Isolation of Coralmycins A and B, Potent Anti-Gram Negative

Sep 6, 2016 - ABSTRACT: Two new potent anti-Gram negative compounds, coralmycins A (1) and B (2), were isolated from cultures of the myxobacteria ...
0 downloads 0 Views 1010KB Size
Download PDF
Article pubs.acs.org/jnp

Isolation of Coralmycins A and B, Potent Anti-Gram Negative Compounds from the Myxobacteria Corallococcus coralloides M23 Yu Jin Kim,† Hyun-Ju Kim,† Geon-Woo Kim,† Kyungyun Cho,‡ Shunya Takahashi,§ Hiroyuki Koshino,§ and Won-Gon Kim*,† †

Superbacteria Research Center, Korea Research Institute of Bioscience and Biotechnology, Yusong, Daejeon 305-806, Korea Department of Biotechnology, Hoseo University, Asan 336-795, Korea § RIKEN Center for Sustainable Resource Science, Hirosawa 2-1, Wako, Saitama 351-0198, Japan ‡

S Supporting Information *

ABSTRACT: Two new potent anti-Gram negative compounds, coralmycins A (1) and B (2), were isolated from cultures of the myxobacteria Corallococcus coralloides M23, together with another derivative (3) that was identified as the very recently reported cystobactamid 919-2. Their structures including the relative stereochemistry were elucidated by interpretation of spectroscopic, optical rotation, and CD data. The relative stereochemistry of 3 was revised to “S*R*” by NMR analysis. The antibacterial activity of 1 was most potent against Gram-negative pathogens, including Escherichia coli, Pseudomonas aeruginosa, Acinetobacter baumanii, and Klebsiella pneumoniae, with MICs of 0.1−4 μg/mL; these MICs were 4−10 and 40−100 times stronger than the antibacterial activities of 3 and 2, respectively. Thus, these data indicated that the β-methoxyasparagine unit and the hydroxy group of the benzoic acid unit were critical for antibacterial activity.

M

anti-Gram-negative metabolites from Korean myxobacteria, we isolated two new potent compounds, coralmycins A (1) and B (2), from Corallococcus coralloides M23, together with another derivative (3) that had been recently reported as cystobactamid 919-2. Cystobactamid 919-2 has been reported to act as a DNA gyrase inhibitor with potent antibacterial activity and was isolated from another myxobacterial strain, Cystobacter sp. Cbv34.12 In this study, we report the producing strain, fermentation, isolation, structural determination, and antibacterial activities of compounds 1−3. Compound 1 was identified as a new derivative hydroxylated at C-2 of the 4-amino-3-isopropoxybenzoic acid moiety of 3, whereas compound 2 is a new βmethoxyaspartic acid derivative with the diastereoisomeric configuration of 3. However, based on the findings of this study, the stereochemistry of cystobactamid 919-2 should be redefined. These data provide important insights into the development of novel antibacterial compounds.

ultidrug-resistant (MDR) pathogens are a major health problem worldwide, particularly since the emergence of vancomycin-resistant Staphylococcus aureus in 2002. The “ESKAPE” pathogens (i.e., Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.) cause significant mortality.1 During the last few decades, efforts to address MDR microorganisms have mainly focused on Gram-positive bacteria, including methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococci, resulting in the development of novel compounds with new modes of action, including linezolid and daptomycin. Recently, infections caused by MDR Gram-negative bacteria have become a growing problem.2 However, no new classes of antibiotics that are effective against Gram-negative bacteria have been developed since fluoroquinolone was introduced in the 1980s.3 Thus, there is an urgent need for new agents to treat infections caused by Gramnegative bacteria resistant to currently available agents.4 Myxobacteria are a group of Gram-negative bacteria that produce a diverse range of bioactive secondary metabolites.5,6 Myxobacteria have received attention as a source of novel antiinfective natural products,7,8 and thousands of myxobacteria have been isolated in Korea.9−11 In the course of screening for © XXXX American Chemical Society and American Society of Pharmacognosy

