Lecanicillium uredinophilum, a potential biological control mycoparasite of phakopsora pachyrhizi, the soybean rust pathogen.

Abstract

Soybean [Glycine max (L.) Merr.] is the largest and most significant contributor to global oilseed production, which is driven by its high oil and protein composition. Although the crop is native to China, Brazil, the USA and Argentina are the three largest producers in the world. South Africa is the top soybean producer in Africa. Soybean production globally is affected by plant pathogens and insect pests. Plant pathogens pose the greatest and most significant threat to soybean production and yield, and the Asian soybean rust (ASR) disease caused by the biotrophic fungus Phakopsora pachyrhizi Syd. & P. Syd., is the most significant among the foliar pathogens of soybeans. ASR disease became a pandemic over two decades ago, and to date, soybean producing areas have experienced epidemics despite the widespread application of fungicides. The costs associated with fungicides for managing the ASR pathogen are substantial, with Brazil alone spending approximately US$3 billion annually. Efforts on breeding for durable ASR genetic resistance, despite significant efforts, have not been successful as the ASR pathogen quickly overcomes single gene resistance. The reliance on chemical fungicides has led to the development of pathogen resistance to these fungicides. Therefore, this study sought to isolate indigenous fungal hyperparasites of rusts, such as Lecanicillium spp., which could serve as biological control agents, as alternatives to chemical control. The dearth of current and scientifically updated information on the mechanism of action of soybean rust hyperparasites is a gap addressed in this thesis. Understanding the mechanisms of action of hyperparasites may assist in the development of effective formulations of biocontrol agents. Hyperparasitic fungi were isolated from ASR pustules on soybean. This was followed by identification of the fungi using morphological characterisation, followed by genomic DNA extraction and Sanger sequencing of the internal transcribed spacers (ITS). These identification steps classified the isolated hyperparasites as two isolates of Akanthomyces muscarium, a Lecanicillium spp. (Aphanocladium araneurum) and a Simplicillium lanosoniveum isolate. Further phylogenetic analyses could not conclusively resolve the taxonomy of the suggested species, leading to additional marker genes being added to enhance the species level identification of the isolates. Two isolates, PP2018-001 and PP2018-003, were selected on the basis of their superior pathogenicity on ASR. The Isolate PP2018-001, and another isolate of Lecanicillium (PP2018-005) obtained from a local biological control company (Andermatt- PHP) were further characterized for their identity. The multi-locus phylogenetic analysis with the additional DNA-polymerase II second subunit (RPB2), the transcription elongation factor- 1α (TEF) identified the two isolates as being Lecanicillium uredinophilum. The PP2018-001 isolate, originally isolated from wild strawberry rust pustules, showed potential as a biocontrol agent of ASR. Ten locally marketed commercial adjuvants (Break-Thru®, Bond®, Aquawet®, Designer®, Ballista®, Tronic®, Summit Super®, Wetcit®, Nufilm P® and Sprayfilm 10®) as well as seven vegetable oils (canola, macadamia, olive, peanut, organic sesame, and sunflower) were sourced from the local market, and were subjected to bioefficacy evaluations for their compatibility with the target fungi. This was measured using mycelial radial growth and colony forming units (CFU) counts of Lecanicillium isolates PP2018-001 and PP-2018-003. The compatibility of the fungi varied significantly with different adjuvants and vegetable oils, compared with the control treatments, for both fungal isolates. The vegetable oils showed consistency in enhancing both radial growth and CFU counts with an increase in concentration compared to the control treatment. All the vegetable oils can be evaluated in future glasshouse and field trials. However, amongst the commercial adjuvants there were significant differences in their effects on the radial mycelial growth and CFU counts, with a consistent decrease in fungal growth with an increase in concentration for all adjuvants, compared to the control treatment. The best commercial adjuvant was Break-Thru® at 0.01% and 0.05%. Ballista® and Bond® showed promising compatibility at 0.01%, and further studies on reduced concentrations might reveal better compatibility. Wetcit® drastically reduced radial growth and CFU counts, and inhibited growth at the 1% concentration or higher. The results of the study would be crucial in the selection of adjuvants for bio-formulations and practical applications in either controlled or field environments for these two biocontrol agents. They may also be applicable to other fungal propagules. The mechanism of action of the L. uredinophilum Isolate (PP2018-001) was investigated. Ultrastructural examinations of fungus-to-fungus interactions (P. pachyrhizi to L. uredinophilum) were done through high throughput, high-end microscopy. Investigations used confocal laser-scanning microscopy (CLSM) employing a green fluorescent protein (GFP) transformant of PP2018-001. Tracking the hyperparasite’s infection process was successful through the CLSM, which revealed the ability of L. uredinophilum to penetrate and to colonise P. pachyrhizi urediniospores. In Scanning Electron Microscopy (SEM) studies, L. uredinophilum first attached and germinated on P. pachyrhizi urediniospores, forming an intense mycelial network and coiling around the urediniospores. Furthermore, L. uredinophilum both directly penetrated urediniospores, and entered through germ pores, after which there was a loss of cellular integrity, as evidenced by the breakdown and multiple perforations of the infected urediniospores. In Transmission Electron Microscopy (TEM) studies at the cellular level, the mode of entry of L. uredinophilum into the urediniospores was confirmed and showed the growth of L. uredinophilum hyphae inside the urediniospores and the loss of cellular integrity inside infected urediniospores, compared with uninfected urediniospores. This study confirmed that L. uredinophilum employed direct mycoparasitism as its mode of action against P. pachyrhizi urediniospores. The L. uredinophilum Isolate PP2018-001 was evaluated for colonization of the P. pachyrhizi urediniospores using three conidial concentrations of L. uredinophilum in greenhouse studies. Three concentrations (1.5 x 102, 1.5 x 104 and 1.5 x 106 conidia.ml-1) of L. uredinophilum were employed for this study, which investigated the level of colonization of the urediniospores by L. uredinophilum, as well as its effect on the ASR disease severity. Colonization of urediniospores by L. uredinophilum mycoparasite was visible 3-10 days post inoculation (dpi). All treatments achieved some level of colonization of the P. pachyrhizi urediniospores, with the highest concentration achieving almost 100% colonization at 10 dpi, whilst the other two concentrations achieved approximately 45% and 34%, in order of their concentration levels. All the treatments significantly impacted ASR severity as assessed at 0 dpi and 10 dpi. The result of this study provided evidence that L. uredinophilum conidial suspension at the highest concentration could be used to stop any further development of P. pachyrhizi and could be used as a curative strategy to control ASR. A framework for evaluating biocontrol deployment in soybean fields was conducted through the physical installation of automatic weather stations (AWS), which facilitated the measurement of various climatic parameters such as relative humidity, wind speed, wind direction, air temperature, dew temperature, and wet bulb temperature in two soybean fields. This study evaluated six machine learning (ML) algorithms to predict leaf wetness (LW) and leaf wetness duration (LWD). The algorithms also estimated the degree of importance of each environmental parameter on the prediction of LW and LWD. While LW and LWD prediction models are mainly developed for disease early warning systems, they could also be used to assess the suitability of field conditions for the deployment of biological control agents. The results presented in this thesis reinforce the importance of studies on biological control in the development of an alternative control strategy for the Asian soybean rust pathogen, P. pachyrhizi. Lecanicillium uredinophilum provides a curative approach for control of ASR.