QM-HiFSA-Aided Structure Determination of Succinilenes A–D, New Triene Polyols from a Marine-Derived Streptomyces sp.

Based on profiles of secondary metabolites produced by marine bacteria obtained using LC/MS, succinilenes A–D (1–4), new triene polyols, were discovered from a culture of a Streptomyces strain SAK1, which was collected in the southern area of Jeju Island, Republic of Korea. The gross structures of 1–4 were primarily determined through analysis of NMR spectra. The double bond geometries of the succinilenes, which could not be established from conventional 1H NMR spectra because of the highly overlapped olefinic signals, were successfully deciphered using the recently developed quantum-mechanics-driven 1H iterative full spin analysis (QM-HiFSA). Succinilenes A–C (1–3) displayed inhibitory effects against lipopolysaccharide (LPS)-induced nitric oxide (NO) production, indicating their anti-inflammatory significance. These three compounds (1–3) commonly bear a succinic acid moiety, although succinilene D (4), which did not inhibit NO production, does not have this moiety in its structure. The absolute configurations of succinilenes A–D (1–4) were established through J-based configuration analysis, the modified Mosher’s method following methanolysis, and CD spectral analysis.


Introduction
Precise interpretation of the spectroscopic data of molecules is the critical step in natural product research for drug discovery. Nuclear magnetic resonance (NMR) spectroscopy is a predominant method used to elucidate the structures of natural products. However, NMR spectra, even when acquired at a high magnetic field, are sometimes misinterpreted or hardly interpretable due to a lack of signal dispersion or the occurrence of higher order effects. Quantum-mechanics-driven 1 H iterative full spin analysis (QM-HiFSA) [1] is an emerging NMR interpretation tool that can reduce errors and ambiguities in the analysis of highly overlapped 1 H NMR spectra. HiFSA allows for the complete and precise interpretation of 1 H NMR parameters such as coupling constants (J HH ), chemical shifts (δ H ), and line widths (∆ν 1/2 ), on the basis of both structural and spectral line shape considerations [2,3].
Similar to the importance of using combinations of chemical and computational analyses, it is critical to focus on prolific sources of structurally and biologically novel compounds because of constant clinical need for new chemical entities with pharmaceutical potential [4]. As part of our efforts to discover new bioactive natural products, we have been investigating the chemistry of actinobacteria derived from marine and saline environments including salterns [5][6][7], intertidal zones [8][9][10][11], and deep-sea sediments [12]. In our previous reports, novel bioactive secondary metabolites were mainly identified through the chemical profiling of bacterial secondary metabolites by LC/MS analysis. Based on this approach and focusing on actinobacterial chemical profiles, we have continuously isolated actinomycete strains from the southern area of Jeju Island, Republic of Korea. Our comprehensive time-course chemical screening of the marine actinomycete strain collection resulted in the discovery of Streptomyces strain SAK1, which was revealed to produce a series of previously-unreported secondary metabolites, each bearing a triene moiety (UV λmax at 270 nm), in the exponential phase of the culture (in 2~3 days after inoculation). Because the strain SAK1 interestingly produced tripartilactam [13] and sceliphrolactam [14], macrocyclic lactams originally reported from insect-associated Streptomyces strains, and the triene-bearing compounds disappeared as reaching the stationary phase of the culture on day 4, a precise time-course study of the culture was performed to maximize the production of these new metabolites. Through a large-scale fermentation of this strain and the isolation of these compounds using High Performance Liquid Chromatography (HPLC), four new triene polyol metabolites, succinilenes A-D (1)(2)(3)(4), with inhibitory activities against lipopolysaccharide (LPS)-induced nitric oxide (NO) production were obtained. The gross structures of these new compounds were elucidated by conventional 1D and 2D NMR spectroscopy. However, their overlapped 1 H NMR signals, especially in the olefinic region, required the application of QM-HiFSA to determine the geometries of their triene moieties. Herein, we report the isolation, structure elucidation, and biological activities of succinilenes A-D (1-4) (Figure 1), emphasizing a successful application of the QM-HiFSA technique for the complete assignment of 1 H-1 H coupling constants, even from complicated high-order 1 H peaks.

