Dragonamide E, a modified linear lipopeptide from Lyngbya majuscula with antileishmanial activity

Article abstract of DOI:10.1021/np900622m

Tropical parasitic and infectious diseases, such as leishmaniasis, pose enormous global health threats, but are largely neglected in commercial drug discovery programs. However, the Panama International Cooperative Biodiversity Group (ICBG) has been working to identify novel treatments for malaria, Chagas' disease, and leishmaniasis through an investigation of plants and microorganisms from Panama. We have pursued activity-guided isolation from an extract of Lyngbya majuscula that was found to be active against leishmaniasis. A new modified linear peptide from the dragonamide series was isolated, dragonamide E (1), along with two known modified linear peptides, dragonamide A (2) and herbamide B (3). Dragonamides A and E and herbamide B exhibited antileishmanial activity with IC₅₀ values of 6.5, 5.1, and 5.9 μM, respectively. Spectroscopic and stereochemical data for dragonamide E (1) and herbamide B (3; the spectroscopic and stereochemical data for this substance is incomplete in the literature) are presented as well as comparisons of biological activity within the dragonamide compound family. Biosynthetic differences among marine compounds with a terminal free amide are also discussed.

Full text of DOI:10.1021/np900622m

60
J. Nat. Prod. 2010, 73, 60–66
Dragonamide E, a Modified Linear Lipopeptide from Lyngbya majuscula with Antileishmanial
Activity
Marcy J. Balunas,†,‡,§ Roger G. Linington,†,‡,§,| Kevin Tidgewell,† Amanda M. Fenner,‡,§ Luis-David Uren˜a,‡
Gina Della Togna,‡ Dennis E. Kyle,‡,⊥ and William H. Gerwick*,†,§
Center for Marine Biotechnology and Biomedicine, Scripps Institution of Oceanography and Skaggs School of Pharmacy and Pharmaceutical
Sciences, UniVersity of California San Diego, La Jolla, California 92093, Instituto de InVestigaciones Cient´ıficas y SerVicios de Alta
Tecnolog´ıa, Clayton, Apartado 0816-02852 Panama´, Smithsonian Tropical Research Institute (STRI), Anco´n, Apartado 0843-03092 Panama´,
and Department of Global Health, College of Public Health, UniVersity of South Florida, Tampa, Florida 33612
ReceiVed October 19, 2009
Tropical parasitic and infectious diseases, such as leishmaniasis, pose enormous global health threats, but are largely
neglected in commercial drug discovery programs. However, the Panama International Cooperative Biodiversity Group
(ICBG) has been working to identify novel treatments for malaria, Chagas’ disease, and leishmaniasis through an
investigation of plants and microorganisms from Panama. We have pursued activity-guided isolation from an extract of
Lyngbya majuscula that was found to be active against leishmaniasis. A new modified linear peptide from the dragonamide
series was isolated, dragonamide E (1), along with two known modified linear peptides, dragonamide A (2) and herbamide
B (3). Dragonamides A and E and herbamide B exhibited antileishmanial activity with IC50 values of 6.5, 5.1, and 5.9
µM, respectively. Spectroscopic and stereochemical data for dragonamide E (1) and herbamide B (3; the spectroscopic
and stereochemical data for this substance is incomplete in the literature) are presented as well as comparisons of
biological activity within the dragonamide compound family. Biosynthetic differences among marine compounds with
a terminal free amide are also discussed.
Tropical diseases such as leishmaniasis, malaria, and Chagas’
disease affect millions of people in equatorial countries each year.1
Inhabitants of these developing-world tropical countries are at high
risk from infection and are often those with the least socioeconomic
ability to obtain proper treatments; therefore, they are the most likely
to develop serious and often fatal illnesses.2 Many tropical diseases
have developed resistance to existing therapeutics, thus limiting
the effectiveness of these treatments. A majority of the existing
agents for treating tropical diseases also have serious side effects
that reduce patient compliance with treatment regimens or, in some
cases, prohibit any treatment at all.1 Tropical diseases are largely
overlooked in drug discovery programs by major pharmaceutical
companies due to the lack of significant financial return on this
expensive and time-consuming process (only 13 of the 1300 new
drugs introduced during the period 1975 through 1999 were for
treating tropical diseases).1
The International Cooperative Biodiversity Group (ICBG) pro-
gram focused in Panama has been searching among marine
cyanobacteria and tropical plant endophytes for new, more effica-
cious, and less expensive treatments for tropical diseases.3-5 It is
hoped that these novel agents will show activity against resistant
strains of these parasites. During the course of this research, a new
modified linear lipopeptide, dragonamide E (1), as well as two
related compounds previously reported in the literature, dragona-
mide A (2) and herbamide B (3), were isolated from a field-collected
marine cyanobacterium and found to be active against leishmaniasis
in an in Vitro screening assay.
