Kurahamide, a cyclic depsipeptide analog of dolastatin 13 from a marine cyanobacterial assemblage of Lyngbya sp.

Article abstract of DOI:10.1246/bcsj.20140008

Kurahamide, a new dolastatin 13 analog, was isolated from a marine cyanobacterial assemblage, consisting mostly of Lyngbya sp. Its gross structure was elucidated by spectroscopic analysis, and the stereochemistries were assigned based on a chiral HPLC analysis of hydrolysis products. Kurahamide strongly inhibited elastase and chymotrypsin in vitro. In addition, kurahamide moderately inhibited the growth of human cancer cells, including HeLa and HL60 cells.

Full text of DOI:10.1246/bcsj.20140008

© 2014 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 87, No. 5, 609-613 (2014)
609
Selected Papers
Kurahamide, a Cyclic Depsipeptide Analog of Dolastatin 13
from a Marine Cyanobacterial Assemblage of Lyngbya sp.
Arihiro Iwasaki,1 Shinpei Sumimoto,2 Osamu Ohno,1 Shoichiro Suda,2 and Kiyotake Suenaga*1
1Department of Chemistry, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa 223-8522
2Faculty of Science, University of Ryukyus, 1 Senbaru, Nishihara, Okinawa 903-0213
Received January 11, 2014; E-mail: suenaga@chem.keio.ac.jp
Kurahamide, a new dolastatin 13 analog, was isolated from a marine cyanobacterial assemblage, consisting mostly
of Lyngbya sp. Its gross structure was elucidated by spectroscopic analysis, and the stereochemistries were assigned based
on a chiral HPLC analysis of hydrolysis products. Kurahamide strongly inhibited elastase and chymotrypsin in vitro.
In addition, kurahamide moderately inhibited the growth of human cancer cells, including HeLa and HL60 cells.
Marine microorganisms are considered to be good sources of
new biologically active substances. Marine cyanobacteria, in
particular, produce a variety of important secondary metabo-
lites.1 Several of these compounds, such as dolastatin 102 and
biselyngbyaside,3 have received attention due to their remark-
able physiological activities.4 In our continuing search for new
bioactive substances from marine cyanobacteria,5 we inves-
tigated the constituents of a marine cyanobacterial assemblage,
consisting mostly of Lyngbya sp., and isolated a cyclodepsi-
peptide, kurahamide (1, Figure 1). The cyclic moiety of 1 is
common to dolastatin 136 and its analogs. Several dolastatin 13
analogs have been isolated from marine cyanobacteria, includ-
ing lyngbyastatins 4-10,7 somamides A and B,8 symplostatin
2,9 and molassamide,10 all of which differ with respect to the
pendant side chain.
rial obtained from the aqueous MeOH portion was subjected
to fractionation with reversed-phase column chromatography
(ODS silica gel, MeOH-H2O) and reversed-phase HPLC
(Cosmosil 5C18-MS-II, MeOH-H2O; Cosmosil Cholester,
MeCN-H2O) to give kurahamide (1) (3.7 mg) as a colorless oil.
The molecular formula of kurahamide (1) was found to be
C55H77N9O15 by ESIMS (m/z 1126.5458, calcd for C55H77-
N9O15Na [M + Na]+ 1126.5437). The NMR data for 1 are
1
summarized in Table 1. The H NMR data suggested that 1 is
a depsipeptide, with several exchangeable protons (¤H ca. 6.5-
7.5) and ¡-protons (¤H ca. 4-5). Two low-field methine pro-
tons connected to the corresponding low-field shifted carbon
signals (¤H/¤C 5.13/73.6, 5.19/71.7), as shown in the HMQC
spectrum (Table 1), indicating methines adjacent to an ester
linkage. Further analysis of the 1H NMR, 13C NMR, COSY,
HMQC, and HMBC spectra recorded in CD3CN revealed the
presence of valine, N-methyltyrosine, phenylalanine, two
threonines and two alanines. In Thr-1, there was no visible
coupling between H-2 and H-3, as in other compounds with the
same cyclic core.7,9,10 HMBC and NOESY (Table 1) corre-
lations established the partial structure of Thr-1. Additionally,
the presence of residues derived from 3-amino-6-hydroxy-2-
piperidone (Ahp), 2-amino-2-butenoic acid (Abu), and butanoic
acid (Ba) were also established based on the detailed analysis of
the COSY and HMBC spectra.