Structural Elucidation
Succinilene A (1) was isolated as a yellow gum, and its molecular formula was deduced as C24H38O7 based on HR-FAB mass spectrometry (obsd. [

Structural Elucidation
Succinilene A (1) was isolated as a yellow gum, and its molecular formula was deduced as C 24  Given this information, all 13 C-1 H one-bond correlations were assigned by analysis of the gHSQC NMR spectrum ( Table 1). Interpretation of COSY and HMBC NMR spectroscopic data readily led to the elucidation of the gross structure of succinilene A (1). First, the COSY signals from a terminal triplet methyl group (H 3 -17 at δ H 1.1619) to aliphatic methylene protons (H 2 -16 at δ H 1.8578 and 1.8105) connected C-17 (δ C 10.9) to C-16 (δ C 27.5). The aliphatic carbon C-16 was elucidated to be directly connected to the oxygenated carbon C-15 (δ C 75.5) based on the COSY correlations between H-15 (δ H 3.7579) and H-16. The homonuclear correlation between H-15 and H-14 (δ H 3.9176) revealed that an oxygen-bound methine carbon (C-14 at δ C 74.2) was connected to C-15, indicating the existence of a diol moiety. Further COSY analysis verified the connectivity between H-14 and H 2 -13 (δ H 2.7253 and 2.6745), identifying the C-14-C-13 bond. The aliphatic carbon C-13 was connected to the sp 2 carbon C-12 (δ C 132.8) based on a 3-bond 1 H-1 H coupling signal between H-12 (δ H 6.1280) and H-13. Additional COSY correlations among the six olefinic protons from H-12 to H-7 (δ H 5.9506) indicated the construction of a triene moiety from C-12 to C-7 (δ C 132.8 to 132.0), which is consistent with the UV absorption maximum at 270 nm. The triene moiety was located next to C-6 based on the 3 J HH couplings between H 2 -6 (δ H 2.6519 and 2.5631) and H-7. The homonuclear correlations of H 2 -6 with the oxygenated proton (H-5 at δ H 4.3745) showed C-6-C-5 (δ C 76.8) connectivity. The HMBC correlations from a singlet methyl proton (H 3 -4-Me; δ H 1.8616) to C-5, C-4 (δ C 140.1), and C-3 (δ C 127.2) indicated expansion of the chain from C-5 to C-3 through C-4. The olefinic methine C-3 was assigned next to C-2 (δ C 37.1), with a branched methyl group (H 3 -2-Me; δ H 0.9973/δ C 16.6) attached to C-2. The COSY correlation from H-1 (δ H 5.0898) to H-2 and H 3 -1-Me identified the connectivity of C-1 (δ C 74.1) and C-2, establishing the first partial structure from C-17 to C-1, a chain composed of 20 carbons bearing three hydroxyl groups and a treiene moiety. Based on additional 2D NMR analysis, the aliphatic H 2 -2 (δ H 2.8516 and 2.8436) displayed 3 J HH and 3 J CH correlations to adjacent aliphatic H 2 -3 peaks (δ H 2.9015, and 2.8990) and two carbonyl carbons C-1 and C-4 (δ C 172.7 and 175.7), completing the second partial structure as a succinic acid. These two partial structures were connected based on a key HMBC correlation from H-1, of which the deshielded chemical shift (δ H 5.0898) was indicative for the existence of an ester linkage, to C-1 , resulting in the gross structure of 1 (Figure 2a).   Table 1). The molecular formula and the UV spectrum of 2 were identical to those of 1, indicating that succinilene B (2) is structurally analogous to succinilene A (1). The 1D and 2D NMR data of 2 also displayed very similar patterns to those of 1, as expected. Further analysis of NMR spectroscopic data resulted in the construction of the gross structure of 2, which was identical to that of 1. Actually, most of the 13 C chemical shifts of 2 were very similar to those of 1, within 0.2 ppm. However, the 13 C peaks of C-13 to C-16 around the oxymethine group located adjacent to the terminal ethyl group differed by 0.5~1 ppm compared to the same peaks in 1, indicating that succinilene B (2) is a diastereomer of 1 that possesses different stereochemistry at C-13 or/and C-14.
