Altertoxins with Quorum Sensing Inhibitory Activities from The Marine-Derived Fungus Cladosporium sp. KFD33

Five new perylenequinone derivatives, altertoxins VIII–XII (1–5), as well as one known compound cladosporol I (6), were isolated from the fermentation broth of the marine-derived fungus Cladosporium sp. KFD33 from a blood cockle from Haikou Bay, China. Their structures were determined based on spectroscopic methods and ECD spectra analysis along with quantum ECD calculations. Compounds 1–6 exhibited quorum sensing inhibitory activities against Chromobacterium violaceum CV026 with MIC values of 30, 30, 20, 30, 20 and 30 μg/well, respectively.


Introduction
The unremitting discovery of antibiotics showed a promising prospect for the bacterial infections, but also resulted in the serious resistance of antibiotics. Bacteria have evolved a variety of resistance mechanisms [1,2]. Quorum sensing (QS) is the regulation of gene expression in response to fluctuations in cell-population density among diverse bacterial species [3]. QS controls the production of virulence factors in bacteria in a population density dependent manner through intercellular communication mechanism [3]. QS inhibitors (QSIs) can inhibit the QS mechanism and attenuate virulence without influencing bacterial growth. Thus, QSIs can be used to disarm pathogens in the host and are not easy to cause bacterial resistance compared to conventional antibiotics [4,5]. The search for efficient QSIs is supposed to be an effective method to solve problems of bacterial infection and antibiotic resistance. Based on this, a screening system has been established for searching of QSIs [6]. Due to the special environmental conditions, marine-derived fungi, as a rich source of various compounds with complex structures and excellent activities, have attracted more and more attentions [7]. Our previous research on new bioactive metabolites from the marine-derived fungi has led to the isolation and identification of many new QSIs, such as aculene E and penicitor B, aculene C, aculene D, aspergillumarins A-B [8], asperochrin D, asperochrin F, (3R,4R)-4,7-dihydroxymellein, asperochrin A, and asteltoxin [9]. As our ongoing research, the fungus Cladosporium sp. KFD33 was isolated from a blood cockle from Haikou Bay, China. The EtOAc extract of the fermentation broth of this fungus showed obvious QS inhibitory activity against Chromobacterium violaceum CV026. Subsequent chemical investigation on the EtOAc extract of the fermentation broth had led to the isolation of five new compounds, named altertoxins VIII-XII (1)(2)(3)(4)(5), as well as a known one, cladosporol I (6) [10] (Figure 1). All of the new compounds showed obvious QS inhibitory activities. Herein, the isolation, structure elucidation, and QS inhibitory activity of compounds 1-6 are described.
Mar. Drugs 2019, 17, x FOR PEER REVIEW 2 of 7 fermentation broth had led to the isolation of five new compounds, named altertoxins VIII-XII (1-5), as well as a known one, cladosporol I (6) [10] (Figure 1). All of the new compounds showed obvious QS inhibitory activities. Herein, the isolation, structure elucidation, and QS inhibitory activity of compounds 1-6 are described.
Compounds 2 and 3 were obtained as yellow powder, exhibiting two totally overlapped peaks over Rp-18 column and two independent peaks with a peak area ratio of 1:2 over a chiral column (Supporting information), suggesting that they are pair of enantiomers, which were different to each other at the absolute configuration in C-1′. Their formulas were determined as C20H18O2 on the basis of HRESIMS, indicating 12 degrees of unsaturation. Analysis of the 1D NMR data of 2 and 3 revealed that their planar structures are similar to that of 1, except that the C-4 carbonyl in 1 was replaced by a methylene in 2 and 3. This deduction was further confirmed by COSY correlations of H2-2/H2-3/H2-4. In order to determine the absolute configuration of the stereocenter C-1′ of compounds 2 and 3, the ECD spectra of (R)-2 and (S)-2 were calculated and compared with the experimental ECD spectra of 2 and 3. As can be seen in Figure 3, the calculated ECD spectra for (R)-2 and (S)-2 matched well with the experimental ECD spectra of 2 and 3, respectively, indicating that the absolute configuration of C-1' in compound 2 is R and in compound 3 is S.
