Fungal Biomass Load and Aspergillus flavus in a Controlled Environment

2.1 Cultures of fungi

The aflatoxigenic A. flavusKSM014 isolate was cultivated and maintained as described previously [14] and thereafter stored as spore suspension in 15% glycerol for short term storage at −20°C or for long term storage at −80°C prior to DNA/RNA extraction.

2.2 Maize cultivars

GAF4 maize lines and KDV1 varieties were obtained from Kenya Agricultural and Livestock Research Organization (KALRO), Kenya. The selection of the varieties are focused mainly on their drought tolerance and the agro-ecological zones in which they were grown. Striga tolerant variety (GAF4) is produced by KALRO Kibos, Kisumu County. GAF4 is cultivated in Homa Bay, Kisumu and Busia counties [15]. The Kenya Dryland Varieties 1 is an open pollinated hybrid recommended for medium to low altitude areas. KDV1 is drought tolerant, matures early and produce flowers after germination between 45 and 52 days. It is mainly cultivated in Homa Bay and Makueni regions (http://drylandseed.com).

2.3 Media preparation and reagents

Phytagel, Nicotinic acid, Glycine, Thiamine hydrochloride, Murashige and Skoog medium (MS), Potassium hydroxide, Pyridoxine hydrochloride and Myo-inositol were from Sigma-Aldrich (USA). MS vitamins; 5 g myo-inositol, 500 mg Thiamine-HCl, 500 mg pyridoxine-HCl, 250 mg nicotinic acid and 100 mg glycine were filter sterilized after preparation in distilled water and thereafter stored at −20°C according to the instructions of the manufacturer’s (Sigma-Aldrich, USA). The modified MS media was briefly prepared, 2.15 g MS salt was dissolved in sterile H₂O, thereafter, 10 ml MS vit. added and pH 5.7 adjusted using 1 M KOH and volume further adjusted to 1 l using sterile H₂O. 5 g of phytagel was added to MS media and heated in microwave to dissolve the salts. Fifty milliliters of the media was dispensed into tissue culture bottles, autoclaved and thereafter, cooled in the level 2 biosafety cabinet for approx. 1 h prior to inoculations as previously described [14].

2.4 Seed sterilization and Aspergillus flavusinfection

The seeds were sterilized in a biosafety cabinet, level 2 [Contained Air Solutions (CAS) BioMAT2, UK]. Twenty milliliters of 95–100% ethanol was used for sterilization of viable seeds for 1 min and briefly shaken for 15 s. The alcohol was replaced with 20 ml of sodium hypochlorite (2.5%). After 15 min of reaction at room temperature, the mixture was shaken for 30 s and thereafter, the liquid discarded. 30 ml of sterile H₂O was used 5× to wash the seeds with intermittent shaking after every wash. 50 ml of sterile H₂O was added and left to stand for 1 hr at rmt. The H₂O was replaced with 20 ml of 2% Tween 20 and shaken for 30 s. Conidia suspensions adjusted to 1 × 10⁶ conidia ml⁻¹ using a hemocytometer was used to inoculate the seeds. The seeds in the tubes were para filmed after sealing and kept for 30 min in a shaking incubator at 30°C. Controls were treated with sterile H₂O instead of spores of conidia and thereafter, incubated following the same conditions. The seeds were left to dry in Petri dishes after inoculations overlaid with filter paper overnight (Whatman No. 1). The seeds were germinated in a plant growth chamber, Conviron (Winnipeg, Manitoba, Canada) set at 28°C after subsequent inoculations onto tissue culture bottles. The germination and growth were observed for a 14-day period, tissues of the plant (roots and shoots) were separately harvested, flash frozen in liquid nitrogen prior to DNA/RNA extraction and stored at −80°C.

