Antibacterial Meroterpenoids, Merochlorins G–J from the Marine Bacterium Streptomyces sp.

Four new chlorinated meroterpenoids, merochlorins G−J (1−4), and 10, a dihydronaphthalenedione precursor, along with known merochlorins A (5) and C−F (6−9), were obtained from cultivation of the bacterium strain Streptomyces sp. CNH-189, which was isolated from marine sediment. The planar structures of compounds 1−4 and 10 were elucidated by interpretation of MS, UV, and NMR spectroscopic data. The relative configurations of compounds 1−4 were determined via analysis of nuclear Overhauser effect (NOE) spectroscopic data, after which their absolute configurations were established by comparing the experimental electronic circular dichroism (ECD) spectra of compounds 1−4 to those of previously reported possible enantiomer models and DP4 calculations. Compound 3 displayed strong antibacterial activities against Bacillus subtilis, Kocuria rhizophila, and Staphylococcus aureus, with MIC values of 1, 2, and 2 μg/mL, respectively, whereas compound 1 exhibited weak antibacterial effects on these three strains, with a 16−32 μg/mL MIC value range.

Lastly, the connection of these two fragments was secured by the interpretation of the HMBC correlations. Correlations from the methylene protons H-11 (δ H 2.61, 1H, d, J = 14.0 Hz, δ H 2.37, 1H, d, J =14.0 Hz) to C-1 and from an exchangeable proton 10-OH to C-9, C-10, and C-11 allowed for the attachment of C-10/C-11 and the placement of the hydroxy group at C-10. The HMBC correlations from a methyl singlet H 3 -26 (δ H 1.51, 3H, s) to C-8, C-9, and C-10 allowed for the placement of the methyl substituent at C-9. Carbon chemical shifts at δ C 74.5 and 66.4; the isotope ratio (9:6:1) of three pseudomolecular ion peaks [M + H]   Merochlorin H (2) was isolated as a pale-yellow oil, and its molecular formula was assigned as C 26 [M + H + 4] + in the LR-ESI-MS spectrum also supported the presence of two chlorine atoms in the molecule. The 1 H NMR spectrum of 2 was almost identical to that of 1, except for the terminal structure of the rearranged sesquiterpene moiety in the terminal fragment. The largest difference in the 1 H NMR spectrum was attributed to the methyl singlet proton signal H 3 -20 (δ H 1.12, 3H, s) instead of the geminal olefinic protons in 1. The long-ranged HMBC correlations from H 3 -20 and H 3 -21 (δ H 1.18, 6H, s) to carbons C-18 (δ C 69.8) and C-19 (δ C 71.7) and the carbon chemical shift of C-19 indicated that 2 has a hydroxy group at C-19 (Table 1). Merochlorin I (3) was isolated as a pale-yellow oil, and its molecular formula was assigned as C 26 H 33 35 Cl 3 O 5 based on the observation of a pseudomolecular ion peak at m/z 553.1281 [M + Na] + in the HRESIMS spectroscopic data (calcd for C 26 H 34 35 Cl 3 O 5 Na, 553.1291). The 1 H and 13 C NMR data for 3 were almost identical to those of 2, suggesting that they share the same carbon backbone. Comparisons and interpretation of the 2D NMR spectroscopic data of 3 to those of 2 indicated that 3 also possesses dihydronaphthalenedione and rearranged sesquiterpene moieties. The isotope ratio (27:27:9:1) Table 1).
The relative stereochemistry of 1-4 was assigned by analysis of NOE spectroscopic data. The NOE crosspeak between 10-OH and the methyl singlet protons H 3 -26 suggested that these protons should be located at the same side on the dihydronaphthalenedione moiety. The chemical structures of 1−4 indicated that they are modified biosynthetic products of 7. This was supported by the conclusion that the C-9 and C-10 stereochemistry of 1−4 was likely the same as that of 7. To determine the absolute configurations of C-9 and C-10 in 1-4, the experimental ECD spectral data of 1-4 were measured (Figure 4). The ECD spectra of 1-4 exhibited almost identical patterns, suggesting that C-9 and C-10 of the dihydronaphthalenedione moiety in 1-4 have the same absolute configurations. Further, comparing their ECD spectra to those of previously reported enantiomer models by our group allowed us to suggest that C-9 and C-10 in 1-4 are in 9S, 10R configurations [19]. Furthermore, compounds 1, 2, 3, and 6 shared the same chemical scaffold and were isolated from the same bacterial strain. These observations suggested that the four compounds likely share the same absolute configuration for C-18. The absolute configurations of C-18 in 1 and of C-15 and C-18 in 4 were determined via DP4 calculations [21]. DP4 calculation is a computational method based on quantum mechanics for confirmation of the absolute configuration using NMR chemical shifts. It was required to design the models of two possible diastereomers of 1 (18R and 18S) and four possible diastereomer models of 4 (15R/18R, 15R/18S, 15S/18R, and 15S/18S). As a result, DP4 analysis indicated that 1 had an 18R configuration with 86% probability (Figure S22), whereas 4 had 15R, 18R configurations with a 100% probability ( Figure S23). Previous studies have proposed plausible biosynthetic pathways for tetrahydroxynaphthalene (THN)-derived meroterpenes from strain CNH-189, such as merochlorins, ansalactams, and meroindenon [17][18][19]. These pathways started from THN coupled with a C 15 isoprene unit with atypical modifications, including a Baeyer-Villiger-style oxidation, a Paterno-Büchi-type 2 + 2 cycloaddition, and a pinacol-type contraction. All instances of atypical enzymatic biosynthesis are good examples of the biosynthetic potential of this marine microorganism.
There are two classes of bacterial THN-derived meroterpenoids, depending on the attachment of an isoprene chain to the THN frame [22]. Class I includes neomarinone, merochlorin A (5), and merochlorin B, which have a modified isoprene unit at the C-2 or C-4 carbon of the THN [23]. Class II compounds include merochlorins G-J (1)(2)(3)(4) and are characterized by the attachment of an isoprene unit at the C-3 carbon of THN. Compounds 1, 2, and 3 are chlorinated products of merochlorin D (7), and compound 4 is an intramolecular cyclization product with further amination at C-19. Interestingly, chlorination at the isoprene chain of 1-3 occurred specifically at C-18. A tetrahydrofuran ring moiety identified in 4 was likely a biosynthesis product of an epoxide-opening reaction that starts with the selective enzymatic epoxidation of the double bonds [23][24][25]. Based on an AntiMarin2013 database search, the amine group at the tail of the isoprene chain of 4 was the first reported marine microbial natural product to possess this modification. It is also a rare case of an amination product from a tertiary alcohol that was produced by an epoxide-opening reaction. There are some synthetic examples of amination in the isopropyl group [26]. Lastly, compound 10 could be synthesized from the reduction of THN to hydroxynaphthoquinone, followed by methylation at C-3 [18,27].
Compounds 1-10 were tested for their antibacterial activities against six pathogenic bacteria, including three Gram-positive bacteria (Bacillus subtilis KCTC 1021, Kocuria rhizophila KCTC 1915, andStaphylococcus aureus KCTC 1927) and three Gram-negative bacteria (Escherichia coli KCTC 2441, Salmonella typhimurium KCTC 2515, and Klebsiella pneumoniae KCTC 2690). Merochlorin I (3) displayed strong antibacterial activities against B. subtilis, K. rhizophila, and S. aureus, with MIC values of 1, 2, and 2 µg/mL, respectively. Merochlorin G (1) was found to exhibit moderate antibacterial activities against the pathogenic strains, with a 16-32 µg/mL MIC range. Compound 10 exhibited strong antibacterial activity against K. rhizophila and moderate activity against B. subtilis, with MIC values of 2 and 32 µg/mL, respectively. Meanwhile, merochlorins H (2) and J (4) did not show any significant antibacterial activities against the six tested pathogens, with MIC values up to 128 µg/mL ( Table 3). The presence of a polar moiety at the isoprene chain (C-19) abrogated the antibacterial properties of merochlorins, which was demonstrated with the hydroxy group in 2 and the amine group in 4. This was also supported by the strong antibacterial activity of 7-9, which possess nonpolar isoprene chain moieties. Therefore, the antibacterial activity against Gram-positive bacteria of these compounds depended on the hydroxy group at the THN core [28,29]. Interestingly, merochlorin A (5) exhibited no antibacterial activity against K. rhizophila, with MIC values up to 128 µg/mL, but exerted a strong antibacterial activity against two other Gram-positive bacteria (B. subtilis and S. aureus) [30]. Table 3. Antibacterial activities of merochlorins (1-10) a .

