Chromolaena laevigata (Asteraceae) as a source of endophytic non-aflatoxigenic Aspergillus flavus: chemical profile in different culture conditions and biological applications

Abstract

Endophytes are microorganisms that form symbiotic relationships with their host. These microorganisms can produce a variety of secondary metabolites, some of which have inhibitory effects on pests and pathogens or even act to promote plant growth. Due to these characteristics, these microorganisms are used as sources of biologically active substances for a wide range of biotechnological applications. Based on that, the aim of this study was to evaluate the production of metabolites of the endophytic Aspergillus flavus CL7 isolated from Chromolaena laevigata, in four different cultivation conditions, and to determine the antimicrobial, cytotoxic, antiviral, and antioxidant potential of these extracts. The multiphasic approach used to identify this strain was based on morphology and ITS gene sequence analysis. The chemical investigation of A. flavus using potato dextrose and minimal medium, using both stationary and agitated methods, resulted in the isolation of kojic acid, α-cyclopiazonic acid, and 20,25-dihydroxyaflavinine. Another 18 compounds in these extracts were identified by UHPLC-HRMS/MS, of which dideacetyl parasiticolide A has been described for the first time from A. flavus. Aflatoxins, important chemomarkers of A. flavus, were not detected in any of the extracts, thus indicating that the CL7 strain is non-aflatoxigenic. The biological potential of all extracts was evaluated, and the best results were observed for the extract obtained using minimal medium against Trichophyton rubrum and Mycobacterium tuberculosis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-021-00502-6.

Introduction

Endophytic microorganisms colonize intra- and intercellular plant tissues. This colonization/association is asymptomatic, causing no damage to the host plant. Among the microorganisms capable of this association, the most common are bacteria and fungi [1]. Our research group has been studying the chemical and biological diversity of the Eupatorieae tribe [2–5]; as part of our investigation of species belonging to this tribe, we selected Chromolaena laevigata (basionym: Eupatorium laevigatum) to characterize its foliar endophytic fungi. Chromolaena laevigata (Lam.) RM King & H. Rob is a shrub native from Mexico to Argentina [6], widely distributed in Brazil. This species is usually found in the Cerrado, Atlantic Forest, Amazon, and Caatinga [7] and is considered an invasive species prone to infesting pastures, crops, roadsides, and vacant lots [8]. Previous studies on the aerial parts of C. laevigata have shown the presence of cadinene, clerodane, steroids, and flavonoids [5, 9]. Aspergillus flavus was identified here as an endophytic fungus from this species.

Aspergillus is widespread in nature and includes more than 250 species. Several Aspergillus species are classified as pathogens of harvested fruits and seeds, especially A. flavus which is well known for its production of toxins such as aflatoxins and cyclopiazonic acid [10–12]. Among the various classes of secondary metabolites produced by Aspergillus species, aflatoxins play an important role as chemomarkers. The word “aflatoxin” is derived from the expression “Aspergillus flavus toxin”, since A. flavus and A. parasiticus, as well as A. nomius, A. bombycis, and A. pseudotamarii, are the species predominantly responsible for aflatoxin contamination of agricultural crops before harvest or during storage [13]. In addition to aflatoxins, A. flavus can also produce a range of other compounds of different classes, such as cyclopiazonic acid, β-nitropropionic acid, aspertoxin, aflatrem, gliotoxin, aspergillic acid, 20,25-dihydroxyaflavinine, and versicolorin A [14].

There are promising possibilities in exploring the biotechnological potential of endophytic fungi to produce a multitude of known and new biologically active specialized metabolites. For example, using controlled fermentation conditions and changing the process parameters, such as the type of medium, aeration, pO₂, pCO₂, pH, temperature, agitation, sampling, and harvest points, the production of specialized metabolites can be optimized [15–17], and the variation of medium conditions has been shown to have variable effects on the production of these metabolites [16, 18, 19]. Often, the variation of culture conditions is used to optimize the yields of a specific compound, such as an active metabolite or drug produced through microorganisms [20–23].

Herein, we evaluated the influence of different cultivation conditions on the production of specialized metabolites from an A. flavus strain isolated from a new host, C. laevigata. First, the multiphasic approach, based on morphology and ITS gene phylogenetic analysis, used to identify endophytic A. flavus is described. Subsequently, the metabolites produced when this endophyte was submitted to four different cultivation conditions were identified using 1D and 2D NMR spectroscopic experiments, and a dereplication technique using UHPLC-HRMS/MS. In vitro antimicrobial, antiviral, and antioxidant activity was also evaluated. Three compounds were isolated, kojic acid, cyclopiazonic acid and 20,25-dihydroxyaflavinine, and 18 compounds were identified by dereplication; however, no aflatoxins were detected in any of the culture media, indicating a non-aflatoxigenic A. flavus strain, with biotechnological potential for use as a biocontrol tool.

Materials and methods

General experimental procedures

Chromatography separations were performed on Sephadex LH-20 (Sigma) chromatography columns (CC). Thin layer chromatography (TLC) was performed on normal-phase pre-coated silica gel 60G or 60GF₂₅₄ plates (Merck). Visualization of the compounds on TLC was accomplished by UV irradiation at 254 and 366 nm and/or by spraying with H₂SO₄/anisaldehyde/acetic acid/methanol (5 : 0.5 : 10 : 85 mL) solution followed by heating at 150 °C or Dragendorff’s solution.

