Lagunamides A and B: Cytotoxic and antimalarial cyclodepsipeptides from the marine cyanobacterium Lyngbya majuscula

Article abstract of DOI:10.1021/np100442x

Lagunamides A (1) and B (2) are new cyclic depsipeptides isolated from the marine cyanobacterium Lyngbya majuscula obtained from Pulau Hantu Besar, Singapore. The planar structural characterization of these molecules was achieved by extensive spectroscopic analysis, including 2D NMR experiments. In addition to Marfey's method and ³J_H-H coupling constant values, a modified method based on Mosher's reagents and analysis using LC-MS was deployed for the determination of the absolute configuration. Lagunamides A and B displayed significant antimalarial properties, with IC₅₀ values of 0.19 and 0.91 μM, respectively, when tested against Plasmodium falciparum. Lagunamides A and B also possessed potent cytotoxic activity against P388 murine leukemia cell lines, with IC₅₀ values of 6.4 and 20.5 nM, respectively. Furthermore, these cyanobacterial compounds exhibited moderate antiswarming activities when tested against Pseudomonas aeruginosa PA01.

Full text of DOI:10.1021/np100442x

1
810
J. Nat. Prod. 2010, 73, 1810–1814
Lagunamides A and B: Cytotoxic and Antimalarial Cyclodepsipeptides from the Marine
Cyanobacterium Lyngbya majuscula
†
‡
‡
§
,†
Ashootosh Tripathi, Jonathan Puddick, Mich e` le R. Prinsep, Matthias Rottmann, and Lik Tong Tan*
Natural Sciences and Science Education, National Institute of Education, Nanyang Technological UniVersity, 1 Nanyang Walk,
Singapore 637616, Singapore, Department of Chemistry, UniVersity of Waikato, PriVate Bag 3105, Hamilton 3205, New Zealand, Swiss
Tropical and Public Health Institute, Parasite Chemotherapy, Socinstrasse 57, PO Box, CH-4002 Basel, Switzerland, and UniVersity of Basel,
CH-4003 Basel, Switzerland
ReceiVed July 1, 2010
Lagunamides A (1) and B (2) are new cyclic depsipeptides isolated from the marine cyanobacterium Lyngbya majuscula
obtained from Pulau Hantu Besar, Singapore. The planar structural characterization of these molecules was achieved by
3
extensive spectroscopic analysis, including 2D NMR experiments. In addition to Marfey’s method and JH-H coupling
constant values, a modified method based on Mosher’s reagents and analysis using LC-MS was deployed for the
determination of the absolute configuration. Lagunamides A and B displayed significant antimalarial properties, with
IC50 values of 0.19 and 0.91 µM, respectively, when tested against Plasmodium falciparum. Lagunamides A and B also
possessed potent cytotoxic activity against P388 murine leukemia cell lines, with IC50 values of 6.4 and 20.5 nM,
respectively. Furthermore, these cyanobacterial compounds exhibited moderate antiswarming activities when tested against
Pseudomonas aeruginosa PA01.
Marine cyanobacterial strains belonging to the Lyngbya genus
Results and Discussion
are bountiful producers of structurally intriguing and biologically
1
Collections of a shallow water strain of L. majuscula were made
during low tides from the western lagoon of Pulau Hantu Besar,
Singapore, in June 2007. The cyanobacterial biomass was extracted
active secondary metabolites. Some of the important biological
activities associated with these secondary metabolites include
2,3
antimicrobial, antimalarial, cytotoxic, and neurotoxic properties.
3
repeatedly with CHCl /MeOH (1:1) and fractionated by VFC with
Among the diverse classes of compounds being discovered from
this genus, a substantial number belong to either the polypeptide
increasing polarity of organic solvents. Preliminary brine shrimp
lethality bioassay (BSLA) of the 100% EtOAc-eluted fraction
showed a high incidence of toxicity (100% at 10 ppm). The
bioactive fraction was then subjected to SEP PAK C18 solid-phase
fractionation followed by reversed-phase HPLC, yielding la-
gunamides A (1) and B (2) as colorless, amorphous solids.
1
-3
or hybrid polyketide-polypeptide structural classes.
As part of the drug discovery program in our laboratory, we chanced
upon a persistent strain of the marine cyanobacterium Lyngbya
majuscula from the intertidal region in the western lagoon of Pulau
Hantu Besar, Singapore. Subsequent chemical workup of its organic
extracts yielded a number of new, as well as known, cytotoxic
71 5 10
Lagunamide A (1) possesses a molecular formula of C45H N O
+
4,5
as determined from HRESIMS based on the [M + Na] ion peak
compounds. Further purification of the bioactive polar fraction
obtained from the vacuum flash chromatography (VFC) of the organic
extract led to the isolation of two new cyclic depsipeptides, la-
gunamides A (1) and B (2). These compounds are structurally related
to a series of potent cytotoxic marine cyanobacterial compounds
1 13
3
at m/z 864.5093. The H and C NMR data, recorded in CDCl ,
of the apparently chromatographically homogeneous material
revealed complex overlapping of signals, which may be due to the
existence of two or more conformers of 1. Nevertheless, the H
NMR spectrum indicated the peptidic nature of 1 and suggested
the presence of at least two secondary amide proton signals.
