Diversity in Secondary Metabolites Including Mycotoxins from Strains of Aspergillus Section Nigri Isolated from Raw Cashew Nuts from Benin, West Africa

Abstract

In a previous study, raw cashew kernels were assayed for the fungal contamination focusing on strains belonging to the genus Aspergillus and on aflatoxins producers. These samples showed high contamination with Aspergillus section Nigri species and absence of aflatoxins. To investigate the diversity of secondary metabolites, including mycotoxins, the species of A. section Nigri may produce and thus threaten to contaminate the raw cashew kernels, 150 strains were isolated from cashew samples and assayed for their production of secondary metabolites using liquid chromatography high resolution mass spectrometry (LC-HRMS). Seven species of black Aspergilli were isolated based on morphological and chemical identification: A. tubingensis (44%), A. niger (32%), A. brasiliensis (10%), A. carbonarius (8.7%), A. luchuensis (2.7%), A. aculeatus (2%) and A. aculeatinus (0.7%). From these, 45 metabolites and their isomers were identified. Aurasperone and pyranonigrin A, produced by all species excluding A. aculeatus and A. aculeatinus, were most prevalent and were encountered in 146 (97.3%) and 145 (95.7%) isolates, respectively. Three mycotoxins groups were detected: fumonisins (B₂ and B₄) (2.7%) ochratoxin A (13.3%), and secalonic acids (2%), indicating that these mycotoxins could occur in raw cashew nuts. Thirty strains of black Aspergilli were randomly sampled for verification of species identity based on sequences of β-tubulin and calmodulin genes. Among them, 27 isolates were positive to the primers used and 11 were identified as A. niger, 7 as A. tubingensis, 6 as A. carbonarius, 2 as A. luchuensis and 1 as A. welwitschiae confirming the species names as based on morphology and chemical features. These strains clustered in 5 clades in A. section Nigri. Chemical profile clustering also showed also 5 groups confirming the species specific metabolites production.

About the Authors:

Yendouban Lamboni

Affiliations International Institute of Tropical Agriculture, Cotonou, Benin, Department of Biotechnology and Bioengineering, Technical University of Denmark, Lyngby, Denmark, Food Quality and Design Group, Wageningen University, Wageningen, The Netherlands, Laboratory of Food Microbiology, Wageningen University, Wageningen, The Netherlands

Kristian F. Nielsen

* E-mail: [email protected]

Affiliation: Department of Biotechnology and Bioengineering, Technical University of Denmark, Lyngby, Denmark

Anita R. Linnemann

Affiliation: Food Quality and Design Group, Wageningen University, Wageningen, The Netherlands

Yüksel Gezgin

Affiliations Department of Biotechnology and Bioengineering, Technical University of Denmark, Lyngby, Denmark, Department of Bioengineering, Ege University, Izmir, Turkey

Kerstin Hell

Affiliation: International Institute of Tropical Agriculture, Cotonou, Benin

Martinus J. R. Nout

Affiliation: Laboratory of Food Microbiology, Wageningen University, Wageningen, The Netherlands

Eddy J. Smid

Affiliation: Laboratory of Food Microbiology, Wageningen University, Wageningen, The Netherlands

Manuele Tamo

Affiliation: International Institute of Tropical Agriculture, Cotonou, Benin

Martinus A. J. S. van Boekel

Affiliation: Food Quality and Design Group, Wageningen University, Wageningen, The Netherlands

Jakob Blæsbjerg Hoof

Affiliation: Department of Biotechnology and Bioengineering, Technical University of Denmark, Lyngby, Denmark

Jens Christian Frisvad

Affiliation: Department of Biotechnology and Bioengineering, Technical University of Denmark, Lyngby, Denmark

Introduction

Aspergillus section Nigri also known as black Aspergilli are among the most common fungi responsible for food spoilage and bio-deterioration of other materials [1], also causing substantial impact on food safety due to their mycotoxins production. They are known to produce the mycotoxins ochratoxin A [2], fumonisins B₂, B₄ and B₆ [3,4] as well as numerous other compounds with poorly investigated activities [4,5]. On the other hand, black Aspergilli are also reported to be of biotechnological importance due to their use in the fermentation industry, for example in their ability to produce hydrolytic enzymes and organic acids [6]. Aspergillus luchuensis is reported to be extensively used in Asia for koji production [7]. Moreover, many A. niger processes have been classified as GRAS (Generally Recognized As Safe) by the Food and Drug Administration of the US government [1] despite the ability of A. niger to produce ochratoxin A and fumonisins. However these mycotoxins seem not to be produced under submerged conditions [8].

Black Aspergilli are one of the most complicated species complexes to classify and identify, and the taxonomy of strains in the A. section Nigri has been studied and debated for decades. In 1934, Mosseray described 35 species of black Aspergilli [9]. Later, that number was reduced to 12 species by Raper and Fennell [10]. In 1984, based on morphological features, Al-Musallam [11] revised the taxonomy of niger group to 7 species: A. japonicus, A. carbonarius, A. ellipticus, A. helicothrix, A. heteromorphus, A. foetidus and A. niger. While, in 2009, Nielsen et al. [4] reported 18 species in the black Aspergilli group with A. niger, A. tubingensis, A. brasiliensis, A. acidus, A. carbonarius and A. ibericus as the most common ones. In 2012, Jurjević et al. [12] added A. floridensis and A. trinidadensis as new species to the A. section Nigri. Recently, Varga et al. [13] revisiting the species in A. section Nigri, added 4 other new species and concluded that the black Aspergilli group includes 26 taxa. Therefore, a polyphasic taxonomic approach [14], has been used to accurately identify black Aspergilli at species level. These include morphological, physiological and biochemical characteristics of the isolates, e.g. using high performance liquid chromatography mass spectrometry (HPLC-MS) as well as DNA sequence analysis. The latter is presently based on the use of β-tubulin [15] and calmodulin [16] genes [7,13], as the ITS regions does not provide sufficient resolution [17].

