Chemical diversity from the Tibetan Plateau fungi Penicillium kongii and P. brasilianum

ABSTRACT Two new secondary metabolites, kongiilines A and B (1, 7), and two asperphenamate derivatives, asperphenamates B and C (5–6), together with 16 known compounds (2–4, 8–20), were isolated from Tibetan Plateau fungi Penicillium kongii and Penicillium brasilianum. This is the first report on asperphenamates B and C as naturally occurring compounds, and that aspterric acid is isolated from P. brasilianum for the first time. Their structures were elucidated by different spectroscopic techniques including high-resolution electrospray ionisation mass spectrum, 1D nuclear magnetic resonance (NMR), and 2D NMR as well as electronic circular dichroism. Compounds 4, 5, and 10 exhibited cytotoxicity activities against human colon carcinoma HCT116 cell line with IC50 values of 88.16, 77.68, and 36.92 μM, respectively. Fungi from Tibetan Plateau represent important and rich resources for the investigation of new chemicals.


Introduction
The genus Penicillium is a major contributor to the production of bioactive molecules. A number of compounds from this genus have been characterised and used for drugs and mycotoxins. For example, the antibiotic penicillin is produced by Penicillium rubens (Houbraken et al. 2011), the immunosuppressive drug mycophenolic acid is produced by Penicillium brevicompactum (Regueira et al. 2011). There were many new compounds that have been found constantly in the following years with impressive anticancer and antifungal activities (Bladt et al. 2013;Kozlovskii et al. 2013;Tang et al. 2015;Koul et al. 2016). And many mycotoxins causing human and animal diseases were produced in some species of Penicillium, such as citreoviridin, citrinin (CIT), ochratoxin A (OTA), patulin (PAT), penitrem A, and penicillic acid (PA) (Lee & Ryu 2015;Oh et al. 2017). Because of the importance of this genus, chemical assessments on more and more Penicillium species have been investigated from different environmental sources. The endophytic fungi (Arunpanichlert et al. 2010;Zhu et al. 2015) and soil fungi (Tansakul et al. 2014;Daengrot et al. 2015) were the most studied fungi in this genus and a number of natural products have been reported to exhibit diverse activities including anti-inflammatory, cytotoxic (Xue et al. 2014), or anti-hepatitis C virus (Kozlovskii et al. 2013;Nishikori et al. 2016). Other strains from extreme conditions also raised the great interest to mine novel structures, such as halotolerant or extremophilic strains (Lu et al. 2008;Stierle et al. 2012)， deep sea-derived strains , and marine animal-derived strains (Qi et al. 2013). The resulted compounds cover different biological activities including cytotoxic activity against cancer cell lines Zhang et al. 2016), potent inhibitory activity against bacteria (Zheng et al. 2016), antifungal activity (Daengrot et al. 2015), inhibiting LPSinduced inflammation (Shin et al. 2016), and inhibitory activity against protease (Sun et al. 2014). In comparison to the above environment-derived Penicillium strains, research on the Penicillium strains from Tibetan Plateau is limited.
Tibetan Plateau, as the highest plateau in the world, is exposed to strong ultraviolet radiation, and has low-temperature and low-oxygen environments.
In the past few years, a series of bioactive molecules or new skeletons has been isolated from Tibetan Plateau origin fungi, such as phaeolschidins, with antioxidant activity, from Phaeolus schweinitzii (Abbas et al. 2013;Han et al. 2013), new skeletons, sterhirsutins, with cytotoxic and immunosuppressant activities from Stereum hirsutum (Qi et al. , 2015, anthraquinone derivatives, with antitumour activities, from an Alternaria species (Chen et al. 2014), sarcoviolins, with antioxidative and α-glucosidase inhibitory activities, from Sarcodon leucopus (Ma et al. 2014a), Gloeophyllins A-J, as cytotoxic ergosteroids, from Gloeophyllum abietinum (Han et al. 2015). Therefore, fungi from Tibetan Plateau represent an important fungal resource for the discovery of novel chemical molecules.
In order to probe the chemical diversity from Tibetan Penicillium, 24 Penicillium strains from Tibetan Plateau were screened and evaluated for secondary metabolite production (data not shown). Based on the high-performance liquid chromatography (HPLC)-UV and/or liquid chromatography-mass spectrometry (LC-MS) analysis, two of them, P. kongii and P. brasilianum with abundant chemical profiles, were selected as targeted strains for further studies. Herein, we described the compound isolation, structural elucidation, and bioactivity evaluations for new compounds.

