3.5.3
THE USE OF DISRUPTED MUTANTS AND THE GFP ENCODING GENE TO STUDY THE BIOCONTROL MECHANISM OF TRICHODERMA

M LORITO1, SL WOO1, B DONZELLI2, RL MACH3, S ZEILINGER3 GE HARMAN2, CP KUBICEK3 and F SCALA1

1Dipartimento di Arboricoltura, Botanica e Patologia Vegetale - sezione Patologia Vegetale, UniversitÓ degli Studi di Napoli ôFederico IIö, 80055 Portici (NA), Italy; 2Department of Horticultural Sciences and Plant Pathology, Cornell University-NYSAES, Geneva, NY, USA; 3TU Wien, Institute of Biochemical Technology and Microbiology, Section Microbial Biochemistry, A-1060 Wien, Austria.

Background and objectives
Biocontrol strains of the genus Trichoderma have been studied for more than 60 years to develop biological tools as an alternative or a complement to chemical fungicides for the control of plant diseases. However, full scale application of these fungal biocontrol agents has been delayed also because there is a little knowledge about their antifungal mechanism in vivo. For instance, lytic enzymes such as the endochitinase are considered to be involved in the mycoparasitic ability of Trichoderma, but a definitive proof that they play a major role in biocontrol is still lacking. Further, little data is available on the mechanism of induction of the genes involved in mycoparasitism and on the behavior of these fungi in the soil or on the phyllosphere during the interaction with the plant and other microorganisms. The objectives of our work are 1) to disrupt genes encoding for chitinolytic enzymes in Trichoderma biocontrol strains, characterize the resulting mutants and compare their biocontrol activity with that of the wild type; and 2) to study gene induction during biocontrol and monitor the colonization of the various habitats, by using the GFP (green fluorescent protein) gene as a reporter system under the control of both constitutive and mycoparasitism-inducible promoters.

Results and conclusions
We obtained mutants of T. harzianum strain P1 having the endochitinase gene and the corresponding enzyme activity impaired and other mutants with both the endochitinase (ThEn-42) and the N-acetyl-▀-glucosaminidase (nag1) genes altered. These mutants were characterized genetically, biochemically and biologically in comparison with the wild type and with another mutant in which the endochitinase activity was enhanced. Except for the endochitinase and the N-acetyl-▀-glucosaminidase, other enzyme activities such as chitobiosidases and glucanases were not affected. The strong antifungal activity towards Botrytis cinerea of the culture filtrates was slightly affected by the absence of the endochitinase activity in the mutants ThEn-42-. In contrast, when the endochitinase was altered and the N-acetyl-▀-glucosaminidase was lacking, a very little antifungal activity was observed in the culture filtrates of the mutants ThEn-42- nag1-. Experiments carried out in the soil also indicated that when at least two genes with different chitinolytic functions are altered, biocontrol may be substantially affected. Mutants expressing the GFP gene with the pki constitutive promoter or the mycoparasitic promoters of the ThEn-42 [1] or nag1 [2] gene were obtained and used in vitro and in vivo to study the Trichoderma-plant and the Trichoderma-host interactions, and to monitor the colonization of the soil and the rhizosphere by the biocontrol agent. The green fluorescent Trichoderma strains were detected and recovered from roots of several plant species and from soil. We observed different levels and modes of root colonization depending on the particular plant and strain tested, although extensive root colonization was usually detected in vitro. In addition, strains expressing the GFP gene under the control of chitinase gene promoters inducible by the presence of the host, were used to isolate compounds enzymatically released from the host cell walls and capable of inducing the mycoparasitic and biocontrol activity of T. harzianum. Our results demonstrated the occurrence of a complex induction mechanism modulated by the host and the usefulness of a non-disruptive reporter system to study the interactions of plants, pathogens and antagonistic fungi.

References
1. Lorito M, Peterbauer C, Sposato P, Mach RL, Strauss J, Kubicek CP, 1996. Proceedings of the National Academy of Sciences of USA 93, 14868-14872.
2. Peterbauer CK, Lorito M, Hayes CK, Harman GE, Kubicek CP, 1996. Current Genetics 30, 325-331.