Halogenated Fatty Acid Amides from Lyngbya majuscula
Journal of Natural Products, 2009, Vol. 72, No. 9 1577
Grenadamide C (2): light yellow oil; [R]23D -17 (c 0.40, CH2Cl2);
UV (MeOH) λmax (log ꢀ) 207 nm (3.84); IR νmax (film on NaCl) 2962,
2932, 2872, 1709, 1641, 1546, 1460, 1378, 1265, 738 cm-1; 1H (CDCl3,
500 MHz) and 13C (CDCl3, 125 MHz) NMR data (Table 1); ESIMS/
MS (m/z 404.2 fragmented using 50% collision energy) m/z 368, 232,
215; ESI-TOF-MS m/z 404.2112, [M + H]+ (calcd for C21H3635Cl2NO2,
404.2118).
order to resolve the FDLA-Met(O2) products that co-eluted under the
above conditions, the derivatized hydrolysate and standards were
similarly chromatographed on a second reversed-phase column [Luna
C18, Phenomenex, 5 µm, 4.6 × 250 mm; MeCN/H2O gradient with
0.10% formic acid: 20:80 to 90:10 (0-40 min), 90:10 (40-45 min);
0.8 mL/min]. For authentic carriebowmide (5), only the configuration
of the AMHA residue was analyzed. As determined by reconstructed
SIC of [M + H]+ ions, the retention times of the L-FDLA-amino acids
in each hydrolysate reaction product were compared with retention times
of standard FDLA-amino acids (see Supporting Information).
Advanced Marfey’s Analysis of MePhe-Ala1 from the Hydroly-
sate of 3 and MePhe-Val1 and MeAla1-Val2 from the Hydrolysate
of 4. The dipeptide MePhe-Ala1 was isolated from the hydrolysate of
3, and the dipeptides MePhe-Val1 and MeAla1-Val2 were isolated from
that of 4, both by reversed-phase HPLC-PDA [Waters Atlantis dC18 5
µm, 3.0 × 250 mm; MeCN/H2O gradient with 0.1% formic acid: 0:100
(0-10 min), 0:100 to 50:50 (10-23 min); 0.3 mL/min]. Each dipeptide
was hydrolyzed for 72 h in 0.4 mL of 6 N HCl at 110 °C. Workup of
the hydrolysate product, preparation of L-FDLA derivatives, and LC-
ESIMS analyses of the Marfey’s products were conducted as described
above (see Supporting Information).
Cytotoxicity Testing. Resazurin is a nonfluorescent dye that is
reduced by viable cells to the fluorescent resorufin. The protocol
used was adapted from O’Brien and co-workers.31 Human embryonic
kidney (HEK-293) cells were grown in culture flasks with DMEM
media (Dulbecco’s modified Eagle’s medium, Cellgro), supple-
mented with 10% heat-inactivated fetal bovine serum (FBS), 2.0
mM L-glutamine, 100 U/mL penicillin, and 0.10 mg/mL strepto-
mycin, in a humidified incubator (5% CO2 at 37 °C). The cells were
plated (6000 cells/well) on poly-L-lysine-coated black/clear-bottom
96-well plates (Greiner). Following the addition of test substances
or vehicle, plates were incubated in growth media for 48 h, at which
time media was completely removed, and wells were washed with
Krebs Ringer Hepes (KRH) buffer (consisting of 1.5 mM CaCl2,
5.6 mM glucose, 20 mM Hepes, 5 mM KCl, 1.5 mM MgCl2, and
135 mM NaCl) and incubated for 2.5 h (5% CO2 at 37 °C) in 13
µM resazurin (Sigma) in KRH. The production of resorufin was
measured using a scanning fluorometer (FLEXstation 3, Molecular
Devices; 9 points/well; emission read at 590 nm following excitation
at 530 nm). Controls included wells receiving 0.1% saponin (MP
Biomedicals), which serve as an “all dead” (background) and wells
receiving 400 nM daunorubicin (Sigma) as a positive control.
Duplicate serial dilutions for each compound were tested on two
separate days. Data were normalized by expressing compound-treated
well fluorescence as a percent of vehicle control (% vehicle) after
subtracting average fluorescence of background (saponin-treated)
wells from vehicle control and compound-treated wells. For each
serial dilution, IC50 values were determined from dose-response
curves fitted with Prism statistical software (GraphPad Software,
version 5.01; log of inhibitor concentration vs normalized response,
variable slope). Independent IC50 values were averaged and are
reported in Table 3.
Itralamide A (3): colorless oil; UV [PDA, MeOH/formic acid (1000:
1
1)] λmax 219 nm; H (CDCl3, 500 MHz) NMR data and 13C chemical
shifts derived from HMBC and HSQC (CDCl3) NMR data (Table 2);
ESIMS/MS (m/z 769.4 fragmented using 33% collision energy) m/z
737, 733, 680, 519, 434, 250; ESI-TOF-MS m/z 769.3457, [M + H]+
(calcd for C36H55Cl2N6O8, 769.3458).
Itralamide B (4): colorless oil; UV [PDA, MeOH/formic acid (1000:
1
1)] λmax 219 nm; H (CDCl3, 500 MHz) NMR data and 13C chemical
shifts derived from HMBC and HSQC (CDCl3) NMR data (Table 2);
ESIMS/MS (m/z 797.4 fragmented using 33% collision energy) m/z
765, 761, 720, 680, 519, 434; ESI-TOF-MS m/z 797.3772, [M + H]+
(calcd for C38H59Cl2N6O8, 797.3771).