2.5 The extraction of gDNA from maize tissues and Aspergillus flavus

A 100 mg of each of the following samples: control healthy maize tissues, infected and A. flavusKSM014 mycelia was used to extract DNA following the method of Möller et al. [16] with some modification. Briefly, 100 mM Tris pH 8.0, 2% SDS, modified TES buffer, 2% (w/v) polyvinylpyrrolidone (PVP) and 10 mM EDTA was prepared. 5 μl RNase (10 mg/ml) and 450 μl of TES buffer was added to microtube (2 ml) containing the tissues and thereafter, homogenized by vortexing for 15 min or with a microtube pestle. Twenty microlitres of Proteinase K (1 μg/μl) was added and vortexed for 1 min, thereafter, incubated for 1 h at 60°C. Seventy microliters of 10% CTAB (0.1 vol.) and 160 μl of 5 M NaCl (0.3 vol. was added and incubated at 65°C for 10 min). Seven hundred and fifty microliters of chloroform/isoamyl-alcohol (24:1) was added, vortexed for 5 min, incubated on ice for 30 min and centrifuged at 14,000 rpm for 10 min. The aqueous phase was transferred cautiously onto a new microtube (2 ml) and 300–350 μl isopropanol (0.55 vol.) added and left to stand for 30 min at RT after mixing gently for 30 s. The mixture was centrifuged for 10 min at 14,000 rpm. Supernatant was cast-off, the pellets rinsed with chilled 700 μl of 70% ethanol twice and centrifuged again at 14,000 rpm for 2 min after mixing gently. The pellets were air dried and dissolved in 40 μl TE buffer (10 mM Tris-Cl pH 8, 1 mM EDTA pH 8) or nuclease free water after discarding the ethanol. The integrity of DNA was assessed on a 1% agarose/EtBr gel and the concentration quantified on a Nano-Drop^TM 1000 spectrophotometer (Nano Drop Technologies, USA). DNA was diluted to 10 ng/μl for further analysis.

2.6 Designing of primers

Sets of 3 primers (Table 1); elongation factor 1 alpha (Ef1ɑ), β-tubulin, and membrane protein (MEP) were used in the current research. MEPand Ef1ɑwere obtained from Dr. Shane Murray (pers. comm) whereas, β-tubulin was designed in Primer3 ver. 4.0 programme [19]. Secondary structure formation was evaluated in DNAMAN software ver. 6.0 (Lynnon LLC., USA) and further verified in OligoAnalyzer Tool (Integrated DNA Technologies). The melt curve and PCR analysis were used to identify both non-specific and specific amplification.

2.7 PCR amplification

Conventional polymerase chain reaction amplification was carried out in volumes of 25 μl and consisted of 0.5 μl of 10 μM dNTPs (Bioline), 10× reaction buffer with MgCl₂, 1 μl of 10 μM forward and reverse primers, 0.2 μl Kapa Taq, 1 μl of 10 ng DNA template, and sterile H₂0. Protocol performed and followed for cycling conditions were: 1 cycle for 5 min at 94°C followed by 35× (for 30 s at 94°C, for 45 s at 60°C, for 90 s at 72°C). Elongation step was achieved at 72°C for 7 min and finally at 4°C for 1 min. The products of PCR were assessed on 2% agarose/EtBr gel in TAE1 X buffer (Tris–acetate 40 mM and EDTA 1.0 mM). Fermentas (100 bp DNA ladder) was used as a molecular size marker.

2.8 Standard curves and fungal quantification

Ten-fold serial dilutions of pooled 10 ng gDNA extracts from A. flavusand control plants were used to create standard curves. The threshold cycle (Ct) values for each dilution were plotted against the logarithm of the starting quantity of the template. The amplification efficiencies were created from the std. curve slopes according to the methods [13, 20]. Additionally, linear regression curves were drawn, and the qPCR efficiency was calculated as: E=10−1Slope.

The quantity of targeted DNA in an unknown sample was inferred from the respective std. curves.

Ten nanograms of DNA isolated from infected and healthy maize roots and shoots respectively were used to assess primer specificity. For the exclusion of false negative results, template DNA samples from fungi were assessed for polymerase chain reaction amplification with primer pairs EF1ɑand β-Tub. DNA extracted from pure fungal cultures (A. flavus) and control plant tissues were pooled, diluted to 10 ng/μl and used to evaluate the quantity of fungal DNA template in the infected plant tissue. The final fungal DNA template concentrations were 1, 5 × 10⁻¹, 2.5 × 10⁻¹, 1.25 × 10⁻¹, 6.25 × 10⁻², 3.125 × 10⁻² ng/μl. These dilutions were used to estimate the detection limits of the EF1ɑand β-Tub primer pair in the infected plant tissues. Serial dilutions of extracted DNA from healthy maize tissue were prepared to gauge the detection limits of the MEP. For normalization and quantification of gene between different samples, the amount of fungal DNA as calculated by the Ctvalue for β-Tub and/or EF1ɑwas divided by the amount of maize DNA as calculated by the Ctvalues for MEP. Rotor Gene 6000 2 plex HRM (Corbett Life Science Research, Australia) was used to assess the profiles of gene expression. Kapa SYBRFast Kit, Master mix (Kapa BioSystems, South Africa) containing DNA polymerase, reaction buffers, dNTPs and 3 mM MgCl₂ were used for each polymerase chain reaction. Final concentrations of 10 μM gene specific primers (0.4 μl reverse and 0.2 μl forward), 1× Kapa SYBR green and 1 μl gDNA template were prepared to 20 μl total volume using nuclease free H₂O. Primer sets of specific genes (Table 1) were used in separate reactions which were performed in triplicate.

