revised December 16, 2022

There is an increasing variety of bioprocesses that are used to produce an ever-widening array of products using a similarly diverse group of organisms. These bioprocesses leverage native or engineered genetics and physiology of an organism and specific cultivation strategies to produce the product of interest. The overall performance (whether it be titer, rate, or yield) of the organism is an interplay between these factors; the genetics and physiology of an organism as well as the bioprocess conditions all impact cell line performance. One important aspect of many bioprocesses is how nutrients and substrates are delivered to the culture.

There has long been an appreciation for the importance of bioprocess feed strategies. Over half a century ago, before the advent of modern genetic tools for targeted cell line engineering, several theories were introduced regarding the impact of substrate availability on cell growth. Jacques Monod, a french biochemist, described in his 1949 paper The Growth of Bacterial Cultures, the impact of substrate limitation in an aqueous culture on bacterial growth rates.¹ The Monod equation describes this phenomenon in a quantitative manner, but the two-parameter model does not account for other factors such as microbial maintenance. The concept of maintenance, described as energy utilized for functions other than creating new cell material has been described by others as a requirement to provide an adequate understanding of microbial kinetics.^2,3 Given the complexity and variety of production cell lines and the overflow metabolism characteristic of many microbes, a single maintenance parameter in a model cannot adequately describe all non-growth related functions in order to provide a comprehensive assessment of cell line performance.  

As a potential solution, the Herbert-Pirt substrate distribution relation provides a mathematical framework to describe the kinetics of substrate utilization by a microbe (Figure 1). According to this framework, substrate is taken up and can then be used for several discrete purposes: growth, maintenance (a constant), product, or byproducts.

‍

‍

How a substrate is distributed within a bioprocess provides important information on how substrate delivery may be optimized in order to improve key performance metrics. With the use of this model, it is straightforward to visualize how cell line performance is intimately related to the feed strategy. 

An optimized feed strategy will deliver the correct amount of feed in order to maximize the output of the desired performance metrics, whether they are titer, rate, yield, or a combination of the three. These performance metrics are informed by a Techno-economic analysis (TEA). On the other hand, disrupting the delicate balance between the cells and optimal substrate delivery can result in subpar performance. In the examples below, the impact of the feed strategy can be visualized in the context of the Herbert-Pirt substrate distribution relation and its impact on cell line performance metrics. 

In the first example (Figure 2a), the optimal performance of a cell is delivered using a feed strategy designed specifically for this cell type. Substrate is delivered to generate biomass, support maintenance  requirements, and allow optimal flux through the pathway, which results in high product yield. In the second example (Figure 2b), the culture is underfed. In this scenario, a disproportionate amount of substrate is dedicated to supporting the fixed costs of maintenance, leaving less substrate-to-product flux and consequently a lower yield. In the third example (Figure 2c), the culture is overfed. This results in higher substrate uptake rates than are required to support metabolism for growth, maintenance, and product pathway flux combined. This results in overflow metabolism which results in decreased product yield and increased production of metabolites that could have toxic effects on the culture. These examples demonstrate the importance of avoiding both overfeeding and underfeeding, which could lead to inaccurate conclusions that mislead bioprocess development and cell line screening.  

‍

‍

The core of bioprocess development consists of varying process parameters to identify those that have a positive impact on performance. However, many of the setpoints that can be varied in process development not only impact the factor being varied, but also impact the optimal substrate delivery rates. For example, a change in temperature setpoint may significantly alter the growth rate of a cell, and consequently the substrate demand. Without optimizing the feed rates in parallel with other process changes, results may be confounded due to suboptimal feed rates in combination with the other conditions. 

Similarly, when screening different cell lines for performance improvements, different genotypes are tested within a baseline process. These genetic changes intentionally alter the physiology of an organism to produce the product of interest, and naturally these modifications can also change the substrate demand. Similar to bioprocess development, screening a cell line in a process without adjusting feed rates accordingly can result in an incorrect assessment of the cell line's performance. 

There are a number of different approaches to optimize substrate delivery concurrently with other variables. Some of the most elegant strategies employ dynamic methods for substrate delivery, leveraging physiological cues from the cell line to signal an increase in demand for feed or for signs of overfeeding. These methods require knowledge of aspects of each organism’s metabolism, sensitive online measurements, and adjustments to identify optimal setpoints for the bioprocess. While the upfront investment in the development of a dynamic bioprocess can take both time and bioreactor capacity, the benefit of optimal substrate delivery in the context of changes in other variables can lead to simplified experimental conclusions and accelerated bioprocess development timelines. Conversely, failure to spend the time upfront to invest in developing these strategies can result in confounded conclusions both in cell line screening and process development, resulting in subpar cell line performance and extended project timelines. 

For more on feed strategies for process development, including key challenges to keep in mind and best practices, check out this resource.

References

1. Monod, J. The growth of bacterial cultures. (1949)

2. S.J. Pirt. The maintenance energy of bacteria in growing cultures Proc R Soc Lond Ser B Biol Sci, 163 (1965), pp. 224-231

3. P. Bodegom. Microbial maintenance: A critical review on it’s quantification. Microbial Ecology, 53 (2007), pp. 513-523