Bioproduction occurs in tanks, but protein synthesis and secretion happen in the single cell. Thus, it is of great value to understand protein secretion, at the single-cell level. To do this, we deployed SEC-seq, a single-cell RNA-sequencing method that simultaneously quantifies protein secretion at the single-cell level. Using this, we investigated the transcriptional differences of single cells in IgG-producing pools, and derivative high and low producer clones. This work provides insights into clone heterogeneity, and adds to the body of knowledge regarding what really drives productivity.

Leveraging two proprietary technologies, we developed a novel antibody expression vector that increases yield, accelerates cell-line development, and improves stability. It features a mutated glutamine synthetase (GS) selection marker with reduced enzymatic activity to enhance selection stringency and generate highly productive, stable clones. It also incorporates an enhanced CMV promoter containing a proprietary DNA insert to boost expression. Together, these innovations increase bulk pool antibody expression four- to five-fold.

Genome engineering of CHO cells is a powerful lever to improve biomanufacturing performance, yet access to flexible nuclease systems remains limiting. This talk presents a streamlined 39-day workflow leveraging the open-source OpenCRISPR-1 nuclease to enable efficient and scalable genome editing in CHO cells. We demonstrate high editing efficiency across multiple targets, including biallelic and multiplexed gene knockout. This approach highlights the potential for next-generation genome editors to accelerate cell line engineering and expand accessible tools for biopharmaceutical development.

As molecule complexity increases, expression platforms must be engineered to support specialized biosynthetic requirements. For certain fusion peptide molecules, biological activity depends on efficient c-terminal amidation catalyzed by peptidylglycine 𝛼-amidating monooxygenase (PAM), which was limited to ~50% by endogenous PAM activity for this molecule. To address this, engineered PAM was co-expressed using a site-specific integration (SSI) platform, enabling controlled PAM expression, and increasing peptide amidation to as high as ~90%.

We have developed a novel, synthetic, tunable promoter system for recombinant protein expression in CHO cells. It delivers a threefold productivity increase over the CMV promoter with enhancer and shows good long-term stability. This promoter can significantly boost biologics manufacturing, enabling higher yields and consistent performance. Its tunability also supports optimized expression for difficult-to-express molecules such as multispecific antibodies.

Correct heavy and light chain pairing is a critical quality attribute for bispecific antibodies, yet early clone selection in cell line development often relies primarily on titer. We present a high-throughput Bio-Layer Interferometry (BLI)–based assay to screen chain pairing directly from culture supernatant, enabling rapid, orthogonal assessment of assembly during fed-batch production and improving clone elimination decisions in bispecific cell line development workflows.

Current methods for isolating high-producing CHO clones employ antibiotic and/or metabolic selection (e.g., DHFR/MTX, GS/MSX). These approaches are time-consuming, labor-intensive, and yield variable productivities. We developed a novel strategy that integrates a miRNA into the expression vector, co-transcribed with the gene of interest. The miRNA enables translation of a reporter mRNA (e.g., GFP or CD4), permitting efficient selection of high producers by FACS or magnetic beads.

Transposase-mediated semi-targeted transgene integration (STI) offers substantial advantages over random integration for the development of stable, high productivity CHO cell lines. Despite increasing implementation of transposases in monoclonal antibody Cell Line Development (CLD), their performance for other recombinant proteins has not been widely demonstrated. Presented here is the successful establishment of a transposase-mediated STI CLD platform for vaccine antigens.

Lactate accumulation in CHO bioreactors can impair growth, productivity, and product quality. Using a lactate-runaway model and comparative bottom-up proteomics across multiple antibody-producing cell lines, consistent differences between high- and low-lactate clones revealed actionable biomarkers. Incorporating HK2 knockout and ASNS overexpression into the host cell line may help prevent lactate runaway, supporting more robust processes and reduced biomanufacturing variability.

• ​Predictive clone selection and productivity modeling
• AI-guided genetic engineering of CHO cells
• Linking cell line genotype to manufacturability
  
• Building robust training data: integrating CLD, bioprocess and developability data
  

Manufacturing AAV vectors via transient transfection of HEK293 cells depends on timely delivery of transfection complexes to bioreactors. We found plasmid–reagent complex size, which changes with incubation, influences productivity overall. To address this constraint, we developed a method to arrest complex growth at the optimal size, enabling storage up to 1 day with less than 10% change in vector genome titer and less than 5% change in full capsids.

In recombinant Adeno-Associated Virus (rAAV) delivered gene therapy, continuous improvement of established manufacturing processes to optimize product quality and production yields is desirable to further reduce manufacturing costs and regulatory risks. Using cell line development and engineering, plasmid and process optimization, plus technological advancements and innovation, we demonstrate how we achieved our current state of the art NAVXpress platform process, as we continue to innovate further for future enhancements of our next generation production processes, while maintaining high product purity.

The increasing demand for faster and animal-free approaches to therapeutic antibody development underscores the importance of advancing AI-guided, in vitro platforms for the discovery of single-chain variable fragments (ScFvs). By integrating cell-free display technologies with machine learning–driven sequence analysis, the approach enables high-throughput screening and characterization of ScFvs with improved affinity, stability, and developability, accelerating therapeutic antibody discovery while reducing costs and ethical constraints. This integrated strategy could transform biologics development by enabling rapid, data-driven discovery and extending to other antibody fragments such as nanobodies.

A deeper understanding of upstream process parameters modulating recombinant protein assembly outcomes enables targeted bioprocess development, thereby improving process robustness and yield. We developed an image‐driven toolkit, which includes high content confocal microscopy, super resolution microscopy, and cryoET. These techniques were applied to yeast expressing L1 from different HPV types and sampled over a time course under different growth conditions. The results uncovered distinct patterns between protein localization, intracellular changes, cell growth conditions, and protein yield.