Nuts are nutritious human foodstuffs [18] because of their high content of protein, carbohydrates, vitamins, essential minerals and especially unsaturated fatty acids. Nuts are consumed in both developing and developed countries by all age groups and across all social strata [19]. Among tree nuts, cashew nuts are known for their high minerals content (e.g. copper, iron and phosphate) and vitamins (e.g. thiamine, vitamin E and pyridoxine) [18]. In tropical regions of Africa, 48% of the world’s cashew nuts are produced, making them crops of high economic importance [20], and in 2011, cashew nut export contributed about 150 million US dollars to the gross domestic product of Benin [21] accounting for 8% of the national export revenues.

Since stored nuts generally have a low water activity, their spoilage association consists mainly of fungi and members of A. section Nigri have been reported to contaminate cashew nuts [22].

Although strains of A. section Nigri have been found on cashew nuts, very little is known about the risk of mycotoxins contamination from black Aspergilli on cashew kernels. Therefore, the objective of this study was to screen the mycotoxins and other metabolites that can be produced by A. section Nigri strains isolated from raw cashew kernels and, based on their metabolite production, to determine which species are prevalent on kernels from Benin. To accomplish these goals, isolated black Aspergilli were assayed for their mycotoxins and other secondary metabolites diversity by LC-HRMS on a LC-time-of-flight mass spectrometry instrument (LC-TOFMS) and representative strains were identified using molecular methods.

Materials and Methods

Chemicals and reagents

All solvents used for chemical analysis were LC-MS grade. Methanol, acetonitrile, 2-propanol, formic acid were LC-MS grade, while ethyl-acetate and dichloromethane were HPLC grade. All were purchased from Sigma Aldrich (Fluka Analytical, Denmark). Purified water was obtained by using a Milli-Q water purification system (Millipore Synergy® UV, Molsheim, France).

Fungal isolates and growth conditions

One hundred and fifty strains belonging to Aspergillus section Nigri isolated from cashew nut samples from northern Benin were used in this study. These strains were obtained from Lamboni et al. [23]. Cashew nuts were sampled in the main cashew production area covering two agro-ecological zones lying within latitudes 8°1’ and 12°3’ N and longitudes 0°8’ and 3°8’ E, with an unimodal rainfall distribution averaging 900 mm to 1000 mm annually and maximum temperatures varying from 28°C to 40°C. Based on the agreement between the International Institute of Tropical Agriculture located in Benin and the Beninese Government, any other specific permissions were not required for sampling cashew nut within the study area. In total, 70 nuts samples were randomly selected in fourteen different locations.

After collection, the cashew shell was cut and the kernel (2 cotyledons) aseptically extracted, plated on dichloran 18% glycerol agar (DG18, Oxoid, Basingstoke, Hampshire, UK) [24] and incubated at 25°C in the dark for 7 days. Four cotyledons were plated per Petri dish, either in five replicates for surface sterilization (SS) (0.4% aqueous solution of sodium hypochlorite) or in two replicates for direct plating (DP), giving a total of 1960 cultured cotyledons. Both culturing methods were used to enable the growth of conidia present in the inner and the outer part of the cotyledons. According to taxonomic schemes and illustrations in Samson et al. [25], colonies belonging to A. section Nigri were first isolated on Czapek yeast autolysate agar (CYA) [24] and later 3 point inoculated on Yeast extract sucrose agar (YES) [24]. The plates were incubated at 25°C in the dark for 5 days. From the centre of fugal colonies, three 5-mm agar plugs were taken with an aseptic steel drill and pooled together into the same vial and stored at 4°C for further extraction.

Plug extraction

A one step extraction method was used by adding 0.5 ml of a mixture of ethyl acetate-dichloromethane-methanol (3:2:1, v/v/v) with 1% (v/v) formic acid to the vials containing the agar plugs. The plugs were then extracted in an ultrasonic bath for 60 min. The supernatant was transferred to a new vial, evaporated to complete dryness using N₂ flow, and re-dissolved in 500 μl of methanol assisted by ultrasonication for 20 min, and the aliquots filtered into an HPLC vial using a 0.45 μm polytetrafluoroethene (PTFE) filter.

UHPLC-QTOF-MS analysis

Analyses were performed using ultra-high-performance liquid chromatography (UHPLC) with diode array detector and maXis 3G QTOF mass spectrometer (MS) (Bruker Daltonics, Bremen, Germany) equipped with an electrospray source (ESI) and connected to an Ultimate 3000 UHPLC system (Dionex, Sunnyvale, USA) equipped with a Kinetex 2.6-μm C₁₈, 100 mm × 2.1 mm column (Phenomenex, Torrance, CA) [26]. A linear water-acetonitrile gradient was used (buffered with 20 mM formic acid) starting from 10% (v/v) acetonitrile and increased to 100% in 10 min, maintaining for 3 min before returning to the starting conditions. MS was performed in ESI⁺, the scan range m/z 100–1000, with a mass accuracy < 1.5 ppm [26]. UV/VIS spectra were collected at wavelengths from 200 to 700 nm. Data processing was performed using DataAnalysis 4.0 and Target Analysis 1.2 (Bruker Daltonics) by the aggressive dereplication approach [26], using a database of 495 known and putative compounds from black Aspergilli, tentatively identifying them based on accurate mass (deviation < 1.5 ppm) and isotopic pattern (isotope fit < 50) [26]. For saturated peaks (>10⁶ counts/sec) a manual verification of the accurate mass was made in the front and the tail of the peak. A further database of 1500 reference standards, tentatively identified compounds were also used along with a small 50 compounds database of peaks observed in sample blanks. All major peaks (observed in the BPC chromatograms) not tentatively identified by the approach were added to the search list as unknown compounds for mapping.