NMR spectra were recorded on a VARIAN Mercury Plus spectrometer operating at 300 and 75.5 MHz, and a Bruker advance III HD spectrometer operating at 500 and 125 MHz, using CDCl₃ and DMSO-d₆ as solvents. The UHPLC analysis was performed on a Shimadzu Nexera X2 instrument, equipped with a CBM-20A system controller, two LC-30AD pumps, a CTO-30A column oven, and a SIL-30AC autosampler. The mass spectra were recorded on a Bruker IMPACT II mass spectrometer, with an electrospray ionization source (ESI) in the positive and negative ion modes, quadrupole-time of flight (Q-TOF) analyzer, and multichannel plate (MCP) detector.

Biological material

The endophyte CL7 was previously isolated from Chromolaena laevigata (Lam.) R. M. King & H. Rob. (Asteraceae) leaves and was retrieved from the Collection of Endophytic and Environmental Microorganisms at the Laboratory of Microbial Biotechnology (CMEA/LBIOMIC-UEM), Universidade Estadual de Maringá, Brazil. This endophyte was registered at the National System for the Management of Genetic Heritage and Associated Traditional Knowledge (SISGEN) under code AEE079A. Chromolaena laevigata (Lam.) R. M. King & H. Rob. (Asteraceae) is a native specimen; it was collected at Ponta Grossa city, Paraná State, Brazil (25° 05′ 16″ S, 50° 05′ 43″ W) in March 2016 and identified by Drª. Marta Regina Barrotto do Carmo. A voucher specimen was deposited at the herbarium at Universidade Estadual de Ponta Grossa (HUEPG 21697); in addition, this plant was registered at SISGEN under code A6E6D08.

Microculture

For morphological analysis, strain CL7 was submitted to the microculture technique. A circular piece of filter paper was placed in a sterile Petri dish, and a sterile glass slide (26 × 76 mm) was put on the filter paper. A block of lemon-leaf agar (100 g L⁻¹ crushed lemon leaves, 15 g L⁻¹ agar, 5 g L⁻¹ peptone, and 15 g L⁻¹ glucose) was cut with a sterile scalpel and transferred to the glass slide. Using a sterile wire needle, the fungus was then inoculated from the culture plate to the four sides of the agar block. A sterile coverslip was put over the block, and 1 mL of sterile water was put onto the filter paper. The plate cover was replaced, and it was incubated at 28 °C for 5 days. After 5 days, the slide was removed and the block was deposited on another slide containing a drop of Cotton Blue dye. The sample was analyzed using a ZEISS Axioskop 2 Plus optical microscope equipped with an Axiocam HRc camera.

DNA extraction, amplification, and phylogenetic analysis

The genomic DNA was extracted using a PowerSoil® DNA isolation Kit (MO BIO Laboratories, Inc., Carlsbad, CA, USA), with 200 mg of mycelia initially added to the tubes, before then following the manufacturer’s instructions.

Amplification of the ribosomal DNA region, ITS1-5.8S-ITS2, was performed using primers V9G (5′-TTACGTCCCTGCCCTTTGTA-3′) and ITS4 (5′-TCCCCGCTTATTGATATGC-3′) described by White et al. [24]. The PCR technique was performed under the following conditions: 50 ng of DNA, 1× PCR buffer (50 mM KCl, 200 mM Tris-HCl, pH 8.4), 1.5 U of Taq polymerase (Invitrogen, Carlsbad, CA, USA), 0.25 μM of each primer, 0.2 mM dNTP (Amersham Bioscience, Freiburg, Germany), and 3 mM MgCl₂ in a final volume of 25 μL. Amplification was performed on a TC1000-G thermocycler (DLAB®) using the following conditions: initial denaturation at 94 °C for 5 min, followed by 35 cycles of denaturation at 94 °C for 30 s, annealing at 48 °C for 50 s, and extension at 72 °C for 2 min, followed by a final extension at 72 °C for 5 min.

PCR products were purified using the enzyme alkaline phosphatase (shrimp; Sigma-Aldrich®) and exonuclease I (E. coli; BioLabs®). The extracted DNA was then subjected to quantification on 1% agarose gel. Amplification products were sequenced by ACTGene molecular analysis using an ABI-PRISM 3100 Genetic Analyzer (Applied Biosystems). The sequencing results were evaluated using BioEdit Sequence Alignment Editor v. 7.2.2. The results were compared with other sequences deposited at the NCBI (National Center for Biotechnology Information) database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Identification was based on the best percentage identity, followed by phylogenetic analysis.

Alignment and phylogenetic analysis were conducted using Geneious prime software (v. 2020.0.4). Based on the BLAST similarity to the CL7 sequence of the A. flavus clade, sequences of nearby species were rescued for phylogenetic reconstruction [12]. The gene sequences were aligned using MAFFT v. 1.4.0 [25]. For phylogenetic analysis based on maximum likelihood and Bayesian inference, MrModelTest v. 2.3 was used to choose the best evolutionary model [26]. The phylogenetic tree was constructed using MrBayes v. 2.2.4 [27], taking into consideration the parameters generated by MrModelTest, with a Markov chain Monte Carlo (MCMC) simulation which lasted until the average standard deviation of the split frequencies was below 0.01. The Bayesian probability was demonstrated on the nodes between each individual. The tree was edited with FigTree v. 1.4.2 and Adobe Illustrator v. 23.0.3 [28].

The DNA sequence was deposited in GenBank under accession number MN955851.1.

