New cyclic depsipeptides from the green alga Bryopsis species; application of a carboxypeptidase hydrolysis reaction to the structure determination

Article abstract of DOI:10.1016/j.tet.2005.10.079

New cyclic depsipeptides, kahalalides P (1) and Q (2), were isolated from the Hawaiian green alga Bryopsis sp. The sequential positions of the dl anti-podal amino acids were determined by a carboxypeptidase hydrolysis reaction. This enzymatic method will

Full text of DOI:10.1016/j.tet.2005.10.079

Tetrahedron 62 (2006) 1301–1308
New cyclic depsipeptides from the green alga Bryopsis species;
application of a carboxypeptidase hydrolysis reaction
to the structure determination
Andrey Dmitrenok,a,† Takashi Iwashita,a Terumi Nakajima,a,‡ Bryan Sakamoto,b,§
Michio Namikoshic and Hiroshi Nagaic,
*
aSuntory Institute for Bioorganic Research, 1-1-1, Wakayamadai, Shimamoto, Mishima, Osaka 618-8503, Japan
bDepartment of Chemistry, University of Hawaii at Manoa, 2545 The Mall, Honolulu, Hawaii, HI 96822, USA
cDepartment of Ocean Science, Tokyo University of Marine Science and Technology, 4-5-7, Konan, Minato-ku, Tokyo 108-8477, Japan
Received 15 September 2005; accepted 19 October 2005
Available online 23 November 2005
Abstract—New cyclic depsipeptides, kahalalides P (1) and Q (2), were isolated from the Hawaiian green alga Bryopsis sp. The sequential
positions of the DL anti-podal amino acids were determined by a carboxypeptidase hydrolysis reaction. This enzymatic method will be
applicable to the structure determination of other non-ribosomal peptides. The absolute chemistry of 3-hydroxy-9-methyldecanoic acid in
kahalalides P and Q were determined by the recently introduced convenient Mosher ester procedure.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
Kahalalides, cyclic, and acyclic peptides, have
been previously characterized from the sacoglossan mollusk
Elysia rufescens and its diet, the green alga Bryopsis sp.1–6
Kahalalide F exhibits selective activity against solid tumors
and is currently undergoing phase II clinical trials.7
Kahalalide A exhibits anti-mycobacterium tuberculosis
activity.8,9 Re-investigation of a Hawaiian alga Bryopsis
sp. extract led to the isolation of two new cyclic
depsipeptides kahalalide P (1) and kahalalide Q (2). We
report herein the determination of the absolute stereo-
chemistry of these compounds. A carboxypeptidase
hydrolysis reaction was utilized for determining the
sequential position of the anti-podal DL amino acids.
2.1. Isolation of kahalalides P and Q
Bryopsis sp. (1.0 kg wet wt) was collected in Hawaii. The
alga was lyophilized and stored at K30 8C. The freeze dried
alga was extracted with methanol. After drying, the extract
(31 g) was mixed with Celite powder and subjected to low-
pressure flash chromatography on an ODS column. The
MeOH/H2O 9:1 fraction was further separated by reverse
phase HPLC, which yield kahalalide P (1) [5.2 mg
(0.0005%)], kahalalide Q (2) [1.8 mg (0.0002%)] as well
as previously described kahalalides G and F.
2.2. Structures of kahalalides P and Q
2.2.1. Planar structure of kahalalide P. The molecular
formula of kahalalide
P (1) was established as
C66H99N11O17 on the basis of the HRFABMS data, m/z
1318.7319 [MCH]C(D C2.1 mmu). It was corroborated
by the 13C NMR spectrum, which displayed signals for
66 carbons. NMR experiments were performed in
DMSO-d6 with addition of 0.05% of TFA, as peak
broadening was observed in C5D5N and pure DMSO-d6
solvents. Detailed analysis of the 2D NMR data enabled
us to assign all the signals for kahalalide P and revealed
a structural framework consisting of peptidal and fatty
acid moieties. Examination of the 1H NMR spectra
Keywords: Marine natural product; Green alga; Depsipeptide;
Carboxypeptidase.
*
Corresponding author. Tel./fax: C81 3 5463 0454;
e-mail: nagai@s.kaiyodai.ac.jp
† Present address: Pacific Institute of Bioorganic Chemistry, pr. 100 let
Vladivostoka, 159, Vladivostok 690022, Russia.
‡ Present address: Hoshi University, 2-4-41 Ebara, Shinagawa, Tokyo
142-8501, Japan.
§ Present address: Department of Anesthesia, Indiana University School of
Medicine, Fesler Hall Rm#204, Indianapolis, IN 46202, USA.
