Construction of synthetic protein‐based library by combinatorial approach

Abstract

Protein‐capture reagents have been shown to be a useful tool for scientific studies and clinical applications. We have constructed a repeat protein‐based combinatorial library for further structural and functional study of membrane proteins. This library was constructed by an ordered assembly of repeat DNA fragments containing random amino acid codons at certain positions. Theoretically, it should contain a diversity of ~2 × 1022. We obtained 45 % error‐free full‐length proteins, reducing the theoretic diversity to ~1 × 1022. Although feasible diversity size is limited by a screening method, a fraction of such library is still sufficient for many applications. By DNA sequencing analysis, 48 out of 49 correct clones are unique in protein sequence; among them, 90 % individuals have >10 positions that are different with one and another. Biophysical analyses of purified proteins from four randomly selected clones suggested correct Mr values and protein folding. Thermostability study using circular dichroism spectroscopy revealed that the melting temperatures for all four proteins are > 95 °C. Taken together, the diversity and stability of this library provide a pool of candidates for further screening of capture reagents that specifically bind to a protein of interest with high affinity.This work was supported by National Institutes of Health Grants R01 GM095538 to LG.