2.3. Structure Elucidation
Chemical screening, monitored by thin-layer chromatography (TLC), of the bacterial extract exhibited several bands with a wide range of polarities. Non-UV-absorbing compounds were detected as intensive pink to violet bands with anisaldehyde/sulfuric acid. UV absorbing bands exhibited pink, yellow, or brown coloration with anisaldehyde/sulfuric acid. Separation of the produced metabolites by the strain was carried out using a series of chromatographic techniques. The physicochemical properties of terretonin N (1) are listed in Table 1.
Compound 1, a moderately polar colourless solid, appeared on TLC as an intensive pink spot that changed to violet on spraying with anisaldehyde/sulfuric acid and heating. As it did not show UV absorption or fluorescence, characteristic for conjugated systems, we supposed it to be a terpenoid . The molecular formula of 1 was determined as C26H38O7 by negative mode ESI-HR-MS (Table 1), indicating the presence of eight double bond equivalents (DBE).
The 1H-NMR spectrum together with HSQC data (Table 2) exhibited six methyl singlets at δH 2.03 (s, H3-23), 1.96 (s, H3-27), 1.68 (s, H3-20), 1.62 (s, H3-24), 1.28 (s, H3-22), 0.86 (s, H3-21), and one methyl doublet at δH 1.38 (d, 6.4 Hz, H-25). An oxymethine group resonating at δH 4.81 (m, H-17) was shown to be attached to a methyl group (H-25) based on the coupling constant (6.4 Hz) and H,H COSY crosspeaks. A pair of two singlets at δH 5.06 (s, H-19a) and 5.16 (s, H-19b) suggested a vinylidene moiety. Signals of further three oxymethine groups were detected; two of them are multiplet at δH 4.81 (H-11) and triplet (3.0 Hz) at δH 4.55 (H-3), while the third one resonated as triplet at δH 4.35 (H-6). A singlet of a non-oxygenated methine group δH 2.58 (s, H-14), as well as two doublets of doublets at δH 2.74 (dd, 14.5, 2.7 Hz, H-7a) and 1.19 (m, H-7b) corresponding to a methylene group, were detected. Further multiplet signals were detected at δH 2.08–0.95 with an integral of 6H, corresponding to two additional methylene groups (H2-1, H2-2) and two methine groups (H-5, H-9).
The 13C/DEPT/HSQC NMR spectra showed 26 resonance signals, corresponding to seven methyl groups (δC 24.2–14.6 [C-20–25,27]), three methylene groups (δC 50.3–22.5 [C-1,2,7]), and seven methine groups, among them four sp3 oxymethine groups (δC 79.4–76.0 [C-3,6,11,17]) and three non-oxygenated ones (δC 61.3–51.5 [C-5,9,14]). Nine quaternary carbons were found, among them one ketone carbonyl (206.8, C-18), two ester carbonyls of an α-lactone (δC 169.4, C-15) and an acetate residue (δC 170.6, C-26), and two sp2 carbons for an exocyclic double bond (δC 147.9 [C-12], 114.5 [C-19]). Further four quaternary carbons resonate in the aliphatic region (δC 50.0–37.1 [C-4,8,10,13]). The evidence mentioned above and the eight double bond equivalents deduced from the molecular formula suggested that compound 1 is a tetracyclic compound.
