Figure Legends

Figure 1. GPEx® Technology Overview

A. The GPEx® cell line development process utilizes retrovector technology to insert the product expressing gene of interest (GOI) into transcriptionally active sites in the host genome and can be used for most mammalian cell lines. B. GPEx® Lightning combines the benefits of GPEx® with GS selection and targeted recombination technology. GPEx® technology was leveraged to place >200 Dock sites, which contain recombinase attachment sequences (att), into transcriptionally active sites throughout the genome, creating a parental GPEx Lightning Dock cell line. Transfection of this parental line is performed with a Boat construct containing att, GS, and the GOI, plus a separate recombinase construct (not shown). After transfection, selection results in the rapid and efficient insertion of up to approximately 200 copies of the GOI into transcriptionally active sites in the genome.

Figure 2. GPEx® Lightning Boat Insertion Process

Dock sites containing the attP site flanked by modified retroviral long terminal repeats (mLTR) sequences were placed in transcriptionally active sites throughout the CHO genome using GPEx® technology. See Figure 1 and main text for detailed description of Dock site placement and GPEx® Lightning Parental Dock cell line development. The parental Dock cell line is co-transfected with a Boat construct (containing attB, GS, and the GOI) and a construct expressing recombinase. Recombinase protein mediates the insertion of the entire Boat construct into the attP site in the Dock; this process occurs simultaneously across many dock sites. Following integration, the upstream att site, now called attR, is comprised of the upstream portion of attP (light blue) and the downstream portion of attB (light purple). Conversely, the downstream att site, attL, is comprised of the upstream portion of attB (light purple) and the downstream portion of attP (light blue). The location of forward (green) and reverse (red) primers used in the attR QPCR assay to quantify integration are indicated.

Figure 3. Pool Selection

A. Percent cell viability over time for pools expressing four different monoclonal antibodies (mAb) or an Fc fusion protein. B. Approximate number of integrated Boat copies per cell in each pool. Error bars represent standard deviation of triplicate transfections.

Figure 4. Pool Generic Fed-Batch Production

Results of fed-batch production from pools using the GFB1 strategy. Viable cell density, viability, and titer for A) three replicate pools expressing mAb1, B) average of replicates for pools expressing mAb2 (n=2), mAb3 (n=3), mAb4 (n=3), and an Fc Fusion protein (n=1). Average standard deviation for each molecule across the time course for each of the 4 mAbs were less than 10%, 3%, and 13% for VCD, viability, and titer, respectively (not displayed).

Figure 5. Pool Ambr® 250 Bioreactor Production

Viable cell density, viability, and titer for mAb1 pp3 production using GFB2 and the optimal GFB2-based production strategy identified in an Ambr® 250 process optimization campaign. Results from the GFB1 strategy (from Figure 4) are shown for comparison.

Figure 6. Pool Stability

A. Final titer and integrated Boat copy number from a stability study of 40+ generations of 3 independent pools expressing an Fc fusion protein. Each pool stability study was performed in 2 different media. Error bars intentionally omitted for clarity. B. Final titer and integrated Boat copy number from a 40+ generation stability study on 2 pools expressing mAb 1 and mAb2. Production was performed using the GFB1 strategy in duplicate. Error bars indicate standard deviation.

Figure 7. Clone Generic Fed-Batch Production and Stability

A. Linear regression of final titer from a 16-day productivity using GFB1 plotted against percent Dock fill for 26 clones isolated from Fc Fusion pp2. B. Final titer from a 40+ generation stability study on the top 3 clones from Fc Fusion pp2. Productivity was performed using the GFB1 strategy. Error bars indicate standard deviation.

Figure 8. Released N-glycan Profile Stability and Pool to Clone comparison

A. Overlay of UPLC traces of fluorescently labeled and released N-glycans from two different Fc-fusion pools. Terminal samples from productivity of the indicated generations (Gen) were analyzed. B. Overlay of UPLC traces of fluorescently labeled and released N-glycans from three different monoclonal antibodies. Terminal samples from productivity of a pool and a clone derived from that pool were analyzed.

Table 1. Released N-glycan Profile Stability

A comparison of individual N-glycan species released from Figure 9. Numbers shown are percentages.

Table 2. N-glycan Species from Pools and Clones.

A comparison of individual N-glycan species released from Figure 10. Numbers shown are percentages. ND = not detected or below 0.4% area reporting threshold.