Fungal materials
Plant leaf samples were collected from growing trees and kept in sterilised plastic bags. Isolation of phylloplane fungi was according to Nakase and Takashima (1993). Penicillium strain XZ135R was identified as P. kongii (isolated from leaves of Cotoneaster sp., Gongbujiangda, Linzhi, Tibet, China), and XZ94 was identified as P. brasilianum (isolated from leaves of an unidentified plant, Cuona, Shannan, Tibet, China) according to beta-tubulin gene (BenA, GenBank No. MF036174) and calmodulin gene (CaM,GenBank No. MF039288). The method of DNA extraction, the primers for amplifying the aforesaid two genes (BenA and CaM), and the polymerase chain reactions followed the methods of Wang and Wang (2013). The phylogenetic trees were drawn using maximum likelihood method and subjected to 1000 bootstrap replications (Frisvad & Samson, 2004) and supplied as Supporting Information Figures S1 and S2. The strains were stocked on slants of potato dextrose agar (PDA) at 4°C and deposited in China General Microbiological Culture Collection of Institute of Microbiology, Chinese Academy of Sciences, Beijing, China as AS3.15332 and AS3.15722, respectively, and they were also maintained at the authors' laboratory and will be supplied upon request for educational or scientific purpose. The two strains were activated and cultured on plates of PDA at 25°C for 7 d. Then, cultures were transferred to the modified YES solid medium (2% yeast extracts (BD Bacto TM ), 15% sucrose, and 1.5% agar) with 20.0 mL on each plate at 25°C for 7 d (Frisvad 1981). For the large-scale fermentations, 12 L YES solid cultures of P. kongii and 3 L YES solid cultures of P. brasilianum were performed, respectively. They were growing at 25°C for 7 d before harvest.

Analysis methods
Analytical HPLC was conducted with a Waters HPLC system (Waters e2695, Waters 2998, Photodiode Array Detector) using an ODS column (C18, 250 mm × 4.6 mm, YMC Pak, 5 μM) with a flow rate of 1 mL/min. Fresh extracts were dissolved in methanol before separated on a linear gradient of MeOH:H 2 O (0.1% formic acid) at a flow rate of 1 mL/min. Fresh extracts from screened strains were detected for 30 min using a linear gradient of 20-100% (0-20 min), 100% MeOH (20-25 min), 20% MeOH (25-30 min). LC-MS analysis method was used as keeping consistent with analytical HPLC.

General experimental procedures
The optical rotations were measured on a Perkin-Elmer 241 polarimeter (Waltham, USA) and UV spectra were determined on a Thermo Genesys-10S UV-Vis spectrophotometer (Madison, USA). Electronic circular dichroism (ECD) spectra were recorded on a JASCO J-815 spectropolarimeter (Tokyo, Japan). NMR spectra were recorded on a Bruker Avance-500 spectrometer using tetramethyl silicane (TMS) as internal standard, and chemical shifts were recorded as δ values. High-resolution electrospray ionisation mass spectrum (HR-ESI-MS) and LC-MS were utilised on an Agilent Accurate-Mass-QTOF LC-MS 6520 instrument. Sephadex LH-20 was purchased from GE Healthcare. TLC was carried out on Silica gel HSGF254 and the spots were visualised by spraying with 10% H 2 SO 4 and heating. Silica gel (Qingdao Haiyang Chemical Co., Ltd., Qingdao, China) and ODS (Lobar, 40-63 mm, Merck, Darmstadt, Germany) were used for column chromatography (CC). RP-HPLC separations were conducted using a Shimadzu LC-6AD liquid chromatograph with a YMC-PACK ODS-A column (250 mm × 10 mm, 5 µm) and a Shimadzu SPD-20A VP UV-Vis detector with a flow rate of 3 mL/min. Solvents used for extraction and chromatographic separation were of analytical grades.

