Undescribed Metabolites from an Actinobacteria Acrocarpospora punica and Their Anti-Inflammatory Activity

In an effort to explore bioactive anti-inflammatory compounds from natural Actinobacteria resources from all over Taiwan and various ecological environments, an active strain of Acrocarpospora punica was collected at Taitung County in Taiwan, prepared from soil origin. A bioassay-guided fractionation of the BuOH extract of a culture broth of a new strain of the actinomycete Acrocarpospora punica led to the isolation of five previously undescribed compounds: acrocarpunicains A–F (1–6). The structures were elucidated by 1D and 2D Nuclear Magnetic Resonance (NMR) spectroscopy and mass spectrometry. Furthermore, the isolated compounds were subjected to in vitro testing to evaluate their anti-inflammatory activity. Of these isolates, acrocarpunicains A (1), B (2), C (3) and F (6) showed NO inhibitory activity with IC50 values of 9.36 ± 0.25, 10.11 ± 0.47, 5.15 ± 0.18, and 27.17 ± 1.87 μM, stronger than the positive control, quercetin (IC50 = 35.95 ± 2.34 μM). To the best of our knowledge, this is the first report on azaphilone and phenanthrene-type metabolites from the genus Acrocarpospora.


Introduction
Actinobacteria (Actinomycetes) can produce various metabolites, which can be used as antibiotics (streptomycin, cycloheximide), anticancer drugs (bleomycin), dextran synthetase inhibitors, immunomodulators (rapamycin), etc. In addition to occupying a pivotal position in agriculture, they also play an increasingly important role in animal husbandry, the food industry, and in environmental protection. They are also well known as an outstanding and fascinating source of commercially valuable bioactive compounds, particularly the aforementioned antibiotics. Almost a half of the known microbial bioactive secondary metabolites are derived from actinomycetes [1][2][3][4][5][6][7], of which more than 70% were obtained from the genus Streptomyces. However, the active ingredients of many new Taiwanese Actinobacteria and their mechanisms of actions are still unknown. Thus, it is necessary to study the active compounds of these Actinobacteria using scientific methods.
In exploring the actinomycetes via preliminary screening, we recently isolated a novel strain, named AC18001 (04107M-2), from the soil of Taitung County with a unique morphology and possessing anti-inflammatory activities. This strain was determined to be Acrocarpospora punica based on their phenotypic and genotypic data.
The genus Acrocarpospora was first described by Tamura et al. [8] and composed of the following three species: A. corrugatum, A. macrocephala and A. pleiomorpha [8]. This microorganism was isolated and identified by our research team. On the basis of phenotypic and genotypic data, it is proposed that the strain should be classified as a new species, named A. punica. Previous chemical investigations of the genus Acrocarpospora have received less attention, and only a few articles have reported on the classification of molecular biology [9][10][11]. species, named A. punica. Previous chemical investigations of the genus Acrocarpospora have received less attention, and only a few articles have reported on the classification of molecular biology [9][10][11].
Recently, over 600 species of microorganisms have been screened for in vitro anti-inflammatory activities, and A. punica has been found to be one of the active species. In a continuation of our studies, aimed at finding new chemical and anti-inflammatory constituents from this genus, we firstly describe the isolation and structural elucidation of five previously undescribed compounds: acrocarpunicains A-F (1-6), from the EtOAc extract of a culture broth of A. punica (Figure 1). A bioassay-guided fractionation of the BuOH extract of the whole broth has led to the isolation of A (1), B (2), C (3) and F (6) with anti-inflammatory activities, along with the isolation of two inactive ones. We herein report the isolation and anti-inflammatory activity properties of these compounds.

The Taxonomic Identification (Phenotypic and Genotypic Data) of Acrocarpospora punica
Strain 04107M-2 T produced branched and non-fragmented substrate mycelia, club-shaped structures borne on the tips of the aerial mycelium. The spores were non-motile, rod and smooth-surfaced ( Figure 2). The growths on various media were poor, and a reddish with golden luster crystal had been observed on oatmeal agar. No soluble pigment was produced in all of the media tested. The results of these physiological and biochemical tests are indicated in Table 1.

The Taxonomic Identification (Phenotypic and Genotypic Data) of Acrocarpospora punica
Strain 04107M-2 T produced branched and non-fragmented substrate mycelia, clubshaped structures borne on the tips of the aerial mycelium. The spores were non-motile, rod and smooth-surfaced ( Figure 2). The growths on various media were poor, and a reddish with golden luster crystal had been observed on oatmeal agar. No soluble pigment was produced in all of the media tested. The results of these physiological and biochemical tests are indicated in Table 1.