Results and Discussion
L. majuscula was collected using snorkeling in shallow waters
and fractionated using normal-phase flash chromatography. One
from Bocas del Toro, Panama, in 2006 and subsequently extracted
of the fractions (60% EtOAc/hexanes) exhibited strong antileish-
manial activity (7.2% of control parasite growth at 10 µg/mL,
* To whom correspondence should be addressed. Tel: +1-858-534-0578.
experimental detail in Supporting Information), while a second more
polar fraction eluting with 100% EtOAc exhibited less potent
antileishmanial activity (38.2% of control parasite growth at 10
µg/mL). This more polar (100% EtOAc) fraction was successively
purified by RP-SPE column chromatography and RP-HPLC, and
two compounds were ultimately isolated, dragonamides E (1) and
Fax: +1-858-534-0529. E-mail: wgerwick@ucsd.edu.
† University of California San Diego.
‡ INDICASAT, Panama´.
§ STRI, Panama´.
⊥ University of South Florida.
| Current address: Department of Chemistry and Biochemistry, University
of California, Santa Cruz.
10.1021/np900622m  2010 American Chemical Society and American Society of Pharmacognosy
Published on Web 12/23/2009

Lipopeptide from Lyngbya majuscula
Journal of Natural Products, 2010, Vol. 73, No. 1 61
A (2). Metabolite 2 was the major compound of this fraction and
exhibited 1H and 13C NMR signals indicative of phenylalanyl and
valinyl residues, along with a fatty acyl chain. Combined with a
prominent [M + H]+ peak by APCIMS at m/z 654, these data were
fully consistent with literature values for the known metabolite
dragonamide A (2).6,7
herbamide B (3) showed an intense [M + H]+ peak by HR APCIMS
for C17H24Cl3N2OS, thus confirming the structure assigned previ-
ously by NMR analysis alone.6
The stereoconfigurations of herbamide B (3) and the closely
related compound herbamide A have never been determined. The
configuration of C-9 in the modified valine residue of 3 was
determined using sequential ozonolysis, hydrolysis, and Marfey’s
analysis using the FDVA Marfey’s reagent. This fragment was
compared with commercially available valine standards that were
also transformed by Marfey’s reagent and identified this substituent
in herbamide B (3) to have an L configuration.
The second eluting and less abundant compound from this
fraction, named dragonamide E (1), gave an [M + H]+ peak by
HRESI-TOFMS that calculated for C37H58N5O5 (12 degrees of
unsaturation) and by database searches was determined to be an
1
unknown substance. The H NMR spectra of 1 and dragonamide
A (2) were very similar, and both included signals for a terminal
NH2 (δ 5.99), an N-methyl phenylalanyl, and three N-methyl valinyl
residues. While both 1 and 3 existed in a single major conformation
in solution, the presence of multiple N-methyl amide functionalities
resulted in the presence of several minor conformers as well (see
Supporting Information). Compound 1 possessed 13C NMR absorp-
tions at δ 84.0 and 69.0, consistent with a terminal acetylene as in
dragonamides A (2) and B (4). However, dragonamide A (2)
possesses two more protons and only 11 degrees of unsaturation
The remaining stereocenter at C-2 of 3 was determined using
the semisynthetic strategy shown in Figure 1. The planar structures
of the fragment that results from ozonolysis and oxidative workup
of 3 and that resulting from ozonolysis and cleavage of the methyl
ester of barbamide are identical. The methine adjacent to the
trichloromethyl in barbamide has previously been determined to
have the S configuration,9 and an abundant supply of this pure and
fully characterized compound was available. After reaction of the
barbamide fragment with both enantiomers of 2-phenylglycine
methyl ester (PGME), two diastereomeric compounds were obtained
with distinctive chemical shifts for the H-1 methyl and H-3
methylene (Figure 2). Herbamide B (3), with unknown configuration
at C-2, was subjected to ozonolysis and reaction with S-PGME and
found to have the identical chemical shifts to the barbamide
fragment reacted with S-PGME (Figure 2), indicating that herba-
mide B and barbamide have the same configuration at this methine
center. The final, stereochemically complete structure for 3 was
therefore determined to have the S configuration at C-2 and an
L-valine residue at C-9.
1
compared to dragonamide E. In the H NMR spectrum of 1, the
distinctive signal corresponding to the methine proton on C-35 (δH
2.68; δC 36.2) in 2 was not present. Instead, in 1 the C-35 carbon
resonated significantly downfield (δ 133.0) compared to dragona-
mide A and, by DEPT and HSQC analysis, was identified as a
quaternary carbon. These observations suggested that the extra
degree of unsaturation in 1 relative to 2 was the result of a double
bond between C-35 and C-36. This deduction was confirmed by
1
location of H-36 at δ 5.41 in the H NMR and observation of
HMBC correlations between H-36 and C-42 (δ 14.4) as well as
C-34 (δ 173.9). Correlations observed between the H-42 methyl
protons and the H-37 methylene protons by NOESY indicated an
E geometry for the C-35-C-36 olefinic bond.
The absolute configurations of the four amino acid residues in
dragonamide E (1) were determined using Marfey’s analysis.