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.10.079

1302
A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
suggested 1 was a peptide with aromatic and aliphatic
residues. The low field portion of the spectrum showed
eight doublets of amide proton signals at 7.4–8.8 ppm,
one broad peak (two protons from the NH2 group of
lysine) at 7.55 ppm and two sets of signals of protons of
monosubstituted aromatic moieties, each of the five
protons were between 7.0–7.4 ppm. Investigation of the
2D NMR data from TOCSY, DQF-COSY, HSQC, and
HMBC experiments led to identification of the corre-
sponding ten amino acid residues: Asp, Val, Lys, Leu, 4-
trans-hydroxy-Pro (Hyp), Pro, two Ser, and two Phe. The
Asp spin system could be traced by TOCSY cross peaks
between signals 7.47 ppm (NH), 4.54 ppm (Ha), 2.52 and
2.25 ppm (Hb); Val protons showed correlation in
TOCSY spectra between 7.81 ppm (NH), 4.64 ppm
(Ha), 1.98 ppm (Hb) and two methyl doublets at 0.93
and 0.88 ppm; Leu protons were connected from TOCSY
cross peaks of NH at 7.74 ppm, Ha at 4.54 ppm, two
methylene protons of Hb at 1.65 and 1.30 ppm, Hg at
1.47 ppm and two methyl doublets at 0.83 and 0.79 ppm.
TOCSY spectrum showed correlation for Ser1 protons at
8.75 ppm (NH), 5.54 ppm (Ha) and 3.58 (2Hb); and for
Ser2 at 8.13 ppm (NH), 4.24 ppm (Ha), 3.71 and
3.56 ppm (2Hb). Two spin systems, 8.32 ppm (NH),
4.79 ppm (Ha), 2.91 and 2.59 ppm (2Hb); and 8.67 ppm
(NH), 4.99 ppm (Ha) and 2.81 ppm (2Hb) were assigned
to Phe1 and Phe2 residues. Lys residue was traced by
TOCSY and COSY cross peaks between NH at
8.41 ppm, Ha at 4.52 ppm, two Hb protons at 1.68 and
1.38 ppm, two Hg at 1.20 ppm, two Hd at 1.50 ppm, two
H3 at 2.75 ppm and two NH2 protons at 7.55 ppm.
and 4.04 ppm,3 whereas related protons in the cyclic peptide
kahalalides E and K resonate at 5.11 and 5.16 ppm2,4).
D-Phe-1
D-Val
L-Ser-1
H
R
O
H
1 L-Hyp
2 L-Pro-1
N
N
D-Asp
HN
N
H
H
NH
HO
O
O
O
O
O
OH
OH
O
HN
1
L-Ser-2
HN
(3R)-9-Me-3-Decol
O
4
2
O
3
9
O
HN
O
O
H
D-Leu
N
N
1 L-Pro-1
2 L-Pro-2
O
L-Phe-2
L-Lys
NH2
1 R=OH
2 R=H
The sequence of the amino acids was established utilizing
NOESY and HMBC experiments and ESI QTOF MS/MS
analysis. Sequential HMBC correlations from the NH proton
to the neighboring carbonyls were seen between Asp NH/9-
Me-3-Decol CO, Leu NH/Ser2 CO, Val NH/Ser1 CO, and Lys
NH/Phe2 CO. Other HMBC cross peaks from the NH to the
carbonyls were inconclusive due to overlaps in the spectra.
Sequential NOESY correlations from the NH proton to the
neighboring a proton were seen between 8.67 ppm (Phe2 NH)
and4.536 ppm (LeuHa), 8.32 ppm (Phe1NH) and 4.542 ppm
(Asp Ha), 8.13 ppm (Ser2 NH) and 4.20 ppm (Hyp Ha),
8.75 ppm (Ser1 NH) and 4.79 ppm (Phe1 Ha), 8.41 ppm (Lys
NH) and 4.99 ppm (Phe2 Ha), 7.81 ppm (Val NH) and
5.54 ppm (Ser1 Ha), and 7.74 ppm (Leu NH) and 4.24 ppm
(Ser2 Ha). NOESY correlations were also observed between
Val Ha at 4.64 ppm and proton Hd of Hyp at 3.55 ppm, and
between Lys Ha at 4.52 ppm and proton Hd of Pro at
3.13 ppm. The NH proton of the Asp residue at 7.47 ppm
showed NOESY peaks with H-2 protons (2.33, 2.38 ppm) and
H-3 proton (5.34 ppm) of the 9-Me-3-Decol. However, no
correlations were observed between H-3 proton of this fatty
acid residue and the proline protons or carbons.
The signal of the oxymethyne proton at 4.32 ppm, which
correlated with the carbon at 68.2 ppm in the HSQC
spectrum was considered as Hg of the hydroxyproline
residue. TOCSY and COSY correlations led to assignment
of other hydroxyproline signals: two Hb protons at 2.11 and
1.88 ppm, Ha at 4.20 ppm and two Hd protons at 3.77 and
3.55 ppm. Relative stereochemistry of 4-hydroxy-Pro (Hyp)
was determined from the NOESY spectrum. The cross
peaks Hyp Ha (4.20)/Hyp Hba (2.11) and Hyp Hg (4.32)/
Hyp Hbb (1.88) appeared stronger than the cross peaks Hyp
Ha/Hyp Hbb and Hyp Hg/Hyp Hba, indicating that the
relative stereochemistry between Hyp Ha and Hg is trans.
