Quinoline, a well-known heterocyclic compound, itself has few applications; however, many of its derivatives are useful in diverse therapeutic applications,1, 2 such as antimalarials (Quinine and Quinidine), antiviral (Saquinavir) and anticancer (Camptothecin, Irinotecan and Topotecan).3 Over the past decades, the syntheses and the biological activities of quinoline-5,8-diones have been reported.2, 4, 5 In addition, various functionalized quinoline-5,8-dione moieties are present in many classes of biologically active compounds, such as antitumor agents lavendamycim,6 streptonigrin7, 8 and styrylquinoline derivatives with perspective HIV integrase inhibitors.9, 10 In the course of screening for bioactive metabolite from microbial sources, a strain Saccharopolyspora sp. YIM M13568 isolated from the mangrove rhizosphere was selected for (1) its excellent cytotoxic activity; (2) kinds of compounds with MWs in the range of 200−300 amu possessing the UV chromophores at ~350, 300, 260 and 220 nm. In this study, we reported fermentation, isolation and structural elucidation of four quinoline alkaloids (1–4) and a new naphthalene derivative (5) from Saccharopolyspora sp. YIM M13568, and their cytotoxic activities.

The strain YIM M13568, isolated from the mangrove rhizosphere samples, was identified as Saccharopolyspora sp. according to its 16 S rDNA sequence. The strain was inoculated on MP agar media (10 l) and cultured for 12 days at 28 °C, and the fermented cultures were extracted with EtOAc-MeOH (85:15, v/v) to obtain a crude extract. The crude extract was partitioned between EtOAc and H2O, and the EtOAc extract was further partitioned between 95% MeOH and petroleum ether to afford the MeOH extract (4.0 g). The MeOH extract was subjected to column chromatography over Sephadex LH-20, MPLC and finally purified by HPLC to afford 1–5 (Figure 1), respectively.

Figure 1
figure 1

Structures of compounds 1–5.

Full size image

The molecular formula of 1 was established as C12H12N2O2 according to the HRESI-MS at m/z=217.0964 [M+H]+ (calcd 217.0972). The 1H NMR (in CDCl3) data of 1 showed two aromatic singlets at δ 8.03 and 5.84, one methyl doublet at δ 2.95 (d, J=5.4 Hz) and two methyl singlets at δ 2.70 and 2.42 (Table 1). The 13C NMR and HSQC revealed the presence of three CH3 (δ 29.2, 23.5 and 19.4), two aromatic CH (δ 134.3 and 101.3) and seven quaternary C-atom (δ 164.5, 135.3, 181.9, 148.4, 181.8, 146.9 and 125.5; Table 1). The NMR data were almost identical with those of sannanine.11 The HMBC and 1H, 1H COSY (Figure 2) correlations further confirmed the structure. Eventually, the structure of 1 was corroborated to be 2,3-dimethyl-6-(methylamino)quinoline-5,8-dione by the X-ray diffraction analysis (Cambridge Crystallographic Data Centre (CCDC) 1438216) of a single crystal obtained from MeOH-CH2Cl2 (1:1), the same as that of reported sannanine (Figure 3).11

Table 1 1H (400 MHz) and 13C NMR (100 MHz) data of 1−5a (δ in p.p.m., J in Hz)
Full size table
Figure 2
figure 2

Selected 1H-1H COSY (—) and HMBC (H→C) correlations for 1, 4 and 5. A full color version of this figure is available at The Journal of Antibiotics journal online.

Full size image
Figure 3
figure 3

X-ray crystal structures of 1 and 3. A full color version of this figure is available at The Journal of Antibiotics journal online.

Full size image

Compound 2 exhibited a quasi-molecular ion at m/z=203.0811 [M+H]+ (calcd 203.0815) in the HRESI-MS experiment, suggesting the molecular formula C11H10N2O2. The NMR data of 2 were similar to those of 1 (Table 1), except for the absence of the C-11 methyl group. Therefore, 2 was determined to be 6-amino-2,3-dimethylquinoline-5,8-dione.