Compounds 2 and 3 were obtained as yellow powder, exhibiting two totally overlapped peaks over Rp-18 column and two independent peaks with a peak area ratio of 1:2 over a chiral column (Supporting information), suggesting that they are pair of enantiomers, which were different to each other at the absolute configuration in C-1 . Their formulas were determined as C 20 H 18 O 2 on the basis of HRESIMS, indicating 12 degrees of unsaturation. Analysis of the 1D NMR data of 2 and 3 revealed that their planar structures are similar to that of 1, except that the C-4 carbonyl in 1 was replaced by a methylene in 2 and 3. This deduction was further confirmed by COSY correlations of H 2 -2/H 2 -3/H 2 -4. In order to determine the absolute configuration of the stereocenter C-1 of compounds 2 and 3, the ECD spectra of (R)-2 and (S)-2 were calculated and compared with the experimental ECD spectra of 2 and 3. As can be seen in Figure 3, the calculated ECD spectra for (R)-2 and (S)-2 matched well with the experimental ECD spectra of 2 and 3, respectively, indicating that the absolute configuration of C-1 in compound 2 is R and in compound 3 is S.  The molecular formula of compound 4 was determined as C21H20O2 based on HRESIMS, indicating 12 degrees of unsaturation. The 13 C NMR data of 4 are similar to those of 2 and 3 except for the presence of a methoxy carbon (δC 55.2) in 4. In the HMBC spectrum, correlation from the protons of this carbon to C-1′ (δC 76.8) was observed, indicating that the OH-1′ in 2 or 3 was methylated in 4. The remaining substructure of 4 was determined to be the same as those of 2 and 3 by detailed analysis of its 2D NMR data ( Figure 2). The absolute configuration of the stereocenter C-1′ in 4 was determined to be the same as that of 2 by their similar ECD curves ( Figure 3) and same sign of optical rotation values. Compound 5 was obtained as yellow powder, and its formula was determined as C20H18O4 on the basis of HRESIMS data, indicating 12 degrees of unsaturation. Comprehensive analysis of the NMR data of 5 revealed that its structure is very similar to that of cladosporol I (6), except for the absence of the C-1′ carbonyl and the presence of a double bond which was located at C-1′ (δC 121.9) and C-2′ (δC 125.9) in 5, as deduced from the COSY correlations of H-1′/H-2′/H-3′/H-4′ as well as the HMBC correlations from H2-3′ (δH 2.29, 2.47) to C-1′ and C-4′a (δC 139.4) and from H-2′ (δH 5.89) to C-4′ (δC 37.5) and C-8′a (δC 121.2). The absolute stereochemistry of the asymmetric C-4′ was estimated from the ECD spectrum. In the ECD spectrum of 5, there was a strong splitting cotton effect centered at 216 nm, which should be originated from exciton chirality caused by interaction between two phenolic chromophores [12]. Thus, according to the exciton chirality rule, the absolute stereochemistry of C-4′ was determined to be S. In addition, the absolute configuration of the C-1 stereocenter in cladosporol I (6) has been determined by comparison of the experimental ECD curve with the calculated ECD curves of the two C-1 epimers (1R,4S)-6 and (1S,4S)-6 [8]. According the results, the calculated ECD curve for (1S,4S)-6 showed an intense positive CE around 225 nm, which matched well with the experimental one and was absent in the calculated ECD curve for (1R,4S)-6. Thus, the intense positive CE around 225 nm can be used to differentiate (1R,4S)-6 and (1S,4S)-6. The experimental ECD curve of 5, with an intense positive CE around 225 nm, is very similar to that of (1S,4S)-6, indicating that the absolute configuration of C-1 is S, same as that of 6.  