This result was further supported by Marfey’s analysis. The
last amino acid, proline was revealed from TOCSY
correlations of Ha at 4.17 ppm, two Hb at 2.01 and
1.83 ppm, two Hg at 2.10 and 1.78 ppm and two Hd protons
at 3.56 and 3.13 ppm.
Peptide 1 was subjected to base hydrolysis, which yielded a
linear product, 3. Product 3 was analyzed by nanoelectro-
spray MS/MS measurement. MS/MS spectrum (Fig. 1),
b- and y-type ions as well as several prominent peaks of
internal ions clearly confirmed the sequence of amino acids
and fatty acid of the acyclic kahalalide P (3).
The presence of a 3-hydroxy-fatty acid residue in 1 was
indicated by analysis of the NMR and MS data. Sequential
COSY correlations were observed between the methylene
signals at 2.38, 2.33 ppm (H-2), H-3 at 5.34 ppm, two H-4
protons at 1.58 ppm and methylene signals at 1.24 ppm, and
between the signals of the two terminal methyl groups
(0.85 ppm, 6H), H-9 (1.48 ppm), two H-8 (1.13 ppm) and
methylenes at 1.24 ppm. Exact length of the fatty acid chain
was confirmed by HRFAB MS and QTOF MS/MS analysis.
Chemical shift of the oxymethyne proton H-3 (5.33 ppm)
indicated an acyloxy nature of this bond and cyclic structure
of the peptide (the chemical shifts of related protons of the
fatty acid residues in the acyclic kahalalides H and J are 3.95
2.2.2. Absolute stereochemistry of kahalalide P. The
absolute stereochemistry of amino acids in 1 was
determined by Marfey’s method,10 which showed Asp,
Val, and Leu to be D, and Ser, Hyp, Pro, and Lys to be L.
Both D- and L-Phe enantiomers were present in the peptide.
Enzymatic cleavage by carboxypeptidase to determine the
positions of D- and L-Phe in 3 was performed (Fig. 2).
Although peptides containing D-amino acid in the second
position are resistant to hydrolysis by endopeptidases,11–13
to the best of our knowledge, this property has never been
utilized in the structure elucidation of natural products. The
compound (1, 0.2 mg) was subjected to base hydrolysis to

A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
1303
Figure 1. Positive ESI QTOF MS/MS spectrum of the acyclic kahalalide P.
yield the linear peptide 3. After HPLC purifications, 3 was
subjected to enzymatic cleavage with carboxypeptidase P.
This enzyme non-specifically releases L-amino acids from
the carboxy terminal of proteins and peptides. The reaction
was monitored by MALDI TOF and QTOF MS and MS/MS
measurements. The reaction completely terminates after
two residues, Pro and Lys, are cleaved from the parent
peptide, thus indicating a -D-Leu-L-Phe-L-Lys-L-Pro–COOH
sequence at the C-terminus of the peptide. Sequencing of 3
with carboxypeptidase Y showed identical results as with
carboxypeptidase P. Furthermore, we synthesized two linear
model peptides with anti-podal positions of D- and L-Phe, 5a
(H2N-D-Asp-D-Phe-L-Ser-D-Val-L-4-trans-hydroxy-Pro-L-
Ser-D-Leu-L-Phe-L-Lys-L-Pro–COOH) and 5b (H2N-D-Asp-
L-Phe-L-Ser-D-Val-L-4-trans-hydroxy-Pro-L-Ser-D-Leu-D-
Phe-L-Lys-L-Pro–COOH), which were subjected to enzy-
matic sequencing in the same conditions as 3. Two
C-terminal residues, Pro and Lys, were cleaved off in the
reaction with 5a (mw of final product [MCH]C 927.27 Da,
[MCNa]C 949.29 Da) similar to the reaction observed
with 3. However, only one C-terminal (Pro) residue was
released in the reaction with anti-podal 5b (mw of final
CP
9-Me-3-Decol - D-Asp - D-Phe - L-Ser - D-Val - L-4-OH-Pro - L-Ser - D-Leu - L-Phe - L-Lys - L-Pro-COOH
3
9-Me-3-Decol - D-Asp - D-Phe - L-Ser - D-Val - L-4-OH-Pro - L-Ser - D-Leu - L-Phe-COOH
4
CP
H2N-D-Asp - D-Phe - L-Ser - D-Val - L-4-OH-Pro - L-Ser - D-Leu - L-Phe - L-Lys - L-Pro-COOH
5a
CP
H2N-D-Asp - L-Phe - L-Ser - D-Val - L-4-OH-Pro - L-Ser - D-Leu - D-Phe - L-Lys - L-Pro-COOH
5b
Figure 2. The carboxypeptidase (CP) hydrolysis reaction of acyclic kahalalide P (3) and synthesized model compounds.

1304
A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
product [MCH]C 1055.87 Da, [MCNa]C 1077.86 Da).