The sequence of each residue for 1 was determined on the
basis of HMBC data and supported by NOESY data (Table 1).
Two HMBC correlations from H-2 of Val (¤ 4.08) and NH
(¤ 7.48) to C-1 of N-Me-Tyr (¤ 172.4) connected these two
residues. A strong HMBC correlation between the N-Me of N-
Me-Tyr (¤ 2.58) and C-1 of Phe (¤ 172.9) connected these two
units. Additionally, an HMBC correlation between H-2 of Phe
(¤ 4.63) and C-5 of Ahp (¤ 76.2) revealed the connectivity
between these two residues. Similarly, an HMBC correlation
from NH of Ahp (¤ 6.84) to C-1 of Abu (¤ 165.1) and NOESY
correlations observed at NH of Abu (¤ 7.49)/H-2 of Thr-1 (¤
Results and Discussion
The marine cyanobacterial samples11 were collected at
Kuraha, Okinawa, and were extracted with methanol. The
extract was filtered, concentrated, and partitioned between
EtOAc and H2O. The EtOAc-soluble material was further parti-
tioned between 90% aqueous MeOH and hexane. The mate-
OH
7
N-Me-Tyr
Val
Thr-1
4
Thr-2
Ala-1
4
O
O
Ba-1
3
O
O
4
H
1
O
N
N
H
1
N
N
1
1
1
4
Phe
7
H
H
1
O
O
N
O
HN
O
1
O
H
N
4
H
4
1
1
1
N
1
4
4
N
O
Abu
O
3
O
5
HO
Ba-2
Ala-2
Ahp
kurahamide (1)
Figure 1. Structure of kurahamide (1).
Published on the web February 21, 2014; doi:10.1246/bcsj.20140008

610
Bull. Chem. Soc. Jpn. Vol. 87, No. 5 (2014)
Kurahamide, a Cyclic Depsipeptide
Table 1. NMR Data for Kurahamide (1) in CD3CN
a)
b)
Unit
Val
Position ¤C
¤
H
(J in Hz)
COSY
3, NH
HMBC
NOESY
1
2
175.4
59.9
1, 3, 4, 5, 1
(N-Me-Tyr)
2, 4, 5
2, 3, 5
2, 3, 4
4.08, m
1.85, m
3, 4, 5, NH
3
4
5
NH
31.8
2, 4, 5
2, 4, 5, NH
2, 3, N-Me (N-Me-Tyr)
2, 3, NH, N-Me (N-Me-Tyr)
2, 3, 5, 2 (N-Me-Tyr), N-Me (N-Me-Tyr)
20.07c) 0.69, d (7.0)
3
3
2
19.3
0.62, d (7.0)
7.48, br
2, 1 (N-Me-Tyr)
N-Me-Tyr 1
172.4
63.2
34.4
2
3a
4.74, dd (11.5, 2.8) 3a, 3b
3.10, dd (14.6, 2.8) 2, 3b
2.51, dd (14.6, 11.5) 2, 3a
1, 3, 4, N-Me, 1 (Phe) 3a, 2 (Phe), 5/9, NH (Val)
2, 4, 5/9
2, 4, 5/9
2, 3b, 5/9
3a, 5/9
3b
4
5/9
6/8
7
N-Me
1
2
3a
3b
4
5/9
6/8
7
130.2
132.1
117.0
157.4
32.1
172.9
52.7
36.4
6.84, d (8.6)
6.61, d (8.6)
6/8
5/9
3, 4, 9/5, 6/8, 7
4, 8/6, 7
2, 3a, 3b, 6/8
5/9
2.58, s
2, 1 (Phe)
2 (Val), 4 (Val), 5 (Val), NH (Ahp), NH (Val)
Phe