DNA sequencing of Aspergillus section Nigri strains

Aspergillus isolates and growth conditions.

Thirty strains of Aspergillus section Nigri isolated from raw cashew nuts were randomly selected for diagnostic PCR and sequencing. The strains were 3 point inoculated on separate Petri dishes containing Czapek yeast autolysate (CYA) agar and incubated in micro perforated plastic bags at 25°C for 7 days in the dark, to ensure extensive conidiation of the colonies. From these cultivations on solid media, we prepared stock suspensions for further inoculations and harvested conidia to make suspensions in 5 ml glass tubes containing autoclaved milli-Q water supplemented with 0.05% Tween 80. Conidia were inoculated at 3 points equidistant from the centre, on CYA and incubated in micro perforated plastic bags in the dark at 25°C for 3 days to favour mycelial growth and reduce the total conidiation as this would inhibit tissue-PCR. The Petri dishes were kept at 4°C for sampling.

Tissue-PCR for molecular identification of fungal isolates.

Tissue PCR alleviates the need for genomic DNA extraction, as fungal mycelial tissue was the direct source for template DNA in PCR reactions amplifying partial genes encoding calmodulin and β-tubulin. PCR tubes containing a total volume of 40 μl had the following components mixed in milli-Q H₂O; 1X Phire PCR buffer (ThermoFisher Scientific, USA), 200 μM dNTP mix (Invitrogen, Merelbeke, Belgium), 0.25 μM forward and reverse primers and 0.7 U Pfu X7 polymerase [27]. We based molecular identification of the thirty strains on the amplification of two partial genes encoding β-tubulin and calmodulin. The selected β-tubulin primers were T10-F-ACGATAGGTTCACCTCCAGAC [28], and Bt2b-R-ACCCTCAGTGTAGTGACCCTTGGC [15], and for calmodulin Cmd5-F CCGAGTACAAGGARGCCTTC and Cmd6-R CCGATRGAGGTCATRACGTGG [29]. A sterile pipette tip was used to streak 1 to 3 mm of peripheral mycelium in two replicates. Distribution of the fungal tissue on the pipette tip to PCR tubes resulted in two tubes with different amounts of biomass and thereby template DNA for PCR.

The amplification was performed in Agilent SureCycler 8800 Thermal Cycler (Agilent Technologies Inc., Santa Clara CA, USA). The amplification process consisted of an initial denaturation step of 30 min at 98°C to release template DNA from fungal debris, followed by 35 cycles of touch-down PCR with 10 s at 98°C (denaturation), 30 s at 61–52°C (primer annealing) and 1 min at 72°C (extension), and a final extension step of 5 min at 72°C. We verified purity of the amplification products by agarose gel electrophoresis in 1% TAE buffer (tris-acetate-EDTA (Ethylenediaminetetraacetic acid)) stained by SYBR Safe DNA Gel Stain (ThermoFisher Scientific, USA) and visualized by UV-light. If more than one product was observed after electrophoresis, PCR products were purified prior to DNA sequencing using the GFX PCR DNA and Gel Band Purification Kit (GE Healthcare UK limited, Buckinghamshire, UK). PCR products were sequenced by GATC Biotech (Constance, Germany).

Phylogenetic analysis of sequence data.

The identity of the β-tubulin and calmodulin gene sequences was determined using Basic Local Alignment Search Tool for nucleotide (BLASTN) algorithm in the National Centre for Biotechnology Information (NCBI) GenBank database (http://blast.ncbi.nlm.nih.gov/Blast.cgi). They were then transformed into multi FASTA format using DNA Baser software. Phylogenetic analyses and molecular evolutionary were conducted using MEGA (Molecular Evolutionary Genetics Analysis) version 6.0 [30]. Sequences were pairwise aligned by Clustal W method [31] and trimmed both sides up to the same nucleotide position. Phylogenetic trees were prepared using the maximum likelihood method. Evolutionary distances were calculated by using the Jukes-Cantor [32] model embedded in the MEGA package. Bootstrap values were calculated from 1000 replications after complete deletion of all positions containing gaps or missing data. To compare with cluster output of DNA Baser, secondary metabolites of strains were grouped using MultiExperiment Viewer (MeV v4.2).

Results

Mycotoxins and other metabolites diversity from Nigri group on cashew nuts

The mycotoxins and other metabolites produced by strains of Aspergillus section Nigri on YES agar are presented in Table 1. From the 150 isolates used for metabolites profiling, 66 strains (44%) belonged to A. tubingensis, 48 strains (32%) to A. niger (with a chemical profile similar to A. welwitschiae), 15 strains (10%) to A. brasiliensis, 13 strains (8.7%) to A. carbonarius, 4 strains (2.7%) to A. luchuensis (synonyms to A. kawachi or A. acidus), 3 strains (2%) to A. aculeatus and 1 strain (0.7%) to A. aculeatinus.

[Figure omitted. See PDF.]