1
6-8
including the aurilides, kulokekahilide-2, and pulau’amide.
3
Changing the solvent to CD OD (Table 1) provided reasonably well
1
13
dispersed H and C NMR signals, which could be attributed to
the preponderance of a single conformer for compound 1 in
1
deuterated methanol. The H NMR spectrum further showed the
presence of at least three N-methyl amide groups at δ 2.90, 3.06,
1
3
3
and 3.32. Furthermore, the C NMR spectrum (in CD OD) of 1
indicated the presence of seven carbonyl carbons attributable to
ester/amide functionalities and a monosubstituted phenyl ring (δ
1
37.4, 129.6, 128.0, and 126.4) system (Table 1).
Detailed analysis of the 1D and 2D NMR data of 1 revealed a
structural framework consisting of peptide and polyketide sections
(
depicted as substructures A and B in Figure 1). Extensive NMR
analysis of 1 allowed the construction of five standard amino acid
and one hydroxy acid moieties in substructure A. Five standard
amino acid residues, N-methylalanine (N-Me-Ala), isoleucine (Ile),
N-methylglycine (N-Me-Gly), N-methylphenylalanine (N-Me-Phe),
and alanine (Ala), were deduced by following the extension of the
1
1
spin system of each residue by H- H COSY (Figure 1) and
*
To whom correspondence should be addressed. E-mail: liktong.tan@
TOCSY experiments. A sixth residue in substructure A (Figure 1)
nie.edu.sg. Tel: +65 6790 3820. Fax: +65 6896 9432.
1
†
exhibited H NMR resonances at δ 4.92 (H-28), 1.68 (H-29), and
Nanyang Technological University.
University of Waikato.
Swiss Tropical and Public Health Institute and University of Basel.
‡
0.97-1.51 (H-30 to H-32), suggesting an isoleucine residue.
However, the HSQC spectrum indicated the attachment of H-28 to
§
1
0.1021/np100442x  2010 American Chemical Society and American Society of Pharmacognosy
Published on Web 10/11/2010

Cytotoxic Cyclodepsipeptides from Lyngbya majuscula
Journal of Natural Products, 2010, Vol. 73, No. 11 1811
Table 1. NMR Spectroscopic Data for Lagunamides A (1) and B (2) in CD
3
OD (400 MHz)
lagunamide A (1)
lagunamide B (2)
a
C
b
c
a
C
b
unit
C/H no.
δ
δ
H
(J in Hz)
HMBC
δ
H
δ (J in Hz)
N-Me-Ala
1
2
3
4
5
6
7
8
8
9
1
171.7
59.2
12.8
36.5
172.0
53.7
38.4
23.7
171.6
60.0
12.5
36.5
170.6
53.0
38.5
23.9
3.96, q (6.9)
1.43, d (6.9)
3.32, s
1, 3, 4
1, 2
2, 5
3.79, m
1.38, d (6.8)
3.36, s
Ile
5.06, d (6.2)
1.83, m
1.64, m
1.32, m
0.93, m
5, 7, 8, 11
5.08, d (7.1)
1.86, m
1.74, m
1.43, m
0.97, m
a
b
10.3
15.0
7, 8
6, 7, 8
10.9
14.8
0
1.06, d (6.8)
1.04, d (6.8)
N-Me-Gly
N-Me-Phe
11
170.2
51.6
170.2
51.5
1
2
4.22, d (18.3)
11, 13
11, 13
12, 14
4.19, d (18.4)
3.58, d (18.4)
2.93, s
3
.59, d (18.3)
1
3
35.5
171.5
54.0
2.90, s
35.8
171.9
54.0
14
1
1
1
1
1
1
2
2
5
5.47, dd (10.3, 5.2)
3.04, m
2.94, dd (14.3, 5.2)
14, 16, 17, 23
15, 17, 18, 19
15, 17, 18, 19
5.44, dd (10.3, 5.0)
3.02, m
2.95, dd (14.1, 5.0)
6a
6b
7
8/22
9/21
0
34.8
34.9
137.4
128.0
129.6
126.4
29.5
137.4
128.0
129.6
126.4
29.4
7.20, m
7.18, m
7.28, m
3.06, s
16, 19, 21
20, 18, 22
19, 21
7.21, m
7.15, m
7.31, m
3.05, s
3
15, 24
Ala
24
173.8
45.3
14.7
173.8
45.6
14.4
2
2
5
6
4.52, q (6.9)
0.87, d (6.9)
24, 26, 27
24, 25
4.47, q (6.9)
0.87, d (6.9)
Hila
27
171.7
76.7
37.3
172.0
76.6
37.4
2
2
3
3
3
3
8
9
0a
0b
1
4.92, m
1.68, m
1.51, m
1.30, m
0.97, m
1.14, m
27, 29, 30, 32, 33
30, 31
29, 31
4.92, d (11.4)
1.86, m
1.54, m
1.34, m
0.96, m
26.3
26.4
10.6
13.5
28, 29, 30
28, 29
10.3