Succinilene C (3) was purified as a yellow gum and determined to possess a molecular formula of C24H36O7 on the basis of HR-FABMS data (obsd. [M + H] + at m/z 437.2562, calcd. [M + H] + 437.2545) combined with 1 H and 13 C NMR spectral data ( Table 2). Based on the molecular weight and UV spectrum of 3, it was expected that one of the hydroxyl groups in succinilene C (3) had been oxidized to a ketone group. According to the comparison of the 1D and 2D NMR spectroscopic data of 3 with those of 1, the oxygen-bound proton (H-15, δH 3.7579) in 1 was absent in 3, and a ketone carbon signal (δC 214.0) was detected instead, clarifying that the hydroxyl group at the C-15 position in 1 had been oxidized to the ketone group in 3. This assignment was also supported by the 3-bond HMBC correlations from H3-17 (δH 1.0890) and H2-13 (δH 2.7803 and 2.6591) to C-15 and H3-17/H-16 and H2-13/H-14 COSY correlations.
Succinilene D (4) was isolated as a yellow gum, and its molecular formula was deduced to be C20H34O4 based on the corresponding HR-FABMS data (obsd. [M + Na] + at m/z 361.2347, calcd. [M + Na] + 361.2355) and 1 H and 13 C NMR spectroscopic data ( Table 2). The molecular weight of 4 was 100 Da lower than those of 1 and 2, possibly indicating that the terminal succinic acid moiety in 1 and 2 might have been cleaved. Through the careful analysis of the 1D and 2D NMR spectral data, the gross structure of succinilene D (4) was elucidated to be the left part of 1 or 2, which is the methanolysis product of 1 or 2 that lacks the succinic acid moiety.  Table 1). The molecular formula and the UV spectrum of 2 were identical to those of 1, indicating that succinilene B (2) is structurally analogous to succinilene A (1). The 1D and 2D NMR data of 2 also displayed very similar patterns to those of 1, as expected. Further analysis of NMR spectroscopic data resulted in the construction of the gross structure of 2, which was identical to that of 1. Actually, most of the 13 C chemical shifts of 2 were very similar to those of 1, within 0.2 ppm. However, the 13 C peaks of C-13 to C-16 around the oxymethine group located adjacent to the terminal ethyl group differed by 0.5~1 ppm compared to the same peaks in 1, indicating that succinilene B (2) is a diastereomer of 1 that possesses different stereochemistry at C-13 or/and C-14.
Succinilene C (3) was purified as a yellow gum and determined to possess a molecular formula of C 24 (Table 2). Based on the molecular weight and UV spectrum of 3, it was expected that one of the hydroxyl groups in succinilene C (3) had been oxidized to a ketone group. According to the comparison of the 1D and 2D NMR spectroscopic data of 3 with those of 1, the oxygen-bound proton (H-15, δ H 3.7579) in 1 was absent in 3, and a ketone carbon signal (δ C 214.0) was detected instead, clarifying that the hydroxyl group at the C-15 position in 1 had been oxidized to the ketone group in 3. This assignment was also supported by the 3-bond HMBC correlations from H 3  [M + Na] + 361.2355) and 1 H and 13 C NMR spectroscopic data ( Table 2). The molecular weight of 4 was 100 Da lower than those of 1 and 2, possibly indicating that the terminal succinic acid moiety in 1 and 2 might have been cleaved. Through the careful analysis of the 1D and 2D NMR spectral data, the gross structure of succinilene D (4) was elucidated to be the left part of 1 or 2, which is the methanolysis product of 1 or 2 that lacks the succinic acid moiety. The relative configurations of 1 were established by analyzing the 1 H-1 H coupling constants and hetero half-filtered total correlation spectroscopy (HETLOC) NMR data [15]. To determine the relationship between C-1 and C-2, J-based configuration analysis was performed with 3 J HH , 3 J CH , and 2 J CH values and ROESY NMR data [16], selecting the most suitable rotamer between C-1 and C-2  Figure 2b, thus proposing 1R* and 2R* configuration.