Table 1. Mycotoxins and other secondary metabolites of Aspergillus section Nigri isolated from cashew nuts.

https://doi.org/10.1371/journal.pone.0164310.t001

In total, 45 metabolites including their isomers were identified during UHPLC-QTOF-MS analysis within retention times (RT) ranging from 1.56 min (nigragillin) to 10.1 min (aflavinine) (S1 Table). Aurasperone C (positive in 97.3%), aurasperone F (96.7%), pyranonigrin A (96.7%), and fonsecin (96%) were the metabolites identified in most of the strains of A. section Nigri. The metabolites that were rarely produced by strains of A. section Nigri were secalonic acids (2%), tubingensins (2%), antafumicins (2.7%), fumonisins (2.7%), kotanin (4%) and ochratoxin A (5.3%). The detection of orlandin, kotanin and fumonisin B₂, B_4, B₆ was specific for A. niger whereas the presence of antafumicin A and B was specific for A. luchuensis. Secalonic acids were specific for A. aculeatus.

Secalonic acids, atromentin, asperazine and aurasperone C were produced consistently by all the strains in a species. The mycotoxin fumonisin B₂ (2.7%) was detected in strains belonging to A. niger whereas ochratoxin A (13%) and ochratoxin B (5.3%) were produced by strains of both A. niger and A. carbonarius.

For example, in Table 1 with UHPLC-QTOF MS, nigragillin was produced by 129 (86%) of the 150 strains studied. Nigragilin was produced by 95% of the strains of A. tubingensis (64/66), 94% of A. niger (46/48) and 100% of both the strains of A. brasiliensis (15) and A. luchuensis (4).

An example of Base Peak Chromatogram (BPC) of A. niger extract is shown in Fig 1 where several compounds were identified including fumonisin B₂ and fumonisin B₄.

[Figure omitted. See PDF.]

Fig 1. Base Peak Chromatogram (BPC) of Aspergillus niger extract.

The Analysis was done by reversed phase ultra-high-performance liquid chromatography quadrupole time of flight mass spectrometry. Positive electrospray ionization (ESI⁺, m/z 100–1000). A. niger was cultured on yeast extract sucrose agar for 5 days in dark. The BPC showed the production of fumonisin B₂ and fumonisin B₄ and other secondary metabolites.

https://doi.org/10.1371/journal.pone.0164310.g001

Phylogenetic analysis

We examined the genetic relatedness of 30 randomly taken strains of Aspergillus section Nigri using nucleotide sequences of β-tubulin and calmodulin genes. Table 2 summarizes the species names based on their metabolite production which was confirmed by sequencing data. Eleven isolates were identified as A. niger, 7 as A. tubingensis, 6 as A. carbonarius, 2 as A. luchuensis, 2 as A. aculeatus, 1 as A. aculeatinus and 1 as A. welwitschiae. Three of the isolates (2 of A. aculeatus and 1 of A. aculeatinus) did not match with the β-tubulin and calmodulin genes used.

[Figure omitted. See PDF.]

Table 2. Aspergillus section Nigri strains isolated from cashew nuts characterized by sequence analysis.

https://doi.org/10.1371/journal.pone.0164310.t002

The phylogenetic relationship between the isolates of A. section Nigri species was illustrated in the maximum likelihood method analysis. Twenty seven isolates matched either with β-tubulin and/or calmodulin genes during PCR amplification. The 27 sequences were aligned and resulted in the formation of 5 clades (Fig 2). All the A. niger strains (11) clustered together, as did the A. tubingensis strains (7) and A. carbonarius strains (6). A. welwitschiae strain was separated from A. niger sequences with 98% bootstrap value. Two strains of A. luchuensis clustered to form a clade.

[Figure omitted. See PDF.]

Fig 2. Phylogenetic trees based on combined sequences data of β-tubulin and calmodulin of 27 strains of Aspergillus section Nigri, with clustering based on metabolites profile.

Phylogenetic analyses and molecular evolutionary were conducted using MEGA version 6.0. The identity of the gene sequences was determined using BLASTN algorithm. The alignment was performed using the Clustal W program. Nucleotide divergences were estimated according to the Jukes-Cantor model. The numbers above branches are bootstrap values. The evolutionary history was inferred using the maximum likelihood method. The chemical clustering was performed using MultiExperiment Viewer (MeV v4.2).

https://doi.org/10.1371/journal.pone.0164310.g002

The hierarchical clustering based on chemical compounds of the 27 strains, showed 5 major clusters similar to the genetic clades (Fig 2). Some of the strains in A. tubingensis group clustered (A) whereas the others clustered together with strains of the A. luchuensis clade (E). Strains of A. niger clustered in 2 groups (B and C) whereas A. welwitschiae joined strains of A. carbonarius to form cluster D (Fig 2). The hierarchical clustering revealed aflavinin, tubingensin, orlandin, secalonic acids, fumonisins and ochratoxins as key compounds for chemical clustering (S1 Fig).

Discussion

One hundred and fifty strains of Aspergillus section Nigri were used for mycotoxin and other secondary metabolite profiling using ultra-high-performance liquid chromatography. All the 45 chemical compounds identified pertained to black Aspergilli as previously described by Nielsen et al. [4]. Some of these natural products are known to be toxic to human and animals. These where classified in Table 1 as group A and included fumonisins and ochratoxin A, which were reported for Aspergillus niger previously[33]. According to Mogensen et al. [34] and Noonim et al. [35], up to 75% of A. niger isolates produce fumonisins and 41% produce ochratoxins. Also, Massi et al. [36] reported 74% of A. niger to be fumonisin B₂ producers while 32% were ochratoxin A producers. In our results, among the 48 strains of A. niger isolated, only 9% (4 strains) produced fumonisins and 15% (7 strains) produced ochratoxins. The difference could be due to the type of commodities from which the A. niger strains were isolated. The effect of the food matrix was demonstrated by Vaamonde et al. [37] who studied the variability of aflatoxin production by strains of A. section Flavi in peanut, wheat and soybean. The isolates of Mogensen et al. [34] were from raisins and those from Noonim et al. [35] from dried coffee bean samples. The food matrix of raisins (16% of moisture, 25 to 30% of sugars and ˂1% of lipids) is different from that of dried green coffee (12–13% of moisture, ˂1% of sugars and 4 to15% of lipids) and much more different from that of cashew nuts (8 to 9% of moisture, 1 to 8% of sugars and 60 to 64% of lipids). Isolates from Massi et al. [36] were from different food commodities: dried fruits, Brazil nuts, coffee beans, grapes cocoa and onions. In addition, cashew nuts also have a thick shell that constitutes a first barrier to microbial contamination [38]. It is known that cashew shells contain tannins that are able to suppress mycotoxin formation [39] and probably alter the gene expression by the fungi. More so, on Pistachio nuts, Marin et al. [40] noticed only 5% of A. niger to be ochratoxin A positive.