Gram-Positive Bacteria
Gram-Negative Bacteria

General Experimental Procedures
Optical rotations were acquired using a Kruss Optronic P-8000 polarimeter with a 5 cm cell. UV spectra were recorded in a Varian Cary UV-visible spectrophotometer with a path length of 1 cm, and IR spectra were recorded on a Perkin-Elmer 1600 FT-IR spectrometer. Low-resolution LC/MS measurements were performed using the Agilent Technologies 1260 quadrupole and Waters Micromass ZQ LC/MS system using a reversedphase column (Phenomenex Luna C18 (2) 100 Å, 50 mm × 4.6 mm, 5 µm) at a flow rate of 1.0 mL/min at the National Research Facilities and Equipment Center (NanoBioEnergy Materials Center) at Ewha Womans University. CD spectra were recorded using an Applied Photophysics Chirascan-Plus circular dichroism spectrometer (Applied Photophysics Ltd., Leatherhead, Surrey, UK). 1 H and 2D NMR spectra data were recorded at 400 and 800 MHz in DMSO-d 6 and CDCl 3 solutions containing Me 4 Si as an internal standard on Varian Inova spectrometers. 13 C NMR spectra were acquired at 100 or 200 MHz on a Varian Inova spectrometer. High-resolution EI-MS spectra were acquired using a JEOL JMS-AX505WA mass spectrometer at Seoul National University.