Phylogeny
The almost-complete 16S rRNA gene sequence (1511 nt) of strain 04107M-2 T was determined. A preliminary comparison of the sequence against the GenBank database revealed high sequence similarity values with members of the genus Acrocarpospora. The phylogenetic tree based on the 16S rRNA gene sequences of the strain 04107M-2 T , other valid published Acrocarpospora species, and other related species are shown in Figure 3. The binary similarity values ranged between 96.5% (A. pleiomorph IFO 16266 T ) and 98.2% (A. corrugata NBRC 13972 T ). The DNA-DNA hybridization rates determined with the new isolate 04107M-2 T to its closest type strains of A. corrugata BCRC 16357 T was 2.9%; A. macrocephala was 0.4%, and A. pleiomorpha was 1.0% (Table 2). It is clear from the DNA-DNA relatedness (<70%) study that strain BC 44T-5T and these two species belong to separate species [12]. The distinctiveness of the isolate also comes from phenotypic evidence compared with the nearest phylogenetic neighbors (Table 2). On the basis of phenotypic and genotypic characters, it was evident that the isolate should be classified as a new species of the genus Acrocarpospora. The name Acrocarpospora punica sp. nov., with the type strain 04107M-2 T (=BCRC AC 18001 T ).

Structure Elucidation of Compounds
Compound 1 was obtained as optically active oil with [α] 25 D = +916 (c 0.01, CHCl 3 ), and its molecular formula was deduced as C 23 H 27 NO 4 from HR-ESI-MS data, implying 11 degrees of unsaturation. The UV spectrum depicted the typical absorption of azaphilone bearing a nitrogen atom at λ max (log ε) 245, 279, 424, 501 nm [13,14]. Its IR spectrum showed bands corresponding to the absorptions of a conjugated carbonyl group at 1720 and 1622 cm −1 , respectively. The CD spectrum showed a positive Cotton effect at 217, 283, and 407 nm, and showed a negative Cotton effect at 237, 341, and 534 nm. The 1 H-NMR spectrum of 1 (  15)]. Comparing the NMR data of 1 with those of rubropunctamine [15] revealed they share a similar structure. The differences between them were that 1 showed another ethyl group [δ H 1.44 (3H, t, J = 7.2 Hz, H-2b) and 3.91 (2H, qd, J = 7.2, 1.2 Hz, H-2a)] at the N-2 position. This was confirmed by the HMBC correlation from H-2a to C-1. Other COSY, HMBC, and NOESY correlations of 1 (Figures 4 and 5) can confirm the completed structure and named acrocarpunicain A.  Compound 1 was identified by comparison with literature data of monascorubramine but showed a dextrorotatory optical activity with [α] D = +916 (c 0.01, CHCl 3 ). Without an enantiomeric antipode for comparison, but with the reference to monascorubramine ((7R)-configuration; [α] D −2600 • (c 0.12, CHCl 3 )) [16,17], the stereochemistry at C-7 of 1 would appear to be of the (7S)-configuration.
Compound 2 was obtained as reddish oil with an optical rotation +336 (c 0.01, CHCl 3 ), and the molecular formula was determined as C 25 H 31 NO 4 based on the HRESIMS peak at m/z 432.21542 [M+Na] + (calcd. for C 23 H 27 N 1 O 4 Na 432.21508) with 11 degrees of unsaturation. An analysis of the NMR (Table 3), UV, and IR data suggested that the structure of 2 is similar to that of compound 1. The UV spectrum depicted the typically absorption of red pigments azaphilone bearing a nitrogen atom at λ max (log ε) 202 (4.12), 247 (4.00), 286(4.00), 424 (3.96), 503 (4.05) nm [12,13]. The IR spectrum revealed the presence of a conjugated carbonyl group (1724, 1624 cm −1 ). The CD spectrum showed a positive Cotton effect at 220, 285, and 408 nm, and showed a negative Cotton effect at 237, 340, and 547 nm. The 1 H-NMR spectrum of 2 (Table 3) showed four methyl groups, including one terminal methyl at the side chain group (δ H 0.86), one primary methyl (δ H 1.44), one secondary methyl (δ H 1.68) connected with a pair of trans-olefinic groups (δ H 6.23 and 6.47), one tertiary methyl group (δ H 2.00), three specific azaphilone olefinic protons (δ H 6.64, 6.73, and 7.77), and five methylene groups. The NMR data of 2 was similar to those of 1, except that the integrate of the high-field methylene groups (δ H 1.33) of 2 is two more protons than 1 and 2 depicted an octanoyl side chain group at C-13. Therefore, 2 was supported by the COSY, HMBC, and NOESY correlations showed in Figures 4 and 5. As evidenced by the above data, the entire structure of 2 was confirmed and named acrocarpunicain B.

