Abstract: Few plant peptides involved in intercellular communication have been
experimentally isolated. Sequence analysis of the Arabidopsis thaliana genome
has revealed numerous transmembrane receptors predicted to bind proteinacious
ligands, emphasizing the importance of identifying peptides with signaling
function. Annotation of the Arabidopsis genome sequence has made it possible to
identify peptide-encoding genes. However, such annotational identification is
impeded because small genes are poorly predicted by gene-prediction algorithms,
thus prompting the alternative approaches described here. We initially
performed a systematic analysis of short polypeptides encoded by annotated
genes on two Arabidopsis chromosomes using SignalP to identify potentially
secreted peptides. Subsequent homology searches with selected, putatively
secreted peptides, led to the identification of a potential, large Arabidopsis
family of 34 genes. The predicted peptides are characterized by a conserved
C-terminal sequence motif and additional primary structure conservation in a
core region. The majority of these genes had not previously been annotated. A
subset of the predicted peptides show high overall sequence similarity to Rapid
Alkalinization Factor (RALF), a peptide isolated from tobacco. We therefore
refer to this peptide family as RALFL for RALF-Like. RT-PCR analysis confirmed
that several of the Arabidopsis genes are expressed and that their expression
patterns vary. The identification of a large gene family in the genome of the
model organism Arabidopsis thaliana demonstrates that a combination of
systematic analysis and homology searching can contribute to peptide
discovery.
Keywords: In silico peptide discovery, plant peptide, peptide family, peptide signaling, secreted peptide, cationic peptide, intercellular communication, Arabidopsis thaliana, model organism, rapid alkalinization factor (RALF), phytosulfokine (PSK), genome annotation, small open reading frame, homology search, proregions, RT-PCR,
