Natural products are considered to be good sources for the screening of lead compounds of clinical drugs. We performed a large number of drug screenings by using a variety of assay systems with crude extracts of microbial cultures as a traditional natural product library. In some assay systems, our crude extract library could not work effectively. With this in mind, we started to construct a purified natural compound library possessing various skeletons from cultures of microorganisms. As actinomycetes are known to produce pharmaceutically useful compounds,1, 2 we focused mainly on secondary metabolites from cultures of actinomycetes. To achieve this, we established a high-throughput system for the detection of secondary metabolites from the cultures of actinomycetes with the retention-time and HR-MS data of known compounds by using a UPLC-TOF-MS system (Waters, Milford, MA, USA).3 We analyzed the secondary metabolites, including potential novel compounds, present in the cultures of the strains isolated from a variety of resources by this system.3 In the course of our chemical screening program, a new furaquinocin derivative named JBIR-136 (1, Figure 1a), together with the known compound kujimycin A,4 was isolated from a culture broth of Streptomyces sp. 4963H2. This paper describes the fermentation, isolation, structural elucidation and, in brief, the biological activity of 1.

Figure 1
figure 1

(a) Structure of JBIR-136 (1). (b) Key correlations of DQF-COSY (bold line) and HMBC (arrow; dashed arrow is four-bond coupling) for 1. (c) Selected NOESY correlations and 1H–1H coupling constants of 4-hydroxy-5-methoxycyclohex-2-enone moiety of 1 (dashed arrow: NOESY correlation). (d) Relative configurations of C-2–C-3 and C-3–C-10 axes of 1 (dashed arrow: NOESY correlation). (e) Δδ values (Δδ (in p.p.m.)=δS–δR) obtained for the (S)- and (R)-methoxy(trifluoromethyl)phenylacetic acid (MTPA) diesters of the methyl ester derivative of 1.

Streptomyces sp. 4963H2 was isolated from a soil sample collected in Kimitsu, Chiba Prefecture, Japan. The strain was cultivated in 50-ml test tubes, each containing 15 ml of a seed medium consisting of 1.0% starch (Kosokagaku, Tokyo, Japan), 1.0% polypeptone (Nihon Pharmaceutical, Tokyo, Japan), 1.0% molasses (Dai-Nippon Meiji Sugar, Tokyo, Japan) and 1.0% meat extract (Extract Ehlrich, Wako Pure Chemical Industry, Osaka, Japan) at pH 7.2 (adjusted before sterilization). The test tubes were shaken on a reciprocal shaker (320 r.p.m.) at 27 °C for 2 days. Aliquots (2.5 ml) of the broth were transferred to 500-ml baffled Erlenmeyer flasks containing 100 ml of a production medium consisting of 2.0% glycerol (Nacalai Tesque, Kyoto, Japan), 1.0% molasses (Dai-Nippon Meiji Sugar), 0.5% casein (Kanto Chemical, Tokyo, Japan), 0.1% polypeptone (Nihon Pharmaceutical) and 0.4% CaCO3 (Kozaki Pharmaceutical, Tokyo, Japan) at pH 7.2 (adjusted before sterilization), and were cultured on a rotary shaker (180 r.p.m.) at 27 °C for 5 days.

The fermentation broth (2 l) was separated by centrifugation. The supernatant was partitioned between EtOAc and H2O (2 l × 3), whereas the mycelial cake was extracted with acetone (50 ml) and filtered, and the filtrate was concentrated in vacuo. The residual aqueous concentrate was extracted with EtOAc (equal volume × 3). The combined EtOAc layers were dried over Na2SO4, and then evaporated to dryness. The residue (733 mg) was subjected to normal-phase medium pressure liquid chromatography (Purif-Pack SI-30, Shoko Scientific, Yokohama, Japan) and developed successively with a gradient system of n-hexane–EtOAc (0–15% EtOAc) followed by the stepwise solvent system of CHCl3–MeOH (0, 2, 5, 10, 20, 30 and 100% MeOH). The 5% MeOH-eluted fraction (91.3 mg) was then subjected to preparative reversed-phase HPLC on a CAPCELL PAK C18 MGII column (5.0 μm, 20 i.d. × 150 mm; Shiseido, Tokyo, Japan) with 60% aqueous MeOH containing 0.1% formic acid (flow rate: 10 ml min−1) to give 1 (7.2 mg; retention time: 10.6 min).

Compound 1 was a pale yellow amorphous solid ([α]24D –38, c 0.36 in MeOH; UV λmax (ɛ) in MeOH: 233 (17 000), 281 (12 000) and 316 (5600) nm). The IR spectrum (νmax (attenuated total reflectance) 3380 and 1652 cm−1) indicated the presence of hydroxy and carbonyl groups. The molecular formula was determined as C21H28O6 by high-resolution ESI MS (m/z 375.1790 [M–H]–, calcd for C21H27O6 375.1808). The direct connectivity between protons and carbons was established by a heteronuclear single quantum coherence spectrum; the tabulated 13C and 1H NMR spectroscopic data for 1 are listed in Table 1. A planar structure was elucidated through the analysis of the DQF-COSY and constant-time (CT-HMBC)5 data, as described below.

