Background: Microsatellite markers are valuable in genotyping and evaluation of genetic resources (WU and TANKSLEY 1993; SAGHAI MAROOF et al. 1994; GARLAND et al. 1999: MAGUIRE et al. 1999). ITMI initially focused on the mapping of genomes in the Triticeae using RFLP markers. More recently microsatellite or SSR markers have been more widely used because of their ease of analysis. These microsatellites have mostly been isolated from enriched genomic libraries. This process is laborious and expensive. The growing availability of DNA sequence information provides new opportunities for the development of molecular markers for use in the Triticeae. The recent completion of the sequence of the rice genome provides an important source of data. The International Triticeae EST Consortium (ITEC) provides a new source of species-specific sequence information from which markers can be derived. We have investigated the identification of microsatellite and single nucleotide polymorphism (SNP) markers in EST sequences. Two main options for the development of microsatellite markers are to source them directly from the target species or to identify them based upon information from a closely related species. Research at the Centre for Plant Conservation Genetics at Southern Cross University has addressed the isolation and characterisation of microsatellite loci from many plant species. Large numbers of loci have been identified in some species by producing enriched genomic libraries (ROSSETTO et al. 1999). In some cases such as sugarcane this has required optimisation of the microsatellite enrichment technique (CORDEIRO et al. 1999). Other species, such as pine trees have required screening of enriched libraries to allow efficient recovery of microsatellite loci (SCOTT et al. 2000a). Microsatellites derived from genomic libraries have been difficult to transfer between species in many cases, although exceptions have been found in some plant groups. For example, microsatellites appear to be widely transferable within the Myrtaceae, possibly due to limited sequence divergence in this family (ROSSETTO et al. 2000). Microsatellite transfer between Pinus species has also proven to be a useful option. Transfer of microsatellites within the cereals has been more limited (reference). One advantage of microsatellites derived from ESTs is the potential that they may be more transferable between species. Recently we demonstrated the potential to transfer microsatellite derived from grape ESTs to related species (SCOTT et al. 2000b). We are currently investigating transfer of sugarcane microsatellites derived from ESTs to sorghum and other related species compared with those from the International Consortium for Sugarcane Biotechnology (ICSB).

Microsatellites: The ITEC database was searched for microsatellite sequences and the loci identified are being evaluated for wheat and barley. Preliminary results with wheat indicate low levels of polymorphism for the microsatellites in wheat sequences. This may be due to the types of cDNA libraries represented currently in the ITEC database. We have evaluated 29 SSRs from barley EST sequences. These barley SSRs have also been evaluated in wheat. Most barley SSRs performed well, in contrast to a similar set of wheat SSRs derived from ESTs.

Single nucleotide polymorphisms: Single nucleotide polymorphisms have been detected by analysis of the Genbank DNA database. Comparison of EST sequences from a cDNA library derived from the variety Alexis and the sequences in the database allowed the detection of possible SNP loci. In other cases, over 50 SNP primer pairs have been designed and used to amplify from 10 barley genotypes to reveal potential SNPs. The initial application of these results has been in the design of variety specific assays for use in identification of barley genotypes.

Molecular beacons: The molecular beacon technique was demonstrated to work effectively in the analysis of a transgene in barley (KOTA et al. 1999). This work is being extended to the application of molecular beacons to the detection of SNPs in barley. Variety specific molecular beacons are likely to have wide application in industry.

DNA analysis of grain samples has often required germination of the sample so that DNA can be extracted from the grain. We now use ground grain samples directly as a source of DNA for analysis. Microsatellite analysis of DNA from a bulk sample has been shown to detect minor contaminants. Detection of alleles for the main variety establish the identity of the sample and minor peaks corresponding to alleles from contaminating varieties can reveal contamination levels of 1% or less. This avoids the need to test large numbers of samples individually as have been required for techniques such as protein electrophoresis.

We have employed DNA fingerprinting techniques to distinguish contamination of Harrington grain by other varieties, namely Stirling, Fitzgerald and Gairdner. Unlike morphological and protein-based tests, DNA techniques rely on a biochemical which is invariant in all tissues. The advantage of this is that the assays are not influenced by crop growth conditions, or seed treatment and storage environment. Also with the huge numbers of DNA markers available, DNA-based techniques particularly those involving microsatellite markers, are rapidly becoming the most widely accepted tools for plant variety identification.

Microsatellite markers have been developed that successfully distinguish Stirling, Fitzgerald and Gairdner from Harrington. PCR with primers to a microsatellite produced a band with Fitzgerald, and a band with Stirling, which are both markedly different from the bands produced by Harrington and Gairdner. A second microsatellite can also be used to distinguish Stirling, Fitzgerald and Gairdner from Harrington. These band sizes are all highly reproducible, and comparison of both markers can be used to distinguish each of these varieties from the other three. Additional markers are available to distinguished these varieties from 22 barley cultivars tested so far.

Results from defined mixes of Stirling, Fitzgerald, and Gairdner with Harrington suggest we can detect as little as 1% contamination by Fitzgerald, 2.5 % contamination by Stirling, and 5% contamination by Gairdner. Further development is needed to produce a reliable assay for detecting lower levels of Gairdner. Initial screening of 57 samples supplied to us suggested all have less than 5% contamination with Stirling or Fitzgerald. Some samples had peaks corresponding to 2.5% contamination with Stirling. The 24 samples tested had no major contamination (>10%) with Gairdner.