The molecular formula of compound 4 was determined as C21H20O2 based on HRESIMS, indicating 12 degrees of unsaturation. The 13 C NMR data of 4 are similar to those of 2 and 3 except for the presence of a methoxy carbon (δC 55.2) in 4. In the HMBC spectrum, correlation from the protons of this carbon to C-1′ (δC 76.8) was observed, indicating that the OH-1′ in 2 or 3 was methylated in 4. The remaining substructure of 4 was determined to be the same as those of 2 and 3 by detailed analysis of its 2D NMR data ( Figure 2). The absolute configuration of the stereocenter C-1′ in 4 was determined to be the same as that of 2 by their similar ECD curves ( Figure 3) and same sign of optical rotation values. Compound 5 was obtained as yellow powder, and its formula was determined as C20H18O4 on the basis of HRESIMS data, indicating 12 degrees of unsaturation. Comprehensive analysis of the NMR data of 5 revealed that its structure is very similar to that of cladosporol I (6), except for the absence of the C-1′ carbonyl and the presence of a double bond which was located at C-1′ (δC 121.9) and C-2′ (δC 125.9) in 5, as deduced from the COSY correlations of H-1′/H-2′/H-3′/H-4′ as well as the HMBC correlations from H2-3′ (δH 2.29, 2.47) to C-1′ and C-4′a (δC 139.4) and from H-2′ (δH 5.89) to C-4′ (δC 37.5) and C-8′a (δC 121.2). The absolute stereochemistry of the asymmetric C-4′ was estimated from the ECD spectrum. In the ECD spectrum of 5, there was a strong splitting cotton effect centered at 216 nm, which should be originated from exciton chirality caused by interaction between two phenolic chromophores [12]. Thus, according to the exciton chirality rule, the absolute stereochemistry of C-4′ was determined to be S. In addition, the absolute configuration of the C-1 stereocenter in cladosporol I (6) has been determined by comparison of the experimental ECD curve with the calculated ECD curves of the two C-1 epimers (1R,4S)-6 and (1S,4S)-6 [8]. According the results, the calculated ECD curve for (1S,4S)-6 showed an intense positive CE around 225 nm, which matched well with the experimental one and was absent in the calculated ECD curve for (1R,4S)-6. Thus, the intense positive CE around 225 nm can be used to differentiate (1R,4S)-6 and (1S,4S)-6. The experimental ECD curve of 5, with an intense positive CE around 225 nm, is very similar to that of (1S,4S)-6, indicating that the absolute configuration of C-1 is S, same as that of 6. The molecular formula of compound 4 was determined as C 21 H 20 O 2 based on HRESIMS, indicating 12 degrees of unsaturation. The 13 C NMR data of 4 are similar to those of 2 and 3 except for the presence of a methoxy carbon (δ C 55.2) in 4. In the HMBC spectrum, correlation from the protons of this carbon to C-1 (δ C 76.8) was observed, indicating that the OH-1 in 2 or 3 was methylated in 4. The remaining substructure of 4 was determined to be the same as those of 2 and 3 by detailed analysis of its 2D NMR data ( Figure 2). The absolute configuration of the stereocenter C-1 in 4 was determined to be the same as that of 2 by their similar ECD curves ( Figure 3) and same sign of optical rotation values.
Compound 5 was obtained as yellow powder, and its formula was determined as C 20 H 18 O 4 on the basis of HRESIMS data, indicating 12 degrees of unsaturation. Comprehensive analysis of the NMR data of 5 revealed that its structure is very similar to that of cladosporol I (6), except for the absence of the C-1 carbonyl and the presence of a double bond which was located at C-1 (δ C 121.9) and C-2 (δ C 125.9) in 5, as deduced from the COSY correlations of H-1 /H-2 /H-3 /H-4 as well as the HMBC correlations from H 2 -3 (δ H 2.29, 2.47) to C-1 and C-4 a (δ C 139.4) and from H-2 (δ H 5.89) to C-4 (δ C 37.5) and C-8 a (δ C 121.2). The absolute stereochemistry of the asymmetric C-4 was estimated from the ECD spectrum. In the ECD spectrum of 5, there was a strong splitting cotton effect centered at 216 nm, which should be originated from exciton chirality caused by interaction between two phenolic chromophores [12]. Thus, according to the exciton chirality rule, the absolute stereochemistry of C-4 was determined to be S. In addition, the absolute configuration of the C-1 stereocenter in cladosporol I (6) has been determined by comparison of the experimental ECD curve with the calculated ECD curves of the two C-1 epimers (1R,4S)-6 and (1S,4S)-6 [8]. According the results, the calculated ECD curve for (1S,4S)-6 showed an intense positive CE around 225 nm, which matched well with the experimental one and was absent in the calculated ECD curve for (1R,4S)-6. Thus, the intense positive CE around 225 nm can be used to differentiate (1R,4S)-6 and (1S,4S)-6. The experimental ECD curve of 5, with an intense positive CE around 225 nm, is very similar to that of (1S,4S)-6, indicating that the absolute configuration of C-1 is S, same as that of 6.

