Purification of Bacillus amyloliquefaciens lipopeptides for postharvest disease control

[摘要] ENGLISH SUMMARY: Africa is the largest exporter of fresh fruit (by volume) in the southern hemisphere and as such, the quality of fruit must be maintained throughout the global supply chain. Several fungal pathogens contribute to significant fruit loss and wastage due to disease during postharvest storage. The use of chemical fungicides as a postharvest disease control strategy has become limiting due to negative environmental concerns (toxicity, non-specificity and non-biodegradability) while the efficient use of bacterial cells/spores as biocontrol agents is restrictive as the viability is dependent on suitable postharvest environments. An alternative and novel biocontrol strategy proposed in this study is to use antifungal lipopeptides, namely fengycin and iturin, produced by Bacillus amyloliquefaciens DSM 23117, as second-generation biocontrol agents for postharvest disease control. However, to be sufficiently effective, concentration and purification of the lipopeptides is necessary due to the low yields produced and the presence of metabolic products such as lipid and protein impurities that are not desired in the final product. The aim of the study was to develop an appropriate downstream concentration and purification programme for antifungal lipopeptides using acid precipitation, solvent extraction and macroporous adsorption as downstream unit operations.To concentrate the lipopeptides, while effecting some degree of purification, acid precipitation studies were conducted by acidification of the cell-free supernatant to pH values 1 – 4 to determine the effect of pH on the recovery and purity of fengycin and iturin. Reverse-phase high performance liquid chromatography (RP-HPLC) quantification of acid precipitate showed high recoveries of 78% and 62% for fengycin and iturin respectively at pH 2 with an optimal fengycin purity of 64% obtained at pH 3. Acid precipitation of the cell-free supernatant at pH 3 would be suitable as the initial concentration step, due to the percentage purity of fengycin and iturin.To improve the purity of fengycin and iturin above that obtained from acid precipitation, solvent extraction was the next unit operation in the purification programme. Organic solvents of varying polarity indices were screened for their efficiency in extracting fengycin and iturin. Following screening experiments, a three-stage methanol extraction and a three-stage diethyl ether-methanol extraction were conducted to further improve the purity. Lipopeptides in solvent extracts were quantified by thin layer chromatography (TLC) which showed methanol to be the best solvent which extracted 100% of lipopeptides with a purity of 74% for fengycin. A final purity of 89% and 16% for fengycin and surfactin respectively was obtained after the 3-stage methanol extraction, while final purities of 74% and 13% were obtained for fengycin and surfactin respectively, with the diethyl ether-methanol extraction. The extraction efficiency of fengycin by the organic solvents was found to be related the functional groups of the solvents, in addition to the polarity. Further analysis of the lipopeptide solvent extract by liquid chromatography electrospray ionisation mass spectroscopy (LC-ESI-MS) was conducted to identify the two bands that appeared on the plate when the solvent extracts were analysed by TLC, and which were assumed to be fengycin. LC-ESI-MS indicated inconclusive results and it could not be confirmed with certainty that both bands were indeed fengycin.Adsorption was the final unit operation used to improve the purity of fengycin and iturin. Batch adsorption experiments were designed through a central composite design (CCD) to elucidate the optimal adsorption conditions for fengycin and iturin on a polymeric macroporous resin HP-20. Batch adsorption experiments showed the lipopeptide to resin (LP/R) ratio and pH to be significant (α = 0.05) parameters in the adsorption of fengycin with an optimal LP/R of 0.5 and optimal pH of 10. Although the LP/R ratio and temperature were found to be equally significant (α = 0.05) in the adsorption of iturin, no optimal conditions could be deduced as the LP/R ratio ranges may have existed outside the experimental region and thus further optimisation studies for iturin are required.To better understand the mechanism of fengycin and iturin adsorption, adsorption kinetics and adsorption isotherm studies were conducted, and the experimental adsorption data modelled through pseudo-kinetic-order rate models and the Langmuir and Freundlich isotherm models. Equilibrium times of 22 h and 24 h were obtained for fengycin and iturin respectively, at the optimal fengycin adsorption conditions. The experimental data was individually fitted to the pseudo-first-order rate model and pseudo-second-order rate model to determine whether the rate limiting step of fengycin and iturin adsorption could be explained by physisorption or chemisorption. Neither of the pseudo-order rate models could explain the kinetics of antifungal LP adsorption on HP-20 due to poor fit, suggesting that the kinetic data obtained in this study is inconclusive and further experimental work is required.After the kinetics of fengycin and iturin was studied, adsorption isotherms were conducted to determine the saturation concentration of fengycin and iturin at a constant temperature of 43°C and pH 10. A saturation concentration of 3 g/L and adsorption percentage of 76.5 ± 0.37% was obtained for fengycin while 0.5 g/L was determined to be the saturation concentration of iturin, with a % adsorption of 54.5 ± 3.99%. The experimental data was individually fitted to the Langmuir and Freundlich isotherm models to determine which model best represented the data and thus characterised the adsorption behaviour of the lipopeptides. As with the adsorption kinetic modelling, neither of the isotherm models could explain the adsorption mechanism of fengycin and iturin on HP-20, implying that other isotherm models such as the Redlich–Peterson model may be used in further experimental work to explain the adsorptive behaviour of the antifungal lipopeptides. To determine the effectiveness of the partially purified lipopeptides and purified lipopeptide mixtures, in vitro efficacy studies were conducted on six common fungal phytopathogens that cause diseases of fruit during postharvest storage. The fungal phytopathogens were grown separately on PDA plates containing the cell-free supernatant, resolubilised acid precipitate, methanol extract and diethyl ether-methanol extract. The growth inhibition was determined after a 5-day incubation period. 12 g/L fengycin resolubilised acid precipitate was the most effective treatment with fungal growth inhibition ranging between 30 – 100%, depending on the target phytopathogen. It was discovered that A. brassicicola was the most susceptible phytopathogen to all the treatments used while A. sclerotiorum was found to be the most resistant phytopathogen of the six phytopathogens studied. It was concluded that the effectiveness of the lipopeptides in fungal growth inhibition was the lipopeptide concentration and phytopathogen species dependent.It can be concluded from this study that an effective purification programme for fengycin and iturin would entail acid precipitation at pH 3, followed by a 3-stage extraction procedure involving methanol as the solvent, a programme which would potentially produce purities of 89% and 17% for fengycin and iturin respectively, with recoveries of 100% and 53% for fengycin and iturin respectively. To date, no study has systematically investigated existing downstream unit operations for the development of an appropriate concentration and purification programme for fengycin and iturin lipopeptides produced by B. amyloliquefaciens.Recommendations for further work include the following. While the optimal batch equilibrium adsorption conditions of 3 g/L fengycin at pH 10 and 43°C would ensure maximum fengycin adsorption on HP-20, further optimisation studies for iturin adsorption are required. Refinement in the adsorption kinetic and isotherm modelling studies is also required to elucidate the mechanism of fengycin adsorption.

[发布机构] Stellenbosch University