Risk assessment of fungal biocontrol agents producing toxic metabolites: a case study with the bindweed biocontrol agent Stagonospora convolvuli LA39

by Désirée Boss

Abstract

Field bindweed (Convolvulus arvensis) and hedge bindweed (Calystegia sepium) are noxious perennial weeds which can only partially be controlledby chemicalor mechanical methods.
Biological control as an alternative is thereforeof increasingpublic and commercial interest. However, to ensure the safe use of biocontrol methods, strategies to assess the possible risks need to be developed. Even though there are regulations about placing biocontrol products based on microorganisms on the market and about the risk assessment of biocontrol agents (BCAs), these regulations are, especiallyin the EU, not clearly defined regarding exactly what kind of information should be included in a comprehensive risk assessment, in particular with respect to relevant metabolites. The high amountof risk assessment data demanded by the EU today is a major hurdle in the commercialization of microbial biocontrol products.
The most important aim of this thesis was to developa general strategy to assess the risks of fungal BCAs which produce toxic metabolites and to implement this strategy using the bindweed biocontrol fungus Stagonospora convolvuli LA39 and its most important toxic metabolites as a model. The strategy developed in this thesis helps to clarify what risk assessment procedures are necessary for registration and can be used as a guideline for the registration process of other fungal BCAs producing toxic metabolites. For a comprehensive risk assessment it is also necessary to gain information about spread and survival of the BCA. Another aim of this thesis was therefore to develop molecular tools to monitorLA39 after its release into the environment. A further objeetive was to optimize the application strategy and efficacy of an S. convolvuli LA39 based mycoherbicide under field conditions.
In Chapter Two a newly developed strategy for the risk evaluation of toxic metabolites is presented, by means of which the entire risk scenario is calculated for a specific case. In the process, the risks of elsinochromeA (ELA) and leptosphaerodione production by the fungus S. convolvuli LA39 were evaluated. First of all, the toxicity of the two metabolites to bacteria, protozoa, fungi, and plants was evaluated in in vitro assays. The most sensitive bacteria and fungi were already affected at 0.01-0.07 u-M ELA, whereas plants were far less sensitive. Leptosphaerodione was less toxic than ELA. Subsequently tests were carried out to determine whether ELA is present in the applied biocontrol product or in bindweeds and crop plants treated with LA39. In plants ELA was never detected and in the biocontrol product the ELA concentration was far too low to have toxic effects on even the most sensitive organisms. Worst case calculations of possible environmental toxin concentrations were made using the data obtained. It can be conclusively stated that the production of ELA by biocontrol strain LA39 does not pose a risk to the environmentor to the consumer. The results obtained in Chapter Two allowed the risks of one specific toxic metabolite produced by one specific fungal BCA to be estimated.The strategy presented is easily applicable to other BCAs. Model studies like this will make it easier to fulfil the EU's requirements for registration procedures of BCAs.
In Chapter Three a molecular identification method was developed which allows tracking of S. convolvuli LA39 after field release. This method is based on similarity coefficients using ISSR fingerprints. The spread and survivalof LA39 in the field were monitored in two experiments. First, in a release-recapture experiments defined area with natural bindweed infestation was treated with LA39 and the fungus was re-isolated from infected bindweed leaves at the end of the growing season. The frequency of recovery of LA39 decreased quickly as the distance to the inoculation plot increased.The maximum distance where an isolate identical to LA39 could be found was 4 m. Second, the environmental persistence of LA39 was assessed in a survival (overwintering) experiment. Although it was possible to reisolate the fungus from infected bindweed in the following year, the frequency of occurrence was very low. The developed ISSR fingerprinting method proved to be suitable to distinguish LA39 from resident Stagonospora spp. and therefore to track the environmental fate of this fungal strain after field application. The results strongly suggest that LA39 is a suitable BCA for Controlling bindweed becauseit has only minimal environmental impact due to its restricted mobility, poor proliferation, and poor persistence over seasons.
In Chapter Two and Three it has been demonstrated that the use of S. convolvuli LA39 as a BCA for bindweed control is free of any risks for the consumer and the environment. Consequently the last part of the thesis consisted in the further development of a biocontrol product with LA39 as the active ingredient, and the verification of its efficacy in the field. As demonstrated in Chapter Four, an LA39 based mycoherbicide efficiently controls bindweed infestations in three non-cropfield trials conducted at different locationsin two different years. The application strategy of the mycoherbicide was optimized regarding the number and date of applications and the application rate. The results suggest that spores of LA39 should be applied two to five times from the beginning of May to the beginning of July. The number of applications needs to be adapted to the size and density of bindweed infestation each year. For efficient bindweed control the mycoherbicide has to be applied at 40-160ml m" and 5 x 106 spores ml"1. Since pathogenic fungi often lose their aggressiveness upon continuous in vitro subcultivation, the influence of subcultivation on the virulence of LA39 was monitored over 30 vegetative generations. LA39 lost aggressiveness against both bindweed species between the third and the ninth vegetative generation. However, the virulence of the fungus was fully restored after a single host plant contact. Thus continuous subcultivation should be avoided when using LA39 as a BCA.
Within this thesis, a strategy for a comprehensive risk assessment for fungal BCAs producing toxic metabolites was developed in order to impact on the simplificationof the EU registration process for microbial biocontrol agents. The feasibility and suitability of this strategy was verified by applying it to the bindweed biocontrol agent S. convolvuli LA39. The results showed that an LA39 based mycoherbicide is a safe alternative to chemical herbicides regarding product toxicity and environmental impact. Moreover, LA39 is highly effective against field and hedge bindweeds, as demonstrated in several field trials, and is therefore a promising candidate for the development of a successful commercial mycoherbicide.

Get the Abstract as PDF