Characterization of Mating Type Genes in Aspergillus flavus Populations from Two Locations in Kenya

In this study, the possibility of sexual reproduction in sampled Aspergillus flavus strains was evaluated by assessing the distribution of mating type (MAT) genes, which are known to control sexual character among fungi, for two counties in Kenya. Forty-four isolates fromNandi andMakueni counties were genotyped byMAT using a multiplex polymerase chain reaction assay.The primer pair for the MAT1-1 amplified a 396 base pair (bp) fragment containing an α-box motif, and MAT1-2 primers targeted a 270 bp segment with a high mobility group protein. The MAT1-2 genes dominated in both regions although the frequency was higher in Nandi (75%) than in Makueni (54.17%). There were no MAT1-1 genes sampled in Nandi, and in Makueni their proportion was 15.91%.Thepercentage of isolates that amplified for bothMATgenes inMakueniwas 9.09%, while inNandi it was 11.36%. Currently, use of aggressive aflatoxin non-producingA. flavus strains as biocontrol is the most promising preharvest aflatoxin control strategy in Kenya. However, we address the possibility of introduced biocontrol strains to breed with existing aflatoxin producing strains in nature, which could lead to the generation of A. flavus offspring capable of aflatoxin production while also being aggressive colonizers and possibly increasing the burden of aflatoxin exposure in food.


Introduction
Aspergillus spp.belongs to phylum Ascomycota and family Trichocomaceae.Species in the fungus genus' section Flavi have potential to infect maize, peanuts, cotton, and tree nuts among other crops and may contaminate with aflatoxins, which are Group 1 carcinogens according to International Centre for Cancer Research [1,2].The growth of A. flavus is usually favored by hot dry conditions with optimum temperature of 37 ∘ C, but the fungus readily grows between temperatures of 25 and 42 ∘ C and will even grow at temperatures from 12 to 48 ∘ C. Its ability to grow at such high temperatures contributes to its pathogenicity in humans [2,3].
Many fungi have the potential to reproduce both sexually and asexually.Successful sexual reproduction requires the presence of functional and compatible mating type genes (MATand MAT -genes).In homothallic fungi, selffertility is possible when both mating type genes are present and functional within the same organism, either on different chromosomes or adjacent to the same chromosome.In heterothallism, only one mating type gene is present in a strain, so it is considered self-infertile and requires the presence of a compatible strain having a functional gene representative of the opposite mating type [4,5] (Horn et al., 2013).Aspergillus flavus is considered functionally heterothallic, and it is uncommon for strains to contain both MAT genes.Horn et al. [6] reported A. nomius strain that contained both MAT genes, which were self-infertile but functionally bisexual, meaning they could mate with both MAT -and MAT -strains.
Since 1981, yearly cases of aflatoxicosis have been reported in the Eastern parts of Kenya following consumption of maize contaminated with A. flavus and aflatoxins.Agriculture is the economic backbone in Rift Valley and Eastern Kenya with maize being the staple food, although aflatoxicosis has not yet been reported in the Rift Valley.Makueni is one of the areas in Eastern Kenya that has reported history of aflatoxicosis [2,[7][8][9].Despite Nandi being a maize growing area, cases of aflatoxicosis have not been recorded, although oesophageal cancer cases have been reported [7,8].In young children, between 0 and 5 years of age, Kangethe et al. [8] reported high aflatoxin levels in Makueni compared to Nandi.Aflatoxin exposure through milk for children younger than 5 years was 4 × 10 −4 and 1 × 10 −4 g/kg per day in Makueni and Nandi, respectively; the exposure of nursing children through breast milk was 6 × 10 −3 and 1 × 10 −6 g/kg per day in Makueni and Nandi, respectively.Children below 30 months of age in Makueni had 1.4 times higher aflatoxin levels in their urine than those of the same age in Nandi.
The use of biological control method is presently the most favorable strategy for lowering preharvest contamination of cereals, groundnuts, and tree nuts with aflatoxin.Nonaflatoxin producing A. flavus is introduced into the environment and outcompetes naturally occurring aflatoxin producing strains in the soil [10].The technology is under trial in Kenya in the aflatoxin hot spots [11][12][13].However, A. flavus populations show high genetic variations due to its potential to outcross by sexual recombination under special conditions in the soil.When such outcrossing occurs, the biocontrol strain could recombine with the indigenous populations of aflatoxigenic strains and generate offspring that not only inherit the aggressiveness of the non-aflatoxigenic parent, but could also inherit the ability to produce one or more serious mycotoxins, such as aflatoxin B1 or cyclopiazonic acid (CPA) (Horn et al., 2013) [14].Therefore, there is a need to understand the potential of introduced biocontrol strains to outcross with the existing aflatoxin producing A. flavus in the soil [14].This study aimed to determine the distribution of MAT genes for A. flavus strains originally isolated from soil and maize samples from Makueni and Nandi as described by Nyongesa et al. [15] and Okoth et al. [16].This is important to be assessed for two reasons: (1) to ascertain the level of recombination in the field population, which is relative to the proportions of MAT genes present, and (2) to better select a candidate biocontrol strain that will lessen the opportunity for recombining with the indigenous population [17].

