Design, synthesis, and evaluation, derivatives of the fat-accumulation inhibitor ternatin: Toward ternatin molecular probes

Article abstract of DOI:10.1016/j.tetlet.2014.06.036

Ternatin, a cyclic heptapeptide derived from mushroom, strongly inhibits fat accumulation in 3T3-L1 adipocytes. However, its mechanism of action remains unclear. In this Letter, we designed, synthesized, and evaluated its derivatives for use as molecular probes to isolate its target protein. Finally, we successfully established a pair of ternatin molecular probes.

Full text of DOI:10.1016/j.tetlet.2014.06.036

Tetrahedron Letters 55 (2014) 4445–4447
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Design, synthesis, and evaluation, derivatives of the
fat-accumulation inhibitor ternatin: toward ternatin
molecular probes
b
a
a,b,
Yoshinori Kawazoe a, , Yoko Tanaka , Sachikazu Omura , Daisuke Uemura
⇑
⇑
a Research Institute of Natural Drug-Leads, Kanagawa University, 2946 Tsuchiya, Hiratsuka, Kanagawa 259-1293, Japan
b Department of Chemistry, Faculty of Science, Kanagawa University, 2946 Tsuchiya, Hiratsuka, Kanagawa 259-1293, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Ternatin, a cyclic heptapeptide derived from mushroom, strongly inhibits fat accumulation in 3T3-L1
adipocytes. However, its mechanism of action remains unclear. In this Letter, we designed, synthesized,
and evaluated its derivatives for use as molecular probes to isolate its target protein. Finally, we
successfully established a pair of ternatin molecular probes.
Received 9 April 2014
Revised 4 June 2014
Accepted 9 June 2014
Available online 14 June 2014
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Ternatin
Cyclic peptide
Fat accumulation
Molecular probe
3T3-L1 cell
Obesity is one of the most common problems societies face
worldwide and is a strong risk factor for several lifestyle-related
diseases, such as hypertension, hyperlipidemia, and diabetes.
Although a few medicines for treating obesity have been approved
and brought to the market, no fundamental treatment for obesity
has yet been established.1
Ternatin (1, see Fig. 1), an unusual amino acid-containing cyclic
heptapeptide, was isolated from the mushroom Coriolus versicolor
as a strong fat-accumulation inhibitor.2 The natural product blocks
lipid accumulation in 3T3-L1 adipocytes with an EC50 value of
their activities. As a result, we successfully created chemical probes
for ternatin.
If we wish to use bioactive molecules as molecular probes, they
should contain a free reactive group such as an amino or a carboxyl
group to introduce functional molecules, including fluorescent
reagents or biotin. We decided to conjugate ternatin analogues to
functional molecules via N-hydroxysuccinimide. Thus, ternatin
molecular probes must contain a free amino group. To introduce
a free amino group to ternatin, we checked previous structure–
activity relationship studies. Ternatin was expected to have a rigid
structure due to three intramolecular hydrogen bonds: between
0.14 lg/mL. In addition, compound 1 decreased the blood sugar
level in KK-A(y) mice, which is an animal model of type 2 diabe-
tes.3 These results suggest that ternatin may be a potent drug for
treating obesity or diabetes. However, a developmental study has
not yet been initiated because the mechanism of action of ternatin
remains unclear. Thus, it is important to determine how this com-
pound works within cells. To this end, researchers usually try to
identify a binding protein or the intracellular localization of the
bioactive molecule by using its derivatives as molecular probes.
In this letter, we describe the design and synthesis of ternatin
derivatives for use as molecular probes. Furthermore, we evaluated
the carbonyl group of D
-allo-Ile1 and the hydroxyl group of
Me-L-ALa2
N
D
-allo-Ile1
NMe-L
-Leu3
Me
N
O
N
HN
H
O
β-OH-
D
-Leu7
HO
Me
Me HN
O
-Leu4
O
L
O
H
HN
N
N
O
Me
O
⇑
-Ala6
Corresponding authors. Tel.: +81 463 59 4111x2525; fax: +81 463 58 9684
Me-L-Ala5
NMe-
D
N
(Y.K.); tel.: +81 463 59 4111x2719; fax: +81 463 58 9684 (D.U.).
