Article pubs.acs.org/jnp
Haliclonadiamine Derivatives and 6-epi-Monanchorin from the Marine Sponge Halichondria panicea Collected at Iriomote Island Delfly B. Abdjul,†,‡ Hiroyuki Yamazaki,*,† Syu-ichi Kanno,† Ohgi Takahashi,† Ryota Kirikoshi,† Kazuyo Ukai,† and Michio Namikoshi† †
Faculty of Pharmaceutical Sciences, Tohoku Medical and Pharmaceutical University, Aoba-ku, Sendai 981-8558, Japan Faculty of Fisheries and Marine Science, Sam Ratulangi University, Kampus Bahu, Manado 95115, Indonesia
‡
S Supporting Information *
ABSTRACT: Four new haliclonadiamine analogues, (10Z,12E)haliclonadiamine (1), (10E,12Z)-haliclonadiamine (2), and halichondriamines A (3) and B (4), were isolated from the Okinawan marine sponge Halichondria panicea together with haliclonadiamine (5) and papuamine (6). The structures of 1−4 were elucidated on the basis of their spectroscopic data by comparisons with those for 5 and 6. Further separation of the remaining fraction led to the isolation of a new bicyclic guanidine alkaloid, 6-epi-monanchorin (7), along with monanchorin (8). Compound 7 is the epimer of 8 at the 6 position. Compounds 1−6 inhibited the growth of Mycobacterium smegmatis with inhibition zones of 12, 7, 8, 7, 16, and 12 mm at 10 μg/disc, respectively. Compounds 2−4 exhibited weak cytotoxicities against the Huh-7 (hepatoma) human cancer cell line and were 2fold less active than 5 and 6. Compounds 7 and 8 were not active against M. smegmatis at 20 μg/disc or the cancer cell line at 10 μM.
M
column) to yield compounds 1 (3.9 mg), 2 (11.9 mg), 3 (10.3 mg), 4 (1.7 mg), 5 (21.7 mg), 6 (18.6 mg), 7 (2.0 mg), and 8 (2.5 mg). Compounds 5, 6, and 8 were identified by comparing their spectroscopic data with those of the reported values for haliclonadiamine,5,8 papuamine,6 and monanchorin,7,9 respectively. Compound 1 was obtained as a yellow oil. HREIMS and NMR data for 1 (Table 1) deduced the molecular formula as C25H40N2, which was the same as those for 5 and 6. The 1H and 13C NMR spectra of 1 resembled those of 5 and 6, and a 2D NMR analysis (COSY, HMQC, and HMBC) indicated that compound 1 possessed the same pentacyclic diamine moieties as 5 and 6. NOESY correlations between H-1 (δH 3.34)/H-8 (1.06), H-1/H-10 (5.48), H-3 (1.55)/H-9 (2.99), H-12 (6.77)/ H-15 (1.79), H-14 (2.51)/H-20 (1.22), and H-14/H-22 (3.68) revealed the relative configurations of the A/B and C/D rings (Figure 1), which were identical to those of haliclonadiamine (5).5 Structural differences between 1 and 5 were detected in the coupling constants of olefinic proton signals (H-10/H-11 and H-12/H-13) in their 1H NMR spectra. Haliclonadiamine (5), the (10E, 12E)-isomer, showed the coupling constants of J10,11 = 14.4 Hz and J12,13 = 14.4 Hz, while J10,11 and J12,13 of 1 were 10.1 and 16.7 Hz, respectively. On the basis of the NOE data for 1, the most stable conformer was calculated using
arine sponges are a rich source of unique metabolites including nitrogen-containing substances with potent biological activities and high chemical diversities.1 Sponges of the genus Halichondria have provided many structural types of bioactive compounds to date.2 An antitumor agent for the treatment of breast cancer, eribulin, was recently developed from halichondrin B, a lead compound that was isolated from Halichondria okadai.3 In the course of our search for antimycobacterial substances from marine invertebrates and microorganisms, we have reported new streptcytosines and agelasine derivatives.4 In further investigations on the extracts of marine sponges, we found that the EtOH extract of the Okinawan Halichondria panicea exhibited antibacterial activity against Mycobacterium smegmatis with an inhibition zone of 10 mm at 50 μg/disc. Bioactivity-guided separation led to the isolation of four new alkaloids, named (10Z,12E)-haliclonadiamine (1), (10Z,12E)haliclonadiamine (2), and halichondriamines A (3) and B (4), together with two known analogues, haliclonadiamine (5)5 and papuamine (6),6 and a new bicyclic guanidine alkaloid, 6-epimonanchorin (7), along with a known compound, monanchorin (8).7 We herein describe the isolation, structure elucidation including absolute configurations, and biological activities of new compounds 1−4 and 7.