Materials and Methods
. .Source of Fungal Isolates.Forty-four Aspergillus flavus isolates used in this study were originally isolated from maize and soil samples collected from Eastern (Makueni) and Rift Valley (Nandi) regions of Kenya and were recorded to have capacity to produced aflatoxins [15,16].The culture collection was maintained at Mycology Laboratory, University of Nairobi, Kenya.
Nandi lies within latitudes 0 ∘ and 0 ∘ 34  North and longitude 34 ∘ 44  and 35 ∘ 25  East, altitude between 1300 m and 2500 m above sea level.The area receives approximately 1200-2000 mm of rainfall and average temperatures of 20 ∘ C annually.Makueni area is 1218 m above sea level with average annual temperatures of 24 ∘ C and rainfall between 200 and 1200 mm.The rainfall is unreliable with frequent droughts [2,[7][8][9].

. . Molecular Characterization of Aspergillus Isolates
. . .Extraction of DNA.Fungal isolates from Eastern and Rift Valley parts of Kenya, with determined aflatoxin accumulation levels in yeast extract sucrose (YES) according to Okoth et al. 2012 [16] and Nyongesa et al. 2015 [15] (are shown in Table 1), preserved on silica gel beads at 4 ∘ C were used.Two to three silica gel beads were transferred on potato dextrose agar (PDA) plates under sterile conditions.Each A. flavus strain was cultured in replicates.The isolates were grown under 37 ∘ C for 7 days, and DNA was extracted using a Zymo Research fungal/Bacterial DNA Mini Prep Kit (Epigenetics, Hatfield, South Africa) according to the manufacturer's instructions.

