This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Introduction
Aspergillus producing AF from contaminated animal feed, a secondary metabolite, is a global problem with high economic impacts on animal and human health. Contaminations occur before harvest, and during the storage and distribution of seeds. In sub-Saharan Africa, the global agricultural loss associated with AF contamination was estimated at 40% of the global production, corresponding to $450 million (Gbashi et al. 2018). Aflatoxins refer to a family of over 18 related compounds, principally B1, B2, G1, and G2 (Ibrahim 2019), produced primarily by the closely related fungi, Aspergillus flavus and Aspergillus parasiticus (Rao et al. 2020). Aflatoxins are polyketide-derived secondary metabolites. AflR encodes a protein containing a zinc-finger binding domain that interacts with target sequence DNA (Lee et al. 2006). AflS encodes a costimulatory regulatory protein that enhances the transcription activity (Liao et al. 2020).
A. flavus and A. parasiticus are dominant species found in crops (Kim et al. 2017) that share morphological similarities and closely cluster in phylogeny (Godet and Munaut 2010). Thus, the distinction between these species is difficult because microscopic identification requires experts in filamentous fungi taxonomy (El Khoury et al. 2011). In addition, the morphological methods of identification are inaccurate because of close inter-species similarities. In contrast, molecular techniques save labor time and their sensitivity and specificity, provide distinct species closeness and distance estimations without sophisticated culture and further confirmation steps (Ahmad et al. 2010).
Recently, several molecular approaches have been used to differentiate among many Aspergillus species, including random amplified polymorphic DNA (RAPD) (Yin et al. 2009; Daghriri et al. 2018), amplified fragment length polymorphism (AFLP), with specific diagnostic PCR primers, polymorphic microsatellite marker analysis, microsatellite length polymorphism (Healy et al. 2004), and DNA sequencing (Nicholson et al. 1998).
The commonly used molecular approaches often target the analysis of the coding sequences for rRNA that contain both conserved and variable regions (Paterson 2006). Fungal rDNA operons have been shown to contain variability regions within genera (Levy et al. 2001). However, PCR-RFLP has been used to identify variations in DNA sequences, and it was shown to recognize closely related organisms using inter-phylogenetic relations (Zarrin and Erfaninejad 2016). This technique has been proven to be fast, sensitive, and reliable in determining the genetic variability among A. section Flavi species (Bagyalakshmi et al. 2007; Mohankumar et al. 2010; Lavkor 2020). Currently, there is a limited understanding of molecular variability among populations of A. flavus isolated from feed sources. Characterization of A. flavus requires an expert polyphasic approach to fully confirm its taxonomic status and develop an effective control strategy (Singh et al. 2018; Frisvad et al. 2019; Martinez-Miranda et al. 2019). Using molecular biology methods combined with morphology has revealed the diversity and genetic variability within the A. flavus species (Okoth et al. 2018).
In the present study, we used the PCR-RFLP targeting the aflR-aflS(J) intergenic spacer (IGS) of the AF biosynthesis cluster to complement the findings of the morphological characterization. The main objectives were to establish and assess the genetic relationships and variability patterns of A. flavus strains isolated from feeds in the southwestern Saharan region of Algeria using PCR and sequence analysis of the IGS region, followed by phylogenetic analysis.
