One of the complexities of multispecifics is their requirement for expression of multiple chains in similar ratios for correct molecule pairing. As expected, having any chain underrepresented can greatly impact molecule pairing and titer. The talk will discuss the use of Targeted Dual Selection as an expression tool to help drive the stable production of correctly paired multispecifics. Use of this technology and strategy early in research can enable simplified purification strategies as well as increased production yields critically required for project progression.

Cellular models of disease have the potential to facilitate and/or accelerate drug discovery and development. Recently, while evaluating a cell line for disease model generation, we realized it was intractable for gene editing to generate the desired homozygous edit. However, upon further process development and optimization, we were able to preferentially generate several homozygous cell line clones for subsequent testing for the required disease model under consideration. These disease cell line model clones may have a role in facilitating the robust advancement of gene therapy products from the lab bench to the clinic.

Transient transfection is used to generate research-grade material to support discovery and early development. Our goal is to develop high titer, automation-friendly transient platforms in CHO and HEK293 cells. We are able to achieve mAb titers of >1 g/L in HEK293 cells and titers of >2 g/L in CHO cells in a 7-day process in scales from 1 mL (96 deep well plates) to 10 L (wave bag).

One serious limitation in the production of new modalities is demand for plasmid DNA, which forms the basis for many. The use of cell-free technologies enables a manufacturing process which, in our hands, is substantially faster, more productive, and more robust with significantly lower space and energy requirements. This presentation explores our experience with cell-free DNA production and infers its potential applications to streamlining of the DNA production process.

Manufacturers can consider different production parameters to meet desired criteria, like improved process yield, during the development process of monoclonal antibodies (mAbs). We have performed a comprehensive and systematic study on the impact of different codon optimization and cell culture methods on product quality, biochemical features, and functional characteristics. We report changes to glycosylation profiles, charge variants, aggregation, fragmentation, and function among purified protein from combinations of these different production parameters.

Many recombinant therapeutic proteins remain challenging to produce in CHO cells. Using a cumate-inducible system permitting reduced DTE protein expression during stable pool selection, we show that pools generated without cumate are significantly more productive compared to selection in the presence of cumate. Reducing expression is associated with higher cell-viability and faster pool-recovery, suggesting reduced cellular stresses and metabolic burden, likely leading to better survival of high-expressing cells.

ER stress mechanisms are poorly understood in CHO cells and are a major bottleneck in improving the efficiency of production of high-cost recombinant biopharmaceuticals. To gain insights into these mechanisms, we have used artificial inducers of the unfolded protein response (UPR) and ER-stress in recombinant CHO cell to characterize the proteome and the ubiquitinated proteome of CHO cells. We have also characterized ER stress response in CHO cells subjected to bioprocess-related stress such as waste build up and culture longevity. Proteins of interest have the potential to be cell engineering targets to improve efficiency of recombinant protein production.

The main goal of process development for the expression of recombinant proteins in E. coli systems is usually to obtain a high accumulation of the target product with proper folding, usually in a soluble form. Our results show that by growing cells under osmotic stress and heat shock, the proportion of soluble protein can be increased from less than 10% to 50%.

Chinese hamster ovary (CHO) cells—commonly used in biopharmaceutical manufacturing—exhibit production instability. We evaluated five DNA repair genes in over 40 cell lines. Lig4 and Xrcc6 showed higher expression in unstable lines with copy number loss, while lower gene expression correlated with increased cell age. These insights may help predict CHO cell line stability.

In recent years, innovative advancements in cell line development have led to significant improvements in the biopharmaceutical landscape. Integrated with high-throughput screening methodologies, automated platforms have streamlined the traditionally time-consuming steps. However, the continued pressure to reduce timelines and costs while delivering on quality has created unique challenges. This talk will discuss how we adapted our CLD platform to address these challenges and increase productivity in different case studies.

Advances in upstream cell culture processes have increased cell densities and productivity but added challenges to cell clarification. In this case study, extensive work was done to develop a centrifugation and depth filtration harvest for a CHO culture process, but these typical cell clarification methods alone were unable to meet process scale-up needs. Two alternative harvest processes using acid precipitation and cationic flocculation were developed to enable process scale-up with disc stack centrifugation and depth filtration. This presentation will focus on the benefits and challenges of developing and implementing these alternative harvest methods.

This presentation explores strategies to significantly increase the yield of recombinant adeno-associated viral vectors (rAAVs) used in gene therapy by using multivariate optimization to fine tune both the bioreactor environment and the transfection process for maximum productivity.

Process analytical technologies (PAT) using Raman spectroscopy in biopharmaceutical development creates opportunities for advanced real-time process monitoring and control. We demonstrate a novel workflow for faster chemometric model development with increased precision using Raman spectral data for several biomolecules of interest from mammalian bioreactors. Resulting Raman models enable more accurate inline monitoring of glucose, lactate, and several amino acids aiding upstream process development for producing monoclonal antibodies/fusion proteins.

An end-to-end automated system used mid-infrared spectroscopy to quantitate metabolites in complex media and biomass probes to control transcription triggers. This enabled continuous control of feed pumps that maintained nutrient levels as well as induction agent input during fed-batch stirred-tank fermentation. The method is adaptable to other systems and enables soft sensing. The ability to quickly develop in-line quantitative metabolite templates and automated transcription triggers is instrumental for project acceleration.

This presentation explores computational modeling and mathematical approaches that leverage data-driven insights and quantitative models to enhance our ability to predict and optimize cell culture processes. By bridging experimental biology with computational tools, we can unlock new avenues for improving yield, reproducibility and scalability in cell-based applications.