Novel Antioxidants and α-Glycosidase and Protein Tyrosine Phosphatase 1B Inhibitors from an Endophytic Fungus Penicillium brefeldianum F4a

Oxidative stress plays a very important role in the progression of diabetes and its complications. A therapeutic agent that is both antidiabetic and antioxidant would be the preferred choice for the treatment of diabetes. The crude extract of the endophytic fungus Penicillium brefeldianum F4a has significant antioxidant and α-glycosidase and protein tyrosine phosphatase 1B (PTP1B) inhibition activities. Chemical investigation of P. brefeldianum F4a using an activity-guided isolation led to the discovery of three new compounds called peniorcinols A–C (1–3) along with six known compounds: penialidins A (4), penialidin F (5), myxotrichin C (6), riboflavin (7), indole-3-acetic acid (8), and 2-(4-hydroxy-2-methoxy-6-methylphenyl) acetic acid (9). Their chemical structures were established by their NMR and HRESIMS. The absolute configurations of 1 and 3 were determined by experimental and calculated electronic circular dichroism (ECD). Their antioxidant activities were evaluated by DPPH• and ABTS•+ scavenging assays. Compounds 1–6 and 8–9 showed moderate to strong free radical scavenging activities. Significantly, 4–6 exhibited more potent ABTS•+ scavenging activity than that of the positive control. Their α-glycosidase and PTP1B inhibition activities were tested. Among them, compound 3 showed α-glucosidase inhibition activity, and compounds 7 and 8 showed PTP1B inhibitory activity for the first time. It is worth noting that 3 and 8 displayed both antioxidant and α-glycosidase or PTP1B inhibition activities. These finding suggest that compounds 3 and 8 could be used as lead compounds to generate new potent drugs for the treatment of oxidative stress-related diabetes.


Introduction
Reactive oxygen species are produced by living organisms because of normal cellular metabolism and environmental factors, which can damage cell structures and alter their functions [1]. The balance between oxidant and antioxidant in favor of oxidants is termed "oxidative stress" [2], which may cause many diseases, including diabetes mellitus [3] and cancer [4]. Notably, much evidence indicates that oxidative stress plays a very important role in the progression of diabetes and its complications [3,5,6], for instance, diabetic nephropathy, cardiomyopathy, and diabetic neuropathy [7,8]. In addition, uncontrolled chronic hyperglycemia is directly related to free radical overload [9]. Therefore, a therapeutic agent that possesses both antioxidants and antidiabetics would be the preferred choice for diabetes [7,8]. 2 of 10 Endophytic fungi have proven to be an important source of secondary metabolites with novel structures and various biological effects [10,11]. The secondary metabolites from endophytic fungi were isolated from two main genera, Aspergillus and Penicillium [12,13]. Penicillium is a widely explored versatile genus due to its chemical diversity and associated biological properties [14,15]. Recently, some novel antidiabetic and antioxidant secondary metabolites isolated from Penicillium have been continuously reported. For instance, two new isocoumarins, penicimarins L and M, with antioxidant and α-glucosidase inhibitory activities were isolated from fungus Penicillium sp. MGP11 [16]. Three xanthones as α-glucosidase inhibitors were isolated from an endophytic Penicillium canescens [17].
During our ongoing search for natural products with antioxidant and antidiabetic activities from endophytic fungi, the crude extract of Penicillium brefeldianum F4a showed significant ABTS •+ scavenging activity (EC 50 = 0.03 mg/mL) and DPPH • scavenging activity (EC 50 = 0.34 mg/mL), along with α-glycosidase inhibition activity (EC 50 = 0.13 mg/mL) and PTP1B inhibition activity (EC 50 = 0.03 mg/mL). Therefore, in order to investigate dualactive secondary metabolites with antioxidant and potential antidiabetic activities from the endophyte strain F4a, three new compounds, peniorcinols A-C (1-3), and six known ones (4-9) were isolated. Herein, we outline their isolation and structure elucidation, together with antioxidant and α-glycosidase and PTP1B inhibition activities.