13.5
2
0.96, m
Dtea
33
169.3
127.6
145.6
29.6
169.5
127.8
145.8
29.4
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
5
6a
6b
7
8
9
0
1a
1b
2
3
4
7.32, brd
2.27, m
2.08, m
3.70, brd (3.0)
2.18, m (3.0, 3.8)
4.84, m (3.8, 10.0)
1.82, m (10.0)
1.32, m
1.16, m
0.91, m
1.93, s
0.90, m
34,36, 43
34, 35, 37
7.31, brd
2.25, m
2.04, m
3.79, brd
2.19, m
4.90, m
70.4
40.4
78.0
37.5
27.2
35, 44
70.6
40.4
82.8
132.4
126.4
1, 38, 40, 44
39, 41, 45
42
45
41a
33, 34, 35
37, 38, 39
40
5.62, m
11.3
11.4
9.0
9.7
11.4
8.5
1.63, m
1.93, s
0.76, d (7.0)
1.64, s
5
11.8
0.91, m
12.1
a
b
c
Recorded at 100 MHz. Recorded at 400 MHz. Coupling constants (Hz) are in parentheses. Protons correlated to carbon resonances in δ
C
column.
hydroxy acid residues in 1 was deduced from HMBC correlations
between H-2/C-1, H-4 and H6/C-5, H-6/C11, H-12/C-11, H -13/
C-12 and C-14, H-15/C-14, H -23/C-24, H-25 and H -26/C-24,
H-25/C-27, and H-28/C-27 (Table 1) to generate substructure A
3
3
3
(
Figure 1).
The structure elucidation of substructure B was accomplished
by COSY analysis of proton signals from the olefinic proton, H-35,
via the allylic methylene protons, H-36a/H-36b, the oxymethine
proton at H-37, and the methine proton at H-38. The continued
COSY correlations can be traced forward between methyl protons
at H
protons at H-39 and H-40, and further to the methyl proton signal
at H -45. Additional COSY correlations between H-40/H -41ab and
-41ab/H -42 completed substructure B. This substructure was
supported by HMBC long-range correlations observed from the
resonances of H -43 to C-33 and C-35, H -44 to C-37 and C-39,
and H -45 to C-39 and C-41 (Figure 1).
In a similar manner, the HMBC correlations observed from H
43 to C-33, C-34, and C-35, as well as correlations from H -44 to
3
-44 to the methine proton at H-38, extending to the methine
3
2
H
2
3
Figure 1. Planar structure of lagunamide A (1) with key COSY
and HMBC correlations indicated.
3
3
an oxymethine carbon (δ 76.7) and thus was more consistent with
a 2-hydroxyisoleucic acid (Hila) moiety (Figure 1). The sequence
N-Me-Ala-Ile-N-Me-Gly-N-Me-Phe-Ala-Hila of these amino/
3
3
-
3

1
812 Journal of Natural Products, 2010, Vol. 73, No. 11
Tripathi et al.
-2
Figure 2. ∆δ(S-R) values (×10 ppm) of the MTPA esters of
lagunamide A (1).
C-37 and C-39, acted as the missing link to complete substructure
B (Figure 1) of 1. Furthermore, the presence of a hydroxy group at
C-37 was suggested on the basis of the molecular formula and the
characteristic chemical shifts of H-37 (δ 3.70) and C-37 (δ 70.4).
3
4,35
The E-geometry of the double bond ∆
was assigned on the basis
13
of the C NMR chemical shifts of the methyl group CH -43 at δ
1.4.
Substructures A and B were connected on the basis of HMBC
3
9
1
data. The R-proton (H-28) of Hila showed a cross-peak to the C-33
carbonyl carbon in substructure B. Furthermore, the correlation of
H-39 in substructure B with the C-1 carbonyl carbon in substructure
A through an ester bond satisfies the final degree of unsaturation
to complete a 26-membered ring in lagunamide A (1).
Several chemical and NMR techniques were employed to
determine the configuration of 1. Lagunamide A was hydrolyzed
with 6 N HCl and was subjected to the advanced Marfey’s method,
revealing the absolute configurations of Ala, N-Me-Phe, N-Me-
1
0-12
Ala, and Ile to be L, D, L, and L-allo, respectively.
