Protein Selections and Genomics: What Next?

• Generate de novo proteins
• Functional genomics
• Engineer extant proteins
  
  

mRNA Display

Our main technology for protein selections is mRNA-display. (Roberts and Szostak 1997).

• Large diversity for protein selection techniques (>10^12)
• Covalent linkage between Protein and the RNA that encodes it allows for stability in low salt conditions
• Completely in-vitro -- no phase of selection needs to go through an organism of any type.

Can we select functional proteins from a random sequence? Yes. The example of ATP binders shows that protein aptamers that bind to ATP are about as frequent as RNA ATP aptamers (1 in 10^11 random sequence olecules). (Keefe and Szostak, 2001).

Dissociation constants of ATP binding protein selected from completely random sequence

m-RNA-display can be easily applied to genomic applications

Genomic Selections Using mRNA-Display

Proteomic Selection of CaM-binding Proteins Using mRNA-Protein Fusions

High Throughput Assay of Potential CaM-binding Proteins

Construction of a PDZ cDNA microarray

DNA chip analysis of PDZ selection

Engineering Proteins

We are interested in exploring the capabilities of polymers with nucleic-acid like properties that have more plausible routes to prebiotic synthesis than either RNA or DNA. Threose Nucleic Acid (TNA) has been suggested as a candidate because of its ability to adopt A-form geometry and base pair with RNA and DNA. To explore the functionality of TNA we are developing a system to do in vitro selections. Key to this undertaking is engineering a protein enzyme that can polymere TNA strands on DNA templates and vice versa.