Sclerotinia stem rot disease on canola in Western Australia: Understanding of the pathogen and exploring biological control agents

Sclerotinia sclerotiorum (Lib.) de Bary is fungal pathogen that attacks over 400 plant species worldwide, causing Sclerotinia stem rot (SSR), an important disease in canola (Brassica napus L.). In Australia, SSR disease is one of the major soil-borne diseases in the canola industry. Furthermore, SSR has emerged as a serious problem on canola production in Western Australia (WA) over the past few years where crop losses can be up to 40% in the worst affected crops. Current management of Sclerotinia disease mostly relies on cultural and chemical control options that often only provide partial and/or sporadic control and can be cost prohibitive. Biological control is one of the alternatives to control soil-borne fungal diseases on brassica and there is an increased interest in biological control encouraged by public awareness about issues related to the use of chemical pesticides and biological control will likely become more important in agricultural systems in the future. The research in this thesis addresses major gaps in knowledge on S. sclerotiorum on canola in WA through examination of the biological characterization of WA isolates of S. sclerotiorum, their pathogenicity on canola, and the possibility of applying biological control in the field. The research involves plant and soil sampling to collect isolates of the pathogen and potential local Biological Control Agents (BCAs); laboratory, glasshouse and field experiments to identify biological characteristics and pathogenicity of S. sclerotiorum and to test the effectiveness of potential local BCAs. One hundred and forty isolates of S. sclerotiorum were collected from WA canola growing areas between 2009 and 2014. Isolate growth on PDAA (Potato Dextrose Agar + Aureomycin) medium at 24 and 48-hours showed highly significant (P≤0.001) differences between isolates. The number of sclerotia produced by isolates over 14 days in-vitro varied from 0 to 50 sclerotia per colony with highly significant differences between isolates (P≤0.001). More than 25 isolates (17.9%) formed an average of 20 sclerotia while 15 isolates (10.7%) did not form sclerotia and only one isolate (0.7%) formed 50 sclerotia. The general colour of the mycelia of the WA isolates of S. sclerotiorum was white but variation was observed from white to grey. Pathogenicity on 10 day-old canola seedlings varied among isolates, both after 48 hours incubation in a misting chamber and after another 48 hours in a growth room. Pathogenicity of fungal pathogens is one of the indi-cators used to determine variability among isolates. S. sclerotiorum has a wide host range and many studies have shown that it is differentiated in pathogenicity. Results of mycelial agar plug inoculations on 10-day old seed-lings indicated that there was variation in pathogenicity among the isolates of S. sclerotiorum from WA and these were categorized as having low, medium or high level of pathogenicity. There were highly significant differences (P≤0.001) in seedling mortality among isolates at 48 hours after incubation in a misting chamber. In addition, there were highly significant differences (P≤0.001) in seedling mortality at 48 hours after being placed into a growth room. At 48 hours after incubation in the misting chamber, only 5 isolates of S. sclerotiorum caused 100% seedling mortality, while 26 isolates caused 0% mortality. The highest frequency of seedling mortality was 10% caused by 44 isolates. Pathogenicity was further investigated after seedlings were returned to the growth room for another 48 hours. Here 33 isolates caused 100% seedling mortality. Research on variability of fungal pathogens has been conducted for many species, including S. sclerotiorum, all over the world. One of the conventional methods used is Mycelial Compatibility Groups (MCGs). Mycelial compatibility is an indirect measure of genetic diversity in fungal populations, and can show the degree of relatedness between or within MCGs. A total of 31 WA isolates of S. sclerotiorum, representative of a range of pathogenicity levels, were chosen for MCG tests. The 31 isolates were classified into 9 my-celial compatibility groups. Fifteen potential fungal biological control agents (F-BCAs) and three potential bacterial biological control agents (B-BCAs) were isolated from canola growing regions in WA. The potential F-BCAs were grown in Petri dishes on PDAA. Mycelial colour showed wide variation, from dark green to white. Radial mycelial growth at 24 and 48 hours differed (P≤0.001) be-tween isolates. At 24 and 48 hours after incubation, isolate F-BCA9 had the highest radial mycelial growth rate with a diameter of 3.2 and 8.5 cm, re-spectively. All potential F-BCA isolates showed some capacity to inhibit the radial mycelial growth of