PCR amplification using SSR primers: The results of PCR amplification with 15 wheat SSR primers are summarized in Table 1 and a representative amplification profile is shown in Fig.1. In hexaploid wheat, 12 of these primer pairs amplified each a single locus, while each of the three remaining primer pairs amplified two different loci located on two different chromosomes. The chromosomes carrying the above 18 different loci were distributed in all the three genomes of hexaploid wheat (Prasad et al. 2000; Varshney et al. 2000a).

It may be noted that an individual wheat SSR primer pair amplified loci in 2 to 10 of the 14 species, which also included species containing genomes other than those present in bread wheat (A, B and D). Therefore, each of the wheat SSR seems to be derived from a corresponding SSR in the presumed ancestral Triticeae genome and is conserved in several diploid and tetraploid species of Triticeae having varied genomic constitutions. In the past, a high proportion of SSRs developed. from T. aestivum and Ae. tauschii were also shown to be functional in related diploid species containing either A or B or D genome confirming their transferability and conservation across Triticeae species (Sourdille et al. 2001; Guyomarc'h et al. 2002). Such conserved wheat SSRs may be used in studies on polymorphism, genetic diversity, gene mapping and synteny conservation across different species of Triticeae.

The cases of failures of amplification of microsatellite loci in some of the species examined during the present study were classified as null alleles. It was assumed that such null alleles might have resulted either due to modification of the primer-binding site or to the loss of corresponding SSR during the course of evolution. Fourteen of the 15 microsatellite primer pairs detected null alleles in one or more (1 to 10) different species containing different genomes. Interestingly, some of the primer pairs that amplified loci on chromosomes of A/B genomes of bread wheat, were found to amplify loci in Ae. tauschii, which is the diploid progenitor of D genome. Similarly, there were SSR loci that belonged to the D genome of bread wheat, but were amplified in diploid species with an A genome (T. urartu) (Tables 1 and 2). This means that during the evolution of bread wheat, some of the SSR loci found in a particular genome of a diploid progenitor species were either lost, or carried a mutation in the primer binding site leading to the origin of a null allele. In several recent studies involving artificially synthesized and naturally occurring allopolyploids of Triticeae, it was shown that allopolyploidization either induced elimination or caused cytosine methylation of certain unique and repetitive DNA sequences (Ozkan et al. 2001; Shaked et al. 2001). However, in studies conducted in hexaploid wheat and related species, Southern hybridization with probes carrying sequences corresponding to SSR primers gave a positive signal in related species having null alleles. This suggested that the locus specificity of SSRs in bread wheat probably originated due to mutations in primer binding sites rather than due to loss of SSRs themselves in related genomes during polyploidization (Guyomarc'h et al. 2002).

The average number of alleles per locus in 14 species was 6.6 with a range of 4 to 11 (Table 2). The maximum number of 11 alleles (120 bp to 177 bp) was observed at Xwmc25-2D carrying (GT)n. The polymorphic information content (PIC) varied from 0.386 for WMC44 to 0.780 for WMC149 (Table 2). The average number of alleles in the present study was slightly lower than an average of 7.5 alleles per locus reported by us within bread wheat in an earlier study, where 20 WMC SSRs were tried on 55 genotypes (Prasad et al. 2000). We believe that fewer alleles per locus in the present study could be due to small sample, so that a bigger sample having many more species each represented by several accessions should resolve many more alleles on each SSR locus.

In the present study no single primer pair was adequate to discriminate all the 14 species studied. However, two different pairs of SSRs, one consisting of WMC243 and WMC415, and the other consisting of WMC35 and WMC404, each discriminated all the 14 species. In our earlier study on bread wheat it was shown that 12 microsatellite markers discriminated. 48 genotypes out of 55 genotypes studied (Prasad et al. 2000). The present study thus suggested that in addition to their use in discriminating accessions belonging to a particular species like bread wheat, SSR markers can also be used for discriminating different species of Triticeae.