Computationally designed, high specificity inhibitors delineate the roles of BCL2 family proteins in cancer

ABSTRACT: In the related work, proteins were computationally designed and experimentally optimized to bind with high affinity and specificity to each human pro-survival BCL2 homolog. Site-directed saturation mutagenesis libraries were generated based on partially-specific computational designs 2CDP06 (for Bcl-2-targeting Computationally Designed Protein), XCDP07 (Bcl-xL), WCDP03 (Bcl-w), and FCDP01 (Bfl-1), and BCDP01 (Bcl-B) which were then transformed into yeast for surface display, as described in Chao et al (Nat Protoc. 2006). Binding to labeled target homolog was detected via FACS after co-incubation with unlabeled competitors to favor specific interactions. Similarly, an SSM library was generated based on partially-specific Mcl-1-targeting design MCDP02 and was sorted for specific interactions toward each Bcl-2 proteins (data labeled MCDP02_[homolog]). DNA was harvested from naïve libraries and sorted populations and deep sequenced. The frequency of each mutation in a sorted population was compared to its frequency in the naïve population to determine enrichment ratios for all possible single amino acid substitutions (frequency calculations based on counts, which can be found in analyzed data file along with enrichment ratios). The most enriching mutations were then combined in combinatorial libraries, sorted as described above, and variants present after four or five rounds of sorting exhibited excellent specificity. Please see our corresponding publication for additional detail.

ORGANISM(S): Saccharomyces cerevisiae

PROVIDER: GSE80194 | GEO | 2016/05/09

SECONDARY ACCESSION(S): PRJNA318254

REPOSITORIES: GEO