least 300 million adults are clinically obese. Obesity
can often trigger serious lifestyle-related diseases such
as cancer, cardiovascular disease, hypertension, hyperli-
pemia, and diabetes. Recently, a variety of foods have
been highlighted as being low-calorie and may confer
some benefits to the body. From the point of view,
mushrooms may be one of these perfect candidates. In
our continuing search for bioactive metabolites from
mushrooms,2 we found that the highly N-methylated
cyclic heptapeptide (ꢀ)-ternatin (1), which were isolated
from the mushroom Coriolus versicolor, potently inhib-
ited fat accumulation against 3T3-L1 murine adipocytes
(Fig. 1). We report here the structure and synthesis of 1.
O
N
N
O
HN
N
H
O
H
O
O
N
HO
HN
HN
O
O
Ternatin (1)
Figure 1. Structure of ternatin (1).
rated by silica-gel and Sephadex LH-20 gel permeation
column chromatography followed by reversed-phase
HPLC to afford a cyclic heptapeptide, (ꢀ)-ternatin (1)
20
[11.5 mg; ½aꢁD ꢀ 23.5 (c 0.13, EtOH)]. Compound 1
The mushroom C. versicolor (5 kg) was homogenized
and extracted with 80% aqueous ethanol followed by
partition with EtOAc and H2O. The EtOAc layer po-
tently inhibited fat accumulation against 3T3-L1 murine
adipocytes, which was partitioned with aqueous metha-
nol and hexane. The aqueous methanol layer was sepa-
inhibited fat accumulation against 3T3-L1 murine adi-
pocytes that were differentiated from preadipocytes by
treatment with insulin (EC50 = 0.14 lg/mL).4–6 Mean-
while, cell viability was not greatly affected at the EC50
value.3 Thus, ternatin (1) was considered to be a poten-
tial anti-obesity drug.
The molecular formula of
1 was found to be
Keywords: Cyclic heptapeptide; Inhibitory effect on fat accumulation;
3T3-L1 murine adipocytes; Solid phase peptide synthesis.
C37H67N7O8 [(M+Na)+, m/z 760.4945, D ꢀ0.4 mmu]
by HRFABMS. Detailed analysis of 1H, 13C NMR,
HMBC, and COSY spectra in benzene-d6 gave the
*
Corresponding author. Tel./fax: +81 52 789 3654; e-mail: uemura@
chem3.chem.nagoya-u.ac.jp
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.04.073

4446
K. Shimokawa et al. / Tetrahedron Letters 47 (2006) 4445–4448
L-(NMe)Ala2
OH
OBn
Me
H
4 steps
a
OHC
EtO2C
EtO2C
D-allo-Ile1
Me
s-Bu
H
O
Ref. 14
N3
R
N
Me
N
(J = 9.4 Hz)
5 : R = N3
6 : R = NH2
3
4
NH
H
O
O
b
H
i-Bu
O
O
β-OH-D-Leu7
L-(NMe)Leu3
H
OBn
H
c
N
H
HO2C
Me
H
N
H
Me
O
O
NHR
7 : R = H
2 : R = Fmoc
i-Bu
Me
L-Leu4
H
N
N
d
Me
Me
H
Me
O
Scheme 1. Reagents and Conditions: (a) benzyl-2,2,2-trichloroacetim-
idate, TfOH, CH2Cl2, rt, 57% (recovered 43% of 4); (b) H2, Pd/C(en),
MeOH, rt, 70%; (c) 1 M NaOH aq, THF, rt, 89%; (d) Fmoc–ONSu,
9% Na2CO3 aq, DMF, 0 °C to rt, 89%.
D-(NMe)Ala6
L-(NMe)Ala5
Figure 2. Plausible stereostructures of ternatin (1) based on 2D NMR
spectra. Arrows indicate ROESY correlations. Possible intramolecular
bonds between OH or NH protons and carbonyl oxygen atoms are
shown by dotted lines.
O
N
planar structure of 1. Amino acid analysis using the
Marfey method7,8 revealed that 1 was composed of L-
H2N
HN
N
H
O
H
a
O
O
N
(NMe)Leu3, L-Leu4, and both L- and D-(NMe)Ala2,5,6
.