Following hydrolysis, dragonamide E (1) was derivatized using
FDVA Marfey’s reagent and compared with commercially available
N-methylphenylalanine and N-methylvaline standards that were
similarly transformed by Marfey’s reagent. All stereogenic centers
in dragonamide E (1) were determined to have the L configuration.
The identical configuration of the trichloromethyl for both
barbamide and herbamide B is potentially indicative of the
conservation of the BarbB1 and BarbB2 enzymes in the herbamide
B producer and their chiral preference during trichloromethyl
biosynthesis.10,11 However, the subsequent biosynthetic steps for
herbamide B likely involve the incorporation of at least one extra
PKS module as well as utilization of alternative optional PKS
domains (e.g., dehydratase and C-methyltransferase).12 Further
studies on the biosynthesis of herbamide B should help illuminate
the regions of similarity and divergence from the barbamide
pathway.
The less polar (60% EtOAc/hexanes) yet strongly antileishmanial
fraction from the primary chromatography was subjected to
reversed-phase solid-phase extraction (SPE) and reversed-phase
HPLC to yield herbamide B (3), a previously reported but
incompletely characterized natural product.6 Several partial struc-
tures were determined from NMR analysis of 3, which aided its
identification. Specifically, the presence of a modified valine residue
was determined on the basis of observation of two methyl doublet
signals (δ 0.99 and 0.94), a double-doublet R-proton signal (δ 5.33),
a methine (δ 2.36), and an N-H (δ 6.68). The presence of a thiazole
ring was determined by the distinctive proton doublet signals at δ
7.27 and 7.75 and was connected to the valine residue through
observation of a three-bond HMBC correlation between C-15 (δ
170.1) and H-10 (δ 2.36). The constellation of a downfield shifted
doublet methyl at δ 1.31 with an HMBC correlation to a uniquely
downfield shifted quaternary carbon (δ 105.5), and a high-field
methine proton at δ 2.60 with HMBC correlations to the carbon
atom of the aforementioned downfield shifted doublet methyl (δ
16.2), suggested a terminus with methyl and trichloromethyl groups,
as in barbamide.8 The remaining downfield singlet methyl (δ 2.02)
was assigned to a vinyl position on the basis of its chemical shift
and was placed at C-7 on the basis of a strong HMBC correlation
between this resonance and the unsaturated ester carbonyl absorption
band at δ 168.5. Piecing these partial structures together identified
a molecular framework consistent with that published for herbamide
B, and upon detailed comparison, all of the 1H and 13C NMR signals
closely matched those published for this structure.6 Our isolate of
Compounds 1-3 were tested for activity against three tropical
disease parasites, Plasmodium falciparum (malaria), Leishmania
donoVani (leishmaniasis), and Trypanosoma cruzi (Chagas’ disease).
Compound 4 was previously tested in these assays,7 and those
results are also reported here to allow for direct comparison with
the biological properties of dragonamides A (2), B (3), and E (1).
Dragonamide E (1) exhibited moderately strong in Vitro activity
against leishmaniasis (5.1 µM), with dragonamide A (2) and
herbamide B (3) also showing comparable activity against this
parasite (6.5 and 5.9 µM, respectively). Dragonamide B (4) was
inactive against all of the tested parasites, indicating that the
aromatic ring-containing residue at the peptide terminus is necessary
for activity. Interestingly, dragonamide A (2) previously showed
weak activity against malaria parasites;7 however, during the current
evaluation, 2 was found to be inactive in this assay (maximum test
concentration ) 10 µM). The in Vitro activity of dragonamide E
(1) and herbamide B (3) against the parasite that causes visceral
leishmaniasis supports their further examination through in ViVo
evaluations. This is especially attractive in the case of herbamide
B because it was isolated as a major metabolite of this cyanobac-
terial collection, and hence, analogues could be produced through
natural product derivitization strategies.
The isolation of dragonamide E (1) represents an interesting
addition to the dragonamide family of natural products. This fatty
acid moiety has no precedence in the marine literature. All other
compounds in the dragonamide series have secondary methyl groups

62 Journal of Natural Products, 2010, Vol. 73, No. 1
Balunas et al.
Figure 1. Semisynthetic strategy utilized to determine stereochemical configuration of herbamide B (3) at C-2.
Figure 2. 1H NMR comparison of chemical shifts for the S-PGME derivative of the herbamide B fragment (A), the S-PGME derivative of
the barbamide fragment (B), and the R-PGME derivative of the barbamide fragment (C). The S-PGME derivative of the herbamide B
fragment directly overlaps with the S-PGME derivative of the barbamide fragment and has distinctive shift differences from the R-PGME
derivative of the barbamide fragment, indicating that herbamide B (3) and barbamide have the same configuration at this center.
of S configuration at the C-35 or equivalent position, with the
exception of dragonamide C, which possesses an E-configured
olefinic bond between C-31 and C-32 (corresponds to C-35 and
C-36 in dragonamide E). However, dragonamide C lacks an
R-methyl group (C-42 in compound 1), but rather, possesses a
methoxy at C-32, which causes the fatty acyl moiety of dragonamide
C to have a distinctively different spatial orientation (Figure 3a).