Culture conditions and isolation

Axenic maintenance of the endophytic CL7 strain was performed on potato dextrose broth (PD) (HiMedia M403), potato dextrose agar (HiMedia M096), and minimal medium (MM), at pH 6.8. The MM was prepared in the following concentrations: NaNO₃ 60.00 g L⁻¹, KH₂PO₄ 15.00 g L⁻¹, KCl 5.00 g L⁻¹, MgSO₄·7H₂O 5.00 g L⁻¹, FeSO₄·7H₂O 0.01 g L⁻¹, ZnSO₄·7H₂O 0.01 g L⁻¹, CuSO₄·7H₂O 0.01 g L⁻¹. At the time of use, the MM was diluted 1: 10 with distilled water and 10.00 g L⁻¹ D-glucose was added.

The CL7 strain was subjected to four different axenic cultivation conditions, using PD and MM and stationary (PD-S; MM-S) and agitated methods (PD-A; MM-A). The strain was cultivated in three 1-L Erlenmeyer flasks, each containing 500 mL of medium. A spore solution was prepared at a concentration of 10⁷ spores mL⁻¹ according to Uldahl et al. with modifications. Briefly, spores were harvested when the cultures were fully sporulated, which was achieved after 10 days of incubation. The spores were released by flooding the Petri dishes with 5 mL sterile distilled water containing 0.1% (v/v) of tween 80 and the gently rubbing the culture with a sterile L-shaped spreader, a hemocytometer slide was used to count spore concentration [29]. Subsequently, 500 μL of this solution was added to each culture flask. After inoculation, the flasks were stored in a growth greenhouse at 28 °C for 21 days for the stationary cultures and in a shaker culture incubator (Nova Instruments NI1712) at 28 °C at 110 rpm for 21 days for the agitated cultures.

At the end of the incubation period, the mycelium was separated from the broth by vacuum filtration using a filter membrane, Büchner funnel and Kitassato. The mycelium was stored in a drying oven at 55 °C for 5 days to determine the dry weight of the mycelium for each cultivation condition. After filtration, the broths (approximately 500 mL) were extracted with ethyl acetate (3 × 150 mL). The solvent present was evaporated using a rotary evaporator (Tecnal TE-210) at 37 °C and the resulting extract was submitted to column chromatography, NMR, and UHPLC-HRMS/MS analysis.

PD-S crude extract yielded compound 1 (491.7 mg), with no purification step. Part of the PD-A crude extract (60.0 mg) was subjected to a CC (Ф = 1.00 cm × h = 25.0 cm) in Sephadex LH-20 using methanol as the mobile phase, to give subfractions PD-A1 to PD-A18. Compound 1 (8.0 mg) was reisolated from subfraction PD-A6, and subfraction PD-A15 afforded compound 2 (16.0 mg).

An aliquot of MM-S (28.0 mg) was subjected to a CC (Ф = 1.00 cm × h = 25.0 cm) in Sephadex LH-20, using methanol as the mobile phase, to give subfractions MM-S1 to MM-S15. Subfraction MM-S6 yielded compound 3 (5.1 mg), and subfraction MM-S10 afforded compound 2 (3.5 mg), previously isolated from PD-A15. MM-A extract was not obtained in an amount sufficient for chromatographic purification.

UHPLC-HRMS/MS

The samples were prepared in MeOH (1.0 mg mL⁻¹) and chromatographic separations were performed using UHPLC on a Symmetry C18 column (75 × 2.0 mm i.d.; 1.6 μm Shim-pack XR-ODS III), maintained at a temperature of 40 °C. The mobile phase used in both the positive and negative ionization modes consisted of H₂O (solvent A) and CH₃CN (solvent B), with the addition of 0.1% formic acid when analyzed in positive ionization mode. The gradient program was as follows: initial 0–1 min, using elution A–B (95 : 5, v/v), 1–3 min (30 : 70 v/v), 3–12 min (5 : 95 v/v), and kept at 95% B for 16 min at a flow rate of 0.2 mL min⁻¹. The injection volume was 3 μL. High-resolution mass spectrometry (HRMS) analysis was carried out in a Q-TOF mass spectrometer with an electrospray ionization interface. The capillary voltage was operated in positive and negative ionization modes, set at 4500 V, using sodium formate (10 μM) as calibrant. The dry gas parameters were set to 8 L min⁻¹ at 200 °C with a nebulization gas pressure of 4 bar. Collision-induced dissociation (CID) fragmentation was performed using argon (Ar) collision gas and collision energy of 20–45 eV. Spectra data of the investigated compounds were collected from m/z 50–1300 with a resolution of 50,000 at an acquisition rate of 5 spectrums per second. The ions of interest were selected by auto MS/MS scan fragmentation. The data processing software used was DataAnalysis 4.3 (Bruker). To identify the compounds, a library of already identified compounds isolated from the genera Aspergillus and Chromolaena was made, with annotations based on the values of mass error and fragmentation profile/characteristic fragment ions for the compounds. The mass error value was calculated using the following equation: Error (ppm) = [(M_calc − M_exp)/M_calc] × 10⁶, where M_calc is the exact theoretical mass and M_exp is the exact experimental mass obtained. Only molecular formulas with ≤ 10 ppm of error were considered in this study.

Antimicrobial activity

Antimicrobial assays were performed by the microdilution method in sterile flat-bottom microplates, according to CLSI [30–32]. The bacteria Escherichia coli ATCC 25922, Pseudomonas aeruginosa ATCC 27853, Bacillus subtilis ATCC 6623, and Staphylococcus aureus ATCC 25923 were cultivated in Mueller–Hinton broth (MHB) (Difco) at 37 °C. The yeasts Candida albicans ATCC 10231 and C. parapsilosis ATCC 22019 were cultivated in Sabouraud dextrose broth (SDB) (Difco) at 37 °C. And the dermatophyte fungi Trichophyton rubrum ATCC 28289 and Microsporum gypseum ATCC 14683 were cultivated in SDB (Difco) at 28 °C.