E-mail addresses: ss199164bn@kanagawa-u.ac.jp (Y. Kawazoe), uemurad@
kanagawa-u.ac.jp (D. Uemura).
Figure 1. Chemical structure of ternatin (1).
http://dx.doi.org/10.1016/j.tetlet.2014.06.036
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.

4446
Y. Kawazoe et al. / Tetrahedron Letters 55 (2014) 4445–4447
b-OH-
amino group of
and the amino group of b-OH-
D
-Leu7, between the carbonyl group of b-OH-
-Leu4, and between the carbonyl group of
-Leu.7 Analogues that could not
form these hydrogen bonds such as [NMe-
-Leu4]ternatin showed
much lower activity than that of ternatin.4 In addition, the hydro-
xyl group in b-OH-
-Leu7 was shown to be important for activity.5
On the other hand, NMe-
-Ala6 could be replaced by other amino
D
-Leu7 and the
did not inhibit fat accumulation at all.7 Thus, the stereoisomer of
the molecular probe can be used as a negative control. Finally,
we designed four ternatin derivatives: an amine (2) and one with
L
L
-Leu4,
D
L
NMe-D
-Lys6 (4) as positive controls and a stereoisomeric amine
(3) and a stereoisomeric one with NMe-L
-Lys6 (5) as negative
controls (Fig. 2).
D
D
The synthesis of ternatin has been established,8 and we fol-
lowed this previous study to synthesize our molecular probes.
Detailed procedure is given in Supplementary material. Briefly,
building blocks are condensed to afford a right fragment and a left
fragment, which are then coupled, cyclized, and deprotected to
give molecular probes (Scheme 1 only shows positive controls.
For the negative controls, we used stereoisomers). All of the
peptide synthesis was performed in a liquid phase.
acids.6 These results suggest that position 6 is suitable for the
introduction of an amino group-containing linker. Based on this
idea, we designed two kinds of ternatin derivatives. The first type
incorporates an amino acid, lysine, at position 6 because it contains
a free amino group in its side chain. However, we were afraid that a
methylation reaction of the amino group might simultaneously
methylate the side-chain amino group. To account for this possibil-
ity, we used another approach that involved the reduction of azido
derivatives to give free amino group-containing analogues.
For the amine derivatives (2 and 3), we used
D- or L-homoserine
(Hse: 6 for -isomer) as a starting material. The hydroxyl group in
D
Stereoisomers in which
were replaced by
D
-allo-Ile1, NMe-
D
L
-Ala5, and NMe-
D
-Ala6
the side chain of Hse was replaced by an azido group by NaN3
(Scheme 2). The resulting azido derivatives were used for peptide
synthesis. Resulting azido-containing ternatin derivatives (7 and
8) were subjected to Staudinger reduction to give free amino
group-containing molecules9 2 and 3, which were obtained in
respective yields of 2.3% and 2.3%, in a total of 14 steps.
