352
J. Nat. Prod. 2010, 73, 352–358
Micropeptins from an Israeli Fishpond Water Bloom of the Cyanobacterium Microcystis sp.† Ella Zafrir and Shmuel Carmeli* Raymond and BeVerly Sackler School of Chemistry and Faculty of Exact Sciences, Tel-AViV UniVersity, Ramat AViV, Tel-AViV 69978, Israel ReceiVed September 6, 2009
Seven new natural products, micropeptin MZ845 (1), micropeptin MZ859 (2), micropeptin MZ939A (3), micropeptin MZ925 (4), micropeptin MZ939B (5), micropeptin MZ1019 (6), and micropeptin MZ771 (7), as well as two known micropeptins, cyanopeptolin S (8) and cyanopeptolin SS (9), were isolated from the hydrophilic extract of the cyanobacterium Microcystis sp. that was collected from a fishpond in Kibbutz Ma’ayan Tzvi, Israel, in July 2006. The structures of the pure natural products were elucidated using spectroscopic methods, including UV, 1D and 2D NMR, and MS techniques. The absolute configuration of the chiral centers of the compounds was determined using Marfey’s method for HPLC. The inhibitory activity of the compounds was determined for the serine proteases: trypsin, chymotrypsin, thrombin, and elastase. These micropeptins inhibited trypsin with IC50’s that varied between 0.6 and 24.2 µM. The SAR of these micropeptins is discussed. Cyanobacteria produce numerous natural products that are not used for primary metabolism, and their role in the producing organism is usually unknown. Microcystis spp. are responsible for water blooms that frequently produce potent hepatotoxins, and thus their environmental, toxicological, biological, and chemical properties have been extensively studied.1 Extracts of toxic strains of Microcystis have been shown to be a rich source of unique modified peptides, such as microginins,2 micropeptins,3 aeruginosins,4 anabaenopeptins,5 and microviridines,6 all of which are protease inhibitors. The micropeptins are the most abundant group of serine protease inhibitors from cyanobacteria, composed of more than 100 different members.7 The selectivity of the micropeptins for the inhibition of different serine proteases is mainly influenced by the nature of the amino acid occupying the fifth position from the carboxy terminus. Basic amino acids (i.e., arginine and lysine) at this position select for inhibition of trypsin-like serine proteases, while aliphatic, aromatic, and other neutral amino acids select for inhibition of chymotrypsin-like serine proteases.8 As part of our ongoing research on the chemistry and chemical ecology of cyanobacterial blooms in water bodies, a biomass (TAU strain IL361) of the cyanobacterium Microcystis sp. was collected, in the summer of 2006, from a fishpond in Kibbutz Ma’ayan-Tzvi, Israel. The isolation and structure elucidation of micropeptins isolated from this cyanobacterial bloom biomass, as well as their biological activity and some SAR, is discussed below.
Results and Discussion The cyanobacterium biomass was freeze-dried and extracted with 70% MeOH in H2O. The extract that inhibited trypsin was flashchromatographed on an ODS column. Two fractions eluted from the column with 40% and 50% MeOH in H2O exhibited protease inhibitory activity and were further separated on a reversed-phase HPLC column. Seven new protease inhibitors, micropeptin MZ845 (1) (first identified by Martin Welker at al.9 using an HPLC-MS method), micropeptin MZ859 (2), micropeptin MZ939A (3), micropeptin MZ925 (4), micropeptin MZ939B (5), micropeptin MZ1019 (6), and micropeptin MZ771 (7), along with two previously described natural products, cyanopeptolin S10 (8) and cyanopeptolin SS11 (9), were isolated from the cyanobacterial extract, their structures were elucidated, and their biological activities were studied. † Dedicated to the late Dr. Richard E. Moore of the University of Hawaii at Manoa for his pioneering work on bioactive natural products. * To whom correspondence should be addressed. Tel: ++972-3-6408550. Fax: ++972-3-6409293. E-mail:
[email protected].
