Fungal biodiversity on maize kernels in an insecticide evaluation trial

The European corn borer (ECB) Ostrinia nubilalis and Fusarium ear rot Fusarium spp. pose a continuous threat to maize production worldwide. There are several reports indicating that ECB damage to maize ears promotes Fusarium ear rot infection. The aim of this study was to monitor the influence of different insecticide treatments (a.i. chlorantraniliprole, indoxacarb, and chlorantraniliprole+lambda-cyhalothrin) on the ECB and fungal diversity on maize kernels in the field in a four-year trial (2013-2016). A total of 16 different fungal genera were isolated from maize kernels, and Fusarium species were confirmed to be the dominant pathogens, present in all treatments, throughout the four years of experiments. The incidence of Aspergillus spp. and Penicillium spp. was established to be low. Apart from Fusarium species, the most frequent genera were: Aspergillus spp., Mortierella spp., Mucor spp., Penicillium spp., Acremonium spp. and Rhizopus spp. chlorantraniliprole chlorantraniliprole+lambda-cyhalothrin showed higher efficacy, though not statistically significant, compared to indoxacarb, in reducing the number of ECB larvae and damage they cause. However, no direct effect on the number of isolated fungal genera has been observed in any of the three insecticide treatments.


INTRODUCTION
Maize (Zea mays L.) is arguably the most widely grown crop in the world. It is mainly used for animal feed (82%), while only 4% of the global production is destined for human consumption (ISTAT, 2018). In Serbia, maize is cultivated on approximately 1.000.000 ha with an annual production of 4.000.000 t, which makes this plant the most important crop in Serbia (FAOSTAT, 2018).
The European corn borer (ECB) -Ostrinia nubilalis and Fusarium ear rot (Fusarium spp.) are two major yield reducing factors in maize production. The maize silk-channel is the most common pathway for ear rot infection. However, Fusarium species can also infect the ear through kernel damage created by insects or birds.
It has been well documented that second generation ECB larvae can enhance fungal infections (Mencarelli et al., 2013;Blandino et al., 2015). It has also been reported that the ECB promotes Fusarium verticillioides infection of maize kernels in temperate areas (Blandino et al., 2010).
Fusarium species are capable of producing a vast number of mycotoxins (fumonisins, fusaric acid, fusaproliferin, beauvericin, moniliformin, deoxynivalenol, zearalenone, diacetoxyscirpenol, etc.), which are able to cause serious problems to both animal and human health. This makes them one of the most important maize pathogens worldwide (Leslie & Summerell, 2006;Blandino et al., 2015). In recent years, fumonisins, deoxynivalenol, zearalenone and diacetoxyscirpenol have been detected on maize kernels in Serbia (Krnjaja et al., 2012;Tančić, 2009;Jajić et al., 2007;. Considering the fact that maize is consumed both as unprocessed and processed food or feed, while mycotoxin contamination starts in the field, food quality and safety should be improved through agronomic prevention strategies. Several factors can facilitate Fusarium infection of maize ears, and injuries caused by insect feeding create suitable infection sites for the fungus (Munkvold & Desjardins, 1997). There are reports which indicate that ECB damage to maize ears is the most responsible factor for Fusarium ear rot infection caused by several Fusarium species Alma et al., 2005). It has also been documented that insecticide treatments against the ECB can also have a positive effect on maize production by decreasing the incidence of Fusarium ear rot development (Blandino et al., 2009). There is, however, much less evidence of the effects of ECB damage on the occurrence of fungi from other genera.
The aim of this study was to monitor the influence of different insecticide treatments (a.i. chlorantraniliprole, indoxacarb, and chlorantraniliprole+lambdacyhalothrin) on the ECB and fungal diversity on maize kernels in the field in a four-year trial (2013-2016).