Preparation of N-Methylthreonine Standards. A mixture of
N-methylthreonine and N,N-dimethylthreonine was prepared using the
method from Bowman and Stroud.28 Separately L-threonine (0.84
mmol) and L-allo-threonine (0.43 mmol) were dissolved in distilled
water. To each solution were added 10% Pd/C (0.10 and 0.05 g,
respectively) and 38% (w/w) formaldehyde (0.14 and 0.07 mL,
respectively). The reaction solutions were bubbled with H2 (g) for
1.5-2.5 h, while the reaction progress was monitored using direct
injection ESIMS. Once the concentration of the monomethyl derivative
plateaued, the solutions were filtered and dried under vacuum. The
desired monomethylated product was separated from starting material
and dimethylated product by reversed-phase HPLC (Atlantis dC18,
Waters, 3.0 × 250 mm) using 0.10% aqueous formic acid as the mobile
phase. The resulting N-methyl-L-threonine (98% purity by ELSD)
demonstrated 1H NMR and 13C NMR signals consistent with literature
values.29 MS analysis confirmed the identification: ESI-TOF-MS m/z
134.0805 [M + H]+ (calcd for C5H12NO3, 134.0812); ESIMS/MS (m/z
134.1 fragmented using 50% collision energy) m/z 98, 88, and 70. The
specific rotation was [R]27 -12 (c 0.093 6 N HCl). The resulting
D
N-methyl-L-allo-threonine (91% purity by ELSD) gave the following
NMR data: 1H NMR (D2O, 300 MHz) δ 4.34 (1H, m, H-3); 3.63 (1H,
d, J ) 3.7 Hz, H-2); 2.75 (3H, s, NMe); 1.21 (3H, d, J ) 6.7 Hz, H-4);
13C NMR (D2O, 300 MHz) δ 171.5 (C-1); 69.3 (C-2); 65.8 (C-3); 33.5
(NMe); 17.5 (C-4). MS analysis confirmed the identification: ESI-TOF-
MS m/z 134.0801 [M + H]+ (calcd for C5H12NO3, 134.0812); ESIMS/
MS (m/z 134.1 fragmented using 50% collision energy) m/z 98, 88,
70. The specific rotation was [R]28 +13 (c 0.10 6 N HCl).
D
Acid Hydrolysis of 3-6. Itralamides A (3, 50 µg) and B (4, 21
µg), authentic carriebowmide (5, 75 µg), and carriebowmide sulfone
(6, 200 µg) were hydrolyzed in 0.4 mL of 6 N HCl (Pierce, 24308) at
110 °C for 18.5 h. The samples were dried under a N2 stream at 60 °C,
then twice reconstituted with 100 µL of H2O and redried. The
hydrolysate was eluted through a C18 SPE column (Phenomenex, Strata
C18-E) using MeOH/H2O (10:90, v/v) to yield free amino and hydroxy
acids.
Beet Armyworm Activity (BAW: Spodoptera exigua). Six repli-
cates of each of 1 and 2 were overlaid in 15% EtOH on top of a wheat
germ/casein-based artificial diet in 96-well plates. The screening diet
was augmented with antibiotics to control bacterial overgrowth.
Template-inoculated filter papers containing temporally synchronized
armyworm eggs were then matched to the well orifice of the diet/extract
plate, and both were covered with a ventilated plate lid. The eggs
hatched and the neonates dropped to the treated diet. After five days
the percent mortality was determined in comparison with control
(vehicle) wells. A positive control was included as serial dilutions of
Javelin (Thermo Trilogy’s Bacillus thuringiensis).
Advanced Marfey’s Method for the Analysis of the Acid
Hydrolysates of 3-6. The acid hydrolysates of 3-6 were reconstituted
with 90, 40, 80, and 100 µL of H2O, respectively. One-half volume of
each resulting solution was reacted with 30/20/40/50 µL of aqueous
1.0 M NaHCO3 and 30/20/40/50 µL of a 0.20% (w/v) solution of
L-FDLA in acetone, respectively. The reaction mixtures were heated
at 80 °C for 3 min and allowed to cool. The reaction solutions for 3-6
were then acidified with 60/40/80/100 µL of 2.0 N HCl and diluted
with 30/10/30/250 µL of MeCN, respectively. Standard FDLA deriva-
tives were prepared for each constituent R-amino acid by reacting 50
µL aliquots of a 5 mM aqueous solution of the standard L-isomer using
separately both L- and DL-FDLA reagents. Standard FDLA derivatives
of AMHA were prepared similarly by separately reacting 50 µL of 3.5
mM solutions of the (2R,3R)- and (2S,3R)-diastereomers with both L-
and DL-FDLA reagents. (The AMHA standards were reported in a
previous study30 and provided by Dr. Philip Williams, University of
Hawaii.) Each reaction product mixture was chromatographed by
reversed-phase HPLC (Eclipse XDB-18, Agilent, 4.6 × 150 mm) using
a gradient of MeCN/H2O with 0.10% formic acid [20:80 to 80:20 (0-30
min), 80:20 (30-35 min); 0.8 mL/min) and detection by ESIMS. In
Acknowledgment. This publication was made possible by NIH Grant
P20 RR-16467 from the National Center for Research Resources and
by the HPU Trustees’ Scholarly Endeavors Fund. Funds for the upgrade
of the Varian Unity INOVA 500 were provided by the CRIF program
of the National Science Foundation (CH E9974921) and the Elsa U.
Pardee Foundation. The purchase of the Agilent LC-TOF-MS was
funded by Grant W911NF-04-1-0344 from the Department of Defense.
We thank Dr. G. M. L. Patterson for the taxonomic identification, Dr.
P. Williams for providing FDLA reagent and reference AMHA samples,