Biophysical investigations of protein consensus sequence design and other sequence-based design strategies

Proteins are versatile biomolecules used for many important medical, industrial, and biotechnological purposes. Most natural proteins are only marginally stable, which can limit their effectiveness for these uses. A major goal of protein design is to generate proteins that have high stability and maintain biological activity. Protein consensus sequence design is a promising protein design strategy that has demonstrated success in achieving this goal. In protein consensus sequence design, a “consensus” sequence for a protein family is made by combining the most frequent amino acid at each position in an alignment of naturally-occurring sequences. Since residues important for both protein stability and function are often highly conserved, stability and activity are likely to be imprinted on the consensus protein. While there are a number of cases in the literature of consensus proteins possess high stability and maintain biological activity, consensus designs have so far been limited to a small number of protein families, and little is understood about the biophysical mechanisms underlying consensus stability enhancement. This thesis explores how and why consensus sequence design works. Chapter 2 presents a consensus sequence design for the homeodomain family. The consensus homeodomain shows a significant increase in stability (~4.5 kcal mol-1) and DNA-binding affinity (~100-fold) relative to a naturally-occurring homeodomain. Chapter 3 presents consensus designs for six additional protein families that differ in size, secondary structure, and function. All consensus proteins adopt well-behaved folded structures in solution, most are more stable than their naturally-occurring counterparts, and for a subset where biological activity can be measured in vitro, expected activities are maintained. Chapter 4 presents a mutational analysis of the consensus homeodomain to understand the sequence and structural features that give rise to the consensus stability enhancement. Consensus substitutions on the protein surface contribute the most to the consensus stabilization, and stabilizing effects from favorable non-additive interactions among consensus residues are identified. Chapter 5 presents a sequence design strategy that explicitly incorporates pairwise couplings that may be missed in consensus design. Pairwise couplings are found to be destabilizing relative to position-specific sequence biases, but are important for maintaining biological activities at levels similar to naturally-occurring proteins