However, it is possible that dragonamide C is an isolation artifact,

Lipopeptide from Lyngbya majuscula
Journal of Natural Products, 2010, Vol. 73, No. 1 63
this point. Isolation of dragonamide E with an alkene between C-35
and C-36 of the fatty acyl moiety is potentially important in that it
suggests a possible mechanism by which other modified linear
peptides are created with both R- and S-configured R-methyl groups
at this center (Figure 3b), perhaps through enoyl reductases of
different facial selectivity, or E versus Z double bonds at this
location.
The isolation of dragonamide E (1) as the fifth compound in the
series led us to consider the biosynthetic relationships of these
compounds, all isolated from Lyngbya species, as well as for several
related compounds with a free amide terminus (Table 2). A total
of 18 compounds with a primary amide terminus have been reported
from various marine organisms, including cyanobacteria (11
compounds), mollusks (2), sponges (3), and shrimp (2). These have
been isolated from collections in both the Pacific Ocean (Japan,
Hawaii, and Ecuador) and the Caribbean Sea (Panama, Florida,
and Curac¸ao), indicating no apparent specialization based on
geography. A range of biological activities have been reported for
these compounds including antiparasitic,6,7 anticancer,13 antifun-
gal,14 diuretic,15 myotropic,15 inhibition of serine protease,16 and
cardioexcitatory activity.17
Figure 3. (A) Distinctive spatial arrangements of the fatty acyl
moieties of dragonamide E (1) and dragonamide C, and (B)
dragonamide E as a precursor for either the R or S configuration of
the fatty acyl moiety, present in dragonamides A and B, dragom-
abin, and majusculamide B (S moiety) and almiramides A-C and
majusculamide A (R moiety).
One of the more striking differences between the structures of
these various primary amide containing natural products lies in the
high level of N-methylation of the cyanobacterial compounds; the
metabolites of other classes of marine life have little or no
created during the methanol extraction process from the corre-
sponding ꢀ-keto compound; further studies are needed to explore
Table 1. NMR Data for Dragonamide E (1) in CDCl3
a
position
δCb, mult.
δH (J in Hz)
COSY
selected HMBC (HfC)
selected NOESY
1
NH2
1
2
3
4
5
5.99, br s
N-MePhe
171.9, qC
56.6, CH
N
5.57, dd (11, 5)
5
3
2, 5
33.4, CH2
3.03, dd (ob)
3, 6, 7, 11
5
3.25, dd (15, 5)
6
7
8
136.9, qC
128.9, CH
128.8, CH
127.2, CH
128.8, CH
128.9, CH
32.0, CH3
171.5, qC
58.7, CH
7.16, m
7.28, m
7.24, m
7.28, m
7.16, m
2.90, s
8
5, 9
6
1, 3, 5, 8
1, 7
7, 9
8, 10
9, 11
10
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
6
1, 11
1, 5, 10
17, 18
5, 9
3, 13
N-MeVal-1
N-MeVal-2
N-MeVal-3
FA moietyc
5.08, d (11)
16
13, 16, 17, 19
12, 17
27.2, CH
20.2, CH3
17.6, CH3
29.9, CH3
169.9, qC
58.2, CH
2.26, m
17, 18
16
16
18
18
16, 17
21
0.88, d (6)
0.66, d (7)
2.43, s
14, 16, 18
14, 16, 17
14, 20
4.94, d (11)
23
20, 23, 25, 26
19
27.3, CH
17.9, CH3
19.9, CH3
30.5, CH3
170.6, qC
58.0, CH
2.26, m
21, 24, 25
23
23
0.74, d (7)
0.71, d (6)
2.98, s
21, 23, 25
21, 23, 24
21, 27
28
26
5.12, d (11)
30
30, 33, 34
26.9, CH
18.1, CH3
19.7, CH3
30.9, CH3
173.9, qC
133.0, qC
129.8, CH
26.6, CH2
27.8, CH2
18.2, CH2
84.0, qC
2.35, m
28, 31, 32
28
28
28, 32
31
30
0.85, d (ob)
0.84, d (7)
2.93, s
28, 30, 32
28, 30, 31
28, 34
42
5.41, dt (7, 2)
2.23, m (ob)
1.63, q (7)
37
34, 37, 42
37
36, 38
37, 39
38
35, 36, 38, 39
36, 37, 39, 40
37, 40, 41
36, 38
37, 39
38
2.19, dt (7, 2)
69.0, CH
14.4, CH3
1.97, t (2)
1.82, s
34, 35, 36
33
a Measured at 400 MHz. b Measured at 100 MHz. c FA moiety ) (E)-2-methyloct-2-en-7-ynoic acid.

64 Journal of Natural Products, 2010, Vol. 73, No. 1
Balunas et al.
N-methylation (Table 2). Compounds from organisms other than
cyanobacteria also have a high level of unusual peptidic modifica-
tions, whereas cyanobacterial compounds with terminal amides have
little modification beyond N-methylation. Another distinct aspect
of these cyanobacterial compounds lies in their extended polyketide
tail, which is largely absent in terminal amide compounds from
other organisms.