Due to the unavailability of pure Hila standards, we decided to
synthesize the isomers by diazotization of the corresponding amino
1
3
acids in dilute perchloric acid. The replacement of the amino
group by a hydroxy group is known to occur with retention of
configuration at the R-carbon owing to anchiomeric participation
14
of the neighboring carboxy function. The synthesized Hila isomers
were subjected to derivatization of the secondary alcohol with
Mosher’s acids (S- and R-MTPA) and subsequent analysis by LC-
MS (refer to scheme in Supporting Information). The modified
technique successfully enabled the assignment of the absolute
configuration of Hila as D-allo.
The absolute configuration of C-37 was defined by preparation
1
5
of the S- and R-MTPA esters of the hydroxy group at C-37 of 1.
The ∆δ(S-R) values (Figure 2) showed unambiguously that C-37
possesses the S-configuration. The relative configurations of the
other three carbons (C-38, C-39, and C-40) were determined using
3
J
H, H values as well as NOESY and ROESY correlations (Figure
).
The protons at H-37 and H-38 displayed a small coupling
3
Figure 3. Newman projections for (a) C-37/C-38; (b) C-38/C-39;
3
constant ( JH-37, H-38 ) 3.0 Hz, obtained through HOM2DJ
experiment in CD OD at 400 MHz), indicating them to be in a
and (c) C-39/C-40. Labels below projections denote predicted size
3
3
of the JH-H coupling constant between protons displayed. Predicted
values highlighted by a circle are consistent with observed JH-H
coupling constant values. Observed NOESY correlations are
presented as double-headed arrows.
syn-conformation. Having the absolute configuration at C-37
determined as S allowed the construction of four (out of six)
possible conformations for C-37 and C-38. A NOESY correlation
3
observed between H
possible relative conformations; however only one, 37S and
8S*, can satisfy the NOESY correlation between H -36 and
3
-44 and H-37 would be satisfied by two
confirmed the absolute configuration of C-37 as S, the absolute
configurations of 37S, 38S, 39S, and 40S were established for
the polyketide moiety in lagunamide A (1).
Lagunamide B (2) was isolated by RP-HPLC from the same
fraction containing compound 1. The HRESIMS of the molecule
3
2
H-38 (Figure 3a). Likewise, the protons H-38 and H-39 displayed
3
a small coupling constant ( JH-38, H-39 ) 3.8 Hz, obtained through
HOM2DJ experiment), which is consistent with a gauche-
conformation and produced four possible conformations. Of
these, however, only 38S* and 39S* can satisfy the NOESY
69 5
provided a molecular formula of C45H N O10 showing a [M +
+
Na] ion peak at 862.4937. Similarly to 1, compound 2 adopted at
1
correlation between H
3
-44 and H-40 (Figure 3b).
least two or more conformers in the H NMR spectrum when
The continued stereochemical analysis of the spin system
through C-39 to C-40 revealed a large coupling constant between
H-39 and H-40 ( JH-39, H-40 ) 10.0 Hz), indicating the protons to
be in an anti-orientation. Moreover, the NOESY correlation
measured in CDCl
to 1, as evidenced by nearly identical H and C NMR chemical
shifts when measured in CD OD (Table 1). However, it displayed
subtle differences within the polyketide portion, which could be
traced in the spin system from CH -42 to the olefinic protons at
H-41 in comparison to methylene H -41 in 1.
3
. Lagunamide B (2) had high structural similarity
1
13
3
3
between H
conformations of either 39S*,40S* or 39S*,40R*. However a
NOESY correlation observed between H -45 and H-38 suggested
the relative configuration to be 39S*,40S* (Figure 3c). Having
3
-45 and H-39 would satisfy two possible relative
3
2
3
Hydrolysis and stereoanalysis of the peptide portion of 2 were
undertaken as described above for lagunamide A (1). The absolute

Cytotoxic Cyclodepsipeptides from Lyngbya majuscula
Journal of Natural Products, 2010, Vol. 73, No. 11 1813
configurations of the five amino/hydroxy acid residues Ala, N-Me-
Phe, N-Me-Ala, Ile, and Hila were determined to be L, D, L, L-allo,
and D-allo, respectively. The absolute configurations from C-37 to
C-39 were proposed to be identical to those of 1 on the basis of
highly comparable NMR spectroscopic data. An E-geometry of the
detection at 230 nm) to obtain lagunamides A (1, 20.1 mg, 2% of
extract, t
min).
R
) 30.1 min) and B (2, 11.2 mg, 1.1% of extract, t
R
) 21.2
2
5
Lagunamide A (1): white, amorphous powder; [R]
D
-36 (c 0.5,
MeOH); UV(MeOH) λmax 229 nm (log ε 2.89); IR (neat) 3436, 2929,
-
1
1
2
599, 2341, 2054, 1739, 1634, 1519, 1246 cm ; H NMR (400.13
double bond ∆4
0, 41
in 2 was assigned on the basis of the C NMR
13
1
3
3 3
MHz, CD OD) and C NMR (100.62 MHz, CD OD) data, see Table
9
chemical shift of C-45 at δ 11.8.