■
RESULTS AND DISCUSSION The EtOH extract of the marine sponge was separated by solvent extraction, an ODS column, and repeated HPLC (ODS © XXXX American Chemical Society and American Society of Pharmacognosy
Received: February 2, 2016
A
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Table 1. 1H (400 MHz) and 13C (100 MHz) NMR Data for (10Z,12E)-Haliclonadiamine (1) and (10E,12Z)Haliclonadiamine (2) in CD3OD 1
Spartan’14 by a Monte Carlo conformational analysis with an MMFF94 force field as shown in Figure 1. Consequently, the structure of 1 was assigned and named (10Z,12E)-haliclonadiamine. The molecular formula of compound 2, C 25 H 40 N 2 , determined from HREIMS, was the same as that of 1. An analysis of the 1D and 2D NMR spectra of 2 (Table 1) revealed that compound 2 had the same planar structure as 1. NOESY correlations between H-1 (δH 4.24)/H-8 (1.04), H-1/H-11 (6.51), H-3 (1.54)/H-9 (2.29), H-9/H-10 (6.25), H-13 (5.42)/ H-15 (1.24), H-14 (2.85)/H-20 (1.24), H-14/H-22 (4.29), and H-20/H-22 established the relative configurations of the A/B and C/D rings (Figure 2). In the 1H NMR spectrum of 2, the coupling constants of the olefinic protons (H-10/H-11 and H12/H-13) were J10,11 = 15.7 Hz and J12,13 = 10.4 Hz. The most stable conformer of 2 (Figure 2) was predicted based on the NOE data for 2 using Spartan’14 by a Monte Carlo conformational analysis with an MMFF94 force field. Thus, the structure of 2 was elucidated and named (10E,12Z)haliclonadiamine. The molecular formula of halichondriamine A (3) was decided as C22H36N2 from HREIMS and NMR data (Table 2), which was 40 Da (C3H4) less than those of 1 and 2. Three methylene signals assigned to the 24−26 positions in 1 and 2 were not detected in the 1H and 13C spectra of 3. The 2D NMR spectra of 3 confirmed that two bicyclic moieties were connected by a conjugated diene. NOESY correlations between H-1 (δH 3.38)/H-10 (5.28), H-3 (1.59)/H-9 (2.68), H-8 (1.03)/H-10, H-11 (6.27)/H-13 (5.63), H-12 (6.52)/H-14 (2.46), H-13/H-15 (1.25), and H-14/H-20 (1.28) revealed the relative configurations of the A/B and C/D rings in 3 (Figure 3). The configurations at C-10 and C-12 were determined as (10Z, 12E) based on the coupling constants of the olefinic proton signals (J10,11 = 10.7 Hz and J12,13 = 15.0 Hz). Figure 3 shows the most stable conformer of 3, predicted using
2
no.
δC, type
δH mult. (J in Hz)
δC, type
δH mult. (J in Hz)
1 2
64.2, CH 36.1, CH2
3.34, m 1.82, m
57.2, CH 36.2, CH2
3 4a 5b
44.7, CH 30.3, CH2 26.7, CH2
45.6, CH 30.3, CH2 26.7, CH2
6c
27.0, CH2
7d
32.0, CH2
1.55, m 1.77, m 1.23, m 1.78, m 1.23, m 1.78, m 1.26, m 1.89, m 1.06, m 2.99, m 5.48, dd 10.1) 6.28, dd 10.1) 6.77, dd 10.1) 5.97, dd 7.3) 2.51, m 1.79, m 1.31, m 1.23, m 1.78, m 1.23, m 1.78, m 1.26, m 1.89, m 1.22, m 1.39, m 2.43, m 3.68, m 2.90, m 3.13, m 1.77, m 2.08, m 3.03, m 3.34, m
55.7, CH 42.4, CH2
4.24, m 1.75, m 1.93, m 1.54, m 1.87, m 1.24, m 1.75, m 1.24, m 1.75, m 1.09, m 1.87, m 1.04, m 2.29, m 6.25, dd 7.1) 6.51, dd 10.4) 6.35, dd 10.4) 5.42, dd 5.8) 2.85, m 1.24, m 1.24, m 1.24, m 1.75, m 1.24, m 1.75, m 1.09, m 1.87, m 1.24, m 1.24, m 2.37, m 4.29, m 3.15, m
18.6, CH2
2.05, m
42.3, CH2
3.06, m
8 9 10
52.5, CH 50.0, CH 135.3, CH
11
131.9, CH
12
132.0, CH
13
131.3, CH
14 15 16a 17b
50.4, 47.7, 30.8, 26.8,
18c
27.3, CH2
19d
32.2, CH2
20 21
45.3, CH 36.7, CH2
22 24
59.6, CH 46.1, CH2
25
23.7, CH2
26
45.4, CH2
a,b,c,d
CH CH CH2 CH2
27.0, CH2 32.0, CH2
(10.6,
55.1, CH 53.0, CH 138.5, CH
(10.1,
126.7, CH
(16.7,
134.8, CH
(16.7,
128.7, CH 46.3, 51.9, 30.9, 26.9,
CH CH CH2 CH2
27.1, CH2 32.2, CH2 44.6, CH 36.9, CH2
(15.7, (15.7, (10.4, (10.4,
The assignments may be interchanged in the same letter.