These results verify the proposed position of D- and L-Phe in
1. This enzymatic method will be applicable to the DL
determination of anti-podal amino acids in other non-
ribosomal peptides, including kahalalides E,2 J,3 and O5
(stereochemistry of these compounds has not been
elucidated).
products of the reactions were measured without purifica-
tion following the recently described convenient Mosher
ester procedure.16 We used model compounds, methyl
(3R)-3-hydroxytetradecanoate (7) and methyl (3RC3S)-3-
hydroxytetradecanoate (7C8, racemic body) to obtain
MTPA derivatives. (R)- and (S)-MTPA esters of these
compounds were obtained and the differences in proton
chemical shifts between (S)-MTPA esters and (R)-MTPA
esters were measured in C5D5N. MTPA esters of 7 have
OH
O
9
1
3
1
distinctive differences in H NMR (Fig. 3). Furthermore,
OH
6
proton spectrum of the MTPA ester of the methyl 3(RCS)-
hydroxytetradecanoate (7C8, racemic body) was almost the
same as the sum of the spectra of (R)-MTPA ester of 7 and
(S)-MTPA ester of 7.
OH
O
3
1
14
14
OCH3
OCH3
7
8
9
OH
O
3
Two milligram of 1 was subjected to acid hydrolysis and the
hydrolyzate was partitioned between CHCl3 and H2O. The
organic layer was evaporated and the free fatty acid
obtained was converted to methyl 3-hydroxy-9-methyl-
decanoate (9) with diazomethane. Products of the reaction
were divided into two parts, and then converted to MTPA
esters directly in NMR tubes as described above.16 While
the spectra of the MTPA derivatives of the fatty acid residue
contains signals from impurities, multiplets of two H-2
protons and H-3 proton are clearly visible (Fig. 3).
Difference in chemical shift of the protons in the (S)-
MTPA ester of methyl 3-hydroxy-9-methyldecanoate and
(R)-MTPA ester of methyl 3-hydroxy-9-methyldecanoate
were found to be almost identical to the MTPA esters of 7
(Fig. 3). Thus, the absolute configuration of 9-Me-3-Decol
from 1 was proved to be (3R) (6).
1
OH
O
9
3
1
OCH3
Although the 3-hydroxy-9-methyldecanoic acid (9-Me-3-
Decol) moiety in kahalalides E, J, H, and K, and 3-hydroxy-
7-methyloctanoic acid moiety in kahalalide D have been
reported, the absolute stereochemistries of these fragments
were not determined due to the paucity of sample
available.2–4 We applied the Mosher ester procedure14,15
to determine the absolute stereochemistry of 9-Me-3-Decol
(6) in 1. MTPA esters were obtained directly in NMR tubes
with deuterated pyridine. The 1H NMR spectra of the
1
Figure 3. Partial H NMR spectra of (R)-and (S)-MTPA esters of (3R)-methyl-3-hydroxytetradecanoate (7) and (3R)-methyl-3-hydroxy-9-methyldecanoate
(9). (A) (R)-MTPA ester of 9; (B) (R)-MTPA ester of 7; (C) (S)-MTPA ester of 9; (D) (S)-MTPA ester of 7. (A) and (C) were measured with 750 MHz
spectrometer. (B) and (D) were measured with 500 MHz spectrometer.

A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
1305
2.2.3. Structure of kalahalide Q. Kahalalide Q (2) showed
5.2. Isolation
[MCH]C ion 1302.7336 (C66H99N11O16, D K1.3 mmu)
from HR FAB MS data, which differ by one oxygen atom
from kahalalide P (1). 1H HMR showed similar data for all
moieties (Table 2) except for the signals of the hydroxy-
proline residue. Careful analysis of the NMR data showed 2
possessed a proline instead of a hydroxyproline residue.
This result was further supported by QTOF MS/MS analysis
of 2, and its linear analog obtained from 2 by base
hydrolysis. Determination of absolute stereochemistry of
amino acid residues of 2 by Marfey’s method showed D- and
L-Phe, D-Asp, D-Val, D-Leu, L-Lys, two L-Ser, and two L-Pro.
Since the NMR data of the peptides from 1 and 2 are almost
indistinguishable, the stereochemistry of the fatty acid residue
and positions of the D- and L-Phe in 2 are identical to 1.
Bryopsis sp. (1.0 kg wet wt) was collected at Kewalo Basin,
Oahu, Hawaii (1998). The alga was lyophilized and stored
at K30 8C. Freeze dried alga was extracted with methanol.
The combined extracts were evaporated to dryness to give
31 g of green powder. The residue was mixed with Celite
and purified by two-step ODS flash chromatography
(column 2!25 cm) (first elution with 20–50–70–90–100%
aqueous MeOH, second with 50–70–80–90–95–100%
aqueous MeOH). The MeOH/H2O 9:1 fraction was further
purified by repeated reverse phase HPLC, (Cosmosil 5C18-
MS-II, 20!250 mm with aqueous CH3CN gradient
80–100% and then Cosmosil 5C18-AR-II, 4.6!250 mm
with aqueous CH3CN gradient 60–70%) to yield kahalalide
P (5.2 mg, 0.0005%) and kahalalide Q (1.8 mg, 0.0002%) as
well as previously described kahalalides G and F.