Fungus Material
The fungal strain Cladosporium sp. KFD33 with yellow mycelium was isolated from a blood cockle in the Haikou Bay, Hainan province, in China in August 2018. After grinding, the sample (1 g) was diluted to 10 −2 g/mL with sterile H 2 O, 100 µL of which was spread on a PDA medium plate containing chloramphenicol as bacterial inhibitor. It was identified by its morphological characteristics and 18S rRNA gene sequences (GenBank accessing No. MN737504, Supporting Information), the used primers of which were NS1 (GTAGTCATATGCTTGTCTC) and NS6 (GCATCACAGACCTGTTATTGCCTC). A reference culture of Cladosporium sp. KFD33 is deposited in our laboratory and which maintained at −80 • C.

Culture Conditions
Plugs of agar, supporting mycelial growth, were cut from solid culture medium and transferred aseptically to a 1000 mL Erlenmeyer flask, containing 300 mL liquid medium (peptone 5 g/L, yeast extract 2 g/L, glucose 20 g/L, MgSO 4 0.5 g/L, KH 2 PO 4 2 g/L, pH 6.5). The fungus was cultured under static conditions at room temperature for 30 days on the shelf of our laboratory.

Extraction and Isolation
The whole culture broth (20 L) was harvested and filtered to yield the mycelium cake and liquid broth. The mycelium cake and liquid broth were extracted by EtOAc for three times, respectively. The EtOAc solution was evaporated under reduced pressure. A total of 10 g EtOAc extract was obtained. The EtOAc extract was submitted to silica gel vacuum liquid chromatography using step gradient elution with PE/EtOAc

QS Inhibitory Assays of C. Violaceum CV026
An overnight culture of C. violaceum CV026 [9] in LB broth (OD 600 ≈ 1.0) was prepared. Then 15 mL LB agar plate were flooded with 1 mL of this culture and C6-HSL at concentration of 500 nM to prepare an agar plate as a lawn. 5-mm diameter wells were bored in the agar plate using a flame-sterilized glass tube and the compounds dissolved in DMSO were added to the wells respectively. These plates were then incubated at 30 • C for 18 h. Inhibition of QS in C. violaceum is manifested as the inhibition of purple pigmentation around the wells containing the compounds

Conclusions
In conclusion, we found five new altertoxins with QS inhibitory activities against Chromobacterium violaceum CV026 from the marine-derived fungus Cladosporium sp. KFD33. These compounds represent a new type of QSIs, and can be used as lead compounds for developing new QSIs drugs, which can disarm pathogens without causing resistance of bacteria.

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