Use of Metabolic Modeling in Industrial Microbiology Settings
Learning Objectives
● Describe the process of metabolic modeling and how models are created.
● Provide examples of how metabolic models are currently used in industrial applications.
● Discuss future directions for metabolic modeling in industrial microbiology.
Metabolic modeling has come far since the early 21st century, as metabolic models expand outwards from model organisms such as Escherichia coli and Saccharomyces cerevisiae, as pipelines become more accessible, and as metabolic models become more complete due to advances in gene-sequencing technology that allow for better views into the genomes of organisms and the functions of those genes. Thanks to these advances, metabolic models have expanded to become genome-scale. These genome-scale metabolic models, or GEMs, have caught the attention of industrial biotechnology as a result of their ability to model whether organisms of interest can generate products of interest in fermentation tanks or whether known organisms can be metabolically engineered to generate new products or perform their functions more efficiently.
Creation of industrial GEMs usually begins with a genome sequence from the organism of interest, which is annotated to identify the genes that are present in the organism and what their potential function may be. In metabolic modeling pipelines such as KBase, this can be done using RASTtk or DRAM annotation software. These annotated genomes are then fed into metabolic modeling pipelines such as the aforementioned KBase to create metabolic models, with databases such as KEGG and UniProt used to fill in gaps left by the software and to correct errors. These models are then experimentally verified through in vivo testing to ensure the organism grows as predicted by the model.1 The stoichiometry of these models is often represented by matrices generated using databases such as BioCyc, to ensure that all reactions and involved molecules are accounted for.1 Once these models have been successfully created, their growth on different media can be simulated using flux balance analysis.2 Additionally, models can be further manipulated to look for useful industrial processes–for example, whether an organism produces desirable polysaccharides and in what yield, or whether genetic manipulations can be performed that will increase the yield of desirable amino acids.1,2 The process of creating a metabolic model has become automated thanks to software pipelines, making it easier than ever for industrial scientists to create useful metabolic models.
Metabolic modeling has been examined as a potential tool for expanding current microbial fermentation operations in the biotechnology industry. Here, scientists have used GEMs to examine the use of new organisms to produce industrially-valuable compounds, as well as to evaluate metabolic engineering techniques that could be used to increase yields or to identify potential genes to target for genetic engineering techniques. To do this, focus has shifted towards building metabolic models of new organisms. For example, a model of Brevibacillus thermoruber 423 was constructed and detailed in a 2019 paper. This bacterium is of interest since it can produce many useful compounds, such as the biopolymer EPS, isoprenoids, ethanol, butanol, and glucoamylases, and metabolic modeling techniques were able to produce a model with 1454 reactions that could be used in the future to examine metabolic fluxes surrounding EPS production as well as to potentially identify targets for metabolic engineering.1 Additionally, researchers have also examined the bacteria Corynebacterium glutamicum, building a 1207-reaction metabolic model to identify genetic modifications that would allow the organism to overproduce compounds such as L-proline, L-lysine, propanediol, and isobutanol.2 Metabolic modeling can also be used to determine what to include in growth media to maximize the growth and biomass production of probiotic lactic acid bacteria, such as Limosilactobacillus reuteri KUB-AC5.3 Thus, metabolic modeling has begun to expand in scope in the biotechnology industry, allowing interested scientists to identify useful metabolic products and to determine which alterations to the metabolism, growth conditions, or genome of a target organism could help to increase yields or growth efficiency.
