RNA-protein interactions are of central biological importance as they provide the means to connect genetic information to cellular biochemistry. Work over the last decade has elucidated a number of protein motifs that are capable of sequence and/or structure specific RNA recognition. However, it is neither possible to design a protein sequence that will bind a chosen RNA target nor predict which motif would be most appropriate for a given recognition problem. The investigator has therefore chosen to pursue an in vitro genetic strategy to isolate RNA binding proteins. The in vitro strategy will be pursued using RNA-protein fusions, protein molecules covalently attached to the mRNA which encodes them. The investigator proposes to explore and compare random sequence protein libraries to two protein motifs that have the potential to act as general frameworks for RNA recognition: 1) the arginine rich (ARM) motif, and 2) a structural analog of the RNP domain, the B1 domain of protein G. The selection experiments will provide a wealth of examples that will either support or challenge the current ability to predict protein structure. Biochemical and physical analysis of these examples should provide insight into which scaffolds are the most general. In addition, the information from the selection experiments will be fed back into the design process in an iterative fashion to improve the overall design of the frameworks and selections.

The design of new proteins is one of the fundamental challenges facing biochemists. This is because it is not currently possible to design a protein with a set of rational steps analogous to the construction of a building. An alternative approach is to construct many different proteins (millions to trillions) and select the few that have desired functional properties. The immune system uses precisely this strategy for our defense, making billions of proteins based on a common framework--an antibody. The investigator proposes to examine new frameworks for their ability to recognize RNA molecules. Using newly developed techniques, the investigator's lab will examine trillions of sequences in an effort to develop general reagents to improve our understanding of protein function.

Agency
National Science Foundation (NSF)
Institute
Division of Molecular and Cellular Biosciences (MCB)
Application #
9876246
Program Officer
Parag R. Chitnis
Project Start
Project End
Budget Start
1999-02-01
Budget End
2004-01-31
Support Year
Fiscal Year
1998
Total Cost
$520,000
Indirect Cost
Name
California Institute of Technology
Department
Type
DUNS #
City
Pasadena
State
CA
Country
United States
Zip Code
91125