Figure 1. Key NOESY correlations of 1 based on the energyminimized conformer.
Spartan’14 by a Monte Carlo conformational analysis with an MMFF94 force field based on NOE data for 3. B
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 3. Key NOESY correlations of 3 based on the energyminimized conformer.
Figure 2. Key NOESY correlations of 2 based on the energyminimized conformer. 1
(5.27)/H-15 (1.23), and H-14/H-20 (1.28) revealed the same relative configurations of the A/B and C/D rings as those of 3 (Figure 4). The coupling constants between H-10/
13
Table 2. H (400 MHz) and C (100 MHz) NMR Data for Halichondriamines A (3) and B (4) in CD3OD 3
4
no.
δC, type
δH mult. (J in Hz)
δC, type
1 2 3 4a
57.5, 36.8, 45.1, 30.8,
53.8, 38.7, 45.6, 30.8,
5b
26.9, CH2
6c
27.1, CH2
7d
30.3, CH2
8 9
53.2, CH 51.0, CH
10
131.7, CH
11
132.6, CH
12
131.6, CH
13
131.2, CH
3.38, m 1.77, m 1.59, m 1.28, m 1.77, m 1.25, m 1.77, m 1.25, m 1.77, m 1.06, m 1.93, m 1.03, m 2.68, br dd (17.4, 10.1) 5.28, dd (10.7, 10.4) 6.27, dd (11.1, 10.7) 6.52, dd (15.0, 11.1) 5.63, dd (15.0, 9.2)
CH CH2 CH CH2
14 15 16a
51.8, CH 50.0, CH 30.9, CH2
17b
27.0, CH2
18c
27.2, CH2
19d
32.3, CH2
20 21
45.6, CH 38.5, CH2
22 a,b,c,d
53.7, CH
2.46, 1.25, 1.28, 1.77, 1.25, 1.77, 1.25, 1.77, 1.06, 1.93, 1.28, 1.25, 2.39, 3.69,
m m m m m m m m m m m m m m
CH CH2 CH CH2
27.0, CH2 27.1, CH2 32.3, CH2 50.6, CH 51.6, CH 132.7, CH 131.5, CH 134.7, CH 127.5, CH 46.5, CH 51.9, CH 30.8, CH2 27.0, CH2 27.1, CH2 32.3, CH2 45.5, CH 38.8, CH2 53.6, CH
δH mult. (J in Hz) 3.68, 1.92, 1.59, 1.28, 1.77, 1.25, 1.77, 1.25, 1.77, 1.11, 1.94, 1.11, 2.50,
m m m m m m m m m m m m m
5.67, dd 9.8) 6.58, dd 11.0) 6.38, dd 10.7) 5.27, dd 10.5) 2.92, m 1.23, m 1.28, m 1.77, m 1.25, m 1.77, m 1.25, m 1.77, m 1.11, m 1.94, m 1.28, m 1.20, m 2.40, m 3.73, m
Figure 4. Key NOESY correlations of 4 based on the energyminimized conformer.