Each well contained appropriate test samples, culture medium and approximately 10⁵ cells mL⁻¹ for bacteria, 10³ cells mL⁻¹ for yeasts, and 10⁴ spores mL⁻¹ for dermatophyte fungi. Serial three-fold dilutions of each extract were done in a microdilution plate (96 wells) containing 100 μL of culture medium. Next, the inoculum was added to each well. The microplates were incubated at 37 °C for 24 h for bacteria and 48 h for yeast and at 28 °C for 72 h for dermatophyte fungi. The MIC was defined as the lowest concentration which resulted in inhibition of visual growth. Minimum bactericidal and fungicidal concentrations (MBC and MFC) were determined by subculturing 10 μL of the culture from each negative well and from the positive control.

Anti-M. tuberculosis activity was performed using a resazurin microtiter assay plate (REMA) [33]. Firstly, crude extract PD-S, PD-A, MM-S, and MM-A stock solutions in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and two-fold serial dilutions were prepared in supplemented Middlebrook 7H9 to obtain final concentrations ranging from 31.25 to 1000 μg mL⁻¹. Then, 100 μL of each standardized inoculum was added to each well of the microplate containing the crude extract dilutions. The microplates were covered with their lids and incubated at 35 °C, in a normal atmosphere, for 7 days. After this period, 30 μL of fresh 0.01% resazurin solution (Acros, Morris Plains, NJ, USA) was added to each well and the microplates were re-incubated at 35 °C for an additional 24 h before visual reading was carried out. A color change from blue to pink indicated mycobacterial growth and the MIC was interpreted as the lowest crude extract concentration that prevented the color change from blue to pink. The medium, drug sterility, and bacterial growth with and without 2.5% (v/v) DMSO controls were included in all tests.

Antioxidant activity

The antioxidant activity of the crude extracts PD-S, PD-A, MM-S, and MM-A was evaluated using three different methodologies: DPPH, ABTS, and evaluation of phenolic compounds. To perform the tests, the samples were solubilized at a concentration of 1 mg mL⁻¹ in distilled water with 5% methanol [34].

To evaluate the antioxidant activity of crude extracts of metabolites by the DPPH (2,2-diphenyl-1-picrylhydrazyl) radical method, the methodology of Ma et al. was used [35]; 25 μL of the samples and 2 mL of 6.25 × 10⁻⁵ mol L⁻¹ DPPH solution was added into light-protected glass tubes kept in the dark for 30 min. Absorbance was measured at 517 nm, and the results were expressed as millimoles of Trolox per milligram of sample (mmol Trolox mg sample⁻¹).

The ABTS (2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt) method was performed according to the method described by Rajurkar and Hande [36]. Prior to the test, a solution of the ABTS⁺ radical containing 5 mL of ABTS solution (7 mM) and 88 μL of potassium persulfate (140 mM) was prepared, remaining in reaction for 16 h in the dark. After this period, 3 mL of the ABTS⁺ solution was transferred into tubes protected from the light; 30 μL of the extract from each sample was added and the solution was kept in the dark for 6 min. Absorbance was measured at 734 nm, and the results were expressed as millimoles of Trolox per milligram of sample (mmol Trolox mg sample⁻¹).

The total phenolic content was determined using the Folin–Ciocalteu method as described by Singleton and Rossi, with modifications [37]. One aliquot (125 μL) of each sample was mixed with 125 μL of the Folin–Ciocalteu reagent (1 : 1 v/v distilled water) and 2250 μL of 3.79 M sodium carbonate solution. This mixture was kept in the dark for 30 min. Absorbance was measured at 725 nm, and the results were expressed as micrograms of gallic acid equivalent per milligram of extract (μg GAE mg sample⁻¹).

Cytotoxic and antiviral activity

The cytotoxicity of the crude extracts PD-S, PD-A, MM-S, and MM-A from fungus A. flavus CL7 at concentrations from 31.25 to 1000 μg mL⁻¹ was evaluated on a monolayer of Vero cells (CCL-81, American Type Culture Collection) prepared in 96-well plates at an initial density of 2.5 × 10⁵ cells mL⁻¹ and incubated at 37 °C for 72 h in a humid oven with 5% CO₂. To determine antiviral activity, the cells were pre-infected with HSV-1 TCID₅₀ (KOS strain) and incubated for 1 h at 37 °C for adsorption and viral penetration, and subsequently treated with different concentrations of the crude extracts (3.125–100 μg mL⁻¹) for 72 h at 37 °C. In both cases, cell viability was determined by the MTT method, and the CC₅₀ (cytotoxic concentration for 50% of cells) and EC₅₀ (effective concentration that protects 50% of viral infection) were calculated by linear regression analysis [38]. Acyclovir (Sigma-Aldrich, 99% purity) was used as a positive control for anti-HSV-1 activity. All concentrations were tested in triplicate in at least three independent experiments and the results were expressed as mean ± standard deviation.

Statistical analysis

Statistical analysis was carried out using GraphPad Prism 8.0.1. The differences among treatments in each experiment were compared using one-way analysis of variance (ANOVA) followed by Tukey’s test. In all cases, the threshold for significance was 5%.

Results

Multiphasic identification of CL7

A multiphasic approach was used to identify strain CL7, based on morphology and phylogenetic analysis. The ITS sequence of strain CL7 showed higher values of identity with A. flavus species by BLAST analysis in GenBank (100% identity). Therefore, phylogenetic analysis was conducted with the aim of confirming the identification at species level, considering the A. flavus clade according to Frisvad et al. [12]. The results show that strain CL7 has greater phylogenetic proximity to A. flavus and A. oryzae with 100% Bayesian probability (Fig. 1a).