3. Bioactivity
5.2.1. Kahalalides P (1). White powder; [a]2D5 C4.5 (c 0.22,
1
At a concentration of 100 mg/mL, kahalalides P (1) and Q
(2) showed 40 and 30% inhibition of HL-60 cancer cell
lines, respectively. At a concentration of 40 mg (using a
8 mm diameter paper disk), both compounds showed no
anti-microbial activity toward the marine bacterium
Ruegeria atlantica TUF-D.17 No hemolytic activity was
observed (0.8% sheep red blood cell suspension assay)18
with either compound at a concentration of 100 mg/mL.
MeOH); H and 13C NMR data were shown in Table 1;
HRFABMS C66H100N11O17 as [MCH]C m/z 1318.7319 (D
C2.1 mmu).
5.2.2. Kahalalides Q (2). White powder; [a]2D5 C10.5
1
(c 0.033, MeOH); H and 13C NMR data were shown in
Table 2; HRFABMS C66H100N11O16 as [MCH]C m/z
1302.7336 (D K1.3 mmu).
5.3. Acid hydrolysis of kahalalides
4. Conclusion
Peptides were hydrolyzed with 6 N HCl at 100 8C for 12 h.
The hydrolyzate was dried with centrifugation in vacuo.
New cyclic depsipeptides, kahalalides P (1) and Q (2), were
isolated from the Hawaiian green alga Bryopsis sp. The
sequential positions of DL anti-podal amino acids were
determined by a carboxypeptidase hydrolysis reaction. This
enzymatic method will be applicable to the structure
determination of other non-ribosomal peptides. The absol-
ute chemistry of 3-hydroxy-9-methyldecanoic acid in
kahalalides P and Q were determined by the recently
introduced convenient Mosher ester procedure. Interest-
ingly, the structure of the acyclic kahalalide P (1) compared
to kahalalides H and J3 suggest that kahalalide P may
possible be an intermediate in the metabolic pathway or end
product of these kahalalides. Kahalalide P without the
C-terminal Pro and Lys would be kahalalide H and without
the C-terminal Pro might be kahalalide J (DL determination
of Phe in kahalalide J has not been elucidated).
5.4. Base hydrolysis of kahalalides
Peptides (0.2 mg) were hydrolyzed with 0.1 N NaOH in
aqueous methanol (50 mL) at 37 8C for 90 min. After
neutralization with 0.1 N HCl, ODS HPLC of the
hydrolyzate was performed (Cosmosil 5C18-AR-II, 4.6!
250 mm, aqueous CH3CN 0–60%) to yield the acyclic
peptides.
5.5. Marfey analysis of 1 and 2
Part of the acid hydrolyzate was added to a 1% 1-fluoro-2,4-
bis(nitrophenyl)-5-L-alanine amide (FDAA) solution in
acetone (10 mL) and 1 M NaHCO3 (20 mL). The sample
was incubated at 40 8C for 1 h. The reaction mixture was
neutralized with 2 N HCl (20 mL) after cooling to room
temperature. The mixture was dried with centrifugation in
vacuo. The residue was dissolved in DMSO and subjected to
HPLC analysis. The FDAA derivatives of standard amino
acids were prepared by the same procedure. HPLC analysis
was carried out on a Cosmosil 5C18-AR-II, 4.6!250 mm
column with two different solvent system (system I: solvent
A (0.1 M ammonium acetate, pH 3.0) and solvent B
(CH3CN), samples were eluted as follows: A, 85–55% in
45 min, system II: solvent A (0.1 M ammonium acetate,
pH 3.0), 10% and solvent B (CH3CN), 90%). The retention
times (min) of the hydrolyzate FDAA derivatives and the
standard derivatives in the solvent system I were as follows:
L-Ser (8.00–8.04), D-Asp (10.76–10.81), L-Pro
(14.24–14.30), L-Phe (25.72–25.74), D-Val (28.05–28.10),
5. Experimental
5.1. General experimental procedures
Optical rotations were measured on a digital spectro-
polarimeter JASCO DP-100. 1H and two-dimensional
NMR spectra were recorded on Bruker DMX-750 spectro-
meter; 13C NMR spectra were obtained on Bruker DRX-500
spectrometer. MALDI TOF measurements were performed
on Perceptive Biosystem Voyager Elite MALDI TOF mass
spectrometer. Nanoflow electrospray ionization time of
flight MS/MS and MS were run on Micromass Q-TOF mass
spectrometer. HRFAB MS data were obtained on JEOL
JMX HX/HX-110A mass spectrometer.