Experimental
Materials and Methods
Fungal isolates and morphological identification. Thirty fungal isolates were obtained from various animal feeds. Out of these, twenty were identified as Aspergillus, and 15 out of 20 were identified as A. section Flavi. Samples were collected in the livestock feed manufacturing unit’s storage facilities in Southwestern Algeria. As described earlier, the strains were isolated using Potato Dextrose Agar (PDA) (Pitt and Hocking 2009). Strains identification included macroscopic (colony color and morphology) and microscopic (head seriation, conidia morphology, and size) characteristics. The identification was performed according to the criteria defined for the taxonomic keys and guidelines for the Aspergillus genus (Pitt and Hocking 2009). All isolates were cultured on Aspergillus flavus/parasiticus agar (AFPA) (bacteriological peptone 10 g; yeast extract 20 g; ferric ammonium citrate 0.5 g; bacteriological agar 15 g; dichloran 2 mg (0.2% in ethanol, 1.0 ml); distilled water 1 l). One hundred microliters of spore suspension were inoculated in the center of 60 mm petri dishes and then incubated at 25°C for 48 h in the absence of light (Abriba et al. 2013; Hossain et al. 2018) as a particular growth condition for rapid identification of Aspergillus belonging to A. section Flavi (Pitt et al. 1983). At 48 h of culture on AFPA, the colonies of A. flavus and A. parasiticus strains appear with an orange color in the bottom of the culture, whereas the other species of Aspergillus section Flavi do not (Bothast and Fennell 1974; Hamsa and Ayres 1977; Pitt et al. 1983).
Molecular identification of fungal strains. Since the isolation and characterization by culture fail to differentiate between A. parasiticus and A. flavus, we used the molecular tools to perform this analysis.
Culture preparation. Fungal mycelia were produced by inoculating the fungal conidia with a glass Pasteur pipette in 250 ml Erlenmeyer flasks containing 150 ml Potato Dextrose Broth (PDB) and incubated for 48 hours at 25°C under orbital agitation at 150 rpm. Mycelia were harvested following filtration on 0.45 μm pore size filters and then snapped frozen into liquid nitrogen. Frozen mycelia were grounded into a fine powder using a mortar and pestle and then stored at –80°C until use.
DNA extraction. One hundred mg of powdered, frozen mycelia were transferred into a 1.5 ml micro-centrifuge tube, thawed, and resuspended into 800 μl of DNA extraction buffer (100 mM TRIS-HCl (pH 7.4), 20 mM EDTA (pH 8.0), 1.4 mM NaCl, 2% sodium dodecyl sulfate) as described by (Lee et al. 2006). Subsequently, 10 μl of Proteinase K (20 mg/ml) was added, and the solution was incubated for 20 min at 60°C. The cell lysate was supplemented with 800 μl of phenolchloroform-isoamyl alcohol (25/24/1, v/v/v) and then mixed by inversion until an emulsion was obtained. Following centrifugation for 5 min at 13,000 rpm at 20°C, the upper (aqueous) phase was transferred into a new microfuge tube. Four hundred microliters of chloroform/isoamyl alcohol (24:1) were added and mixed by inversion, and then the mixture was centrifuged for 10 min at 13,000 rpm at 4°C. The upper aqueous phase was transferred into a new microfuge tube and supplemented with 0.7 ml of cold 2-propanol for instant DNA precipitation. The precipitate was centrifuged for 10 min at 4°C, 13,000 rpm, and then the supernatant was discarded. The pellet was rinsed with 500 μl of 70% ethanol to remove residual salts. Following the last centrifugation at 13,000 rpm for 5 min at 4°C, ethanol was discarded, and the DNA pellet air dried. DNA pellet was then resuspended into 50 μl of sterile ultrapure water. To remove the contaminant RNA, 5 μl of RNase (20 mg/ml) was added, incubated for 15 min at 37°C, and then heat-inactivated for 20 min at 70°C. The quality of DNA was checked following DNA separation by electrophoresis in a 0.8% agarose gel in TAE (40 mM Tris, 20 mM sodium acetate, and 1 mM EDTA, pH = 8.6) buffer and a high molecular weight marker. DNA concentration was measured at the optical density (OD) of 260 nm and the purity at 240 and 280 nm wavelengths using a UV spectrophotometer.
Primers. The non-coding intergenic sequences (IGS) that separate the aflR and aflS(J) genes of fungal ribosomal DNA operon were used as a target for PCR amplification to detect A. flavus and A. parasiticus. Sequences of the primers used were as follows: IGS-F, 5’-AAGGAATTCAGGAATTCTCAATTG-3’ and IGS-R, 5’-GTCCACCGGCAAATCGCCG-TGCG-3’ (Ehrlich et al. 2003; 2007; El Khoury et al. 2011).