H-11 and H-12/H-13 (J10,11 = 14.6 Hz and J12,13 = 10.7 Hz) assigned the structure of 4 as (10E,12Z)-3. On the basis of the NOE data for 4, a Monte Carlo conformational analysis was performed with an MMFF94 force field utilizing Spartan’14 as shown in Figure 4. The absolute configurations of 1−4 were considered to be the same as those of 5 and 6 because compounds 1−6 were isolated from the same marine sponge. The specific rotations of haliclonadiamine (5) and papuamine (6) isolated in this study were similar to the reported values. 5,6 The absolute configurations of haliclonadiamine and papuamine have been determined by total syntheses.8 In order to confirm the absolute structure of 1, the electronic circular dichroism (ECD) spectrum of 1 was compared with the calculated ECD spectrum (Figure 5). The experimental ECD spectrum of 1 (green line) matched well with the calculated ECD spectrum for the (1S, 3S, 8R, 9S, 14S, 15R, 20S,
(14.6, (14.6, (11.0, (10.7,
The assignments may be interchanged in the same letter.
HREIMS and NMR data (Table 2) for halichondriamine B (4) showed the same molecular formula (C22H36N2) as 3. The 1 H and 13C NMR spectra of 4 were similar to those of 3, and the analysis of 2D NMR spectra of 4 indicated that compound 4 had the same skeletal structure as 3. NOESY correlations between H-1 (δH 3.68)/H-8 (1.11), H-3 (1.59)/H-9 (2.50), H10 (5.67)/H-12 (6.38), H-11 (6.58)/H-14 (2.92), H-13
Figure 5. (a) Experimental ECD spectrum of 1 and (b) calculated ECD spectrum of (1S,3S,8R,9S,14S,15R,20S,22R)-1. C
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
conformational analysis with the semiempirical AM1 molecular orbital method utilizing Spartan’14 (Figure 6). Compounds 7 and 8 may be biosynthesized via the same pathway because both compounds were obtained from the same marine sponge. The specific rotation of 8 was similar to the reported value of monanchorin.7 Because the absolute configuration of monanchorin has been established by total synthesis,9 the absolute configuration of 7 was suggested as (1R, 5R, 6R). The antibacterial activities of 1−8 against M. smegmatis NBRC 3207 were examined using the paper disc method,4,10 and the results obtained are listed in Table 4. Compounds 1−6
22R)-isomer (black line). Thus, the absolute structures of the four new compounds 1−4 were assigned as shown in the graphic. Although light-induced isomerization might explain the presence of the three double-bond isomers 1, 2, and 5 (and separately 3 and 4), compounds 1−6 were detected by HPLC in the original EtOH extract, and there were no signs of isomerization during the routine treatments of the purification process. The molecular formula of 7, C11H21N3O, deduced from HREIMS and NMR data (Table 3), was the same as that of Table 3. 1H (400 MHz) and 13C (100 MHz) NMR Data of 6epi-Monanchorin (7) and Monanchorin (8) in DMSO-d6 7 no. 1 2-NH 3 4-NH 5 6 8 9
δC, type 77.1, CH
8 δH mult. (J in Hz)
4.81, br s 8.40, br s
157.4, C 49.5, CH 77.4, CH 28.2, CH2 26.4, CH2
10-NH2 11
32.6, CH2
12
24.2, CH2
13 14 15
31.1, CH2 22.0, CH2 13.8, CH3
Table 4. Biological Activities of Compounds 1−8 against Mycobacterium smegmatis and a Human Cancer Cell Line (Huh-7 Hepatoma)
δC, type 75.7, CH
δH mult. (J in Hz) 4.82, br t (6.0) 8.68, br s
compound
157.9, C 8.25, 3.40, 3.87, 2.04, 2.14, 2.08, 2.10, 7.12, 1.43, 1.53, 1.26, 1.36, 1.26, 1.26, 0.86,
br d (5.3) br s br t (6.3) m m m m s m m m m m m t (6.8)
50.3, CH 78.6, CH 28.2, CH2 23.3, CH2
32.8, CH2 24.7, CH2 31.0, CH2 22.0, CH2 13.9, CH3
8.56, 3.28, 4.13, 1.86, 2.19, 1.91, 2.02, 7.37, 1.45, 1.53, 1.24, 1.35, 1.25, 1.25, 0.85,