Received: April 2, 2016

A

DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

5‴, and between NH-6‴ and NHa/b-4‴ (Figures 2, S11, and S12) indicated that the relative configuration of the βmethoxyasparagine moiety of 3 should be “S*R*”. This relative configuration is different from the reported configuration (“SS”) of cystobactamid 919-2. Since the 1H chemical shifts and coupling constants in the β-methoxyasparagine moiety of 3 in both DMSO-d6 and CD3OD were completely identical to those of cystobactamid 919-2 (Table 2) and because the specific rotation [−14.4 (c 0.08, MeOH)] of 3 was the same as that previously reported for cyctobactamid 919-2 [−14.3 (c 0.057, MeOH)],12 the absolute configuration of 3 should be the same as that of cyctobactamid 919-2. Thus, these data indicated that the relative stereochemistry of cyctobactamid 919-212 should be revised from “SS” to “S*R*”. The molecular formula of 1 was determined to be C46H45N7O15 on the basis of HRESIMS (Figure S13) in combination with 1H and 13C NMR data. The IR absorptions at 1645 cm−1 suggested the presence of a carbonyl moiety. The 1 H and 13C NMR data for 1 were similar to those of 3. The major differences in 1H and 13C NMR (Table 1) with COSY and HMQC data (Figures S14−S16) were that one more orthocoupling aromatic proton signal (δH 7.81, 1H, d, J = 9.0 Hz; δC 111.3 and δH 7.63, 1H, d, J = 9.0 Hz; δC 126.4) appeared in 1 instead of the proton signals of the 1,2,4-trisubstituted benzene moiety of 3 and an oxygenated sp2 carbon at δ 156.1 (C-3) appeared in 1 instead of the aromatic carbon at δ 115.2 (C-3). Together with the molecular formula, these data suggested the hydroxylation at C-3 in the 4-amino-3-isopropoxybenzoic acid unit of 3. Indeed, in the HMBC spectrum (Figures 1 and S17), one proton at δ 7.63 (H-7) in the benzoic acid unit had longrange couplings with C-1 and two sp2carbons at C-3 (δ 156.1) and C-5 (δ 137.0). The other proton at δ 7.81 (H-6) had HMBC correlations with two sp2 carbons at δ 117.0 (C-2) and 136.2 (C-4). Additionally, the O-linked methine proton at δ 4.80 of the isopropoxy group in the benzoic acid moiety was long-range coupled to C-4. These data clearly indicated the hydroxylation at C-3 of the 4-amino-3-isopropoxybenzoic acid unit. The remaining structure was also confirmed by HMBC data (Figure 1). Thus, the planar structure of 1 was determined as a new derivative hydroxylated at C-3 of 3. The relative configuration of the β-methoxyasparagine moiety of 1 was determined by comparison with those of 3 and was confirmed by 3JHH coupling constants and NOESY spectra. The large coupling constant (7.5 Hz) between H-2‴ and H-3‴ indicated the anti configuration. NOE correlations were observed between H-3‴and NH-8‴/NH-6‴ and between NH-6‴ and NHa/b-4‴ in DMSO-d6 (Figures 2, S20, and S21 and Table S1), similar to 3. Additionally, although the specific rotation [+16.5 (c 0.1, MeOH)] of 1 was different from that [−14.4 (c 0.08, MeOH)] of 3, the Cotton effect [[θ]25(nm) (MeOH): −4367 (251), 4666 (311)] of 1 was almost the same as that [[θ]25(nm) (MeOH): −5344 (251), 3999 (305)] of 3 in the circular dichroism (CD) spectra (Figures S34 and S36). These data clearly indicated that 1 had the same absolute stereochemistry as 3, but indicated that determination of the absolute stereochemistry by optical rotation was difficult in this class of compounds.12 The molecular formula of 2 was determined to be C46H44N6O15 on the basis of HRESIMS (Figure S22) in combination with 1H and 13C NMR data. The IR absorptions at 1684 cm−1 suggested the presence of a carbonyl moiety. The 1 H and 13C NMR data of 2 were similar to those of 3. The differences in 1H and 13C NMR (Table 1) with COSY and