The geographic origin of a strain can reportedly influence its mycotoxin production. Isolates of A. flavus from various geographic regions have revealed differences in the proportions of isolates that produce low, medium and high amount of aflatoxins [41]. This could also apply to fumonisin and ochratoxin production by black Aspergilli. Samson et al. [42] reported ochratoxin production from species of A. section Nigri isolated from different food matrices collected from various regions. Moreover, Perrone et al. [43] reported that 33% of A. niger isolated from grapes in Italy produced ochratoxin A. In our study we did not notice simultaneous production of both fumonisins and ochratoxins from the same strains of A. niger, where Frisvad et al. [8] reported that up to 10% of A. niger strains may produce both mycotoxins. Ochratoxin A production rate can be overestimated in some studies as HPLC with fluorescence detection (even using immunoaffinity purification), can provide false positives [44] which unfortunately has been extensively reported for A. tubingensis. [43,45]. Ochratoxin A production from A. tubingensis was not detected during our screening process, confirming the report of Nielsen et al. [4] and Storari et al. [44] concerning this. Also, in accordance with Frisvad et al. [8], strains of A. tubingensis, A. brasiliensis and A. luchuensis did not produce fumonisins or ochratoxins.

A. carbonarius isolates always produced ochratoxin A as reported by several authors [4,46,47]. Our result was in accordance with this consistent production since all the 13 isolates of A. carbonarius showed ochratoxin A production.

Secalonic acids, reported as toxic metabolites of A. aculeatus [48] were noticed during our analysis, confirming their production by A. aculeatus as mentioned by Parenicová et al. [49]. These toxic compounds were not produced by A. aculeatinus from cashew nuts which is in contrast to the report by Noonim et al. [50].

Some secondary metabolites detected during our analysis, such as Aurasperones, Nigragillin, Malformins and Nigerazine and grouped as B in Table 1 are reported to be toxic compounds to plants, bacteria, and mice [5,51]. Asperazine was reported to have significant in vitro cytotoxicity against human leukemia [52] but no in vivo including bioavailability studies confirmed this. Malformins are currently being investigated for anti-cancer drug potential [53]. Altogether there are very few studies on other effects than in-vitro or in older studies intraperitoneal injection of compounds. However these do not include degradation in the body nor bioavailability of the compounds, and with a definition of mycotoxin being toxic through a natural route of exposure, such studies can only be considered indicative, but also highlights the need for testing these compounds in relevant animal models under relevant exposure conditions. Similar problems are reported in the Aspergillus glaucus group (and formerly Eurotium) [54,55]

DNA sequencing using β-tubulin and calmodulin genes was performed to validate morphology and extrolite profile based on identification of our isolates and their association to A. section Nigri. The use of a polyphasic approach, to identify and validate to species level isolates of fungi, was described by Frisvad [56], Oliveri et al. [14]. Our DNA sequences confirmed the species name identified by morphological and chemical characteristics and the phylogenetic tree shows that the main clades belong to the black Aspergilli. Perrone et al. [43] in a cluster analysis of 94 isolates of A. section Nigri identified the same clades confirming the fact that A. tubingensis and A. niger are the main clades of A. section Nigri as reported by Nielsen et al. [4]. Moreover, Samson et al. [57] confirmed the presence of these 4 different clades in A. section Nigri and grouped them as biseriate group of Aspergillus section Nigri in contrast to uniseriate group of Aspergillus including A. aculeatus.

The cluster analysis of the 27 strains using their metabolite profiles was similar to the clustering based on sequencing data. The secondary metabolites have been previously used most often in species recognition due to their high species specificity [58]. Samson et al. [57] mentioned that isolates of Aspergillus species usually produce a diverse range of secondary metabolites that are characteristic of the different groups of section of Aspergillus. They also reported that the production of a particular secondary metabolite is an efficient identification aid for allocating a species to section while profiles of secondary metabolites can be very effective in identifying an Aspergillus isolates to species. With few exceptions, this was effective during our analysis where the combined production of orlandin, fumonisins and kotanin was specific to A. niger, and the production of antafumicin A and B was specific to A. luchuensis.

Conclusion

The diversity in secondary metabolites including mycotoxins from isolates of Aspergillus section Nigri, analysed using UHPLC-QTOF-MS, revealed several metabolites produced by 7 different species that contaminated cashew nuts samples from Benin. In pure cultures on a laboratory medium, ochratoxin A and fumonisins, the 2 main toxic compounds from black Aspergilli, were produced by strains of 2 predominant species in A. section Nigri, namely A. niger / A. welwitchiae and A. carbonarius, although A. carbonarius is unable to produce fumonisins. Ochratoxins and fumonisins were produced by a relatively little proportion of the isolates of A. niger and A. carbonarius, but it is well know that species of A. section Nigri are the most isolated on cashew kernels, given a substantial number of species that may produce mycotoxins in cashew nuts. Even though the presence of fungi has not always meant the presence of mycotoxins, the production of ochratoxin A and fumonisins by isolates on A. section Nigri on cashew nuts could constitute an additional and hidden problem in term of mycotoxins content, and can negatively affect cashew nut safety and the nutritional quality of the nuts.

