Advancements in gene editing technology have enabled engineering Chinese hamster ovary (CHO) hosts with improved growth, viability, or productivity for recombinant monoclonal antibody (mAb) expression. One approach involved a knockout (KO) of the BCAT1 gene, which encodes the first enzyme in the branched chain amino acid (BCAA) catabolism pathway. The BCAT1 KO accumulated less BCAA byproducts including growth-inhibitory short chain fatty acids (SCFAs), ultimately improving growth and titer. However, because SCFA accumulation induces a metabolic shift to higher cell-specific productivity, SCFA supplementation during production aided in further translating the higher growth into higher titers. Here we describe knocking out BCKDHA and BCKDHB genes, which act downstream of BCAT1, to reduce SCFA accumulation. We found that a partial KO of BCKDHA and BCKDHB introduced in the background of an apoptosis-resistant BAX-BAK host can achieve higher viabilities and titers. This was evident when SCFAs were added to boost productivity, as such additives negatively impacted culture viability in the wildtype but not BAX-BAK KO cells during batch production. Altogether, our findings suggest that SCFA addbacks can significantly increase productivity and mAb titers in the context of apoptosis-attenuated CHO cells with partial KO of BCAA genes. Such engineered CHO hosts can offer productivity advantages for expressing biotherapeutics in an industrial setting.

Chinese hamster ovary (CHO) cells have been the primary workhorse for biotherapeutic protein production for nearly 40 years. Despite advances in process intensity and efficiency, universal mammalian cell phenotypes such as lactate and ammonia production, as well as the obligate requirement to supply essential amino acids, have led to challenges in process optimization without a one-size-fits-all solution. Over the last 9 years, we have developed genetic engineering strategies fundamentally reimagining these ubiquitous mammalian cell phenotypes and will present case studies around each in turn. Lactate: Multiplex knockout of lactate dehydrogenase(s) and pyruvate dehydrogenase kinase(s) is sufficient to eliminate lactate production without impacting growth, protein production, or product quality. Intriguingly, this genetic engineering strategy appears to be generalizable to other mammalian cell lines such as HEK293. Ammonia: Simultaneous knockout of asparaginase and glutaminases—responsible for the first step of catabolism for the respective amino acids—entirely eliminates ammonia production while cells are growing and reduces total ammonia levels over the full culture. This strategy can be combined with the lactate strategy described above, leading to cell lines with no lactate production and decreased ammonia production. Again, this strategy did not impact growth or product quality, while product titer actually improved using these cell lines as host cells for standard CLD workflows. Essential amino acids: We have been able to restore biosynthesis for 2 essential amino acids (including 1 never before shown in a mammalian cell line) by introducing heterologous genes from a multiple microbes. While the resulting heterogenous pools show minor decreases in growth (15-25%), we are exploring if subcloning and/or additional optimization around pathway gene expression is sufficient to restore normal growth. We will also share the results of efforts to introduce these strategies into the engineered lactate/ammonia cell lines and an initial assessment of suitability for biotherapeutic production. Collectively, these engineered cells have the potential to serve as next-generation manufacturing hosts, bypassing many of the existing limitations of mammalian cell lines.

Engineering biology can transform bioproduction by enabling precise, programmable control of cellular systems. This interdisciplinary approach combines synthetic biology, systems biology and -omics analytics to drive improvements in bioprocess phenotypes such as yield, cell line stability and product quality. We have utilized synthetic biology strategies to enhance gene expression stability and productivity. Using a CRISPR-dCas9 system to deliver the catalytic domain of a histone acetyltransferase to the CMV promoter boosted mAb titres and cell line stability [1]. Complimentary to this, tRNA gene barriers have proven more effective than traditional UCOE elements in maintaining stable and high-level transgene expression [2,3]. Systems biology has provided deeper insight into CHO cell metabolism revealing ATF4 as a key regulator of lactate shifts induced by glutamine depletion and manipulating its expression improves cell performance [4]. Flux sampling of genome-scale metabolic models, constrained by transcriptomic data, has further identified metabolic signatures unique to high-producing CHO clones, suggesting nutrient supplementation strategies to enhance productivity [5]. Cell death pathway dissection shows non-apoptotic cell death dominate during fed-batch bioreactor conditions [6]. These findings demonstrate how engineering biology enables rationale CHO design. As the toolkit expands, future CHO expression systems will be increasingly modular, predictable and adaptable – designed from the genome up for next-generation biotherapeutics. 1. Butterfield SP et al. (2024) Biotechnol J, 19: e202400474. 2.Sizer RE et al. (2024) Biotechnol J, 19: e2400196. 3.Sizer RE et al. (2025) Biotechnol J, 20: e202400455. 4.Torres M et al. (2025) Metab Eng, 88: 25-39. 5.Meeson KE et al. (2025) Biotechnol Bioeng, on line ahead of print. 6.Mentlak DA et al. (2024) Biotechnol J, 19:e2300257.

Despite evidence that they are not functional or infective, retrovirus-like particles (RVLPs), originating from endogenous proviral sequences in Chinese hamster ovary (CHO) cells, present a safety risk for biotherapeutics manufactured using this cell line due to their resemblance to other mammalian leukemia viruses. The presence of RVLPs contributes significantly to the complexity and cost of therapeutic protein manufacturing using CHO cells: dedicated chromatography, low-pH or detergent-based viral inactivation and viral filtration steps are often necessary during DSP primarily for the purpose of reducing RVLP levels. Furthermore, these viral clearance/inactivation steps may be incompatible with some proteins and more complex biologics such as enveloped VLPs. Here, we demonstrate that CRISPR- and shRNA-based cell engineering strategies can be used to disrupt RVLP production by targeting the RVLP nucleotide sequences. Additionally, specific antibodies were generated to monitor RVLP protein expression, including RVLP envelope (Env) protein localized on the surface of CHO cells, greatly facilitating selection of RVLP-deficient clones. These modified CHO cells showed reduced RVLP production while maintaining or enhancing the ability to produce recombinant virus-like particles (VLPs), highlighting their potential application in biomanufacturing, especially for complex biologics that are incompatible with standard RVLP mitigation procedures, namely viral inactivation and nanofiltration.