Within the cyanobacterial compounds, the first two amino acid
residues are almost always N-methylated, with the exception of
dysidenamide (Table 2). N-Methylvaline is the most common first
and second residue, with N-methylphenylalanine, tyrosine (also
O-methylated in the case of the majusculamides), or alanine also
being found in these two residues. The third amino acid residue is
either alanine or N-methylvaline (or absent as in the majusculamides
and dysidenamide). The fourth residue is always N-methylated and
is either valine or phenylalanine. The polyketide tails of these
cyanobacterial compounds all end in terminal acetylene function-
alities except for carmabin B. A 10-carbon PKS chain is the most
common, but chain length is variable and ranges from 9 to 13
carbons. Dragonamides A and B, as well as dragomabin, all have
the same PKS tail comprising a 2-methyloct-7-ynoyl residue with
an S-configured secondary methyl group.7 The carmabins are both
chain extended, containing two additional carbons in the polyketide
chain and an additional methyl substituent. The structural differences
in the lipophilic chain likely represent variations in the biosynthetic
assembly process, specifically, the incorporation of additional PKS
domains in the cluster. As deduced from Table 2, the close structural
relationship of these primary amide-containing cyanobacterial
peptides suggests an evolutionary relationship in their biosynthetic
gene clusters and further exemplifies a strategy by which these
marine prokaryotes diversify their suite of expressed natural
products.
Experimental Section
General Experimental Procedures. Optical rotations were obtained
with a Jasco P-2000 polarimeter. UV and IR spectra were recorded on
Beckman Coulter DU-800 and Nicolet IR100 FT-IR spectrophotom-
eters, respectively. NMR spectra were recorded on a JEOL Eclipse 400
MHz spectrometer or a Varian Inova 600 MHz spectrometer and
referenced to residual solvent signals (δH 7.26, δC 77.2 for CDCl3).
Low-resolution APCI mass spectra were obtained on a JEOL LC-mate
mass spectrometer. High-resolution mass spectra were acquired on
Agilent ESI-TOF or JEOL LC-Mate APCI mass spectrometers. HPLC
purifications of natural product isolates were carried out on a Merck
Hitachi LaChrom HPLC system with a L-7100 pump, L-7614 degasser,
and L-7455 diode array detector using a Prontosil-120 C18 (4.6 × 250
mm, 5 µM) RP-HPLC column, with solvent systems as indicated below.
HPLC purifications of semisynthetic compounds were carried out on a
Waters HPLC system with a Waters 515 binary pump and a Waters
996 PDA detector using a Phenomenex Synergi Fusion 4µ column (10
× 250 mm), with solvent systems as indicated below.
Collection, Extraction, and Isolation. Lyngbya majuscula was
collected in July 2006 by hand using snorkeling methods from
mangrove roots in the Bastimentos National Park (9°14.689 N,
82°08.366 W) in Bocas del Toro, Panama. After straining through
a mesh bag to remove seawater, the sample was stored at -4 °C.
Voucher specimen number PAB-28-JUL-06-1 is deposited at the
INDICASAT sample respository, Panama City, Panama. The sample
(63.4 g dry weight) was thawed and repetitively extracted seven
times with 2:1 CH2Cl2/MeOH to afford after solvent evaporation
3.2 g of crude organic extract. The extract was fractionated using
flash Si gel column chromatography (Aldrich, Si gel 60, 230-400
mesh, 40 × 180 mm) using 300 mL each of 100% hexanes (A), 9:1
hexanes/EtOAc (B), 4:1 hexanes/EtOAc (C), 3:2 hexanes/EtOAc (D),
2:3 hexanes/EtOAc (E), 1:4 hexanes/EtOAc (F), 100% EtOAc (G),
3:1 EtOAc/MeOH (H), and 100% MeOH (I). Fraction E showed
strong antileishmanial activity (only 7.2% of control growth at 10
µg/mL), while fraction G showed good antileishmanial activity
(38.2% of control growth at 10 µg/mL). Fraction E was subjected
to further fractionation using a Burdick & Jackson C18 RP-SPE
cartridge using a MeOH/H2O gradient (1:1, 3:2, 7:3, 4:1 MeOH/

Lipopeptide from Lyngbya majuscula
Journal of Natural Products, 2010, Vol. 73, No. 1 65
EtOAc, 100% MeOH, 100% EtOAc). The 7:3 eluting fraction was
subjected to RP-HPLC purification (63% MeOH/37% H2O, 209 nm,
1.0 mL/min) to yield herbamide B (3, 11.2 mg, tR 49.5 min, 0.35%
of extract). Fraction G was passed through a Burdick & Jackson
C18 RP-SPE cartridge with a 0.45 µm nylon filter using 100% MeOH.
The eluent was chromatographed using RP-HPLC (69% MeOH/
39%H2O, 209 nm, 1.0 mL/min), yielding dragonamide E (1, 3.0
mg, tR 20.9 min, 0.09% of extract) and dragonamide A (2, 22.3 mg,
tR 24.8 min, 0.70% of extract).
