Mapping binding energetics

Although most phage display studies have focused on either improving natural protein functions or developing novel protein functions, the method can also be used to understand the molecular basis of protein structure and function. Libraries with restricted diversity have proven to be highly effective for studying protein function, because reducing the chemical complexity of the library greatly simplifies the identification of key interactions within a binding interface (Morrison and Weiss 2001; Avrantinis and Weiss 2005; Runyon et al. 2007; Zhang et al. 2007).

Shotgun alanine scanning (described in more detail in Chapter 4) was developed as a high-throughput, phage display version of traditional alanine scanning for mapping binding energetics at protein binding interfaces (Weiss et al. 2000). Libraries are constructed using binary degenerate codons that ideally encode for only the wild-type and alanine at each position to be scanned, and dozens of positions can be scanned in a single library. Following selection for either structure or function using appropriate immobilized ligands, individual clones are sequenced, and the importance of each residue for the selected trait is determined from the ratio of wildtype relative to alanine at each position. The wild-type sequence is conserved at positions that are important for the selected trait, and the analysis is quantitative, so the energetic contributions of individual side chains can be assessed with accuracies approaching those for traditional biophysical analyses with purified proteins.

Shotgun alanine scanning was first applied to study interactions between human growth hormone (hGH) and its receptor (Weiss et al. 2000). This combinatorial method successfully reiterated previous results obtained by conventional alanine scanning, and enabled subsequent homolog and serine scans that provided deeper insights into the molecular details of the interaction (Pal et al. 2005). The method was also used to establish the structural basis for the improved affinity of an hGH variant derived from phage display selections (Opalka et al. 2003; Pal et al. 2005). In addition, the hGH system showed that shotgun alanine scanning data can be used to detect cooperative interactions between interface side chains by analysis of the frequencies of double-alanine mutations (Pal et al. 2005). Following the success of shotgun alanine scanning, a "quantitative saturation" scanning approach was developed where diversity is restricted spatially rather than chemically (Pal et al. 2006). This method allowed for the quantitative assessment of the effects of all possible mutations at all positions in the hGH interface for receptor binding. The interface was found to be highly tolerant to mutations, but the nature of the tolerated mutations challenged generally accepted views about the biophysical and evolutionary pressures governing protein-protein interactions.

In recent years, shotgun scanning approaches have been applied to many other proteins. The method was applied to the phage display platform itself to determine how coat proteins assemble into phage particles (Roth et al. 2002; Weiss et al. 2003). Several antibodies have been scanned to shed light on the details of antigen recognition (Vajdos et al. 2002; Fellouse et al. 2006; Lee et al. 2006), and peptide-protein interactions have also been analyzed (Murase et al. 2003; Schaffer et al. 2003; Skelton et al. 2003). The method has been extended beyond protein-protein interactions in studies of proteins that recognize DNA (Sato et al. 2004) or small molecules (Avrantinis et al. 2002). Shotgun scanning principles also apply in systems other than phage display, and combinatorial alanine scanning has been adapted for in vivo selections with protein complementation assays (Phillips et al. 2006). Overall, these numerous studies validate the use of combinatorial libraries as effective tools for the rapid yet quantitative analysis of the principles underlying protein structure and function.

0 0

Post a comment