There are no regulations on ochratoxin A and fumonisins for raw and processed cashew nuts like those of EU and WHO on aflatoxins. Nevertheless, these findings suggest more investigations in order to detect the presence and the levels of ochratoxin A and fumonisins and to evaluate their exact contribution to the total level of mycotoxins in cashew kernels. But immediate actions should emphasize on the prevention by strengthening post-harvest practices that can lower fungal contamination along the cashew nut value chain, mainly during nut storage, where high contamination of species belonging to black Aspergilli are noticed.

Supporting Information

[Figure omitted. See PDF.]

S1 Fig. Hierarchical clustering based on metabolites profile of strains of Aspergillus section Nigri.

NB: The chemical clustering was performed using MultiExperiment Viewer (MeV v4.2); In brackets (“”) are the chemicals with high probability of being species specific metabolites.

https://doi.org/10.1371/journal.pone.0164310.s001

(DOCX)

S1 Table. Metabolites profile of Aspergillus section Nigri isolated from cashew nuts from Benin.

https://doi.org/10.1371/journal.pone.0164310.s002

(XLSX)

Acknowledgments

The first author (YL) gratefully thanks Wageningen University, the International Institute of Tropical Agriculture in Benin, and the EEC project MycoRed (KBBE-2007-222690-2). The DTU Metabolomics Platform, Department of Systems Biology, Technical University of Denmark is acknowledged for access to UHPLC- QTOF instrument.

Author Contributions

1. Conceptualization: YL KFN KH.

2. Data curation: YL KFN YG.

3. Formal analysis: YL KFN YG.

4. Funding acquisition: KFN MT MAJSB JBH JCF.

5. Investigation: YL KFN YG JBH JCF.

6. Methodology: YL KFN YG JBH JCF.

7. Project administration: KFN MAJSB JCF.

8. Resources: YL KFN YG KH EJS MT JBH JCF.

9. Software: YL KFN EJS.

10. Supervision: KFN MT JBH JCF.

11. Validation: YL KFN JBH JCF.

12. Visualization: YL KFN EJS JBH JCF.

13. Writing – original draft: YL.

14. Writing – review & editing: YL KFN ARL KH MJRN EJS MAJSB JBH JCF.

Citation: Lamboni Y, Nielsen KF, Linnemann AR, Gezgin Y, Hell K, Nout MJR, et al. (2016) Diversity in Secondary Metabolites Including Mycotoxins from Strains of Aspergillus Section Nigri Isolated from Raw Cashew Nuts from Benin, West Africa. PLoS ONE 11(10): e0164310. https://doi.org/10.1371/journal.pone.0164310

References

1. Schuster E, Dunn-Coleman N, Frisvad J, van Dijck P. On the safety of Aspergillus niger–a review. Appl Microbiol Biotechnol. 2002; 59(4–5):426–35. pmid:12172605

2. Chiotta ML, Ponsone ML, Sosa DM, Combina M, Chulze SN. Biodiversity of Aspergillus section Nigri populations in Argentinian vineyards and ochratoxin A contamination. Food Microbiol. 2013; 36(2):182–90. pmid:24010597

3. Månsson M, Klejnstrup ML, Phipps RK, Nielsen KF, Frisvad JC, Gotfredsen CH, et al. Isolation and NMR characterization of fumonisin B2 and a new fumonisin B6 from Aspergillus niger. J Agric Food Chem. 2010; 58(2):949–53. pmid:20028011

4. Nielsen KF, Mogensen JM, Johansen M, Larsen TO, Frisvad JC. Review of secondary metabolites and mycotoxins from the Aspergillus niger group. Anal Bioanal Chem. 2009; 395(5):1225–42. pmid:19756540

5. Blumenthal CZ. Production of toxic metabolites in Aspergillus niger, Aspergillus oryzae, and Trichoderma reesei: justification of mycotoxin testing in food grade enzyme preparations derived from the three fungi. Regul Toxicol Pharmacol. 2004; 39(2):214–28. pmid:15041150

6. Varga J, Kevei F, Hamari Z, Tóth B, Téren J, Croft JH, et al. Genotypic and phenotypic variability among black aspergilli. In: Samson RA, Pitt JI, editors. Integration of modern taxonomic methods for Penicillium and Aspergillus classification. Amterdam, The Netherlands: Harwood Academic; 2000. pp. 397–411.

7. Hong S-B, Lee M, Kim D-H, Varga J, Frisvad JC, Perrone G, et al. Aspergillus luchuensis, an industrially important black Aspergillus in East Asia. PLOS ONE. 2013; 8(6):e63769.

8. Frisvad JC, Larsen TO, Thrane U, Meijer M, Varga J, Samson RA, et al. Fumonisin and ochratoxin production in industrial Aspergillus niger strains. PLOS ONE. 2011; 6(8):e23496. pmid:21853139

9. Mosseray R. Les Aspergillus de la section "Niger" Thom & Church. La Cellule. 1934; 43:203–85.

10. Raper K, Fennell DI. The genus Aspergillus. Baltimore, MD, USA: Williams and Wilkins; 1965. 686 p.

11. Al-Musallam A. Revision of the black Aspergillus species. Utrecht: Univ Utrecht.; 1980. 92 p.

12. Jurjević Ž, Peterson SW, Stea G, Solfrizzo M, Varga J, Hubka V, et al. Two novel species of Aspergillus section Nigri from indoor air. IMA fungus. 2012; 3(2):159–73. pmid:23355969

13. Varga J, Frisvad JC, Kocsubé S, Brankovics B, Tóth B, Szigeti G, et al. New and revisited species in Aspergillus section Nigri. Stud Mycol. 2011; 69:1–17. pmid:21892239

14. Oliveri C, Torta L, Catara V. A polyphasic approach to the identification of ochratoxin A-producing black Aspergillus isolates from vineyards in Sicily. Int J Food Microbiol. 2008; 127(1):147–54.

