Potential Antiviral Xanthones from a Coastal Saline Soil Fungus Aspergillus iizukae

Five new (1–5) and two known xanthones (6 and 7), one of the latter (6) obtained for the first time as a natural product, together with three known anthraquinones, questin, penipurdin A, and questinol, were isolated from the coastal saline soil-derived Aspergillus iizukae by application of an OSMAC (one strain many compounds) approach. Their structures were determined by interpretation of nuclear magnetic resonance (NMR) and high-resolution electrospray ionization mass spectroscopy (HRESIMS) data, as well as comparison of these data with those of related known compounds. Antiviral activity of xanthones 1−7 was evaluated through the cytopathic effect (CPE) inhibition assay, and compound 2 exhibited distinctly strong activity towards influenza virus (H1N1), herpes simplex virus types 1 (HSV-1) and 2 (HSV-2) with IC50 values of 44.6, 21.4, and 76.7 μM, respectively, which indicated that it was worth to further investigate it as a potential lead compound. The preliminary structure-activity relationship of the xanthones is discussed.


Introduction
Aspergillus iizukae is present in various environments, such as leaves of Silybum marianum [1], sponges [2], guts of a cricket species Gryllus testaceus [3], earthworm casts and different kinds of soil [4]. Recently, Özkaya et al found that the ethyl acetate extract of a sponge-associated A. iizukae showed inhibitory effects against the aquaculture pathogens Lactococcus garvieae and Vagococcus salmoninarum [2]. Additionally, a novel flavin adenine dinucleotide-dependent glucose dehydrogenase was separated from the culture of A. iizukae, which was capable of catalyzing the oxidation of glucose to glucono-δ-lactone [5]. Nevertheless, there are only a couple of reports on its metabolites. Up to now, four new and three known aromatic butenolides were isolated from the insect symbiont A. iizukae [3]. It was also reported that the endophytic A. iizukae from S. marianum could produce silybin A, silybin B and isosilybin A [1].
In our continuing search for new biologically active metabolites from fungi [6,7], the prolific fungus A. iizukae was isolated from coastal saline soil in Kenli, China. Previously, two new diphenyl derivatives and nine known compounds were obtained from it in a shaken fermentation, and six of them contain chloride atoms, that demonstrating the strain was able to utilize chlorine in the biosynthesis of its metabolites [8]. In order to obtain more halogen-containing compounds, the working strain was fermented statically in liquid culture medium supplemented with sodium bromide applying an OSMAC strategy. Chemical exploration of the fungal extract led to the isolation of seven xanthones (1−7) (Figure 1), including five new xanthones (1−5), one new natural product (6) previously reported as a semisynthetic compound, and a known one (7), along with three known anthraquinones. Herein, the isolation and structure elucidation of the new metabolites, along with the antiviral activity of 1−7 are reported.
Mar. Drugs 2018, 16, x FOR PEER REVIEW 2 of 9 bromide applying an OSMAC strategy. Chemical exploration of the fungal extract led to the isolation of seven xanthones (1−7) (Figure 1), including five new xanthones (1−5), one new natural product (6) previously reported as a semisynthetic compound, and a known one (7), along with three known anthraquinones. Herein, the isolation and structure elucidation of the new metabolites, along with the antiviral activity of 1−7 are reported.
Compound 2 was isolated as a light-yellow powder. The HRESIMS data (m/z 335.0319 [M + H] + ) gave the same molecular formula C16H11ClO6 as 1. The IR and UV data of 2 resembled those of 1, indicating that they have same scaffold. Detailed analysis and comparison of their 1 H and 13 C NMR data revealed a slight difference in the chemical shifts of 1 H and 13 C atoms in the rings A, especially those of C-1 and C-4a, while chemical shifts in the rings B were almost identical between 1 and 2. The aforementioned data suggested that the chlorine atom was at C-4 in 2, which was confirmed by the HMBC correlations ( Figure 2) from H3-11 to C-2, C-3 and C-4, and from 1-OH to C-1, C-2 and C-1a. The other HMBC correlations shown in Figure 2 supported that the structure of 2 is methyl-(4-chloro-l,6-dihydroxy-3-methylxanthone)-8-carboxylate.
Compound 3, a light-yellow solid, was isolated in small amount, and showed a protonated ion in its HRESIMS at m/z 349.0476 [M + H] + , indicating a molecular formula of C17H13ClO6 having one CH2 unit more than that of 2. The 1 H and 13 C NMR data were very similar to those of 2, except for the