10.1021/np900546u
Inspection of the NMR spectroscopic data of 1-7 reveals signals indicative of the micropeptin family. In the 1H NMR spectrum a methyl singlet at ca. 2.70 ppm of an NMe moiety, a hydroxy of the amino hydroxy piperidone (Ahp) moiety at 6.10 ppm, a quartet signal at 5.40 ppm, and a doublet methyl at 1.21 ppm of the O-substituted Thr were all visible. Comparison of the 1H NMR spectra of all of the nine members of this micropeptin family revealed their similarity: a doublet methyl (IIIle-6) at ca. -0.25 ppm, a triplet amide proton signal at ca. 7.50 ppm (Arg-δ-NH), a very broad signal of the guanidine moiety of Arg around 6.50-7.00 ppm, and overlapping signals of a monosubstituted phenyl moiety in the aromatic region (7.15-7.30 ppm) of the spectrum. The 13C NMR spectrum revealed some other characteristic signals including a guanidine carbon at 156 ppm of the Arg moiety and three (two in the case of 7) methine carbons adjacent to oxygen at around 73 ppm, in accordance with Ahp, Thr, and glyceric acid. The NMR data suggested a structure composed of the same seven acid units (except for 7, which contains six acid units). The differences between the nine micropeptins were located at the amino hydroxy piperidone moiety and in the substitutions of the glyceric acid moiety. Five of the new micropeptins had an amino methoxy piperidone (Amp) moiety instead of an Ahp moiety, which was characterized by the appearance of a singlet methoxy group at 3.00 ppm, instead of the hydroxy group at 6.10 ppm. One of the micropeptins (7) was truncated and did not contain the glyceric acid moiety, while the remaining eight micropeptins differed in the substitution pattern of the glyceric acid (GA) by sulfate groups; two of the compounds possessed a sulfate substituent on position 2 of the glyceric acid. Such derivatives have not yet been described in the literature. Micropeptin MZ845 (1) was isolated as an amorphous, white solid. The molecular formula of 1, C40H63N9O11, was deduced from the high-resolution MALDI-TOF-MS measurements of its protonated molecular cluster ion at m/z 846.4641. Examination of the NMR spectra of 1 in DMSO-d6 revealed its peptide nature, i.e., seven carboxylic carbons in the 13C NMR spectrum (see Table 2) and four amide doublet protons in the 1H NMR spectrum (see Table 1). Taking into account the NMe-aromatic amino acid and the N,N-disubstituted-amino acid of the micropeptins, this micropeptin was composed of six amino acids. The Ahp and glyceric acid structure was suggested on the basis of the COSY and HSQC spectra, which showed a singlet of a hydroxy group at 6.20 ppm of the Ahp moiety and two hydroxy signals at 4.76 ppm (t) and 5.80 ppm (brd) for GA. Analysis of the COSY, TOCSY, and HSQC 2D NMR spectra allowed the assignment of the side chains of two isoleucine units, a threonine, an arginine, a glyceric acid, and two
2010 American Chemical Society and American Society of Pharmacognosy Published on Web 12/22/2009
Micropeptins from the Cyanobacterium Microcystis sp.
Journal of Natural Products, 2010, Vol. 73, No. 3 353
Chart 1
Table 1. Comparison of the 1H NMR Data of Compounds 1-7 in DMSO-d6a position I
Ile
NMePhe
II
Ile
Ahp/Amp
Arg
Thr
GA
a
2 3 4a 4b 5 6 NH 2 3a 3b 5,5′ 6,6′ 7 NMe 2 3 4a 4b 5 6 3 4pax 4peq 5peq 5pax 6 NH OH/OMe 2 3a 3b 4 5 5-NH NH 2 3 4 NH 2 3a 3b 2-OH 3-OH
1c
2b
3c
4b
5c
6c
7c
δH, mult.
δH, mult.
δH, mult.
δH, mult.
δH, mult.
δH, mult.
δH, mult.