PCR amplification. PCR was performed using 200 ng of a fungal DNA template (5 μl), 1 μl of each of the IGS-R and IGS-F (20 pM) primers (Eurofins Genomics, France), 25 μl of Econo Taq Plus Green 2 × Master Mix (Euromedex, France), and 18 μl of sterile H2O. The Master Mix contained: 0.1 U/μl of Econo Taq DNA polymerase – reaction buffer (pH 9.0), 400 μM each (dATP, dGTP, dCTP, and dTTP); 3 mM MgCl2 and blue and yellow tracking dyes. Microtubes containing the PCR mix were transferred into a thermocycler and submitted to 34 cycles of amplification following an initial denaturation of DNA at 94°C for 5 min. Each cycle consisted of a denaturation step at 95°C for 1 min, an annealing step at 46°C for 1 min, and an extension step at 56°C for 1.5 min, with a final extension at 60°C for 10 min added to the last cycle. After amplification, the PCR products were stored at 4°C until used. PCR products were analyzed by electrophoresis separation on a 1.5 % agarose gel, and PCR-specific bands visualized by UV trans-illumination. The size of the amplicons was estimated using a 1 kb DNA ladder. The PCR products were purified using the clean-up Nucleospin kit (Machery Nagel, France), using the manufacturer’s instructions.
Restriction enzyme digestion of PCR-IGS amplicons: PCR-IGS-RFLP. PCR-RFLP using Bgl II restriction endonuclease to digest the IGS product was shown to distinguish between A. parasiticus and A. flavus (El Khoury et al. 2011). Here we used this technique to examine whether the strains identified as A. section Flavi could be identified as A. parasiticus and/or A. flavus. The PCR-IGS amplicons were digested with the restriction endonuclease BglII (Promega, France). The reactions were performed using a total of 0.5 μg of purified PCR product in 10 μl, 2 μl of 10 × Bgl II incubation buffer, 1 μl of Bgl II (10 U/μl), and 7 μl ultrapure H2O in a total volume of 20 μl. The mixture was incubated at 37°C for 2 h, and then separated by electrophoresis in a 2% agarose gel. Restriction bands were revealed using a UV trans-illumination and a Gel-doc Image Lab system (Bio-Rad, France).
Sequencing, sequence alignment, and phylogenetic analysis. A total of 1 μg of purified PCR product of each amplicon was shipped to Eurofins Genomics (Germany) for sequencing using IGS‑F primer. Sequence data were sent as electronic files to the lab for examination and use for bioinformatic analyses.
DNA BLAST analysis of obtained sequences was done online using the NCBI/NIH website to identify the closest sequences that match our new sequences. Multiple sequence alignment was performed using Clustal Omega software (The EMBL-EBI search and sequence analysis tools APIs in 2019). Sequences were submitted to GenBank on the NCBI website (http:// www.ncbi.nlm.nih.gov) under the reported accession numbers (Table I). Available sequences of aflR-aflS(J) intergenic region of reference strains were obtained from the GenBank database. The phylogenetic trees of isolates and their close relatives based on aflR-aflS(J) intergenic region were derived from sequence alignment using Clustal Omega.
Aspergillus flavus isolates identity based on the BLAST NCBI data.
The flavus group of Aspergillus was reported to be a significant cause of yearly economic losses reaching over 25% (Klingelhöfer et al. 2018), because of contamination of agriculture and feed components. This type of aflatoxin-producing fungi are opportunistic pathogens that contaminate corn, wheat, peanuts, and other food crops in many places of the world and causes severe diseases when ingested. Out of 30 fungal isolates, 20 were identified as Aspergillus, and 15 out of the 20 were characterized as A. section Flavi. Macroscopic and microscopic examinations of these 15 strains showed yellow-green colonies after growth on PDA, with spherical and rough spores/smooth to finely rough and globose conidia.