4.63, dd (11.7, 4.4) 3a, 3b
2.68, dd (14.7, 11.7) 2, 3b
1, 3, 1 (Ahp), 5 (Ahp) 3b, 5/9, 2 (N-Me-Tyr), NH (Ahp)
2, 4, 5/9
3b, 5/9, 5 (Ahp)
2, 3a, 5/9
1.78, m
2, 3a
138.5
131.0
129.4
127.8
170.78d)
50.1
6.64, dd (8.2, 1.7) 6/8, 7
3, 7
4, 8/6
5/9
2, 3a, 3b, 5 (Ahp), 6/8, 4b (Ahp)
5/9
7.00, m
6.95, m
5/9, 7
5/9, 6/8
Ahp
1
2
3.64, ddd (2.4, 6.7,
12.7)
3a, 3b, NH
1
3b, NH
3a
3b
4a
4b
5
OH
NH
1
22.9
30.8
76.2
2.08, m
1.44, m
1.57, m
1.49, m
4.99, br s
4.06, m
6.84, d (6.7)
2, 3b, 4a, 4b
3a, 4a, 4b
3a, 3b, 4b, 5 2
3a, 3b, 4a, 5
4a, 4b, OH
3b, OH, NH
2, 3a, 4a
3b, 5, OH
5, 5/9 (Phe)
4a, 4b, OH, 3a (Phe), 5/9 (Phe)
3a, 4a, 5, NH (Abu)
2, 3a, 2 (Phe), NH (Abu), N-Me (N-Me-Tyr)
1, 5
2
5
2
1 (Abu)
Abu
165.1
129.4
136.9
14.9
2
3
4
5.66, qd (7.1, 0.8) 4, NH
1.79, dd (7.1, 0.7) 3, NH
1, 2
1, 2, 3
4
3
NH (Ahp), 2 (Thr-1), 3 (Thr-1), NH (Thr-1),
OH (Ahp)
NH
7.49, br s
3, 4
1, 2, 3, 1 (Thr-1)
Thr-1
Thr-2
1
2
3
4
NH
1
170.81d)
57.5
73.6
4.27, d (9.4)
5.13, q (7.3)
20.10c) 1.05, d (7.3)
7.11, d (9.4)
171.3
NH
4
3
1, 3, 4
1, 4, 1 (Val)
2, 3
3, 4, NH, NH (Abu)
2, 4, NH (Abu)
2, 3, NH
2
2, 3, 1 (Thr-2)
2, 4, 2 (Thr-2), 3 (Thr-2), NH (Abu)
2
3
4
NH
1
2
3
NH
1
2
58.2
71.7
17.9
4.37, dd (8.0, 4.6) 3, NH
5.19, dq (4.6, 6.4) 2, 4
1, 3, 4, 1 (Ala-1)
4, 1 (Ala-2)
3, 4, NH, NH (Thr-1)
2, 4, NH, NH (Thr-1)
2, 3, NH
1.02, d (6.4)
7.04, d (8.0)
3
2
2, 3, 1 (Ala-1)
2, 3, 4, 3 (Ala-1), 2 (Ala-1)
Ala-1
Ala-2
174.8
50.5
18.1
4.17, dq (7.3, 7.0) 3, NH
1, 3
1, 2
1, 2
3, NH, NH (Thr-2)
2, NH, NH (Thr-2)
2, 3, 2 (Ba)
1.09, d (7.0)
6.71, d (7.3)
2
2
173.8
49.8
18.1
4.07, m
3, NH
1, 3
1, 2
2, 3, 1 (Ba-2)
3, NH
2, NH
2, 3, 2 (Ba)
3
NH
1.08, d (7.5)
6.59, m
2
2
Continued on next page:

A. Iwasaki et al.
Continued:
Bull. Chem. Soc. Jpn. Vol. 87, No. 5 (2014)
611
a)
b)
Unit
Position ¤C
¤
H
(J in Hz)
COSY
HMBC
NOESY
Ba-1
1
2
3
4
1
2
3
4
174.4
39.01e) 1.91, m
3
2, 4
3
1, 3, 4
1, 2, 4
2, 3
3, NH (Ala)
2, 4
3
20.26f) 1.35, tq (7.7, 7.7)
14.47g) 0.68, t (7.7)
174.4
Ba-2
38.77e) 1.91, m
3
2, 4
3
1, 3, 4
1, 2, 4
2, 3
3, NH (Ala)
2, 4
3
20.36f) 1.35, tq (7.7, 7.7)
14.55g) 0.67, t (7.7)
a) Measured at 100 MHz. b) Measured at 400 MHz. c), d), e), f), g) These carbon signals are interchangeable.
