New cyclic tetrapeptides from Nonomuraea sp. TA-0426 that inhibit glycine transporter type 1 (GlyT1)

Article abstract of DOI:10.1016/j.bmcl.2008.10.104

In the course of our search for natural antipsychotic agents, we isolated five new cyclic tetrapeptides from the fermentation broth of Nonomuraea sp. TA-0426. These compounds turned out to be analogues of WSS2220, which had been produced by the same actin

Full text of DOI:10.1016/j.bmcl.2008.10.104

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6321–6323
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
New cyclic tetrapeptides from Nonomuraea sp. TA-0426 that inhibit
glycine transporter type 1 (GlyT1)
b
b
a
a
Yuichi Terui a, , Chu Yi-wen , Li Jun-ying , Tsutomu Ando , Takuya Fukunaga ,
*
Takeshi Aoki a, Yoshihisa Toda a
a Medicinal Chemistry Laboratories, Taisho Pharmaceutical Co., Ltd, 1-403 Yoshino-cho, Kita-ku, Saitama-shi, Saitama 331-9530, Japan
b Sichuan Industrial Institute of Antibiotics, 9 Shabanqiao Road, Chengdu, Sichuan 610051, China
a r t i c l e i n f o
a b s t r a c t
Article history:
In the course of our search for natural antipsychotic agents, we isolated five new cyclic tetrapeptides from
the fermentation broth of Nonomuraea sp. TA-0426. These compounds turned out to be analogues of
WSS2220, which had been produced by the same actinomycete and showed strong inhibitory activity
against GlyT1. Four of the present peptides exhibit more potent GlyT1 inhibitory activities than
WSS2220.
Received 25 August 2008
Revised 17 October 2008
Accepted 24 October 2008
Available online 26 October 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
GlyT1
GlyT1 inhibitor
WSS2220
WSS2219
Cyclic tetrapeptide
Based on the glutamatergic hypothesis of schizophrenia, glycine
transporter type 1 (GlyT1) is a promising target for antipsychotic
agents.1 Several GlyT1 inhibitors have been reported, and some
of these agents are now being investigated in clinical trials.1 We re-
cently isolated a cyclic tetrapeptide, named WSS2220 (1), as a po-
tent and selective inhibitor of GlyT1.2 Compound 1 was produced
by Nonomuraea sp. TA-0426 that was isolated from a soil sample
collected at Henan Province, China. In our successive exploration
of GlyT1 inhibitors, we were able to isolate five new tetrapeptides,
2–6, from a fermentation broth of TA-0426. In this paper, we report
the isolation, structure determination and biological activity of
these new peptides. The preliminary structure–activity relation-
ship (SAR) based on the structures of the isolated compounds
and their synthetic analogues is also described.
CHCl3–MeOH 19:1 eluent from SCC was purified by reversed-phase
HPLC, eluting with CH3CN–H2O 65:35, to give WSS2218
(6, 15.2 mg, rt = 13.6 min) as a colorless powder (Fig. 1).
The n-BuOH extract (6.0 g) of TA-0426 was suspended in dis-
tilled water and extracted with EtOAc. The EtOAc extract was then
separated by silica gel column chromatography (SCC), eluting with
a CHCl3–MeOH solvent system. The cyclic peptide-containing
material, eluted with CHCl3–MeOH 9:1, was further purified by re-
versed-phase HPLC (column; YMC-Pack ODS-AM 250 ꢀ 10 mm; id,
5 lm; YMC, flow rate; 2.5 mL/min), eluting with CH3CN–H2O
45:55, to give WSS2219 (2, 13.4 mg; rt = 15.6 min), WSS2221
(3, 3.7 mg, rt = 11.4 min), WSS2222 (4, 5.7 mg, rt = 12.0 min) and
WSS2217 (5, 3.5 mg, rt = 9.7 min) as colorless powders. The
* Corresponding author. Tel.: +81 48 669 3107; fax: +81 48 652 7254.
E-mail address: yuichi.terui@po.rd.taisho.co.jp (Y. Terui).