Biological Studies
The six isolates, in sufficient amounts, were evaluated by examining their inhibitory effects on LPS-induced inducible nitric oxide synthase (iNOS)-dependent NO production in the murine macrophage cell line RAW 264.7 ( Table 5). The inhibitory activity data of the six isolated compounds on NO generation by macrophages are shown in Table 3. From the results of our abovementioned tests, the following conclusions can be drawn: (a) Compared to quercetin (IC50 value 35.95 ± 2.34 μM), which was used as a positive control in this study, acrocarpunicains A, B, C, and F (1, 2, 3 and 6) exhibited NO inhibi-     (Figure 2) of H-1 to C-4a and C-8a; and H-12 to C-7 and C-8, these two singlet methyl groups were assigned to C-8a and C-7, respectively. The side chain group from C-15 to C-19 used C-14 to connect with the furanone ring based on the HMBC correlation from H-15 to C-14 and from H-13 to C-14. The 2-oxopentyl moiety at C-4a in monascuspurpurone [19] was replaced by trans-propenyl in 4, certainly. The trans-fused ring linkage of these two rings was revealed by the absence of a NOESY correlation between H-12 and H-6. Moreover, the H-12 showed a NOESY correlation to H-13, referring to their suprafacial orientation. As a result, the relative configuration of 4 was assigned as rel-(6R, 7R, 13S) [19]. Based on these results, 4 was identified and named acrocarpunicain D.  (Table 4), 39 resonances were subclassified by DEPT experiments into 7 methyls, 11 methylenes, 13 methines (1 oxygenated methine and 5 sp 2 methines), and 8 quaternary carbons (1 ester, 3 oxygenated and 1 sp 2 carbons). All signals in the above spectra showed that compound 5 had a β-sitosterol skeleton with one O-linkage isoferulate group. From the signal of the HMBC spectrum, it can be observed that δ H 7.49 (H-7 ) is correlated to δ C 108.7 (C-2 )) and 102.5 (C-6 ), and δ H 6.22 (H-8 ) is correlated to δ C 166.0 (C-9 ) and 126.5 (C-1 ). Two hydroxyl groups (δ H 5.40 and 5.98 (each 1H, br s)), and a methoxyl group (δ H 3.92 (3H, s, OMe-4 )) located at the aromatic ring were determined by 13 C-NMR signals at C-3 (δ C 143.9), C-4 (δ C 147.4), and C-5 (δ C 134.0) and the HMBC correlations of OMe-C(4)/C-4. From the above, it can be inferred that 5 is a steroid and takes a trans-3-hydroxyisoferulate at the C-3 position. It was compared with β-sitosteryl trans-isoferulate, and hence, compound 5 was determined to be β-sitosteryl trans-3-hydroxyisoferulate and designated as acrocarpunicain E.   [20]. The IR spectrum revealed the presence of aromatic rings (1626, 1483 cm −1 ). Eight indices of hydrogen deficiency (IHD) were determined from the molecular formula, 13 C-NMR (Table 3), and DEPT spectra.  2H, m, H-3) were determined by the two-dimensional spectra HMBC and COSY, respectively. Thus, the structure of 6 was determined to be 6-isopropyl-7-methoxy-1,1-dimethyl-1,2,3,4-tetrahydrophenanthrene and was designated as acrocarpunicain F.

Discussion
In summary, actinomycetes have been recognized as a large microbial reservoir that can be expected to provide a variety of structurally unique and biologically potent natural metabolites. Continuing our previous chemical and biological studies of microbially produced metabolites, a new actinomycete strain Acrocarpospora punica, isolated from soil samples collected in Taitung County, Taiwan was determined to be able to produce bioactive metabolites during its liquid fermentation process according to our system screening plan.
Secondary metabolites of the genus Acrocarpospora have been rarely studied in the past. The A. punica strain in this study has only 10 components reported by our team in the past [21]. After modifying the fermentation conditions, we obtained six components from the active layer of BuOH, five of which were new compounds whose backbones included azaphilone, dihydrobenzofuran, isochromene-3,4(1H)-dione, and tetrahydrophenanthrene. These components were originally discovered from the chemotaxonomically significant genus Acrocarpospora. These results demonstrate that Acrocarpospora produces unique and diverse metabolites in different fermentation conditions and soil-derived collections. Therefore, in a special ecological environment, more natural products with biological activity may be found by searching for Acrocarpospora species.