■

RESULTS AND DISCUSSION Because the one-dimensional (1D) and two-dimensional (2D) NMR spectra of 3 were much better in both DMSO-d6 and CD3OD than those of 1 and 2, the chemical structure of 3 was first determined. The NMR data of cystobactamid 919-2 in DMSO-d6 have not been well studied due to signal broadening effects.12 The molecular formula of 3 was determined to be C46H45N7O14 on the basis of HRESIMS (Figure S1) in combination with 1H and 13C NMR data. The 1H and 13C NMR (Table 1), DEPT, COSY, and HMQC data (Figures S2− S5) of 3 demonstrated the presence of three 1,4-disubstituted benzenes, a 1,2,4-trisubstituted benzene, a 1,2,3,4-tetrasubstituted benzene, two isopropoxyl groups, two methines coupled to each other, a methoxy, seven carbonyls, and nine exchangeable protons. In the 1H−15N HMQC and HMBC spectra (Figures S7 and S8), six amide nitrogens (δ 109.0, 110.4, 119.1, 123.8, 131.6, and 135.5) and a nitro nitrogen (δ 369.5) were detected. From the 1H−13C and 1H−15N HMBC spectra (Figure 1), 3 consisted of five units, including two paraaminobenzoic acids, a para-nitrobenzoic acid, a 4-amino-3isopropoxybenzoic acid, a 4-amino-2-hydroxy-3-isopropoxybenzoic acid, and a β-methoxyasparagine. The connectivity of the five units was determined by 1H−13C and 1H−15N HMBC and NOESY spectral analyses. Thus, the planar structure of 3 was identified as cystobactamid 919-2.12 The relative configuration of the β-methoxyasparagine moiety of 3 was determined based on the 3JHH and nJCH coupling constants and NOESY spectrum. The large coupling constant (8.0 Hz) between H-2‴ and H-3‴ indicated the anti conformation (Table 2). The nJCH values were measured in DMSO-d6−CD3OD (4:1) to remove complicated splitting patterns from NH protons (Figure S9). The large coupling constant (6.7 Hz) between H-2‴ and C-3‴ indicated that H-2‴ and 3‴-OMe are gauche.13 The nJCH values to the carbonyl carbon were measured by an LSPD experiment (Figure S10). The small coupling constants 3JH‑2‴,C‑4‴ = 3.2 Hz and 3JH‑3‴,C‑1‴ < 2 Hz indicated that H-2‴/C-4‴ and H-3‴/C-1‴ are gauche, supporting the anti relationship of H-2‴ and H-3‴. Since it is difficult to determine the relative configuration by the JBCA method only, we analyzed NOE data. The NOE correlations between H-3‴ and NH-8‴/NH-6‴, between NH-8‴ and H3B

DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Table 1. NMR Spectroscopic Data for 1, 2, and 3 3 (400 MHz, DMSO-d6) position 1 2 3 4 5 6 7 8 9, 10 11 1′ 2′ 3′ 3′−OH 4′ 5′ 6′ 7′ 8′ 9′, 10′ 11′ 1″ 2″ 3″, 7″ 4″, 6″ 5″ 8″ 1‴ 2‴ 3‴ 4‴ 4‴ NHa NHb 5‴ 6‴ 1⁗ 2⁗ 3⁗, 7⁗ 4⁗, 6⁗ 5⁗ 8⁗ 1⁗′ 2⁗′ 3⁗′, 7⁗′ 4⁗′, 6⁗′ 5⁗′ 5⁗′ NO2 a

δC, type 166.9, 125.7, 113.9, 146.3, 133.3, 119.6, 122.6, 71.7. 21.6,

δN

C C CH C C CH CH CH CH3

δH (J in Hz)

7.56 brs

123.8

8.50, 7.58, 4.75, 1.37, 10.97,

d (8.3) d (8.3) m d (6.0) brs

163.6, C 116.4, C 150.3, C

3 (900 MHz, CD3OD) δC, type 170.5, 128.3, 115.2, 148.5, 134.1, 121.5, 124.1, 73.4, 22.2,

δH (J in Hz)

C C CH C C CH CH CH CH3

7.68, brs

8.48. 7.67, 4.78, 1.46,

d (8.8) d (8.8) m d (6.1)

1 (500 MHz, CD3OD) δC, type 175.6, 117.0, 156.1, 136.2, 137.0, 111.3. 126.4, 75.9, 22.6,

C C C C C CH CH CH CH3

δH (J in Hz)

7.81, 7.63, 4.80, 1.33,

d (9.0) d (9.0) m d (6.0)

2 (800 MHz, CD, OD) δC, type 170.0, 127.8, 115.3, 148.5, 134.5, 121.4, 124.2, 73.5, 22.5,