General Experimental Procedures
Semipreparative HPLC was recorded using a Dionex UltiMate 3000 HPLC system equipped with multiple wavelength detectors utilizing YMC-Pack-ODS-A column (250 × 10 mm, 5 µm). Optical rotations were acquired by using an SGW-2 digital polarimeter. IR spectra were measured with a Thermo fisher Nicolet 6700 FT-IR spectrometer in KBr discs. HRESIMS data were obtained on a Thermo Scientific Q Exactive mass spectrometer. 1 H NMR (600 MHz), 13 C NMR (150 MHz), and 2D NMR were tested by a Bruker-AV-600 NMR spectrometer using solvent signals as references. The ECD spectra were measured using a Science MOS-450 spectrometer. Column chromatography was performed with silica gel (200-300 mesh; Qingdao Ocean Chemical Co. Ltd., Qingdao, China), Sephadex LH-20 gel (GE Healthcare, Uppsala, Sweden), and ODS-A reversed-phase silica gel (50 µm; YMC Co. Ltd., Kyoto, Japan). The supernatant was treated with macroporous resin HP20 (Mitsubishi Chemical Co. Ltd., Kyoto, Japan).

Fungal Material
The fungus F4a was isolated from the roots of H. cordata and identified as P. brefeldianum (a former synonym of Eupenicillium brefeldianum) [18], and it has been stored at the Institute of Applied Ecology, Chinese Academy of Sciences.

Fermentation and Extraction
The fungus F4a was incubated on a culture medium (0.25% barley extract, 0.1% yeast extract, 0.05% soybean cake, 2% glucose, 1% starch, 0.4% CaCO 3 , 0.3% KH 2 PO 4 , 0.25% MgSO 4 ·7H 2 O, 0.1% (NH 4 ) 2 SO 4 , 0.08% KNO 3 , and 0.002% CuSO 4 , pH = 7.0) at 28 • C for 7 days at 180 rpm. The fermentation broth was centrifuged (4000 rpm for 20 min) to obtain supernatant and mycelium. The supernatant was treated with macroporous resin HP20 and stirred for 2 h (28 • C, 180 rpm). The resin was then filtered and eluted three times with methanol. The methanol was collected and dried by a rotary evaporator to obtain extract A. The mycelium was extracted three times by acetone and put in an ultrasonic water bath for 40 min. The acetone was collected and dried by a rotary evaporator to obtain extract B. A total of 35 g was obtained by combining extracts A and B.

ECD Calculations
The CONFLEX analyses of compounds 1 and 3 were performed with Merck Molecular Force Field (MMFF). Subsequently, the conformations that showed >1% Boltzmann population were subjected to geometry optimization by density functional theory (DFT) at the B3LYP/6-311+G (d, 2p) level. Then, the time-dependent DFT (TDDFT) method at the CAM-B3LYP/TZVP level was used on the lowest energy conformations for each configuration with a PCM model in a methanol solution [19]. Finally, the overall theoretical ECD curves were simulated using SpecDis 1.71 software (University of Wurzburg, Wurzburg, Germany) on the basis of the Boltzmann weighting of each conformer [20]. All calculations were performed with the Gaussian 09 program package [21].

DPPH Radical Scavenging Assay
The DPPH free radicals scavenging assay was tested according to the previously reported methods [22]. First, 100 µL of sample at different concentrations in ethanol (the blank control used 100 µL of ethanol instead of the sample) was added to 96 wells, followed by 100 µL of 0.15 mM DPPH. After incubating for 30 min in darkness at room temperature, the reduction of DPPH radical was determined by measuring the absorbance at 517 nm with a microplate reader. L-Ascorbic acid was used as the positive control.