Field trial design
The experiment was carried out on a commercial maize dent hybrid (NS 6030, NS Seme, FAO group 600) on an experimental site at Rimski Šančevi near Novi Sad, Serbia, from 2013 to 2016. Three insecticide treatments were compared to an untreated control: T1 -chlorantraniliprole (100 ml ha -1 ); T2 -indoxacarb (250 ml ha -1 ); and T3 -chlorantraniliprole +lambda cyhalothrin (200 ml ha -1 ). The experiment setup consisted of a completely randomized block design in four replicates. Each plot consisted of four rows of maize, separated from other plots by one untreated row on each side. The length of each plot was 10 m, with a spacing of 2 m between blocks.
ECB flight activity was monitored using a light trap (model RO Agrobečej, Serbia), which was active from 20:00 to 7:00 h every day during the maize growing season. The sampled specimens were removed and counted on a daily basis. The insecticides were applied during peak flight of the ECB, which usually occurred in the first week of August, using a backpack sprayer unit with a high clearance attachment with 6 nozzle booms (model 315-HCB-4) from Bellspray Inc, USA. The working height of the sprayer was manually adjustable (0.6-4.2 m) and the spray volume was 400 l ha -1 at a pressure of 200 kPa with an operation speed of 4-6 km h -1 .
Assessments of insecticide effectiveness consisted of counting the number of surviving larvae and the number of injuries on 80 ears per treatment.

Analyses of mycobiota biodiversity
Ten ears with fungal disease symptoms were randomly collected from each plot (totaling 40 ears per treatment) for biodiversity assessment. Ten kernels per ear were chosen from infected area. In total, 400 kernels per treatment were analyzed for fungal species diversity. The kernels were surface sterilized in 1% sodium hypochlorite solution for 1 minute, rinsed three times with sterile distilled water, dried and placed on 2% water agar (WA) medium amended with streptomycin (50 mg L -1 ). After 7 days of incubation at room temperature (25±2°C) in daylight, the kernels were examined and morphological identification of different species was performed according to Watanabe (2010).

Data analysis
The incidence of species (I) was calculated as: I (%) = [Number of one species colonies which occurred in a kernel sample/total number of all fungal colonies found in one kernel sample] × 100.
The effects of treatments on the average number of fungal species, number of ECB larvae and number of injuries were determined using an analysis of variance (ANOVA) and Tuckey's test. The analyses were performed in Minitab 17 (trail version).

