Screening for aerobic endospore-forming bacteria as biocontrol agents for powdery mildew disease of cucurbits.
Tredgold, Heather Rayne.
MetadataShow full item record
Powdery mildew of cucurbits costs the South African cucurbit-growing industry millions of Rands per year in reduced yields and compromised fruit quality. Amongst the many bacterial and fungal antagonists of cucurbit powdery mildew, certain aerobic endospore-forming bacteria (AEFB) species show promise as biocontrol agents of this disease. When embarking upon biocontrol agent selection, multifaceted screening strategies are crucial. A study was undertaken with the aim of isolating AEFB from the cucurbit phylloplane for evaluation as potential antagonists of cucurbit powdery mildew using various screening approaches. Three hundred and nine AEFB isolates were isolated from cucurbit leaf material sourced from eight locations in the greater Msunduzi, KZN region. Dual-culture antifungal bioassays were performed using surrogate phytopathogenic fungi Botrytis cinerea and Rhizoctonia solani in place of the obligately biotrophic Podosphaera spp.. Two PCR-based genotyping methods were used to differentiate and group 55 antifungal AEFB isolates: internal-transcribed spacer region (ITS) PCR and randomly amplified polymorphic DNA (RAPD) PCR. The RAPD-PCR distinguished greater levels of genetic polymorphisms amongst isolates than did the ITS-PCR, revealing 14 different profiles as opposed to the three obtained from ITS-PCR; with 42% of isolates associated with a single RAPD-PCR banding profile. Phylogenetic relationships between representatives of each of the RAPD-PCR fingerprint groupings were determined by sequence analysis of 16S rRNA and gyrase subunit A (gyrA) gene fragments. In each instance, several distinct clusters were discernable, though gyrA sequences displayed higher levels of strain-level sequence heterogeneity. Comparisons of both gene sequence types with reference strains from the GenBank database revealed similarities to several known plant-associated strains of AEFB, including B. amyloliquefaciens subsp. plantarum and B. subtilis. Matrix-assisted laser deionisation-desorption time-of-flight mass spectrometry (MALDI-TOF-MS) based identification of selected AEFB was evaluated by comparing spectral data from AEFB isolates with reference strains in a Bruker BDAL Biotyper database. Only three out of the 14 isolates evaluated were identified to species level with acceptable confidence levels. This poor taxonomic resolution was ascribed to a paucity of applicable reference strains in the BDAL library. Nevertheless, mass spectra profiles of each isolate allowed for the clustering of related isolates to be achieved when dendograms were created. Antifungal compounds were extracted from 14 isolates using an acid-precipitation and methanol extraction protocol. Detection and identification of lipopeptide compounds in these extracts was assessed using thin-layer chromatography (TLC) and MALDI-TOF-MS. PCR-based screening for lipopeptide production potential using selected lipopeptide gene markers (viz. surfactin, iturin, bacillomycin, and fengycin) was also evaluated for the selected 14 isolates. These isolates were found to produce multiple lipopeptide compounds; including homologues of surfactin, iturin, and fengycin. However, disparities that emerged between PCR, TLC, and MALDI-TOF-MS data suggest that some PCR primers, the ituD marker in particular, showed limited specificity amongst the AEFB strains screened. Based on the overall findings, nine isolates proceeded to in vivo screening against Podosphaera spp. using an agarised detached cotyledon assay and a biocontrol pot trial. Isolates achieving the most effective antagonism of Podosphaera spp. differed in each respective assay. Isolate cce175 provided the highest antagonism in the biocontrol pot trial, and isolate sqo279 provided the best results in the detached cotyledon assay. The impacts of inoculum preparation were assessed using isolate cce175 in a biocontrol pot trial. Treatments varied in cell growth phase and assessed cell-free supernatant, whole broth, and cell-only fractions on biocontrol efficacy compared to a Tebuconazole (430 g/l) fungicide control. None of the treatments were found to impact disease at a statistically significant level. The merits and limitations of the various screening approaches used, and issues surrounding the isolation and assessment of biocontrol efficacy in plant-associated AEFB, are discussed.