Cell-free protein synthesis decouples cell growth from protein production, enabling high-throughput production of mg quantities of protein in only 5–8 hours. This open system supports site-specific incorporation of multiple non-natural amino acids and efficient synthesis of complex molecules from diverse scaffolds. Scalable from bench to commercial scale, cell-free platforms accelerate high-throughput protein expression and offer precision conjugation strategies and a powerful, flexible alternative to traditional cell-based biologics discovery and manufacturing.

Cell-free gene expression, CFE (a.k.a. cell-free transcription-translation, TXTL), is a rapidly expanding technology that enables a wide variety of applications. CFE is characterized by extensive scalability and portability and a fast experimental turnover that enables screening and synthesizing biologics such as proteins, enzymes, and phages, in a few hours. I will present the most recent biomanufacturing capabilities of an all-E. coli CFE system.

Numerous factors during protein synthesis can lead to misfolding and aggregation. While various molecular chaperone classes are activated to alleviate or clear this occurrence natively, recombinant protein production frequently overwhelms this network. While cotransformation chaperone plasmids are commercially available to assist with bacterial recombinant protein expression, limited tools are available to aid cytosolic or secreted recombinant protein production in baculovirus infected insect cells. To that end, our group has engineered coexpression bacmids and utilizes alternative promoters to improve final target production for our clients.

A lack of consistent, complete, and standardised experimental reporting limits computational prediction of protein expression outcomes and requires extensive dataset curation. We propose a structured template for capturing protein expression and purification data to improve reproducibility, support machine learning applications, and reduce empirical construct screening. The framework prioritises metadata into critical, highly enabling and optional categories, and includes negative data to improve dataset quality. By generating diverse, machine-usable datasets, this approach aims to support generalisable predictive models, establish a trusted protein production repository and accelerate the path from digital design to purified protein.

The production of monoclonal antibodies, fragment antibodies, and bispecifics has become a large portion of protein production in the past several years. While we have made tremendous advancements in the cell culture methodologies and protein purification, many are still doing the upfront work of testing ratios of heavy and light chains to determine optimal ratios during expression. We have created a new novel bicistronic vector that not only increases protein expression but also eases downstream work into making stable cell lines. 

High-titer antibody clones do not always translate into developable, manufacturable candidates. This talk presents an analytical perspective on bridging expression screening and CMC, highlighting how early LC-MS-based characterization, paired with orthogonal methods such as SEC, CE-SDS, and cIEF, can flag product-quality liabilities before candidate selection. General best practices for fit-for-purpose method design, data triage, and cross-functional alignment will be shared to reduce late-stage risk.

Production of recombinant proteins involves maintaining their integrity through purification and downstream applications. This is especially challenging for autolysis-prone peptidases. We developed an activity-guided purification strategy that suppresses autolysis, thus enabling production of intact KLK7, a serine peptidase, without resorting to catalytic mutations or inhibitors.  Further investigation helped identify novel biological substrates of KLK7 and inhibitory antibodies, demonstrating a robust approach for purifying labile recombinant enzymes.