Article pubs.acs.org/jnp
Protease Inhibitors from Microcystis aeruginosa Bloom Material Collected from the Dalton Reservoir, Israel Simi Adiv and Shmuel Carmeli* Raymond and Beverly Sackler School of Chemistry and Faculty of Exact Sciences, Tel-Aviv University, Ramat Aviv, Tel-Aviv 69978, Israel S Supporting Information *
ABSTRACT: Nine new metabolites, aeruginosins DA495A (1), DA511 (2), DA642A (3), DA642B (4), DA688 (5), DA722 (6), and DA495B (7), microguanidine DA368 (8), and anabaenopeptin DA850 (9), were isolated along with the known micropeptins MZ924, MZ939A, and MZ1019, cyanopeptolins S and SS, microcin SF608, and aeruginazoles DA1497, DA1304, and DA1274 from bloom material of the cyanobacterium Microcystis aeruginosa collected from the Dalton reservoir, Israel, in October 2007. Their structures were elucidated by a combination of various spectroscopic techniques, primarily NMR and MS, while the absolute configurations of the asymmetric centers were determined by Marfey’s and chiral-phase HPLC methods. Two of the new aeruginosins, DA511 (1) and DA495A (2), contain a new Choi isomer, (2S,3aS,6S,7aS)-Choi. The structure elucidation and biological activities of the new metabolites are described.
D
type serine proteases (i.e., microcin SF608 inhibits trypsin with an IC50 value of 0.5 μg/mL15). It has been shown that extracts of Microcystis aeruginosa, and other toxic bloom-forming cyanobacteria, are more toxic than the pure microcystins themselves, and the enhanced toxicity is attributed to the presence of the different groups of protease inhibitors (i.e., aeruginosins, micropeptins, anabaenopeptins, and microviridins) in the extracts.16,17 In the past decade attempts have been made to reveal the purpose(s) for the production of the microcystins and the accompanying protease inhibitors in cyanobacteria and to study the relationships between these groups of compounds. Inhibition of trypsin-type proteases and lowering the amounts of active glutathione S-transferase (GST)18 in the water flea Daphnia magna, a cyanobacteria grazer, may imply that the areuginosins are produced in order to prevent the detoxification of the microcystins in the environment.19 As part of our continuing interest in the chemical ecology of cyanobacterial water blooms and the search for novel drugs for human diseases, we examined the extracts of M. aeruginosa bloom material collected in October 2007 from the Dalton reservoir in northern Israel.20 The extract of this bloom material afforded 18 natural products: eight aeruginosins, aeruginosins DA495A (1), DA511 (2), DA642A (3), DA642B (4), DA688 (5), DA722 (6), and DA495B (7) and the known microcin SF608,15 two of which, aeruginosins DA495A (1) and DA511 (2), contain a new Choi isomer, 2S,3aS,6S,7aS-Choi, microguanidine DA368 (8), anabaenopeptin DA850 (9), aeruginazoles DA1497, DA1304, and DA1274,21 and the known micropeptins MZ924, MZ939A, and MZ101922 and cyanopeptolins S23 and SS.24 In this paper,
uring the past three decades, marine, freshwater, and terrestrial cyanobacteria have been documented as prolific producers of biologically active natural products.1−3 Freshwater bloom-forming genera of cyanobacteria have been extensively studied, mainly due to their competence to biosynthesize toxins, which poses a serious threat to the health of humans and domesticated and wild animals.4 The microcystins are the most abundant family of hepatotoxic metabolites found in freshwater bloom-forming cyanobacteria.5 These toxins are usually accompanied by other groups of metabolites, mainly by the following five groups of protease inhibitors: micropeptins (139 isolated variants),6 anabaenopeptins (38 isolated variants),7 aeruginosins (38 isolated variants),8 microginins (32 isolated variants),9 and microviridins (15 isolated variants).10 The aeruginosins are a group of linear