. . Analysis of Aspergillus flavus Isolates by Mating Type.
The MAT genes were established by a diagnostic polymerase chain reaction (PCR) using primers M1F and M1R for the MAT -gene and M2F and M2R for the MAT -gene [18,19] (see Table 2).The PCR was performed in 20 l reactions, which included 1l of a 1:10 or 1:100 DNA dilution, 1 U REDTaq DNA polymerase (Sigma-Aldrich Company, Milan, Italy), 2 l REDTaq buffer supplemented with 1.7 l of 22 mM MgCl 2 for a final concentration of 3.0 mM, 10 mM deoxyribonucleotide triphosphates, 0.5% Bovine Serum Albumin (BSA), and 0.5 M of each of the 4 primers (M1F, M1R, M2F, and M2R) [18,19].Reactions were run in a Mastercycler ep gradient (Bio-Rad, California, USA) with a thermal profile of 5 min at 95 ∘ C followed by 40 cycles of 30 s at 95 ∘ C, 60 s at 54 ∘ C, and 45 s at 72 ∘ C. The amplified DNA was electrophoresed in 1.5% (w/v) Tris-acetate-Ethylenediaminetetraacetic acid (EDTA) agarose gels, and amplicons were designated as MAT -and MAT -using a 100 bp DNA ladder (exACTGene, Fisher Scientific International) as a size standard.
A concentration of 1% agarose gel was made.The agarose was boiled at 100 ∘ C for 5 minutes in a conical flask and left to cool to 55 ∘ C, and 0.3 l of ethidium bromide was added while swirling the flask to enable the gel mix with ethidium bromide.The mixture was poured into a gel tank with the combs on and left to solidify.Molecular marker (2 l) was added to one well and DNA (4 l) to the other wells and the arrangements were noted.The gel was run for 45 minutes at 80 voltage and viewed under gel doc (Bio-Rad, Molecular Imager Gel Doc6 XR-CLASS, Imaging System, California, USA).
In this study, sampled A. flavus strains from Makueni and Nandi include those that contain at least fragments of both MAT genes and those that contain a single MAT gene.Fungi that produce functional MAT -and MAT -genes on the same thallus are self-fertilizing while those with a single MAT gene require compatible MAT -or MAT -nuclei from two different individuals.There were 20.45% isolates with fragments of both MAT genes though Nandi had 11.36%, while in Makueni, the percentage was lower (9.09%).In both regions, MAT -isolates frequency was dominant (61.36%),  although the frequency in Nandi was higher (75%) than in Makueni (54.17%).The isolates from Nandi had no MATgenotypes, while in Makueni their percentage was 15.91%.
Laboratory crosses between strains of A. flavus of opposite mating type genes have successfully been applied to induce sexual recombination between MAT -and MATisolates [20][21][22].In our study, most of the isolates from Makueni were either MAT -or MAT -, indicating that they are likely heterothallic.Four of the isolates sampled in Makueni contained fragments of both MAT genes (Table 3), indicating the potential for a homothallic existence.The nearly equal distribution of single mating types, as well as the presence of isolates exhibiting evidence of two MAT genes, in this region could be an indication that there is active recombination occurring in the fields where these isolates were sampled.Five of the Nandi isolates exhibited evidence for containing both MAT genes.Of the isolates that contained one or the other MAT gene, no MATisolates were identified (Table 3).From this we can infer that recombination is not occurring at a rate comparable to Makueni.Mating tests would allow us to confirm or refute the functionality of these loci.
We may be able to infer a correlation between mating type distribution and aflatoxin outbreaks based on our findings.For example, aflatoxin outbreaks in in Makueni (having a more equal distribution of mating types) have been reported since 1981, while no aflatoxin outbreaks have been reported in Nandi, which has a predominance of MAT -genes [7-9] ( Okoth et al., 2016).Sexual recombination between toxigenic and non-aflatoxigenic A. flavus strains in the soil is viewed as a major cause of diversity enabling some of the A. flavus progenies to inherit the ability to produce aflatoxin [5,14,17,20,[23][24][25].
Currently, biocontrol is the most favorable technique for lowering preharvest aflatoxin contamination.The strategy includes spreading non-aflatoxigenic A. flavus strain spores, which result in greatly reduced levels of aflatoxin.Both aflatoxin and non-aflatoxin producing strains occupy the same niches in the soil [26].In our study, the 44 A. flavus strains characterized are agricultural isolates [15,16].Existence of a nearly equal distribution of MAT genes in Makueni indicates potential for sexual recombination (Table 3).Atoxigenic MAT -IVM300566 (0 g/kg) as biocontrol could outcross with toxigenic MAT -in the two regions which could facilitate recombination.Therefore, it is advisable to search for an atoxigenic MAT -strain as a candidate biocontrol strain.Ehrlich [10] suggests that it is important to test frequency of genetic recombination in agricultural environments where non-aflatoxin producing biocontrol has been introduced.Based on our findings and extrapolation to the field, it is important to assess this distribution for a field prior to release of fertile biocontrol strains that can recombine with the indigenous population.We must also verify the fecundity of strains that contain both MAT genes to determine if they are functionally bisexual as Horn et al. did with A. nomius [6].

Conclusion
This study underscores the importance of investigating the mating type distribution in the field, prior to biocontrol selection and release, to minimize the potential for sexual recombination while promoting efficacy of the biocontrol strain.Our findings indicate that A. flavus strains in Kenya have the potential to harbour both MAT genes, although we are uncertain of their functionality.Also, we show that not all fields will have the same distribution of mating types.The Makueni field population may have higher genetic and chemotype diversity, and potential for sexual recombination, due to the observed distribution of mating types.Selection of a naturally infertile atoxigenic strain as biocontrol would be better here.For Nandi, the distribution is less diverse, and if the MAT -+MAT -isolates are incapable of self-fertilizing or outcrossing, then use of MAT -biocontrol strain would be effective.Another characteristic that would be helpful would be to ensure that the biocontrol strain lacks aflatoxin cluster genes since it might be more difficult to inherit the entire cluster during recombination.

Table 2 :
Sequences of nucleotide primers used in the study.

Table 3 :
Mating type distribution for A. flavus isolates sampled in two counties in Kenya.