The Use Of Proteomics To Improve The Annotation Of Genes In Banana: Pushing The Limits Of Proteogenomics To Understand Stress and Tolerance

One attractive way to perform gene expression profiling and to understand gene function is proteome research. Higher eukaryotes - and especially plants with their flexibility towards polyploidy - tend to have several protein isoforms and post-translational modifications. This generates a lot of redundancy: all isoforms share the same sequences and differ only in a few aminoacids. Proteogenomics predicts possible gene models in a newly sequenced genome and confirms it in a biological sample using mass spectrometry. The genome of a generated double haploid banana variety has recently been published (D'Hont et al., 2012) but all almost all edible bananas are allopolyploid and have a complex genome structure with genomic rearrangements (Carpentier et al., 2011; De Langhe et al., 2010). As such, protein polymorphism analysis in allopolyploid varieties is important in understanding biodiversity and to improve gene annotation. The combination of two-dimensional electrophoresis 2D DIGE with de novo MS/MS sequence determination is still the workhorse for proteomics in non-model plants (Carpentier et al., 2008) and is an excellent tool to quantify and analyze the different protein isoforms and to identify variety specific proteins and amino acid polymorphisms (Carpentier et al., 2011). However, proteins with multiple membrane spanning domains –a class of proteins extremely important in stress reactions- are hardly detected via 2-DE and the throughput is limited (Vertommen et al., 2010, 2011). We will present here a combined peptide based shot gun /2DE approach to contribute to the further annotation of the banana genome.