Research into metabolic modeling for uses in the biotechnology industry looks to continue to expand the uses of the technique to improve its use in genetic and metabolic engineering. In addition, future work with metabolic modeling could be useful in improving process development and sustainability in the industry. Using metabolic modeling to predict the outcomes of genetic and metabolic engineering experiments would help both researchers and companies save time, money, and resources when finding new directions to explore, as long as the metabolic models generated are accurate. This could help improve sustainability, since models could be used in place of resource-heavy in vivo experiments to reduce waste.4 Metabolic models can also be used in the process development stage of industrial fermentation to develop media or growth conditions that improve the yield of a desired product or overall biomass of desired organisms or to monitor reaction conditions when combined with existing or future real-time monitoring of fermentation reactors.5 Finally, algorithms could be integrated with metabolic models to streamline them so that they are practical for industrial use and potentially to reduce error, as current models are too complex.5 Additional work still needs to be done to modify current metabolic models so that they can be used efficiently in industrial fermentation; however, should these efforts succeed, metabolic models hold a great deal of promise for industrial biotechnology researchers, as they help to elucidate the functioning of new species and allow for a more streamlined approach to developing genetic and metabolic engineering techniques.
Questions and Answers
1. What are the most common applications of metabolic modeling in industry currently?
Some of the most common applications of metabolic modeling for industrial applications are identifying the metabolic processes and products of new organisms for industrial culture, modeling potential yields under different sets of growth conditions, and examining the impact of genetic or metabolic engineering on the production of desirable metabolites. These answers were chosen because they are all discussed in the text deliverable, although a student is welcome to include another explanation if they know of one from their own work in the course.
2. What are some of the limitations of current metabolic modeling techniques?
Current models are not always accurate due to software inaccuracies, and metabolic models are not yet available for every organism. Additionally, some models are too complex to be implemented in streamlined process development plans and need to be ‘trimmed down’so that they are easier to understand. These answers were chosen based on information provided in the text.
3. How are scientists planning to expand on metabolic modeling in the biotechnology industry?
Metabolic modeling can be expanded to include new organisms, to monitor fermentations in real time, or to replace some in vivo tests that use up many valuable resources and thus to reduce a company’s carbon footprint. The models could also be made more accurate given the imperfect nature of the software used to create them, or integrated with algorithms to make them more efficient. These answers were chosen based both on answers discussed in the text and conclusions drawn from the frequency of errors in the software used in the KBase modules discussed elsewhere in the BIT495 course.
References
1. Yaşar Yildiz S, Nikerel E, Toksoy Öner E. Genome-Scale Metabolic Model of a Microbial Cell Factory ( Brevibacillus thermoruber 423) with Multi-Industry Potentials for Exopolysaccharide Production. OMICS: A Journal of Integrative Biology. 2019 [accessed 2022 Apr 9];23(4):237–246. https://www.liebertpub.com/doi/10.1089/omi.2019.0028. doi:10.1089/omi.2019.0028 2. Zhang Y, Cai J, Shang X, Wang B, Liu S, Chai X, Tan T, Zhang Y, Wen T. A new genome-scale metabolic model of Corynebacterium glutamicum and its application. Biotechnology for Biofuels. 2017 [accessed 2022 Apr 9];10(1):169.
3. Namrak T, Raethong N, Jatuponwiphat T, Nitisinprasert S, Vongsangnak W, Nakphaichit M. Probing Genome-Scale Model Reveals Metabolic Capability and Essential Nutrients for Growth of Probiotic Limosilactobacillus reuteri KUB-AC5. Biology. 2022 [accessed 2022 Apr 9];11(2):294. https://www.mdpi.com/2079-7737/11/2/294. doi:10.3390/biology11020294
4. Stalidzans E, Dace E. Sustainable metabolic engineering for sustainability optimisation of industrial biotechnology. Computational and Structural Biotechnology Journal. 2021 [accessed 2022 Apr
5. Richelle A, David B, Demaegd D, Dewerchin M, Kinet R, Morreale A, Portela R, Zune Q, von Stosch M. Towards a widespread adoption of metabolic modeling tools in biopharmaceutical industry: a process systems biology engineering perspective. npj Systems Biology and Applications. 2020 [accessed 2022 Apr 9];6(1):6. http://www.nature.com/articles/s41540-020-0127-y. doi:10.1038/s41540-020-0127-y
Creative Commons Attribution 4.0 International License (CC-BY 4.0)