OH
2.08 ppmH
N-Me-Tyr
7
4.06 ppm
Val
Thr-1
Thr-2
Ala-1
OH
4
O
H
Ba-1
O
1
O
3
H
O
1
N
1
2
1
3
N
H
2
1
N
N
2
5
H
H
4.99 ppm,
br s
3
H
1
O
N
O
Phe
O
H
N
O
1
HN
O
4
O
NOESY
O
H
N
H
N
4
2
H
COSY
HMBC
NOESY
1
4
1
1
N
1
O
2
1
N
O
O
HO
5
Ba-2
Abu
Ala-2
H
coupling constant
Ahp
2.4 Hz
12.7 Hz
H
Figure 2. Structure of kurahamide (1), based on 2D NMR
analyses.
3.64 ppm
Figure 3. Selected NOESY correlations and coupling
constants showing the relative stereochemistry of the
Ahp residue in kurahamide (1).
4.27) and NH of Abu (¤ 7.49)/H-3 of Thr-1 (¤ 5.13) expanded
the sequence to Val-N-Me-Tyr-Phe-Ahp-Abu-Thr-1. Addi-
tionally, HMBC correlations from NH of Thr-1 (¤ 7.11) to C-1
of Thr-2 (¤ 171.3), from H-2 of Thr-2 (¤ 4.37) to C-1 of Ala-1 (¤
174.8), and from H-3 of Thr-2 (¤ 5.19) to C-1 of Ala-2 (¤ 173.8)
revealed the linear sequence for all nine amino acids. Further-
more, an HMBC correlation between H-3 of Thr-1 (¤ 5.13) and
C-1 of Val (¤ 175.4) indicated the presence of an ester bond
between these amino acids. The low-field chemical shift of H-3
of Thr-1 also supported the presence of an ester bond at this
position and thereby established the cyclic partial structure of 1.
Finally, a NOESY correlation between NH of Ala-1 (¤ 6.71) and
H-2 of Ba-1 (¤ 1.91) and an HMBC correlation between NH of
Ala-2 (¤ 6.59) and C-1 of Ba-2 (¤ 174.4) allowed us to deter-
mine the location of two butanoic acid moieties, thereby com-
pleting the gross structure of kurahamide, as shown in Figure 2.
The absolute configurations of amino acids except for the
Ahp moiety were determined by Marfey’s analysis13 and a
chiral HPLC analysis of the acid hydrolysates of 1. These ana-
lyses revealed that all of the amino acids had L-configurations.
With regard to the Ahp moiety, the absolute configuration was
determined by chiral HPLC analysis of the acid hydrolysates of
a PCC oxidation product. Oxidation followed by acid hydrol-
ysis liberated glutamic acid, the configuration of which was
assigned to be L. In the Ahp ring system, the magnitude of the
coupling constant (3JH2-H3a = 12.7 Hz, determined by a decou-
pling experiment) suggested that H-2 and H-3a are located in
axial positions. Additionally, a NOESY correlation between H-
3a (¤ 2.08) and 5-OH (¤ 4.06), and the observation of two small
coupling constants for H-5 of Ahp (¤ 4.99, br s) revealed that
5-OH is also located in an axial position of the piperidone ring.
Thus, the absolute configuration of the Ahp unit was deter-
mined to be 2S, 5R (Figure 3). A four-bond HMBC correlation
between H-4 (¤ 1.79) and C-1 (¤ 165.1) of Abu indicated a “w”
Table 2. IC50 Values of Kurahamide for Serine Protease
Activities In Vitro
IC50/¯M
Compounds
Elastase
Chymotrypsin
Trypsin
Kurahamide (1)
PMSF
0.10
860
9.0
200
>100
1100
configuration for bonds between these atoms and, therefore,
a Z geometry for the double bond.10,14 The absolute stereo-
chemistry of kurahamide was determined as shown in 1.
Because analogs of dolastatin 13 were reported to exhibit
serine protease-inhibitory activity,7a,7b,10 1 was evaluated for
this activity. The elastase, chymotrypsin, and trypsin activities
were tested in vitro. As shown in Table 2, 1 showed strong
inhibitory activity against elastase and chymotrypsin, with IC50
values of 0.10 and 9.0 ¯M, respectively. The inhibition activity
of 1 was much stronger than that of phenylmethylsulfonyl fluo-
ride (PMSF) which was used as a positive control. Meanwhile,
there was no apparent inhibition of trypsin at 100 ¯M, the
highest concentration tested. This is a similar selectivity profile
to that previously reported for other dolastatin 13 analogs.
