In the realm of biomanufacturing, space-time yield (STY) is increasingly recognized as an essential metric for evaluating upstream process efficiency. STY facilitates the direct comparison of various protein production methodologies—batch, fed-batch, and continuous fermentation—providing researchers and process developers with critical insights into process yields, production timelines, and facility constraints.
This comprehensive approach aids in the formulation of efficient and goal-oriented manufacturing strategies. We discuss here the methodology for calculating STY using model organisms Pichia pastoris and Chinese hamster ovary cells (CHO), highlighting the unique attributes of fed-batch versus continuous fermentation processes and their respective impacts on STY. By optimizing for higher STY, organizations can achieve greater protein output with minimized spatial and temporal requirements, thereby reducing commercial production expenses.
Traditionally, product concentration (titer) has been employed to describe the productivity of fed-batch and batch fermentations aimed at generating substantial quantities of heterologous protein for downstream processes. Titer, quantified as a concentration (e.g., g/L), represents the protein amount present in the bioreactor at a given time. In fed-batch fermentation, the final titer signifies the total protein produced, making it a useful metric for comparing similar processes.
Continuous fermentation, however, presents a different scenario. Nutrients are continually supplied to the bioreactor, while secrete5d proteins and waste products are constantly removed, resulting in a dense biomass that sustains high protein expression levels over extended periods.
Space-time yield: a comprehensive metric for productivity
In perfusion fermentation proteins are harvested continuously from the biomass. Consequently, the protein titer within the bioreactor does not accurately reflect the overall protein yield due to the ongoing harvest. Relying solely on titer can thus misrepresent the true productivity of continuous fermentation processes.
STY offers a normalized measure of productivity applicable to any cultivation mode or protein-expressing organism. Defined as the total mass of protein produced per bioreactor volume per cultivation day, STY normalizes protein quantity based on cultivation scale and duration, enabling straightforward comparisons across different processes.
Normalized comparison of processes of different lengths is important when comparing continuous fermentation with fed-batch, because continuous fermentation campaigns are often longer and more easily extended than fed-batch campaigns. Fed-batch processes are limited by nutrient depletion and waste accumulation, which eventually lead to unhealthy biomass and halted protein production, whereas the continuous input of nutrients and removal of waste products in perfusion fermentation enables healthier biomass that can thrive longer.
In fact, perfusion fermentation campaigns can often run twice as long as a fed-batch campaign, minimizing labor and materials required for campaign turnovers. STY assists researchers and developers in selecting optimal processes through normalized comparisons and in evaluating process enhancements, including intensification and facility layout optimization.
Comparing cultivation methods and host systems
Using the mathematical framework from Bausch et al.1, we assessed STY for various cultivation methods involving different manufacturing strategies and hosts. Table 1 outlines the key parameters and assumptions for the model cultivations.
The comparison includes a traditional fed-batch fermentation (Process 1) and a perfusion fermentation process (Process 2) with P. pastoris. Both processes assume a two-stage model: initial biomass accumulation at the maximum growth rate, followed by protein production after reaching maximum viable cell density. Fermentation lengths are set at 6 days for fed-batch and 12 days for perfusion, based on literature and operational experience
Specific productivity, measured in micrograms of product per gram of cells per day during production, is assumed constant across operations. The assumed maximum viable wet cell weight in the perfusion process is higher than that of the fed-batch process due to healthier cell environments, and the perfusion rate and cell bleed rate are reported in vessel volumes per day (vvd). Protein removed via the cell bleed line in the continuous process is excluded from yield calculations as it is typically discarded.
Fed batch cultivation of CHO Cells
Parameters for CHO fed-batch cultivation was sourced from Bausch et al.1 Growth and production were assumed to occur simultaneously, with a maximum viable cell density of 20 x 10^6 cells/mL and specific productivity of 25 pg/gcell/day, converted to µg/gcells/day using an average mass of 2 ng per cell. This framework allows for a comprehensive comparison of fed-batch CHO cultivation with P. pastoris fermentation methods.
Comparative analysis of fed-batch and continuous fermentation with Pichia pastoris
A detailed comparison between fed-batch and continuous fermentation using P. pastoris (Figure 1) reveals significant differences in biomass achieved during production and overall fermentation length. Continuous fermentation maintains a healthier cell environment through continuous nutrient replenishment and waste removal, resulting in higher biomass sustained for longer compared to fed-batch processes (Figure 1A).
Figure 1A & B: Biomass and Titer P. pastoris, Fed-Batch ( ) vs P. pastoris, Continuous (…….). Figure 1C & D: Cumulative Protein and STY P. pastoris, Fed-Batch ( ) vs. P. pastoris, Continuous (…….)