1 2 3 4 5 6 7 8 streptomycin sulfateb doxorubicinc
br s m br t (6.7) m m m m m m m m m m m t (6.6)
M. smegmatis (inhibition zone, mm)
cytotoxicity (IC50, μM)
10 μg/disc
Huh-7
12 7 8 7 16 12
>10 6.7 7.2 7.8 3.6 3.9 >10 >10 0.4
a a
30
Not active at either 10 or 20 μg/disc. bPositive control for the antimycobacterial assay (5 μg/disc). cPositive control for cytotoxicity against Huh-7 cell line. a
exhibited antimycobacterial activities with inhibition zones of 7−16 mm at 10 μg/disc, while 7 and 8 were not active at 20 μg/disc (Table 4). Compound 5 exhibited the most potent activity (16 mm at 10 μg/disc). Therefore, the 13-membered ring (E ring) and the (10E, 12E) configuration appear to be favorable structural features for antimycobacterial properties. Papuamine (6) was previously reported to show antimicrobial activity against M. tuberculosis,11 and, thus, compounds 1−5 may also be active against M. tuberculosis. Haliclonadiamine (5) and papuamine (6) were originally isolated as antimicrobial substances from the marine sponge Haliclona sp,5,6 and we recently revealed the cytotoxicities of these compounds against several human solid cancer cell lines. 12 Papuamine (6) reduced cell survival through mitochondrial damage and the activation of JNK.12b Therefore, the cytotoxicities of 1−8 were evaluated against a human cancer cell line, Huh-7 (hepatoma), and the IC50 values obtained are listed in Table 4. Similar to the results observed for antimycobacterial activities, compound 5 exerted more potent cytotoxic effects on Huh-7 cell line than the geometric isomers (1 and 2) and open-chain derivatives (3 and 4). Compounds 7 and 8 did not show apparent activity against the proliferation of the cancer cell line.
monanchorin (8). The 1H and 13C NMR data for 7 were very similar to those for 8, except for the signals due to the 5, 6, and 9 positions (Table 3). An analysis of the COSY and HMBC spectra of 7 revealed that the planar structure of 7 was the same as that of 8 (Figure 6a). A marked difference between 7 and 8
Figure 6. (a) COSY and key HMBC correlations of 7 and (b) key NOESY correlations of 7 based on the energy-minimized conformer of 7.
was detected in their specific rotations; namely, 7 ([α]22D −16.4) and 8 ([α]22D +50.7) showed opposite signs. These results suggest that compounds 7 and 8 are stereoisomers of each other. A NOESY correlation between H-6 (δH 3.87) and H-9a (2.08) revealed that compound 7 was the epimer at the 6 position of 8. On the basis of the relative configuration, the most stable conformer of 7 was calculated by the systematic
■
EXPERIMENTAL SECTION
General Experimental Procedures. Specific rotations were determined with a JASCO P-2300 digital polarimeter. UV spectra were measured on a U-3310 UV−visible spectrophotometer (Hitachi), and IR spectra on a PerkinElmer Spectrum One Fourier transform D
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Haliclonadiamine (5): yellow oil; [α]25D −35.8 (c 0.10, CH3OH); lit. [α]D −18.2,5 lit. [α]D −5.0 (c 0.12, CHCl3).7 Papuamine (6): yellow oil; [α]20D −119.8 (c 0.10, CH3OH); lit. [α]D −150.0 (c 1.5, CH3OH).6 6-epi-Monanchorin (7): yellow oil; [α]22D −16.4 (c 0.20, CH3OH); IR (KBr) νmax 3370, 3256, 2959, 2938, 2863, 1694, 1610, 1208, 1136, 1068 cm−1; 1H and 13C NMR (DMSO-d6), Table 3; EIMS m/z 211 [M]+; HREIMS m/z 211.1681 (calcd for C11H21N3O, 211.1685). Monanchorin (8): yellow oil; [α]22D +50.7 (c 0.35, CH3OH); lit. [α]25D +39.0 (c 3.90, CH3OH);8 1H and 13C NMR (DMSO-d6), Table 3. Conformational Analyses and Calculations of ECD Spectra. The most stable conformers