According to a previous paper, the adjacent residue on the N-
terminal side of Ahp unit should bind in the S1 specificity
pocket of trypsin,15 and a basic residue (L-Arg or L-Lys) at this
position in the inhibitor is necessary for trypsin inhibition at
concentrations under 10 ¯M.16 In the case of kurahamide (1), the
present residue is Abu, and this fact can explain why 1 is not an
efficient trypsin inhibitor. Meanwhile, the inhibitory activities
of 1 against elastase and chymotrypsin can come from the
hydrophobicity of the Abu unit.16,17 To evaluate the growth-

612
Bull. Chem. Soc. Jpn. Vol. 87, No. 5 (2014)
Kurahamide, a Cyclic Depsipeptide
inhibitory activities against cancer cells by 1, MTT assays with
HeLa cells and HL60 cells were used. The cells were treated in
96-well plates with various concentrations of the compounds for
72 h. The data from these assays revealed that 1 inhibited the
growth of both HeLa cells and HL60 cells, with IC50 values of
16 and 2.5 ¯M, respectively.
component, except for a mixture of Ala and Thr [Conditions for
HPLC separation: column, Cosmosil 5C18-PAQ (º20 © 250
mm); flow rate, 5.0 mL min¹1; detection at 215 nm; solvent
H2O. Retention times (min) of components: Ala and Thr (tR =
10.5 min), Val (tR = 12.6 min), N-Me-Tyr (tR = 31.2 min), Phe
(tR = 42.4 min)].
In conclusion, kurahamide (1), a novel analog of dolastatin
13, was isolated from a marine cyanobacterial assemblage, con-
sisting mostly of Lyngbya sp. The structure of 1 was established
by spectroscopic and chiral HPLC analysis of acid hydrolysates.
Kurahamide (1) exhibited strong protease-inhibitory activity
against elastase and chymotrypsin, with IC50 values of 0.10 and
9.0 ¯M, respectively. Compound 1 was also shown to inhibit
the growth of HeLa cells and HL60 cells. Several dolastatin 13
analogs have been reported as protease inhibitors, and their
modes of action have recently been reported.17
Each fraction was dissolved in H2O (50 ¯L) and analyzed by
chiral HPLC, and the retention times were compared to those of
authentic standards [DAICEL CHIRALPAK (MA+) (º4.6 ©
50 mm); flow rate, 1.0 mL min¹1; detection at 254 nm; solvent
2.0 mM CuSO4, 2.0 mM CuSO4-MeOH (95:5) and 2.0 mM
CuSO4-MeCN (90:10)]. With 2.0 mM CuSO4, the retention
times (tR min) for authentic standards were D-Val (3.8) and
L-Val (6.1). With 2.0 mM CuSO4-MeOH (95:5), the retention
times for authentic standards were N-Me-D-Tyr (10.8) and
N-Me-L-Tyr (13.2). With 2.0 mM CuSO4-MeCN (90:10), the
retention times for authentic standards were D-Phe (5.9) and L-
Phe (8.6). The retention times (min) (and the respective HPLC
conditions) of the amino acids in the hydrolysate were 6.1
(100:0), 13.2 (95:5), and 8.6 (90:10), indicating the presence of
L-Val, N-Me-L-Tyr, and L-Phe in the hydrolysate.
The Ala- and Thr-containing fraction was dissolved in H2O
(100 ¯L). A 1.0% 1-fluoro-2,4-dinitrophenyl-5-L-alaninamide
(Marfey’s reagent) solution in acetone (200 ¯L) and 50 ¯L of
1 M NaHCO3 were added, and the mixture was heated at 80 °C
for 3 min. The solution was cooled to room temperature, neutral-
ized with 1 M HCl, and evaporated to dryness. The residue
was resuspended in 50 ¯L of MeCN-H2O (1:1), and the solu-
tion was analyzed by reversed-phase HPLC. [Cosmosil 5C18-
AR-II (º4.6 © 250 mm); flow rate 1.0 mL min¹1; detection, UV
340 nm; solvent 0.02 M NaOAc-MeOH (45:55)]. The retention
times (tR min) of the derivatized amino acids in the hydrolysate
matched those of L-Thr (6.4) and L-Ala (9.1), but not L-allo-Thr
(6.6), D-allo-Thr (11.1), D-Thr (18.2) or D-Ala (23.6).