O
BnO
O
HN
A hydroxyl proton and two amide NH protons in the
L-Leu4 and b-OH-Leu7 moieties were observed in the
1H NMR spectrum both in CD3OD and in C6D6, which
suggested that (ꢀ)-ternatin (1) had a single conforma-
tion that was fixed by three intramolecular hydrogen
bonds in these solvents. Detailed analysis of ROESY
correlations and coupling constants elucidated a plausi-
ble stereostructure shown in Figure 2. The absolute ste-
reochemistry of (NMe)Ala2 was confirmed to be L. The
coupling constant of the two methine protons H-2 and
H-3 in b-OH-Leu7 was 9.4 Hz, which suggested an anti
arrangement. Thus, the absolute stereochemistry of b-
OH-Leu7 was confirmed to be 2R,3R. As for the
(NMe)Ala5,6 moiety, the stereoisomer L-(NMe)Ala5-D-
(NMe)Ala6 was considered as a plausible conformer
based on ROESY correlations. Likewise, the stereo-
chemistry of Ile1 was suggested to be D- or D-allo-form.
As a result, the structure of 1 was found to be identical
to that of ternatin, which was previously isolated as an
antibacterial or antimicrobial compound from the fun-
gus Didymocladium ternatum9 and Cladobotryum sp.10
The stereochemistry of Ile1 moiety of ternatin was previ-
ously described as D-form,9,11 but its X-ray crystal struc-
ture has suggested D-allo-form.9,12 Furthermore, the
sign of optical rotation of ternatin was not clearly men-
tioned.13 Thus, it was still ambiguous whether (ꢀ)-1 was
identical to authentic compound. To confirm the stereo-
chemistry and biological activities of 1, synthetic study
was carried out.
HN
NH
HOOC
O
8
9
O
N
HN
N
H
H
O
O
b
O
RO
O
HN
HN
N
N
O
O
10 : R = Bn
1 : R = H
c
Scheme 2. Reagents and Conditions: (a) solid-phase peptide synthesis
(Fmoc deprotection, 20% piperidine in DMF; coupling reaction,
HATU, DMF; cleavage from the resin, 0.5% TFA/CH2Cl2, rt), 70%;
(b) HATU, HOAt, i-Pr2NEt, CH2Cl2, 0 °C to rt, 73%; (c) H2, 10% Pd/
C, MeOH, rt, 81%.
loaded with L-Leu4 [H-L-Leu-Trt(2-Cl) resin (8)], incor-
poration was elaborated in the [C] to [N] direction by
sequential deprotection with 20% piperidine followed
by HATU-mediated coupling with synthesized leucine
derivative 2 and the corresponding Fmoc amino acids.
Deprotection of the final Fmoc group and cleavage from
the resin under a mild acidic condition gave linear hep-
tapeptide 9 in 70% yield. Macrolactamization of 9 was
performed using HATU, HOAt and i-Pr2NEt in dilute
solution (0.77 mM).16 Finally, removal of the Bn group
in lactam 10 afforded cyclic heptapeptide 1.17
First, Fmoc-b-OBn-D-Leu-OH (2) was prepared
(Scheme 1). Isobutyraldehyde (3) was converted to the
chiral secondary alcohol 4 as described in the litera-
ture.14 The hydroxyl group in 4 was protected with a
Bn group followed by the Pd/C(en)-mediated selective
hydrogenation15 of azide to afford amine 6. Hydrolysis
of the ester and Fmoc-protection of the amino group
gave 2.
The 1H NMR data of synthetic ternatin was identical to
that of natural (ꢀ)-1. The optical rotation of 1 was
ꢀ35.5° (c 0.40, EtOH), which agreed with that of natu-
ral compound. Synthetic ternatin (1) potently inhibited
the adipogenesis in 3T3-L1 cells (Fig. 3). Two stereoiso-
mers cyclo [D-Ile1-L-(NMe)Ala2-L-(NMe)Leu3-L-Leu4-
L-(NMe)Ala5-D-(NMe)Ala6-(2R,3R)-3-hydroxy-Leu7]-
and cyclo [D-Ile1-L-(NMe)Ala2-L-(NMe)Leu3-L-Leu4-D-
A cyclic peptide 1 was then synthesized in the solid
phase (Scheme 2). Starting from Trt(2-Cl) resin pre-

K. Shimokawa et al. / Tetrahedron Letters 47 (2006) 4445–4448
4447
G. D.; Wang, Y. X. Nat. Med. 2004, 10, 355; (e)
Kopelman, P. G. Nature 2000, 404, 635.
2. Watanabe, R.; Kita, M.; Uemura, D. Tetrahedron Lett.
2002, 43, 6501.
3. (a) Green, H.; Kehinde, O. Cell 1973, 1, 113; (b) Green,
H.; Meuth, M. Cell 1974, 3, 127.