According to H,H-COSY and HMBC experiments (Figure 6), compound 1 shows the following structural features: In ring A, the acetate residue is connected to the hydroxyl group at position 3 due to a visible 3J HMBC correlation between H-3 and C-26 (170.6). The gem-methyls (C-21,22) are attached to C-4. In ring B, the clear HMBC correlations from H-6 (4.35) to C-10 (38.1) and C-8 (37.1), and the H,H-COSY crosspeaks between H-6 and both neighboring protons (H-5 [1.25] and H-7 [2.74, 1.19]), completed the assignment. The HMBC correlations between H3-20 (1.68) and carbons C-1, 5, 9, and 10 established the fusion between rings A and B via C-5 and C-10. Alternatively, the direct fusion between rings B and C was proven through carbons C-8 and C-9 as the existence of an HMBC correlation from H3-23 (2.03) versus carbons C-7,8,9, and 14. The assignment of the vinylidene moiety at C-12 is based on the 3J HMBC correlations of corresponding methylene protons (H-19a/19b [5.06, 5.16]) with C-11 and C-13. The connectivity between rings C and D through C-13 and C-14 was verified by the HMBC correlation directed from the methyl H3-24 (1.62) towards C-12, 13, 14 and 18. The keto-lactone of ring D was identified by the 3J HMBC crosspeak between H-17 (4.81) and the lactone carbonyl C-15 (169.4), while the doublet methyl H3-25 (1.38) exhibited a 3J correlation towards the ketone carbonyl C-18 (206.8). According to the chromatographic behavior and based on the spectroscopic and mass spectrometric analyses together with data base searches, structure 1 was confirmed as a new meroterpenoidal/sesterterpenoidal compound and was given the trivial name terretonin N.
The relative stereochemistry of compound 1 was deduced from NOESY cross-peaks: H-3, CH3-21, CH3-20, CH3-23, CH3-24, and CH3-25 are located on one side of the molecule, while H-5, H-9, H-14, H-11, H-17, and CH3-22 point to the opposite side (Figure 7). Furthermore, the coupling constant of H-3 (J ~ 3.0 Hz) is indicative for its equatorial configuration, and, hence, the acetoxy residue has an axial orientation. The absolute configuration of terretonin N was established by X-ray structure determination (anomalous dispersion) of a single crystal grown from an acetone solution. The chiral carbons C-3, C-5, C-6, C-8, C-9, C-10, C-11, C-13, and C-17 have (S) configuration, while C-14 is the only stereocenter that has an (R) configuration (Figure 8).
Since the discovery of terpenes (more than 150 years ago), researchers have reported approximately 50,000 different terpenes derived from plants and fungi. Bacteria and other microorganisms are known to produce terpenes as well. However, they have received much less attention [31,32]. Cane et al. (Brown University, Providence, RI, USA) have proven the genetic capacity of bacteria to create terpenes with structural diversity [33,34,35,36]. During their studies 15 years ago, they have concluded, through the genome data gathered from many Streptomyces spp., that the latter have gene sequences that are similar to those encoding terpene synthases in plants and fungi. Consequently, they proved that actinobacteria indeed have genes for terpene synthases and that those enzymes could be used to make terpenes [37,38].
Terretonins are meroterpenoids biosynthetically derived from 3,5-dimethylorsellinic acid (DMOA), which is cyclized by terpene cyclase to form a terpenoid . Recently, it has been reported that bacteria harbor numerous genes coding for terpene cyclases , in addition to terpene synthases, which are widely distributed in bacteria as well . This underscores the eminent capability of bacteria to produce terpenoids with high similarity to those obtained from fungi. For example, merochlorin A, a sesterterpenoid, has been isolated from Streptomyces sp. . Further intensive search in the literature and in the National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/protein/) revealed that thousands of gene cluster homologs encoding enzymes proposed to be involved in terretonin biosynthesis  have been identified in actinobacteria. These enzymes include terpene synthase and terpene cyclase, aromatic prenyltransferase, monooxygenase, polyketide synthases, methyltransferase, phytanoyl-CoA dioxygenase, epoxidase, cytochrome P450 monooxygenases, and short chain dehydrogenases. Such kinds of enzymes commonly exist in the bacterial systems as well. Additional studies and approaches are required to support the production of terretonins from bacteria, including the study of the full bacterial genome to characterize the gene cluster for terretonin biosynthesis.
Infectious diseases are the leading cause of death worldwide. Emerging infections due to methicillin resistant Staphylococcus aureus (MRSA) pose a significant threat to patients [1,2]. It has been estimated that in the United States more people die from MRSA related infections than from HIV . Infections involving drug resistant bacteria are more difficult to treat due to increased costs and decreased efficacies [4,5]. One important approach to drug discovery for the treatment of MRSA is through natural products research.
