PHJ F VAN DEN Boogert1, PJ M Bonants1 and JH Schneider2 lDLO Research Institute for Plant Protection (IPO-DLO), PO Box 9060, NL-6700 GW Wageningen, the Netherlands; 2Institute of Sugar Beet Research, PO Box 32, NL-4600 AA Bergen op Zoom, The Netherlands

Background and objectives
Rhizoctonia solani is a soil-borne basidiomycete comprising at least 12 non-interbreeding anastomosis groups (AGs). AG 2 is extremely variable in pathogenicity. Distinct subgroups are recognized associated with specific crop species, e.g. AG 2-2 in sugar beet and AG 2t in tulip. Pathogen detection as part of soil sample analysis is a new option in disease management with perspectives for crop choice and risk assessment. Natural variation in field populations and co-inhabiting nonpathogenic AGs or subgroups in soil necessitates a discriminative detection method. For pathogenspecific detection three hurdles need to be overcome: i) development of subgroup-selective markers that cover the full range of variation, ii) efficient DNA extraction methods for complex matrices (soil, plant tissue), and iii) sensitive PCR technology for multiplication and detection of selected DNA fragments. Objectives of this study were to investigate the pathogenic and genetic variation of R. solani from sugar beet in the field, and to develop pathogen-specific markers for relevant subgroups of the pathogen.

Results and conclusions
We randomly collected sugar beet plants from fields throughout the Netherlands, and performed a structural sampling of a number of naturally developing disease patches. Below-ground root pieces were plated out on water agar and Rhizoctonia-like cultures were grown to purity. AG and AGsubgroup determination was done microscopically and by isozym analysis from pectin culture filtrates. The majority of Rhizoctonia-like isolates from healthy-looking and diseased sugar beet plants were assigned to AG 2-2 subgroup IIIB. Representatives of AG 5 and 4 occurred at low frequencies (< 1 0 %). Two thus far unknown subgroups of AG 2-2 were encountered. Selected isolates were also tested for cytoplasmatic compatibility (perfect hyphal fusion) in anastomosis tests, and for pathogenicity in bioassays. Perfect hyphal fusion was a rare (< 1 %) event within AG 2-2111B populations upon pairing isolates from distant plants. In contrast, isolates from single or neighboring plants fused perfectly at relatively high frequency. This suggests the occurrence of nonisogenic populations within single disease patches. Pathogenicity (in terms of aggressiveness) showed relatively low variation within single fields, but differed significantly between fields. Field populations high and low in aggressiveness could thus be distinguished. The genetic variation of the field isolates was assessed using amplified fragment length polymorphism (AFLP) and DNA-sequences of the internal transcribed spacers (ITS) of RDNA genes. Analogous to pathogenicity, AFLP profiles and ITS sequences showed relatively low variation within single fields, but differed significantly between fields. Subgroupspecific PCR primers were developed from DNA-sequences of the ITS of RDNA genes. Variation in ITS sequences among random chosen AG 2-2IIIB isolates matched variation in AFLP, and in pathogenicity. We were successful in selecting specific PCR primers that cover the full range of variation of AG 2-2IIIB.
Our results show that AG 2-2IIIB is the dominant Rhizoctonia pathogen in Dutch sugar beet fields. There is substantial pathogenic and genetic variation between field populations of AG 22IIIB. The high frequency of non-isogenic isolates within single disease patches suggests an operative sexual cycle as a possible source of pathogenic and genetic variation. ITS regions of RDNA offer good prospects for developing pathogen-specific primers. Subgroup-specific detection methods for R. solani and for other soil-borne fungi will be developed at IPO-DLO in the near future.