AFPA was used to identify both A. parasiticus and A. flavus (Hossain et al. 2018). Four colonies with yellow to green color with a white border surrounding the yellow to the greenish surface and an intense yellow-orange reverse color were selected as A. flavus and A. parasiticus (Fig. 1). This orange color is due to the reaction of ferric citrate with aspergillic acid, forming a colored complex. The eleven remaining colonies produced creamy reverse color on AFPA, characteristic of Aspergillus oryzae species (Pitt et al. 1983; Rodrigues et al. 2009; Frisvad et al. 2019).
Fig. 1
Phenotypic characterization of A. flavus isolates grown on A. flavus/parasiticus agar (AFPA) medium. Colony color: white-green colony diameter 18 mm, colony reverse color: yellowish orange after seven days of incubation at 25°C.
In previous studies, AFPA alone was used to identify A. flavus and A. parasiticus (Abriba et al. 2013; Fakruddin et al. 2015; Hossain et al. 2018; Krulj et al. 2020). However, based only on the phenotypic characteristics, it was impossible to distinguish between the two closely related species A. flavus and A. parasiticus because of overlapping morphological features (Rodrigues et al. 2009). Thus, molecular tools are necessary to differentiate these two species further. In the present study, we performed a molecular genetic characterization of the isolates selected to examine whether the four isolates could be A. parasiticus or A. flavus.
DNA samples from the four isolates were used as templates for PCR amplification using IGS-specific primers. These primers were highly specific for aflR-aflS(J) IGS fragments. DNA samples from all four fungal strains produced the expected 674 bp IGS amplicon with no additional or non-specific bands (Fig. 2). However, this 674 bp product was identical for both A. flavus and A. parasiticus. Therefore, we needed to examine the PCR product by RFLP further.
Fig. 2
PCR amplification of intergenic sequences of the four isolates. IGS-F and IGS-R primers were used for PCR amplification of the intergenic sequences in total DNA isolated from the four fungal isolates as described in Material and Methods. A total of 5 μl of each PCR product was separated in a 1.5% agarose gel by electrophoresis. Lane 1 kb ladder (L) DNA marker, Aspergillus flavus isolates are indicated on the top of each lane. The sizes of selected DNA fragments are indicated on the sides of the panels.
Previous studies indicated that PCR-IGS-RFLP using restriction enzymes that cleave the PCR products into sub-fragments are valuable tools for detecting and differentiating between A. flavus and A. parasiticus (El Khoury et al. 2011; Nikolic et al. 2018; Lavkor 2020). Therefore, we used the restriction endonuclease BglII to digest the four PCR products. Results shown in Fig. 3 clearly show that digestion of all the PCR products with BglII resulted in an identical profile yielding three fragments of 362, 210, and 102 bp. This typical profile demonstrated that all species were A. flavus. In contrast, BglII digestion should produce a profile with only two fragments of 311 bp and 362 bp with an IGS amplicon of A. parasiticus (El Khoury et al. 2011). Our analysis failed to detect this last profile among the four analyzed.
Fig. 3
PCR and RFLP analyses of fungal amplicons. A total of 5 μl of PCR products of each of the isolates (FZM1, FAK45, FSZ47 and FDY50) was digested with Bgl II restriction endonuclease as described in Materials and Methods. Nondigested (ND) and digested (D) products were loaded and separated by electrophoresis in a 2% agarose gel. A 100 bp DNA ladder (L) was used as molecular weight marker, the positions of 100, 500 and 1,000 bp fragments are indicated. Positions of the 362, 210 and 102 bp fragments generated by BglII digestion are indicated.
Our findings were similar to those reported by (Kana et al. 2013), which indicate that all A. flavi isolates recovered from all the commodities (feedstuffs and animal feed) were A. flavus. In contrast, other studies (Stanković et al. 2015; Nikolic et al. 2018) found the presence of A. parasiticus on wheat kernels and corn. This difference may result from the geographic isolation and the local atmosphere conditions (temperature, humidity), which might cause shifts in mycobiota distribution and composition of cereals observed among different regions.