Experimental
General Experimental Procedures. Chemicals and sol-
vents were the best grade available and were used as received
from commercial sources. All NMR spectral data were recorded
1
on a JEOL JNM-ECX400 spectrometer for H (400 MHz) and
13C (100 MHz). 1H NMR chemical shifts (referenced to residual
CHD2CN observed at ¤ 1.93) were assigned using a combina-
tion of data from COSY and HMQC experiments. Similarly,
13C NMR chemical shifts (referenced to CD3CN observed at
¤ 118.2) were assigned based on HMBC and HMQC experi-
ments. ESI mass spectra were obtained on an LCT premier EX
spectrometer (Waters). Optical rotations were measured with a
JASCO DIP-1000 polarimeter. IR spectra were recorded on a
JASCO RT/IR-4200 instrument.
Collection, Extraction, and Isolation. The marine cyano-
bacterial samples were collected at Kuraha, Okinawa Prefecture,
Japan, at a depth of 0-1 m in March 2013. The collected cyano-
bacteria (2.3 kg) were extracted with methanol (4 L) for one
week. The extract was filtered, and the filtrate was concentrated.
The residue was partitioned between ethyl acetate (3 © 0.4 L)
and water (0.4 L). The material obtained from the organic
layer was partitioned between 90% aqueous methanol (0.4 L)
and hexane (3 © 0.4 L). The aqueous methanol fraction (1.3 g)
was first separated by column chromatography on ODS (13 g)
eluted with 40% methanol, 60% methanol, 80% methanol,
and methanol. The fraction (313 mg) eluted with 60% methanol
was subjected to HPLC [Cosmosil 5C18-MS-II (º20 © 250
mm); flow rate 5 mL min¹1; detection, UV 215 nm; solvent
65% MeOH] in six batches to give a fraction that contained
kurahamide (7.5 mg, tR = 33.0 min). This fraction was further
separated by HPLC [Cosmosil Cholester (º20 © 250 mm); flow
rate 5 mL min¹1; detection, UV 215 nm; solvent 40% MeCN] to
give kurahamide (1) (3.7 mg, tR = 40.0 min).
Oxidation, Acid Hydrolysis and Chiral HPLC Analysis.
Kurahamide (1) (0.5 mg) was dissolved in CH2Cl2 (1.0 mL),
and PCC (2.0 mg) was added. The reaction mixture was
allowed to stand for 12 h at room temperature, after which it
was washed with H2O. The CH2Cl2 portion was dried under
N2. The PCC-oxidized kurahamide was heated with 9 M HCl
(100 ¯L) for 23 h at 110 °C. The hydrolyzed product was
evaporated to dryness and could be separated to give a Glu-
containing fraction [Conditions for HPLC separation: column,
¹1
Cosmosil 5C18-PAQ (º20 © 250 mm); flow rate, 5.0 mL min
;
detection at 215 nm; solvent TFA-H2O (0.1:99.9). Reten-
tion time (tR min): Glu (10.9)]. The Glu-containing-fraction
was dissolved in H2O (50 ¯L) and analyzed by chiral HPLC,
and the retention time was compared to those of authentic
standards [DAICEL CHIRALPAK (MA+) (º4.6 © 50 mm);
flow rate, 1.0 mL min¹1; detection at 254 nm; solvent 2.0 mM
CuSO4]. The retention times (tR min) for authentic standards
were D-Glu (11.3) and L-Glu (16.4). The retention time (min)
of Glu in the hydrolysate was 16.4, indicating the presence of
L-Glu in the hydrolysate.
Gene Sequencing. A cyanobacterial filament was isolated
under a microscope. Genomic DNA was extracted using the
DNeasy Plant Mini kit (Qiagen) following the manufacturer’s
specifications. The 16S rRNA genes were PCR-amplified from
isolated DNA using the primer set 16S 27F, a universal primer,
28
Kurahamide (1): colorless oil; ½ꢀꢀD ¹28.4 (c 0.2, CH3-
OH); IR (film): -max 3288, 2965, 1735, 1653, 1636, 1534,
1
1446, 1382, 1202 cm¹1; H NMR, 13C NMR, COSY, HMBC,
and NOESY data, see Table 1; HRESIMS m/z 1126.5458
[M + Na]+ (calcd for C55H77N9O15Na, 1126.5437).
Acid Hydrolysis, Chiral HPLC Analysis, and Marfey’s
Analysis.
