Navigating biomanufacturing challenges in the use of CHO cells for recombinant adeno-associated virus production

The initial goal of this project was to investigate different production methods for recombinant adeno-associated virus (rAAV), with a special emphasisonChinese hamster ovary(CHO) cells due to their successful track record in complex biopharmaceutical production. Firstly, the current manufacturing system, based on triple transfection in human embryonic kidney 293 (HEK293) cells, was reviewed to pinpoint existing knowledge gaps and bottlenecks (Introduction, Section 1.3). The system’s complexity in identifying essential elements and the apoptotic response triggeredbysometransfected geneswere identified as significant challenges. To address the latter, we studied the potential of generating anti-apoptotic cell lines.

Given the project’s vision, CHO cells were selected for this genetic engineering approach. However, the literature provided conflicting studies regarding the impact of anti-apoptotic genes on productivity and culture lifespan. We developed different cell lines via targeted integration to ensure that only a single copy of the gene of interest and erythropoietin (EPO) were present in our cells at a known locus.

The generation and characterization of CHO-S-derived anti-apoptotic cell lines are presented in Chapter 1. In this chapter, we evaluated relevant anti-apoptotic genes in batch mode and further characterized the most promising cell lines. Additionally, the most suitable phenotypes were selected for rAAV production tests in Chapter 2.

Despite significant efforts,we faced considerable challenges in validating rAAV production in CHO cells. All the approaches attempted are detailed in Chapter 2. In parallel, to address the challenges in the current production system via triple transfection in HEK293 cells, a designof- experiments (DOE) approach was used to optimize the plasmid ratio in the production process. The optimization results are presented in Chapter 3.

Lastly, we collaborated with VVector Bio, a CDMO inMontréal, Canada, to design an efficient strategy for separating full from empty capsids. The method is described in Chapter 4.