**4. Monitoring strategies: The importance of reliable analysis as controlling guidance**

Monitoring of the corn chain in northern Paraná showed that 81 % (n = 435, crop 2003) and 98.8 % (n = 435, crop 2004) of corn was safe for human consumption, in regard to fumonisin. The decreasing trend in fumonisin contamination, when compared to previous studies [8, 9, 12], could suggest a conscious monitoring procedure at the quality control level, in accordance with the strict guidelines imposed by importing countries.

The main approaches in corn phytosanitary control involve pesticides and agricultural practices with an emphasis on tillage and crop rotation. Efforts have been focused on novel fungicides for *Fusarium* sp. control to maximize grain yield [51, 52]; however, several studies have shown that fungicide application can increase mycotoxin levels [53, 54]. The recom‐ mended dose of fludioxonil + metalaxyl-M (2.5 + 1.0 %) was insufficient to inhibit *F. verticillioides* growth *in vitro*, but it increased FB1 production from 3.5-fold to 12.5-fold (2.58 μg mL-1 when compared with 0.72 μg mL-1 in control), with an alteration in mycelial morphology [55, 56]. A scanning electron microscopy analysis showed that the fungicide caused the inhibition of hyphal growth and defects of hyphae, such as excessive septa‐ tion, cell wall disruption, and withered hyphae, and extracellular material around the hyphae was rarely observed (Figure 3) [57].

Therefore, efforts to reduce mycotoxin levels should be focused on sustainable production. In uninterrupted planting in tropical regions, non-drastic management of cropping systems using culture rotation in no-tillage areas under different fertilizations emphasizing nitrogen rate, and low cost organic waste remain concerns in the protection of grain and soil conserva‐ tion [58, 59, 60, 28].

**Figure 3.** Electron micrographs of *F. verticillioides* 103 F mycelia cultured in defined liquid media in the absence (con‐ trol) and presence (treatment) of fludioxonil + metalaxyl - M at the dose recommended by the manufacturer (1.5 μL mL-1) showing the fibrillar extracellular material present in the control cultures (A and C) and withered hyphae and disruption of cell walls in the treatments (B and D).

The effect of conventional and no-tillage cropping systems in corn cultivated in summer following either oats or fallow in winter on natural fumonisin levels (2006 and 2007 growing seasons) has been assessed [60]. No-till corn following oats showed stronger fumonisin contamination patterns than the other treatments (2006 season, *P* < 0.05). Although no-till could be advantageous from a soil conservation standpoint, it may enhance fumonisin contamination in the tropics, contrasting another report [61] that there was no significant difference between conventional and no-till in fumonisin in monoculture corn in Northern Italy. When the nitrogen fertilizer rates (0 to 90.0 kg ha-1 N) on fumonisin contamination was evaluated, higher fumonisin levels were detected in plots with lower N (≤ 22.5 kg ha-1) than ≥ 45.0 kg ha-1 N, indicating a negative correlation between fumonisin and N rates [60]. Both N stress due to deficiency and excessive rates can increase the FB1 level in corn [62].

The nitrogen-fixing potential of *Azospirillum* sp. in the rhizosphere can increase yields, reduce costs and improve the nutritional quality of corn kernels. An experiment was conducted matching the inoculums of the *Azospirillum brasilense* Ab-V5 and Ab-V6 strains in corn seeds with N doses in Northern and Central-Southern Paraná in the 2010/2011 and 2011/2012 seasons [63]. Although the seed inoculation associated with N doses showed non-significant effects on fungal count, the inoculum treated plots showed lower fumonisin levels (p < 0.05) than the non-treated plots, indicating favourable trend towards agricultural practices with inoculants. Fumonisins were detected in 90 % of samples in 2010/2011 (mean, 0.62 μg g-1) and 97.5 % in 2011/2012 (mean, 4.34 μg g-1) in the northern region, whereas its occurrence was 45 % in 2010/2011 (mean, 0.14 μg g-1) and 100 % in 2011/2012 (mean, 2.67 μg g-1) in central-southern Paraná.

