Biography

Frances Arnold’s research group studies how proteins evolve and develops methods of directed evolution, which they apply to engineering novel enzymes, biosynthetic pathways, and synthetic genetic regulatory circuits. Frances received her B.S. in Mechanical and Aerospace Engineering from Princeton University in 1979 and her Ph.D. in Chemical Engineering from the University of California, Berkeley, in 1985. Following postdoctoral research in chemistry at UC Berkeley and the California Institute of Technology, she joined the faculty of Caltech's Division of Chemistry and Chemical Engineering in 1987. Dr. Arnold has co-authored more than 200 publications and edited several books on protein engineering and laboratory protein evolution. A member of the U.S. National Academy of Engineering and the Institute of Medicine, she serves on the Advisory Panel to the David and Lucile Packard Foundation’s Fellowship program and the science advisory boards of several corporations. Her recent awards include the 2007 FASEB Excellence in Science Award and the 2005 Olin-Garvan Medal of the American Chemical Society. She has more than 25 patents issued or pending.
 

Frances Arnold, Ph.D.

Dick and Barbara Dickinson Professor of Chemical Engineering and Biochemistry
California Institute of Technology

www.che.caltech.edu/faculty/arnold_f/index.html
www.che.caltech.edu/groups/fha/index.html

Lessons from Synthetic Protein Families

Abstract:
We are investigating ways to emulate evolution in the laboratory in order to create new proteins with desirable properties. This approach circumvents our profound ignorance of how the amino acid sequence encodes protein function and exploits the ability of biological systems to evolve and adapt. Here I will describe recent efforts to accelerate the discovery of novel proteins through a combination of evolutionary and computational design approaches. I will also discuss what we can learn from the resulting protein sequences and the functions they encode.

We have developed computational tools to identify elements of sequence and structure that can be swapped among related proteins while minimizing structural disruption (SCHEMA). Structure-guided SCHEMA recombination of homologous proteins generates diverse sequences which still have a high probability of retaining the parental fold and function. We have used this approach to construct synthetic families of beta-lactamases and cytochrome P450 heme domains which differ from the parents by many dozens of amino acid substitutions on average. Analysis of these laboratory-generated protein families provides new insights into what it takes to make stable, functional enzymes, free from many of the filtering effects of natural selection. Unlike datasets of natural protein sequences, those generated by high throughput sequencing and functional analysis of the laboratory-generated proteins include sequences with nonnatural functions (e.g. not-folded and not-functional sequences, particularly useful for testing folding/function predictions) that explore what is physically possible within a given protein family.