Fig. 1.

Phylogenetic analysis and macro- and micromorphological structures of Aspergillus flavus CL7. a Cladogram obtained from a Bayesian inference based on the alignment of ITS gene. Bayesian probability was demonstrated at the nodes between each organism. The strain Aspergillus tamarii CBS 104.13 was used as external group; b 7-day-old culture colony on potato dextrose agar medium; c Conidiophores and conidia; d large sclerotia

The macro- and microstructures corroborate this identification (Fig. 1b–d). The colonies are a yellowish-green color, with conidiophores and large sclerotia characteristic of these species. In addition, it may be observed in subsequent sections that the chemical compounds produced by these strains match the literature data. Thus, to the best of our knowledge, this is the first time that A. flavus is described as an endophyte from the Eupatorieae tribe.

Culture conditions

Observing the macroscopic characteristics of A. flavus (CL7) cultured in different conditions (Fig. 2), it was possible to notice the formation of a mycelium with an intense yellow color, and the presence of a large amount of exudate when the PD-S methodology was used (Fig. 2a). On the other hand, when A. flavus CL7 was cultivated in MM using the stationary method (MM-S), the mycelium of the fungus started to present a predominantly dark green color, and the amount of exudate on the surface was reduced (Fig. 2c). In the agitated cultures, with both PD and MM media, the fungus developed in the form of pellets, very similarly in both media (Fig. 2b and d).

Fig. 2.

Micrograph of the mycelium of the endophytic fungus Aspergillus flavus CL7 under different conditions after 21 days of cultivation. a) Potato Dextrose Stationary (PD-S); b) Potato Dextrose Agitated (PD-A); c) Minimum Medium Stationary (MM-S); d) Minimum Medium Agitated (MM-A)

The greatest dry biomass production was obtained when MM-S conditions were used (3.407 ± 0.088 g), followed by PD-A and PD-S that had statistically the same production, 2.601 ± 0.186 g, and 2.589 ± 0.035 g, respectively. On the other hand, the highest production of total metabolites dissolved in ethyl acetate was obtained using PD-S methodology, 163.8 ± 20.7 g of metabolite extract, and only 11.8 ± 2.0 g was obtained using MM-S (Table S1).

Chemical constituents

Chemical investigation of the endophytic fungus A. flavus CL7, associated with the leaves of C. laevigata, yielded three compounds, identified by comparison of their spectroscopic 1D and 2D NMR data with those reported in the literature.

Compound 1 was identified as kojic acid [39], and it was obtained from the culture of A. flavus CL7 grown on PD medium using the stationary and agitated methods (Table S2; Figs. S1–S5). Compound 2 was isolated from the A. flavus CL7 culture grown on PD medium under agitation and basal MM using the stationary method, and it was identified as cyclopiazonic acid (2) (Table S3; Figs. S6–S17) [40]. Compound 3, 20,25-dihydroxyaflavinine, was isolated only from culture on basal MM using the stationary method (Table S4; Figs. S18–S32) [41].

To provide comprehensive coverage of the A. flavus metabolome, fungal extracts PD-S, PD-A, MM-S, and MM-A were analyzed in both positive and negative ion modes by UHPLC-HRMS/MS. This strategy allowed the identification of 21 compounds (Table 1; Table S5).

Table 1.

Specialized metabolites identified of Aspergillus flavus CL7 extracts by UHPLC-HRMS/MS

Compounds	Extracts
	PD-S	PD-A	MM-S	MM-A
α-Cyclopiazonic acid (2)	+	+	+	+
2-Oxo-cyclopiazonic acid	+	+	+	+
Kojic acid (1)	+	+	+
Phomaligin A	+	+	+
Asparasone A	+		+	+
Aspergillic acid	+		+	+
Aspyrone	+	+
20,25-dihydroxyaflavinine (3)	+		+
Itaconic acid	+		+
Paspaline	+		+
Hydroxysydonic acid		+		+
Terrein		+		+
14-Deacetyl parasiticolide A	+
Dideacetyl parasiticolide A	+
Spinulosin	+
Citreoisocoumarin		+
Gregatin B		+
Podophyllotoxin		+
Flufuran			+
Speradine F			+
α-Cyclopiazonic acid imine				+

The identification of the compounds was done by comparison of their exact masses and fragmentation patterners to reported literature (see Online Resource 1, Table S5)

PD-S potato dextrose stationary, PD-A potato dextrose agitated, MM-S minimum medium stationary, MM-A minimum medium agitated

Antimicrobial assay

The antimicrobial activity of crude extracts obtained from the different cultures of A. flavus CL7 was evaluated against four pathogenic bacterial strains, two Gram-negative (P. aeruginosa and E. coli) and two Gram-positive (B. subtilis and S. aureus), and four fungal strains (Candida albicans, C. parapsilosis, T. rubrum, and M. gypseum).

The crude extracts were inactive against the Gram-negative bacteria; however, they were active against the Gram-positive bacteria, with the exception of PD-S which was inactive against all microorganisms evaluated. The best results were observed for the extract obtained using MM against the dermatophyte T. rubrum (Table 2).

Table 2.