D
-Ile1, NMe-
-Ala5, and NMe- -Ala6, respectively,
L
D
-allo-Ile 1
D
-Ile 1
For the lysine derivatives, we used Boc-
D-Lys(Cbz)-OH (9) or
Me
N
O
N
Me
N
O
Boc- -Lys(Cbz)-OH as a starting material. We expected that the ter-
L
minal Boc-protected amino groups were preferentially methylated
because of higher acidity. However, Cbz-protected side-chain
amino groups were also simultaneously converted to give Boc-
HN
HN
N
H
O
H
O
O
O
Me
Me HN
Me
Me HN
O
O
O
O
HO
HO
H
H
HN
HN
N
N
NMe-
D-Lys(NMe)(Cbz)-OH
or
Boc-NMe-
L-Lys(NMe)(Cbz)-OH,
N
N
respectively (Scheme 3 shows
D-isomer, positive control.), in all
O
Me
O
O
Me
O
of the conditions we tested. At this moment, we decided to change
synthetic target molecules 4 and 5 to N-methylated-Lys side chain-
containing derivatives, 10 and 11. Although the side chain was
converted to a secondary amino group, it could be coupled with
NHS-functional molecules. After cyclization, catalytic hydrogena-
tion with Pd/C was performed to remove a Cbz group, however,
the reaction would not take place. Next, we applied catalytic trans-
fer hydrogenation with HCO2NH4. The reaction proceeded, albeit
slowly and in low yield (13% and 13% for the positive and negative
controls, respectively). We prepared the lysine-derived ternatin
molecular probes 10 and 11 in respective yields of 0.74% and
0.79%, respectively, in a total of 10 steps.
Finally, we evaluated the fat-accumulation inhibitory effects of
our molecular probes. 3T3-L1 cells were induced to undergo adi-
pose differentiation by insulin in the presence of various concen-
trations of compounds 1, 2, 3, 10, and 11. After one week of
induction of adipogenesis,10 cellular triglyceride levels were mea-
sured using a LabAssay™ Triglyceride kit (WAKO) according to
NH2 N-Me-D-Ala 5
NH2
N-Me-L
-Ala 5
Amine derivative6
Amine derivative6
3
2
Me
O
Me
N
O
N
HN
H
N
HN
H
N
O
Me
Me HN
O
O
Me
Me HN
O
O
N
O
O
HO
H
O
HO
H
HN
HN
N
N
N
O
Me
O
O
Me
O
NH2
NH2
N-Me-L-Lys 6
N-Me-
D
-Lys 6
5
4
Figure 2. Design of molecular probes for ternatin.
Me
N
O
NH2
BocHN
N
MeO
H
Me
N
O
O
Me
O
HN
O
HN
N
MeO
H
O
O
Me
Me HN
O
N
O
O
HO
H
Right Fragment
HN
N
O
O
Me
O
Me
O
O
HO
H
NBoc
Me
N
NBoc
NH2
H
R
EtO
H
N
EtO
2: R=(CH2)2NH2
4: R=(CH2)4NH2
7: R=(CH2)2N3
O
O
OH
Me
OH
R
R
10: R=(CH2)4NH(Me)
Left Fragment
Scheme 1. Plan for the synthesis of molecular probes for ternatin.

Y. Kawazoe et al. / Tetrahedron Letters 55 (2014) 4445–4447
4447
N3
OH
NH2
OH
1) SOCl2 / MeOH
1) DPPA, DBU / DMF
MeO
MeO
HO
NHBoc
NHBoc
2) Boc2O, DIPEA / DMF
2) NaN3 / DMF
O
O
O
2 steps
61 (89) %
6
2 steps
70 (75) %
N3
NaH, CH3I
dry THF
HO
NBoc
Me
O
79 (70) %
Scheme 2. Conversion of HSE to azido derivative. The
D-isomer for compound 2 is shown. A yield for L-isomer is indicated in parentheses.
is almost same as that in the highly active derivative, [NMe-D-
(Cbz)HN
HO
Aoc6]ternatin. In addition, compound 2 should be much more reac-
tive than compound 10 because of the free amino group in the side
chain at position 6. The reactivity of molecules usually affects their
kinetics, including binding to serum proteins, cell permeability,
and intracellular distribution. Among these, penetration of the cell
membrane is highly associated with free polar functional groups.
Thus, the combination of length and polarity of the side chain at
position 6 may have resulted in the difference in activity between
compounds 2 and 10.
NBoc
Me
(Cbz)HN
O
HO
NHBoc
(Cbz)N
Me HO
NaH, CH3I
dry THF
O
NBoc
Me
9
O
Scheme 3. Methylation of Lys derivative.