Marine sponges of the genus Dysidea (order Dictyoceratida, family Dydideidae) have proven to be prolific producers of structurally diverse secondary metabolites, such as sesequiterpene quinones [6,7,8], sesquiterpenoids , diterpenoids , sterols , and polychlorinated compounds [12,13,14]. These metabolites showed a spectrum of interesting biological activities, including antifungal , antibacterial , antitumor [17,18], anti-inflammatory [15,19], and antioxidative activities .
In our efforts to search for new anti-MRSA agents from marine sponges collected from the South China Sea, chemical investigation of an active fraction from the sponge Dysidea sp. resulted in the isolation of a novel meroterpenoid, dysidinoid A (1) (Figure 1). It is the first meroterpenoid from Nature bearing a 9,4-friedodrime skeleton and a 2,5-dionepyrrole. Antibacterial evaluation showed that dysidinoid A showed potent antibacterial activity against two strains of pathogenic bacteria MRSA with MIC90 values of 8.0 μg/mL against both. Details of structural elucidation and antibacterial activity of dysidinoid A (1) were reported herein.
Figure 1. The chemical structure of dysidinoid A (1).
Figure 1. The chemical structure of dysidinoid A (1).
2. Results and Discussion
Dysidinoid A (1) was obtained as colorless needles with +35.4 (c 0.50, MeOH). Its IR spectrum showed absorption bands assignable to amide (3276 cm−1) and carbonyl (1775 and 1714) functionalities. The positive ESIMS of 1 exhibited quasimolecular ion peaks at m/z 302.2 [M+H]+ and 324.2 [M+Na]+, respectively. The molecular formula of C19H27NO2 with seven degrees of unsaturation, was deduced from HRESIMS at m/z 324.1941 [M+Na]+ (calcd. for C19H27NO2, 324.1939), which was supported by the 1H- and 13C-NMR data (Table 1). The 1H-NMR spectrum of 1 showed resonances attributable to two olefinic protons at δH 5.16 (H-3) and 6.26 (H-18), three tertiary methyl groups at δH 1.55 (H3-11), 1.00 (H3-12), and 0.88 (H3-14), a secondary methyl group at δH 0.95 (H3-13). In addition, the spectrum showed resonances due to an exchangable amine proton at δH 7.33 (20-NH), as well as partially overlapping signals with complex coupling patterns between δH 1.08 and 2.61 that could be attributed to several aliphatic methylene and methine units. The 13C-NMR and DEPT spectra of 1 showed 19 carbon resonances, corresponding to two carbonyl groups (δC 171.7 and 170.4), two olefinic quaternary carbons (δC 143.9 and 147.9), two aliphatic quaternary carbons (δC 38.3 and 42.4), two olefinic methine carbons (δC 120.5 and 130.4), two aliphatic methine carbons (δC 37.4 and 47.0), five aliphatic methylene carbons (δC 19.0, 26.3, 36.2, 27.4, and 32.5), and four methyl carbons (δC 17.7, 19.8, 16.3, and 18.0). The above spectroscopic signatures suggested the presence of a 9,4-friedodrime sesquiterpene moiety and accounted for four degrees of unsaturation, indicating three rings in the structure of 1.