To determine the absolute configurations of the stereogenic centers at C-1, C-2, C-5, C-14, and C-15 in succinilene A (1), the modified Mosher's method was applied [17]. Methanolysis was performed to detach the succinic acid moiety, which enabled derivatization of the hydroxyl group at C-1 with α-methoxy trifluoromethyl-phenylacetic acid (MTPA) chloride. After purifying the methanolysis product (5) of succinilene A, the secondary alcohols in 5 were derivatized using Rand S-MTPA-Cl.
Analysis of the 1 H and COSY NMR spectroscopic data for these Sand R-MTPA esters (8 and 9) led to the assignment of the ∆δ S-R values, establishing 1R, 5S, 14R, and 15R configurations (Figure 3a). Based on the established relative configuration, the absolute configuration of C-2 was determined to be 2R. The relative configurations and absolute configurations of succinilene B (2) were determined to be 1R, 2R, 5S, 14R, and 15S using the same procedure applied for 1 (Figure 3b). The relative and absolute configurations of succinilene C (3) were analogously established as 1R, 2R, 5S, and 15R through application of the modified Mosher's method to the methanolysis product (7) of 3 (Figure 3c). Because the gross structure of succinilene D (4) was identical to the methanolysis products (5 and 6) of 1 and 2, the 1 H NMR and CD spectra of 4 were compared with those of 5 and 6. The 1 H NMR data for 6 was identical to those of 4 but was different from the 1 H NMR spectrum of 5, indicating that the relative configuration of 4 is identical to that of 6. In addition, the CD spectra of 4 and 6 were identical, thus leading to the assignment of 1R, 2R, 5S, 14R, and 15S configurations for 4 ( Figure S43). determined to be 2R. The relative configurations and absolute configurations of succinilene B (2) were determined to be 1R, 2R, 5S, 14R, and 15S using the same procedure applied for 1 (Figure 3b). The relative and absolute configurations of succinilene C (3) were analogously established as 1R, 2R, 5S, and 15R through application of the modified Mosher's method to the methanolysis product (7) of 3 (Figure 3c). Because the gross structure of succinilene D (4) was identical to the methanolysis products (5 and 6) of 1 and 2, the 1 H NMR and CD spectra of 4 were compared with those of 5 and 6. The 1 H NMR data for 6 was identical to those of 4 but was different from the 1 H NMR spectrum of 5, indicating that the relative configuration of 4 is identical to that of 6. In addition, the CD spectra of 4 and 6 were identical, thus leading to the assignment of 1R, 2R, 5S, 14R, and 15S configurations for 4 ( Figure S43). Even though the absolute configurations of the stereogenic centers in 1-4 were completely assigned, the full assignments of the double bond geometries were not straightforward. The geometries of the double bonds of C-7/C-8 and C-11/C-12 in the succinilenes (1-4) were determined as 7E and 11E based on the H-7/H-8 and H-11/H-12 trans-coupling constant values. However, it was virtually impossible to clarify the double bond geometry of C-9/C-10 in 1-4 because the H-9 and H-10 olefinic protons were greatly overlapped, displaying second-order peaks in the 1 H NMR spectra and, thus, hampering the accurate deconvolution of the 1 H-1 H coupling constants between H-9 and H-10, even at higher magnetic field (900 MHz).