Kurahamide (1) (0.5 mg) was treated with 9 M
HCl (100 ¯L) for 23 h at 110 °C. The hydrolyzed product
was evaporated to dryness and could be separated into each

A. Iwasaki et al.
Bull. Chem. Soc. Jpn. Vol. 87, No. 5 (2014)
613
and 23S 30R, a cyanobacterial-specific primer. The PCR reac-
tion contained DNA derived from a cyanobacterial filament,
12.5 ¯L of GoTaq (Promega), 1.0 ¯L of each primer (10 pM),
and 10.5 ¯L of H2O for a total volume of 25 ¯L. The PCR
reaction was performed as follows: initial denaturation for
10 min at 95 °C, amplification by 35 cycles of 1 min at 94 °C,
1 min at 60 °C, and 1 min at 72 °C, and final elongation for
7 min at 72 °C. PCR products were analyzed on agarose gel
(1%) in TAE buffer and visualized by ethidium bromide
staining. The obtained DNA was sequenced with 16S 27F, 16S
1494R, and 23S 30R primers. The 16S rRNA gene sequence
is available in the DDBJ/EMBL/GenBank databases under
accession number AB857842.
Supporting Information
1H NMR, 13C NMR, NOESY, HMQC, and HMBC spectra in
CD3CN for kurahamide (1). HPLC chromatograms for deter-
mination of absolute configurations. This material is available
free of charge on the web at http://www.csj.jp/journals/bcsj/.
References
1
J. K. Nunnery, E. Mevers, W. H. Gerwick, Curr. Opin.
Biotechnol. 2010, 21, 787.
2
H. Luesch, R. E. Moore, V. J. Paul, S. L. Mooberry, T. H.
Corbett, J. Nat. Prod. 2001, 64, 907.
3
T. Teruya, H. Sasaki, K. Kitamura, T. Nakayama, K.
Suenaga, Org. Lett. 2009, 11, 2421.
M. Costa, J. Costa-Rodrigues, M. H. Fernandes, P. Barros,
V. Vasconcelos, R. Martins, Mar. Drugs 2012, 10, 2181.
a) A. Iwasaki, T. Teruya, K. Suenaga, Tetrahedron Lett.
Protease Inhibition Assays.
Serine protease-inhibitory
activities were determined as previously reported with slight
modifications.7a Elastase inhibitory activity was assessed using
high-purity porcine pancreatic elastase (Sigma, E0258). The
assay buffer used was 50 mM Tris-HCl (pH 8.0). Assay buffer
(73.4 ¯L), elastase solution (75 ¯g mL¹1 in assay buffer, 5 ¯L),
and various concentrations of 1 (6.6 ¯L, dissolved in MeOH)
were preincubated for 15 min at room temperature in a 96-well
plate. After this time, 15 ¯L of substrate solution was added
(2 mM N-succinyl-Ala-Ala-Ala-p-nitroanilide in assay buffer)
to each well, and the reaction was followed by measuring
the absorbance at 405 nm for 30 min. Inhibitory activities
against chymotrypsin and trypsin were determined using α-
chymotrypsin from bovine pancreas (Sigma, C4129) and
trypsin from porcine pancreas (Sigma T0303), with 2 mM N-
succinyl-Gly-Gly-Phe-p-nitroanilide as a substrate solution for
chymotrypsin and 2 mM N¡-benzoyl-L-arginine 4-nitroanilide
hydrochloride for trypsin. The assay buffer was 50 mM Tris-
HCl, 100 mM NaCl, and 1 mM CaCl2 (pH 7.8). Assay buffer
4
5
2010, 51, 959. b) T. Teruya, H. Sasaki, H. Fukazawa, K. Suenaga,
Org. Lett. 2009, 11, 5062. c) M. Morita, O. Ohno, T. Teruya, T.
Yamori, T. Inuzuka, K. Suenaga, Tetrahedron 2012, 68, 5984.
6
G. R. Pettit, Y. Kamano, C. L. Herald, C. Dufresne, R. L.
Cerny, D. L. Herald, J. M. Schmidt, H. Kizu, J. Am. Chem. Soc.
1989, 111, 5015.