Table 1. The 1H- (600 MHz) and 13C- (150 MHz) NMR data of compound 1 in CDCl3. a
|Position||δC||δH (J in Hz)||HMBC (H→C)||NOESY|
|1α||19.0, CH2||1.83, m||C-2, 3, 5, 9, 10||H-1β, 2β, 10|
|1β||1.53, m||C-2, 5, 10||H3-12, 14, H-1α, 2β|
|2α||26.3, CH2||1.93, m||H-1α, 1β, 10|
|2β||2.07, m||C-3, 4, 10||H-1α, 1β, 2α|
|3||120.5, CH||5.16, br s||C-5, 11||H3-11, H-2α, 2β|
|6α||36.2, CH2||1.08, m||C-8||H-6β, 7a, 8, 10|
|6β||1.68, dt (12.8, 3.4)||C-7, 8, 10, 12||H3-11, 12, H-6α, 7b|
|7a||27.4, CH2||1.41, m||C-5, 6, 9, 13||H-6, 8|
|7b||1.40, dd (6.9, 3.5)||H-6, 8, H3-12, 13, 14|
|8||37.4, CH||1.28, m||C-7, 9, 13||H-7b, 10, H3-13|
|10||47.0, CH||1.12, dd (12.4, 1.6)||C-2, 4, 5, 9, 12, 14, 15||H-1α, 2α, 8, 15α, 15β|
|11||17.7, CH3||1.55, br s||C-3, 4, 5||H3-12, H-3|
|12||19.8, CH3||1.00, s||C-4, 5, 6, 10||H3-11, 14, H-6β, 7β|
|13||16.3, CH3||0.95, d (6.7)||C-7, 8, 9||H3-14, H-7β, 8|
|14||18.0, CH3||0.88, s||C-8, 9, 10, 15||H3-12, 13, H-1β, 7β|
|15α||32.5, CH2||2.61, d (14.1)||C-8, 9, 10, 14, 16, 17, 18||H3-14, H-1α, 10|
|15β||2.43, dd (14.1, 1.2)||C-8, 9, 10, 14, 16, 17, 18||H3-13|
|18||130.4, CH||6.26, d (1.0)||C-15, 16, 19||H3-13, H-10, 15α, 15β|
|20-NH||7.33, br s|
Unambiguous assignment of NMR data of 1 was achieved by a combination of COSY, HSQC, and HMBC experiments, as depicted in Figure 2. In the 1H-1H COSY spectrum, the correlations of H2-1/H2-2/H-3, H2-6/H2-7/H-8/H3-13, and allylic coupling correlations of H-3/H3-11 revealed the presence of two fragments (thick lines in Figure 2). The two spin systems and their connectivity with the remaining atoms enabled assembly into the final planar structure based upon the HMBC spectrum of 1. The HMBC correlations from H3-11 to C-3, C-4, and C-5, from H3-12 to C-4, C-5, C-6 and C-10, from H3-13 to C-7, C-8, and C-9, and H3-14 to C-8, C-9, C-10, and C-15 indicated the presence of 9,4-friedodrime sesquiterpene skeleton with four methyl groups at C-4, C-5, C-8, and C-9, respectively. This assignment was confirmed by the HMBC correlations from H-10 to C-2, C-4, C-5, C-9, C-12, C-14, and C-15. Furthermore, the olefinic proton H-18 showed HMBC correlations with C-15, C-16, and C-19, in combination with the chemical shifts of the proton and carbon resonances, suggested the presence of a 2,5-dionepyrrole substructure. In addition, HMBC correlations from the methylene protons H2-15 to C-8, C-9, C-10, C-14, C-16, C-17, and C-18 supported the linkage of C-9 and C-16 via the methylene CH2-15 between the 9,4-friedodrime sesquiterpene moiety and 2,5-dionepyrrole substructure. Therefore, the gross structure of 1 was determined as shown in Figure 2.
Figure 2. Key COSY, HMBC, and NOESY correlations of dysidinoid A (1).
Figure 2. Key COSY, HMBC, and NOESY correlations of dysidinoid A (1).
The relative configuration of 1 was deduced from NOESY correlations in combination with coupling constant values. The large coupling constant between H-1β and H-10 (J = 12.4 Hz) and the NOESY correlations of H-1β/H3-12 and H3-14 indicated the axial orientations of these protons and methyls and also revealed the trans fusion of the two six-numbered rings [15,18]. The NOESY correlation of H3-13/H3-14 and H3-12/H3-14 revealed the three methyl groups are all β-orientation, while NOESY correlations from H-8 to H-6α, and H-10 suggested the three protons were α-orientation.
Fortunately, crystals of 1 suitable for single crystal X-ray diffraction analysis were obtained from a methanol solution. The relative configuration of 1 was unambiguously established by its X-ray crystal structure (Figure 3). Besides, a final refinement of the CuKa diffraction data resulted in the assignment of the absolute configuration of 1 as 5S, 8S, 9R, and 10S.
Minimal inhibitory concentration (MIC) was detected to evaluate the antimicrobial activities of dysidinoid A (1) toward two strains of hospital-acquired methicillin-resitant Staphylococcus aureus (MRSA H0556 and MRSAH0117). Dysidinoid A (1