NaCl and then dried over NaSO4. This was filtered using Celite and
then concentrated under reduced pressure to afford 7.3 mg of impure
material. This was subjected to HPLC purification using a Phenomenex
Synergi Fusion 4µ column (10 × 250 mm) with an isocratic solvent
system of 61% CH3CN/39% H2O and a flow of 2 mL/min (retention
time 23.7 min) to yield 1.0 mg of the R-PGME derivative of barbamide
fragment (2.84 µmol, 16%): [R]D -405 (c 0.59, CHCl3); 1H NMR
(CDCl3, 600 MHz) δ 7.33-7.39 (m, 5H), 6.49 (d, J ) 6.6 Hz, 1H),
5.58 (d, J ) 7.1 Hz, 1H), 3.75 (s, 3H), 3.18 (m, 1H), 3.02 (dd, J )
14.9, 2.9 Hz, 1H), 2.29 (dd, J ) 15.0, 9.8 Hz, 1H), 1.38 (d, J ) 6.5
Hz, 3H); HRESIMS m/z 374.0094 (calcd for C14H16Cl3NO3Na,
374.0088).
Dragonamide E (1): white, amorphous solid; [R]25D -220 (c 0.15,
CHCl3); UV (MeOH) λmax (log ε) 203 (5.69), 228 (5.16) nm; IR νmax
(film) 3251, 2964, 1688, 1632, 1463, 1393, 1259, 1097 cm-1; 1H NMR
and 13C NMR data (CDCl3, 400 MHz), see Table 1; APCIMS m/z (%)
652.5 (80, [M + H]+), 474.3 (100), 361.2 (82), 248 (91); HRESI-
The S-PGME derivative of the barbamide fragment was synthesized
as described for the R-PGME derivative using S-PGME (100.7 mg,
0.50 mmol) to afford 0.6 mg (1.70 µmol, 9%): [R]D +343 (c 0.35,
CHCl3); 1H NMR (CDCl3, 600 MHz) δ 7.33-7.39 (m, 5H), 6.53 (d, J
) 6.3 Hz, 1H), 5.59 (d, J ) 7.0 Hz, 1H), 3.74 (s, 3H), 3.18 (m, 1H),
3.08 (dd, J ) 14.9, 2.5 Hz, 1H), 2.26 (dd, J ) 15.0, 9.8 Hz, 1H), 1.28
TOFMS [M + H]+ m/z 652.4418 (calcd for C37H58N5O5, 652.4437).
1
Herbamide B (3): yellow glass; [R]25 -27 (c 0.20, CHCl3); H
D
NMR and 13C NMR data (CDCl3, 400 MHz) were as reported;6 HMBC
NMR (HfC) (1f2, 3, 13), (2f1, 3, 4, 13), (3f1, 2, 4, 5, 13), (4f2,
3, 6), (5f3, 6, 7), (6f4, 7, 8, 14), (9f10, 11, 12, 15), (10f9, 11, 12,
15), (11f9, 10, 12), (12f9, 10, 11), (14f6, 7, 8), (16f15, 17),
(17f15, 16), (NHf8, 9); APCIMS m/z (%) [M + H]+ 409.5 (41),
375.5 (7), 339.5 (21), 303.4 (69), 265.3 (100); HR APCIMS [M +
H]+ m/z 409.0689 (calcd for C17H24Cl3N2OS, 409.0675).
(d,
J ) 6.4 Hz, 3H); HRESIMS m/z 374.0091 (calcd for
C14H16Cl3NO3Na, 374.0088).
Ozonolysis was separately performed on herbamide B using O3(g)
(4% in O2, 1/16 L/min) bubbled through 1.0 mg of herbamide B in
200 µL of CH2Cl2 at -78 °C for 5 min. This was dried using N2 gas
and oxidized using H2O2, then acidified with 1 N HCl and partitioned
sequentially with ether (2 × 10 mL) and ethyl acetate (2 × 10 mL).
The organic layers were combined and concentrated and used without
purification for the next reaction.
The herbamide B fragment was reacted only with S-PGME (100.6
mg, 0.50 mmol) as described above to yield 0.7 mg (1.99 µmol, 70%):
[R]D +302 (c 0.24, CHCl3); 1H NMR (CDCl3, 600 MHz) δ 7.33-7.39
(m, 5H), 6.53 (d, J ) 6.4 Hz, 1H), 5.59 (d, J ) 7.1 Hz, 1H), 3.74 (s,
3H), 3.17 (m, 1H), 3.08 (dd, J ) 15.0, 2.6 Hz, 1H), 2.26 (dd, J )
14.9, 9.8 Hz, 1H), 1.28 (d, J ) 6.6 Hz, 3H); HRESIMS m/z 374.0093
(calcd for C14H16Cl3NO3Na, 374.0088).
