Table of Contents

  1. The Importance of Strain Screening
  2. Challenges with Strain Screening
  3. Best Practices for Strain Screening
  4. How Strain Screening Works at Culture

The Importance of Strain Screening

Strain screening is the process of testing different genetically distinct organisms within a single fermentation process in order to identify high performing strains. With advances in synthetic biology, it is routine to make many different genetic edits, and combinations of edits, when developing an industrial strain. While inserting on-pathway enzymes in a heterologous pathway is fairly straightforward, it is unlikely to advance a strain to commercial relevance. To get there, strategies such as making improvements to enzyme activity, balancing enzyme expression or addressing toxicity are usually required. Biology is complex, and it is difficult to predict the exact impact of a genetic change in a strain; they need to be tested empirically.
 
Scale-down models can be constructed to rapidly screen thousands of strains in microtiter plates, however fundamental differences in the environmental conditions can impact strain physiology and, consequently, performance. Control constraints at the microtiter plate scale include: process pH, dissolved oxygen, and substrate availability.  Because of this, a high proportion of false positives are expected from plate cultivations. Additionally, at such low process volumes, these models are not well suited for generating an in-depth analysis of strain performance since they often only yield a single endpoint measurement for titer and other interesting metabolites.
 
Therefore, even with a microtiter plate scale-down model of a fermentation, screening of strains in bench-scale bioreactors is required to have an accurate and detailed assessment of strain performance. Bench-scale bioreactors, such as Culture’s 250mL reactors, provide the process control required to screen for improved strain performance in the context of a fermentation where parameters such as pH, substrate concentration and oxygen transfer can be controlled in a manner similar to larger sized reactors. Bioreactor screening also gives a detailed assessment of strain physiology through online and time point measurements taken over the course of a fermentation. This allows different phenotypes to be visualized to capitalize on incorporating multiple desired traits into a strain or even realizing synergistic effects of combining different phenotypes. 

Challenges with Strain Screening

The overall goal of strain screening is to identify strains with improved performance in the bioprocess. While screening for improved performance may seem straightforward, there are a number of factors that can make this complex.

Changing microbe physiology
One of the largest challenges in strain screening is that genetic changes, often by design, also makes changes to the physiology of the organism. As overall performance in a bioprocess is a result of the interaction of the microbe with the process, changing the microbe without concurrent changes to the process can result in microbes being screened not performing to their full potential.
Generating statistically meaningful results
Bioreactor capacity is a finite resource, and there is always variability in fermentation KPIs between replicate runs, both within and between experiments. Structuring experiments with sufficient replicates to have statistical confidence in strain performance can limit the number of strains that can be screened. Conversely, structuring experiments with insufficient replicates can hinder the ability to make conclusions. 
Identifying nuanced but impactful phenotypes
Endpoint fermentation metrics only provide a limited understanding of the full performance of a strain. Two strains may have the same endpoint KPIs, but have drastically different performance kinetics within the fermentation. These may be of interest for different reasons, and combining phenotypes may have a synergistic effect on performance. Failure to capture these subtleties within the fermentation data can result in missed opportunities for strain improvement. 

Best Practices for Strain Screening

Best practices are centered around acknowledging and compensating for the challenges that are covered above. By accounting for the common reasons where strain screening fails, overall strain improvement timelines can be accelerated. 
 
Use dynamic bioprocesses 
Dynamic bioprocesses build upon a baseline fermentation process that adapts in response to the physiology of different strains. Features of a dynamic process can include automated feed triggers to transition from the batch to the fed-batch portion of the fermentation or using an adaptive feed rate that is optimal for each strain based on physiological cues.
 
Statistically driven replicate structure
A statistically informed approach to strain screening can ensure meaningful conclusions are possible from data generated. An examination of strain performance variability both within and between experiments as well as defined improvement detection thresholds for performance improvements and tolerance for false positive and negative errors can inform a statistically sound screening replicate structure. 
 
In-depth data analysis
Bench-scale bioprocesses can generate incredibly rich datasets, with both continuous online measurements and offline time point measurements. Online data can detect subtle changes within the fermentation such as metabolic shifts from RQ data obtained from continuous off-gas measurements on every bioreactor. Overlaying online trends with discrete time point measurements can yield deep insights into strain performance and, importantly, opportunities for improvement.

How Strain Screening Works at Culture

Strain screening is one of the most common categories of bioprocess work that is carried out at Culture. Fundamentally, the types of screening processes that can be run in Culture’s cloud bioreactors are no different than those that could be performed at other facilities. However, there are a few advantages to screening strains at Culture. 
 
Variety of adaptive control schemes
Culture has a variety of fermentation process control methods that can be incorporated into screening processes. These include automated feed triggers and dynamic control during the fed-batch based on parameters such as pH, OUR or RQ. Customers can specify an adaptive control scheme, and Culture will build it into the bioprocess recipe. 
 
Capacity 
Culture’s high-throughput bioreactor capacity means that there does not need to be a tradeoff between structuring replicates based on statistics and the throughput of strains. Culture’s capacity enables screening larger numbers of strains in a shorter period of time. 
 
Precision fermentation control 
Each of Culture’s reactors is equipped with several features to ensure precision fermentation control, reducing biological variability as a consequence of imprecise process control. Lower biological variability between replicates can also increase the statistical confidence in results. These features include gases delivered via mass flow controllers and up to 5 feeds delivered via peristaltic pump with scale feedback control. Published mass balance data on every run also ensures quality data. 
 
Accessible data and time to analyze it
Culture’s Cloud Console allows you to design, monitor and analyze experiments from your laptop. Console enables access to data from single or multiple runs and allows for grouping and graphing of data for easy analysis. Graphs can be saved and shared in workspaces to allow rapid analysis of strain performance and collaboration across groups. In addition, as Culture manages the execution of experiments, time can be spent analyzing data rather than troubleshooting reactors.