THE VAST MAJORITY OF METABOLIC FUNCTIONS IN CONTEMPORARY BIOLOGY IS MADE POSSIBLE ONLY BY THE CATALYTIC AND REGULATORY ACTIVITIES OF PROTEINS. THE CORRESPONDING DIVERSITY OF PROTEIN FUNCTIONS THAT EVOLVED OVER BILLIONS OF YEARS IS SIMPLY ASTOUNDING. WHILE THE RESULT OF THIS EVOLUTIONARY PROCESS CAN BE STUDIED THROUGH ANALYZING MODERN PROTEINS THE MECHANISMS THAT HAVE LIKELY LED TO THE EMERGENCE OF THIS ABUNDANT VARIETY OF PROTEIN STRUCTURES AND FUNCTIONS ARE STILL POORLY UNDERSTOOD. WE WILL EXPERIMENTALLY STUDY THE PROCESS OF HOW PROTEINS CAN ACQUIRE NOVEL FUNCTIONS. WE WILL ADDRESS THE TWO KEY ASPECTS OF THIS FUNDAMENTAL BIOLOGICAL QUESTION: (I) HOW CAN AN EXISTING PROTEIN EVOLVE TO ADOPT NOVEL FUNCTIONS? (II) HOW COULD THE EARLIEST FUNCTIONAL PROTEINS HAVE FIRST APPEARED ON THE STAGE OF DARWINIAN EVOLUTION? WE ANTICIPATE TO DEMONSTRATE POTENTIAL SCENARIOS FOR THE EMERGENCE OF FUNCTIONAL PROTEINS THAT INTERFACE WITH LIVING ORGANISMS. IN OUR APPROACH WE WILL SCREEN COMBINATORIAL LIBRARIES OF TENS OF MILLIONS OF RANDOMIZED PROTEINS FOR VARIANTS WITH NEW FUNCTIONS USING AN IN VIVO SELECTION TECHNIQUE. WE WILL EVALUATE AND COMPARE TWO VERY DIFFERENT PROTEIN LIBRARIES FOR THEIR CAPACITY TO YIELD NOVEL PROTEINS THAT ENABLE BACTERIAL CELL GROWTH THROUGH NEW ENZYMATIC OR REGULATORY ACTIVITIES. OUR SEARCH WILL BEGIN WITH NATURE'S MOST COMMON ENZYME FOLD THE (ALPHA/BETA)8 BARREL. THIS UBIQUITOUS FOLD IS ADOPTED BY ABOUT 10% OF ALL KNOWN ENZYME STRUCTURES UTILIZED IN FIVE OF THE SIX ENZYMATIC CLASSES AND CATALYZES A WIDE ARRAY OF DIFFERENT REACTIONS. DESPITE ITS DOMINANT PRESENCE IN BIOLOGY IT IS NOT FULLY UNDERSTOOD HOW NATURE IS ABLE TO SO READILY REPURPOSE THIS FOLD. WE ENGINEERED A LIBRARY CONTAINING TRILLIONS OF RANDOMIZED YET SOLUBLE (ALPHA/BETA)8 BARREL VARIANTS. A LIBRARY WITH SUCH A HIGH COMPLEXITY AND QUALITY IS UNPRECEDENTED. CONSIDERING THE UNIVERSAL ROLE OF THIS FOLD AMONG NATURAL ENZYMES WE ARE POISED TO SCREEN OUR LIBRARY FOR EMERGING DE NOVO CATALYTIC ACTIVITIES. OUR SECOND LIBRARY CONSISTS OF ENTIRELY RANDOM POLYPEPTIDES OF 80 AMINO ACIDS IN LENGTH. THIS LIBRARY WILL BE USED TO EMULATE THE CONCEIVABLE SCENARIO IN WHICH THE EARLIEST FUNCTIONAL PROTEINS ORIGINATED FROM MIXTURES OF RANDOM-SEQUENCE POLYPEPTIDES. WHILE THIS IS A PLAUSIBLE SCENARIO EXPERIMENTAL DATA IN SUPPORT ARE SURPRISINGLY SPARSE. BOTH LIBRARIES WILL BE SCREENED FOR FUNCTIONAL PROTEINS USING A BACTERIAL GROWTH SELECTION. THIS TECHNIQUE EMPLOYS STRAINS OF THE MODEL ORGANISM ESCHERICHIA COLI WITH SINGLE-GENE DELETIONS RENDERING THEM CONDITIONALLY AUXOTROPHIC - CAPABLE OF GROWTH ON NUTRIENT RICH MEDIUM AND INCAPABLE OF GROWTH ON MINIMAL MEDIUM. THESE AUXOTROPHIC STRAINS ARE DEFICIENT IN A FUNCTION ESSENTIAL FOR CORE METABOLISM SUCH AS A GENE NECESSARY FOR AMINO ACID NUCLEOTIDE OR COFACTOR BIOSYNTHESIS. EACH OF THESE STRAINS WILL BE TRANSFORMED WITH OUR PROTEIN LIBRARIES. GROWTH OF SELECT CLONES ON MINIMAL MEDIUM WILL IDENTIFY PROTEIN VARIANTS WHOSE FUNCTION RESCUED THE WILD TYPE PHENOTYPE. THIS NEW PROTEIN FUNCTION COULD EITHER EFFECTUATE REGULATORY CHANGES IN THE METABOLIC NETWORK OR EVEN PROVIDE A NEW CATALYTIC ACTIVITY. THIS METHOD ALLOWS TO TEST FOR ABOUT 100 DIFFERENT BIOLOGICAL FUNCTIONS. IN A PILOT STUDY WE HAVE ALREADY IDENTIFIED ARTIFICIAL PROTEINS FROM BOTH LIBRARIES THAT COMPENSATE FOR A MISSING ENZYME.