C C CH C C CH CH CH CH3

167.0, C 117.2, C 153.2, C

167.0, C 116.9, C 153.1, C

167.0, C 117.2, C 153.0, C

139.1, 137.8, 115.4, 125.7, 77.4, 22.6,

C C CH CH CH CH3

138.9, 137.5, 115.0, 125.4, 77.2, 22.6,

C C CH CH CH CH3

139.1, 137.8, 115.3, 125.8, 77.5, 22.8,

C C CH CH CH CH3

167.3, 131.0, 129.6, 121.2, 143.5,

C C CH CH C

167.1, 131.0, 129.5, 121.2, 143.4,

C C CH CH C

167.4, 131.8, 129.5, 121.6, 143.3,

C C C CH C

170.0, 57.6, 82.5, 174.8,

C CH CH C

169.9, 57.6, 82.5, 174.7,

C CH CH C

171.1, 57.1, 84.6, 173.5,

C CH CH C

δH (J in Hz)

7.67, brs

8.49, 7.68, 4.78, 1.46,

d (8.8) d (8.8) m d (5.6)

7.73, 7.79, 4.52, 1.35,

d (8.8) d (8.8) m d (6.4)

11.21, brs 138.4, 136.2, 115.3, 124.9, 75.6, 22.0,

C C CH CH CH CH3

164.3, 128.6, 128.4, 118.8, 142.1,

C C CH CH C

168.6, 55.7, 79.7, 170.8,

C CH CH CH

119.1

57.7, CH3 109.0

164.2, 140.4, 129.3, 123.5, 149.2,

C C CH CH C

7.47, 7.54, 3.31, 8.46,

brs brs s d (8.1)

7.90, da 7.90, da 131.6

7.73, 7.79, 4.53, 1.35,

d (8.3) d (8.3) m d (6.0)

7.97, d (8.6) 7.84, d (8.6)

7.74, 7.77, 4.55, 1.35,

d (9.0) d (9.0) m d (6.0)

7.97, d (8.8) 7.84, d (8.8)

7.95, d (8.8) 7.87, d (8.8)

10.56, s 4.92, dd (8.0, 8.1) 4.09, d (8.0)

110.4

C C CH CH C

d (8.5) d (8.5) m d (6.1) s

7.97, d (8.6) 7.83, d (8.6) 135.5

165.4, 128.9, 128.2, 119.6, 141.7,

7.50, 7.80, 4.32, 1.26, 9.58,

5.06, d (8.0) 4.18, d (8.0)

59.6, CH3 169.4, 130.9, 129.7, 121.4, 143.5,

C C CH CH C

167.0, 142.0, 130.3, 124.9, 151.9,

C C CH CH C

3.50, s

7.92, d (8.0) 7.89, d (8.0)

59.6, CH3 169.4, 130.9, 129.7, 121.3, 143.3,

C C CH CH C

166.9, 141.8, 130.2, 124.7, 151.4,

C C CH CH C

5.08, d (7.5) 4.18, d (7.5)

3.50, s

7.92, d (9.0) 7.90, d (9.0)

60.0, CH3 169.1, 130.7, 129.5, 121.5, 143.2,

C C CH CH C

166.9, 142.0, 130.9, 124.8, 151.4,

C C CH CH C

5.16, d (3.2) 4.42, d (3.2)

3.59, s

7.91, d (8.8) 7.88, d (8.8)

10.8, s

8.21, d (8.6) 8.38, d (8.6)

8.16, d (8.6) 8.38, d (8.6)

8.16, d (8.8) 8.38, d (8.8)

8.15, d (8.6) 8.37, d (8.6)

369.5

Overlapped signal.

Thus, the planar structure of 2 was determined as a new βmethoxyaspartic acid derivative of 3. The relative configuration of the β-methoxyaspartic acid moiety of 2 was determined by 3JHH and nJCH coupling constants and ROESY and NOE differential spectra. The small coupling constant (3.2 Hz) between H-2‴ and H-3‴ indicated the gauche conformation. The nJCH values were measured in DMSO-d6−CD3OD (4:1) by a HETLOC experiment (Figure S27). The small coupling constant (4.4 Hz) between H-2‴ and C-3‴ indicated that H-2‴ and 3-OMe are in an anti