Antimicrobial activity of Aspergillus flavus CL7 extracts (MIC, MBC, or MFC)

Microorganisms	MIC/MBC or MFC (μg mL−1)
	Extracts				Ref. antibiotics
	PD-S	PD-A	MM-S	MM-A
Pseudomonas aeruginosa	> 1000	> 1000	> 1000	> 1000	3.125a
Escherichia coli	> 1000	> 1000	> 1000	> 1000	1a
Bacillus subtilis	1000/> 1000	125/500	250/500	125/250	0.2b
Staphylococcus aureus	1000/> 1000	250/500	500/1000	250/> 1000	0.02c
Candida albicans	1000/> 1000	250/500	> 1000	250/> 1000	3.9d
Candida parapsilosis	> 1000	500/> 1000	> 1000	500/> 1000	3.9d
Trichophyton rubrum	> 1000	500/> 1000	31.2/> 1000	31.2/1000	2.5d
Microsporum gypseum	> 1000	125/125	125/1000	250/1000	2.5d
Mycobacterium tuberculosis	250	125	62.5	125	0.125e

The highest concentration tested was 1000 μg mL⁻¹. Values expressed as > 1000 μg mL⁻¹ indicate that there was no inhibition of growth of the microorganisms at the concentrations tested. Tests were performed in triplicate

Data from three independent experiments

PD-S potato dextrose stationary, PD-A potato dextrose agitated, MM-S minimum medium stationary, MM-A minimum medium agitated

^aTetracycline

^bVancomycin

^cPenicillin

^dNystatin

^eRifampicin

The crude extracts were also evaluated against M. tuberculosis H37Rv, using the REMA technique. As observed for the other microorganisms, PD-S was the least active extract, and the best result was observed for MM-S, with an MIC of 62.5 μg mL⁻¹.

Antioxidant assay

The antioxidant potential of A. flavus CL7 extracts, obtained in different culture media, was determined using DPPH and ABTS methods, together with the total phenolic content (TPC) (Table 3). Crude extracts PD-S (93.85 μg GAE mg sample⁻¹) and PD-A (84.22 μg GAE mg sample⁻¹) presented the highest levels of phenols and, consequently, they showed potent antioxidant activity; PD-S exhibited higher scavenging potential for the ABTS method (2.29 mmol Trolox mg sample⁻¹; 98.97%) and PD-A exhibited higher scavenging potential for the DPPH method (0.23 mmol Trolox mg sample⁻¹; 8.85%).

Table 3.

Antioxidant activity by DPPH, ABTS, and total phenolic compounds of Aspergillus flavus CL7 extracts

Extracts	DPPH		ABTS		Total phenolic compounds
	mmol trolox/mg	% activity	mmol trolox/mg	% activity	μg GAE/mg	% activity
PD-S	0.07 ± 0.00a	2.16 ± 0.10a	2.29 ± 0.01a	98.97 ± 0.22a	93.85 ± 16.27a	13.71 ± 2.38a
PD-A	0.23 ± 0.01b	8.85 ± 0.34b	0.95 ± 0.03b	41.64 ± 1.24b	84.22 ± 17.32a	12.71 ± 2.61ac
MM-S	0.21 ± 0.01b	8.15 ± 0.49b	0.67 ± 0.07c	29.57 ± 2.92c	49.78 ± 6.67b	9.09 ± 1.22ac
MM-A	0.05 ± 0.01a	1.29 ± 0.23c	0.41 ± 0.04d	18.28 ± 1.90d	8.67 ± 2.22c	3.46 ± 0.89bd

Data expressed as mean ± standard deviation of three independent experiments. In the same raw, the values marked with the same lowercase letter are similar (p > 0.05), whereas those marked with different ones are significantly different (p < 0.05)

PD-S potato dextrose stationary, PD-A potato dextrose agitated, MM-S minimum medium stationary, MM-A minimum medium agitated, GAE gallic acid equivalent

Cytotoxic and antiviral assays

The evaluation of cytotoxicity showed that among all crude extracts tested, PD-S was the least toxic to the cells. However, all crude extracts were inactive, at the concentrations evaluated, for cells infected with HSV-1 TCID₅₀ (KOS strain) (Table 4).

Table 4.

Cytotoxicity and antiviral activity of Aspergillus flavus CL7 extracts by the MTT method

Extracts	CC50 (μg mL−1)	EC50 (μg mL−1)
PD-S	> 1000	> 100
PD-A	60.17 ± 7.02a	> 100
MM-S	40.17 ± 2.75b	> 100
MM-A	21.08 ± 1.38c	> 100
Acyclovir	> 1000	0.1115 ± 0.005

The highest concentration tested was 1000 μg mL⁻¹ for cytotoxicity assay and 100 μg mL⁻¹ for antiviral activity. Tests were performed in triplicate

Data expressed as mean ± standard deviation of three independent experiments. In the same raw, the values marked with the same lowercase letter are similar (p > 0.05), whereas those marked with different ones are significant

PD-S potato dextrose stationary, PD-A potato dextrose agitated, MM-S minimum medium stationary, MM-A minimum medium agitated, CC₅₀ cytotoxic concentration for 50% of cells, EC₅₀ effective concentration that protects 50% of viral infection

Discussion

As mentioned before, our research group has been studying species from the Eupatorieae tribe, especially from the genus Eupatorium sensu lato; based on the fact that there are few reports on endophytic fungi in species of this genus, Chromolaena laevigata (basionym: Eupatorium laevigatum) was selected for this study [2–5].

An endophytic fungus from the genus Phomopsis has been described from Chromolaena arnottiana (basionym: Eupatorium arnottianum) [42]. Six morphologically different endophytes have been found to inhabit the leaves of Ageratina adenophora (basionym: Eupatorium adenophorum); four morphotypes are close to Alternaria, Cladosporium, Pestalotiopsis, and Didymella, while two morphotypes are close to unidentified fungi [43]. Three endophytic fungi have been identified from Eupatorium buniifolium; they were identified as Fusarium solani, Alternaria alternata, and Neofusicoccum sp. [44].

