A full-length enriched cDNA library and expressed sequence tag analysis of the parasitic weed, Striga hermonthica

BACKGROUND: The obligate parasitic plant witchweed (Striga hermonthica) infects major cereal crops such as sorghum, maize, and millet, and is the most devastating weed pest in Africa. An understanding of the nature of its parasitism would contribute to the development of more sophisticated management methods. However, the molecular and genomic resources currently available for the study of S. hermonthica are limited. RESULTS: We constructed a full-length enriched cDNA library of S. hermonthica, sequenced 37,710 clones from the library, and obtained 67,814 expressed sequence tag (EST) sequences. The ESTs were assembled into 17,317 unigenes that included 10,319 contigs and 6,818 singletons. The S. hermonthica unigene dataset was subjected to a comparative analysis with other plant genomes or ESTs. Approximately 80% of the unigenes have homologs in other dicotyledonous plants including Arabidopsis, poplar, and grape. We found that 589 unigenes are conserved in the hemiparasitic Triphysaria species but not in other plant species. These are good candidates for genes specifically involved in plant parasitism. Furthermore, we found 1,445 putative simple sequence repeats (SSRs) in the S. hermonthica unigene dataset. We tested 64 pairs of PCR primers flanking the SSRs to develop genetic markers for the detection of polymorphisms. Most primer sets amplified polymorphicbands from individual plants collected at a single location, indicating high genetic diversity in S. hermonthica. We selected 10 primer pairs to analyze S. hermonthica harvested in the field from different host species and geographic locations. A clustering analysis suggests that genetic distances are not correlated with host specificity. CONCLUSIONS: Our data provide the first extensive set of molecular resources for studying S. hermonthica, and include EST sequences, a comparative analysis with other plant genomes, and useful genetic markers. All the data are stored in a web-based database and freely available. These resources will be useful for genome annotation, gene discovery, functional analysis, molecular breeding, epidemiological studies, and studies of plant evolution.