1306
A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
Table 1. 1Ha and 13C NMRb data for kahalalide P (1) in 0.05% TFA/DMSO-d6
Unit
Position
13C NMR (ppm)c
1H NMR (ppm)d
Mult
J (Hz)
9Me3Decol
1
2
168.2
40.4
—
Ha 2.38
Hb 2.33
5.34
1.58, 2H
1.24, 2H
1.24, 2H
1.24, 2H
1.13, 2H
1.48
0.85, 6H
7.47
4.54
Ha 2.52
Hb 2.25
—
—
8.32
4.79
Ha 2.91
Hb 2.59
—
7.37, 2H
7.29, 2H
7.25
dd
dd
m
m
m
m
m
m
m
d
10.1, 14.7
3.4, 14.7
3
4
5
6
7
8
9
71.0
33.6
e
e
28.8
38.3
27.3
22.5, 2C
—
10, 11
NH
a
6.6
7.7
Asp
d
49.3
35.0
m
dd
dd
b
10.6, 16.2
4.0, 16.2
C]O
COOH
NH
a
169.3
171.9
—
53.9
39.3
Phe1
d
8.9
dt
dd
dd
9.5, 4.5
4.5, 13.0
10.2, 13.0
b
10
137.8
129.8, 2C
128.0, 2C
126.3
171.7
—
53.9
63.7
169.9
—
54.8
32.0
19.5
17.7
170.9
60.5
37.8
20, 60
30, 50
40
d
t
t
7.4
7.4
7.4
C]O
NH
a
—
8.75
5.54
3.58, 2H
—
7.81
4.64
1.98
Ser1
Val
d
8.3
m
d
b
5.5
—
C]O
NH
a
d
t
7.1
7.1
b
m
d
d
g
0.93, 3H
0.88, 3H
—
6.8
6.8
g0
C]O
a
b
Hyp
4.20
t
8.5
Ha 2.11
Hb 1.88
4.32
Ha 3.77
Hb 3.55
—
8.13
4.24
Ha 3.71
Hb 3.56
—
7.74
4.54
Ha 1.65
Hb 1.30
1.47
Ha 0.83, 3H
Hb 0.79, 3H
—
8.67
4.99
m
ddd
m
dd
m
4.4, 9.4, 13.4
4.0, 10.5
g
d
68.2
55.6
C]O
NH
a
171.8
—
55.7
60.9
Ser2
Leu
d
8.5
4.0, 8.5
4.0, 11.3
dt
dd
m
b
C]O
NH
a
169.0
—
50.3
40.7
d
8.9
m
m
m
m
d
b
g
23.8
23.2
21.8
171.9
—
52.9
d
6.7
6.7
d0
d
C]O
NH
a
Phe2
d
dd
d
9.0
8.0, 9.0
8.0
b00
2 , 600
300, 500
400
C]O
NH
a
38.3
2.81, 2H
—
7.14, 2H
7.10, 2H
7.07
—
8.41
4.52
Ha 1.68
Hb 1.38
1.20, 2H
1.50, 2H
2.75, 2H
7.55, 2H
—
137.1
129.2, 2C
127.7, 2C
126.1
169.8
—
100
d
t
t
7.4
7.4
7.4
Lys
d
8.8
48.8
31.9
m
m
m
m
m
m
br s
b
g
d
3
21.2
26.4
38.7
—
171.9
58.5
24.4
NH2
C]O
a
Pro
4.17
dd
m
m
m
m
m
m
4.2, 8.5
b
Ha 2.01
Hb 1.83
Ha 2.10
Hb 1.78
Ha 3.56
Hb 3.13
—
g
28.7
46.6
171.9
d
C]O
a
At 750 MHz.
At 188 MHz.
b
c
Reference of chemical shift was DMSO-d6 as 39.5 ppm.
Reference of chemical shift was DMSO-d6 as 2.49 ppm.
d
d
e
13C 26.6 or 24.8 ppm.

A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
Table 2. Ha and 13C NMRb data for kahalalide Q (2) in 0.05% TFA/DMSO-d6
1307
1
Unit
Position
13C NMR (ppm)c
1H NMR (ppm)d
Mult
J (Hz)
9Me-3decol
1
2
168.2
40.0
—
Ha 2.38
Hb 2.33
5.33
1.58, 2H
1.24, 2H
1.24, 2H
1.24, 2H
1.13, 2H
1.48
0.85, 6H
7.48
4.54
Ha 2.52
Hb 2.25
—
—
8.34
4.78
Ha 2.92
Hb 2.59
—
7.38, 2H
7.29, 2H
7.25
dd
dd
m
m
m
m
m
m
m
d
9.8, 14.5
3.6, 14.5
3
4
5
6
7
8
9
71.0
33.6
e
e
28.7
38.3
27.4
22.5, 2C
—
10,11
NH
a
6.6
7.4
Asp
d
49.2
35.0
m
m
dd
b
4.0, 16.4
C]0
COOH
NH
a
169.2
f
Phe1
—
53.4
38.9
d
8.8
dt
dd
dd
8.8, 4.5
4.5, 13.0
11.0, 13.0
b
10
137.8
20,60
30,50
40
129.7, 2C
128.0, 2C
126.3
d
t
t
7.4
7.4
7.4
f
C]O
NH
a
—
8.77
5.54
3.59, 2H
—
7.82
4.68
1.99
Ser1
Val
—
d
9.0
53.9
63.7
169.9
—
55.0
32.1
19.0
17.9
170.7
61.5
24.4
29.5
m
m
b
C]O
NH
a
b
g
—
d
t
7.0
7.0
m
d
d
0.91, 3H
0.89, 3H
—
6.8
6.8
C]O
Pro1
a
b
g
4.13
dd
m
m
m
m
m
4.1, 8.6
1.88, 2H
Ha 2.16
Hb 1.79
Ha 3.72
Hb 3.61
—
8.08
4.28
Ha 3.70
Hb 3.56
—
7.66
4.54
Ha 1.66
Hb 1.30
1.44
0.83, 3H
0.79, 3H
—
8.66
5.01
d
47.7
f
C]O
NH
a
Ser2
Leu
—
55.6
60.9
d
8.5
4.0, 8.5
dt
m
m
b
C]O
NH
a
169.0
—
50.3
