PEDS Advance Access published online on October 4, 2008
Protein Engineering Design and Selection, doi:10.1093/protein/gzn050
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Protein engineering of improved prolyl endopeptidases for celiac sprue therapy
1Department of Chemical Engineering, 2Department of Chemistry and, 3Department of Biochemistry, Stanford University, Stanford, CA 94305 and 4DNA2.0 Inc., Menlo Park, CA, USA
5 To whom correspondence should be addressed. E-mail: khosla{at}stanford.edu
Due to their unique ability to cleave immunotoxic gluten peptides endoproteolytically, prolyl endopeptidases (PEPs) are attractive oral therapeutic candidates for protecting celiac sprue patients from the toxic effects of dietary gluten. Enhancing the activity and stability of PEPs under gastric conditions (low pH, high pepsin concentration) is a challenge for protein engineers. Using a combination of sequence- and structure-based approaches together with machine learning algorithms, we have identified improved variants of the Sphingomonas capsulata PEP, a target of clinical relevance. Through two rounds of iterative mutagenesis and analysis, variants with as much as 20% enhanced specific activity at pH 4.5 and 200-fold greater resistance to pepsin were identified. Our results vividly reinforce the concept that conservative changes in proteins, especially in hydrophobic residues within tightly packed regions, can profoundly influence protein structure and function in ways that are difficult to predict entirely from first principles and must therefore be optimized through iterative design and analytical cycles. Incubation with whole wheat bread under simulated gastric conditions also suggests that some variants have pharmacologically significant improvements in gluten detoxification activity.
Keywords: celiac sprue/prolyl endopeptidase
Received July 15, 2008; revised August 25, 2008; accepted August 25, 2008.