3.5.14
A PCR/RFLP ANALYSIS OF 16S rDNA FROM PSEUDOMONAS SPP. SUPPRESSING FUNGAL DISEASES IN CEREALS


JJ BOROWICZ, M HOKEBERG, DJ VAN DER GAAG and B GERHARDSON

Plant Pathology and Biocontrol Unit, Swedish University of Agricultural Sciences, Box 7035, SE-750 07 Uppsala, Sweden

Background and objectives
One of the factors limiting broader application of biocontrol in agricultural practice is the selection of highly effective disease suppressing biocontrol agents. A specific screening to find new bacterial strains effective against diseases caused by seed-borne pathogens such as Drechslera teres, was developed in our laboratory [1]. The reliability and consistency of this biotest was confirmed for the isolate MA 342 in repeated field experiments [1]. In this study, this biotest was used to select new bacterial isolates with higher then 60% disease suppressive effect against barley net blotch, caused by D. teres. A complex characterisation of these isolates was then done by means of various identification methods and biochemical tests. A 16S rDNA/RFLP analysis with ten restriction enzymes was also performed. The 16S rDNA/RFLP patterns and other characters of these isolates were then compared to the restriction patterns and other characters of the isolate MA 342 and of 11 reference strains of fluorescent pseudomonads.

Materials and methods
All bacterial isolates were isolated from roots of wild or cultivated plants from the mountain area in Sweden. The 11 reference strains of fluorescent pseudamonads were purchased from different type culture collections.

Results and conclusions
Out of 250 Swedish mountain isolates tested, 25 have shown over 60% disease suppression against net blotch in the barley greenhouse test. In this test, none of the reference strains showed a disease suppressive effect higher than 55%. On the basis of 16S rDNA/RFLP patterns from two restriction endonucleases, Ajul and Rsal, the 25 disease suppressing isolates were classified to three major restriction pattern groups. The 16S rDNA/RFLP patterns of the Swedish mountain isolates contained in these three major groups differed from the 16S rDNA/RFLP patterns of most of the reference strains. The Aju and Rsa1 restriction patterns of only a few of the Swedish mountain isolates differed from the patterns in the three major groups. Ten of the Swedish mountain isolates, including the isolate MA 342 have shown identical AjuI-C; RsaI-A restriction pattern and their disease suppressive effect was in all cases higher than 70%. The isolates in the three major 16S rDNA/RFLP pattern groups differed also by some other characters such as secretion of proteases, starch degradation, and the RP-HPLC metabolise profile of the culture filtrate. The isolates classified to a specific 16S rDNA/RFLP pattern groups have, however, in general shown the same biochemical characters.

References
1. Hokeberg M, Gerhardson B, Johnsson L, 1997. European Journal of Plant Pathology 103, 25-33.