Optimization of physicochemical parameters for lipopeptides surfactin production by Bacillus halotolerans under oat straw medium using Box-Behnken design and artificial neural networks

Abstract

The primary objective of this study was to produce lipopeptide surfactin using Bacillus halotolerans (GenBank accession: ON204239), which was isolated from the rhizosphere of pine trees in the El Baaraouia forest in Eastern Algeria. B. halotolerans demonstrated the capacity to biosynthesize surfactants when cultured on a blood agar medium, resulting in the formation of clear zones around the colonies. Antagonism tests were conducted using the culture supernatant on two liquid media: LB (Luria-Bertani) and a natural medium (OSM) with oats as the sole carbon source. The results showed significant inhibition zones (15 mm) on LB medium compared to larger zones (25 mm) on OSM after 24 h. In vivo, antagonistic activity under field conditions after 12 weeks revealed that B. halotolerans inoculated in OSM effectively enhanced and protected tomato plants against F. napiforme compared to the LB medium and the control group. The results exhibited notable differences in plant size, number of leaves, number of discolored leaves, and root mass, all favoring the OSM-inoculated plants. The study utilized a combination of Box-Behnken design within the response surface methodology (BBD-RSM), artificial neural networks (ANNs), and the genetic algorithm (GA) to optimize culture conditions and achieve maximum surfactin production. The optimal conditions for surfactin production included a substrate concentration of 7 g, incubation time of 30 h, pH of 2.31, and a temperature of 37.4 °C. Under these conditions, the antagonism test against phytopathogenic F. napiforme (GenBank accession: ON204349) using the culture supernatant revealed a significant inhibition zone of 18.5 mm after 24 h. UPLC-MS analysis of the lipopeptides produced by B. halotolerans cultured on an oat straw medium confirmed the production of various surfactin analogs. Seven analogs (C12, C13, C14, C15, C16, C17, and C18) were identified. Notably, the production of surfactin analogs by B. halotolerans in the Luria-Bertani medium yielded fewer analogs (C12, C13, C14, C15, and C16) in comparison. In summary, this study successfully optimized surfactin production conditions using BBD-RSM, ANNs, and GA, leading to enhanced antagonistic activity against F. napiforme and the identification of multiple surfactin analogs.