Establishing the Foundation for the Characterization of Non-Model Plastid Proteomes

The focus of plastid development research in model organisms, specifically Arabidopsis thaliana and Solanum lycopersicum (Tomato), has often produced the misconception that plastid transition is identical across all tissues.  Investigation of fruit plastids using Transmission Emission Microscopy (TEM), however, has revealed an extreme diversity in plastid structure and morphology.   Such differences suggest that the composition of fruit plastid proteomes likely differ and suggest undiscovered plastid functions.  In an effort to characterize the predicted plastid proteome of non-model organisms, predicted gene sets from Vitis vinifera (Grape), Malus X Domestica Borkh. (Apple), Prunus persica (Peach), Fragaria vesca  (Woodland Strawberry),  and Populus trichocarpa (Black cottonwood) as well as the model Solanum lycopersicum were analyzed with TargetP to predict genes encoding plastid targeted proteins.  All genes predicted to encode plastid targeted proteins were clustered along with Arabidopsis plastid-targeted proteins derived from the Chloroplast 2010 project.  This led to the identification of large portions of each dataset lacking appropriate homology to any targeted sequence ranging from 9.5% in Populus trichocarpa to 20.8% in Vitis vinifera.  Functional analysis using Blast2GO reveals that a majority of these predicted proteins possess protein binding, nucleotide binding, and DNA binding functionalities.  These analyses provide for a foundation in which future studies may validate and further characterize plastid targeted proteins in non-model systems.