Repository logo

Doctoral Degrees (Microbiology)

Permanent URI for this collectionhttps://hdl.handle.net/10413/8007

Browse

Browsing Doctoral Degrees (Microbiology) by Subject "Aerobic bacteria."

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Screening of aerobic endospore-forming bacterial isolates as candidate biocontrol agents against rhizoctonia solani.
    (2016) Hunter, Charles Haig.; Schmidt, Stefan.; Laing, Mark Delmege.; Wallis, Frederick Michael.
    Bacterial-based biocontrol of soil-borne phytopathogens has gained prominence as a promising technology for developing sustainable agricultural pest control practices. Aerobic endospore-forming bacteria are seen as potential candidates for biocontrol applications due to various ecological and physiological traits which have been shown to influence plant health and disease control. Their ability to produce endospores also provides a major commercial advantage over non spore-forming bacteria. Appropriate screening methods are central to the discovery of successful biocontrol agents and should ideally be both ecologically relevant and able to evaluate a large number of isolates. A study was therefore undertaken with the aim of establishing screening methods that facilitate the selection of aerobic endospore-forming bacteria as candidate biocontrol agents against Rhizoctonia solani, an economically important fungal pathogen exhibiting a wide host range. Aerobic endospore-forming bacteria were isolated from rhizosphere material of five crop types grown in composted pine bark medium and screened for R. solani antagonism using traditional in vitro dual-culture bioassays. Isolates exhibiting antifungal activity were then evaluated in vivo for biocontrol activity against R. solani in cucumber seedling trials. Selected isolates were evaluated further using several screening approaches including: genomic fingerprinting; characterization of, and PCR-based screening for genes involved in the biosynthesis of bioactive lipopeptide compounds; and, the use of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) as a means of rapidly screening bacterial isolates. Approximately 6% of the bacterial isolates (n=400) showed antagonism towards R. solani in vitro. Dual-culture bioassays against R. solani, Fusarium oxysporum, Botrytis cinerea and Pythium arrhenomanes revealed that the antagonistic activity amongst isolates varied considerably and was influenced by the duration of the assay. From these assays it was possible to rank isolates based on the extent and stability of the inhibitory response in vitro as well as by the spectrum of antifungal activity observed. Twenty-four isolates were selected for in vivo screening for biocontrol activity against R. solani, using susceptible cucumber plants grown under greenhouse conditions. In preliminary experiments the pathogen loading rates were shown to have a marked influence on disease severity. In experiments where R. solani was seeded in the form of colonized agar plugs, significant differences between treatments and controls were recorded and several potential biocontrol candidates were identified. A general observation was that isolates that achieved high rankings in vitro performed better in the in vivo trial than those with lesser rankings; although some exceptions were noted. These findings support the notion that fungal antagonism is an important determinant of biocontrol potential that can be used in preliminary biocontrol screening programmes. Internal-transcribed spacer region (ITS) PCR and randomly amplified polymorphic DNA (RAPD) PCR were evaluated as methods to differentiate isolates exhibiting antifungal activity in vitro. ITS-PCR distinguished three major groupings, but proved to be limited in its ability to detect inter- and intra-specific variation amongst closely related organisms. Based on 16S rRNA gene sequence analysis, two of the groups were identified as members of the “Bacillus subtilis” and “Bacillus cereus” clusters; while, the third group consisted of a single isolate identified as a strain of Brevibacillus laterosporus. RAPD-PCR revealed further levels of genetic diversity within each ITS grouping. The “Bacillus subtilis” cluster was distinguished further into four distinct groups, which based on gyrA gene fragment sequence analysis, were identified as strains of B. amyloliquefaciens subsp. plantarum and B. subtilis respectively. Sequence matches were consistent with the RAPD-PCR groupings, indicating that this method was appropriate for differentiating related isolates at the strain and possibly the sub-species level. Clonal similarities were evident for a number of strains isolated from different plant species suggesting that these may reflect populations of rhizosphere competent strains and/or plant adapted ecotypes. Strains of B. amyloliquefaciens subsp. plantarum and B. subtilis were amongst the best performers in the in vivo biocontrol seedling trial and generally performed better than the “Bacillus cereus” group of isolates. RAPD-PCR of the “Bacillus cereus” isolates