Therefore, a recently developed computational analysis system, QM-HiFSA [1], was utilized to extract the H-9/H-10 coupling constants of the succinilenes using PERCH software. After geometry optimization and dynamic simulation of molecules using the MMS module, the basic 1 H NMR parameters were predicted based on the 3D structural information. Then, the obtained initial parameters were optimized by iterative quantum mechanical simulation against experimental spectra. The iteration was repeated until both the simulated and experimental spectra were in excellent agreement with each other, with root mean square (RMS) values below 0.1 (RMS = 0.075 for 1; RMS = 0.066 for 2; RMS = 0.093 for 3; RMS = 0.091 for 4) (Figures S44-S47). Thus, the precise 1 H NMR parameters were determined with high precision (chemical shifts: δH, 0.1 ppb; coupling constants: J, 10 mHz) (Tables 1 and 2) [18]. The J couplings of the highly overlapped olefinic proton signals (H-9 and H-10) of 1-4 were calculated to be 14.03 Hz, 14.79 Hz, 15.05 Hz, and 14.86 Hz (cis J couplings ca. 10-12 Hz; trans J couplings ca. 14-18 Hz) [18,19], respectively. These results confirmed the 9E geometry of the trans-olefinic protons (H-9 and H-10) in 1-4 ( Figure 4). Even though the absolute configurations of the stereogenic centers in 1-4 were completely assigned, the full assignments of the double bond geometries were not straightforward. The geometries of the double bonds of C-7/C-8 and C-11/C-12 in the succinilenes (1-4) were determined as 7E and 11E based on the H-7/H-8 and H-11/H-12 trans-coupling constant values. However, it was virtually impossible to clarify the double bond geometry of C-9/C-10 in 1-4 because the H-9 and H-10 olefinic protons were greatly overlapped, displaying second-order peaks in the 1 H NMR spectra and, thus, hampering the accurate deconvolution of the 1 H-1 H coupling constants between H-9 and H-10, even at higher magnetic field (900 MHz).
Therefore, a recently developed computational analysis system, QM-HiFSA [1], was utilized to extract the H-9/H-10 coupling constants of the succinilenes using PERCH software. After geometry optimization and dynamic simulation of molecules using the MMS module, the basic 1 H NMR parameters were predicted based on the 3D structural information. Then, the obtained initial parameters were optimized by iterative quantum mechanical simulation against experimental spectra. The iteration was repeated until both the simulated and experimental spectra were in excellent agreement with each other, with root mean square (RMS) values below 0.1 (RMS = 0.075 for 1; RMS = 0.066 for 2; RMS = 0.093 for 3; RMS = 0.091 for 4) (Figures S44-S47). Thus, the precise 1 H NMR parameters were determined with high precision (chemical shifts: δ H , 0.1 ppb; coupling constants: J, 10 mHz) (Tables 1 and 2) [18]. The J couplings of the highly overlapped olefinic proton signals (H-9 and H-10) of 1-4 were calculated to be 14.03 Hz, 14.79 Hz, 15.05 Hz, and 14.86 Hz (cis J couplings ca. 10-12 Hz; trans J couplings ca. 14-18 Hz) [18,19], respectively. These results confirmed the 9E geometry of the trans-olefinic protons (H-9 and H-10) in 1-4 ( Figure 4).

Bioactivities of Succinilenes A-D
The biological activities of the succinilenes (1-4) were primarily evaluated for cytotoxicity against human cancer cells. In this assay, succinilene A showed moderate cytotoxicity against the gastric carcinoma cell line SNU638, with an IC 50 value of 12.1 µg/mL (27.6 µM), whereas succinilenes B-D did not display significant inhibitory activities against the tested cancer cell lines. As part of the efforts to search for biological activities, a nitrate assay was performed for 1-4. Overproduction of nitric oxide (NO) is highly associated with inflammation, and the pathway that regulates NO production is considered a useful target for anti-inflammatory agents [20]. RAW 264.7 cells were treated with LPS (1 µg/mL), which induces inflammatory responses [21,22]. When the cells were pretreated with various concentrations of the succinilenes (0, 20, or 40 µM) 30 min prior to LPS stimulation, NO production was gradually inhibited in a concentration-dependent manner, except for treatment with succinilene D (4). No significant effect on cell viability, as determined by the MTT assay (>90% cell survival), was observed at the tested concentrations, indicating that the inhibition of NO production by the tested compounds was not mediated by a cytotoxic effect. As a result, among the four compounds, succinilene C (3) exhibited the most remarkable inhibitory activity against NO production. By contrast, succinilene D, without succinic acid connected through an ester linkage, resulted in the overproduction of NO ( Figure 5). After 20 h, the cultured media were collected, and the nitrite concentrations were measured using the Griess reaction. ** p < 0.01, *** p < 0.001 are considered statistically significant compared to the control group.