Figure 1. Structures of cyclic tetrapeptides 1–6.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.104

6322
Y. Terui et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6321–6323
The HRTOF-MS of 2 showed a quasi-molecular ion peak at m/z
525.3044 [M+Na]+, corresponding to a quasi-molecular formula
of C27H42N4O5Na (Calcd for C27H42N4O5Na: 525.3053). IR absorp-
tions at 1680 and 1522 cmꢁ1 suggested the presence of amide
functions. Based on an analysis of 1D and 2D NMR spectra, the
presence of one phenylalanine (Phe) and two isoleucines (Ile)
was easily deduced. The structure of the remaining amino acid res-
idue was determined from 2D NMR spectral data (Fig. 2): the COSY
spectrum indicated the connectivities H-200/H-300/H-600 and H-400/
H-500. The chemical shifts of C-400 (d 67.4) and H-400 (d 3.88) showed
that C-400 was an oxymethine carbon. The HMBC correlations
between H-200/C-100 and H-600/C-400confirmed the connections of
C-100/C-200 and C-300/C-400. Thus, this amino acid residue was con-
firmed to be 4-hydroxyisoleucine (4-OH Ile). The HMBC correla-
tions H-200/C-400, H-500/C-300 and H-600/C-200 support the 4-OH Ile
fragment.
We encountered a problem in determining the amino acid
sequences; the four NH signals of 2 appear as an overlapped broad
hump at d 7.40–7.90 (in DMF-d7),3 which made it impossible to use
them for sequencing by HMBC spectroscopy. Therefore, N-methyl-
ation of 2 was performed. Compound 2 was treated with CH3I and
NaH in DMF at 0 °C4 to give pentamethylated compound 7 (30%
after HPLC separation).5 The strong N-Me signals of 7 showed in-
tense HMBC cross-peaks and the following HMBC correlations clar-
ified the amino acid sequences of 2: NCH3-2/C-2 and 1000, NCH3-20/
C-1 and 20, NCH3-200/C-10 and C-200 and NCH3-2000/C-100 and 2000
(Fig. 2).
Scheme 1. Methylation and acid hydrolysis of 2. Reagents and conditions: (a) CH3I,
NaH, DMF, 0 °C, 1 h; (b) 6 M HCl, 110 °C, 12 h. Structure of funebrine (9) is shown.
Table 1
1H NMR (500 MHz) and 13C NMR (125 MHz) data for 2 in DMF-d7.
Position
Ile-1
13C
1H
1
2
3
4
5
6
173.0
62.3
36.1
26.5
10.9
16.3
4.00 (dd, J = 11.0, 11.0)
1.94 (m)
1.20 (m), 1.60 (m)
0.89 (t, J = 7.1 Hz)
0.93 (t, J = 6.6 Hz)
7.40–7.90 (br)
The configurations of Phe and Ile were determined to be
and -Ile by hydrolysis (6 M HCl) followed by the advanced
Marfey’s method.6 At the same time, the stereochemistry at the
-position of 4-OH Ile was also elucidated: HPLC of the - and
-FDLA (1-fluoro-2,4-dinitrophenyl-5-leucineamide) derivatives
of 4-OH Ile in the hydrolysates gave peaks at 6.69 and 10.48 min,
respectively, which established that the -position of 4-OH Ile
L-Phe
L
a
L
2-NH
D
Ile-2
10
172.9
62.0
36.0
26.6
10.8
16.2
20
4.02 (dd, J = 11.0, 11.0 Hz)
1.89 (m)
1.20 (m), 1.60 (m)
0.87 (d, J = 7.1 Hz)
0.91 (d, J = 6.6 Hz)
7.40–7.90 (br)
30
a
had an S-configuration.7 When the hydrolysis mixture was purified
using a strong cation exchange (SCX) resin and SCC, cyclo-4-hydro-
xy isoleucine (8) was obtained in 18% yield (Scheme 1).8 The
coupling constants between H-2/H-3 and H-3/H-4 are 11.4 and
9.6 Hz, respectively, which indicates the anti/anti-configuration
of H-2/H-3/H-4. Thus, the absolute stereochemistry of 8 was deter-
mined to be 2S,3S,4R. (2S,3S,4R)-Cyclo-4-hydroxyisoleucine was
reported as a component of funebrine (9), which was isolated from
the plant Quararibea funebris.9 The coupling pattern of 8 was iden-
tical to the data in the literature.9b Thus, the structure of 2 was
determined to be as shown in Figure 1. The structures of com-
pounds 3–6 were determined essentially in the same manner as
described for 2 and by comparison of their NMR data with those
of 210 (Table 1).