Biological Assays. The three tropical disease bioassays were
performed as previously described. Plasmodium falciparum malaria
parasites obtained from a chloroquine-resistant P. falciparum strain
(Indochina W2) were maintained and assayed in human erythrocytes.18
Axenic amastigotes of a WHO reference strain of Leishmania donoVani
(LD-1S/MHOM/SD/00-strain 1S) were used to assay for antileishmanial
activity.19 For the Chagas’ assay, a transgenic ꢀ-galactosidase-
expressing Trypanosoma cruzi strain was used (Tulahuen strain, clone
C4).20
Marfey’s Analysis. A portion of compound 1 (0.2 mg) was
hydrolyzed at 120 °C with 6 N HCl for 18 h. A 0.1 M NaHCO3 solution
(100 µL) was added to the dried hydrolysate of 1, as well as to standards
of N-Me-L-phe, N-Me-D-phe, N-Me-D,L-val, and N-Me-L-val. A solution
of 1-fluoro-2,4-dinitrophenyl-5-L-valine-amide (FDVA) in acetone (0.25
mg in 50 µL) was added to each vial. Each vial was sealed and
incubated at 90 °C for 5 min. To quench reactions, 50 µL of 2 N HCl
was added and then diluted with 100 µL of CH3CN. The Marfey’s
derivatives of the hydrolysate and standards were analyzed by HPLC
using a Phenomenex Jupiter C18 column (4.6 × 250 mm). The reaction
mixtures were analyzed starting with 30% CH3CN/70% H2O acidified
with 0.02% TFA followed by a gradient elution profile to 70% CH3CN/
30% H2O acidified with 0.02% TFA over 60 min at a flow of 0.8 mL/
min, monitoring at 340 nm. Retention times for the amino acid standards
were N-Me-L-phe 26.3 min, N-Me-D-phe 27.8 min, N-Me-D,L-val 23.8
and 29.6 min, and N-Me-L-val 23.8 min, while the hydrolysate gave
peaks at 23.8 and 26.3 min.
Prior to Marfey’s analysis of compound 3, ozonolysis was performed
to cleave the thiazole ring. O3(g) (4% in O2, 1/16 L/min) was bubbled
through compound 3 (0.1 mg in 200 µL of CH2Cl2) at -78 °C for 5
min. The solvent was removed using N2 (g), and the ozonate was then
subjected to hydrolysis at 120 °C with 6 N HCl for 18 h. Marfey’s
analysis was undertaken as described above, comparing the ozonolyzed
hydrolysate to the D-val and L-val standards. The Marfey’s derivatives
of the ozonolyzed hydrolysate and standards were analyzed by HPLC
using a Phenomenex Synergi Fusion 4µ column (4.6 × 250 mm). The
reaction mixtures were analyzed starting with 15% CH3CN/85% H2O
acidified with 0.1% HCO2H followed by a gradient elution profile to
60% CH3CN/40% H2O acidified with 0.1% HCO2H over 45 min at a
flow of 0.8 mL/min, monitoring at 340 nm. Retention times for the
amino acid standards were D-val 25.3 min and L-val 31.6 min, while
the ozonolyzed hydrolysate gave a peak at 31.7 min.
Acknowledgment. This research was funded by a Fogarty Interna-
tional Center (FIC) International Cooperative Biodiversity Group
(ICBG) grant based in Panama (ICBG U01 TW006634) and by an FIC
International Research Scientist Development Award (IRSDA) (K01
TW008002, P.I. M.J.B.). T. L. Suyama provided assistance with the
degradation of barbamide and chiral analysis, and J. K. Nunnery assisted
with collection of some physical data. We are grateful to Panama’s
Autoridad Nacional del Ambiente (ANAM) for providing access to
biological samples in Panama. K.T. thanks the Growth Regulation &
Oncogenesis Training Program for a postdoctoral fellowship (NIH T32
CA009523).
Semisynthesis of Barbamide and Herbamide B Fragments and
Their PGME Derivatives. Ozonolysis was performed on barbamide
to cleave the alkene. O3(g) (4% in O2, 1/16 L/min) was bubbled through
12.9 mg of barbamide in 200 µL of CH2Cl2 at -78 °C for 5 min. This
was dried, and a mild hydrolysis was performed using 1 mL of 1.0 N
NaOH and 1 mL of THF at RT overnight (16 h). To stop the reaction,
2 mL of 1 M NaHSO4 was added. This was extracted with diethyl
ether (2 × 5 mL). The organic layer was concentrated and was used
without purification for the next reaction.
The barbamide fragment was split into two portions and reacted
separately with either R-PGME or S-PGME. R-PGME (103.1 mg, 0.51
mmol), Et3N (179 µL, 1.27 mmol), (3-dimethylaminopropyl)ethylcar-
bodiimide hydrochloride (199.4 mg, 1.04 mmol), and 4-dimethylami-
nopyridine (11.4 mg, 93.29 µmol) were added sequentially to 1.7 mg
(8.34 µmol) of the barbamide fragment in CH2Cl2 (10 mL) stirring at
0 °C. This solution was left to warm to room temperature overnight
(18 h) and then concentrated under reduced pressure. The residue was
resuspended in EtOAc (75 mL) and filtered. The filtrate was washed
sequentially with 0.2 M HCl, H2O, 1 M NaHCO3, H2O, and saturated
Supporting Information Available: Table of bioassay results,
1
experimental details of bioassays, H, 13C, HSQC, and HMBC NMR
spectra of 1 and 3, and gradient HPLC profile of compound 1. This
material is available free of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Pink, R.; Hudson, A.; Mouries, M. A.; Bendig, M. Nat. ReV. Drug
DiscoVery 2005, 4, 727–740.