Cytotoxicity assays
Cytotoxic activity against HCT116 human colon carcinoma cells was assayed according to the 3-[4,5dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide (MTT) assay method (Ma et al. 2014b). Cells were incubated with tested compounds (DMSO as solvent) at 37°C in a humidified atmosphere of 5% CO 2 95% air for 72 h. Each well was added 50 μL of MTT/medium solution (0.5 mg/mL), and cells were incubated for another 4 h. After removing the MTT/medium, 100 μL of DMSO was added to each well. The plate was shaken to dissolve the precipitates, and activity was measured at 540 nm using a microplate reader. The inhibition rates were calculated and plotted versus test concentrations to afford the IC 50 (± SD) for three independent experiments, each was carried out in triplicate. Taxol was used as the reference substance that showed cytotoxicity against HCT116 human colon carcinoma with IC 50 value of 0.98 ± 0.12 μM.
Asperphenamate C (6) has the same molecular formula C 20 H 30 N 2 O 5 with m/z 523.2231 [M + H] + (calcd for C 32 H 31 N 2 O 5 , 523.2227) as 5 based on the HR-ESI-MS analysis. Comparison of the 1D and 2D NMR data between 6 and 5 indicated that the gross structure of compound 6 is the same as that of 5 with the difference of the location of one of the hydroxyl groups ( Table 2). The hydroxyl group was determined at C-14 from the observation of the HMBC correlations from H-12/H-16 to C-10 (δ 170.3), C-13/C-15 (δ 116.1), and C-14 (δ 162.3). The absolute configuration of 6 was determined by the ECD spectrum, which showed the similar Cotton effect with asperphenamate. Furthermore, the specific rotation of 5 was similar to that of asperphenamate, which was consistent with asperphenamate (McCorkindale et al. 1978). Thus, the 3D configurations of 2 and 2′ position are same between compound 6 and asperphenamate, and compound 6 was defined as N-p-phydroxybenzoylphenylalanine-2-benzoylamino-3-phenyl propyl ester. To the best of our knowledge, compound 6 was isolated from nature for the first time.
The isolated compounds were tested for their cytotoxicities against human cancer cells (Table 3). Hydroxylated asperphenamates asperphenamate B (5) and asperphenamate C (6) displayed very weak cytotoxicity activities against human cancer cell line HCT116. The IC 50 values for 5 and 6 were detected at 88.16 and 77.68 μM, respectively (Table 3). Known compound mycophenolic acid (10) exhibited activities against HCT116 with IC 50 value of 36.92 μM (Table 3). This compound has shown previously outstanding anti-proliferative immunosuppressive activity and was used as drug for organ transplant patients. The other tested compounds were inactive.
Among these compounds, asperphenamate is composed of two amino acids via ester bond connection and has been reported with anticancer activity (Pomini et al. 2006). Subsequently, several derivatives were synthesised for improving the solubility and activity. For example, asperphenamate derivatives 1a and 1c, as a pair of isomers, showed activities against T47D, MDA-MB231, and HL60 cell lines, and 1c exhibited the most potent activities against breast cancer cell lines T47D and MDA-MB231 (IC 50 = 8.2 and 11.9 μM, respectively) (Yuan et al. 2010). Asperphenamate derivative IM23b, benzyl modified at C15 of asperphenamate, showed the greatest potency in human breast cancer MCF-7 cells (Yuan et al. 2012). This compound was also found in P. brevicompactum. The phylogenetic and morphological study showed that P. kongii belongs to Penicillium section brevicompacta and has some kinship with P. brevicompactum . Recently, genome of P. brasilianum has been sequenced (Horn et al. 2015), which provides the opportunity to elucidate its biosynthetic pathway.