S. sclerotiorum as well as reduce the number of sclerotia formed by the pathogen in dual culture tests in Petri dishes. There were significant differences (P≤0.001) in the magnitude of inhibition of radial mycelial growth (40 – 60%) and sclerotial formation (65 - 100% inhibition). Isolates F-BCA12 and F-BCA15 totally inhibited the formation of sclerotia by the pathogen. Colonization of sclerotia in soil indicated that sclerotia were colonized by the spores of each F-BCA and therefore all sclerotia in the presence of F-BCAs could not form any new sclerotia in Petri dishes. The colony colour of the 3 potential B-BCAs ranged from yellow to whitish yellow. Isolate B-BCA3 had the fastest colony growth rate. There was no significant difference among the potential B-BCAs in their ability to inhibit radial mycelial growth (57 - 59%) of the pathogen (P=0.934>0.05) or for-mation of sclerotia (89 - 95%) (P=0.079>0.05). Using Sanger Sequencing molecular tools (ITS regions), the F-BCAs were identified as Trichoderma atroviride (four isolates), T. gamsii (three iso-lates), T. koningiopsis (two isolates), T. longibrachiatum (two isolates), T. paraviridescens (two isolates), T. pseudokoningii (one isolate), and T. viri-descens (one isolate). Identification of B-BCAs through 16S rRNA sequencing revealed that B-BCA1 and B-BCA2 were Serratia proteamaculans while BCA3 was Ochrobactrum anthropi. Three field experiments were undertaken with canola to evaluate effi-cacy of BCAs. Experiment 1, conducted in 2014 was established as a Randomized Complete Block Design. However, no SSR disease symptoms were observed and there were no significant (P=0.2744>0.05) yield differ-ences between treatments at harvest. In field experiment 2 undertaken in 2015, a Randomized Complete Factorial Design (RCFD) was used consisting of 16 treatments with pseudo- replication inside the treatment due to limited space. Because of pseudo-replication, Restricted Maximum Likelihood (REML) analysis was used to evaluate treatments effects. There were differences (P0.05). The protein content of the untreated control treatement was significantly higher compared to the BCA treatments, but for oil content it was significantly lower compared to the other treatments. The best BCA for grain oil content in experiment 2 was B-BCA1. In the second experiment undertaken in 2015, the RCFD consisted of fifteen treatments with pseudo-replication inside the treatments. Clear symp-toms of SSR disease were apparent when the canola crop was sprayed with S. sclerotiorum. The REML analysis revealed significant effects among the BCAs (P0.001). There were also differences between time of BCA application (P0.001) between interation of BCAs and time of spraying the pathogen. Moreover, protein and oil contents based on dif-ferent flowering stage of BCA application showed that seed protein content, when the application of BCAs was made at 50% flowering stage, was signifi-cantly higher compared to application of BCAs at 30% and 10% flowering stages. The seed oil content was lowest when the BCAs were applied at 50% flowering. There were no significant differences in seed protein content be-tween the treatments (P=0.854>0.05), spraying time (P=0.850>0.05), and flowering stage (P=0.828>0.05). Furthermore, there were no significant dif-ferences in oil content between treatments (P=0.855>0.05), timing of BCA application (P=0.624>0.05) and flowering time (P=0.694>0.05). The flowering trial was undertaken to observe after how many days newly formed flowers drop their petals. Data for petal dynamics at first flow-ering, at first petal drop, and at end of flowering, in the field and glasshouse showed that first flower opening occurred in the glasshouse at 74 days after sowing (DAS), while in field plots the first flower opening occurred 2 days later. First petal drop in the glasshouse was at 78 DAS while in the field plots first petal drop was at 79 DAS. Furthermore, flowering finished in the glasshouse on 109 DAS, but in the field plots flowering finished 107 DAS. However, there was no significant difference in timing of petal dynamics be-tween glasshouse and field studies where the mean flowering period in the glasshouse was 35 days and in the field was 32 days. The results from these preliminary experiments indicate that local WA BCAs have some ability to control SSR disease in the field, however multiple trials in grower’s fields are needed in order to observe the stability of the BCAs and maintenance of their virulence against the pathogen. Further investigation is needed to determine other mechanisms of bio-control such as antibiosis and myco-parasitism as well as improving formulations and delivery systems. Untimately, biological control products should be as effec-tive as pesticides, economical, easy to use, non-toxic, environmentally safe and acceptable to regulatory agencies, growers and consumers