15. Glass NL, Donaldson GC. Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes. Appl Environ Microbiol. 1995; 61(4):1323–30. pmid:7747954

16. Carbone I, Kohn LM. A method for designing primer sets for speciation studies in filamentous ascomycetes. Mycologia. 1999:553–6.

17. Balajee SA, Borman AM, Brandt ME, Cano J, Cuenca-Estrella M, Dannaoui E, et al. Sequence-Based identification of Aspergillus, Fusarium, and Mucorales species in the clinical mycology laboratory: where are we and where should we go from here? J Clin Microbiol. 2009; 47(4):877–84. pmid:19073865

18. USDA (United States Department of Agriculture). National Nutrient Database for Standard Reference Release 27, Basic Report: 12087, Nut, cashew nuts, raw. 2015. Available: http://ndb.nal.usda.gov/ndb/foods/show/3677?fgcd=Nut+and+Seed+Products&manu=&lfacet=&format=&count=&max=35&offset=35&sort=&qlookup=. Accessed 13 July 2015.

19. Škrbić B, Živančev J, Godula M. Multimycotoxin analysis of crude extracts of nuts with ultra-high performance liquid chromatography/tandem mass spectrometry. J Food Compost Anal. 2014; 34(2):171–7.

20. Gyedu-Akoto E, Lowor ST, Assuah M, Kumi W, Dwomoh EA. Assessment of post-harvest handling effects on quality of cashew nuts and kernels in Ghana. J Sci Res Rep. 2014; 3(77):953–65.

21. FAOSTAT-Food and Agriculture Organization Statistics Division [Internet]. 2015 [cited October 2015]. Available from: http://faostat.fao.org/site/567/DesktopDefault.aspx?PageID=567#ancor.

22. Adeniyi DO, Adedeji AR. Evaluation of fungal flora and mycotoxin potential associated with postharvest handlings of Cashew Nut Arch Appl Sci Res. 2015; 7(2):30–3.

23. Lamboni Y, Frisvad JC, Hell K, Linnemann AR, Nout RMJ, Tamo M, et al. Occurrence of Aspergillus section Flavi and section Nigri and aflatoxins in raw cashew kernels (Anacardium occidentale L.) from Benin. LWT—Food Sci Technol. 2016; 70:71–7.

24. Pitt JI. Fungi and food spoilage / J.I. Pitt and A.D. Hocking. Hocking AD, editor. London; New York: Blackie Academic & Professional; 1997.

25. Samson RA, Hoekstra ES, Frisvad JC, Filtenborg O. Introduction to food and airborne fungi. 6th ed ed. Utrecht, The Netherlands: Centraalbureau voor Schimmelcultures; 2000.

26. Klitgaard A, Iversen A, Andersen MR, Larsen TO, Frisvad JC, Nielsen KF. Aggressive dereplication using UHPLC–DAD–QTOF: screening extracts for up to 3000 fungal secondary metabolites. Anal Bioanal Chem. 2014; 406(7):1933–43. pmid:24442010

27. Nørholm MH. A mutant Pfu DNA polymerase designed for advanced uracil-excision DNA engineering. BMC Biotech. 2010; 10(1):1–7.

28. O'Donnell K, Cigelnik E. Two divergent intragenomic rDNA ITS2 types within a monophyletic lineage of the fungus Fusarium are nonorthologous. Mol Phylogenet Evol. 1997; 7(1):103–16. pmid:9007025

29. Hong S-B, Cho H-S, Shin H-D, Frisvad JC, Samson RA. Novel Neosartorya species isolated from soil in Korea. Int J Syst Evol Microbiol. 2006; 56(2):477–86.

30. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: molecular evolutionary genetics analysis version 6.0. Mol Biol Evol. 2013; 30(12):2725–9. pmid:24132122

31. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994; 22(22):4673–80. pmid:7984417

32. Jukes TH, Cantor CR. Evolution of protein molecules. In: Munro HN, editor. Mammalian protein metabolism. New York: Academic Press; 1969. pp. 21–132.

33. Varga J, Kocsubé S, Suri K, Szigeti G, Szekeres A, Varga M, et al. Fumonisin contamination and fumonisin producing black Aspergilli in dried vine fruits of different origin. Int J Food Microbiol. 2010; 143(3):143–9. pmid:20826035

34. Mogensen JM, Frisvad JC, Thrane U, Nielsen KF. Production of Fumonisin B-2 and B-4 by Aspergillus niger on Grapes and Raisins. J Agric Food Chem. 2010; 58(2):954–8. pmid:20014861

35. Noonim P, Mahakarnchanakul W, Nielsen KF, Frisvad JC, Samson RA. Fumonisin B2 production by Aspergillus niger in Thai coffee beans. Food Addit Contam Part A. 2009; 26(1):94–100.