40
50
60
20-NH
4-OH Ile
100
172.7
60.3
40.6
67.4
18.0
11.1
200
4.09 (dd, J = 11.2, 9.6 Hz)
2.16 (m)
3.88 (m)
1.01 (d, J = 6.3 Hz)
0.73 (d, J = 6.6 Hz)
7.40–7.90 (br)
300
400
500
600
200-NH
Phe
1000
173.1
58.3
37.7
138.7
130.1
129.4
127.6
2000
4.57 (dt, J = 7.7, 8.8 Hz)
3.14 (m)
3000
4000
5000, 9000
6000, 8000
7000
7.20–7.35*
7.20–7.35*
7.20–7.35*
7.40–7.90 (br)
To investigate the structure–activity relationship, we synthe-
sized H-Ile3-Phe-OH (linear tetrapeptide, 10), cyclo-(Ile4-Phe) (cyc-
lic pentapeptide, 13) and cyclo-(Ile5-Phe) (cyclic hexapeptide, 14)
(Scheme 2). Compounds 13 and 14 were prepared via cyclization
of the linear peptides, 11 and 12, respectively. The glycine-uptake
2000-NH
*
Not assigned precisely because of signal overlapping.
inhibitory activities and the GlyT1/GlyT2 selectivities of the iso-
lated and synthesized peptides were measured (Table 2) as previ-
ously reported.2a T98G cells (2 ꢀ 104 cells/well) were plated in a
96-well plate and incubated at 37 °C in a humidified incubator.
After 24 h, the cells were washed and incubated at room tempera-
ture with 250 nM [3H]-glycine and test sample. After 15 min, the
cells were washed and lysed with 0.5 M NaOH, a scintillation cock-
tail was added, and the cell lysates were counted using a scintilla-
tion counter.11 The GlyT1/GlyT2 selectivities were also measured
in the same way using COS7 cells transfected with rat GlyT1 or
GlyT2. Of the tested samples, 2 showed the most potent inhibitory
activity with high GlyT1 selectivity. The linear tetrapeptide (10)
Figure 2. COSY and HMBC data for 2 and 7.

Y. Terui et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6321–6323
6323
We have isolated five novel cyclic tetrapeptides with selective
GlyT1 inhibitory activities. A preliminary SAR study indicated that
a cyclic tetrapeptide structure was necessary for strong inhibitory
activity against GlyT, and the amino acid components affected the
selectivity against GlyT subtypes.
Acknowledgments
We thank Dr. Yuriko Nozawa for performing LC/MS analyses,
Dr. Atsushi Okada for performing NMR measurements, Mr. Haruaki
Yamamoto for fermenting TA-0426, and Dr. Akira Kawashima, Dr.
Toshiya Noguchi, Mr. Osamu Nozawa, and Mr. Hisashi Adachi for
their helpful suggestions.
References and notes
1. (a) Thomsen, C. Drug Discov. Today Ther. Strateg. 2006, 3, 539; (b) Depoortère,
R.; Dargazanli, G.; Estenne-Bouhtou, G.; Coste, A.; Lanneau, C.; Desvignes, C.;
Poncelet, M.; Heaulme, M.; Santucci, V.; Decobert, M.; Cudennec, A.; Voltz, C.;
Boulay, D.; Terranova, J. P.; Stemmelin, J.; Roger, P.; Marabout, B.; Sevrin, M.;
Vigé, X.; Biton, B.; Steinberg, R.; Françon, D.; Alonso, R.; Avenet, P.; Oury-Donat,
F.; Perrault, G.; Griebel, G.; George, P.; Soubrié, P.; Scatton, B.
Neuropsychopharmacology 2005, 30, 1963; (c) Lowe, J. A., III Expert Opin. Ther.
Patents 2005, 15, 1657.
2. (a) Terui, Y.; Chu, Y.; Li, J.; Nozawa, O.; Ando, T.; Fukunaga, T.; Aoki, T.; Toda, Y.;
Kawashima, A. Tetrahedron Lett. 2008, 49, 3067; (b) Toda, Y.; Chu, Y.; Li, J.;
Terui, Y., Fukunaga, T. WO2003104265.; (c) Chu, Y.; Li, J.; Toda, Y. CN1535980.