of new haliclonadiamine derivatives 1−4 were predicted using Spartan’14 (Wavefunction, Inc., 2014) by a preliminary conformational analysis with the MMFF94 force field followed by geometry optimizations using the density functional theory (DFT) with the B3LYP functional and 6-31G(d,p) basis set. The ECD spectrum of 1 in CH3CN was calculated for the predicted most stable conformer using Gaussian 09 (Gaussian, Inc., 2009) by the time-dependent DFT (TDDFT) with the B3LYP functional and 6311+G(d,p) basis set. No Boltzmann averaging was performed because relative energies with respect to the most stable conformer were 0.84 kcal/mol for the second most stable conformer and >1.2 kcal/mol for the higher ones. The solvent effect was introduced by the polarizable continuum model (PCM). Forty low-lying excited states were calculated corresponding to the wavelength region down to approximately 192 nm, and the calculated spectrum was displayed using GaussView 5.0.9 (Semichem, Inc., 2009) with the peak halfwidth at half-height being 0.333 eV. The most stable conformer of 6-epi-monanchorin (7) was predicted using Spartan’14 by a preliminary conformational analysis with the semiempirical AM1 molecular orbital method followed by geometry optimization using DFT with the B3LYP functional and 6-31G(d) basis set. Antimycobacterial Assay. The antibacterial assay was carried out using M. smegmatis NBRC 3207 with the paper disc method.4,10 Strain NBRC 3207 was obtained from the Biological Resource Center (NBRC), NITE (Chiba, Japan), and maintained in 20% glycerol at −80 °C. The test microorganism was cultured in Middlebook 7H9 broth containing 0.05% polysorbate 80, 0.5% glycerol, and 10% Middlebook OADC at 37 °C for 2 days and adjusted to 1.0 × 106 CFU/mL. The inoculum was spread on the above medium containing 1.5% agar in a square plate. Each sample in CH3OH was adsorbed to a sterile filter disc (6 mm, Advantec), and, after the evaporation of CH3OH, the disc was placed on an agar plate and incubated at 37 °C for 2 days. Streptomycin sulfate (5 μg/disc) and CH3OH were used as positive and negative controls, respectively. WST-1 Assay. Cytotoxicity was assessed utilizing the water-soluble tetrazolium assay (WST-1; sodium 5-(2,4-disulfophenyl)-2-(4-iodophenyl)-3-(4-nitrophenyl)-2H tetrazolium inner salt), which detects metabolically competent cells with an intact mitochondrial electron transport chain.13 Briefly, 1 × 104 cells were seeded on each well of a 96-well plastic plate and cultured overnight. Cells were treated with each test compound followed by an incubation for 72 h, and medium containing WST-1 solution (0.5 mM WST-1 and 0.02 mM 1-methoxy5-methylphenazinium methylsulfate; 1-PMS) was added to each well. Cells were incubated at 37 °C for 60 min, and absorption at 438 nm (reference 620 nm) was measured using an SH-1200 microplate reader (Corona Electric). Control cells were treated with 0.1% EtOH. Cell viability was calculated using the following formula: absorbance in the treated sample/absorbance in the control ×100 (%).