modified peptides, which attracted the attention of synthetic chemists due to their modified peptidic nature and their capacity to inhibit important proteolytic processes.11 They are characterized by the presence of the proline-mimicking 2-carboxy-6hydroxyoctahydroindole (Choi) derivative at the third position, a hydroxyphenyl lactic acid (Hpla) derivative at the N-terminus of the modified peptide, a variable amino acid at the second position, and an arginine derivative (if present) at the fourth position. Analyses of the aeruginosin gene-cluster sequences revealed that these linear modified peptides are biosynthesized in cyanobacteria from amino acid precursors by nonribosmal peptide synthetase (NRPS)-type enzyme assembly.12 To date, three natural Choi stereoisomers (out of 16 possible) have been characterized in the literature. The most frequent Choi isomer, 2S,3aS,6R,7aS-Choi (L-Choi), is contained in 33 known compounds (including those isolated in the present research),8 2S,3aR,6R,7aR-Choi (L-3a,7a-diepi-Choi) in one isolated compound,13 and 2R,3aR,6R,7aR-Choi (D-3a,7a-diepi-Choi) in four published compounds.14 The aeruginosins inhibit trypsin© 2013 American Chemical Society and American Society of Pharmacognosy
Received: August 21, 2013 Published: November 21, 2013 2307
dx.doi.org/10.1021/np4006844 | J. Nat. Prod. 2013, 76, 2307−2315
Journal of Natural Products
Article
Chart 1
mid-spectrum (5.5−3.8 ppm) two pairs of exchangeable doublet signals and four pairs of methine protons next to electronegative atoms were observed. The aliphatic region presented many overlapping signals that were too complex to be interpreted. The 13C NMR spectrum (Table 1) revealed a similar doubling of signals corresponding to three pairs of amide/ester carbonyls, a pair of phenol sp2 carbons, 13 signals of aromatic sp2 carbons consistent with a pair of phenol and a pair of phenyl moieties, five pairs of methines in the midrange of the spectrum (73 to 50 ppm), and several additional methine and methylene signals in the aliphatic region. Both rotamers were fully characterized (see Table 1 for the NMR data of the cis rotamer and Tables S1a and S1b in the Supporting Information for the full NMR data of the cis rotamer and the trans rotamer, respectively), but for clarity only the structure elucidation of the major cis rotamer is discussed below. The nature of the cis and trans rotamers of the amide bond between Phe and Choi was distinguished through the NOEs of Phe-H-2 with Choi-H-2 in the cis rotamer and of Phe-H-2 with Choi-H7a in the trans rotamer.25 The assignment of the Phe and Hpla moieties was readily achieved through the interpretation of the data from the COSY, TOCSY, HSQC, and HMBC 2D NMR experiments (Tables 1, S1a, and S1b). The structure elucidation of the 2-carboxyamide-6-hydroxyoctahydroindole (Choiamide) moiety was more challenging due to the overlapping signals in the aliphatic region. Analyses of the COSY and TOCSY correlations assembled the sequence of H-2 through
we report the isolation, structure elucidation, and biological activities of the nine new metabolites isolated from this bloom material.
■
RESULTS AND DISCUSSION Aeruginosins DA495A (1) and DA511 (2) were eluted closely one after the other from a reversed-phase HPLC column. They were isolated as colorless oils with comparable NMR spectra and molecular weights that differ by 16 mass units. Both appeared as ca. 55:45 mixtures, respectively, of two rotamers characteristic of the aeruginosins.25 Full assignment of the 1H and 13C NMR data suggested that 1 contained a phenylalanine moiety at the second position of its short peptide backbone, while 2 contained a tyrosine moiety, and the rest of their structures were identical. Aeruginosin DA495A (1) exhibited a positive HRESIMS molecular adduct ion at m/z 518.2266 ([M + Na]+), in agreement with the