3. Use of DMSO-d6 resulted in poor separation of the signals.
4. Meyer, W. L.; Templeton, G. E.; Grable, C. I.; Jones, R.; Kuyper, L. F.; Lewis, R. B.;
Sigel, C. W.; Woodhead, S. H. J. Am. Chem. Soc. 1975, 97, 3802.
1
5. NMR data for 7: H NMR (500 MHz, pyridine-d5): d 0.24 (H-600), 0.95 (H-5),
0.95 (H-50), 1.03 (H-500), 1.16 (H-6), 1.40 (H-60), 2.15 (H-3), 2.71 (H-300), 2.81
(20-NCH3), 3.07 (2-NCH3), 3.18 (400-OCH3), 3.25 (200-NCH3), 3.39 (H-400), 3.56
(2000-NCH3), 4.37 (H-200), 4.62 (H-2), 5.30 (H-20), 5.34 (H-2000), 7.34 (H-5000, H-
9
000), 7.64 (H-6000, H-8000), 7.64 (H-7000). 13C NMR (125 MHz, pyridine-d5): d 9.4
(C-600), 10.7 (C-5), 10.9 (C-50), 15.8 (C-60), 15.9 (C-500), 16.9 (C-6), 24.3 (C-40),
28.8 (C-4), 30.5 (20-NCH3), 30.5 (200-NCH3), 31.2 (2-NCH3), 33.3 (2000-NCH3),
33.3 (C-30), 33.5 (C-3), 37.0 (C-3000), 37.7 (C-300), 56.0 (C-200), 56.3 (400-OCH3),
58.0 (C-20), 59.5 (C-2000), 70.3 (C-2), 76.1 (C-400), 127.7 (C-4000), 129.3 (C-6000, C-
8
000), 129.9 (C-5000, C-8000), 131.1 (C-7000), 169.0 (C-100), 169.5 (C-10), 173.1 (C-1),
Scheme 2. Synthesis of cyclic pentapeptide (13) and hexapeptide (14). Reagents
and conditions: (a) HATU, HOAt, i-Pr2NEt, DMF, 0 °C, 1 h. Structure of apicidin (15)
is shown.
173.6 (C-1000).
6. Fujii, K.; Ikai, Y.; Oka, H.; Suzuki, M.; Harada, K. Anal. Chem. 1997, 69, 5146.
7. The FDLA-derivatives were analyzed using a YMC-Pack Pro C18 (100 ꢀ 4.6 mm;
id, 3 lm; YMC) column at 40 °C. CH3CN–H2O containing 5 mM of ammonium
formate and 5 mM of formic acid was used as the mobile phase under a linear-
gradient elution mode (CH3CN, 15–85%, 20 min) at a flow rate of 1.0 mL/min
with UV detection at 215 nm.
Table 2
GlyT inhibition activities (nM).
8. NMR data for 8: 1H NMR (500 MHz, D2O): d 1.32 (3H, d, J = 6.4 Hz), 1.53 (3H, d,
J = 6.0 Hz), 2.42 (1H, qdd, J = 6.4, 9.6, 11.4 Hz), 4.22 (1H, d, J = 11.4 Hz), 4.50 (1H,
qd, J = 6.0, 9.6 Hz). 13C NMR (125 MHz, D2O): d 173.2 (s), 82.2 (d), 55.8 (d), 42.9
(d), 17.3 (q), 12.3 (q).
Compound
T98G
rGlyT1
rGlyT2
1
2
3
4
5
6
10
13
14
15
ALX5407
ALX1393
12.5
3.5
5.5
5.0
4.0
21.2
16,400
655
1680
71.8
4.8
20.4
3.0
2260
278
574
450
542
226
N.T.
N.T.
N.T.
87.7
N.T.
21.1
9. (a) Raffauf, R. F.; Zennie, T. M.; Onan, K. D.; Le Quesne, P. W. J. Org. Chem. 1984,
49, 2714; (b) Tamura, O.; Iyama, N.; Ishibashi, H. J. Org. Chem. 2004, 69, 1475.