Compound 2 was isolated as a light-yellow powder. The HRESIMS data (m/z 335.0319 [M + H] + ) gave the same molecular formula C 16 H 11 ClO 6 as 1. The IR and UV data of 2 resembled those of 1, indicating that they have same scaffold. Detailed analysis and comparison of their 1 H and 13 C NMR data revealed a slight difference in the chemical shifts of 1 H and 13 C atoms in the rings A, especially those of C-1 and C-4a, while chemical shifts in the rings B were almost identical between 1 and 2. The aforementioned data suggested that the chlorine atom was at C-4 in 2, which was confirmed by the HMBC correlations ( Figure 2) from H 3 -11 to C-2, C-3 and C-4, and from 1-OH to C-1, C-2 and C-1a. The other HMBC correlations shown in Figure 2 supported that the structure of 2 is methyl-(4-chloro-l,6-dihydroxy-3-methylxanthone)-8-carboxylate.
Compound 3, a light-yellow solid, was isolated in small amount, and showed a protonated ion in its HRESIMS at m/z 349.0476 [M + H] + , indicating a molecular formula of C 17 H 13 ClO 6 having one CH 2 unit more than that of 2. The 1 H and 13 C NMR data were very similar to those of 2, except for the appearance of an additional methoxyl group and the disappearance of a hydroxyl group. Therefore, 3 was a methylated derivative of 2. The lower resonance frequencies of C-9 (δ C 172.9) suggested that the methoxyl was at C-1 [13]. This hypothesis was proved by the HMBC correlations ( Figure 2) from H 3 -14 to C-1, and from H-2 to C-1, C-4 and C-1a. Based on these evidences, the structure of 3 was established as methyl-(4-chloro-6-hydroxy-1-methoxy-3-methylxanthone)-8-carboxylate. appearance of an additional methoxyl group and the disappearance of a hydroxyl group. Therefore, 3 was a methylated derivative of 2. The lower resonance frequencies of C-9 (δC 172.9) suggested that the methoxyl was at C-1 [13]. This hypothesis was proved by the HMBC correlations ( Figure 2) from H3-14 to C-1, and from H-2 to C-1, C-4 and C-1a. Based on these evidences, the structure of 3 was established as methyl-(4-chloro-6-hydroxy-1-methoxy-3-methylxanthone)-8-carboxylate.  Compound 4 was obtained as an orange powder. The molecular formula was determined as C17H14O6 from the HRESIMS m/z 313.0727 [M − H] − , showing the absence of chlorine atoms and one hydrogen atom more than that of 3. In the 1 H NMR spectrum, four broad singlets (δH 6.94, 6.87 6.82 and 6.73, 1H, s, each) were observed. Careful comparison of the NMR spectra data between 3 and 4 (Tables 1 and 2) revealed that it was a dechlorinated derivative of 3. This hypothesis was confirmed  In the 1 H NMR spectrum, four broad singlets (δ H 6.94, 6.87 6.82 and 6.73, 1H, s, each) were observed. Careful comparison of the NMR spectra data between 3 and 4 (Tables 1 and 2) revealed that it was a dechlorinated derivative of 3. This hypothesis was confirmed by analysis of the HMBC correlations ( Figure 2). Therefore, the structure of 4 was elucidated as methyl-(6-hydroxy-1-methoxy-3-methylxanthone)-8-carboxylate. Compound 5 was isolated as a pale-yellow powder and had the molecular formula C 15 H 9 ClO 6 as determined by the HRESIMS m/z 319.0013 [M − H] − . Therefore, 5 has one CH 2 unit less than 2. In its IR spectrum the broad absorption at 3300−2600 cm −1 , along with the absorption at 1694 cm −1 , revealed the presence of a carboxylate functionality in 5. Its NMR data were very similar to those of 2, except for the absence of a methoxy group and the presence of a broad singlet of a hydroxy group in 5. In the HMBC spectrum, the correlations from H-11 to C-2, C-3 and C-4, and from H-2 to C-1, C-4, C-11 and C-1a were observed, which unequivocally established the substitution of ring A. In spite of no HMBC correlations observed from H-6 and H-7 to any carbons, the ring B and the chemical shift assignment of corresponding carbons and protons could be achieved by comparison of its 1 H and 13 C data with those of 6 and calyxanthone [14]. Interestingly, in the 13 C NMR spectrum, the resonance intensity of carbon atoms in ring B was much weaker than that of carbon atoms in ring A ( Figure S30). Compound 5 was identified as 4-chloro-1,6-dihydroxy-3-methylxanthone-8-carboxylic acid.
Compound 6 was isolated as an orange-red powder and has low solubility in methanol. Its molecular formula C 15 Figure S40), suggesting the existence of two chlorine atoms. Its UV spectrum showed characteristic absorption bands of xanthones. In the 1 H NMR spectrum, the meta-coupled aromatic protons at δ H 6.95 (d, J = 2.2 Hz, H-5) and 6.86 (d, J = 2.1 Hz, H-7) of ring B were supported by the HMBC correlations ( Figure 2) from H-5 to C-6, C-7, C-9a and C-10a, and from H-7 to C-5, C-12 and C-9a, which was confirmed by similar 1 H and 13 C chemical shifts values to those of calyxanthone [14]. The downfield shift of the carbonyl implied the hydrogen bonded phenolic hydroxyl at δ H 13.22 (br s, 1-OH) attached to C-1. Additionally, the HMBC correlations from the aromatic methyl protons to C-2, C-3 and C-4, and from 1-OH to C-2, demonstrated that chlorine atoms were both at C-4 and C-2, respectively. Ultimately, the structure of 6 was established to be 2,4-dichloro-1,6-dihydroxy-3-methylxanthone-8-carboxylic acid. Compound 6 was previously described as a synthetic intermediate using only UV and IR data [15]. This is the first report of its isolation from a natural extract.