Accurate identification and differentiation between this aflatoxigenic species are of great importance in determining toxicological risks due to the difference in the toxic profile of each species in the section Flavi. (Martínez-Culebras and Ramón 2007). It is well known that A. parasiticus produces only “B” and “G” type toxins, whereas A. flavus produces “B” type toxins but also cyclopiazonic acid, versicolorin, and sterigmatocystin (Wilson et al. 2002; Bailly et al. 2018). Metabolite analysis is a multistep process that requires long term culture and mycotoxin production (14 d) followed by extraction and purification for chromatographical analysis. In addition, this method requires multiple and expensive instruments to perform all these processes. In contrast, PCR-RFLP is a rapid and sensitive method requiring minimal investment and was shown to be a reliable method for taxonomic studies of Aspergillus species (Martínez-Culebras and Ramón 2007).
The Aspergillus genome is organized into eight chromosomes where the genes encoding aflatoxins are located in the 54th cluster, 80 kb from the telomere of chromosome 3 (Georgianna and Payne 2009). They include regulatory genes aflR, aflS(J), and several structural genes (Yu et al. 2004; Paterson 2006; Price et al. 2006). The aflR and aflS(J) genes are separated by a 737-bp intergenic region (Yu 2012). Sequence variability within this intergenic fragment has been used to establish the phylogenetic organization of A. flavus genotypes (Al-Wadai et al. 2013).
To confirm the PCR-RFLP data, suggesting that all our four isolates were A. flavus, we sequenced their amplicons using the IGS-F primer. The sequences were deposited in NCBI/Genbank under the following accession numbers (Table I). Sequence data were used for blast analysis against the Genbank DNA sequence database and to produce multiple sequence alignments (Fig. 4). The highest homology of the sequences of the isolates with those of sequences of the GenBank reference strains was used to define the isolates. All four isolates were identified as A. flavus with a similarity percentage ranging from 97.48% to 99.67% to A. flavus RefSeq from NCBI (Table II and III). In addition, sequence analysis confirmed the presence of the two restriction sites for BglII in the amplicon that yielded the three fragments (362, 210, and 102 bp). Thus, the aflR-aflS(J) IGS-RFLP assay with BglII is sufficient to detect A. flavus without sequencing. Therefore, this assay can be used for epidemiological studies to detect the prevalence of A. flavus.
Fig. 4
Nucleotide sequence alignment of the four isolates with four reference strains. The names of the (.) four isolates are in green. Nucleotide sequence alignment was performed using Clustal software (–) represent similarity (.), represent deletion. Nucleotide changes are shown in bold font.
Genbank accession numbers of the Aspergillus flavus isolates.
Isolate
Accession number
Strain
FZM1
OL944584.1
A. flavus strain FZM1 aflR-aflJ intergenic region, partial sequence
FAK45
OL944586.1
A. flavus strain FAK45 aflR-aflJ intergenic region, partial sequence
FSZ47
OL944587.1
A. flavus strain FSZ47 aflR-aflJ intergenic region, partial sequence
FDY50
OL944588.1
A. flavus strain FDY50 aflR-aflJ intergenic region, partial sequence
Genbank accession numbers of the Aspergillus flavus reference strains.
