Dr. Balint et al. have devised a novel solution to two related problems in protein engineering. (1) Most cDNA and random peptide expression libraries are full of sequences which do not express stable proteins. (2) Production costs for many proteins o industrial or pharmaceutical importance are prohibitive due to suboptimal expression of the recombinant protein in heterologous hosts. Their method is based on the testable hypothesis that if any single domain in a multi-domain protein fails to achieve its native fold, the entire protein will turnover and fail to accumulate. If confirmed this fact could be exploited to enrich cDNA expression libraries for stable expressors of native folds in heterologous hosts. The same principles could also be used to pre-enrich random peptide libraries (RPL) for stable expressors, and to select stable, functional, high-level expressors from mutagenic libraries of pharmaceutical or industrial proteins which are otherwise poorly expressed in heterologous hosts. If true, many library of protein domains, whether cDNA, random peptide, or mutagenic, could be expressed as a two-domain fusion via a flexible linker with a selectable marker domain, by which only those members of the library which can achieve stable folds in vivo would be selected. In Phase I they will test this concept.
NOT AVAILABLE