HMQC data (Figures S23−25) were that the chemical shifts of H-2‴, H-3‴, and C-3‴ of the β-methoxyasparagine unit were upfield-shifted and that the 3JHH coupling constant between H2‴ and H-3‴ was reduced to 3.2 Hz in 2 (Table 2). Additionally, NHa/b-4‴ were not detected in 2 in DMSO-d6 (Table S1), suggesting the presence of β-methoxyaspartic acid, which was supported by IR absorption at 1684 cm−1 and the molecular formula. The HMBC spectrum (Table S1 and Figures S26 and S30) confirmed the remaining structure of 2. C

DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

The antibacterial activities of 1−3 in Gram-positive and Gram-negative bacteria were evaluated (Table 3). Compound 1 showed the most potent antibacterial activity against both Gram-positive and Gram-negative bacteria. Compound 1 showed MICs of 0.01−0.2 μg/mL against Gram-positive bacteria, including Staphylococcus aureus, methicillin-resistant Staphylococcus aureus (MRSA), Streptococcus pneumoniae, and Enterococcus faecalis; these MICs were about 100 and 10 times more potent than those of 2 and 3, respectively. Additionally, 1 showed MICs of 0.1−4 μg/mL against Gram-negative bacteria, including Escherichia coli, Pseudomonas aeruginosa, Acinetobacter baumanii, and Klebsiella pneumoniae; these MICs were about 40 and 2 times more potent than those of 2 and 3, respectively. The antibacterial activity of 1 was higher than that of ciprofloxacin, which is commonly used in the clinical setting, against all Gram-positive and Gram-negative bacteria tested except Pseudomonas aeruginosa and Klebsiella pneumoniae. Importantly, 1 still exhibited potent antibacterial activity against ciprofloxacin-resistant bacteria, such as quinolone-resistant S. aureus (QRSA) and E. coli CCARM 1356, with an MIC of 0.1 μg/mL, which was 10 times more potent than that of 3. Since cystobactamid has been reported to inhibit DNA gyrase, the antibacterial target of coralmycins is probably the same, as supported by the reduced activity of 1 and 3 against S. aureus strains resistant to quinolone, a DNA gyrase inhibitor, compared with MRSA and the wild-type bacteria. The cytotoxicity of 3 was evaluated in three human cell lines (HepG2, liver hepatoblastoma cells; MRC-9, fetal lung fibroblasts; and MCF10A, normal breast cells), with adriamycin as a positive control. Compound 3 was inactive against all the tested cells at 50 μg/mL, while adriamycin showed GI50 values of 0.04−0.08 μg/mL. In summary, coralmycins were identified as new potent antibacterial compounds against Gram-positive and Gramnegative pathogens. Coralmycin A, the hydroxylated derivative at C-3 of 3, showed 10 times higher activity than 3. Compound 3 showed 10 times higher activity than coralmycin B, a new βmethoxyaspartic acid derivative of 3. Thus, the β-methoxyasparagine moiety and the hydroxy group of the 4-amino-2hydroxy-3-isopropoxybenzoic acid unit were found to be important for the antibacterial activities of these compounds. The higher antibacterial activity of coralmycin A against quinolone-resistant bacteria than ciprofloxacin indicated that coralmycins could bind to a different site than ciprofloxacin on

Figure 1. Key HMBC, COSY, and NOE correlations of 3 and 1.

conformation.13 The 3JH‑2‴,C‑4‴ and 3JH‑3‴,C‑1‴ values were speculated to be small from the HMBC cross-peak intensity, indicating that H-2‴/C-4‴ and H-3‴/C-1‴ are in a gauche conformation. On the basis of the JBCA approach, the relative configuration of 2 was determined to be “S*S*” with the major conformation of C-1‴ and C-4‴ in an anti relationship. Additionally, strong ROE correlations between H-2‴and H-3‴ and the NOE from NH-6‴ to OMe supported the proposed stereochemistry, although there was an unexpected weak ROE bewteen H-2‴ and OMe′, suggesting the presence of the other rotamer as a minor conformer (Figures 2, S28, and S33). The specific rotation [+18.4 (c 0.1, MeOH)] of 2 was different from that [−14.4 (c 0.08, MeOH)] of 3, and the Cotton effect [[θ]25(nm) (MeOH): 266 (247), 4131 (290) ] of 2 was different from that [[θ]25(nm) (MeOH): −5344 (251), 3999 (305)] of 3 in the CD spectrum (Figure S35). Thus, these data supported that 2 was a new β-methoxyaspartic acid derivative with the diastereoisomeric configuration of 3.