Fungal Material
Aspergillus iizukae KL33 (GenBank accession numbers: HQ717800) was isolated from coastal saline soil in Kenli, Shandong Province of China, in August 2008. The strain was deposited at the Department of Chemistry, Binzhou Medical University, Yantai.
40 L of the fermentation broth was separated into mycelium and filtrate through cheesecloth. The filtrate was extracted three times with ethyl acetate. The mycelium was extracted with methanol for three times. The methanol solution was concentrated under reduced pressure to give an aqueous solution. The aqueous solution was extracted three more times with ethyl acetate. Both the ethyl acetate solutions were concentrated under reduced pressure to give a crude extract (65 g).

Antiviral Activity
The antiviral activity against influenza A virus (H1N1) was carried out by CPE inhibition assay as previously reported [18,19]. First, confluent MDCK cell monolayers and influenza virus (A/Puerto Rico/8/34 (H1N1), PR/8) were incubated together at 37 • C for 1 h. Then, the cells were treated with different test compounds after removing the virus dilution. After incubating at 37 • C for 48 h, the cells were fixed with 4% formaldehyde of 100 µL for 20 min at room temperature. Later on, the cells were stained with 0.1% crystal violet for 30 min after removal of the formaldehyde. Finally, the plates were washed and dried, followed by the measurement of the intensity of crystal violet staining for each well at 570 nm in a microplate reader (Bio-Rad, USA). Ribavirin was used as the positive control.
The anti-herpes simplex virus types 1 (HSV-1) and 2 (HSV-2) activity of 1−7 on Vero cells were conducted in the same way as described above [20], and acyclovir (ACV) was applied as the positive control.

Conclusions
In summary, the culture based on OSMAC strategy of the fungus A. iizukae yielded seven xanthones, including five new and one isolated for the first time as a natural product. Among them, five metabolites contain chlorine, and their structures were different from those previously reported from A. iizukae. From this study, it is clear that the OSMAC strategy is still a powerful tool in producing new metabolites from microorganisms. Compound 2 exhibited a strong antiviral activity against H1N1, HSV-1 and HSV-2 with IC 50 values of 44.6, 21.4, and 76.7 µM, respectively, compared with the positive controls.
The results of antiviral activity of 1−7 indicated that the hydroxy group at C-1 and the methyl carboxylate group at C-8 essentially contributed to the anti-H1N1, anti-HSV-1 and anti-HSV-2 activities, and the position of the chlorine atom in ring A would affect the antiviral activities. Additionally, it seemed that methylation of the hydroxy group at C-1 or replacement of methyl carboxylate at C-8 by carboxylic acid, to a large extent, lower the antiviral effect.
Xathones have attracted considerable interest for their promising biological activities and the interesting structural scaffold, which could be modified by various substituents [21,22]. Our finding suggests that 2 might be a potential anti-H1N1 lead candidate, worthy of a further pharmacological exploration.