Table 1 NMR spectroscopic data (600 MHz) for JBIR-136 (1)

The sequence from the oxymethine proton 6-H (δH 4.63) to the methylene protons 8-H2 (δH 2.89; 2.42) via the oxymethine proton 7-H (δH 3.62) was observed in the DQF-COSY spectrum (Figure 1b). The CT-HMBC spectrum showed 1H–13C long-range correlations from 8-H2 to the conjugated ketone carbonyl carbon C-9 (δC 191.4) and the aromatic quaternary carbon C-9a (δC 110.7), from 7-H to C-9 and the aromatic quaternary carbon C-5a (δC 146.9), from 6-H to C-5a and C-9a and from the methoxy proton 18-H3 (δH 3.39) to the oxygenated methine carbon C-7 (δC 81.1), indicating the presence of a 4-hydroxy-5-methoxycyclohex-2-enone moiety. In addition, strong m-couplings from the aromatic methine proton 5-H (δH 6.59) to the aromatic quaternary carbons C-3a (δC 119.5) and C-9a, together with HMBC correlations to the oxygenated aromatic carbon C-4 (δC 160.2) and the oxymethine carbon C-6 (δC 71.4), and weak four-bond w-shaped couplings from 8-Hα and 6-H to the oxygenated aromatic carbon C-9b (δC 162.0) were observed, indicating the presence of a benzene ring substructure.

The presence of a monoterpene unit was elucidated as follows. A 3-isopropylidenepropanol moiety (from CH-10 to CH3-15) was established by the spin systems from the oxymethine proton 10-H (δH 4.00) to the olefinic methine proton 12-H (δH 5.29) through the aliphatic methylene protons 11-H2 (δH 2.44, 2.28), and 1H–13C long-range correlations from the two sets of allylic methyl protons 14-H3 (δH 1.64) and 15-H3 (δH 1.69) to the olefinic methine carbon C-12 (δC 121.7) and olefinic quaternary carbon C-13 (δC 134.7), as obtained from the DQF-COSY and CT-HMBC spectra, respectively. The assignments of C-14 and C-15 were established because of the γ-effect at C-14 (δC 18.1). 1H–13C long-range couplings from the singlet methyl proton 17-H3 (δH 1.29) to the quaternary carbon C-3 (δC 53.3), the aromatic quaternary carbon C-3a, the oxymethine carbon C-10 (δC 74.9) and the oxymethine carbon C-2 (δC 89.2) revealed that the 3-isopropylidenepropanol and benzene moieties were interconnected through C-3. 1H–1H coupling between the oxymethine proton 2-H (δH 4.58) and methyl protons 16-H3 (δH 1.25) and the 1H–13C long-range correlation from the methyl proton 16-H3 to C-3 were evident in the DQF-COSY and CT-HMBC spectra, respectively. In addition, the HMBC coupling from the oxymethine proton 2-H to the aromatic carbon C-9b established that these carbons were connected through an ether bond. The index of hydrogen deficiency of eight allowed the final determination of the structure of 1, as shown in Figure 1a.

The relative configuration of 1 was determined from NOESY and the J values. From the 1H NMR spectrum, large coupling constants of 3J6-H,7-H=7.2 Hz and 3J7-H,8-Hβ=7.8 Hz and a small coupling value of 3J7-H,8-Hα=3.6 Hz were obtained. The NOESY spectrum showed a correlation between 6-H and 8-Hβ. On the basis of these data, 6-H, 7-H and 8-Hβ are found to be in pseudoaxial positions, whereas 8-Hα and O-7 are in pseudoequatorial positions, as shown in Figure 1c. The NOESY correlation of 2-H/17-H implied that CH3-16 and CH3-17 were in the trans orientation. For the C-3/C-10 axis, NOESY correlations from 11-H2 to 16-H and 17-H showed gauche orientation between C-11 and C-17. Furthermore, a strong NOESY correlation between 10-H and 16-H was seen, whereas no correlation was observed for 10-H/17-H, indicating gauche orientation between C-17 and the oxygen atom O-10. On the other hand, 1H–13C coupling constants of 3J10H–C17=4.8 Hz, 3J10H–C2<2.8 Hz and 3J10H–C3a<2.8 Hz were obtained from the analysis of the J-resolved HMBC-2 spectrum.6 According to the J-based configuration analysis method,7 this result strongly supported the relative configuration described above. Taken together, the relative configurations at C-2, C-3 and C-10 were elucidated to be R*, S* and R*, respectively, as shown in Figure 1d.

The absolute configuration of 1 was determined using a modified Mosher’s method8 after O-methylation with trimethylsilyl-diazomethane in MeOH to afford a methyl ester. The position of the added methoxy group was confirmed by an HMBC correlation from the emerging methoxy protons (δH 3.91, in Me2CO-d6) to the oxygenated olefinic carbon C-4 (δC 161.8). Treatment of the methyl ester derivative of 1 with (R)- and (S)-methoxy(trifluoromethyl)phenylacetic acid chloride in dry pyridine gave the (S)- and (R)-methoxy(trifluoromethyl)phenylacetic acid diesters, respectively. The differences in chemical shift (Δδ=δS–δR) are shown in Figure 1e. From these values, it was found that C-6 and C-10 were arranged in the R configuration. Given this result, the absolute configurations of 1 were assigned as 2R, 3S, 6R, 7R and 10R.

The structure of 1 is related to those of the furaquinocins,9, 10, 11, 12, 13 in particular furaquinocin D,11, 12 isolated as the secondary metabolite of Streptomyces. As furaquinocins show cytotoxicity against cancer cells, we evaluated the cytotoxic activity of 1 against human cervical carcinoma HeLa cells by using a (2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt (WST-8) colorimetric assay (Cell Counting Kit; Dojindo, Kumamoto, Japan) for 72 h. The results showed that 1 exhibited a weak cytotoxicity, with an IC50 value of 44.4 μM.