showed that they exhibited greater levels of genetic heterogeneity and that the groupings detected were not consistent when different primer sets were evaluated. Genomic fingerprinting was found to provide an insight into the prevalence, distribution and possible rhizosphere competency of related strains. Liquid chromatography was used in conjunction with electrospray-ionization time-of-flight (ESI-TOF) mass spectrometry (MS) to characterize bioactive lipopeptides purified from culture supernatants of selected strains that ranked highly in the in vitro/in vivo assays. Phylogenetically related strains produced very similar lipopeptide profiles. Bacillus subtilis strains were found to produce isoforms of surfactin and fengycin. In addition to these lipopeptides, B. amyloliquefaciens subsp. plantarum strains were also found to produce isoforms of bacillomycin D or iturin A. Bacillomycin/iturin and fengycin fractions exhibited antifungal activity in vitro, whereas surfactin fractions did not. Isolates that ranked the highest in the R. solani dual-culture bioassays all produced either isoforms of bacillomycin D or iturin A. Bacillomycin D producing isolates were amongst the best performers in the in vivo biocontrol trials. Gene markers targeting the biosynthetic apparatus of the detected lipopeptide classes were then assessed for screening purposes using PCR. BACC1F/1R primers targeting the bacillomycin D synthetase C (bmyC) gene correlated well with the ESI-TOF MS findings, whereas ITUD1F/1R primers targeting the malonyl-CoA-transacylase (ituD) gene linked to iturin A biosynthesis were unable to distinguish between isolates that produced iturin or bacillomycin in culture. Disparities between some of the PCR and ESI-TOF MS results suggested that primers targeting srfA (surfactin) and fenD (fengycin) biosynthetic genes showed limited specificity amongst the strains screened. Phylogenetic comparisons of srfD and fenD gene sequences from selected strains of B. amyloliquefaciens subsp. plantarum and B. subtilis revealed that these genes clustered according to species with marked heterogeneity between clusters being evident. Using fenD gene sequence data from B. amyloliquefaciens subsp. plantarum FZB42, primers (FENG1F/1R) targeting fengycin synthetase genes of strains of B. amyloliquefaciens subsp. plantarum isolated in this study were successfully established. MALDI-TOF MS was assessed as a means of identifying isolates antagonistic to R. solani in vitro and determining their associated lipopeptide profiles. Mass spectra were obtained in the m/z range 2000 to 20000 for identification and grouping purposes and in the m/z 750 to 2500 range in order to profile lipopeptide production. The available Bruker BDal spectral library allowed for the identification of isolates to the genus level but proved to be limited for identifying environmental isolates to the species level. Extension of the library using “inhouse” mass spectra generated from isolates identified in this study significantly improved the level of isolate identification in subsequent identification runs. Cluster analysis of mass spectra allowed for the relationships between isolates to be established and provided a means of grouping closely related isolates. Strains of B. subtilis and B. amyloliquefaciens were clearly distinguished from one another and the potential for differentiating strains at the subspecies level was also shown. MALDI-TOF MS also provided a convenient means of detecting bioactive lipopeptides directly from whole cell preparations, cell extracts and crude culture filtrates. Lipopeptide profiles varied depending on taxonomic groupings. Results for isolates within the “Bacillus subtilis” group supported the earlier ESI-TOF MS findings and were found to be more reliable than PCR screening for lipopeptide synthesis genes. “Bacillus cereus” group isolates produced distinct spectral profiles with peaks that were consistent with biomarkers previously described in the literature as isoforms of the kurstakin class of lipoheptapeptides. Brevibacillus laterosporus CC-R4 yielded a unique spectral profile in the m/z 750-2000 range with mass fragments which were similar to antimicrobial compounds recently reported in the literature. Overall, MALDI-TOF MS was found to fulfil the requirement for a practical yet robust technique suitable for processing large numbers of aerobic endospore-forming bacteria for biocontrol screening. This study has shown that genomic fingerprinting and MALDI-TOF MS characterization of bacterial isolates are worthwhile additions to preliminary in vitro screening practices. They provide a level of isolate differentiation and characterization that is beneficial for selecting candidate biocontrol agents, which is not possible with traditional screening practices. Effectively, they allow traditional biocontrol screening to move away from empirically based approaches to ones which are “knowledge” based, allowing for representative groups of bacteria with specific traits to be selected for further evaluation.