Microbiologia e inmunologia

ABSTRACT Biological control is an alternative to pesticides for protection against crop diseases. A key to progress in the field of biological control to protect maize against Fusarium verticillioides is to select in vitro the best agent to be applied in the field. This research was undertaken to evaluate the correlation between different screening methods and to suggest an adequate procedure that could be used to select bacterial agents with potential biocontrol against F. verticillioides in the maize rhizosphere. The results show an extensive Pearson correlation coefficient analysis between different screening procedures carried out for Arthrobacter spp., Azotobacter spp., Pseudomonas spp., and Bacillus spp. paired with 13 F. verticillioides strains isolated from maize endorhizosphere. The following screening methodologies were correlated: niche overlap index (NOI), indices of dominance, growth rate, lag phase, antibiosis, and fumonisin production. It was observed that NOI and antibiosis methodologies did not show any correlation. They were used to choose the best bacteria to apply under greenhouse conditions. Azotobacter armeniacus RC2 showed a NOI > 0.9 and inhibited all F. verticillioides strains assayed. Seed maize bacterization with A. armeniacus RC2 at 106 and 107 inoculum level resulted in significantly lower values of F. verticillioides counts compared to the control without bacteria at the root levels assayed. The analysis of variance indicated that there were significant interactions between two fungal repression assays. Antibiosis assays significantly correlated with greenhouse conditions (p < 0.01) at two inoculum levels tested. Therefore, the selection system adopted was adequate to choose the best bacterial biocontrol agent. The application of this methodology shows a way to outline the best biocontrol candidate.

Silage is one of the most important feed sources for bovines. Mycotoxin contamination of feedstuffs is a worldwide concern. The aim of this study was to compare mycobiota and levels of aflatoxin B₁ (AFB₁), fumonisin B₁ (FB₁), deoxynivalenol (DON), zearalenone (ZEA) and patulin (PAT) in corn trench silos and silo bags. Dry matter was higher in trench silos. Counts varied from not detected to 10⁸ CFU g⁻¹ in both trench silos and silo bags. Isolation frequencies of Aspergillus spp. and Fusarium spp. were higher in trench silos, whereas Penicillium spp. was higher in silo bags. Silo bags showed less diversity than trench silos. Strains isolated produced AFB₁, FB₁ and PAT. In trench silos, AFB₁ was the only mycotoxin detected (1-160 µg kg⁻¹). In silo bags AFB₁ levels varied from 5.8 to 47.4 µg kg⁻¹. DON was detected in two silo bag samples. When handling is adequate the reduction of mould and mycotoxin contamination in silo bags is considerable. This study will enable estimation of the mycotoxicological risk of different ensiling practices and determination of the most adequate method to minimize economic losses and reduce hazard to animal and human health.