The molecular formula of 3 was determined to be C12H12N2O2 according to the HRESI-MS at m/z=217.0965 [M+H]+ (calcd 217.0972), the same as that of 1. A methyl amine-substituted 2,3-dimethyl quinolinedione at C-8 was revealed by the 1H, 13C NMR, HSQC and HMBC data. The apparent differences between 3 and 1 were the resonance of C-6 (δ 148.4 in 1; 175.8 in 3) and C-8 (δ 181.8 in 1; 154.5 in 3), which indicated the presence of an o-quinoline, similar to ammosamide D isolated from the marine-derived Streptomyces variabilis.12 Finally, the X-ray diffraction analysis (CCDC 1438217) confirmed the structure as 2,3-dimethyl-8-(methylamino) quinoline-5,6-dione (Figure 2).

The molecular formula of 4 was determined as C15H18N2O3 according to the HRESI-MS at m/z=275.1384 [M+H]+ (calcd 275.1390). Compared with 3, the difference mainly occurs at C-5. The 2-oxopropyl substituted at C-5 was confirmed by the HMBC correlations from H-5c to C-5a and C-5b, and from H-5a to C-4a, C-5, C-6, C-5b and C-5c (Figure 2). The CD experiment suggested that 4 was a racemic mixture, which was consistent with its small optical rotation value. Therefore, 4 was determined to be 5-hydroxy-2,3-dimethyl-8-(methylamino)-5-(2-oxopropyl) quinolin-6-one. Our experiment suggested that 3 can be transformed to 4 in the presence of acetone and silica gel, and therefore 4 should be an artifact, which was consistent with racemic mixture.

Compound 5 exhibited the molecular formula C12H11NO3, as determined by its HRESI-MS at m/z=218.0808 [M+H]+ (calcd 218.0812). The 1H, 13C NMR and HSQC data of 5 showed three aromatic CH signals at δ 7.75, 7.26 and 5.49 (Table 1). According to the HMBC correlations of H-1 to C-2, C-3, C-8 and C-4a, and H-4 to C-2, C-5, C-8a and C-9, and H-7 to C-5 and C-8a, a 1,2-naphthoquinone skeleton was confirmed. The presence of a methyl amine substitute was confirmed by the 1H-1H COSY correlations between the methyl doublet at δ 2.93 (H-10) and the exchangeable singlet at δ 8.27 (8-NH) (Figure 2); its attachment at C-8 was corroborated by the HMBC correlation from H-11 (δ 2.93) to C-8 (δ 156.3). The HO group at C-2 was confirmed by the downfield shift of C-2 (δ 161.6). Therefore, 5 was determined to be 2-hydroxy-3-methyl-8-(methylamino)naphthalene-5,6-dione.

To evaluate the cytotoxic activity of 1–5, we measured their cytotoxic effects on human cancer HeLa, PC3, HCT116 and SW480 cell lines using sulforhodamine B (SRB) methods.13 Compound 1 was observed to be cytotoxic to HeLa, HCT116 and SW480 cells with IC50 value of 11.5, 22.4 and 19.0 μM; compound 2 was observed to be cytotoxic to HeLa, PC3, HCT116 and SW480 with an IC50 of 18.5, 15.3, 23.7 and 14.3 μM; compound 3 was observed to be selectively cytotoxic to SW480 cells with IC50 value of 13.9 μM. However, 4 and 5 show almost no cytotoxicity to test cell lines (Table 2).

Table 2 IC50 values (μM) of compounds 1–5 tested against human tumor cell lines
Full size table

In summary, chemical study on the strain M13568 led to the isolation and characterization of five compounds including two quinoline-5,8-dione (1 and 2), two quinoline-5,6-dione derivatives (3 and 4) and a 1,2-naphthoquinone (5). Compound 1 (named sannanine) was the first example as a quinoline-5,8-dione derivative with methyl substituted on C-3, and our current findings have revealed important new additions to the family of 2,3-dimethyl quinolinedione system. Owing to insufficient biological screening models, the efficient activities of compounds 1−5 need further investigation.