ABTS Radical Scavenging Assay
The ABTS free radical scavenging assay was tested according to the previously reported methods [23]. The ABTS radical cation solution was prepared by mixing 2.45 mM potassium persulphate and 7 mM ABTS (1:1, v/v). The mixture was incubated in darkness at room temperature for 12 h before use. The ABTS cation radical solution was diluted with water to obtain an absorbance of 0.70 ± 0.02 at 734 nm. Then, 100 µL of the sample at various concentrations in ethanol (the blank control used 100 µL of ethanol instead of the sample) was added to 3.0 mL of diluted ABTS •+ solution and mixed vigorously. After incubating for 6 min in darkness at room temperature, the absorbance was measured at 734 nm with a microplate reader. L-Ascorbic acid was used as the positive control.

Antidiabetic Assay In Vitro
The α-glucosidase inhibitory assay was determined by the method described by Apostolidis and Lee with slight modifications [24]. First, 30 µL of the 2 U/mL α-glycosidase in 0.1 M phosphate buffer solution (PBS) and 20 µL of the sample at different concentrations in DMSO (the blank control used 20 µL of DMSO instead of the sample) were mixed in 800 µL of 0.1 M PBS (pH 6.8), incubating for 5 min at 37 • C in a water bath. After preincubation, 150 µL of 10 mM p-nitrophenyl glucopyranoside (p-NPG) was added to the mixture to initiate the reaction. After reacting for 30 min, 650 µL of 1 M Na 2 CO 3 solution was added to terminate the reaction. The enzyme activity was determined by measuring the absorbance of p-nitro phenol at 405 nm with a microplate reader. Acarbose was used as the positive control.

PTP1B Inhibitory Assay
The PTP1B inhibitory activity was established according to the previously reported methods [25]. The reagent para-nitrophenyl phosphate (p-NPP) was used as a substrate for PTP1B. Then, 81 µL enzyme in buffer (pH 7.5), including EDTA (1 mM), Tris-HCl (25 mM), dithiothreitol (DTT) (1 mM), β-mercaptoethanol (2 mM), and 10 µL of sample at various concentrations in DMSO (the blank control used 10 µL of DMSO instead of the sample), was incubated in 96-well plates at 37 • C for 10 min. Then, 4 µL p-NPP (10 mM) in buffer was added and incubated at 37 • C for 30 min. Next, 5 µL of 2 M NaOH solution was added to terminate the reaction. Then, the absorbance was determined at 405 nm wavelength. Sodium orthovanadate (Na 3 VO 4 ) was used as the positive control.

Bioactivity-Guided Compound Isolation
The active extract of the strain P. brefeldianum F4a showed significant DPPH • and ABTS •+ scavenging activities, together with α-glycosidase and PTP1B inhibition activities. Then, the extract was further isolated by silica gel column chromatography to afford nine fractions (A-I), and fractions B and C, together with E, showed bioactivity. Fraction B (4.5 g) displayed significant ABTS •+ scavenging and PTP1B inhibition activities, fraction C (2.2 g) showed antioxidant activity, and fraction E (0.5 g) exhibited significant PTP1B inhibition activity, suggesting that the strain F4a can not only produce secondary metabolites with antioxidant or α-glycosidase and PTP1B inhibition activities but also may produce secondary metabolites with the above two kinds of activities. Compound 8 (3.4 mg) was recrystallized from ABTS •+ scavenging and PTP1B inhibition activities subfraction B 1 (0.1 g).  5 g). Finally, all of the monomer compounds were systematically evaluated for four biological activity assays. Although compound 3 was isolated from the antioxidant subfraction B 2 , compound 3 exhibited weak α-glycosidase inhibition activity ( Table 2). This could be because the content of compound 3 is relatively low, resulting in the α-glycosidase inhibition activity of subfractions B 2 not being detected. The chemical structures of compounds 1-9 are shown in Figure 1. (10.0 mg) was isolated from fractions E (0.5 g). Finally, all of the monomer compounds were systematically evaluated for four biological activity assays. Although compound 3 was isolated from the antioxidant subfraction B2, compound 3 exhibited weak α-glycosidase inhibition activity (Table 2). This could be because the content of compound 3 is relatively low, resulting in the α-glycosidase inhibition activity of subfractions B2 not being detected. The chemical structures of compounds 1-9 are shown in Figure 1.
Compound 2 was obtained as a white powder with the molecular formula of C 15 Figures 2, S12 and S13). Therefore, compound 2 was characterized as a new derivative of 1 and named peniorcinol B. of H-7′/C-3, C-2′, C-6′ and H-11′/C-4, C-5 (Figures 2, S5 and S6). To determine the absolute configuration of 1, the optimized conformations of (12′R)-1a and (12′S)-1b were obtained at the B3LYP/6-311+G (d, 2p) level and used for ECD calculations. The calculated ECD curve showed a high degree of similarity to the experimental ECD spectrum of 1a (Figure 3), which indicated the absolute configuration of 1 is 12′R. Therefore, compound 1 was characterized as a new skeleton compound with orcinol structure and named peniorcinol A.   Table 1). The data of 1D NMR (δH 2.09, s and δC 19.2, CH3), HRESIMS, and the HMBC correlations of H-11′/C-4, C-5 revealed that the 2-hydroxy-propyl group of 1 was replaced by a methyl in 2 (Table 1, Figures 2, S12 and S13). Therefore, compound 2 was characterized as a new derivative of 1 and named peniorcinol B.

