(620aj) Design, Selection and Optimization of Affinity Peptides for Applications in Bioprocessing (Rapid Fire)

Affinity ligands are employed in a range of applications as drugs, drug carriers, purification and imaging tags, biosensing agents, inhibitors of biomolecular interactions and modulators of crystal growth. The specific design and engineering of an affinity ligand depends on the desired application. Typically, the design of affinity ligands for different applications is guided by one or more end-points of affinity, selectivity and biological activity. While high-affinity and selectivity are often the governing design criteria, it is not always the case since certain challenges require additional criteria to be met. In such cases, affinity ligands often need to be tailored to simultaneously possess multiple functionalities and biochemical properties. Peptides are an attractive class of affinity ligands well-suited for such applications due to their amenability to rational design and modifications, ease of manufacture, low-cost, high stability and prolonged shelf-life. In this work, we will demonstrate the rational design and engineering of affinity peptides to meet one such challenge encountered in the bioprocessing industry.             

            Current bioprocessing strategies for the production of non-antibody based biologics are costly and require lengthy purification trains.  While Protein A chromatography has emerged as the “gold-standard” step offering robust, high yield and high selectivity, affinity-based purification of monoclonal antibodies (mAbs), no affinity ligands currently exist for non-mAb based biologics. Further, non-affinity methods typically cannot match the process yields, effectiveness, reduced time and costs offered by affinity chromatography. By employing (and developing) a repertoire of design and screening strategies available in a biomolecular engineer’s toolbox, our goal is to address this shortcoming and create, against any given biologic, robust, highly selective, low-cost, affinity peptides that possess all the desired properties essential for protein purification. These properties include not just high affinity for their target, but also high selectivity (in a milieu of host-cell proteins and impurities encountered with changing product titers and variable feed streams), sufficient dynamic binding capacity at the column scale (when immobilized on a support to create an affinity adsorbent), high elutability (to enable product dissociation and recovery), high stability (i.e. resistance to base and proteolytic cleavage) and good regenerability (for use over multiple purification cycles).

            In this talk, we will focus on the design and optimization of affinity peptides for the purification of two key biologics – human growth hormone (hGH) and interferon α2b (IFNα2b).  Using a diverse array of peptide design and screening strategies such as epitope mapping of natural binding partners, de novo design and in silico screening, mutational analysis, and cyclisation, we have designed libraries of peptides targeting both hGH and IFNα2b. We have tested these peptide libraries under different conditions and on various platforms (microarray screening and in-solution binding assays using fluorescence polarization, followed by batch-scale, resin screening experiments) to identify lead candidates that were carried forward for optimization at the column-scale. Some of the key results of this work are as follows:

• The designed peptide libraries exhibited a range of affinities, selectivities and elution characteristics for the target biologics, thus enabling the selection of the most optimal peptide ligands for the purification of recombinant hGH and IFNα2b.
• We found that short stretches of naturally-occurring amino acid sequences can be effectively re-designed to incorporate and display functionalities not intended for them by nature. For example, a peptide epitope involved in protein-protein complex formation in nature can be engineered by point-mutagenesis to display enhanced selectivity in hGH purification.
• Further, the rational insertion of specific amino acids at the front or tail ends of the same epitope yielded candidates with enhanced capacities and very different elution profiles for hGH. Interestingly, we also find that some peptides can be tailored to impart specificity in purification for a given biologic or a set of biologics.
• Finally, our results with a top-performing, hGH-binding peptide obtained after two rounds of design and optimization, demonstrate that affinity peptides can offer purification performances (good product recovery, effective DNA and host-cell protein clearance) comparable to those obtained in Protein A chromatography.

           Our preliminary studies with the design and selection of IFNα2b-binding peptides have yielded several promising candidates that are now being optimized to provide enhanced purification performance. Based on the lessons learnt from these studies, our ongoing and future efforts will focus on engineering elutability in the ligands a priori and incorporation of non-natural amino acids and cyclisation to further optimize the specific molecular recognition and biochemical properties of these affinity peptides. Overall, this work demonstrates the potential impact and power of rational design of affinity peptides on downstream bioprocessing of therapeutic proteins.