RESEARCH POSTER PRESENTATION DESIGN © 2012 www.PosterPresentation s.com Clone selection is one of several key steps in a successful stable cell line development (CLD) process. To overcome the shortcomings of traditional manual cloning and screening processes, we have successfully integrated two high-throughput cloning methods: FACS and ClonePix TM into our CLD workflow. Both methods are highly robust, and utilize a primary clone screening step based on high expression of the protein of interest. A side by side comparison of ClonePix TM with traditional limiting dilution cloning (LDC) was performed using a same monoclonal antibody expressing cell pool. ClonePix TM is an automated selective clone selection method compared to highly labor-intensive and random LDC method. In addition, we present case studies using FACS cloning in combination with CloneSelect TM Imager which outline an approach for identifying high-producing clones with only one round of cloning while ensuring the monoclonality, which significantly reduces the CLD timeline. Overall, the studies presented here aim at demonstrating high-throughput cloning technologies that reduce cost, labor and timeline for CLD and improve clone selectivity to ensure the identification of high producing clones. Abstract Kitty Agarwal, Rich Harper, Brian Davies, Amy Venturini, Abhinav Shukla and Ying Huang Cell Line Development, KBI Biopharma Inc., Durham, North Carolina 27704. Integrating High-Throughput Cloning into Cell Line Development Traditional vs. High-throughput Cloning www.kbibiopharma.com We used the same mAb producing CHO K1 cell pool for cloning and clone screening for both LDC and ClonePix TM . For LDC, cells were seeded at 0.5cells/well in 96-well plates and wells with >10% confluence were scaled up to 24-well plates in ~ 3 weeks. Monoclonality and growth were monitored using CloneSelect TM Imager (CSI). For ClonePix TM , cells were seeded in semisolid media containing CloneDetect (FITC labeled antibody). 10 days post-seeding high producing clones were picked using CLonePix TM based on shape, size, proximity and exterior mean fluorescence intensity into 96-well plates. Following titer assessment, high producing clones were scaled up to 24-well plate. For both the cloning techniques, clones were further scaled up to 6-well plates and shake flasks based on productivity. Titers were measured using ELISA assay. For FACS we present multiple case studies of single cell cloning and clone screening of mAb producing CHO S cells to select high-producing monoclonal clones (mAb X and Y) using FACSJazz TM cell sorter. For each mAb pool, single cells were sorted based on high fluorescence signal into 96-well plate. Following titer assessment high producing clones were scaled up to 24-well plate. Monoclonality was verified using CSI. The same work flow was followed as LDC and ClonePix TM from here on. Top 3 clones mAb X clones were assessed for productivity with different feeds and varying feed percentages. Titers for FACS studies were measured using high-throughput methods: Octet TM and Pro A HPLC. Conclusions Cell line selection were run in parallel using conventional LDC and high-throughput automated ClonePix TM . The productivity of the selected cell lines from the two methods was comparable. ClonePix TM successfully generated high yielding clones. ClonePix TM and FACS offer the advantage of clone screening based on productivity early on during clone selection, thus increasing the probability of generating high-producing clones by screening more number of clones in a shorter duration compared to LDC. FACS can successfully sort and clone cells with favorable attributes and clonality can be verified using the CloneSelect TM Imager. Further improvement of strategies to reduce time lines and increase productivity are ongoing, which include direct selection, amplification and cloning with ClonePix TM and pool enrichment using FACS, as well as incorporation of other high-throughput methods like ambr TM for clone screening. Single Cell Cloning using FACS Figure 2: Cell growth (i) and monoclonality (ii) of clones recovered from LDC were monitored using CSI. Starting from 2 hours post-seeding, CSI was used to capture cell images on Days 0, 1, 2, 7, 14 and 19. Well Images (ii) of representative clone presented here showing the clone growth. Figure 1: Cloning using ClonePix TM and FACS involves selecting cells based on high expression of mAb opposed to Limiting Dilution Cloning (LDC) in which cells are seeded into 96-well plates to ensure single cells per well. LDC vs. ClonePix TM : Suspension Culture Assessment Figure 4: Top 30 clones were further assessed for productivity in a 7 day batch culture. (i) On average, clones identified by LDC and ClonePix TM had 1.8-fold and 2.5-fold higher mAb productivity compared to the parent pool respectively. (ii) Clone ranking across different screening stages. For ClonePix TM , 9 of the top 10 clones in shake flask were among the top 30 clones at the 6-well plate stage, indicating a good correlation between clone rankings across the screening stages. Top 5 ClonePix TM clones were further evaluated for mAb productivity using different feeding strategies. Shake Flask Batch Culture Fed Batch Productivity Assessment Figure 5: Top 5 clones by ClonePix TM were assessed using two different feeding strategies. mAb productivity for both the schemes was found to be similar. Figure 6: Top mAb pools (mAb X and Y; representative mAb X FACS profile shown in (i)) selected based on titer and FACS profile for single cell cloning. Cell growth and monoclonality were monitored using CSI. CSI cell images were captured on Days 0, 1, 2, 3, 7, and 14. Outgrown clones were assessed for titer. High yielding clones were scaled up to 24-well plate followed by expansion of top ~ 60 clones to 6-well plate and further top 30 clones to shake flask based on absolute and growth-normalized titers. Figure 8: Top 10 clones were assessed for productivity with simple fed batch. Viable Cell Density (data not shown) and harvest day titer were measured. Highest producing clone titers were 0.44g/L (mAb X) and 0.33g/L (mAb Y). Figure 9: Top 3 mAb X clones were assessed for productivity with different feeding strategies. One of the feeding strategies (Scheme 2) provided the highest titer of 0.94g/L for clone2. Scale up of the clones to bioreactors (3L) gave high titer (1.0- 1.3g/L) for clone2 indicating the scalability of clones generated by FACS. Limiting Dilution Cloning ClonePix TM : Screen for High Yielding Clones Figure 3: ClonePix TM images on Day10 post-seeding (i) in white and fluorescent light indicate presence of high yielding clones. Post-picking cell growth was monitored on Days 2 and 6 using CSI (ii and iii). ClonePix TM allows clones to be selected based on high mAb productivity early on in the clone screening process . Titer of mAb X Clones Titer of mAb Y Clones Figure 7: mAb X and mAb Y clones were assessed for productivity at each screening stage in static culture using high-throughput Octet TM system, to select for high producing clones. Top 30 clones were scaled up to shake flask. Fed Batch Assessment of Top 3 mAb X Clones Simple Fed Batch Culture of Top 10 mAb X and mAb Y Clones LDC vs. ClonePix TM : Static Culture Assessment (i) (ii) Static Culture screening of mAb X and mAb Y Clones from FACS mAb X and mAb Y Clone Selection in Shake Flask