+
1
; HRESIMS m/z [M + Na] 864.5106 (calcd for C45
864.5093).
Lagunamide B (2): white, amorphous powder; [R]
MeOH); UV(MeOH) λmax 221 nm (log ε 2.91); IR (neat) 3487 (br),
71 5
H N O10Na,
Lagunamides A (1) and B (2) were tested for their antimalarial
activity against the NF54 strain of the malarial parasite and showed
effective in Vitro activity against Plasmodium falciparum with IC50
values of 0.19 and 0.91 µM, respectively. This is the first report of
2
5
D
-39 (c 0.5,
-1
3
436 (br), 2962, 2566, 2374, 2170, 1736, 1641, 1462, 1199, 1044 cm ;
1
13
6-8
3 3
H NMR (400.13 MHz, CD OD) and C NMR (100.62 MHz, CD OD)
antimalarial activity for this class of aurilide-related molecules.
+
data, see Table 1; HRESIMS m/z [M + Na] 862.4940 (calcd for
10Na, 862.4937).
Advanced Marfey’s Analysis of Amino Acids. Lagunamide A (1,
.0 mg) was hydrolyzed in 6 N HCl (1 mL) in a sealed reaction vial
at 110 °C for 18 h. Trace HCl was removed under a stream of N gas,
and the resulting hydrolysate was redissolved in H O (0.6 mL) and
Interestingly, the only structural difference between 1 and 2 is the
additional olefinic group between C40-C41 in 2, and this minor
difference is reflected in the enhanced (about 4.7-fold increase)
antimalarial activity observed in 1. Furthermore, the antimalarial
activity of 1 is found to be similar to that of dolastatin 15 but less
potent than that of dolastatin 10, which is the most potent
45 69 5
C H N O
1
2
2
divided into two equal portions. Each portion was combined with either
a 1% solution of 1-fluoro-2,4-dinitrophenyl-5-L-alaninamide (L-FDAA,
Marfey’s reagent) (50 µL) in acetone or a racemic mixture of a 1%
solution of 1-fluoro-2,4-dinitrophenyl-5-DL-alaninamide (DL-FDAA, 50
1
6
antimalarial marine cyanobacterial compound known. The la-
gunamides also exhibited potent cytotoxic properties against the
P388 cancer cell line with IC50 values of 6.4 nM for 1 and 20.5
nM for 2.
3
µL) in acetone and 1 M NaHCO (25 µL), and the two mixtures were
heated at 40 °C for 45 min. Both reaction mixtures were cooled to
room temperature (rt), quenched by addition of 2 N HCl (25 µL), dried,
and redissolved in MeCN (500 µL). The aliquots were subjected to
reversed-phase LCMS (Agilent 1100 series) according to the advanced
Marfey’s method (Phenomenex, Luna, 150 × 2.0 mm, 5 µm, 100 Å;
MeCN in 0.1% (v/v) aqueous HCOOH; at 0.20 mL/min) using a linear
Lagunamides A (1) and B (2) displayed antiswarming activity
when tested against the Gram-negative bacterial strain Pseudomonas
aeruginosa PA01. Both compounds, when tested at 100 ppm,
exerted moderate antiswarming activities (62% for 1 and 56% for
2
compared to control). P. aeruginosa is an opportunistic nosoco-
1
0-12
mial pathogen, and the discovery of the lagunamides having
antiswarming activities on P. aeruginosa PA01 is therefore
significant. It is however unknown at this point if the antiswarming
activities of these cyanobacterial compounds operate by interference
of the bacterial quorum sensing system. The discovery of the diverse
biological activities of the lagunamides in this present study once
again demonstrates the importance of marine cyanobacteria as a
source of potential therapeutic agents.
gradient (10-50% MeCN over 60 min).
MSD spectrometer was used for detection in API-ES (negative mode).
The retention times and ESIMS product ions (t
An Agilent 1100 series
R
in min, m/z [M -
-
H] ) of the L-FDAA monoderivatized amino acids in the hydrolysate
of the first portion were observed to be Ile (49.2, 382.0), N-Me-Phe
(
51.9, 430.1), Ala (31.0, 340.0), and N-Me-Ala (37.4, 354.0), while
the reaction with racemic DL-FDAA in the second portion gave rise to
two peaks for each corresponding amino acid moiety. The retention
-
times and ESIMS product ions (tR1/tR2, min, m/z [M - H] ) were
observed to be Ile (49.2/56.1, 382.0), N-Me-Phe (51.1/51.9, 430.1),
Ala (31.0/51.1, 340.0), and N-Me-Ala (37.4/38.1, 354.0). Peaks eluted
Experimental Section
R
with longer t could be attributed to the D-FDAA derivative of the
General Experimental Procedures. Optical rotations were measured
on a Bellingham Stanley ADP 440 polarimeter. UV and IR spectra
were measured on a Varian Cary 50 UV visible spectrophotometer and
a PerkinElmer spectrum 100 FT-IR spectrophotometer, respectively.
amino acids. Consequently, the absolute configurations of the moieties
in the hydrolysate of 1 were confirmed as L-Ile, D-N-Me-Phe, L-Ala,
and L-N-Me-Ala.