Collection and Phylogenetic Analysis of Strain CNH-189
The marine-derived actinomycete strain CNH-189 was isolated from a marine sediment sample collected near Oceanside, California. The strain was identified as a Streptomyces sp. based on 16S rRNA gene sequence analysis (accession no. HQ214120).

Extraction and Isolation
Sterilized XAD7HP resin (20 g/L) was added after 24 h of cultivation, and the culture was incubated for an additional 5 days at 27 • C. Once the bacteria were cultured in the presence of XAD7HP resin, the resin was collected on cheesecloth, washed with deionized water, and eluted with acetone. The acetone was removed under reduced pressure, and the resulting aqueous layer was extracted with ethyl acetate (3 × 500 mL). The ethylacetate-soluble fraction was dried in vacuo to yield 4.5 g of crude extract. The crude extract was fractionated by open-column chromatography on silica gel (25 g) and then eluted with a step gradient of dichloromethane and methanol to obtain seven fractions. The first fraction was purified by reversed-phase HPLC (Phenomenex Luna C-18 (2), 250 × 100 mm 2 , 2.0 mL/min, 5 µm, 100 Å, UV = 254 nm) using an isocratic solvent system to 90% CH 3 CN in water to render merochlorins G (1, 4.2 mg, t R = 26.1 min) and H (2, 3.0 mg, t R = 23.2 min). The fourth fraction was subjected to reversed-phase HPLC chromatography eluting with 82% CH 3 CN in water to obtain merochlorin I (3, 5.8 mg, t R = 30.2 min). The third fraction was purified using 64% CH 3 CN in water to obtain merochlorin J (4, 2.0 mg, t R = 31.2 min). To isolate compound 10, strain CNH-189 was regrown in 80 L scale. The treatment of the cultivation and crude extract fractionation was the same as the first culture. The first fraction was purified by reversed-phase HPLC (Phenomenex Luna C-18 (2), 250 × 100 mm 2 , 2.0 mL/min, 5 µm, 100 Å, UV = 254 nm) using an isocratic solvent system from 82% CH 3 CN in water to obtain compound 10 (3.3 mg, t R = 6.5 min). Merochlorins A (5, 2.6 mg,

Conformational Search and DP4 Calculations
A conformational search of merochlorins G (1) and J (4) was carried out by Macro-Model with the Merck molecular force field (gas phase), a 10 kJ/mol upper energy limit, and a 0.001 kJ (mol Å) −1 convergence threshold on the rms gradient to minimize computational complexity and expense. In the case of 1, 66 conformers were obtained for the 18R diastereomer and 37 conformers were obtained for the 18S model under the 10 kJ/mol limit of molecular potential energy. In addition, in the case of 4, 10 conformers were obtained for the 15R/18R diastereomer, 5 conformers for the 15R/18S diastereomer, 2 conformers for the 15S/18R diastereomer, and 7 conformers for the 15S/18S diastereomer. The Boltzmann population was calculated based on the potential energy of each conformer. Ground-state geometry optimization was performed by density functional theory (DFT) modeling with TurbomoleX 4.3.2. The basis set was def-SV(P) for all atoms, and the level of theory was B3-LYP at the functional level in the gas phase. Calculated chemical shifts of 1 H and 13 C were averaged by the Boltzmann populations. The experimental chemical shifts compared to these Boltzmann-averaged chemical shifts and the DP4 analyses indicated the 18R configuration of 1 with 86.6% and 15R/18R configurations of 4 with 100.0% probability.

Antibacterial Assay
Antibacterial susceptibility was tested against three Gram-positive bacteria (Bacillus subtilis KCTC 1021, Kocuria rhizophila KCTC 1915, andStaphylococcus aureus KCTC 1927) and three Gram-negative bacteria (Escherichia coli KCTC 2441, Salmonella typhimurium KCTC 2515, and Klebsiella pneumoniae KCTC 2690), as described in a previous study with modifications [16]. These bacteria were grown in Mueller-Hinton broth at 37 • C and 225 rpm for 24 h. Compounds 1-10 and positive controls were dissolved in DMSO, and 100 µL of each solution was dispensed into the wells of 96-well plates starting at 128 µg/mL concentration. Compounds 1-10 and positive controls were serially diluted, and Mueller-Hinton broth was added to a final concentration of 0.5% McFarland standard. The 96-well microtiter plates were cultivated for 24 h at 37 • C. The minimal inhibitory concentration (MIC) was defined as the lowest positive control concentration that visibly inhibited bacterial growth [31].

Conclusions
In conclusion, an intensive chemical screening of the Streptomyces sp. strain CNH-189 resulted in the discovery of new merochlorin derivatives (1)(2)(3)(4) and a proposed biosynthetic merochlorin precursor 10, along with known congeners 5-9. Further, compound 3 exhibited strong antibacterial activities against Gram-positive strains such as B. subtilis, K. rhizophila, and S. aureus, with MIC values of 1-2 µg/mL.  Institutional Review Board Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.