In addition to the draft of the human genome sequence, the genome sequences of an increasing number of model organisms are now available. This sequence information is expected to revolutionize the way biological questions can be addressed. Molecular mechanisms should now be approachable on a more global scale in the context of (nearly) complete sets of genes, rather than by analyzing genes individually. However most protein-encoding open reading frames (ORFs) predicted from these sequencing projects have remained completely uncharacterized at the functional level. For example, out of 19,000 ORFs predicted from the C. elegans genome sequence, the function of approximately 1,200 has been experimentally characterized during the last 30 years. Functional genomics and proteomics address this limitation through the simultaneous annotation of large numbers of predicted ORFs. Despite the urgent need for large-scale functional annotation projects, functional genomics approaches have remained relatively undeveloped in multicellular organisms, primarily because of the lack of suitable methods to clone large numbers of protein-encoding ORFs into many different expression vectors. Indeed, most strategies developed in these projects are based upon the expression of large numbers of proteins in exogenous settings and in fusion with relevant tags. In order to facilitate these different proteome-wide projects, a complete set of ORFs (or "ORFeome") will need to be cloned multiple times into many different expression vectors for each model organism of interest. To achieve this goal, one solution is to clone an ORFeome of interest once and for all in a "resource" vector allowing a convenient transfer to various expression vectors. To clone the C. elegans ORFeome into various expression vectors, we use a recombination cloning technique referred to as Gateway. This technique allows both the initial cloning of ORFs and their subsequent transfer into different expression vectors by site-specific recombination in vitro. We have now finished the first part of the C. elegans ORFeome project which was to attempt to clone the ~19,000 predicted ORFs. We will present the success rate in cloning of the ORFs and the overall quality of the ORFeome to date. We will also describe how the ORFeome was used as a new approach to construct a ~100% normalized yeast two-hybrid library. Finally we will show how we could transfer thousands of ORFs from the resource clones into a dozen different expression vectors for uses in large-scale functional genomic and proteomic projects such as gene inactivation by RNAi, protein interaction mapping by yeast two-hybrid, protein production for structural genomics etc.

Authors: Jerome Reboul, Philippe Vaglio, Jean-Francois Rual, Philippe Lamesch, Laurent Jacotot, Monica Martinez, Christopher Armstrong, Nicolas Bertin, Troy Moore, Jim Hartley, Hongmei Lee, Lynn Doucette-Stamm, David E Hill, Marc Vidal, Mike Brasch