Marfey’s Analysis for Isoleucine. Lagunamide A (1, 0.5 mg) was
hydrolyzed in 6 N HCl at 110 °C for 20 h and derivatized with Marfey’s
reagent (L-FDAA) as described above. Two portions each of 0.5 mg
1
13
H, C, and 2D NMR spectra were recorded in CD
Bruker NMR spectrometer using the residual solvent signal (δ
ppm and δ at 49.1 ppm) as internal standards. HRESIMS data were
3
OD on a 400 MHz
H
at 3.31
C
standard L-Ile and L-allo-Ile were dissolved in 100 µL of H
solution of L-FDAA (100 µL) and 1 N NaHCO (20 µL) were added
to one portion each of L-Ile and L-allo-Ile, and to the other portions
were added a 1.0% solution of D-FDAA (100 µL) and 1 N NaHCO
2
O. A 1.0%
obtained using a Bruker Daltonics MicrOTOF mass spectrometer. HPLC
isolation of lagunamides A (1) and B (2) was conducted on a Shimadzu
LC-8A Preparative LC equipped with a Shimadzu SPD-M10A VP diode
array detector, while an Agilent 1100 series coupled with an Agilent
LC/MSD trap XCT mass spectrometer equipped with an ESI interface
system was used for the detection of the Marfey-derivatized alanine,
N-methyl-phenylalanine, N-methyl-alanine, and isoleucine as well as
for Mosher’s derivatized R-hydroxy acids in compound 1. Cell viability
in 96-well plates was measured using a Bio Rad Benchmark Plus
microplate reader.
Biological Material. The filamentous benthic marine cyanobacterium
Lyngbya majuscula (∼1.5 L) was collected from the western lagoon
of Pulau Hantu Besar, Singapore, during low tides on June 25, 2007,
and stored before workup (70% EtOH(aq), -20 °C). A voucher
specimen of this cyanobacterial bloom material is maintained at NIE
under the code TLT/PHB/002.
3
3
(
20 µL). All four mixtures were then heated at 40 °C for 45 min. The
solutions were cooled to rt, neutralized with 2 N HCl (10 µL), and
evaporated to dryness. The residues were then resuspended in MeCN
(
1
1
500 µL). The aliquots were subjected to reversed-phase LCMS (Agilent
100 series) according to the Marfey’s method (Phenomenex, Luna,
50 × 2.0 mm, 5 µm, 100 Å; MeCN in 0.1% (v/v) aqueous HCOOH;
at 0.20 mL/min) using a linear gradient (30-70% MeCN over 40
min).
1
0-12
An Agilent 1100 series MSD spectrometer was used for
detection in API-ES (negative mode). The derivatized Ile residue in
the hydrolysate of 1 eluted at the same retention time as the derivatized
standard L-allo-Ile (15.8 min) but not that of L-Ile (15.0 min), D-Ile ()
D-FDAA-derivatized L-Ile, 18.4 min), and D-allo-Ile () D-FDAA-
derivatized L-allo-Ile, 18.8 min).
Preparation of 2-Hydroxyisoleucic Acid (Hila). L-Ile (100 mg, 0.75
mmol) was dissolved in 0.2 N perchloric acid (50 mL) at 0 °C. To this
2 3 2
was added a cold (0 °C) solution of Na SO (1.4 g, 20 mmol) in H O
Extraction and Isolation. The cyanobacterium (∼169 g dry wt)
was exhaustively extracted using CHCl
3
/MeOH (1:1) to produce an
organic extract (∼1 g). The extract was then fractionated on normal-
phase silica VFC using a stepwise gradient solvent system from hexane,
to EtOAc, and MeOH. The fraction eluted with 100% EtOAc was found
to possess 100% activity in the brine shrimp toxicity assay at 10 ppm
and was passed through a Sep-Pak RP-18 cartridge eluting with 100%
MeOH. The eluent was concentrated in Vacuo, and the resulting dark
green gum was subjected to C18 RP-HPLC (Phenomenex Sphereclone
(20 mL) with rapid stirring. With continued stirring the reaction mixture
was allowed to reach rt until evolution of N subsided (about 30 min).
The solution was then boiled for 3 min, cooled to rt, and saturated
with NaCl before extraction with Et O and drying under vacuum to
2
2
give 2S,3S-Hila (L-Hila). The three other stereoisomers 2R,3R-Hila (D-
Hila), 2R,3S-Hila (D-allo-Hila), and 2S,3R-Hila (L-allo-Hila) were
5
µm ODS, 250 × 10.00 mm, MeOH/H
2
O (78:22) at 3.0 mL/min, UV

1
814 Journal of Natural Products, 2010, Vol. 73, No. 11
Tripathi et al.
synthesized in a similar manner from D-Ile, D-allo-Ile, and L-allo-Ile,
respectively.
