Modeling-based optimization approaches for the development of Anti-Agrobacterium tumefaciens activity using Streptomyces sp TN71. (June 2018)

• Record Type: Journal Article
• Authors: Smaoui, Slim; Ennouri, Karim; Chakchouk-Mtibaa, Ahlem; Sellem, Imen; Bouchaala, Kameleddine; Karray-Rebai, Ines; Ben Ayed, Rayda; Mathieu, Florence; Mellouli, Lotfi
• Is Part Of: Microbial pathogenesis. Volume 119(2018)
• Page Start: 19
• Page End: 27
• Publication Date: 2018-06
• Subjects: Streptomyces sp. TN71 -- Anti A. tumefaciens activity -- Medium optimization -- Response surface methodology -- Artificial neural network; Pathogenic microorganisms -- Periodicals; Pathology, Molecular -- Periodicals; Communicable Diseases -- microbiology -- Periodicals; Communicable Diseases -- parasitology -- Periodicals; Micro-organismes pathogènes -- Périodiques; Pathologie moléculaire -- Périodiques; Electronic journals; 616.9041
• DOI: 10.1016/j.micpath.2018.04.006
• Languages: English
• ISSNs: 0882-4010

Abstract

Abstract: A new aerobic bacterium TN71 was isolated from Tunisian Saharan soil and has been selected for its antimicrobial activity against phytopathogenic bacteria. Based on cellular morphology, physiological characterization and phylogenetic analysis, this isolate has been assigned as Streptomyces sp. TN71 strain. In an attempt to increase its anti- Agrobacterium tumefaciens activity, GYM + S (glucose, yeast extract, malt extract and starch) medium was selected out of five different production media and the medium composition was optimized. Plackett–Burman design (PBD) was used to select starch, malt extract and glucose as parameters having significant effects on antibacterial activity and a Box–Behnken design was applied for further optimization. The analysis revealed that the optimum concentrations for anti- A. tumefaciens activity of the tested variables were 19.49 g/L for starch, 5.06 g/L for malt extract and 2.07 g/L for glucose. Several Artificial Neural Networks (ANN): the Multilayer perceptron (MLP) and the Radial basis function (RBF) were also constructed to predict anti- A. tumefaciens activity. The comparison between experimental with predicted outputs from ANN and Response Surface Methodology (RSM) were studied. ANN model presents an improvement of 12.36% in terms of determination coefficients of anti A. tumefaciens activity. To our knowledge, this is the first work reporting the statistical versus artificial intelligence based modeling for optimization of bioactive molecules against phytopathogens. Highlights: High anti- Agrobacterium tumefaciens activity from Streptomyces sp. TN71. Starch, malt extract and glucose concentrations were optimized for A. tumefaciens activity. RSM and ANN techniques were employed for anti A. tumefaciens activity. Comparatively, ANN outperformed in terms of predictive and estimation capabilities than RSM.