36. Massi FP, Sartori D, de Souza Ferranti L, Iamanaka BT, Taniwaki MH, Vieira MLC, et al. Prospecting for the incidence of genes involved in ochratoxin and fumonisin biosynthesis in Brazilian strains of Aspergillus niger and Aspergillus welwitschiae. Int J Food Microbiol. 2016; 221:19–28. pmid:26803270

37. Vaamonde G, Patriarca A, Pinto VF, Comerio R, Degrossi C. Variability of aflatoxin and cyclopiazonic acid production by Aspergillus section flavi from different substrates in Argentina. Int J Food Microbiol. 2003; 88(1):79–84. pmid:14527788

38. Lund B, Baird-Parker TC. Microbiological safety and quality of food: Springer Science & Business Media; 2000.

39. Molyneux RJ, Mahoney N, Kim JH, Campbell BC. Mycotoxins in edible tree nuts. Int J Food Microbiol. 2007; 119(1–2):72–8. pmid:17719114

40. Marín S, Hodžić I, Ramos AJ, Sanchis V. Predicting the growth/no-growth boundary and ochratoxin A production by Aspergillus carbonarius in pistachio nuts. Food Microbiol. 2008; 25(5):683–9. pmid:18541167

41. Donner M, Atehnkeng J, Sikora RA, Bandyopadhyay R, Cotty PJ. Distribution of Aspergillus section Flavi in soils of maize fields in three agroecological zones of Nigeria. Soil Biol Biochem. 2009; 41(1):37–44.

42. Samson RA, Houbraken J, Kuijpers AF, Frank JM, Frisvad JC. New ochratoxin A or sclerotium producing species in Aspergillus section Nigri. Stud Mycol. 2004; 50:45–61.

43. Perrone G, Mulè G, Susca A, Battilani P, Pietri A, Logrieco A. Ochratoxin A production and amplified fragment length polymorphism analysis of Aspergillus carbonarius, Aspergillus tubingensis, and Aspergillus niger strains isolated from grapes in Italy. Appl Environ Microbiol. 2006; 72(1):680–5. pmid:16391107

44. Storari M, Bigler L, Gessler C, Broggini G. Assessment of the ochratoxin A production ability of Aspergillus tubingensis. Food Addit Contam Part A. 2012; 29(9):1450–4.

45. Medina A, Mateo R, López-Ocana L, Valle-Algarra FM, Jiménez M. Study of Spanish grape mycobiota and ochratoxin A production by isolates of Aspergillus tubingensis and other members of Aspergillus section Nigri. Appl Environ Microbiol. 2005; 71(8):4696–702. pmid:16085865

46. Abarca ML, Accensi F, Bragulat MR, Castella G, Cabanes FJ. Aspergillus carbonarius as the main source of ochratoxin A contamination in dried vine fruits from the Spanish market. J Food Prot. 2003; 66(3):504–6. pmid:12636310

47. Chiotta ML, Susca A, Stea G, Mulè G, Perrone G, Logrieco A, et al. Phylogenetic characterization and ochratoxin A–Fumonisin profile of black Aspergillus isolated from grapes in Argentina. Int J Food Microbiol. 2011; 149(2):171–6. pmid:21723640

48. Andersen R, Buechi G, Kobbe B, Demain AL. Secalonic acids D and F are toxic metabolites of Aspergillus aculeatus. J Org Chem. 1977; 42(2):352–3. pmid:830866

49. Pařenicová L, Skouboe P, Frisvad J, Samson RA, Rossen L, ten Hoor-Suykerbuyk M, et al. Combined molecular and biochemical approach identifies Aspergillus japonicus and Aspergillus aculeatus as two species. Appl Environ Microbiol. 2001; 67(2):521–7. pmid:11157212

50. Noonim P, Mahakarnchanakul W, Varga J, Frisvad JC, Samson RA. Two novel species of Aspergillus section Nigri from Thai coffee beans. Int J Syst Evol Microbiol. 2008; 58(7):1727–34.

51. Ghosal S, Biswas K, Chakrabarti DK. Toxic naphtho-. gamma.-pyrones from Aspergillus niger. J agric food chem. 1979; 27(6):1347–51. pmid:544666

52. Varoglu M, Corbett TH, Valeriote FA, Crews P. Asperazine, a selective cytotoxic alkaloid from a sponge-derived culture of Aspergillus niger. J Org Chem. 1997; 62(21):7078–9. pmid:11671801

53. Wang J, Jiang Z, Lam W, Gullen EA, Yu Z, Wei Y, et al. Study of malformin C, a fungal source cyclic pentapeptide, as an anti-cancer drug. PLOS ONE. 2015; 10(11):e0140069. pmid:26540166

54. Blaser P, Ramstein H, Schmidt-Lorenz W, Schlatter C. Toxizitat und Mutagenitat der xerophilen Schimmelpilze der Gattung Eurotium (Aspergillus glaucus-Gruppe). LWT—Food Sci Technol. 1980; 14:66–71.

55. Greco M, Kemppainen M, Pose G, Pardo A. Taxonomic characterization and secondary metabolite profiling of Aspergillus Section Aspergillus contaminating feeds and feedstuffs. Toxins. 2015; 7(9):3512–37. pmid:26364643

56. Frisvad JC. Rationale for a polyphasic approach in the identification of mycotoxigenic fungi. In: De Saeger S, editor. Determining mycotoxins and mycotoxigenic fungi in food and feed. Oxford: Woodhead Publishing; 2011. pp. 279–97.

57. Samson RA, Visagie CM, Houbraken J, Hong SB, Hubka V, Klaassen CHW, et al. Phylogeny, identification and nomenclature of the genus Aspergillus. Stud Mycol. 2014; 78:141–73. pmid:25492982

58. Larsen TO, Smedsgaard J, Nielsen KF, Hansen ME, Frisvad JC. Phenotypic taxonomy and metabolite profiling in microbial drug discovery. Nat Prod Rep. 2005; 22(6):672–95. pmid:16311630