Métodos alternativos para identificação de Fusarium, quantificação de fumonisinas e controle de fungos fitopatogênicos

Abstract

Maize (Zea mays L.) is one of the most common grains in the human and animal diet and has great economic and social importance. However, this agricultural crop is very susceptible to attack by pathogenic fungi that, in addition to causing losses in production, can produce mycotoxins capable of causing serious damage to health. These pathogens and toxins are difficult to control and detection methods are quite expensive and time-consuming. Thus, a search for alternative methods of control and rapid detection and identification of fungi and mycotoxins is highly demonstrated. Thus, the present work aimed to use hyperspectral imaging technology in the near infrared and multivariate image analysis techniques to develop a method for classifying Fusarium fungi, quantifying fumonisins, as well as analyzing an antifungal activity of phenolic compounds result of sorghum grains on the development of phytopathogenic species. The use of hyperspectral images in the near infrared (HSI-NIR) combined with pattern recognition analysis, discriminant analysis of partial complements (PLS-DA) of images in order to develop a rapid method for the identification of Fusarium verticillioides and F graminearum. The validation was performed with 15 granted by Fusarium spp. The HSI-NIR technique was able to identify and distinguish fungi F. verticillioides and F. graminearum. To quantify the content of total fumonisins (B1 + B2) in maize grains, the medium spectra obtained from the HSI-NIR technique with multivariate regression were used, and a total of 51 ed. The CLAE method was used to obtain the reference results. After the pre-processing of the data, the final model using the medium spectra of HSI-NIR images presents, for the calibration set, R2 of 0.96 and RMSEC 822.45 µg.kg-1; and R2 of 0.94 and RMSEP of 890.35 µg.kg-1 for the set of users used in external validation. The deviation performance ratio (RPD) for the 4.9 model. The antifungal effect of the phenolic extract of sorghum SC319 (1330 mg GAE.g-1) on species of the genera Aspergillus, Fusarium, Penicillium, Stenocarpella, Colletotrichum and Macrophomina was evaluated, by means of MIC analysis, germination rate and growth curve. For the MIC analysis, all samples showed growth above the maximum tested concentration (5000 µg.mL-1), contrary to what was expected, there was a positive effect, that is, higher germination rates in the presence of the phenolic extract for the percentage of germinated conidia. in 4 of the 14 samples, and in the growth curve it was possible to observe that the control had a higher count of CFU's than the samples with EFS. There was no evidence of an antifungal effect of phenolic sorghum extract on toxigenic species that affect corn. However, other tests with different sorghum cultivars are suggested in order to evaluate the antifungal effect of this extract.