Biological Studies
The six isolates, in sufficient amounts, were evaluated by examining their inhibitory effects on LPS-induced inducible nitric oxide synthase (iNOS)-dependent NO production in the murine macrophage cell line RAW 264.7 ( Table 5). The inhibitory activity data of the six isolated compounds on NO generation by macrophages are shown in Table 3. From the results of our abovementioned tests, the following conclusions can be drawn: (a) Compared to quercetin (IC 50 value 35.95 ± 2.34 µM), which was used as a positive control in this study, acrocarpunicains A, B, C, and F (1, 2, 3 and 6 1, 2, 3, and 6 showed about 4-, 3-, 7-and 1.2-fold NO inhibitory activities compared to quercetin, respectively. (c) Compound 4 showed weak NO inhibitory activity, whereas compound 5 displayed no NO inhibitory activity. (d) Compound 1 (acrocarpunicain A) with a hexanoyl side chain exhibited more effective inhibition than its analogue, compound 2 (acrocarpunicain B), with an octanoyl side chain against LPS-induced NO generation. (e) Compounds 3 and 6 belong to liner azaphilone skeleton and naphthenic derivatives, respectively; compound 3 has better anti-inflammatory activity than 6. (f ) Furthermore, the RT-PCR analysis in the present study indicated that LPS treatment increased the level of iNOS mRNA expression . Samples 1, 2, 3, and 6 can inhibit the production of NO in the ELISA experiment (Table 5), and in the RT-PCR experiment, the iNOS gene has a significantly inhibited band (Figure 6), the results show that transcription can be inhibited at the same time. In addition, in its effect on translation, it can be seen that the expression of the protein is inhibited by inhibiting the expression of the gene, indicating that samples 1, 2, 3, and 6 may have the effect of regulating immunity. Further, the iNOS result was proven by RT-PCR, demonstrating that the compounds can inhibit the expression of the iNOS gene, and it was noted that these compounds (1, 2, 3, and 6) inhibited this increase in a concentration-dependent manner. At the highest concentration, none of the compounds tested showed any obvious cytotoxicity toward RAW 264.7 cells. (g) The cytotoxic effects were measured using an MTT assay. The high cell viability (> 90%) indicated that the inhibitory activities of LPS-induced NO production by the active compounds (1, 2, 3, and  6) were not as a result of cytotoxicity. NO generation. (e) Compounds 3 and 6 belong to liner azaphilone skeleton and naphthenic derivatives, respectively; compound 3 has better anti-inflammatory activity than 6.
(f) Furthermore, the RT-PCR analysis in the present study indicated that LPS treatment increased the level of iNOS mRNA expression. Samples 1, 2, 3, and 6 can inhibit the production of NO in the ELISA experiment (Table 5), and in the RT-PCR experiment, the iNOS gene has a significantly inhibited band (Figure 6), the results show that transcription can be inhibited at the same time. In addition, in its effect on translation, it can be seen that the expression of the protein is inhibited by inhibiting the expression of the gene, indicating that samples 1, 2, 3, and 6 may have the effect of regulating immunity. Further, the iNOS result was proven by RT-PCR, demonstrating that the compounds can inhibit the expression of the iNOS gene, and it was noted that these compounds (1, 2, 3, and 6) inhibited this increase in a concentration-dependent manner. At the highest concentration, none of the compounds tested showed any obvious cytotoxicity toward RAW 264.7 cells. (g) The cytotoxic effects were measured using an MTT assay. The high cell viability (> 90%) indicated that the inhibitory activities of LPS-induced NO production by the active compounds (1, 2, 3, and 6) were not as a result of cytotoxicity. Figure 6. Samples 1, 2, 3, and 6 in the RT-PCR experiment. It can also be clearly seen that the iNOS gene has been significantly inhibited. Compounds IC50 (μM) a Figure 6. Samples 1, 2, 3, and 6 in the RT-PCR experiment. It can also be clearly seen that the iNOS gene has been significantly inhibited.