4.57, dd 1.74, m 1.27, m 1.06, m 0.80, t 0.83, d 7.77, d 5.12, dd 3.29, m 2.79, dd 7.26, d 7.22, d 7.19, m 2.73, s 4.37, d 1.74, m 1.07, m 0.60, m 0.60, brd -0.25, d 4.44, m 2.61, m 1.75, m 1.75, m 1.75, m 4.91, brs 7.32, d 6.20, brd 4.29, m 2.02, m 1.47, m 1.47, m 3.08, brq 7.53, t 8.60, d 4.66, d 5.52, q 1.21, d 7.64, d 4.07, q 3.60, m 3.49, m 5.80, brd 4.76, t
4.58, dd 1.72, m 1.34, m 1.13, m 0.84, t 0.87, d 7.00, d 5.16, dd 3.30, m 2.80, dd 7.26, m 7.22, m 7.19, m 2.73, s 4.42, d 1.80, m 1.05, m 0.59, m 0.59, brd -0.28, d 4.47, m 2.43, dd 1.74, m 2.10, m 1.67, m 4.44, m 7.19, m 3.03, s 4.29, m 2.00, m 1.45, m 1.45, m 3.08, m 7.45, t 8.63, d 4.68, d 5.54, q 1.21, d 7.66, d 4.07, m 3.60, m 3.48, m 5.80, d 4.76, t
4.60, dd 1.73, m 1.33, m 1.11, m 0.84, t 0.86, d 6.95, d 5.15, dd 3.32, m 2.79, dd 7.24, m 7.21, m 7.17, m 2.73, s 4.42, d 1.76, m 1.05, m 0.58, m 0.58, brd -0.29, d 4.46, m 2.42, dd 1.74, m 2.06, m 1.66, m 4.43, m 7.15, d 3.02, s 4.31, m 2.05, m 1.42, m 1.41, m 3.07, m 7.44, t 8.58, d 4.65, d 5.50, q 1.20, d 7.64, d 4.24, dd 3.97, dd 3.84, dd 6.09, brs
4.72, dd 1.76, m 1.26, m 1.00, m 0.81, t 0.84, d 7.61, d 5.13, dd 3.29, m 2.81, dd 7.25, m 7.23, m 7.20, m 2.73, s 4.38, d 1.74, m 1.05, m 0.60, m 0.60, brd -0.26, d 4.44, m 2.59, m 1.73, m 1.73, m 1.73, m 4.91, brs 7.30, d 6.06, d 4.28, m 2.02, m 1.46, m 1.46, m 3.06, m 7.46, t 8.53, brs 4.63, m 5.46, q 1.22, d 7.83, d 4.65, m 3.63, dd 3.63, dd
4.70, dd 1.78, m 1.34, m 1.11, m 0.85, t 0.86, d 6.89, d 5.17, dd 3.30, m 2.81, dd 7.25, m 7.22, m 7.19, m 2.74, s 4.44, m 1.79, m 1.07, m 0.60, m 0.60, brd -0.27, d 4.44, m 2.40, dd 1.74, m 2.04, m 1.67, m 4.44, m 7.17, d 3.02, s 4.26, m 2.00, m 1.47, m 1.44, m 3.06, m 7.47, t 8.56, d 4.63, d 5.48, q 1.22, d 7.83, d 4.65, d 3.63, m 3.63, m
4.70, m 1.78, m 1.33, m 1.06, m 0.84, m 0.86, d 6.88, d 5.16, d 3.30, m 2.81, dd 7.25, m 7.23, m 7.20, m 2.74, s 4.44, m 1.80, m 1.06, m 0.60, m 0.59, brd -0.29, d 4.47, m 2.40, m 1.74, m 2.07, m 1.66, m 4.43, m 7.17, d 3.02, s 4.30, m 2.02, m 1.43, m 1.43, m 3.07, m 7.44, t 8.53, d 4.63, d 5.46, q 1.21, d 7.75, d 4.72, m 3.99, m 3.91, dd
4.85, dd 1.77, m 1.30, m 1.10, m 0.84, t 0.87, d 6.89, d 5.16, dd 3.29, m 2.82, dd 7.26, m 7.21, m 7.20, m 2.74, s 4.44, m 1.81, m 1.05, m 0.59, brd 0.59, brd -0.27, d 4.45, m 2.33, m 1.77, m 2.10, m 1.65, m 4.45, m 7.24, m 3.01, s 4.36, m 2.06, m 1.45, m 1.46, m 3.10, m 7.49, m 8.93, d 4.14, brs 5.60, q 1.40, d 8.40, brs
4.77, t
4.73, dd
Complete NMR data are available in the Supporting Information (coupling constants, HMBC and ROESY correlations). b 500 MHz. c 400 MHz.
short fragments in accordance with a phenylalanine moiety and an amino hydroxy piperidone (Ahp). The structure of the side chains of the latter two amino acids and the assignment of the carboxyamide carbons to the side chains were achieved by analysis of the results of a 1H-13C HMBC experiment (see Table 4 in the Supporting Information). The sequence of the amino acids of the peptide, IIle, NMePhe, IIIle, Ahp, Arg, Thr, and GA was assigned on the basis of HMBC correlations between the carboxyl of NMePhe and IIle NH, the carboxyl of IIIle and the NMe of NMePhe,
the C-6 methine of Ahp and H-2 of the IIIle, the carboxyl of Arg and NH of Ahp, the carboxyl of Thr and NH of Arg, the carboxyl of IIle and H-3 of Thr (the lactone linkage), and the carboxyl of GA and the NH of Thr. The amino acid sequence could also be assembled from the ROESY data (see Table 4 in the Supporting Information). The relative configuration of Ahp-C-6 (R*) is based on the J-values of H-6 (