4. The preadipocyte cell line 3T3-L1, purchased from
Human Science Research Resources Bank, Japan Health
Sciences Foundation (Osaka, Japan), was cultured in
DMEM with 10% FCS in two 96-well plates at 37 °C, 5%
CO2 for 4–7 days (see Ref. 5). After the cells reached 100%
confluence, the culture buffer was changed to a differen-
tiation buffer and the samples dissolved in MeOH or water
were added. Differentiation buffer was composed of
DMEM containing 10% FCS, 1 lM dexamethasone, 0.5
mM IBMX, 90 U/mL penicillin, 90 lg/mL streptomycin,
and 10 lg/mL insulin (see Ref. 6). After 7 days, the
differentiated 3T3-L1 adipocytes in a 96-well plate were
treated with 2% Triton-X 100 for 30 min at room
temperature followed by sonication for 1 min. Fat
accumulation was determined by measuring liberated
triglyceride using a Triglyceride E-test Kit (Wako). To
determine the cell viability of differentiated 3T3-L1
adipocytes, another 96-well plate was treated with a Cell
Counting Kit-8 Test (Wako), and the absorbance at 450
Figure 3. Effects of synthetic ternatin (1) on the fat accumulation of
3T3-L1 murine adipocytes. a and c, Differentiated 3T3-L1 murine
adipocytes induced by insulin or rosiglitazone (incubated for 11 days).
The accumulated fat was stained with oil red O. b and d, incubated
3T3-L1 adipocytes as with a and c in the presence of 1 (1 lg/ml).
(NMe)Ala5-L-(NMe)Ala6-(2R,3R)-3-hydroxy-Leu7] were
also synthesized. However, these spectroscopic
data were not identical to that of natural 1. Thus, the
structure of 1 was confirmed to be a cyclo [D-allo-
Ile1-L-(NMe)Ala2-L-(NMe)Leu3-L-Leu4-L-(NMe)Ala5-D-
(NMe)Ala6-(2R,3R)-3-hydroxy-Leu7] as shown in
Figure 1., being identical to the reported X-ray crystal
structure. It is reasonable to suppose that the previous
assignment of D-Ile1 moiety in ternatin was incorrect.
nm was measured. Values are means
SD for duplicate
samples. (ꢀ)-Noradrenaline bitartrate hydrate used as a
standard was less effective (EC50 = 42 lg/mL).
5. (a) Green, H.; Kehinde, O. Cell 1976, 7, 105; (b) Sekiya,
K.; Okuda, H.; Hotta, Y.; Arichi, S. Phytotherapy Res.
1987, 1, 58; (c) Russell, T. R. Method Enzymol. 1981, 72,
720.
6. (a) Green, H.; Kehinde, O. Cell 1975, 5, 19; (b) Rubin, C.
S.; Hirsch, A.; Fung, C.; Rosen, O. M. J. Biol. Chem.
1978, 253, 7570.
7. Marfey, P. Carlsberg Res. Commun. 1984, 49, 591.
8. Dinitrophenyl alanyl amide (DAA) derivatives of b-OH-
Leu and D-allo-Ile were not clearly detected on HPLC
analysis due to other contaminants.
9. Miller, R.; Galitsky, N. M.; Duax, W. L.; Langs, D. A.;
Pletnev, V. Z.; Ivanov, V. T. Int. J. Pept. Protein Res.
1993, 42, 539.
10. Feng, Y.; Blunt, J. W.; Cole, A. L. J.; Cannon, J. F.;
Robinson, W. T.; Munro, M. H. G. J. Org. Chem. 2003,
68, 2002.
11. Dictionary of Natural Products, 3rd ed.; Chapman & Hall:
London, 1997; Vol. 10, p 362.
In summary, (ꢀ)-ternatin (1) was isolated from the
mushroom C. versicolor, which significantly inhibited
fat accumulation against 3T3-L1 murine adipocytes.
The structure of 1 was determined by spectroscopic
analysis and chemical synthesis. The fat accumulation-
inhibitory effect of 1 may be caused by the prevention
of adipogenesis and/or the activation of lipogenesis in
adipocytes. Further studies on the biological activities
of ternatin and its derivatives are in progress.
12. (a) Miller, R.; de Titta, G. T.; Langs, D. A.; Weeks, C. M.;
Hauptman, H. A.; Chang, C.-S.; Lee, C.-W.; Thuman, P.;
Jones, R. Science 1992, 259, 1430; (b) Langs, D. A. Acta
Crystallogr. D 1993, 49, 158.