The use of landfill leachate in agricultural soils as fertilizers has been suggested as an alternative for the disposal of this effluent; however, heavy metals may be a limiting factor [63]. The application of increasing doses of leachate (0 to 130.8 m3 ha-1) increased the yield, protein content, lipid and ash in corn grain, but no effect was observed on fumonisin reduction, which occurred in all samples, with 31.2 of samples with levels over the maximum tolerable limit in Brazilian guidelines (5.0 μg g-1). An increasing trend in lead content was also observed in the 2009/2010 seasons, and in sodium (2011/2012 seasons) when the leachate rate was increased [63].

The management of plant density (60 to 105 thousand plants/ha) with N doses (0 to 240 kg ha-1) showed no effect on corn fungal count, but there was an increasing trend in fumonisin levels when plant density was increased. Total fumonisins (FB1 + FB2) were detected in corn grain at levels ranging from non-detectable to 7.80 μg g-1 (mean, 1.50 μg g-1) in the 2009/2010 season, while it was non-detectable to 23.36 μg g-1 (mean, 1.72 μg g-1) in the 2010/2011 season [63].

Efforts also should be focused on the safety and quality of the wheat chain, one of major universal components in food. Although 90 % of the national crop is centred in southern Brazil, domestic consumption still depends on importation [1, 2]. Table 3 shows that natural contam‐ ination of DON in wheat was non-equally distributed among different crops and was de‐ pendent on local and climatic conditions (the impact of agricultural management practices was evaluated in 2010 and 2011 seasons). Environmental conditions can shift the metabolic route of *Fusarium graminearum*, changing the fusariotoxin profile in grains, ex., leading to an increase of acetylated analogues of DON [64, 65]. Although such acetylated trichothecenes are consid‐ ered less toxic than DON, a rapid deacetylation can take place in the digestive tract of mam‐ mals, turning into it to DON [66]. DON of the group B has been regarded as a unique trichothecene in wheat products under current Brazilian guidelines [41], although studies have indicated that group B trichothecenes, such as nivalenol (NIV) and acetylated analogues (3 acetyl-DON and 15-acetyl-DON), should be included [17].

**Figure 3.** Electron micrographs of *F. verticillioides* 103 F mycelia cultured in defined liquid media in the absence (con‐ trol) and presence (treatment) of fludioxonil + metalaxyl - M at the dose recommended by the manufacturer (1.5 μL mL-1) showing the fibrillar extracellular material present in the control cultures (A and C) and withered hyphae and

The effect of conventional and no-tillage cropping systems in corn cultivated in summer following either oats or fallow in winter on natural fumonisin levels (2006 and 2007 growing seasons) has been assessed [60]. No-till corn following oats showed stronger fumonisin contamination patterns than the other treatments (2006 season, *P* < 0.05). Although no-till could be advantageous from a soil conservation standpoint, it may enhance fumonisin contamination in the tropics, contrasting another report [61] that there was no significant difference between conventional and no-till in fumonisin in monoculture corn in Northern Italy. When the nitrogen fertilizer rates (0 to 90.0 kg ha-1 N) on fumonisin contamination was evaluated, higher fumonisin levels were detected in plots with lower N (≤ 22.5 kg ha-1) than ≥ 45.0 kg ha-1 N, indicating a negative correlation between fumonisin and N rates [60]. Both N stress due to

The nitrogen-fixing potential of *Azospirillum* sp. in the rhizosphere can increase yields, reduce costs and improve the nutritional quality of corn kernels. An experiment was conducted matching the inoculums of the *Azospirillum brasilense* Ab-V5 and Ab-V6 strains in corn seeds

deficiency and excessive rates can increase the FB1 level in corn [62].

disruption of cell walls in the treatments (B and D).

150 Food Production and Industry

In addition, lactic bacteria of the *Lactobacillus plantarum* group have shown versatile profiles concerning genes that can code for special functions such as biodegradation, absorption, and adherence to different surfaces in the vast natural microbiota in food niches [69, 68]. Lactic acid bacteria strains can be isolated from multiple wheat sources (grains, germ, bran, and flour) and have been tested against *F. graminearum* strain IAPAR 2218 (Figure 4). All tested strains demonstrated some DON-reducing potential, and the non-viable autoclaved *L. plantarum* cells (71.19 %) showed a higher effectiveness in DON reduction than viable cells (16.41 %). The probable mechanism of reduction would be the adsorption by cell walls [69, 70]. Lactic acid bacteria can also degrade a range of low molecular weight compounds, carrier compound families, and influx and efflux facilitators and mycotoxin degrading enzymes have been detected [71].

