Cross-species EST alignments reveal novel and conserved alternative splicing events in legumes
1 Department of Plant Pathology, University of Minnesota, St. Paul, MN 55108, USA
2 Department of Genetics, Development and Cell Biology and Department of Statistics, Iowa State University, Ames, IA 50011, USA
3 Pioneer Hi-Bred International, Inc., a DuPont company, 7200 N.W. 62nd Avenue, Johnston, IA 50131, USA
BMC Plant Biology 2008, 8:17 doi:10.1186/1471-2229-8-17Published: 19 February 2008
Although originally thought to be less frequent in plants than in animals, alternative splicing (AS) is now known to be widespread in plants. Here we report the characteristics of AS in legumes, one of the largest and most important plant families, based on EST alignments to the genome sequences of Medicago truncatula (Mt) and Lotus japonicus (Lj).
Based on cognate EST alignments alone, the observed frequency of alternatively spliced genes is lower in Mt (~10%, 1,107 genes) and Lj (~3%, 92 genes) than in Arabidopsis and rice (both around 20%). However, AS frequencies are comparable in all four species if EST levels are normalized. Intron retention is the most common form of AS in all four plant species (~50%), with slightly lower frequency in legumes compared to Arabidopsis and rice. This differs notably from vertebrates, where exon skipping is most common. To uncover additional AS events, we aligned ESTs from other legume species against the Mt genome sequence. In this way, 248 additional Mt genes were predicted to be alternatively spliced. We also identified 22 AS events completely conserved in two or more plant species.
This study extends the range of plant taxa shown to have high levels of AS, confirms the importance of intron retention in plants, and demonstrates the utility of using ESTs from related species in order to identify novel and conserved AS events. The results also indicate that the frequency of AS in plants is comparable to that observed in mammals. Finally, our results highlight the importance of normalizing EST levels when estimating the frequency of alternative splicing.