Figure 1B compares protein titers between the two fermentation modes. In fed-batch processes, titer increases until cell health declines after day 6, resulting in a final titer of 3.7 g/L. In continuous processes, a steady-state titer of 0.73 g/L is achieved after about 4 days and maintained for the remaining 8 days of the 12day campaign.
As exhibited in Figure 1C, cumulative protein in continuous processes is initially higher due to greater cell mass at induction. After 6 days, continuous processes produce 5.6 grams of product, over 40% more than fed-batch (3.7 grams). Given the variability in cumulative protein across cultivation scales and lengths, a normalized metric like STY is essential for direct comparison.
STY, the total protein harvested per bioreactor volume per day, is higher for continuous processes (Figure 1D). The longer the cultivation, the better the STY due to the amortized initial cell growth period.
In fed-batch (Process 1), the final protein titer represents the cumulative protein produced during cultivation (Figure 2). In continuous processes (Process 2), the observed titer does not reflect cumulative protein expressed due to continuous harvesting.
Figure 2. Cumulative Protein Harvest and Space-Time Yield
As shown in Figure 2, cumulative protein in continuous processes is initially higher due to greater cell mass at induction and continues to be higher throughout the campaign. After 6 days, the typical end of a fed-batch campaign, the continuous process produces 5.6 grams of product, over 40% more than fed-batch (3.7 grams). Extending the continuous process to 12 days results in over 13 grams of harvested protein, more than three times that of fed-batch (Figure 2).
Given the variability in cumulative protein across cultivation scales and lengths, a normalized metric like STY is essential for direct comparison. STY, the total protein harvested per bioreactor volume per cultivation day, is higher for continuous processes (Figure 2). This means that even if you were to run two back-to-back fedbatch campaigns in the same 12-day period, the 12-day continuous fermentation would still make more protein. The longer the cultivation, the better the STY due to the amortized initial cell growth period.
Comparing CHO fed-batch cultivation
Figure 3 offers a comprehensive comparison between CHO fed-batch cultivation and P. pastoris fed-batch and continuous fermentation processes.
Figure 3: Process and Productivity Comparisons
A five-fold higher specific productivity is assumed for CHO cells (Bausch et al.1), but the growth rate and maximum cell viability are significantly lower, resulting in slower biomass accumulation and less overall biomass (Figure 3A). P. pastoris processes typically achieve biomass one hundred times higher than CHO, emphasizing the need to consider protein composition, time constraints, and equipment usage when selecting an expression organism.3
As shown in Figure 3B, CHO cell titers in fed-batch campaigns are higher but require longer campaigns, affecting facility utilization and process scheduling. STY (Figure 3D) is higher for Pichia-derived processes, with continuous fermentation of P. pastoris achieving nearly three times the STY of CHO fed-batch, even assuming higher CHO cell productivity.
Space-time yield is a critical metric for normalizing and comparing productivity across various cultivation modes and host organisms.2,3,4 Among the processes modeled, continuous fermentation with P. pastoris demonstrated the highest STY, despite a five-fold lower specific productivity compared to CHO.
Higher STY leads to increased protein production within less space and time, reducing process and facility investments and operational costs. Therefore, STY should be a key consideration when designing processes to achieve translational, clinical, and commercial manufacturing goals for protein products.
Kerry Love, PhD, is the founder, CEO, and president of Sunflower Therrapeutics where Laura Crowell, PhD, is director of R&D. Stacy Martin is a strategic consultant at the company.
References
- Bausch, M., Schultheiss, C., Sieck, J.B. (2019). Recommendations for comparison of productivity between fed-batch and perfusion processes. Biotechnol J., 14: 1700721.
- Kunert, R., Reinhart, D. (2016). Advances in recombinant antibody manufacturing. Microbiol. Biotechnol., 100: 3451–3461.
- Kunert, R., Gach, J., Katinger, H. (2008). Expression of a Fab fragment in CHO and Pichia pastoris: A comparative case study. BioProcess Int., June Supplement, 34–40
- Maccani, A., et al. (2014). Pichia pastoris secretes recombinant proteins less efficiently than Chinese hamster ovary cells but allows higher space-time yields for less complex proteins. Biotechnol J., 4: 526–537.
- Crowell, L., Martin, S. & Love, K. (2024). Perfusion fermentation in the Daisy PetalTM bioreactor.
The post Enhancing Biomanufacturing with Space-Time Yield appeared first on GEN – Genetic Engineering and Biotechnology News.