**Figure 4.** Effect of lactic acid bacteria isolated from different wheat sources (grains, germ, bran, flour) against *F. grami‐ nearum* strain IAPAR 2218. Inhibition Halo scale: (-): no inhibition; (+):1 to 5 mm; (++): 6 to 10 mm; (+++): 11 to 15 mm; and (++++): >15 mm.

Another use of naturally occurring microorganisms in the biocontrol/biodegradation of undesired natural toxins has been assessed for the reduction of cyanobacteria in drinking water. The potential of microcystin (MC) biodegradation has been tested in the following microorganisms: *Sphingosinicella microcystinivorans* (B9) isolated of the Lago Tsukui, Kanaga‐ wa-Japan; water kefir (mixture of lactic and acetic bacteria and yeast) (P4); *L. acidophilus* La-5 (P5); and yeast isolated of sugarcane (L5). The strain B9 degraded 99 % of MCs, while the strains P4, P5 and L5 degraded 44, 43 and 54 % of total MCs, respectively, after 96 h (Figure 5).

Strain B9 (*S. microcystinivorans*) showed the highest MCs degradation capacity and has been evaluated for its anti-cyanobacterial activity against 5 cyanobacteria strains, *Microcystis* sp. (C1), *Microcystis* sp. (C2), *Anabaena ucrainica* (C3), *Phormidium tenue* (C4), and *Synechocystis*(C5). After 96 h, the inhibition percentages (cellular counts) against cyanobacteria strains ranged from 41.4 to 79.3 %, while the inhibition percentages (concerning chlorophyll-a) ranged from 34.4 to 68.9 %.

probable mechanism of reduction would be the adsorption by cell walls [69, 70]. Lactic acid bacteria can also degrade a range of low molecular weight compounds, carrier compound families, and influx and efflux facilitators and mycotoxin degrading enzymes have been

**Figure 4.** Effect of lactic acid bacteria isolated from different wheat sources (grains, germ, bran, flour) against *F. grami‐ nearum* strain IAPAR 2218. Inhibition Halo scale: (-): no inhibition; (+):1 to 5 mm; (++): 6 to 10 mm; (+++): 11 to 15 mm;

Another use of naturally occurring microorganisms in the biocontrol/biodegradation of undesired natural toxins has been assessed for the reduction of cyanobacteria in drinking water. The potential of microcystin (MC) biodegradation has been tested in the following microorganisms: *Sphingosinicella microcystinivorans* (B9) isolated of the Lago Tsukui, Kanaga‐ wa-Japan; water kefir (mixture of lactic and acetic bacteria and yeast) (P4); *L. acidophilus* La-5 (P5); and yeast isolated of sugarcane (L5). The strain B9 degraded 99 % of MCs, while the strains P4, P5 and L5 degraded 44, 43 and 54 % of total MCs, respectively, after 96 h (Figure 5).

Strain B9 (*S. microcystinivorans*) showed the highest MCs degradation capacity and has been evaluated for its anti-cyanobacterial activity against 5 cyanobacteria strains, *Microcystis* sp. (C1), *Microcystis* sp. (C2), *Anabaena ucrainica* (C3), *Phormidium tenue* (C4), and *Synechocystis*(C5). After 96 h, the inhibition percentages (cellular counts) against cyanobacteria strains ranged from 41.4 to 79.3 %, while the inhibition percentages (concerning chlorophyll-a) ranged from

detected [71].

152 Food Production and Industry

and (++++): >15 mm.

34.4 to 68.9 %.

**Figure 5.** Biodegradation of microcystin by *Sphingosinicella microcystinivorans* (strain B9), water kefir (P4), *Lactobacillus acidophilus* La-5 (P5) and yeast (L5).

In summary, adequate agricultural practices based on crop rotation, fertilization, soil biodi‐ versity, resistant crops, and post-harvest management could reduce mycotoxin contamination in field. Further long-term strategies encouraging no-tillage cultivation, and the maintenance of riparian forests in extensive agricultural land would be goals to maintain water quality; and the sustainable production of nutrient-rich high-quality products would still be possible.
