Chemical diversification of peptide phage display libraries

Cyclic peptides possess several properties that make them highly attractive for drug development, including their ability to bind challenging targets, high target specificity, and suitability for powerful discovery and synthesis techniques such as biological display and automated solid-phase peptide synthesis. However, their therapeutic application has so far been largely restricted to extracellular targets, primarily due to difficulties in crossing cell membranes. Developing short peptides with limited polar surface area, and thus higher potential for membrane permeability, remains challenging with methods such as phage display, as maintaining short peptide lengths inherently limits the diversity attainable in the libraries. For example, a 6-amino acid peptide library comprises only 64 million sequences combinatorially and exhibits limited chemical diversity due to the use of only canonical amino acids as building blocks. This thesis aimed to address this limitation by appending chemically diverse fragments to phage-displayed peptide libraries, with the ultimate goal of facilitating the discovery of short, membrane-permeable binders against a challenging target. In a first project, I established a method that enabled automated, parallel chemical modification of phage display peptide libraries with chemical fragments. Using 11 structurally diverse building blocks, I diversified phage-encoded peptide libraries and succeeded in identifying binders to the oncology target Myeloid cell leukemia 1 (MCL-1). Structure-activity relationship analysis showed that the appended fragments contributed to binding. These results validated the concept that appending chemical fragments can drive novel binding interactions and demonstrated the feasibility of enriching short ligands through chemical diversification. In a second project, I extended this approach by establishing a purification-free synthesis workflow in microtiter plates, generating 384 chemically diverse fragments for phage display peptide library diversification. Quality control analysis confirmed the successful preparation of a highly diverse library. Phage display selections against MCL-1 did not yield binders, likely due to the complexity introduced by a new approach in which multiple libraries were screened in pools. The results of this PhD thesis show that biological display systems and high-throughput chemical synthesis methods can be combined to enable the screening of large and chemically diverse libraries for identifying sub-kDa peptidic ligands for potential targeting of intracellular proteins.