Accession number
Strain
CP051037.1
A. flavus strain A9 chromosome 3
CP051085.1
A. flavus strain K54A chromosome 3
CP051045.1
A. flavus strain Tox4 chromosome 3
CP047251.1
A. flavus strain SU-16 chromosome 3
Furthermore, this sequence analysis showed variation among the A. flavus isolates. This variation included single and multiple substitutions, insertions, inversions, and deletions. Sequence comparison of FSZ47 and FDY50 isolates with that of the reference strain SU 16 showed the highest sequence similarity (99,37%, and 99.67%, respectively). Since the reference strains A9, K54A, and Tox4 are 100% similar in the studied segment, their similarity with FZM1 and FAK45 isolates was found to be 97.48% and 98.12% respectively. Therefore, the highest divergence compared to the existing reference strains was seen for the sequences of the FZM1 isolate. A total of 13 positions of divergence involving single nucleotide polymorphisms (SNPs) as well as double and multiple nucleotide polymorphisms (MNPs) were observed (Fig. 4). In contrast, only minor variations were seen in FDY50 isolate corresponding to SNPs at the positions 3, 5 and 28 of the sequence. On the opposite side, the isolate FAK45 showed the highest genetic similarity with A. flavus reference strains A9, K54A, and Tox4 with only minor variations at the positions 50 and 103. Most of the observed polymorphisms were nucleotide substitutions, with only 10% being insertions and inversions, as shown in Table IV. Nucleotide deletions and SNPs in AF coding genes are often associated with the inability of A. flavus to produce aflatoxins. Deletion strains of A. flavus were reported in many studies (Chang et al. 2005; Donner et al. 2010; Adhikari et al. 2016; Hua et al. 2018).
Nucleotide variations in aflR-aflS(J) intergenic region sequences of Aspergillus flavus isolates.
However, not all AF non-producing strains have deletions in the AF biosynthesis gene cluster. The strain NRRL 30797 known as a biocontrol agent has a few substitutions in the aflatoxin biosynthesis gene cluster (Chang et al. 2012). This genetic variability may complicate the diagnosis and, therefore, the control of A. flavus infection and aflatoxin contamination of food and feed. Detecting particular strains of A. flavus, such as NRRL 30797, requires sequencing or a special PCR-RFLP capable of detecting the single substitutions. In the present study, the IGS region has been used as a target for PCR and sequencing for phylogenetic analysis. The selection of this region was motivated by the sequence variability between different species but only minor variations within isolates of the same species. The phylogenetic tree of isolates and their close relatives based on the aflR-aflS(J) intergenic region was constructed using Clustal Omega (Okoth et al. 2018) to illustrate the relationships of isolates among A. flavus species. The phylogenetic tree obtained clearly shows that the A. flavus strains clustered into three major clades. In addition, several sub-clades indicate a wide variation among A. flavus species (Fig. 5). FSZ47 and FDY50 isolates were clustered in one clade and were found to be closely related to A. flavus strain SU 16 forming a mixed cluster. In contrast, FAK45 and FZM1 isolates were found to form distinct sub-clades. Variations exist among A. flavus isolates; those with SNPs share a closer relationship clustered with A. flavus strain SU16; while the isolate FAK45 clustered with a distinct sub-clade. These data agree with the previous finding reported by different groups (Krimitzas et al. 2013; Chiba et al. 2014). Recently, new direct detection methods of A. flavus through AF-B1 and Zearalenone by Raman spectrometry (Yang et al. 2021) or AF-B1 alone using a ortable Raman spectrometer combined with colloidal Au nanoparticles (Wang et al. 2022) were reported. This latter rapid technique provides quantitative measurement of AF-B1 in grain crops.
Fig. 5
Maximum Likelihood phylogenic tree showing the relationships between the examined A. flavus isolates and reference strains, based on aflR/aflS(J) intergenic sequence.
Conclusions
The four diagnosis methods (phenotyping, PCR, PCR-RFLP, and amplicon sequencing) have shown different sensitivity levels of fungal characterization. Phenotypic tools were limited to A. section Flavi identification; PCR alone failed to distinguish A. parasiticus from A. flavus; PCR-RFLP in this study was able to detect A. flavus only. However, it failed to distinguish between different strains of A. flavus. Sequencing and sequence analysis was able to distinguish between the A. flavus strains. These data established the selection criteria of the methods to use depending on the segregation level of fungal strains. New methods based on Raman spectrometry to measure AF-B1 alone or in combination with other fungal components might provide additional help for the detection of the pathogenic A. flavus in grain crops.