7
a) S. Matthew, C. Ross, J. R. Rocca, V. J. Paul, H. Luesch,
J. Nat. Prod. 2007, 70, 124. b) K. Taori, S. Matthew, J. R. Rocca,
V. J. Paul, H. Luesch, J. Nat. Prod. 2007, 70, 1593. c) J. C. Kwan,
K. Taori, V. J. Paul, H. Luesch, Mar. Drugs 2009, 7, 528.
8
L. M. Nogle, R. T. Williamson, W. H. Gerwick, J. Nat.
Prod. 2001, 64, 716.
9
G. G. Harrigan, H. Luesch, W. Y. Yoshida, R. E. Moore,
D. G. Nagle, V. J. Paul, J. Nat. Prod. 1999, 62, 655.
10 S. P. Gunasekera, M. W. Miller, J. C. Kwan, H. Luesch,
V. J. Paul, J. Nat. Prod. 2010, 73, 459.
¹1
(40 ¯L), enzyme solution (5 ¯L, 1 mg mL in assay buffer),
11 Most of the cyanobacterium was morphologically classified
into the genus Lyngbya because it was composed of short cells
with a thick sheath. From the phylogenetic tree inferred from
959 bp of 16S rRNA gene sequences (see Supporting Infor-
mation, S12) it was revealed that the present cyanobacterium
(Maeda 130904A, accession no. AB857842) forms a clade with
Oscillatoria miniata NAC8-50 (GU724208), and is closely related
to Trichodesmium spp.12 The identification described above was
carried out using the same cyanobacterium collected at the same
location in September 2013, and we confirmed that its extract
contained kurahamide (1). Despite our efforts, other minor cyano-
bacteria in the assemblage were not identified by 16S rRNA gene
analysis. A detailed classification will be reported later.
12 S. Suda, R. Moriya, S. Sumimoto, O. Ohno, K. Suenaga,
J. Mar. Sci. Tec.-TAIW. 2013, 21, 175.
13 P. Marfey, Carlsberg Res. Commun. 1984, 49, 591.
14 a) R. Araya-Maturana, T. Delgado-Castro, W. Cardona,
B. E. Weiss-Lopez, Curr. Org. Chem. 2001, 5, 253. b) R.
Araya-Maturana, H. Pessoa-Mahana, B. Weiss-Lopez, Nat. Prod.
Commun. 2008, 3, 445.
15 A. Y. Lee, T. A. Smitka, R. Bonjouklian, J. Clardy, Chem.
Biol. 1994, 1, 113.
16 R. G. Linington, D. J. Edwards, C. F. Shuman, K. L.
McPhail, T. Matainaho, W. H. Gerwick, J. Nat. Prod. 2008, 71, 22.
17 L. A. Salvador, K. Taori, J. S. Biggs, J. Jakoncic, D. A.
Ostrov, V. J. Paul, H. Luesch, J. Med. Chem. 2013, 56, 1276.
and various concentrations of 1 (5 ¯L, dissolved in MeOH)
were preincubated for 10 min at 37 °C, before 25 ¯L of sub-
strate solution was added. The reaction was followed by mea-
suring the absorbance at 405 nm for 30 min. PMSF (Nacalai
tesque, Japan) was used as a positive control. Enzyme activity
in each well was calculated based on the slope of the reaction
curve compared to that of the solvent control.
Cell Growth Analysis. HeLa cells were cultured at 37 °C
with 5% CO2 in DMEM (Nissui, Japan) supplemented with 10%
¹1
heat-inactivated FBS, 100 units/mL penicillin, 100 ¯g mL
¹1
¹1
streptomycin, 0.25 ¯g mL amphotericin, 300 ¯g mL L-glu-
tamine, and 2.25 mg mL NaHCO3. HL60 cells were cultured
¹1
at 37 °C with 5% CO2 in RPMI (Nissui, Japan) supplemented
with 10% heat-inactivated FBS, 100 units/mL penicillin,
¹1
¹1
100 ¯g mL streptomycin, 0.25 ¯g mL amphotericin, 300
¹1
¹1
¯g mL L-glutamine, and 2.25 mg mL NaHCO3. HeLa cells
were seeded at 2 © 104 cells/well in 96-well plates (Iwaki,
Japan) and cultured overnight. HL60 cells were seeded at
1 © 105 cells/well in 96-well plates. Various concentrations
of compounds were then added, and cells were incubated for
72 h. Cell proliferation was measured by the MTT assay.
This work was supported in part by Grants-in-Aid from the
Ministry of Education, Culture, Sports, Science and Technol-
ogy of Japan (Nos. 24310160 and 25750386).