In Vitro Antimalarial Assay. Plasmodium falciparum drug-sensitive
1
3,14
NF54 and chloroquine-resistant K1 strains were cultivated in a variation
1
7,18
Absolute Configuration of Hila Moiety in 1. Determination of the
absolute configuration for the Hila residue in 1 was accomplished by
a modified method based on Mosher’s reagents and analysis using
LCMS. Lagunamide A (1, 1.0 mg) was hydrolyzed in 6 N HCl (1 mL)
in a sealed reaction vial at 110 °C for 18 h. Trace HCl was removed
of the medium previously described,
supplemented with 0.5% ALBUMAX II, Hepes (25 mM), NaHCO
consisting of RPMI 1640
3
(25 mM, pH 7.3), hypoxanthine (0.36 mM), and neomycin (100 µg/
mL). Human erythrocytes served as host cells. Cultures were maintained
in an atmosphere of O :CO :N (3:4:93) in humidified modular
2 2 2
under a stream of N
2
gas, the resulting hydrolysate was divided into
chambers at 37 °C. Compounds were dissolved in DMSO (10 mM),
diluted in hypoxanthine-free culture medium, and titrated in duplicates
over a 64-fold range in 96-well plates. Infected erythrocytes (1.25%
final hematocrit and 0.3% final parasitemia) were added into the wells.
two equal portions (0.5 mg each), and pyridine (0.5 mL) was added to
each. R-Methoxy-R-trifluoromethylphenylacetic acid (R-MTPACl) (2.5
mg) was added to one portion and S-MTPACl (2.5 mg) to the other.
The reaction was carried out for 10 h at rt, and the solvent was
3
After 48 h incubation, [ H]hypoxanthine (0.5 µCi) was added per well
evaporated under N
each) was derivatized with either R- or S-MTPACl (1.0 mg each in
.5 mL of pyridine). All derivatized samples were subjected to reversed-
2
. In a similar manner, each isomer of Hila (0.5 mg
and plates were incubated for an additional 24 h. Parasites were
harvested onto glass-fiber filters, and radioactivity was counted using
a Betaplate liquid scintillation counter (Wallac, Zurich). The results
were recorded and expressed as a percentage of the untreated controls.
0
phase LCMS (Agilent 1100 series) (Phenomenex, Luna, 150 × 2.0
mm, 5 µm, 100 Å; MeCN in 0.1% (v/v) aqueous HCOOH; at 0.20
mL/min) using a linear gradient (30-70% MeCN over 90 min). An
Agilent 1100 series MSD spectrometer was used for detection in API-
1
9
IC50 values were estimated by linear interpolation.
Acknowledgment. The authors would like to acknowledge NIE
AcRF (RI 8/05 TLT) for financial support. In addition, the authors thank
A. L. Ang (NSSE, NIE), S. G. Y. Lee (NSSE, NIE), C. C. Teo (NSSE,
NIE), and P. Gread (University of Waikato) for providing valuable
technical assistance as well as the Swiss Tropical and Public Health
Institute, Switzerland, for performing the antimalarial assay on the
lagunamides.
ES (negative mode). The retention times and ESIMS product ions (t
R
in min) of the S-MTPACl-monoderivatized standard hydroxy amino
acids were observed to be L-Hila (44.1 min), L-allo-Hila (44.5 min),
D-Hila () R-MTPACl-derivatized L-Hila, 42.9 min), and D-allo-Hila
(
) R-MTPACl-derivatized L-allo-Hila, 43.4 min). Consequently, the
absolute configuration of the Hila moiety in the hydrolysate of 1
derivatized with S-MTPACl was confirmed as D-allo-Hila since it eluted
at 43.4 min.
Supporting Information Available: H, 13C, and 2D NMR spectra
1
MTPA (r-Methoxy-r-Trifluoromethylphenylacetic Acid) Esters
of 1. Two portions of lagunamide A (1, 0.5 mg each) were reacted
with R- or S-MTPACl (5.0 mg) in pyridine (0.5 mL) for 10 h at rt, and
in CD OD for lagunamides A (1) and B (2) and biological data. This
material is available free of charge via the Internet at http://pubs.acs.org.
3
the solvent was then evaporated under N
2
. The corresponding esters
References and Notes
were subjected to NMR analysis.
1
S-MTPA Ester: H NMR (400 MHz, CD
H-36a), 2.015 (H-36b), 3.542 (H-37), 1.301 (H-38), 4.839 (H-39),
3
OD) δ 7.201 (H-35), 2.206
(1) (a) Gerwick, W. H.; Tan, L. T.; Sitachitta, N. In The Alkaloids.