infrared spectrometer. ECD spectra were measured with a spectropolarimeter (J-720; JASCO). NMR spectra were recorded on a JEOL JNM-AL-400 NMR spectrometer (400 MHz for 1H and 100 MHz for 13C) in CD3OD (δH 3.30, δC 49.0) or DMSO-d6 (δH 2.46, δC 39.5). EIMS and HREIMS were performed using a JMS-MS 700 mass spectrometer (JEOL). Preparative HPLC was carried out with a Hitachi L-6200 system. Middlebook 7H9 broth, polysorbate 80, and Middlebook OADC were purchased from BD Chemical Industry. Plastic plates (96-well) were purchased from Corning Inc. All other chemicals including organic solvents were purchased from Wako Pure Chemical Industries Ltd. Marine Sponge and Isolation of Compounds 1−8. The marine sponge was collected by scuba diving at Iriomote Island in Okinawa, Japan, in 2013 and identified as Halichondria panicea. A voucher specimen was deposited at the Faculty of Pharmaceutical Sciences, Tohoku Medical and Pharmaceutical University as 13-9-7=22. The frozen sponge (525.6 g, wet weight) was thawed, cut into small pieces, and extracted three times with EtOH (1.0 L). The EtOH extract was evaporated, and the residue (21.2 g) was dissolved in CH3OH−H2O (9:1, 300 mL) and extracted with hexane. The lower layer was evaporated to remove CH3OH, diluted with H2O, and extracted successively with EtOAc and BuOH. The EtOAc and BuOH extracts showed 12 and 18 mm inhibition zones against M. smegmatis at 25 μg/disc, respectively. The BuOH extract (6.3 g) was separated by an ODS column with CH3OH−H2O (stepwise elution) into seven fractions (Frs. B-1−B-7). The active Fr. B-2 (407.3 mg, eluted with 30% CH3OH) was subjected to preparative HPLC [column, PEGASIL ODS (Senshu Sci. Co., Ltd., Tokyo, Japan), i.d. 10 mm × 250 mm; solvent, 50% CH3OH containing 0.05% TFA; flow rate, 2.0 mL/min; detection, UV 210 nm] to give halichondriamine B (4) (1.7 mg, tR = 29.9 min), halichondriamine A (3) (10.3 mg, tR = 35.0 min), (10E,12Z)-haliclonadiamine (2) (11.9 mg, tR = 49.9 min), papuamine (6) (18.6 mg, tR = 57.9 min), and haliclonadiamine (5) (21.7 mg, tR = 67.6 min). The EtOAc extract (0.7 g) was applied to an ODS column and eluted stepwise with CH3OH−H2O to obtain seven fractions (Frs. E-1−E-7). (10Z,12E)-Haliclonadiamine (1) (3.9 mg, tR = 58.8 min) was isolated from active Fr. E-6 (115.2 mg, eluted with 100% CH3OH) by preparative HPLC (column, PEGASIL ODS, 10 mm × 250 mm; solvent, 55% CH3OH containing 0.05% TFA; flow rate, 2.0 mL/min; detection, UV 210 nm). Fr. E-2 (47.0 mg, eluted with 30% CH3OH) was subjected to preparative HPLC (column, PEGASIL ODS, i.d. 10 mm × 250 mm; solvent, 45% CH3OH containing 0.05% TFA; flow rate, 2.0 mL/min; detection, UV 210 nm) and afforded 6epi-monanchorin (7) (2.0 mg, tR = 17.3 min) and monanchorin (8) (2.5 mg, tR = 20.2 min). (10Z,12E)-Haliclonadiamine (1): yellow oil; [α]22D −35.4 (c 0.10, CH3OH); UV (CH3OH) λmax (log ε) 238 (4.1) nm; ECD (c 0.010, CH3CN) λmax (Δε) 228 (−2.3), 262 (+0.6) nm; IR (KBr) νmax 3430, 2930, 2856, 1680, 1205, 1184, 1137, 1034 cm−1; 1H and 13C NMR (CD3OD), Table 1; EIMS m/z 368 [M]+; HREIMS m/z 368.3184 (calcd for C25H40N2, 368.3192). (10E,12Z)-Haliclonadiamine (2): yellow oil; [α]19D +73.3 (c 0.10, CH3OH); UV (CH3OH) λmax (log ε) 240 (4.2) nm; ECD (c 0.010, CH3CN) λmax (Δε) 240 (+4.2) nm; IR (KBr) νmax 3431, 2929, 2856, 1680, 1205, 1184, 1135, 1028 cm−1; 1H and 13C NMR (CD3OD), Table 1; EIMS m/z 368 [M]+; HREIMS m/z 368.3186 (calcd for C25H40N2, 368.3192). Halichondriamine A (3): yellow oil; [α]22D −21.6 (c 0.10, CH3OH); UV (CH3OH) λmax (log ε) 239 (4.2), 234 (4.2) nm; ECD (c 0.010, CH3CN) λmax (Δε) 210 (+0.2), 239 (−0.6) nm; IR (KBr) νmax 3422, 2930, 2857, 1680, 1638, 1205, 1184, 1137, 1028 cm−1; 1H and 13C NMR (CD3OD), Table 2; EIMS m/z 328 [M]+; HREIMS m/z 328.2867 (calcd for C22H36N2, 328.2879). Halichondriamine B (4): yellow oil; [α]21D −9.6 (c 0.10, CH3OH); UV (CH3OH) λmax (log ε) 240 (4.2), 234 (4.2) nm; ECD (c 0.010, CH3CN) λmax (Δε) 237 (−1.0) nm; IR (KBr) νmax 3422, 2929, 2857, 1682, 1651, 1205, 1184, 1137, 1031 cm−1; 1H and 13C NMR (CD3OD), Table 2; EIMS m/z 328 [M]+; HREIMS m/z 328.2867 (calcd for C22H36N2, 328.2879).