The quality and integrity of the isolated DNA, samples from infected and control tissues of the maize, and saprophytic fungi were subjected to polymerase chain reaction analysis with the reference genes under the following amplification conditions: for 10 min at 95°C; 35 cycles for 3 s at 95°C, for 20 s at 64°C, for 1 s at 72°C for MEPand at annealing temperature 62°C for both Ef1ɑand β-tubulin.

2.9 Statistical analysis

The statistical analysis was performed as previously described [14].

3.1 Gene specificity and qPCR assays

To our knowledge, a qPCR assay for the detection and quantification of A. flavusbiomass using extracted fungal DNA from control or infected maize tissues has not been previously reported. Since this is the first report, our discussion will be in comparison with reports for Fusariumspp. and related fungi where this assay is more commonly used.

In this study, the qPCR assay was developed to specifically detect and quantify A. flavusgDNA in maize tissues. Primers were designed, and their specificity was confirmed by testing against control and infected tissues (Figure 1). The fungal biomass in the co-infected shoots differed from the fungal biomass in the roots according to 1-way ANOVA analysis and TMCT test (P < 0.05).

Amplification of the MEP gene (203 bp) was used to detect maize DNA, while amplification of β-tubulin (118 bp) and Ef1ɑ(102 bp) were used to detect A. flavusDNA (Table 1; Figure 1). The specificity of the primer pairs was determined by conventional PCR (Figure 1) after A. flavusKSM014 infection of GAF4 and KDV1 maize lines. A. flavusDNA extracted from infected maize plant tissues, for both lines, gave an amplification product for both β-tubulin (118 bp) and Ef1a(102 bp) (Figure 1). However, there was amplification product for Ef1athan there was for to β-tubulin (Figure 1), especially in the roots. The MEPgene (203 bp) was amplified in both control and infected maize plants for both lines (Figure 1). MEPamplification was plant specific and β-Tub and Ef1ɑwere fungal specific. Based on these results, β-Tub is a better marker for detecting A. flavusin infected maize tissues than was Ef1ɑ(Figure 1) and was used for fungal biomass determination.

3.2 Colonization of plant tissues by Aspergillus flavus

Aspergillus flavusKSM014 infection of both maize lines resulted in changes in maize phenotype with the KDV1 showing more severe symptoms that GAF4 (Figure 2). After 3–14 days post infection, the infected kernels for both maize lines showed stunted growth compared to control kernels (Figure 2). Additionally, the shoots and roots exhibited minimal growth with the A. flavusfungi colonizing the kernels and this could possibly explain the reason for stunted growth or germination. The phenotypic observations suggest that KDV1 maize line grown in Makeuni is more susceptible to fungal infection (A. flavus), whereas GAF4, grown in Kisumu and Homa bay appeared more resistant to the infection (Figure 2).

The observed phenotypic characteristics were further supported by the detection and quantification of fungal biomass load in gDNA extracted from infected and control plant tissues as revealed by the qPCR assay (Figure 3).

Insignificant difference was observed in biomass of fungi between infected plant tissues for the GAF4 and the control maize line (Figure 3a). In contrast, significant differences in biomass of fungi for the KDV1 maize line was exhibited upon infection (p < 0.05) for both the shoot and root tissue (Figure 3b). Fungal gDNA level was observed to be lower in the infected GAF4 maize line tissues compared to KDV1 suggesting that GAF4 was more resistant to A. flavusKSM014 infection than KDV1 (Figure 3).