Although both positive controls 2 and 10 were each 15–30-fold
less active than the original natural product 1, there was a big
difference in activity between the positive controls and negative
controls (Fig. 3). This difference encouraged us to use them as
molecular probes for ternatin. We have since been able to
introduce a fluorescein isothiocyanate molecule to the probes.
Furthermore, we conjugated the probes to resin. Experiments to
demonstrate the intracellular localization of ternatin and to
identify a ternatin-binding protein are now in progress.
100
50
0
11
3
2
Acknowledgments
10
This study was supported in part by Grant-in-Aid for Scientific
Research from Japan Society for the Promotion of Science
(25242069 to D.U. and 24510316 to Y.K.). We are indebted to the
Uehara Memorial Foundation to the financial support.
Ternatin (1)
0
5
10
Concentration (μg/mL)
Supplementary data
Figure 3. Evaluation of the activity of molecular probes. The cells were induced to
differentiate with each probe, 6, 7, 8, 9, or 1 (ternatin) for one week, then cellular
triglyceride levels were measured. Data are expressed as mean values and SEM of
three independent experiments.
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.
06.036.
References and notes
the manufacture’s instructions. Consistent with a previous study,
ternatin (1) strongly inhibited fat accumulation with an EC50 value
1. Campfield, L. A.; Smith, F. J.; Burn, P. Science 1998, 280, 1383.
2. Shimokawa, K.; Mashima, A.; Asai, A.; Yamada, K.; Kita, M.; Uemura, D.
Tetrahedron Lett. 2006, 47, 4445–4446.
3. Kobayashi, K.; Kawashima, H.; Takemori, K.; Ito, H.; Murai, A.; Masuda, S.;
Yamada, K.; Uemura, D.; Horio, F. Biochem. Biophys. Res. Commun. 2012, 427,
299–304.
4. Shimokawa, K.; Miwa, R.; Yamada, K.; Uemura, D. Org. Biomol. Chem. 2009, 7,
777–784.
5. Shimokawa, K.; Iwase, Y.; Yamada, K.; Uemura, D. Org. Biomol. Chem. 2008, 6,
58–60.
6. Shimokawa, K.; Iwase, Y.; Miwa, R.; Yamada, K.; Uemura, D. J. Med. Chem. 2008,
51, 5912–5914.
7. Shimokawa, K.; Mashima, I.; Asai, A.; Ohno, T.; Yamada, K.; Kita, M.; Uemura, D.
Chem. Asian J. 2008, 3, 438–446.
8. Shimokawa, K.; Yamada, K.; Kita, M.; Uemura, D. Bioorg. Med. Chem. Lett. 2007,
17, 4447–4449.
9. Weiqing, X.; Derong, D.; Weiwei, Z.; Guangyu, L.; Dawei, M. Angew. Chem., Int.
Ed. 2008, 47, 2844–2848.
of 0.16
product (0.14
effect (EC50 = 5.57 and 2.40
negative controls showed no activity (EC50 >10
l
g/mL, which was nearly the same as the value of natural
g/mL).2 Positive controls 2 and 10 had a moderate
g/mL, respectively). As expected, the
g/mL) (Fig. 3).
l
l
l
Positive control 10 showed slightly better activity than positive
control 2. This might be explained by the results of our previous
structure–activity relationship study, which suggested that a
longer side chain at position 6 was associated with better activity.
For example, the activity of [NMe-D
-Aoc6]ternatin was 1.7 times
greater than that of original ternatin. If we compare positive con-
trols 2 and 10, compound 10 has a side chain at position 6 that
is two carbon atoms longer than that on 2. Furthermore, when
we consider the terminal methyl group on amine, it seems that 3
carbon atoms are elongated in compound 10. The side chain length
10. Kawazoe, Y.; Tanaka, S.; Uesugi, M. Chem. Biol. 2004, 11, 907–913.