Microorganism, Cultivation, and Preparation of the Actinobacteria Strain
The strain was isolated from a soil sample collected in Taitung County, Taiwan, by using HV agar, and was then incubated at 28 • C for 3 weeks. The strain was maintained on oatmeal agar and a suspension of spores or mycelia fragments of the strain in a broth containing 20% (v/v) glycerol was stored at −20 • C. The inoculum's medium contained: malt extract, 3 g; yeast extract, 3 g; glucose, 5 g; agar, 1.5 g; and distilled water 1 L. The initial pH of the medium was 8. The synthetic culture medium contained: glucose, 20 g; Monosodium Glutamate (MSG), 10 g; K 2 HPO 4 , 5 g; KH 2 PO 4 , 5 g; MgSO 4 •7H 2 O, 1.0 g; KCl, 0.5 g; ZnSO 4 •7H 2 O, 0.01 g; FeSO 4 •7H 2 O, 0.01g; MnSO 4 •H 2 O, 0.003 g per liter of distilled water. The initial pH of the medium was adjusted to 5.5. The slant culture was kept on PDA (potato dextrose agar) Difco. The spores of the strains were prepared by growth on PDA slants for 14 days at 28 • C. The spores were washed with sterile water. A suspension of 10 7 spores was used to incubate a 5 L Erlenmeyer flask containing 2 L inoculum medium, which was incubated at 28 • C on a rotary shaker for 3 days. This inoculum was then transferred to a 50 L fermenter (B. Braun, Germany) containing 30 L of synthetic medium operated at 100 rpm and at 30 • C with an aeration rate of 0.3 vvm. After 21 days of cultivation, the fermentation was stopped, and the liquid culture was separated from the mycelium by filtration.

Determination of NO Production and Cell Viability Assay
Mouse macrophage cell line (RAW 264.7) was obtained from the Bioresource Collection and Research Center (BCRC 60001) and cultured at 37 • C in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (Gibco), 4.5 g/L glucose, 4 mM glutamine, penicillin (100 units/mL), and streptomycin (100 µg/mL) in a humidified atmosphere in a 5% CO 2 incubator. The cells were treated with 10, 25, 50 µM natural products in the presence of 1 µg/mL LPS (lipopolysaccharide, Sigma-Aldrich) for 20 h. The concentration of NO in culture supernatants was determined as nitrite, a major stable product of NO, by a Griess reagent assay [22], and the cell viabilities were determined using the MTT assay as described previously [23].

Reverse Transcription-PCR
The ImProm-II Reverse Transcription System (Promega) was used to take 1 µg of the extracted total RNA, to which was added 1 µL of Oligo(dT)15 Primer, and other supplements with DEPS water. At this time, the total volume was 5 µL. The reaction was immediately placed on ice for at least 5 min, and then 5 × buffer 4 µL, MgCl2 (25 mM) 1.2 µL, dNTP Mix (10 mM) 1 µL, RNasin Ribonuclease Inhibitor 0.5 µL, ImProm-II reverse transcriptase 1 µL and DEPC H2O 6.8 µL was added; the cDNA product was obtained after the reactions of 25 • C, 5 min; 42 • C, 1 h; and 70 • C, 15 min. At the same time, a negative control group was made without reverse transcriptase to confirm whether there was DNA contamination. The obtained cDNA product was then amplified by a PCR reaction to amplify the predicted fragment.

Conclusions
Actinomycetes have potential economic and biotechnological value and have long been recognized as major microorganisms in the medical industry. To date, there are tens of thousands of antibiotics produced by microorganisms, of which more than 70% are derived from Actinobacteria [23]. Secondary metabolites of Actinobacteria have various structures and biological activities, including antibacterial, antifungal, antitumor, insecticidal and herbicidal, enzyme inhibition, immune regulation, etc. [24,25], which indicates that Actinobacteria have great potential for the development of new medicines. As part of our investigations aimed at exploring structurally novel bioactive secondary metabolites from actinomycetes, our chemical research on the fermentation extract of Acrocarpospora punica led to the isolation of five previously undescribed compounds, namely, acrocarpunicain A-F (1-6) ( Figure 2). The structure of these isolates was determined by spectroscopic experiments. The BuOH soluble fraction from the A. punica fermentation broth was tested in vitro and demonstrated anti-inflammatory activity that decreased LPS-stimulated nitric oxide (NO) in RAW 264.7 cells. In addition, compounds 1, 2, 3, and 6 showed potent inhibition with IC 50 values ≤27.17 µM, against lipopolysaccharide (LPS)-induced nitric oxide (NO) generation, stronger than the positive control, quercetin (IC 50 = 35.95 ± 2.34 µM). To the best of our knowledge, this is the first report on azaphilones (1-3) and phenanthrene analogs (6) from the genus Acrocarpospora.