Verspreiding, ecologie en behandeling van rhizogene agrobacteria, oorzaak van gekke wortels bij hydroculturen

Promovendus/a: Lien Bosmans

Promotor(en): Prof. dr. ir. Hans Rediers, Prof. dr. ir. Bart Lievens, Prof. dr. ir. René De Mot

Rhizogenic Agrobacterium biovar 1, harbouring a root-inducing (Ri) plasmid, is the causative agent of hairy root disease (HRD) in the hydroponic cultivation of tomato, cucumber and eggplants. The disease is characterized by extensive root proliferation leading to strong vegetative growth and, in severe cases, substantial losses in marketable yield. The overall objective of this thesis was to investigate the ecology and management of rhizogenic Agrobacterium biovar 1. Whereas previous studies have particularly focused on the taxonomy, identification and infection process, virtually nothing is known about the genetic and phenotypic diversity among rhizogenic agrobacteria and how the disease can be prevented or controlled adequately.

In a first part of this PhD study, we assessed the phenotypic and genetic diversity of rhizogenic Agrobacterium biovar 1. First, a collection of 41 isolates was subjected to a number of phenotypic assays, revealing high phenotypic diversity among the strains tested (e.g. strong differences in biofilm formation capacity and growth characteristics; occurrence of catalase positive and catalase negative strains). Next, isolates were characterized by multilocus sequence analysis (MLSA), targeting four housekeeping genes (16S rRNA gene, recA, rpoB and trpE) and two loci from the root-inducing Ri-plasmid (part of rolB and virD2). Besides a large phenotypic diversity, remarkable genetic diversity was observed, especially for some chromosomal loci such as trpE, encoding an anthranilate synthase. In contrast, genetic diversity was lower for the plasmid borne loci, indicating that the studied chromosomal housekeeping genes and Ri-plasmid borne loci might not exhibit the same evolutionary history. A genomospecies-level identification analysis revealed that at least four genomospecies (G1, G3, G8 and G9) are associated with HRD in hydroponically grown vegetables.

In order to accurately detect, identify and quantify this plant pathogen, a SYBR Green-based quantitative real-time PCR (qPCR) assay was developed. The assay was designed based on all rolB sequences available in GenBank and validated using a large collection of both target and non-target strains, and was found to be highly specific for rhizogenic Agrobacterium biovar 1 strains. Based on a calibration curve derived from artificially contaminated water samples mimicking environmental conditions, unknown bacterial concentrations could be accurately estimated in real-life samples. The detection limit of the assay was as low as 1 cell per mililiter water.

In addition to an accurate diagnosis of HRD there is an urgent need for effective treatment(s) to inactivate agrobacteria that are present in the irrigation system, especially because rhizogenic bacteria are known to form and reside in biofilms in the irrigation circuit. Therefore, the efficacy of hydrogen peroxide (H2O2) was assessed to control Agrobacterium-containing biofilms. Hydrogen peroxide is a commonly used biocide to control plant pathogens in irrigation systems, but little or nothing was known about its efficacy to control rhizogenic agrobacteria. Using lab-scale experiments we found a huge variation between different rhizogenic Agrobacterium biovar 1 strains in EC50 values, ranging from 18.8 to >600 ppm H2O2, representing the lowest and highest concentration tested, respectively. Using pilot-scale experiments in which different H2O2 concentrations were tested, a treatment with 25 ppm H2O2 was found to be ineffective. In contrast, treatment with 50 ppm significantly affected a catalase-negative Agrobacterium population, while a catalase-positive population was only marginally affected. For the catalase-positive Agrobacterium population, a treatment of 100 ppm H2O2 was required to be effective. Finally, H2O2 treatment of HRD in two commercial greenhouses was monitored, and showed that the H2O2 concentration decreased considerably towards the end of the irrigation circuits. Moreover, a clear correlation was found between the actual concentration of H2O2 and the incidence of HRD.

As effective control of HRD may require high disinfectant dosages, that may have phytotoxic effects, there is an increasing interest in alternatives to control HRD. Moreover, several of these chemicals may be converted to unwanted by-products with human health hazards. Therefore, in a last chapter we explored the potential of beneficial bacteria as a sustainable means to control HRD. A large collection of phylogenetically diverse bacteria was screened for antagonistic activity against rhizogenic Agrobacterium biovar 1 using the agar overlay assay. Out of more than 150 strains tested, only closely related Paenibacillus strains belonging to a particular clade showed antagonistic activity, representing the species P. illinoisensis, P. pabuli, P. taichungensis, P. tundrae, P. tylopili, P. xylanexedens and P. xylanilyticus. Assessment of the spectrum of activity revealed that some strains were able to inhibit the growth of all rhizogenic agrobacteria strains tested, while others were only active against part of the collection. Preliminary characterization of the compounds involved in the antagonistic activity of two closely related Paenibacillus strains, tentatively identified as P. xylanexedens, revealed that they are water-soluble and have low molecular weight. Application of a combination of these strains in greenhouse conditions resulted in a significant reduction of HRD, indicating the great potential of these strains to control HRD.

Altogether, this PhD study greatly increased our knowledge on rhizogenic Agrobacterium biovar 1, with an important focus on genetic and phenotypic diversity, the development of reliable detection methods and the development of effective strategies for chemical and biological disease control. Eventually this study should lead to a better management of HRD in the hydroculture of tomatoes, cucumbers and eggplants.