40.3
d
8.9
m
m
m
m
d
b
g
23.8
23.2
21.8
f
d0
6.7
6.7
d
d
C]O
NH
a
Phe2
—
52.8
d
q
9.1
9.0
b00
2 ,600
300,500
400
C]O
NH
a
38.6
2.81, 2H
—
7.14, 2H
7.10, 2H
7.07
—
8.43
4.52
Ha 1.65
Hb 1.38
1.19, 2H
1.50, 2H
2.75, 2H
7.53, 2H
—
m
137.1
129.2, 2C
127.7, 2C
126.0
169.8
—
100
d
t
t
7.4
7.4
7.4
Lys
d
8.3
48.7
31.9
m
m
m
m
m
m
br s
b
g
d
3
21.1
26.4
38.5
—
169.2
58.5
24.3
NH2
C]O
a
Pro 2
4.18
dd
m
m
m
m
m
m
4.0, 8.5
b
Ha 2.01
Hb 1.84
Ha 2.10
Hb 1.77
Ha 3.54
Hb 3.12
—
g
28.7
d
46.5
f
C]O
a
At 750 MHz.
At 188 MHz.
b
c
Reference of chemical shift was DMSO-d6 as 39.5 ppm.
Reference of chemical shift was DMSO-d6 as 2.49 ppm.
d
e
f
d
13C 26.6 or 24.8 ppm.
d
13C 171.9 or 171.8 ppm.

1308
A. Dmitrenok et al. / Tetrahedron 62 (2006) 1301–1308
L-Lys (28.97–29.02), D-Phe (31.97–32.03) and D-Leu
(33.99–34.04). The configuration of Hyp could not be
determined in this system due to poor peak resolution.
Further analysis with solvent system II revealed L-Hyp in
the samples. (retention time: authentic 4-trans-hydroxy-D-
Pro 20.85 min, authentic 4-trans-hydroxy-L-Pro 22.73 min,
the sample 22.90 min).
with centrifugation in vacuo. Hydrolyzate was partitioned
between CHCl3 and H2O. Organic layer was evaporated and
obtained free fatty acid (6) was converted to methyl-3-
hydroxy-9-methyldecanoate (9) by diazomethane. (R)- and
(S)-MTPA esters of 9 were obtained in the same manner as
for the model compounds. Selected signals of the (S)-MTPA
ester of 9: 1H NMR (C5D5N, 750 MHz) d 2.83 (2H, d, H-2a
and H-2b) and 5.82 (m, H-3). Selected signals of the (R)-
MTPA ester of 9: 1H NMR (C5D5N, 750 MHz) d 2.89 (dd,
H-2a), 2.85 (dd, H-2b) and 5.80 (m, H-3).
5.6. Enzymatic hydrolysis
C-terminus of peptide 3 was hydrolyzed with carboxy-
peptidase P (Takara Bio Inc., Japan) in 30 mmol ammonium
acetate buffer (pH 5.0) and carboxypeptidase Y (Oriental
Yeast Co., Osaka, Japan) in 30 mmol ammonium acetate
buffer (pH 6.0). The carboxy end of peptides 4, 5 were
Acknowledgements
We thank Mr. Tsuyoshi Fujita at the SUNBOR for HRFAB
MS measurement, Drs. Hiroyuki Minakata, Miki Hisada,
and Manabu Horikawa at SUNBOR for valuable help and
discussion, Drs. Shigeki Matsunaga and Yoichi Nakao at
The University of Tokyo for discussion and Dr. Inna
Krasikova (Pacific Institute of Bioorganic Chemistry,
Vladivostok, Russia) for providing authentic samples of
3(R)- and 3(RCS)-hydroxy fatty acids.
hydrolyzed with carboxypeptidase
P in 30 mmol
ammonium acetate buffer (pH 5.0). Each sample (1 nM in
10 mL buffer) was mixed with carboxypeptidase P or
carboxypeptidase Y, 1 mL (1 unit), in a capped 0.5 mL
polypropylene tube and kept at 37 8C. Reactions were
monitored by MALDI MS measurement. Reactions completed
approximately in 10 min for 3 and 4 and in 15 min for 5.
5.7. Synthesis of peptides 5a and 5b
The peptides were synthesized using a FastMOCe chemistry
with a solid-phase peptide synthesizer (Model 433A, Applied
Biosystems) and purified by reverse phase HPLC (Cosmosil
5C18-AR-II, 4.6!250 mm) with 70% aqueous CH3CN.