RESULTS AND DISCUSSION
Insecticide efficacy against ECB larvae and the damage they caused were evaluated in a four-year field trial. In 2015, there were no statistically significant differences in the efficacy of insecticides, compared to control treatment (Abbott values, Table 1). This may be attributed to unfavorable weather conditions which resulted in low abundance of ECB larvae on maize kernels in all treated and control plots.
However, the years 2013, 2014 and 2016 were more favorable for pest development and the number of ECB larvae varied from 36-249 in 2013, 3-33 in 2014 and 5-46 in 2016 (Table 1). During those years, significant differences were found in the efficacy of all insecticide treatments compared to the control, and chlorantraniliprole and chlorantraniliprole+lambdacyhalothrin (T1 and T3) showed the highest efficacy (Table 1). Nevertheless, there were no statistically significant differences in the efficacies of different treatments.
Furthermore, the lowest fungal diversity was found in chlorantraniliprole (T1) treatments over all four years of the trial, while the highest diversity of up to eight genera was observed in indoxacarb (T2) treatments. Chlorantraniliprole+lambda-cyhalothrin treatments (T3) resulted in fewer fungal genera than in control plots in all years except 2013 (Table 1). Lower fungal diversity in treatments T1 and T3, compared to T2, can be attributed to the higher efficacy of those insecticides, which was confirmed by a lower number of ear injuries in those treatments in all four years of the trial (Table 1). This is consistent with data reported by Blandino et al. (2009), Saladini et al. (2008, and Scarpino et al. (2015), who showed the importance of ECB larvae as vectors of Fusarium species in Italy. However, based on average species counts, there were no statistically significant differences in fungal diversity among the treatments (Table 1).
Sixteen fungal genera in total were detected in maize seed samples during the four-year trial ( Figure  1). Fusarium species were confirmed as the dominant pathogens of maize kernels in all four years and they were present in all treatments. The incidence of Fusarium species on maize kernels was almost constant with little variation throughout the trial: 82.99% (2013), 83.55% (2014), 85.02% (2015) and 85.01% (2016). In total, 11 Fusarium species were identified on maize kernels: F. equiseti, F. graminearum, F. oxysporum, F. proliferatum, F. pseudograminearum, F. semitectum, F. solani, F. sporotrichioides, F. subglutinans, F. verticillioides, and one unidentified Fusarium species. The number of Fusarium species was more or less constant and varied from 8 to 10 depending on trial year (data not shown). As expected based on earlier research in Serbia, the dominant species were: F. verticillioides, F. graminearum and F. proliferatum. In general, the most common Fusarium species that have been isolated from maize kernels are: F. graminearum, F. verticillioides and F. subglutinans, but also F. culmorum, F. proliferatum and F. equiseti, depending on geographic location (Cotten & Munkvold, 1998;Velluti et al., 2000;Torres et al., 2001). In Serbia, those species were detected on maize kernels in a 1994-1996 study by Lević et al. (1997), and F. verticillioides was the most abundant species in those samples (63%), followed by F. subglutinans (50.6%), F. graminearum (12.2%), F. proliferatum (9.6%) and F. oxysporum (5.8%). Fusarium species which have been sporadically found in maize kernel samples in Serbia are: F. solani, F. equiseti, F. sporotrichioides, F. chlamydosporum, F. crookwellense and F. semitectum with incidences of less than 3 % during the period from 1994 to 1996 (Lević et al., 1997), and F. equiseti, F. poae, F. polyphialidicum, F. solani and F. sporotrichioides in the period from 2004to 2007(Lević et al., 2009. This is consistent with the results of this present study, in which F. equiseti, F. oxysporum, F. semitectum and F. sporotrichioides were sporadically found in maize kernel samples. Apart from the genus Fusarium, species from another 15 genera were also isolated from infected maize kernels over the period from 2013 to 2016: Acremoniella sp., Acremonium spp., Alternaria spp., Aspergillus spp., Botrytis spp., Geotrichum spp., Gliocladium spp., Hyalodendron spp., Mortierella spp., Mucor spp., Paecilomyces spp., Penicillium spp., Rhizopus spp., Staphylotrichum spp. and Trichoderma spp. (Figure 1). Species from the genera Mucor (9%) and Aspergillus (4.27%) were the most common in 2013 (excluding Fusarium), while the genus Mortierella was the most common in 2014, 2015 and 2016 with average incidence frequencies of 9.3%, 5.37% and 4.85%, respectively. Species found in all four trial years in at least one treatment were: Fusarium spp., Gliocladium spp. and Species of the genera Acremoniella, Botrytis, Geotrichum, Staphylotrichum and Trichoderma were detected in one trial year only. The species commonly found together with Fusarium spp. as accompanying species in the examined maize kernel samples were: Penicillium spp., Aspergillus spp., Mortierella spp., Mucor spp., Rhizopus spp., and Gliocladium spp. with incidences of up to 7.33%, 2.45%, 16.09%, 21.33%, 12.50% and 6.54%, respectively ( Figure 1). These accompanying species are common mycobiota on maize kernels in Serbia and worldwide and already confirmed in some previous studies (Krnjaja et al., 2004;Tančić, 2009;Mendoza et al., 2017;Xing et al., 2018;Tóth et al., 2012). The lower incidence of Aspergillus and Penicillium species in the present study is consistent with the findings of Rodríguez Páez et al. (2011), who also detected Penicilium spp. and Aspergillus spp. in maize kernels at low percentages -0.5% and 0.25%, respectively. This apparently unusual finding could be partly attributed to the high incidence of competitive Fusarium species found during all trial years. Lević et al (2013), on the other hand, observed an uncommonly high frequency and incidence of Aspergillus species on maize kernels in 2012. This fact emphasizes the complexity of the factors influencing fungal infections (weather conditions -insect occurrence -fungal development). The same authors suggest that high incidences of Aspergillus species could be attributed to extremely stressful agrometeorological conditions, i.e. high temperatures and severe drought, which occurred in 2012, while the ECB occurrence and damage were listed as a secondary factor.
In conclusion, the results of this study showed that a total of 16 fungal genera were isolated from maize kernels, while Fusarium species were confirmed as the dominant pathogens of maize kernels in all four years, and they were present in all treatments. The incidence of Aspergillus spp. and Penicillium spp. was found to be low. Apart from Fusarium species, the most commonly represented other genera were: Aspergillus spp., Mortierella spp., Mucor spp., Penicillium spp., Acremonium spp. and Rhizopus spp.
Treatments with chlorantrani liprole and chlorantraniliprole+lambda-cyhalothrin showed higher efficacy, though not statistically significant compared to indoxacarb, in reducing the number of ECB larvae and their resulting damage. However, no direct effect on the number of isolated fungal genera was observed in any of the three insecticide treatments. The relationship between insect feeding-caused damage and fungal occurrence indicates that reduction in fungal development, and consequently in mycotoxin production, is a complex problem which requires further research and an integrated pest management approach.