At UCB, our Protein Sciences team is dedicated to producing non-antibody proteins that support drug discovery research across our global organization. In an ever-changing environment, we have successfully navigated both internal and external challenges. Leading a protein science team involves not only balancing resources and ensuring the highest quality protein production but also involves fostering growth opportunities for scientists and pioneering new methods and technologies to achieve more with less. This talk will explore strategies that have enabled us to thrive in a dynamic industry.

We report a versatile approach to remove product-related impurities of multispecific antibodies by altering the binding affinity of light chains to Kappa Select or Protein L resins. These purification-enabling mutations (PEMs) do not impact target binding or stability and we demonstrate the design and application of an entirely affinity-based purification scheme using bispecific and tri-specific antibodies. This purification approach should be adaptable across a range of production scales and workflows.

We designed a workflow to enhance the purity of multispecific antibodies using an advanced automated transfection platform capable of handling up to 1200 transfections per hour. Our strategy focuses on improving product quality during the expression phase to mitigate the contaminant profiles entering the purification stage. By employing mild elution Fc binding resin, we have successfully reduced the purification process of complex antibodies from three steps to one, selectively eluting contaminants at different pH values. This optimized approach not only increases efficiency but also ensures high product quality to enable rapid downstream characterization in supporting early discovery research.

We developed a novel mag bead-based protease elution method to build a high-throughput protein production platform that delivers over 1000 purified proteins per week, ensuring over 95% purity and yields greater than 200 µg per protein. Through optimized automation and E. coli as the expression host, the workflow streamlines the transition from linear double-stranded DNA to pure protein in just 4 days. This adaptable, scalable system is ideal for rapidly evaluating and triaging designed proteins and creating large datasets for machine-learning applications in protein engineering.

• What challenges do we face?
• What improvements do you apply?
• What might address future needs?

Join a group to share and experience and hear what others have learned:

Think Tank 1: Workflow vs technology development?

Facilitators: Jess Williamson, UCB; Monika Musial-Siwek, Be Biopharma Inc.; Pramisha Adhikari, J&J

Think Tank 2: When and how to go from benchtop to R&D?

Facilitators: Bjørn Voldborg, Technical University of Denmark; Karen Betancourt Moreira, BMS; Andrea Partridge, Merck

Think Tank 3: Doing more with less: how do you test new methods and workflows without blowing up your annual budget?

Facilitators: Ruth L. Saxl, Jackson Laboratory; Edson Carcamo Noriega, AI Proteins; Christopher Wassif, AZ

Think Tank 4: Tearing down silos: how do you foster cross-functional collaborations to innovate and improve?

5:30 Think Tank Report Outs: Listen and Learn​

During the Think Tank Table discussions, we shared our experiences and working solutions for end-to-end protein production workflows. Now as a collective community, let’s hear from the table facilitators as they share key discussion points and strategies and provide a wrap-up of their table’s discussion. What can we take away and apply?

• Determine what your end-product could look like
• Consider the full pipeline of your candidate as early as possible
• Select suitable expression systems early on
• Use regulatory acceptable hosts as soon as possible​

The production and purification of recombinant native mammalian proteins in microgram to milligram quantities is a complex and time-consuming process, often presenting a significant bottleneck in protein-based applications. Despite these challenges, the Antigen Engineering and Production Group at IPI effectively supports both the high-throughput antibody discovery and validation platforms. This presentation highlights the key strategies we employ to overcome large-scale antigen production challenges while achieving “high-throughput-like” efficiency.

We have developed a versatile CRISPR/Cas9-based B cell engineering platform, which allows for the generation of B Cell Medicines (BCMs) with broad utility while serving as in vivo protein factories. We demonstrate versatility in three therapeutic areas: enzyme replacement therapy (ERT), immuno-oncology, and coagulopathies. Our first clinical program, BE-101, is a BCM therapy engineered to express human factor IX (hFIX) offering a potential new therapeutic option for people with hemophilia B. US FDA cleared BE-101 IND and granted orphan drug designation and Fast Track Designation. The FIH clinical trial, the BeCoMe-9 Phase 1/2 clinical trial (NCT06611436) is open for enrollment.

Commercial antibodies are notoriously not completely validated making it difficult for investigators to discern if an antibody will work in their desired assay. At JAX, we have developed an antibody validation pipeline. Each antibody is screened by western blot, immuno-precipitation-mass spectrometry, immunohistochemistry, enzyme-linked-immunosorbent-assay, fluorescence-activated cell sorting, and Bio-Layer Interferometry. All collected data and the hybridoma sequence are supplied to our commercial distribution partner and any requesting JAX scientist.

Structure-based drug design is an iterative process that requires the analysis of many DNA constructs along with optimization of expression conditions to yield high-quality target protein. For efficient screening of multivariate parameters, we developed high-throughput expression workflows utilizing Tecan Fluent and ambr15 robotic platforms. We successfully integrated these two systems, each optimized for different tasks, to express target proteins at various scales for downstream analytical needs. 

Production of high-quality biological molecules has historically been a major bottleneck in biologics discovery. While automation in this field has predominantly centered on protein production, labor-intensive tasks such as seed-culture production, workflow and data management, and high-volume bio-waste processing are often overlooked. Here, we present our redefined automation platform and highlight our journey going from a concept to a pipeline capable of generating large amounts of high-quality complex biologics.

This presentation will focus on the convergence of a new high-throughput antibody discovery platform capable of screening 100s of millions of antibodies with machine learning to accelerate the full discovery process. This work is resulting in the identification of high affinity, developable modalities fit for therapeutic use in accelerated time frames, while generating significant amounts of data further refining our algorithms and models.