(2) Nwaka, S.; Hudson, A. Nat. ReV. Drug DiscoVery 2006, 5, 941–955.
(3) Capson, T. L.; Coley, P. D.; Kursar, T. A. Nat. Biotechnol. 1996, 14,
1200–1202.
(4) Kursar, T. A.; Capson, T. L.; Coley, P. D.; Corley, D. G.; Gupta,
M. B.; Harrison, L. A.; Ortega-Barr´ıa, E.; Windsor, D. M. Pharm.
Biol. 1999, 37, 114–126.
(5) Coley, P. D.; Heller, M. V.; Aizprua, R.; Arau´z, B.; Flores, N.; Correa,
M.; Gupta, M.; Solis, P. N.; Ortega-Barr´ıa, E.; Romero, L. I.; Go´mez,
B.; Ramos, M.; Cubilla-Rios, L.; Capson, T. L.; Kursar, T. A. Front.
Ecol. EnViron. 2003, 1, 421–428.
(6) Jime´nez, J. I.; Scheuer, P. J. J. Nat. Prod. 2001, 64, 200–203.

66 Journal of Natural Products, 2010, Vol. 73, No. 1
Balunas et al.
(7) McPhail, K. L.; Correa, J.; Linington, R. G.; Gonza´lez, J.; Ortega-
Barr´ıa, E.; Capson, T. L.; Gerwick, W. H. J. Nat. Prod. 2007, 70,
984–988.
(8) Orjala, J.; Gerwick, W. H. J. Nat. Prod. 1996, 59, 427–430.
(9) Nguyen, V. A.; Willis, C. L.; Gerwick, W. H. Chem. Commun.
(Cambridge, U.K.) 2001, 1934–1935.
(10) Flatt, P. M.; O’Connell, S. J.; McPhail, K. L.; Zeller, G.; Willis,
C. L.; Sherman, D. H.; Gerwick, W. H. J. Nat. Prod. 2006, 69,
938–944.
(11) Galonic, D. P.; Vaillancourt, F. H.; Walsh, C. T. J. Am. Chem. Soc.
2006, 128, 3900–3901.
(12) Chang, Z.; Flatt, P.; Gerwick, W. H.; Nguyen, V. A.; Willis, C. L.;
Sherman, D. H. Gene 2002, 296, 235–247.
(13) Gunasekera, S. P.; Ross, C.; Paul, V. J.; Matthew, S.; Luesch, H. J.
Nat. Prod. 2008, 71, 887–890.
(14) Li, H.-y.; Matsunaga, S.; Fusetani, N. J. Nat. Prod. 1996, 59, 163–
166.
(15) Nieto, J.; Veelaert, D.; Derua, R.; Waelkens, E.; Cerstiaens, A.; Coast,
G.; Devreese, B.; Van Beeumen, J.; Calderon, J.; De Loof, A.; Schoofs,
L. Biochem. Biophys. Res. Commun. 1998, 248, 406–411.
(16) Nakao, Y.; Masuda, A.; Matsunaga, S.; Fusetani, N. J. Am. Chem.
Soc. 1999, 121, 2425–2431.
(17) Yokotani, S.; Matsushima, A.; Nose, T.; Morishita, F.; Shimohigashi,
Y. Pept. Sci. 2005, 41, 539–540.
(18) Corbett, Y.; Herrera, L.; Gonza´lez, J.; Cubilla, L.; Capson, T. L.; Coley,
P. D.; Kursar, T. A.; Romero, L. I.; Ortega-Barr´ıa, E. Am. J. Trop.
Med. Hyg. 2004, 70, 119–124.
(19) Williams, C.; Espinosa, O. A.; Montenegro, H.; Cubilla, L.; Capson,
T. L.; Ortega-Barr´ıa, E.; Romero, L. I. J. Microbiol. Methods 2003,
55, 813–816.
(20) Molinar-Toribio, E.; Gonza´lez, J.; Ortega-Barr´ıa, E.; Capson, T. L.;
Coley, P. D.; Kursar, T. A.; McPhail, K.; Cubilla-Rios, L. Pharm.
Biol. 2006, 44, 550–553.
(21) Hooper, G. J.; Orjala, J.; Schatzman, R. C.; Gerwick, W. H. J. Nat.
Prod. 1998, 61, 529–533.
(22) Marner, F.-J.; Moore, R. E.; Hirotsu, K.; Clardy, J. J. Org. Chem.
1977, 42, 2815–2819.
(23) Sone, H.; Nemoto, T.; Ojika, M.; Yamada, K. Tetrahedron Lett. 1993,
34, 8445–8448.
NP900622M