molecular formula C27H33N3NaO6 and 13 degrees of unsaturation. The interpretation of the NMR data in DMSO-d6 was rather complicated due to the high similarity of the chemical shifts of the proton and carbon signals of the above-mentioned rotamers. The aromatic region of the 1H NMR spectrum of 1 (Table 1) presented pairs of phenol signals and three amide protons (two pairs of primary amide singlet signals and a pair of doublet secondary amides), overlapping signals indicating the presence of a phenyl moiety, and doublet signals of two pairs of para-substituted phenol moieties. In the 2308
dx.doi.org/10.1021/np4006844 | J. Nat. Prod. 2013, 76, 2307−2315
Journal of Natural Products
Article
Table 1. NMR Data of the Major cis Rotamer of Aeruginosins DA495A (1)a and DA511 (2)b in DMSO-d6 aeruginosin DA495A (1) position
a
δC, mult.c
Hpla 1 2 3
173.4, C 72.5, CH 39.9, CH2
4 5,5′ 6,6′ 7 2-OH 7-OH Phe 1 2 3
128.7, 130.3, 115.0, 155.8,
4 5,5′ 6,6′ 7 NH
138.3, 129.6, 128.2, 126.4,
Choi 1 amide 2 3
174.3, C 59.6, CH 33.3, CH2
C CH d C
aeruginosin DA511 (2) δH, mult., J (Hz)
position
3.85, ddd (8.8, 6.1, 3.8) 2.60, dd (14.0, 3.6) 2.29, dd (14.0, 8.5) 6.84, d (8.4) 6.59, d (8.4) 5.29, d (6.3) 9.07, s
170.1, C 51.7, CH 36.7, CH2 C CH CH CH
3a 4
32.8, CH 22.8, CH2
5
29.9, CH2
6 7
67.1, CH 36.3, CH2
7a 6-OH NH2(a) NH2(b)
56.7, CH
4.28, dt (2.5, 8.4) 2.88, m 2.77, m 7.19, 7.22, 7.16, 7.71,
m m m d (8.3)
4.78, 2.32, 1.77, 2.23, 1.62, 1.52, 1.60, 1.21, 3.29, 2.34, 0.84, 4.04, 4.49, 7.73, 7.32,
d (9.0) m dd (12.2, 5.7) m m m m m m m q (12.0) ddd (12.0, 6.3, 6.3) d (4.3) s s
δC, mult.c
Hpla 1 2 3
173.2, C 72.4, CH 39.9, CH2
4 5,5′ 6,6′ 7 2-OH 7-OH Tyr 1 2 3
128.7, 130.2, 114.9, 155.8,
4 5,5′ 6,6′ 7 NH 7-OH Choi 1 amide 2 3
128.7, 130.3, 114.9, 155.7,
C CH d C
δH, mult., J (Hz) 3.84, ddd (9.7, 6.0, 3.7) 2.61, m 2.28, dd (14.0, 9.0) 6.86, d (8.4) 6.61, m 5.29, d (6.0) 9.09, s
170.1, C 51.5, CH 37.0, CH2 C CH CH C
4.19, m 2.79, m 2.63, m 6.99, m 6.62, m 7.64, d (8.0) 9.06, s
174.2, C 59.4, CH 33.2, CH2
3a 4
32.7, CH 22.7, CH2
5
29.8, CH2
6 7
67.0, CH 35.9, CH2
7a 6-OH NH2(a) NH2(b)
56.3, CH
4.76, 2.35, 1.75, 2.25, 1.58, 1.62, 1.56, 1.20, 3.28, 2.37, 0.83, 4.01, 4.51, 7.73, 7.31,
d (8.8) m dd (11.6, 5.5) m m m m m m m (eq) q (11.8) ddd (11.8, 6.7, 6.1) d (4.1) s s
500 MHz for 1H, 125 MHz for 13C. b400 MHz for 1H, 100 MHz for 13C. cMultiplicities and assignments from HSQC experiment.
nitrogen was out of the ring plane, thus explaining the multiplicity of H-2 (doublet of 9.0 Hz). A comparison of the NMR data of Choi in 1 with those of microcin SF608 (containing L-Choi)15 and aeruginosin KT608A (containing D3a,7a-diepi-Choi)14 revealed the similarity of the Choi moiety in 1 with those of the D-3a,7a-diepi-Choi moiety of aeruginosin KT608A. This suggested that the relative configuration of the Choi moiety of 1 was consistent with either 2S,3aS,6S,7aS-Choi or 2R,3aR,6R,7aR-Choi (L-6-epi-Choi or D-3a,7a-diepi-Choi, respectively). The sequence of the subunits that compose 1, Hpla-Phe-Choi-amide, was based on the HMBC correlation of Hpla-CO with Phe-NH, the NOE correlation of Phe-H-2 and Choi-H-2, and the HMBC correlations of Choi-CO with the primary amide protons. These arguments established the relative configuration of the Choi moiety and the planar structure of 1. The application of Marfey’s method27 (L-FDAA) and chiral-phase HPLC to determine the absolute configurations of the amino acids and the hydroxy acid, respectively, revealed the presence of D-Phe and D-Hpla moieties in 1. The retention time of Choi-L-FDAA (35.1 min) from 1 was identical with that of the synthetic L-6-epi-Choi-L-FDAA (35.1
H-7a, while the correlations in the HSQC spectrum assigned the carbons directly attached to these protons (Tables 1 and S1a). H-2 was correlated to the adjacent protons resonating at δH 2.35 (a 9.0 Hz coupling constant) and 1.77 (