General Experimental Procedures
Optical rotations were measured using a Jasco P-1020 polarimeter (sodium light source, JASCO, Easton, PA, USA) with a 1-cm cell. IR spectra were acquired on a Thermo Nicolet iS10 spectrometer (Thermo, Madison, CT, USA). UV spectra were obtained with a Perkin Elmer Lambda 35 UV-VIS spectrometer (Perkin Elmer, Waltham, MA, USA). Electrospray ionization (ESI) low-resolution LC/MS data were recorded on an Agilent Technologies 6130 quadrupole mass spectrometer coupled to an Agilent Technologies 1200 series HPLC (Agilent Technologies, Santa Clara, CA, USA). High-resolution fast-atom bombardment (HR-FAB) mass spectra were collected with a Jeol JMS-600W high-resolution mass spectrometer (Jeol, Tokyo, Japan) at the National Center for Inter-University Research Facilities (Gwanak-gu, Seoul, Korea). 1 H, 13 C, and 2D NMR spectra were obtained on a Bruker Avance 600 MHz spectrometer (Bruker, Billerica, MA, USA) at the NCIRF and a Bruker Avance 900 MHz NMR spectrometer at the Korea Basic Science Institute (Ochang, Chungcheongbuk-do, Korea).

Isolation of Bacteria, Cultivation, and Extraction
A sediment sample was obtained from the southern area of Jeju Island in a 40-mL sterilized plastic tube. The sample (1 g) was dried at room temperature (rt) for 3 h and diluted in 4 mL of sterilized distilled water. The mixture was spread on actinomycete isolation agar, chitin-based agar, SC medium (1 L of distilled water, 10 g of starch, 0.3 g of casein, 2.0 g of KNO 3 , 2.0 g of K 2 HPO 4 , 0.05 g of MgSO 4 ·7H 2 O, 0.02 g of CaCO 3 , 0.01 g of FeSO 4 ·7H 2 O, 16 g of agar, and 100 mg/L cycloheximide), and A4 medium (1 L of distilled water, 18 g of agar, and 100 mg/L cycloheximide). The strain SAK1 was isolated on SC medium. Based on the analysis of the 16S rDNA sequence, SAK1 was clarified to be most closely related to Streptomyces flavogriseus (99% identity, GenBank accession number: KY484916). For the production of the succinilenes, the strain SAK1 was initially cultivated in 50 mL of YEME media (4 g of yeast extract, 10 g of malt extract, and 4 g of glucose in 1 L of distilled water) in a 125 mL Erlenmeyer flask. This seed culture was cultivated for three days on a rotary shaker at 200 rpm at 30 • C. Ten milliliters of the seed culture was inoculated into 1 L of YEME medium in 2.8-L Fernbach flasks (12 ea × 1 L, total volume of 12 L). After two days of incubation, the whole culture (12 L) was extracted twice with 18 L of ethyl acetate. The ethyl acetate layer was separated by a fractionating funnel and dried over anhydrous sodium sulfate. The ethyl acetate extract was concentrated in vacuo to yield 2 g of dried material. This procedure was repeated six times (72 L of culture; total amount of extract: 12 g) to obtain a sufficient amount of the succinilenes for the structure elucidation and bioassays.

