The advent of green-fluorescent protein (GFP) as a means to visualize proteins in living cells has begun a revolution in most fields of cellular biology. Three major drawbacks currently hamper experiments using GFP-protein tagging. First, proteins of interest, once tagged, must generally be over-expressed and this over-expression can generate novel phenotypes that disrupt the processes being studied. In addition, over-expression will disrupt protein localization. Finally, to tag a protein, a full-length cDNA must be available. We have developed a GFP-protein trap in Drosophila that is capable of generating full-length GFP-fusion proteins, at random, throughout the genome. The GFP-trap is nearly random, and independent of protein size. The expression of the resulting GFP-fusion proteins will be controlled by endogenous genomic regulatory sequences. As a result, the correct spatial and temporal expression patterns will be achieved, as well as the endogenous protein expression levels. We propose develop a database of GFP-fusion proteins encompassing more than 50% of the Drosophila genome that will dramatically facilitate experiments aimed at understanding protein dynamics in living cells, in vivo. We anticipate that this technology will enable a new generation of experiments. Protein localization, trafficking, turnover, concentration and translation will be able to be studied in an in vivo genetic system without the caveat of protein over-expression that often precludes reliable experimental interpretation. Ultimately, the database that we will develop will also enable the development of new assays for cell and developmental studies based on GFP-expression in subsets of cells, determined by the expression of single genes. Our database will also facilitate the identification of new genes that function at discrete times and places involved in processes such as cell fate and neuro-development. Finally, we anticipate that this database will enable future proteomic approaches in an in vivo model genetic organism.