10. 13C NMR data for 3 (125 MHz, DMSO-d6): d 9.8 (q; C-600), 10.1 (q; C-6), 15.4 (q;
C-5), 17.0 (q; C-500), 19.1 (q; C-40), 19.5 (q; C-50), 25.0 (t; C-4), 29.0 (d; C-30),
34.6 (d; C-3), 36.2 (t; C-3000), 39.0 (d; C-300), 56.5 (d; C-2000), 58.4 (d; C-200), 60.5
(d; C-2), 62.1 (d; C-20), 65.4 (d; C-400), 126.5 (d; Ar), 128.2 (d; Ar), 128.7 (d; Ar),
137.2 (s; Ar), 171.0 (s; C-100), 171.3 (s; C-20), 171.5 (s; C-1), 171.5 (s; C-1000). 13C
NMR data for 4 (125 MHz, DMSO-d6): d 9.9 (q; C-600), 10.0 (q; C-60), 15.3 (q; C-
50), 17.1 (q; C-500), 19.1 (q; C-4), 19.4 (q; C-5), 25.1 (t; C-40), 29.0 (d; C-3), 34.7
(d; C-30), 36.1 (t; C-3000), 39.2 (d; C-300), 56.4 (d; C-2000), 58.3 (d; C-200), 60.2 (d; C-
20), 62.3 (d; C-2), 65.5 (d; C-400), 126.5 (d; Ar), 128.2 (d; Ar), 128.7 (d; Ar), 137.2
(s; Ar), 171.0 (s; C-100), 171.3 (s; C-10), 171.4 (s; C-1), 171.5 (s; C-1000). 13C NMR
data for 5 (125 MHz, CD3OD): d 10.1 (q; C-600), 16.6 (q; C-500), 19.6 (q; C-40),
19.6 (q; C-50), 20.1 (q; C-4), 20.1 (q; C-5), 30.4 (d; C-30), 30.5 (d; C-3), 37.3 (t; C-
10.2
11.4
11.1
29.6
N.T.
N.T.
N.T.
94.6
3.0
N.T.
N.T.
ALX5407 (Tocris): selective inhibitor of GlyT1.
ALX1393 (Sigma–Aldrich): selective inhibitor of GlyT2.
3
000), 40.5 (d; C-300), 58.7 (d; C-2000), 60.3 (d; C-200), 64.3 (d; C-20), 64.7 (d; C-2),
67.9 (d; C-400), 127.8 (d, Ar), 129.5 (d; Ar), 130.2 (d; Ar), 138.3 (s, Ar), 174.5 (s;
C-100), 175.0 (s; C-10), 175.2 (s; C-1), 175.3 (s; C-1000). 13C NMR data for 6
(125 MHz, CD3OD): d 10.4 (q; C-600), 10.4 (q, C-60), 10.6 (q; C-6), 15.4 (q; C-600),
15.8 (q; C-60), 15.8 (q; C-6), 26.8 (t; C-400), 26.8 (t; C-40), 26.9 (t; C-4), 36.0 (d; C-
300), 36.2 (d; C-30), 36.3 (d; C-3), 37.4 (t; C-3000), 58.5 (d; C-2000), 62.3 (d; C-20),
62.4 (d; C-200), 62.7 (d; C-2), 127.8 (d; Ar), 129.4 (d; Ar), 130.1 (d; Ar), 138.4 (s; Ar),
175.1 (s; C-100), 175.2 (s; C-10), 175.3 (s; C-1000), 175.5 (s; C-1).
showed weak glycine-uptake inhibitory activity in T98G. The cyclic
pentapeptide (13) showed a moderate inhibitory activity, while
cyclic hexapeptide (14) showed only weak activity. Apicidin
(15),12 a cyclic tetrapeptide that was reported as a histone deace-
tylase (HDAC) inhibitor, showed moderate inhibitory activity, but
no selectivity was observed against GlyT subtypes. These results
indicate that the cyclic tetrapeptide structure is important for
strong inhibitory activity against GlyT, and GlyT1/GlyT2 selectivity
is affected by the amino acid components.
11. The test compound concentration producing 50% inhibition of glycine-uptake
(IC50 value) was determined by use of a nonlinear regression curve-fitting
program. Values are means of two experiments.
12. Darkin-Rattray, S. J.; Gurnett, A. M.; Myers, R. W.; Dulski, P. M.; Crumley, T. M.;
Allocco, J. J.; Cannova, C.; Meinke, P. T.; Colletti, S. L.; Bednarek, M. A.; Singh, S.
B.; Goetz, M. A.; Dombrowski, A. W.; Polishook, J. D.; Schmatz, D. M. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 13143.