The fungal biomass of Alternaria dauciwas observed to be equivalent in two carrot cultivars between 1 and 15 days of post-inoculation, whereas it was found to be four-fold higher in the more susceptible cultivar between 21 and 25 days post-inoculation [21]. This suggests that fungal pathogens may colonize both susceptible and resistant cultivars in a similar manner during the first stages of the interaction, but fungal development is subsequently restricted in the partially resistant cultivar due to putative plant defense mechanisms [21].

It must be noted that we measured fungal biomass 14 days after infection when symptoms of the infection was phenotypically visible. However, other fungal biomass studies have shown that specific fungi could be identified even before the development of the symptoms. The presence of Colletotrichum acutatumby qPCR in strawberry leaves was detected by Debode et al. [22] 2 h post-inoculation whereas the initial symptoms of the disease appeared only after 96 h. Similarly, Fusarium langsethiaegDNA was accurately measured by Divon and Razzaghian [23] in oats independently from symptoms of the disease. These findings show the specificity and efficiency of the qPCR assay for the detection and quantification of fungal pathogens upon infection at early stages, before symptomatic appearances.

GAF4 is a Strigaspp. resistant maize line cultivated in Kisumu, Kibos, Homa Bay and some parts of Nandi, while KDV1 is an open pollinated maize variety cultivated in Makueni and the neighboring counties. The observation that KDV1 maize line as more susceptible to aflatoxigenic A. flavus(KSM014) infection could be one of the contributing factors to why Makueni and the neighboring regions are more prone to frequent aflatoxicosis outbreak and high levels of aflatoxin contamination of the maize used for consumption.

The current study relates to the previous findings on Makueni maize samples [18] where they screened the strains of A. flavusisolated from maize kernels obtained from Makueni region on CAM media and found that there was significant variation in production of blue (toxigenic) and green (atoxigenic) fluorescence by most isolates. Seventy eight percent of the isolates from Makueni were observed to produce high amounts of aflatoxin AFB1, AFB2, the most potent carcinogen compared to other regions under study [18]. Additionally, studies conducted by Probst et al. [24] in eastern Kenya, revealed a similar result where they performed culture-based methods to monitor and describe the population structures of aflatoxigenic fungi and its closely associated strains on maize kernels. Moreover, a related study by Lewis et al. [25] and Klich [26] observed that in sub-Saharan Africa, products from subsistence farmers may reach the final consumer without the appropriated monitoring, resulting in critical risks for human health.

Moreover, the current study developed a qPCR assay using A. flavusgDNA and the β-tubulin gene for the quantification of A. flavusin maize tissue. Due to its high sensitivity and specificity, qPCR has been incorporated in official protocols of the European Plant Protection Organization (http://archives.eppo.org/index.htm) for the production, certification and assessment of healthy plant materials [27, 28]. This could therefore, in future, provide a screening strategy for finding African maize cultivars that are resistant to A. flavusinfection or as an assessment of healthy maize plants. Zhao et al. [29] developed a qPCR assay for the detection of Magnaporthe poaeresistant Poa pratensis(Kentucky bluegrass turf), which typically needed 3 weeks to detect using conventional culture-based methods. Further, Montes-Borrego et al. [30] demonstrated that fungal presence can be detected earlier, enabling the selection of resistant plants even when samples are indistinguishable based on visual assessment. Lastly, the early detection of latent infections of rust on leaves of cereals was used to estimate infection levels before the appearance of the disease [2].

The genomic DNA extracted from the co-infected shoots of both maize lines showed varied concentrations of fungal biomass load compared to the roots according to analysis using 1-way ANOVA and TMCT test (p < 0.05). The quantification of Verticillium dahliaegDNA in different tomato cultivars also revealed the concentration of pathogen DNA in plant tissues increased and decreased in susceptible and resistant cultivars, respectively [31]. Similarly, significant differences were found in the amount of F. oxysporumDNA in roots of different chickpea cultivars [32], while the detection of Phomopsis sclerotioidesin pumpkin, melon, cucumber and watermelon showed that infection and rate of disease development of this polyphagous pathogen may vary according to the host [33]. In general, Vandemark and Barker [34], concluded that low levels of pathogen DNA in resistant plants is indicative of a mechanism that inhibits pathogen growth, whereas, the presence of a relatively high amount of pathogen DNA in asymptomatic plants indicates a resistance mechanism based on tolerance rather than on true resistance.