Table 2. Comparison of NMR Data of the β-Methoxyasparagine Moiety in Cystobactamid 919−2, 3, and 2 cystobactamid 919-212 DMSO-d6

CD3OD

position

δC

1‴ 2‴ 3‴ 4‴ 4‴ NHa NHb 5‴ 6‴ l‴ 2‴ 3‴ 4‴ 5‴

168.4 55.4 79.8 170.6

57.4 169.6 57.2 82.1 174.5 59.1

3

δH (J in Hz)

δC

4.92, m 4.09, d (7.9)

168.6 55.7 79.7 170.8

7.48, 7.55, 3.31, 8.46,

s s s d (8.3)

57.7 170.0 57.6 82.5 174.8 59.6

5.08, d (7.4) 4.18, d (7.4) 3.50, s D

2 δH (J in Hz)

δC

4.92, t (8.0) 4.09, d (8.0)

168.7 54.9 82.6 170.1

7.47, 7.54, 3.31, 8.46,

s s s d (8.1)

5.06, t (8.0) 4.18, d (8.0) 3.50, s

58.9 171.1 57.1 84.6 173.5 60.0

δH (J in Hz) 4.87, brs 4.27, brs

3.44, s 8.37, brs 5.16, d (3.2) 4.42, d (3.2) 3.44, s DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 2. NOE correlations in the β-methoxyasparagine moiety of 1−3. solution contained 100 mg/L MnCl2·4H2O, 20 mg/L CoCl2, 10 mg/L CuSO4, 10 mg/L Na2MoO4·2H2O, 20 mg/L ZnCl2, 5 mg/L LiCl, 5 mg/L SnCl2·2H2O, 10 mg/L H3BO3, 20 mg/L KBr, 20 mg/L KI, and 8 g/L EDTA Na−Fe3+ salt (trihydrate). A piece of agar from the mature plate culture of the producing strain was inoculated into a 500 mL Erlenmeyer flask containing 100 mL of sterile seed liquid medium with the above composition and cultured on a rotary shaker (150 rpm) at 28 °C for 3 days. For the production of coralmycins, 25 mL of the seed culture was transferred into 1 L Erlenmeyer flasks containing 250 mL of the above medium containing 10 g of Amberlite XAD16 (Sigma, USA) and then cultivated for 9 days using the same conditions. The resin and cells were recovered and extracted twice with 50% acetone and then twice with 100% acetone. The acetone was evaporated from the extract, and the remaining aqueous phase was extracted successively with chloroform and ethyl acetate. The ethyl acetate extract showed antibacterial activity against P. aeruginosa. The ethyl acetate extract from the 50 L cultures was concentrated in vacuo. The resulting residue (0.67 g) dissolved in MeOH was applied to thinlayer chromatography (TLC) on precoated silica gel 60 F254 plates (Merck No 1.05715.0001, Darmstadt, Germany) with CHCl3−MeOH (5:1) to give two active bands (bands I and II), with Rf values of 0.16 and 0.58, respectively. Band I was further purified by TLC on silica gel 60 RP-18 F254 plates (Merck No. 1.15389.0001) with CH3CN−H2O (50:50) to yield 1 (1.6 mg) and 2 (3.0 mg), with Rf values of 0.21 and 0.19, respectively, as yellowish powders. Band II was also purified by RP-18 TLC developed with CH3CN−H2O (65:35) to give an active band with an Rf value of 0.51. The active band was finally purified using an ODS HPLC column (20 × 150 mm, S-4 μm, YMC C18) with CH3CN−H2O (60:40) containing 0.1% trifluoroacetic acid (TFA) at a flow rate of 4 mL/min to give 3 (8.5 mg) as a yellowish powder with a retention time of 17.2 min. Compounds 1 and 2 were also determined to be single peaks at 210 nm using an analytical HPLC column (4.6 × 150 mm, S-4 μm, YMC C18) with CH3CN−H2O (50:50) containing 0.01% TFA at a flow rate of 0.8 mL/min with retention times of 21.1 and 19.5 min, respectively. Compound 1: yellow powder; [α]25D = +16.5 (c 0.1, MeOH); λmax nm (log ε) in MeOH 212 (4.58), 259 (sh) (4.20), 305 (4.36), 307 (4.36); IR (KBr) 3427, 2926, 1645, 1525, 1274 cm−1; CD (MeOH) [θ]25(nm) (MeOH) −4367 (251), 4666 (311); HRESIMS m/z 934.2902 (M − H)−, C46H44N7O15 requires 934.2895. Compound 2: yellow powder; [α]25D = +18.4 (c 0.1, MeOH); λmax nm (log ε) in MeOH: 202 (4.79), 266 (sh) (4.16), 299 (4.26), 317 (4.26); IR (KBr) 3415, 2920, 2852, 1684, 1599, 1516, 1268, 1187 cm−1; CD (MeOH) [θ]25(nm) (MeOH) 266 (247), 4131 (290); HRESIMS m/z 943.2753 (M + Na)+, C46H44N6O15Na requires 943.2762; m/z 919.2764 (M − H)−, C46H43N6O15 requires 919.2786. Compound 3: yellow powder; [α]25D = −14.1 (c 0.08, MeOH); CD (MeOH) [θ]25(nm) (MeOH) −5344 (251), 3999 (305); HRESIMS m/z 920.3083 (M + H)+, C46H46N7O14 requires 920.3103. Determination of Antibacterial Susceptibility. Whole-cell antimicrobial activity was determined using broth microdilution as described previously.14 Drug-resistant pathogens, including methicillin-resistant S. aureus CCARM 3167, quinolone-resistant S. aureus CCARM 3505, and Escherichia coli CCARM 1356, were obtained from the Culture Collection of Antimicrobial Resistant Microbes of Korea.

