Poster M15 - Purification of Monoclonal Antibodies by Self-Assembling High-Affinity Supramolecular Filaments

Introduction: The purification of monoclonal antibodies (mAbs) from cell culture effluent represents a major challenge in developing efficient and cost-effective production of therapeutic proteins on an industrial scale. Affinity precipitation has emerged as a promising alternative to conventional protein-A chromatography-based separation techniques, which are limited by production capacity and high cost. However, many affinity precipitation technologies still struggle to achieve the high yield and specificity provided by protein-A. Supramolecular peptides and peptide derivatives have been extensively utilized in regenerative medicine, drug delivery, and disease diagnostics due to their high biocompatibility, biodegradability, and specificity. These supramolecular polymers can undergo macroscopic phase transitions, driven by changes in intermolecular or interparticle interactions. Leveraging this typically undesired phenomenon, we have developed peptide amphiphile nanofibers, termed immunofibers (IFs). These IFs are created by co-assembling two different peptide amphiphiles: one containing the Z33 peptide sequence known for its high affinity to mAbs, and a filler molecule that spaces out the ligand molecules along the fiber surface. By optimizing the ligand:filler and ligand:mAb ratios, we can purify two different mAbs from cell culture effluent with high selectivity and purity.

Materials and

Methods: Immunofiber (IF) peptide amphiphiles were synthesized using standard Fmoc solid phase peptide synthesis and purified using preparative RP-HPLC with a C18 column. Peptide identity was confirmed using MALDI-TOF mass spectroscopy and purity was confirmed with analytical RP-HPLC before being aliquoted and lyophilized. Lyophilized IF filler and ligand molecules were first dissolved in hexafluoroisopropanol to remove any pre-existing structures before being mixed at set molar ratios. HFIP was then evaporated out and PBS was added as the final solvent for the co-assembled IF solutions and left to assemble overnight before experimentation or analysis. Pure monoclonal antibodies (mAbs) and mAb clarified cell culture effluent were sourced from Bristol-Meyers Squibb. Fluorescent mAb (FITC) was purchased from Sigma-Aldritch. Fiber structures were verified through CD spectroscopy on a Jasco J-710 spectropolarimeter and Bright-field TEM imaging using a 2% uranyl acetate stain on a carbon film coated copper grid. High-affinity binding between fibers and mAb was verified using isothermal titration calorimetry (ITC) of a 100 μM mAb solution into a 40 μM IF solution on a VP-ITC system. For mAb purification experiments, solutions of assembled IFs and mAbs were mixed with allowed to bind undisturbed for 30 minutes. Ammonium sulfate salt was then added to induce phase separation. After 30 minutes, the mixtures were centrifuged and the supernatant was collected for analysis. RP-UPLC was used to quantify filler and ligand concentrations and ProA-HPLC was used to quantify mAb concentration.

Results, Conclusions, and Discussions: We confirmed that the filler and ligand molecules co-assemble into supramolecular nanofibers and maintain this fibrous structure even in the presence of mAb at neutral pH, regardless of the filler:ligand ratio used in co-assembly. However, as filler molecules assist in spacing out the ligand molecules along the fiber surface, the filler:ligand ratio was found to have a huge impact of the precipitation yield of mAb, with higher filler:ligand ratios leading to higher rates of mAb capture. This suggests that the increased spacing between ligand molecules along the fiber surface leads to more efficient mAb capture. Moreover, IFs alone or mAb alone failed to precipitate out of solution while in cases with IF+mAb, over 95% of IFs also precipitated out of solution, indicating that the precipitation could be attributed to the high affinity binding between the two species. We also determined high ligand:protein ratio to be critical for higher precipitation rates. We also verified the high selectivity of our IFs for mAb, as mixtures of IF and non-mAb proteins failed to phase separate. Finally, we were able to recover high purity of mAb after precipitation. IF-mAb complexes were resuspended in PBS and the solution was dialyzed with a 50 kDa membrane. Under optimum conditions, over 90% of pure mAb could be recovered with minimal IF remaining in the final solution. When we mixed IFs with mAb cell culture effluent for different mAb types, over 85% of pure mAb was recovered for all mAbs tested. These promising results showcase the potential of supramolecular peptides as efficient and cost-effective materials for protein purification.

Acknowledgements (Optional):