Chemistry and Biology; Cordell, G. A., Ed.; Academic Press: San
Diego, 2001; Vol. 57, pp 75-184. (b) Tan, L. T. Phytochemistry 2007,
(
2
4
.248 (H-40), 1.207 (H-41a), 1.413 (H-41b), 0.918 (H-42), 1.907 (H-
6
8, 954–979.
3), 0.901 (H-44), 0.923 (H-45).
1
(2) Tan, L. T. J. Appl. Phycol. 2010, 22, 659–676.
R-MTPA Ester: H NMR (400 MHz, CD
3
OD) δ 7.223 (H-35),
(
3) Linington, R. G.; Clark, B. R.; Trimble, E. E.; Almanza, A.; Urena,
L. D.; Kyle, D. E.; Gerwick, W. H. J. Nat. Prod. 2009, 72, 14–17.
4) Tripathi, A.; Puddick, J.; Prinsep, M. R.; Lee, P. P. F.; Tan, L. T. J.
Nat. Prod. 2009, 72, 29–32.
2
3
.232 (H-36a), 2.047 (H-36b), 3.556 (H-37), 1.283 (H-38), 4.810 (H-
9), 2.236 (H-40), 1.191 (H-41a), 1.401 (H-41b), 0.902 (H-42), 1.939
(
(
H-43), 0.881 (H-44), 0.891 (H-45).
Antiswarming Assay Based on Pseudomonas aeruginosa PA01.
(5) Tripathi, A.; Puddick, J.; Prinsep, M. R.; Lee, P. P. F.; Tan, L. T.
Phytochemistry 2010, 71, 307–311.
(6) Han, B.; Gross, H.; Goeger, D. E.; Mooberry, S. L.; Gerwick, W. H.
J. Nat. Prod. 2006, 69, 572–575.
An antiswarming assay based on the bacteria P. aeruginosa PA01 was
carried out using either lagunamide A or B (0.5 mg) added into 5 mL
of molten STA (soft top agar) [constituted with 200 mL of deionized
(
(
(
7) Nakao, Y.; Yoshida, W. Y.; Takada, Y.; Kimura, J.; Yang, L.;
Mooberry, S. L.; Scheuer, P. J. J. Nat. Prod. 2004, 67, 1332–1340.
8) Williams, P. G.; Yoshida, W. Y.; Quon, M. K.; Moore, R. E.; Paul,
V. J. J. Nat. Prod. 2003, 66, 1545–1549.
9) Couperus, P. A.; Clague, A. D. H.; van Dongen, J. P. C. Org. Magn.
Reson. 1976, 8, 426–431.
H
1
2
O, 1.3 g of agar (Technical agar # 3 Oxoid), 2 g of tryptone (Difco),
g of NaCl (Sigma)] and poured immediately over the surface of a
solidified Luria Bertani agar (40 g/L from MoBio) plate as an overlay.
Once the overlaid agar had solidified, the plate was point inoculated
with the bacterial strain (5 µL) and incubated (37 °C for 72 h). The
experiment was performed in triplicate along with a MeOH-treated plate
as a control, and the extent of swarming was determined by measuring
the area of the colony using a leaf surface meter (area meter AM200,
ADC Bioscientific Ltd.).
Cytotoxicity Assay. Test solutions were added to P388 murine
leukemia cells (ATCC CCL46) in a series of eight 2-fold dilutions,
and cells were then incubated (35 °C, 72 h). Media, solvents, cells,
and positive controls were included in each assay run. 3-(4,5-
Dimethylthiazoyl-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was
added to wells; then cells were incubated for a further 4 h. Cell viability
was determined by measurement of formazan production via a
spectrophotometer at 540 nm. The percentage inhibition of cell growth
was determined by comparison of test well absorbance to that of a
control, and a plot of logarithm of sample concentration versus
absorbance was used to determine the IC50 (ng/mL) value (concentration
of the test sample required to reduce the P388 cell line growth by 50%).
(10) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591–596.
(11) Fuji, K.; Ikai, Y.; Mayumi, T.; Oka, H.; Sujuki, M.; Harada, K. I.
Anal. Chem. 1997, 69, 3346–3352.
(
12) Fujii, K.; Ikai, Y.; Oka, H.; Suzuki, M.; Harada, K. I. Anal. Chem.
1
997, 69, 5146–5151.
(
(
13) Mamer, O. A. Methods Enzymol. 2000, 324, 3–10.
14) Mamer, O. A.; Reimer, M. L. J. J. Biol. Chem. 1992, 267, 22141–
2
2147.
(
15) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092–4096.
(16) Fennell, B. J.; Carolan, S.; Pettit, G. R.; Bell, A. J. Antimicrob.
Chemother. 2003, 51, 833–841.
(
17) Dorn, A.; Stoffel, R.; Matile, H.; Bubendorf, A.; Ridley, R. G. Nature
1
995, 374, 269–271.
(18) Trager, W.; Jensen, J. B. Science 1976, 20, 673–675.
19) Huber, W.; Koella, J. Acta Trop. 1993, 55, 257–261.
(
NP100442X