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.6b00095. E
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
■
Article
Experimental data for known compounds 5, 6, and 8 and 1D and 2D NMR spectra of 1−4 and 7 (PDF)
AUTHOR INFORMATION
Corresponding Author
*Tel/fax (H. Yamazaki): +81 22 727 0218. E-mail: yamazaki@ tohoku-mpu.ac.jp. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported in part by the Foundation for Japanese Chemical Research to H.Y. The calculations by Gaussian 09 were performed using supercomputing resources at the Cyberscience Center, Tohoku University. We are grateful to the Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University, for providing the human cancer cell lines, Dr. K. Ogawa of the Z. Nakai Laboratory for identifying the marine sponge, and Mr. T. Matsuki and S. Sato of Tohoku Medical and Pharmaceutical University for measuring mass and NMR spectra.
■
REFERENCES
(1) (a) Blunt, J. W.; Copp, B. R.; Keyzers, R. A.; Munro, M. H.; Prinsep, M. R. Nat. Prod. Rep. 2015, 32, 116−211 and previous reports in this series. (b) Faulkner, D. J. Nat. Prod. Rep. 2002, 19, 1−48 and previous reports in this series. (2) (a) Abdjul, D. B.; Yamazaki, H.; Takahasi, O.; Kirikoshi, R.; Ukai, K.; Namikoshi, M. Chem. Pharm. Bull., accepted. (b) Tanaka, N.; Suto, S.; Ishiyama, H.; Kubota, T.; Yamano, A.; Shiro, M.; Fromont, J.; Kobayashi, J. Org. Lett. 2012, 14, 3498−3501. (c) Ohno, O.; Chiba, T.; Todoroki, S.; Yoshimura, H.; Maru, N.; Maekawa, K.; Imagawa, H.; Yamada, K.; Wakamiya, A.; Suenaga, K.; Uemura, D. Chem. Commun. 2011, 47, 12453−12455. (d) Chill, L.; Yosief, T.; Kashman, Y. J. Nat. Prod. 2002, 65, 1738−1741. (3) (a) Hirata, Y.; Uemura, D. Pure Appl. Chem. 1986, 58, 701−710. (b) Newman, D. J.; Cragg, G. M. Mar. Drugs 2014, 12, 255−278. (c) Dybdal-Hargreaves, N. F.; Risinger, A. L.; Mooberry, S. L. Clin. Cancer Res. 2015, 21, 2445−2452. (4) (a) Bu, Y. Y.; Yamazaki, H.; Ukai, K.; Namikoshi, M. Mar. Drugs 2014, 12, 6102−6112. (b) Abdjul, D. B.; Yamazaki, H.; Kanno, S.; Takahashi, O.; Kirikoshi, R.; Ukai, K.; Namikoshi, M. J. Nat. Prod. 2015, 78, 1428−1433. (5) Fahy, E.; Molinski, T. F.; Harper, M. K.; Sullivan, B. W.; Faulkner, D. J.; Parkanyi, L.; Clardy, J. Tetrahedron Lett. 1988, 29, 3427−3428. (6) Baker, B. J.; Scheuer, P. J.; Shoolery, J. N. J. Am. Chem. Soc. 1988, 110, 965−966. (7) Meragelman, K. M.; McKee, T. C.; McMahon, J. B. J. Nat. Prod. 2004, 67, 1165−1167. (8) McDermott, T. S.; Mortlock, A. A.; Heathcock, C. H. J. Org. Chem. 1996, 61, 700−709. (9) Yu, M.; Snider, B. B. Org. Lett. 2009, 11, 1031−1032. (10) Ericsson, H. Scand. J. Clin. Lab. Invest. 1960, 12, 408−413. (11) Barker, L. P.; Lien, B. A.; Brun, O. S.; Schaak, D. D.; McDonough, K. A.; Chang, L. C. Planta Med. 2007, 73, 559−563. (12) (a) Yamazaki, H.; Wewengkang, D. S.; Kanno, S.; Ishikawa, M.; Rotinsulu, H.; Mangindaan, R. E. P.; Namikoshi, M. Nat. Prod. Res. 2013, 27, 1012−1015. (b) Kanno, S.; Yomogida, S.; Tomizawa, A.; Yamazaki, H.; Ukai, K.; Mangindaan, R. E. P.; Namikoshi, M.; Ishikawa, M. Int. J. Oncol. 2013, 43, 1413−1419. (13) Berridge, M. V.; Herst, P. M.; Tan, A. S. Biotechnol. Annu. Rev. 2005, 11, 127−152.
F
DOI: 10.1021/acs.jnatprod.6b00095 J. Nat. Prod. XXXX, XXX, XXX−XXX