Antioxidant Activities
The antioxidant activities of compounds 1-9 were assessed using two methods: DPPH and ABTS. In the DPPH • assay (Table 2), compounds 5 and 6 showed moderate activity (EC 50 = 28.42 and 30.07 µM) compared with the positive control (EC 50 = 6.12 µM). To the best of our knowledge, this is the first report of DPPH • scavenging activities of compounds 5 and 6. The DPPH • scavenging activities of 4 have been reported in previous studies (IC 50 = 156.38 µg/mL) [32], which is in agreement with our results (EC 50 > 100 µM). Compared with the same skeleton of chromone (4-6), compound 4 (EC 50 > 100 µM) was found to be less active in DPPH • scavenging than compounds 5 and 6 (EC 50 = 28.42 and 30.07 µM), suggesting that the carboxyl group (-COOH) significantly reduced the DPPH • scavenging activity.
Since the discovery of PTP1B in 1988, it has become a highly plausible candidate for therapeutic inhibitors of obesity and type 2 diabetes [34]. As shown in Table 2, compounds 7 and 8 showed moderate PTP1B inhibitory activity (EC 50 = 8.87 and 11.68 µM, respectively) compared with the positive control (EC 50 = 2.38 µM). To the best of our knowledge, this is the first report of their PTP1B inhibitory activities.
There is a large body of evidence demonstrating that oxidative stress plays a very important role in the progression of diabetes and its complications [3,5,6]. Therefore, a therapeutic agent that has both antidiabetic and antioxidant activities would be the preferred choice for diabetes. Compounds 3 and 8 displayed both antioxidant and αglycosidase or PTP1B inhibition activities, which could be explored as potential natural antioxidant and antidiabetic agents for use in oxidative stress-related diabetes. Further, it is necessary to study the antidiabetic activities of these compounds in vivo.

Conclusions
In summary, three new compounds, peniorcinols A-C (1-3), and six known compounds (4-9) were isolated from the endophytic fungus P. brefeldianum F4a. In the bioassays, compounds 1-6 and 8-9 showed moderate to strong radicals scavenging activity. Significantly, compounds 4-6 exhibited more potent ABTS •+ scavenging activity than that of the positive control. Additionally, compounds 3, 7, and 8 showed potential αglycosidase or PTP1B inhibition activities. Compounds 3 and 8 displayed both radical scavenging activity and α-glycosidase or PTP1B inhibition activity, which suggests that they could be used as lead compounds to generate potent drugs for the treatment of oxidative stress-related diabetes.