13. Voronkino, T. M.; Koreneva, V. M.; Kursanova, G. V.;
Muchak, E. T.; Onoprienko, V. V.; Kolos’ov, M. N.
U.S.S.R. Patent N517 198, 1974.
14. Hale, K. J.; Manaviazar, S.; Delisser, M. Tetrahedron
1994, 50, 9181.
15. Sajiki, H.; Hattori, K.; Hirota, K. J. Org. Chem. 1998, 63,
7990.
Acknowledgements
We thank Drs. T. Tsuji and T. Ohno (FANCL Co.,
Ltd), Dr. K. Yamaguchi (Tokyo University of Marine
Science and Technology), and Drs. M. Ito and J. Ito
(Tsukuba Research Institute, Banyu Pharmaceutical
Co., Ltd) for their valuable discussions regarding the
fat-accumulation-inhibition assay. This study was sup-
ported in part by Grants-in-Aid for Creative Scientific
Research (16GS0206) from JSPS. We are indebted to
Ono Pharmaceutical Co., Ltd for their financial support.
16. (a) Davis, J. S. J. Peptide Sci. 2003, 9, 471; (b) Carpino, L.
A. J. Am. Chem. Soc. 1993, 115, 4397.
24:5
17. Spectroscopic data for synthetic ternatin (1): ½aꢁD ꢀ35.5
(c 0.40, EtOH); IR (CHCl3) 3426, 3344, 3313, 2960, 2933,
1
2871, 1636, 1526, 1506, 1466, 1409 cmꢀ1; H NMR (800
References and notes
MHz, C6D6)
d 0.57 (t, J = 7.4 Hz, 3H), 0.60 (d,
J = 6.9 Hz, 3H), 0.75 (d, J = 7.3 Hz, 3H), 0.79 (m, 1H),
0.90 (m, 1H), 0.90 (d, J = 6.9 Hz, 3H), 0.93 (d, J = 6.9 Hz,
3H), 0.97 (d, J = 6.4 Hz, 3H), 1.01 (ddd, J = 13.3, 9.4,
4.1 Hz, 1H), 1.18 (d, J = 6.9 Hz, 3H), 1.25 (d, J = 6.9 Hz,
3H), 1.25 (m, 1H), 1.27 (d, J = 7.3 Hz, 3H), 1.45 (d,
1. Recent selected reviews on obesity: (a) Campfield, L. A.;
Smith, F. J.; Burn, P. Science 1998, 280, 1383; (b)
Kolonin, M. G.; Saha, P. K.; Chan, L.; Pasqualini, R.;
Arap, R. Nat. Med. 2004, 10, 625; (c) Calle, E. E.; Kaaks,
R. Nat. Rev. Cancer 2004, 4, 579; (d) Evans, R. M.; Barish,

4448
K. Shimokawa et al. / Tetrahedron Letters 47 (2006) 4445–4448
J = 6.9 Hz, 3H), 1.54 (d, J = 6.4 Hz, 3H), 1.64 (m, 1H),
J = 6.4 Hz, 1H), 7.59 (d, J = 8.7 Hz, 1H), 7.95 (d,
J = 9.2 Hz, 1H); 13C NMR (201 MHz, C6D6) d 174.6,
174.5, 174.2, 172.7, 169.9, 168.8 (2C), 76.0, 59.2, 56.3,
55.1, 52.3, 51.3, 49.8 (2C), 40.5, 37.9, 33.7, 30.7, 30.2, 29.7,
29.4 (2C), 26.7, 26.0, 25.3, 23.9, 23.3, 22.6, 21.3 (2C), 15.9,
14.8, 14.2, 13.6, 13.1, 11.6; HRMS (FAB) calcd for
C37H67N7O8Na (M+Na)+ 760.4949, found 760.4922.
1.73–1.84 (m, 2H), 2.12 (m, 1H), 2.23 (m, 1H), 2.36 (m,
1H), 2.60 (s, 3H), 2.68 (s, 3H), 2.79 (s, 3H), 3.30 (s, 3H),
3.98 (dd, J = 9.4, 2.1 Hz, 1H), 4.31 (dd, J = 10.1, 4.0 Hz,
1H), 4.48 (dd, J = 6.9, 3.2 Hz, 1H), 5.17 (dd, 1H), 5.20 (td,
J = 8.0, 2.0 Hz, 1H), 5.66 (q, J = 6.9 Hz, 1H), 5.69 (q,
J = 6.9 Hz, 1H), 5.89 (d, J = 2.7 Hz, 1H), 6.23 (d,