The endophytic fungus isolated from Chromolaena laevigata in this work was identified as Aspergillus flavus. Endophytic A. flavus has been previously isolated from many plant species, for example, from Paspalum maritimum Trin. [45], Aegle marmelos [46, 47], Cephalotaxus fortunei [48], Tylophora ovata [49], Catharanthus roseus, Annona squamosa, and Curcuma xanthorrhiza [50]. So, this is the second study about endophytes from Chromolaena species, and the first time that endophytic Aspergillus has been described from Eupatorium sensu lato.

Characterization of the fungus as Aspergillus flavus was based on BLAST analysis in GenBank; however, the characteristics of the macro- and microstructure were essential for confirming the species. Many studies have indicated that A. oryzae is the domesticated form of A. flavus and can be distinguished from the wild type due to many factors, among them the presence of larger and more smooth conidia, more floccose aerial mycelia and weaker sporulation, the absence of sclerotia, no production of aspergillic acid, and a lack of aflatoxin production [12, 51–53]. According to Frisvad et al., A. oryzae is isolated from fermented products, while strains from other environments, even if they are not aflatoxin producers, are identified as A. flavus [12]. Therefore, based on the phylogenetic analysis performed on the type of host, on the macroscopic characteristics of the culture together with the presence of large sclerotia and the profile of identified secondary metabolites, strain CL7 was confirmed as being the species A. flavus.

Macroscopic differences in cultures were observed when the nutrients in the media were changed, and also when the cultures were agitated (Fig. 2). A culture medium must have a source of carbon and nitrogen, in addition to mineral salts and some other nutrients, and the constitution of the medium directly influences the growth of the mycelium and the production of bioactive metabolites [54–56]. When PD was used as the medium, greater production of spores was observed compared to culture in MM. The highest dry biomass was observed for the MM-S condition; however, it provided the smallest amount of metabolite extract. Although cultivation in PD produced a similar dry biomass using agitated and stationary conditions, production of the metabolite extract was 240% higher when the stationary method was used.

The effect of culture conditions on the production of specialized metabolites by the endophytic fungus A. flavus CL7 was also investigated in this work. First, the metabolite extracts were submitted to a CC, and three metabolites were isolated. Kojic acid (1) was the major compound found in PD-S extract, and it was identified directly in this extract, without purification. This compound is widely described from Aspergillus genus, having already been found for A. aflatoxiformans, A. alliaceus, A. arachidicola, A. aspearensis, A. austwickii, A. bombycis, A. bertholletius, A. caelatus, A. cerealis, A. flavus, A. flocculus, A. hancockii, A. lanosus, A. leporis, A. luteovirescens, A. minisclerotigenes, A. mottae, A. neoalliaceus, A. nomius, A. novoparasiticus, A. oryzae, A. parasiticus, A. parviscierotigenus, A. pseudocaelatus, A. pseudonomius, A. pseudotamarii, A. sergii, A. sojae, A. subflavus, A. tamarii, A. transmontanensis, and A. vandermerwei [12, 57, 58]. Kojic acid has also been identified as a major compound from A. flavus associated with Catharanthus roseus, Annona squamosa, and Curcuma xanthorrhiza [50].

Cyclopiazonic acid (2), an indole tetramic acid, was described for the first time from A. flavus in 1977 [59]. However, this metabolite is briefly described from endophytic Aspergillus, being reported from the marine-derived fungus A. flavus [60–62]. For plant-derived fungus, this metabolite was described from A. flavus isolated from Triticum aestivum, and from A. versicolor isolated from Paris polyphylla [63, 64].

Compound 3, 20,25-dihydroxyaflavinine, is frequently described from the sclerotia of A. flavus; however, this is the first time that this compound has been isolated from an endophytic A. flavus strain [41, 65–68].

Other compounds were putatively identified using the dereplication technique by UHPLC-HRMS/MS. In PD, 18 compounds were identified, and 15 compounds were detected in MM. Only cyclopiazonic acid and its derivative 2-oxo-cyclopiazonic acid were produced in all culture media evaluated. Comparing the metabolites produced when the culture of A. flavus was grown in PD using the stationary method, with those produced in the same medium but under agitation, considerable variation was observed: only five substances were common between them (aspyrone, α-cyclopiazonic acid, 2-oxo-cyclopiazonic acid, kojic acid, and phomaligin A). For culture in MM, the number was even lower: only four compounds were common (asparasone A, aspergillic acid, α-cyclopiazonic acid, and 2-oxo-cyclopiazonic acid).

The principal variation was observed when the agitation method was applied. It was also possible to verify that some compounds were produced in both culture media; however, a variation in the metabolites produced was observed when the culture was carried out under agitation or without agitation. Itaconic acid, paspaline, and 20,25-dihydroxyaflavinine were detected from both culture media when they were kept without agitation. On the other hand, hydroxysydonic acid and terrein were identified when the media were kept under agitation. These results corroborate the variable effects of medium conditions on the production of specialized fungal metabolites [18, 19].

The major compounds produced by A. flavus are mycotoxins, especially aflatoxins and cyclopiazonic acid derivatives, and it has been demonstrated that the ability to produce these toxins can vary considerably [69–71]. It was previously reported that the culture of A. flavus in PD enhances the formation of aflatoxins as compared to CZPY medium, and it was also shown that cyclopiazonic acid is produced in liquid static culture in much greater concentrations than aflatoxins [69–72]. However, in our work, only cyclopiazonic acid was isolated, even using PD under agitation, and aflatoxins were not found.