Methanolysis of Succinilenes A-C
Succinilene A (1) (5 mg) was dissolved in 2 mL of MeOH, and 28 mg of NaOMe was added to the vial to prepare a 0.5 M NaOMe solution. The mixture was stirred at room temperature for 5 h. The reaction was quenched by the addition of 1 N HCl. The methanolysis product was extracted by water-ethyl acetate partitioning after evaporating the MeOH in vacuo. After removing the solvent, the methanolysis product (5) was purified by HPLC using gradient elution conditions (10%-100% acetonitrile/water over 40 min; UV detection at 280 nm; flow rate: 2 mL/min) and a reversed-phase HPLC column (Kromasil C 18 : 250 × 10 mm, 5 µm). The methanolysis product (5)

MTPA Esterification of Succinilenes A-C
The methanolysis products of succinilenes A-C (5-7) were prepared individually in 40-mL vials (two 1-mg samples for each compound) and dried completely under high vacuum overnight. After adding catalytic amounts of crystalline N,N-dimethylaminopyridine (DMAP) to each reaction vial, freshly distilled anhydrous pyridine (1 mL) was added under argon gas. The reaction mixtures were stirred at room temperature for 5 min. After 5 min, Rand S-α-methoxy trifluoromethyl-phenylacetic acid (MTPA) chloride (20 µL) were added, respectively. The reactions were carried out for 3 h at room temperature with stirring. The reactions were quenched by adding 50 µL of MeOH. The reaction products were purified by HPLC using gradient elution conditions of 40% to 100% aqueous acetonitrile over 20 min with a reversed-phase C 18 column (Kromasil C 18 : 250 × 10 mm, 5 µm; flow rate: 2 mL/min; UV detection at 280 nm). The tetra-S-and -R-MTPA esters (8 and 9) of 5 eluted at 34.3 and 35.2 min, respectively. The tetra-S-and -R-MTPA esters (10 and 11) of 6 eluted at 33.5 and 34.7 min. The tetra-S-and -R-MTPA esters (12 and 13) of 7 were isolated at 36.1 and 37.3 min. The ∆δ S-R values around the stereogenic centers of the MTPA esters were assigned based on the analysis of 1 H and 1 H-1 H COSY NMR spectra.

Nitrite Assay
To evaluate the inhibitory activity of succinilenes A-D (1-4) on LPS-induced NO production, RAW 264.7 cells were cultured in 10% FBS-DMEM without phenol red, plated in 24-well plates (3 × 10 5 cells/mL), and incubated for 24 h. Cells were washed with PBS, fresh medium was added, and the cells were incubated with 1 µg/mL LPS in the presence or absence of test compounds. After additional 20 h of incubation, the media were collected and analyzed for nitrite accumulation as an indicator of NO production by the Griess reaction. Briefly, 180 µL of Griess reagents (0.1% N-(1-naphthyl)ethylenediamine dihydrochloride in H 2 O and 1% sulfanilamide in 5% H 3 PO 4 ) was added to 100 µL of each supernatant from LPS or sample-treated cells in 96-well plates. The absorbance was measured at 540 nm, and nitrite concentration was determined by comparison with a sodium nitrite standard curve. Percent inhibition was expressed as [1 − (NO level of test samples/NO levels of vehicle-treated control)] × 100.

Statistical Analysis
Data are presented as the means ± SD for the indicated number of independently performed experiments. Statistical significance (p < 0.01) was assessed by a one-way analysis of variation (ANOVA) coupled with the Dunnett's t-test.

Conclusions
Succinilenes A-D (1-4) were discovered from a marine-derived Streptomyces sp. through chemical profiling of a marine actinomycete strain collection. Spectroscopic analysis of 1-4 was used to assign their structures as new polyene polyol compounds, each bearing a conjugated triene moiety. These compounds were biologically evaluated for inhibitory effects on LPS-induced NO production, indicating their anti-inflammatory significance. Succinilenes A-C, commonly incorporating a succinic acid moiety, displayed significant inhibitory effects on NO production whereas succinilene D without succinic acid did not inhibit NO production, thus proposing that the succinic moiety is important for the inhibitory effects on LPS-induced NO production. Microbial metabolites having succinic acid are not common, but a few microbial polyketide compounds are known to bear succinic acid in their structures, such as macrolactin T from the marine bacterium Bacillus marinus [26], reveromycin A from Streptomyces sp. [27], and macrocyclic lactone A26771B from the fungus Penicillium turbatum [28]. These succinic acid-bearing compounds have not been tested for the inhibitory activities on LPS-induced NO production.
Since the chemical shifts of the olefinic protons belonging to the triene moieties and, thus, coupling with each other, are close enough to generate second-order peaks, the triene geometries of the succinilenes could not be determined using conventional methods. Our application of the spin simulation of QM-HiFSA enabled us to calculate the 1 H-1 H coupling constants of these double bond protons and successfully determine the triene geometries. These results suggest that the QM-HiFSA technique is a useful tool for the detailed structural elucidation of complicated natural products displaying highly overlapped 1 H NMR spectra.