References and notes
1. Hamann, M. T.; Scheuer, P. J. J. Am. Chem. Soc. 1993, 115,
5825–5826.
5.8. Preparation of the (R)- and (S)-MTPA ester
derivatives
2. Hamann, M. T.; Otto, C. S.; Scheuer, P. J.; Dunbar, D. C.
J. Org. Chem. 1996, 61, 6594–6600.
3. Goetz, G.; Nakao, Y.; Scheuer, P. J. J. Nat. Prod. 1997, 60,
562–567.
(R)- and (S)-MTPA esters of the compounds were obtained
following the reported procedure.16 Each model compound,
methyl (3R)-3-hydroxytetradecanoate (7) and methyl (3RC
3S)-3-hydroxytetradecanoate (7C8, racemic body), was
divided into two parts, then transferred into a clean NMR
tubes and dried under the stream of N2 gas. Deuterated
pyridine (0.6 mL) and (R)-(K)-a-methoxy-a-(trifluoro-
methyl)phenylacetyl chloride (6 mL) or (S)-(C)-a-
methoxy-a-(trifluoromethyl)phenylacetyl chloride (6 mL)
were added to NMR tube immediately under a N2 gas
stream, and then NMR tube was shaken carefully to mix the
sample and MTPA chloride evenly. The reaction NMR
tubes were permitted to stand at room temperature and were
4. Kan, Y.; Fujita, T.; Sakamoto, B.; Hokama, Y.; Nagai, H.
J. Nat. Prod. 1999, 62, 1169–1172.
5. Horgen, F. D.; de los Santos, D. B.; Goetz, G.; Sakamoto, B.;
Kan, Y.; Nagai, H.; Scheuer, P. J. J. Nat. Prod. 2000, 63,
152–154.
6. Bonnard, I.; Manzanares, I.; Rinehart, K. L. J. Nat. Prod. 2003,
66, 1466–1470.
7. Jimeno, J.;Lopez-Martin, J. A.;Ruiz-Casado, A.; Izquierdo, M. A.;
Scheuer,P. J.;Rinehart, K. Anti-cancerDrugs2004, 15, 321–329.
8. El Sayed, K. A.; Bartyzel, P.; Shen, X.; Perry, T. L.; Zjawiony,
J. K.; Hamann, M. T. Tetrahedron 2000, 56, 949–953.
9. Bourel-Bonnet, L.; Rao, K. V.; Hamann, M. T.; Ganesan, A.
J. Med. Chem. 2005, 48, 1330–1335.
1
monitored by H NMR. The reaction was completed in
approximately 4 h. Selected signals of the (S)-MTPA ester
of 7: 1H NMR (C5D5N, 500 MHz) d 2.83 (2H, d, JZ6.4 Hz,
H-2a and H-2b), 5.83 (m, H-3), 1.83 (m, H-4a), 1.75 (m,
H-4b) and 1.39 (m, H-5). Selected signals of the (R)-MTPA
ester of 7: 1H NMR (C5D5N, 500 MHz) d 2.89 (dd, JZ8.5,
16.5 Hz, H-2a), 2.85 (dd, JZ5.0, 16.5 Hz, H-2b), 5.81 (m,
H-3), 1.70 (2H, m, H-4) and 1.20 (m, H-5, data from
TOCSY spectrum).
10. Marfey, P. Carlsberg Res. Commun. 1984, 49, 591–596.
11. Hayashi, R.; Bai, Y.; Hata, T. J. Biochem. 1975, 77, 69–79.
12. Cucumel, K.; Bagnol, D.; Moinier, D.; Fisher, J.; Conrath, M.;
Cupo, A. J. Neuroimmunol. 1998, 81, 211–224.
13. Schroeder, W. A. Methods Enzymol. 1972, 25, 138–143.
14. Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512–518.
15. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am.
Chem. Soc. 1991, 113, 4092–4096.
16. Su, B.-N.; Park, E. J.; Mbwambo, Z. H.; Santarsiero, B. D.;
Mesecar, A. D.; Fong, H. H. S.; Pezzuto, J. M.; Kinghorn, A. D.
J. Nat. Prod. 2002, 65, 1278–1282.
Proton spectrum of the MTPA ester of the methyl (3RC3S)-
hydroxytetradecanoate (7C8, racemic body) was almost the
same as the sum of the spectra of (R)-MTPA ester of 7 and
(S)-MTPA ester of 7.
17. Namikoshi, M.; Negishi, R.; Nagai, H.; Dmitrenok, A.;
Kobayashi, H. J. Antibiot. 2003, 56, 755–761.
18. Nagai, H.; Oshiro, N.; Takuwa-Kuroda, K.; Iwanaga, S.;
Nozaki, M.; Nakajima, T. Biochem. Biophys. Res. Commun.
2002, 294, 760–763.
Two milligram of the peptide (1) was hydrolyzed by heating
the sample in 6 N HCl for 12 h. The hydrolyzate was dried