The results from the chemical investigation of A. flavus have revealed constituents that agree with the identification, classification, and chemotaxonomy of this fungal strain. However, no aflatoxins were identified, which are chemomarkers of this species; so, to provide comprehensive coverage of the A. flavus metabolome, all extracts were submitted for analysis by UHPLC-HRMS/MS. Table 1 shows the secondary metabolites identified, which corroborate the classification of our strain as A. flavus. Only compounds already described from this species were identified, except for dideacetyl parasiticolide A, which is described here for the first time in this species [12, 73–76]. However, aflatoxins were not detected in these extracts which were obtained by different methods. The absence of these chemomarkers allows us to indicate strain CL7 as a non-aflatoxigenic A. flavus.

Regarding the possible applications of secondary metabolite extracts obtained by different methods of culturing the endophyte CL7, the antimicrobial activity of extracts was evaluated. Despite the extracts showing higher MIC than the reference antibiotics, some of them showed promising values. According to Holetz et al., extracts with an MIC below 100 μg mL⁻¹ present good antimicrobial activity; between 100 and 500 μg mL⁻¹ indicates moderate antimicrobial activity and from 500 to 1000 μg mL⁻¹ indicates weak antimicrobial activity; above 1000 μg mL⁻¹, the extract may be considered inactive [77]. In this context, PD-S extract did not present antimicrobial activity against the microorganisms tested, since the MIC was greater than 1000 μg mL⁻¹. On the other hand, the PD-A extract, which used the same culture medium but was cultured on a shaker, showed an improvement in antimicrobial activity, with moderate activity against C. albicans, T. rubrum, S. aureus, B. subtilis, and M. gypseum. In addition, for the dermatophyte fungus M. gypseum, it was observed that the microorganism died at the same MIC (MIC = MFC). This activity may be attributed to the podophyllotoxin detected in this extract, which is known to have antimicrobial properties [78].

The best results were observed for the extract obtained using MM against the dermatophyte T. rubrum, with an MIC of 31.2 μg mL⁻¹. This may be related to the presence of aspergillic acid in this extract, which has been described as an antibiotic and antifungal agent that is derived from certain A. flavus strains [79]. Although this metabolite was also detected in PD-S, and this extract did not show antimicrobial activity, this could be related to the concentration of aspergillic acid in the extract as the major compound in the PD-S extract is kojic acid.

In evaluating anti-tuberculosis activity, Tosun et al. considered inactive those extracts that could not prevent the growth of M. tuberculosis up to a concentration of 200 μg mL⁻¹ and, according to Gu et al., an MIC of ≤ 128 μg mL⁻¹ is defined as active against M. tuberculosis. By these measures, only PD-S extract showed no significant activity, and MM-S extract was the most potent inhibitor of M. tuberculosis proliferation [80, 81].

Among the extracts evaluated, PD-S was the least toxic against Vero cells, since its CC₅₀ was higher than the highest concentration tested (1000 μg mL⁻¹). On the other hand, the extracts PD-A, MM-S, and MM-A were highly cytotoxic to Vero cells, with CC₅₀ of 60.17, 40.17, and 21.08 μg mL⁻¹, respectively. The high toxicity of extracts grown in MM and those grown on PD under shaking conditions is probably related to the compounds isolated from these fractions, such as α-cyclopiazonic acid and 20,25-dihydroxyaflavinine [82]. Regarding antiviral activity, the extracts did not show a significant inhibition capacity for HSV-1 virus (KOS strain), since the EC₅₀ were all higher than 100 μg mL⁻¹, whereas for the standard drug (acyclovir), the value determined was 0.11 μg mL⁻¹.

The extract of the metabolites of the endophyte A. flavus CL7 obtained by steady-state culture in PD presented high antioxidant potential by the ABTS method, with almost 99% antioxidant activity. This result was probably due to the major presence of kojic acid (1), an important antioxidant used by the cosmetics industry [83].

In summary, we can highlight the identification of a non-aflatoxigenic strain of A. flavus from a new host C. laevigata. Dereplication analysis by UHPLC-HRMS/MS enabled the identification of 21 compounds, one of which, dideacetyl parasiticolide A, is being reported for the first time in this species. This work also corroborates the direct influence of culture medium on the production of specialized metabolites. In addition, the biological activity of the different extracts was evaluated; the extract obtained from stationary culture on MM proved to be the most active in general, mainly due to the MIC obtained against T. rubrum and M. tuberculosis, two pathogenic microorganisms that affect the population and that are becoming more and more resistant to conventional therapies. Thus, we describe a strain of A. flavus with biotechnological potential for use as a biocontrol tool, as a source of biomolecules active against other pathogenic microorganisms and as a promising producer of kojic acid, an important metabolite used in the food and beauty industries.

Supplementary information

NMR and UHPLC-HRMS/MS data of this work can be found in online version of the paper.

Author contribution

RBB, JAP, and DCB conceived and designed research. RBB, JAMO, DIB, ADP, and JCP conducted experiments. MRBC, MHS, MRPC, and JAP contributed new reagents or analytical tools. EBZ, EHE, JEM, RFC, BDPF, and TUN realized biological activities. RBB, JAMO, and JCP analyzed data. RBB, JAMO, DIB, ADP, JCP, and DCB wrote the manuscript. All authors read and approved the manuscript.

Funding

This work was financially sponsored by Concelho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)-Finance Code 001, Fundação Araucária and Instituto Nacional de Ciência e Tecnologia Biodiversidade e Produtos Naturais–INCTBioNat (CNPq 465637/2014-0).

Declarations

Conflict of interest

The authors declare no competing interests.