Table 3. Antibacterial Activities of 1−3 MIC (μg/mL) test organism Staphylococcus aureus RN 4220 MRSA CCARM 3167 MRSA CCARM 3506 QRSA CCARM 3505 QRSA CCARM 3519 Streptococcus pneumonia KCTC 5412 Enterococcus faecalis KCTC 5191 E. coli CCARM 1356 E. coli KCTC 1682 Pseudomonas aeruginosa KCTC 2004 Acinetobacter baumannii KCTC 2508 Klebsiella pneumoniae KCTC 22057

1

2

3

>16

0.063 0.125 0.125 1 2 2

0.03 0.125 0.125 4

4 16 4 >16

0.125 1 0.5 8

0.5 64 0.003 0.06

0.125

4

0.25

0.125

8

0.006

2

2 1 1

ciprofloxacin

0.015 0.015 0.015 0.125 0.25 0.25

>16

0.125 2 128 128 0.5

DNA gyrase or may inhibit another antibacterial target as well. Coralmycin A has great potential for treatment of multidrugresistant bacteria, including Gram-negative bacteria.

■

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were determined on a JASCO P-1020 polarimeter. CD spectra were determined on a J720 JASCO CD spectropolarimeter. UV spectra were measured on a Shimadzu UV-1601 UV−visible spectrophotometer. IR spectra were obtained using a Bruker EQUINOX 55 spectrophotometer. NMR spectra were recorded on a Bruker Biospin Avance 400, 500, 700, 800, or 900 MHz spectrometer (Korea Basic Science Institute). HRESIMS data were recorded on a JEOL JMSHX110/110A mass spectrometer. Isolation and Identification of Myxobacterium M23. Strain M23 was isolated from soil collected in Incheon City, Korea. M23 was a rod-shaped, Gram-negative bacterium that moved by gliding motility and grew using live Escherichia coli cells as the sole nutrient source. The cells of M23 formed characteristic fruiting bodies that were unique to the myxobacterium C. coralloides on WC medium containing 10 mM 3-(N-morpholino)propanesulfonic acid (pH 7.6), 0.1% CaCl2· 2H2O, and 1.5% agar, suggesting that the isolate was C. coralloides. Phylogenetic analysis, performed using the 16S rRNA sequence of the isolate, also supported that the isolate was C. coralloides. The 1479-bp 16S rRNA sequence (GenBank KX588243) of the isolate was 99.86% identical to that of the C. coralloides type strain DSM 2259(T). Therefore, we concluded that the isolate was C. coralloides. Fermentation and Isolation. Fermentation was carried out in CYS medium containing 0.5% casitone, 0.1% yeast extract, 0.3% soluble starch, 0.1% MgSO4·7H2O, 0.05% CaCl2, 50 mM 4-(2hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 0.4% trace element solution, and 0.5 μg/mL cyanocobalamin. The trace element E

DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

(10) Lee, C.; An, D.; Lee, H.; Cho, K. J. Microbiol. Biotechnol. 2013, 23, 297−303. (11) Hyun, H.; Chung, J.; Kim, J.; Lee, J. S.; Kwon, B. M.; Son, K. H.; Cho, K. J. Microbiol. Biotechnol. 2008, 18, 1416−1422. (12) Baumann, S.; Herrmann, J.; Raju, R.; Steinmetz, H.; Mohr, K. I.; Huttel, S.; Harmrolfs, K.; Stadler, M.; Muller, R. Angew. Chem., Int. Ed. 2014, 53, 14605−14609. (13) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.; Tachibana, K. J. Org. Chem. 1999, 64, 866−876. (14) Zheng, C. J.; Sohn, M. J.; Lee, S.; Kim, W. G. PLoS One 2013, 8 (e78922), 920.3103.

Most of the test strains were grown to mid log phase in MuellerHinton broth and diluted 1000-fold in the same medium. Cells (105/ mL) were inoculated into Mueller-Hinton broth and dispensed at 0.2 mL/well in 96-well microtiter plates. Streptococcus pneumonia and Acinetobacter baumanii were grown in Todd-Hewitt medium and nutrient broth, respectively. The test compounds and ciprofloxacin (Sigma) were soluble in DMSO, the final concentration of which did not exceed 0.05% in the cells. Cells were treated with 0.05% DMSO as a vehicle control or test compounds. The MICs were determined in triplicate by serial 2-fold dilutions of the test compounds. The MIC was defined as the concentration of a test compound that completely inhibited cell growth during a 24 h incubation period at 37 °C. Bacterial growth was determined by measuring the absorption at 650 nm using a microtiter enzyme-linked immunosorbent assay (ELISA) reader. Cytotoxicity Assay. Three human cell lines (HepG2, MRC-9, and MCF10A) were purchased from ATCC (Manassas, VA, USA). The cells (less than passage 20) were grown in RPMI 1640 containing 10% newborn calf serum and seeded in 96-well plates at a concentration of 1 × 104 cells/well 1 day before the start of treatment. Cells were treated with compounds at various concentrations for 48 h. Cell viability was evaluated by the standard sulforhodamine B procedure. GI50 values of the tested compounds were calculated using GraphPad Prism v4.0 software.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.6b00294. 1D NMR, 2D NMR, and CD spectra of coralmycin A, coralmycin B, and compound 3 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel (W.-G. Kim): 82-42-860-4298. Fax: 82-42-879-8103. Email: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2012R1A2A2A01014821) and the KRIBB Research Initiative Program, Republic of Korea.

■

REFERENCES

(1) Rice, L. B. J. Infect. Dis. 2008, 197, 1079−1081. (2) Boucher, H. W.; Talbot, G. H.; Benjamin, D. K., Jr.; Bradley, J.; Guidos, R. J.; Jones, R. N.; Murray, B. E.; Bonomo, R. A.; Gilbert, D. Clin. Infect. Dis. 2013, 56, 1685−1694. (3) Giske, C. G.; Monnet, D. L.; Cars, O.; Carmeli, Y.; ReAct-Action on Antibiotic, R. Antimicrob. Agents Chemother. 2008, 52, 813−821. (4) Boucher, H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E.; Gilbert, D.; Rice, L. B.; Scheld, M.; Spellberg, B.; Bartlett, J. Clin. Infect. Dis. 2009, 48, 1−12. (5) Weissman, K. J.; Muller, R. Bioorg. Med. Chem. 2009, 17, 2121− 2136. (6) Wenzel, S. C.; Muller, R. Mol. BioSyst. 2009, 5, 567−574. (7) Weissman, K. J.; Muller, R. Nat. Prod. Rep. 2010, 27, 1276−1295. (8) Wenzel, S. C.; Muller, R. Curr. Opin. Drug Discovery Dev. 2009, 12, 220−230. (9) Kim, D.; Chung, J.; Hyun, H.; Lee, C.; Lee, K.; Cho, K. J. Microbiol. Biotechnol 2009, 19, 1288−1294. F

DOI: 10.1021/acs.jnatprod.6b00294 J. Nat. Prod. XXXX, XXX, XXX−XXX